aids to drawing technique

It's fair to say that photographs have

become the principal aid to drawing used by artists not a new idea
today. A credible style of contemporary painting
(photorealism) consists of the meticulous copying copying the image
of photographic documents, and this style has
challenged other "realist" painters with a distinct tracing the image
visual world and color sense. Artists also trace
photographs onto the support as the foundation documenting with photographs
drawing for paintings done in a less realistic style.
Even a nonrepresentational artist's "visual culture" creative editing with
is enriched by the sense of light, contrast, color, photographs
saturation, focus, perspective, depth of field, optical
curvature, movement, blur, flare and halo that representing the world
have been ingrained in us all through daily
exposure to the products of optics, photography,
video, film and computer design.

My aim is to fit some of these themes into the

methods artists use as aids to drawing. I describe
four basic activities. Most important is the use of
photographic materials for foundation drawings.
These materials can also be manipulated to model a
painting effect that is not imagined in photorealist
terms. Photographic media can be used in the
experimental or playful search for new visual
effects or visual compositions in painting, by image
collage or image editing. Finally, the artist can use
photographic media to clarify or analyze a specific
visual problem.

not a new idea

The idea of using optical aids in
drawing or painting is quite old. The traditional
account is that the Florentine architect Filippo
Brunelleschi (1377-1446) worked out the
principles of linear perspective by meticulously
painting the view from a room on a windowpane.
And several woodcuts from the 15th century show
artists studying the perspective of objects by
viewing them through grids of wire or scribed
glass; the outlines of objects were copied onto
paper or canvas, square by square, as the
foundation drawing for a finished work.

Traditional Tools. The first practical optical device

was the camera obscura or "dark room,"
mentioned as early as the Roman era (in a treatise
on painting by Pliny, 50 AD), but especially popular
from the 16th to early 19th centuries.

The original camera obscura was literally a small

room with a rotatable metal mirror on the roof. The
mirror reflected the surrounding landscape
downward through a small hole in the roof, where it
could be viewed on a flat white table underneath.
By the 17th century these rooms had been reduced
to a portable wooden pyramid (about the size of a
modern washing machine), with a mirror at the
apex and a opening in one side that allowed an
artist to trace the projected image on a sheet of
paper inside.

These cameras obscura had two important effects

on artistic practice: they provided a tool for tracing
or copying complex images, and they created a
model of visual fact — in the outlines, colors,
values and even optical distortions of two
dimensional images — that affected artists'
expectations of painting and perceptions of reality.

Recent art historical research by Martin Kemp has

expanded our appreciation of the linkages and
overlapping interests of artists development
of linear perspective and scientific studies
through a variety of optical, surveying and "image
recording"

More recently, the expat British artist David

Hockney has marshalled evidence for the use of
cameras obscura, curved mirrors and lenses by
European artists startin with Dutch painters of the
early 15th century. (He suggests many early
painters used a convex mirror to reflect a brightly
lit image seen through a small window.) Paintings
by Jan Vermeer (1632-1675) of Delft were almost
certainly created using a camera obscura (probably
a large booth, using Dutch lenses rather than
mirrors): the highlights in many of these have the
peculiarly rounded, disklike appearance that can
only be produced by the magnifying effect of an
optical lens. More recently, new research
on Thomas Eakins has confirmed he extensively
relied on tracings of photographic images to
compose his paintings.
By the late 18th century, the camera obscura was
the size of a small suitcase (illustration at right)
and the optics had improved sufficiently so that the
device could be used in the field. An adjustable lens
focused the image onto a mirror at the back set at
a 45° angle, which reflected the image upwards
onto a flat sheet of glass, shielded from daylight by
a folding screen, where the image was traced onto
paper. Use of the camera obscura was also much
more out in the open — less secret, and employed a portable camera obscura
in landscape or topographical work. (c.1800)

A different instrument became popular at the

beginning of the 19th century — the camera
lucida, commonly described as an invention of the
English scientist William Hyde Wollaston in 1806
(he patented it as a portable drawing device in
1807). This consisted of a partially silvered prism
that reflected or transmitted the images from two
different directions; the prism was held by a
mechanical arm clamped to the side of a drawing
table or drawing board. The artist sat at the table
making a portrait with
or placed the board in his lap and looked into the a camera lucida
prism with one eye (illustration at right). The prism (c.1810)
was adjusted so that the transmitted image of the
object to be drawn was superimposed over the
reflected image of the drawing hand on the paper;
this allowed the artist to trace the outline of the
image on the paper. Cornelius Varley, the architect
brother of John Varley, patented in 1811 a camera
lucida that embedded the prism within a small
telescope. This presented through a single eyepiece
the reflected image of the paper actual size, and
the transmitted view the image to be drawn under
limited magnification.

These tools were difficult to use in the field, but

hardly impractical. John Cotman used a camera
lucida for his series of English architectural etchings
(actually, a female assistant prepared a basic
drawing for him, as he had trouble using the tool
himself), and many topographical artists used
similar instruments as well. The camera lucida was
not used to trace all the outlines in a drawing, but
rather to notate quickly the major corners, edges
and curves of the important forms (the work of a
minute or less), which were then connected
together freehand.

The New York critic Susan Sontag commented, "If

David Hockney's thesis is correct, it would be a bit
like finding out that all the great lovers of history
have been using Viagra." Cute, but hardly accurate.
Art is the activity of creating compelling imagery as
physically valuable objects — not simply performing
manually difficult graphical tasks. Artists use what
they can as well as they can, including any fruits of
contemporary technology available to them. It's
their imagination, passion and attentive care — not
their tools — that make them great lovers.

David Hockney's visual and documentary evidence for the

use of optical aids in western painting from 1430 onwards is
reviewed in Secret Knowledge: Rediscovering the Lost
Techniques of the Old Masters(Viking Studio, 2001). This
includes his correspondence with other art experts, some of
whom are converts to his theory.

copying the image

Artists today have abandoned these
traditional tools as ineffective and cumbersome.
When optical aids are required, they typically
photograph the scene with a camera, then use any
of the methods described below to transfer this
image in the studio. The drawback to this method is
that it forces reliance on the photograph for most
or all judgments of lighting, value and color: it
eliminates direct sensory impressions from the
creative process.

The most common remedy is to photograph the

scene, but also to make color studies and design
sketches on the spot, then combine these notes
with the photograph to make the final painting.
Full Scale Methods of Copying. Perhaps the most
direct approach is to prepare a full size sketch or
life drawing, then transfer this cartoon to the
support at a 1:1 scale.

A convenient and quick alternative is the "flip"

method: I lightly tape the cartoon to the support at
the edge opposite my drawing hand, and with my
free hand repeatedly flip the drawing over and
away from the support, each time marking with a
pencil the location of important lines, corners or
mass centers. With a little practice this shadowing
(it's not really a tracing) can be done quickly and
accurately. The trick is to fix your eye steadily on a using the "flip" method to copy
specific point in the drawing, then flip the cartoon a full size drawing
away to mark it with your pencil, then check the
location by flipping the drawing over it again.

Neither of these methods should be used to trace

the drawing exhaustively. This leads to a copy that
looks lifeless and stiff. Instead, mark out the most
important landmarks and recreate the drawing
freehand on the support. This allows you to correct
minor disproportions in the cartoon, and bring
everything together with a unified sense of gesture
and control.

"Squaring" Methods. Traditional methods that

still come in handy involve tracing or "squaring" an
opaque image, such as an etching, drawing,
finished painting, magazine illustration or
photographic print.

The simplest method is tracing the image. Lay a

sheet of thin vellum (available at most art stores,
and at many stationery stores in the drafting
supplies section) over the image to copy and trace
it with a pencil. Then transfer this tracing to the
drawing or painting surface. To provide a firm
support and to keep the tracing from slipping
during work, both the drawing paper and the
overlay of vellum should be securely taped or
stapled to a drawing board. The transfer can be
done in one of four ways:

1. The tracing is lifted slightly and the pencil is

moved freehand underneath the sheet.

2. The tracing sheet is laid drawing side down on

the drawing surface, and rubbed with a burnishing
tool over all drawn lines. This transfers a faint
impression of the line to the drawing sheet (but
reversed left to right).

3. The back side of the tracing sheet is rubbed with

a graphite pencil over all the traced lines, then laid
drawing up on the drawing sheet. All the drawn
lines are traced with a pen or pencil, and the
pressure transfers some of the graphite on the back
of the sheet onto the drawing surface.
4. A sheet of carbon paper can be placed between
the vellum and the drawing surface, and the lines
traced the same as in 3.

There are two drawbacks to this method. The major

drawback is that the drawing must be copied
actual size - you cannot enlarge or reduce the
drawing or change its proportions in any way. The
second is that the transfer pressure of the tracing
pencil tip often leaves depressed lines in the
surface of the paper, which can be difficult to cover
completely when the paper is painted (the brush
does not fill the depression, or the depression
appears as a dark line in the painted surface).
The alternative method, used since antiquity, is
to square the picture, and it requires no special
equipment or messy copying medium.

first step: squaring the original picture

The first step is to rule a regular grid over the
painting to be copied. The total number of squares
required will depend on the variety and complexity
of shapes in the picture to be copied, and on the
degree of accuracy you require — smaller squares
make the copying more accurate, but also more
time consuming.

You can also rule a coarse grid over the entire

drawing, and subdivide one or more of the large
squares to capture the details in a local part of the
drawing, as shown above.

Although no special equipment is necessary, you

can save time with an overlay. Many drafting
stores sell vellum sheets preruled with a grid, and
on strongly contrasted designs, one of these sheets
can be overlaid on the drawing to provide the grid.
Some map stores sell sheets of acetate preruled
with black lines for reading the intermediate
distances on topological maps and the like; or you
can buy a sheet of blank acetate and rule or scratch
the lines yourself. These transparent sheets work
best to copy dark, faded or delicately toned images.

Not only do these overlays save time, they avoid

having to draw the grid directly on the surface of
the photograph or drawing, which scars it
permanently. (If it's an image you expect to copy
many times, this may be preferrable.)
 squaring and noting the drawing surface

The second step is to rule the drawing surface.

This grid must be drawn as lightly as possible, so
that it does not emerge through the painting, and
does not visually distract you from drawing the
lines of the image. I sometimes use medium gray
aquarelle pencils to draw this grid, because the
lines dissolve when paint is applied over them and
the color is very faint.

Ruling this grid gives you the opportunity to change

the format of the image. If you want to reduce the
size of the image, rule the squares smaller than
they are on the original; to enlarge the image,
make the squares bigger. You can also change the
proportions. To make the landscape image "wide
screen," just make the grid spacing rectangular,
with the horizontal spacing of lines greater than the
vertical. To reduce the size of the foreground, rule
the bottom row of squares as rectangles, but leave
the rest of the picture in square.

The third step is to note the major line

crossings of the image. I find it's easiest to work
systematically: for example, from top to bottom of
the image, noting the major in the horizontal lines;
then left to right, noting the important vertical
lines, and finally adding any diagonal lines or
details. This helps to avoid omitting important lines,
or counting off squares in the grid to identify where
a line is located.

All this is done freehand, by eye. You simply make

a small mark on the drawing at each point where a
line in the image crosses a line in the grid. You can
also use short lines to indicate the slope and
direction of a line. To improve your accuracy in
estimating the location of a point, lay the point of
your pencil on the center of a line to divide it into
two, and locate the point in the half segment.
 connecting the guidemarks with lines

The final step is to connect the guidemarks in

the drawingas living lines. Again, I work down the
page constructing all the horizontal lines, then
across the page working on the verticals. Unless
accuracy is important, it's best to use the
guidemarks as reminders of how the line is shaped,
but deviate from these in the interests of design,
simpification or emphasis.

tracing the image

Modern artists have adopted a
variety of optical aids to assist in drawing or
painting.
Opaque Projectors. Most popular are a variety
of opaque projectors or magnifying
projectors (which I used and loved in my
childhood under the name Magnajector). The
current brand names include Artograph, Trace-
Master, Seerite and Kopykake; prices range from
around $50 up to $630, depending on the quality of
the optics, magnification power, illumination source
and ventilation system.

Object Projection. Methods for projecting the

whole object. In former times a camera obscura
worked for this purpose.

two opaque projectors

Artjector student portable model (top); Kopykake

tracing the shadow of a botanical
professional table top model (bottom)
specimen
Microscopes have also been used as projection
tools, for example in anatomizing microrganisms,
minerals or living tissue.

Slide projectors. With the new era of digital

cameras and displays the technology of the slide
projector has slipped into antiquity. When these are
available with slide documentation or source
materials, they can be used in the same way as an
opaque projector.

documenting with photographs

Photographs had a significant
impact on artistic perception because they could
visually resolve aspects of the world that were
difficult or impossible to see with the unaided eye.

The watershed texts are the many series of

analytical photographs made by the victorian
photographer Eadweard Muybridge (18-18), who
published, as Animal Locomotion(1887), books of
human figures and animals in motion,
photographed in rapid sequence against an
analytical background grid. It was Muybridge who
definitively settled a longstanding debate as to
whether a trotting horse is ever completely
suspended in the air during its gait (the answer is
yes).
Flowing water, rippling reflections, the colors of
twilight, the wings of the hummingbird and the
shape of exploding fireworks are some of the other
visual delights that have been exactly rendered in
paintings based on the evidence in photographs.
And most textbooks on linear pespective will reprint
a photograph of railroad tracks or urban
skyscrapers as "proof" that perspective is "real."

creative editing of photographs

One of the most exciting
possibilities that has been opened up by digital
image processing — either directly, with a digital
camera, or indirectly, by scanning a photograph or
magazine made in the regular way — is the ability
to use digital media as a creative tool.
 using image editing to reduce detail

The example shows the differences between the

original image (left) and the image processed with
the Adobe Photoshop "pixelate/facet" utility (right).
The pixelation reduces the image to an irregular
mosaic of color, simplifying forms and eliminating
detailed or complex textures (note the difference in
the wisteria blossoms on either side). I find this is
particularly useful to analyze shapes into large
areas of color, and to reduce complex, detailed
textures into simplified, schematic patterns that are
easier to paint and to weave into decorative effects.

If you have access to a powerful image processing

program, and are willing to learn how to use it, you
can greatly expand your compositional creativity.
This next example shows how I revised an
impromptu landscape photograph into a base image
for a landscape painting.

revising image composition: format

proportions

Photoshop permits the insertion of a transparent

layer, which can be used to draw guidelines or
registration marks for shifting image elements; this
layer can be blanked out so the image can be
evaluated without it. The image above shows the
landscape with the grid of format
proportions superimposed on it.

The format proportions suggest that the central

cluster of trees can be broken up, and the far
mountains lowered and reshaped to define the
second horizontal. The foreground darks can be
lowered somewhat, and the image of the road
increased. These changes would all have the effect
of increasing the sense of depth and space in the
picture, and thereby open the composition to a
more emphatic statement of the light flowing from
left to right.

revising image composition: before and

after

These comparison images show the landscape

photo before and after editing. The changes are
subtle, but I see a general strengthening of the
sense of recession and perspective, a more graceful
disposition of masses, better statement of the
relationship between the road and the groups of
trees in the middle distance, and a better pattern of
alternating dark and light bands across the
landscape.

The image processing program can also be used to

alter values or hues, change color schemes, enlarge
or reduce individual objects, reverse objects left to
right or top to bottom, and copy a single individual
object (such as a tree branch) multiple times to
create a complex pattern or texture. Notice that all
these changes are what and artist might naturally
try out in a value sketch or color study: the
difference is that the changes can be made more
quickly, and multiple versions of the same image
can be compared simply by turning on or off the
transparent layers containing each edit.

Landscape Formats. Many contemporary artists

are so accustomed to accepting a photograph as
the painting representation that they don't consider
the many ways that a photo can be altered for
dramatic effect.

Landscape painters traditionally "adjusted" the

proportions of objects and distances, sometimes by
violating or exaggerating linear perspective rules,
to produce a more compelling landscape image.
These exaggerations are part of the landscape
tradition. I often use simple Photoshop
transformations to produce a similar effect.

Here's an example from a photo of the nearby

California coast, made with the 18mm-70mm zoom
lens that comes standard on my Nikon D70
prosumer digital camera. (I adjusted the lens to
about 30mm, or wide angle, to make this shot.)

Most digital photographs are now roughly in the

format proportion 30:20, which is close to the
format of the standard full sheet (30:22), shown as
the added black at the bottom of the image.
However the technique I describe works for any
format, provided you first crop the photo to the
image you want, then take that image width to
equal the width of the paper format you will use for
the painting.

landscape photo against full sheet format

In Adobe Photoshop a simple way to make the

formats match would be to select the entire digital
image (Command+A), then use the Free
Transform utility (in the Edit pulldown menu,
or Command+T) to increase only its height by
110%. However this won't eliminate the basic
shortcomings of the original image. The distant
cliffs appear farther away and lower in height than
they impress the viewer in real life, while the
foreground rocks are proportionately too large. The
viewer seems to look across rather than down onto
the surf, and the surf around the promontory
seems flat and lifeless. The foreground space
seems to flow away on either side, as if running out
the bottom corners of the frame.

Here's the cure. Use the Rectangular Marquee

Tool to select the entire bottom of the image, from
the base of the main middle distance form to the
bottom edge (orange rectangle). Then use Free
Transform to resize only this part of the image
upwards (orange arrow), judging by eye the
amount of foreshortening that looks best in relation
to the height of the middle distance form. (A
reduction of one third, or 66%, is usually pretty
effective.)
 vertically compressing the landscape
foreground

Finally, select the entire image area and use Free

Transformagain to pull it downwards until it fills the
entire full sheet format area. This largely
compensates for the previous foreground
compression, but vertically elongates the middle
distance and background forms.

altered photo resized to full sheet format

Now the distant cliffs rear up and approach nicely,

and have more weight in comparison with the
foreground rocks. The horizon has been lowered,
increasing the proportion of sky to landscape forms.
The foreground beach seems to flow under the
viewer's feet, and the viewer is pulled into the
center of the image. The middle distance surf
appears closer and, for that reason, more lively. All
the landscape elements are pulled into a tighter,
more dynamic relationship, in part by violating the
perspective rules for wide angle images that push
middle distance objects farther away.

To help you visualize what's been done here, I've

performed exactly the same transform on a
perspective gradient and square forms (left),
producing the altered perspective space (right). The
horizon has been dropped and the orthogonals
beyond the transformation boundary have been
tilted upward. Foreground recession and object
proportions have been slightly compressed
vertically.

perspective gradient before and after

transform

This method translates into a simple drawing

procedure when working in the field: elongate
distant forms vertically (to make them appear
larger and closer) and vertically compress the
foreground (increase the foreshortening) by a
proportional amount, without changing (or slightly
flattening) the proportions of foreground objects.

What's most intriguing is that people familiar with

the specific landscape you've painted will prefer the
transformed image but will not notice that it has
been distorted in any way. This is even true for
many buildings or architectural forms, which look
more dynamic or impressive when distored in this
way.

If landscape buildings or figures appear

unacceptably distorted after the transform, simply
select and copy that building or figure from the
original image, make the standard transform on the
surrounding space (landscape), paste the original
shape over the distorted one, then free transform
the original building or figure image until it has the
size and proportions you like best in context.

As you see, it's practical and sometimes powerful to

select different parts of the image, resize,
transform and reposition them in different ways,
then paste them together as a completely new
image. You can also horizontally expand certain
figures, for example only the boxers in a boxing
ring, to make them appear more massive. (Martin
Scorcese shot the boxing scenes in Raging
Bull using boxing rings of different sizes, in order to
produce different perspective and size effects for
the camera.)
Analysis of Form and Light. Photographic media
can also be used to analyze a great variety of visual
problems, in particular by exaggerating or
simplifying a complex image in ways that allow it to
be painted in a nonphotographic style.

Your resources will depend on the kinds of images

and photoediting software available to you. Nearly
all digital cameras now are sold with simple
photoediting software, and professional painters
may find it worthwhile to invest in an industrial
grade program, such as Adobe Photoshop. (Most
important: play around with the program you have
to learn what is possible.)

Here is a simple analysis example that addresses

one of the most basic landscape problems: painting
a tree.
 analyzing a tree using Photoshop image
edit functions

top left: original image; top right: filter > pixelate

> facet; bottom left: filter > brush strokes >
accented edges; bottom right: filter > artistic >
cutout

The image at left is probably legible to an

experienced painter, but for many painters starting
out in landscape painting the tree will be an
amorphous puzzle — they can't see the tree for the
leaves. Photoshop lets us boil the form down to its
essentials.

The specific method I used was: (1) reduce the

original digital image to a suitably small format
(this is some part guesswork), (2) simplify the
image structure using the Pixelate: Facet utility,
then (3) boost both the brightness and contrast
(with the Adjust Image: Brightness/Contrast utility)
to obtain a clear definition of the form. (Blurring
and then contrasting the image can produce similar
results.)

Transformed, the image resolves into interlocking

forms that can be obviously painted as a jigsaw of
flat color areas. The dark "holes" where we can see
into the interior, deeply shaded part of the tree are
clearly located, the terminator between lighted and
shaded exterior leaves is accurately defined, all the
color areas resolve into a basic color containing
random touches of lighter, yellower or redder
accents, and we even discover an unexpected detail
— those near white reflections in the topmost
branches.

Cameras are not the arbiters of your artistic vision

unless you choose to make them so. Yet it's
incorrect to say that these computer manipulations
are different ways of representing reality — reality
is only "represented" within the mind, not outside
it. I've described methods of filtering or
distorting the rich information presented to us
by the world, via made images, in a way that
affects the viewer's perception in artistically
desirable ways.

representing the world

When David Hockney began is
research into the early use of optical projection
devices, he was initially struck by the unique
appearance of drawings and paintings that might
have been traced from the images projected by
lenses or mirrors. Drawings lacked the hesitancy,
trial and error lines and erasures; paintings had
unique qualities of physical detail.

With photographs, too, new areas of visual detail

finally could be realized: the gait of a horse or the
leap of a cat, the weave of watery reflections.
At each step, artistic application of the technology
also inserted clumsy misinterpretations: paintings
showed peculiar lapses in figure proportions or
posture, because the optical devices required
separate, limited projections fitted together. An
early painting by Thomas Eakins, of horses trotting,
is absurdly unreal.

The point is that the artist never surrenders the

responsibility of choosing the informative
moment and view used in the image. A
photorealist painting of a diver at the beginning of
his release would look as comically peculiar as misjudged photorealism in Thomas Eakin's "A May
Eakins' trotting horses: the diving board curving Morning in the Park" (1880)
downward under his feet, both arms out and knees
flexed, as if steadying for a fall.

Actions have their moments of inflection, of

emphasis, of maximum dynamic potential: the dive
becomes visible when the diver in the air, not when
he is in his preparatory squat. At the end of parking
a car, the driver is merely sitting in the car, as if
waiting for someone to arrive; at the end of a fight,
the assailant may approach his prostrate victim in
the same way a good samaritan might approach
someone who has fallen accidentally. Artists choose
the moments that make things clear, or
intentionally prosaic and ambiguous, and this is
something the technology can never do.

It's wrong to say that these are different ways of

representing reality — reality is only "represented"
 within the mind, not outside it. Rather, they are
ways of filtering or distorting the rich
information presented to us by the world,
summarizing and abstracting reality in ways that
provide esthetic impact.

Last revised 08.I.2015 • © 2015 Bruce MacEvoy

elements of perspective technique

This page and the pages linked to it comprise the
most extensive treatment of linear perspective
available online, and one of the most comprehensive
tutorials available in any textbook currently in print.
New edition, revised and expanded 07/2014.

The technique of linear perspective allows artists

to simulate or construct the appearance of three
dimensional space on a two dimensional surface. It is
one of the major innovations of European art, with
an extraordinary impact on western visual culture page index
from the 15th to the 19th centuries.
perspective in the world
Linear perspective is the artist's slide rule, an antique
technology that has been replaced by film or digital central perspective
photography, opaque projectors, and computer
assisted design and animation software. It lives on two point perspective
primarily in art historical monographs, architectural
visualizations and the freewheeling digital mayhem three point perspective
of Grand Theft Auto.
advanced perspective techniques
So why bother with it? Because knowledge of
perspective greatly enhances your perception and shadows, reflections &
understanding of light and space, and attunes you to atmosphere
spatial recession as the power line of visual design. It
is a reliable guide to drawing in all situations, and a
fascinating case study of the ways that a painting is
shaped by purely conceptual considerations. It is also
indispensable to understand the design problems
that inspired and challenged artists of the past.

Although the Greeks and Romans drew foreshortened

limbs and architectural forms, the geometrical tools
necessary to construct a consistent illusion of spatial
depth across the entire picture plane were first
developed and used by 15th century Florentine
artists with strong mathematical skills: the visionary
architect Filippo Brunelleschi (1377-1446) and the
humanist scholar Leon Battista Alberti (1404-
1472). They systematized the Tuscan workshop
methods of pseudoperspective or near perspective
which were already on display in frescos by Giotto
and Duccio. It's interesting that technically advanced
northern Italians were working out the geometry of
linear perspective at about the same time that
optically sophisticated artists in the Low Countries
were adopting projection devices to create more
realistic pictures. Perspective was just one aspect of
a new artistic realism that was created by many
practitioners in separate regions of Europe.

In 1435 Alberti wrote the Latin manuscript De

Pictura (On Painting), which he revised and
translated into Italian as Della Pittura in 1436. This
was the first book to describe perspective
methods as part of the overall design of a painting,
and within a century artists such as Piero della
Francesca (c.1470) and Leonardo da Vinci (c. 1490)
in Italy, Albrecht Dürer (1525) in Germany and the
architect Jean Pélerin (1505) in France wrote more
ambitious perspective studies. During the 16th and
17th centuries linear perspective diffused across
Europe and was honed through the analysis of
specific representational problems, in particular
foreshortened human figures, the visualization of
intricate geometrical solids (often realized as
engravings or as inlaid wood designs called intarsa),
architectural scenery for plays and operas, frescos
representing great spatial depth, and anamorphic
projections (which appear "normal" from an oblique
angle of view, or when viewed in a curved mirror). In
the 17th century these refined LP methods were
absorbed into and thoroughly grounded by the
science of projective geometry, and analytical
geometricians began to investigate perspective
issues unrelated to any artistic practice.

No matter: artists continued using perspective in a

variety of innovative ways, developing practical
shortcuts for constructing perspective drawings, and
refining the perspective rendering of architectural
forms in combination with the free
perspectiverepresentation of figures and objects. By
the 18th century perspective was ensconced as a
core study in art academies, which transformed
perspective techniques into a mechanical orthodoxy
that contributed to the extinction of strict perspective
styles of painting. Early 19th century artists such
as J.M.W. Turner made interesting explorations in
optical color mixing and aerial perspective, preparing
a complete break from perspective traditions late in
the 19th century.
In the traditional account, Brunelleschi either
discovered or verified linear perspective by
literally painting views of buildings onto a
mirror or windowpane. This "draw on glass"
approach was probably not the method Brunelleschi
actually used, but it fits well with the idea that a
perspective painting is a mirror or window view of
the world, and it soon became the standard way to
convince drawing students that perspective really
works, even when the scope of the problem is no
bigger than a table top.
 one more convert to perspective

from Charles Hayter, Introduction to

Perspective (1813)

Want to try it? Punch a small hole in an index card,

staple or tape the card to one end of a wood
yardstick, then tape the other end of the yardstick to
the back of a chair, so that the hole in the card is at
your eye level as you straddle the chair. Put the chair
in front of a large window, sit, and look through the
hole in the card with one eye. Use an erasable
marker pen or grease pencil to draw on the glass the
buildings or objects in view, then hold a white paper
behind the glass to see your finished design. As you
draw, you may notice that this procedure feels
confining. That's because it is — illustrating how far
the assumptions of perspective differ from our fluid,
constantly changing visual experience.
 a renaissance perspective machine

from Albrecht Dürer, Underweysung der Messung mit

Zirckel und Richtscheyt ... [Instruction How to Measure
with Compass and Straight Edge ...] (1525)

This peephole or "peep show" tracing method does

not make clear that linear perspective is really
created by straight lines passing through a
vertical plane. For that, the perspective "machine"
(diagram, above) by the German artist Albrecht
Dürer (1471-1528) is a better example. By looping a
weighted string through an eye ring in the wall, one
artist could hold the opposite end along the contours
of a lute; a second artist could measure the height
and horizontal position of the string where it passed
through a rectangular frame, then transfer this point
to a paper or canvas (shown as a hinged panel). By
repeating the measurements dozens or hundreds of
times, a "connect the dots" perspective view of the
three dimensional lute could be constructed on a two
dimensional surface. Through similar empirical
methods, Renaissance artists worked out the
geometric fundamentals of perspective technique.

Dürer's machine demonstrates that linear

perspective explicitly depends on a single point
of view in space. A perspective drawing of a
building or landscape reveals the location, vantage
and orientation of the viewer as precisely as it shows
the physical form of the objects in view. The eye ring
(diagram, above) represents the eye of the viewer
or center of projection; the stretched string
represents the straight beams of light or visual
rays that converge on the eye from all objects in the
field of view; the hinged surface is the artist's canvas
or image plane. Everything comes down to that
point of view defined by the wall ring or hole in index
card: the point of view, not objects in space, is
the fundamental perspective theme.

Linear perspective is fundamentally a geometrical

method, not a mathematical one — which
means you draw, you don't calculate. All you need
are the standard construction tools from high school
geometry: a straight edge (ruler), pencil and a
compass (or a long strip of cardboard and some push
pins). Unfortunately, these tools are awkward to use
when the perspective view is extreme or the drawing
is very large, and they make quantitative discussion
of perspective issues difficult. So I explain numerical
calculations, reduced drawings and other tricks that
can help you through those problems. But
calculations are not necessary to apply LP methods.

There is a sprawling and stale literature on linear

perspective, but two points deserve mention. As
James Elkins and Martin Kemp point out, the
historical uses of perspective construction have
always been opportunistic and evolving. Many
seemingly "obedient" perspective paintings reveal,
on closer examination, "fuzzy" or multiple vanishing
points — some objects in the image are "in
perspective" and other objects are not. Most artists
chose to make "corrections" to apparently
inconsistent perspective distortions, often in
pursuit of a more pleasing finished composition:
linear perspective often creates as many design
problems as design solutions. In that limited sense,
LP is not any better or worse for making "realistic"
images than the various types of parallel or paraline
perspectives used by architects and engineers —
elevation, section and plan, or military, isometric and
dimetric projections. In all these, similar distortions
appear and are easily ignored. The moral is
that perspective must be used gently: it is a very
cool drawing tool and a compelling representational
convention, not the ultimate code of drawing
correctness.

The best approach is to rely on LP to lay out basic

proportions and shapes, to clarify difficult drawing
problems, and to guide your intuitions about forms
and textures in space. Freehand perspective and
trust in your visual judgment should not be pushed
aside, but nourished and exercised by the insights
linear perspective can provide. As Vasari wrote of
Michelangelo, "he held his compasses, that is to say
his judgment, in his eyes and not his hands."

I've divided these diverse materials into separate

pages, and occasionally use nonstandard terminology
in the interest of clarity. I hope to provide a practical
and provocative reference to the topic.

References. There are many books on linear perspective,

and they are wildly uneven in quality and practicality. For
beginners, Perspective Made Easy by Ernest Norling
(Norton, 1967), first published in 1939, provides a very lucid,
entertaining and highly practical general introduction. John
Raynes's A Complete Guide to Perspective (Collins &
Brown, 2005) introduces perspective problems through an
attractive selection of photographs and diagrams, but is
rather informal about construction solutions, especially for
shadows and reflections. The discussion in Perspective for
Artists by Rex Vicat Cole (Norton, 1976) is a more extensive
and practical presentation, though somewhat out of date. The
most comprehensive practical instructions, tailored to the
tools and methods of architects and draftsmen, can be found
in Perspective Drawing: A Step-by-Step Handbook by
Michael E. Helms (Prentice Hall, 1997), available as a
facsimile reprint. Another useful though occasionally
inaccurate guide is Perspective: From Basic to Creative by
Robert W. Gill (Thames & Hudson, 2006). The elegant "circle
of view" approach adopted here is based on Linear
Perspective: Its History, Directions for Construction,
and Aspects in the Environment and in the Fine Arts by
Willy Bärtschi (Van Nostrand Reinhold, 1981), regrettably now
out of print but probably too formal for most artists.
The following sources expand on the historical and artistic
context for perspective techniques. The Science of Art by
Martin Kemp (Yale University Press, 1990) is a careful study
of the historical development and use of linear
perspective. The Poetics of Perspective by James Elkins
(Cornell University Press, 1994) is a stimulating if academic
study of the development and artistic impact of perspective
techniques from the early Renaissance to early Baroque. E.H.
Gombrich's classic Art and Illusion(Princeton University
Press, 1960) is the most articulate defense of linear
perspective as an accurate description of the visual world. On
that theme, Optics, Painting and Photography by M.H.
Pirenne (Cambridge University Press, 1970) experimentally
(and with clear photographic examples) explores and confirms
the accuracy and limitations of perspective as a description of
natural human vision. The most famous dissenting view is
Erwin Panofsky's Perspective as Symbolic Form(1924;
Zone Books, 1997), treacherous for the uninformed but an
amusement for adepts. There are also many web pages
devoted to perspective, most of them useless; start with the
online translation of Alberti's Della Pittura, which is available
in paperback from Penguin Books. There are many books
describing perspective for use in CAD programs, which are
less useful for a painter.

PAGE INDEX

Page 1. Perspective in the World

The Texture of Space. Perspective originates in the

common appearance of the real world, where texture
is as important as geometrical buildings in our
experience of distance.

Four Perspective Facts. Everything in perspective

arises from four perspective facts — the straight line
path of light, a fixed viewpoint, the visual cone
centered on the direction of view, and the image
cross section through the visual cone.

Creating the Perspective View. The four

perspective facts provide the basis for a framework
to represent the three dimensional world on a two
dimensional surface.

The Perspective Setup. We discover the structure

of the perspective view by "looking at" a metric grid;
this perspective structure becomes the 90° circle of
view framework that allows us to solve any
perspective problem.

Basic Rules of Perspective. Nearly all perspective

constructions can be explained in terms of 15
perspective rules (and a perspective glossary).

Image Plane, Viewpoint & Direction of View.

The orientation of the image plane to the viewer's
head is governed by the display convention, viewing
convention and projection assumption; changing the
direction of view changes the perspective in the
image; the height of the viewpoint is always shown
by the virtual horizon line.

Perspective Distortions. Viewing a perspective

image from any location other than the center of
projection creates distortions in the image objects:
these are "cured" by reducing the circle of view, or
can be manipulated for expressive effect.
 Page 2. Central Perspective

The Visual Ray Method. Artists and

mathematicians of the 18th century discovered the
"double fold" that allows perspective problems to be
analyzed in terms of visual rays from the viewpoint
to the plan view of an object.

One Point Perspective. In 1PP or central

perspective, parallel lines define horizontal and
vertical dimensions and a single vanishing point (the
principal point) defines recession in depth.

Orthogonals & Central Recession. Orthogonals

define the viewer's central recession, which is
unaffected by the orientation of objects in view; they
project a unit dimension in depth and, with the
diagonal vanishing points, can define perspective
distance from the viewer.

Slanting & Sloping Planes. Simply by rotating the

circle of view or raising the horizon line, central
perpective can represent slanting or sloping surfaces.

Perspective Gradients. Perspective controls the

appearance of textures or patterns on all natural and
artificial surface: object distance and ground height
on the viewing plane are inversely proportional.

Distance & Size. Triangular proportions are the

geometric core of perspective; using them we can
define the exact size of objects we want to draw,
provided we know the viewing distance to painting
and the distance or actual size of the objects.

Scaling the Drawing. Discussion of central

recession and format dimensions, with step by step
instructions for scaling the circle of view, object
distance, image size and image location using the
anchor point and measure bar.

Display Geometry & Image Impact. The

geometrical relationships in a perspective image
strongly affect its visual impact and esthetic
qualities.

Anamorphic Images. Images can be distorted,

using projection methods similar to linear
perspective, so that they appear flat or veridical
when viewed obliquely or in a curved mirror.

Constructing a 1PP Cube. The step by step

method for constructing a three dimensional
rectangular solid using 1PP or central perspective
projection methods, with emphasis on the artistic
decisions that are part of the process.

Early Renaissance Methods. The earliest uses of

central perspective were based on practical drawing
methods, not abstract geometry.

Page 3. Two Point Perspective (2PP)

Two Point Perspective. The basic geometry of 2PP
projections, in which there are two vanishing points
for the sides of a rectangular solid seen at an angle.

Rotating the Vanishing Points. The correct

method for locating 2PP vanishing points in relation
to the 90° circle of view.

Locating the Measure Points. The isoceles triangle

that defines the geometry of measure points;
locating measure points from established vanishing
points.

Constructing a 2PP Cube. The step by step

method for constructing a three dimensional
rectangular solid using 2PP projection methods.

Inclined Lines & Inclined Planes. Methods for

finding the vanishing point of an inclined line and the
vanishing line of an inclined plane.

Distance Point Projection. A tradition method for

constructing 2PP drawings relies on the diagonal
vanishing points to project a square and its diagonal
into perspective space.

The Ground Line Framework. The alternative

perspective setup commonly used by draftsmen and
architects.

Who Has a 12 Foot Table? It's not uncommon for

the vanishing points in 2PP to be 12 feet or more
apart. There are three different remedies for this
problem.

VP Spacing from an Object Drawing. There are

inflexible geometrical relationships between the size
of the object, its distance from the viewer, and the
spacing of the vanishing points in 2PP drawing.
These limit your freedom to guess at vanishing point
relationships.

Where Is the Center of Projection? The circle of

view approach allows you to find the center of
projection implied by any 2PP image that contains
defining diagonal elements.

Page 4. Three Point Perspective (3PP)

Three Point Perspective. The basic geometry of

3PP projections, in which there are three vanishing
points for all sides of a rectangular solid that is both
rotated and tilted to the image plane.

The Perspective Sketch Construction Method.

The step by step method for constructing a 3PP
projection of a cube by starting with a sketch of the
desired perspective proportions.

Constructing a 3PP Cube (Sketch Method). The

step by step method for constructing a three
dimensional cube using 3PP projection methods, with
two methods to define the measure points.
The Horizon Line Construction Method. The
geometrical logic for rotating the 3PP vanishing
points, which permits them to be located and the
drawing scaled exactly.

Constructing a 3PP Drawing (Horizon Line

Method). Step by step explanation of how a 3PP
drawing is made, from rotating the vanishing points
to scaling the image size.

Page 5. Advanced Perspective Techniques

Perspective of Complex Plane Figures. Projecting

complex two dimensional figures into perspective
space, including an ellipse, a pentagon, a city map,
and three ways to project a circle.

Perspective of Complex Solid Forms. Complex

solid forms can be projected into perspective space
by using various strategies; examples include
Uccello's chalice, a sphere, the Leaning Tower of
Pisa, two types of cones, a rectangular solid, a series
of arches, a staircase, an octahedron, a
dodecahedron, and the human figure.

Buildings From Blueprints or Plans. Buildings and

large objects can be projected into perspective space
by deriving measure bars and vanishing points from
plans, elevations and profile drawings.

Paraline Perspectives. Perspective systems based

on nonconverging (parallel) lines, which can
represent all three dimensions and all internal right
angles at equal scale and without recession.

Curvilinear Perspectives. Recent perspective

systems based on curved vanishing lines, designed
to represent the subjective curvatures of natural
vision.

Page 6. Shadows, Reflections & Atmosphere

Basic Rules of Perspective Shadows. The basic

geometry and terminology for perspective shadow
constructions.

Shadows from Solar Light. Different construction

problems arise depending on whether the shadow
creating edge and the shadow receiving surface are
vertical, horizontal or inclined.

Shadows from Local Lights. Nearby light sources

complicate shadow perspective because they are not
point sources and produce enlarged or diffuse
shadow shapes.

Perspective of Reflections. Reflections from

mirrors or still water surfaces present simple
construction problems; ripples in water produce more
complex periodic patterns; convex or spherical
surfaces create optical distortions.

Aerial Perspective. Landscape distance is

suggested by the many subtle changes in detail,
color and tone that occur when air filters light across
large distances, or smoke or fog filters light across a
romantic rendezvous.

Rainbows. Rainbows combine color, atmosphere,

landscape and light in panoramic proportions of
beauty that vary with weather, time of day,
geographic location and season. They signal the end
of our perspective studies.

Last revised 07.I.2015 • © 2015 Bruce MacEvoy

perspective in the world technique

Linear perspective originates in the common
appearance of the real world, yet it seems to follow
the abstract constraints of geometry. It can visualize
the infinite reach of three dimensional space by
organizing everything around a single, precisely
located viewpoint. These foundation topics are the texture of space
presented in this page.
four perspective facts
If you already have some perspective training, then
my approach will be unfamiliar. Most perspective creating the perspective view
tutorials are focused on the object you want to draw.
My emphasis is on the viewer: linear perspective the perspective setup
is the two dimensional image of a unique
viewpoint and direction of view. I introduce linear basic rules of perspective
perspective as embedded in our natural view of the
physical world and as connected to basic facts of image plane, viewpoint &
vision, then present a geometrical summary of the direction of view
perspective method, the assumptions behind its
presentation in pictorial art, and the ways its perspective distortions
limitations can be used in effective artistic design.

I postpone the "how to" drawing tutorial because

awareness of the foundation themes can cure a
student's hackneyed or mechanical application of
perspective construction. To get a feel for what
perspective is really about, one must realize that it is
visible everywhere and in everything — even when
architectural edges and corners are nowhere to be
seen.

the texture of space

Vision creates an image of the physical world from
the weave of light around us. How does it do this?
One way to address that question is to answer a
more specific one: how do we "see" that an object is
near or far from us?

Anything that helps us see the relative distance of

objects in space is called a distance cue.
Fundamentally, all distance cues are made possible
by the geometrical regularity of three
dimensional space, and it is this regularity that
linear perspective attempts to simulate.

First, vision takes advantage of the fixed

characteristics of our two eyes to make sense of
what we see. The most powerful distance
cue, binocular parallax, is the disparity between
the images created by the two eyes that arises
because they are located about 5cm to 7cm apart.
This causes near objects to shift back and forth
against a distant background as we close first one
eye and then the other. The mind uses this parallax
to infer the distance of objects in the field of view:
the larger the left to right shift, the closer the object.
We also use motion parallax, which occurs when
we move our head, stoop or turn, walk or run
through the environment. Parallax is a very powerful
and accurate distance cue, and it is effective across
an enormous range of distances — binocular parallax
from the tip of our nose out to about 20 meters, and
motion parallax (depending on the speed of
movement) out to several kilometers.

Parallax cues depend so heavily on the fixed

attributes of space and the location of our eyes that
it takes infants only about four months to learn how
to use parallax to guide reaching and grasping. Other
cues related to eye position, such as lens focusing
(accommodation) and crossing the eyes to see
close objects (convergence), are comparatively
weak — they are only useful within a few feet.

However, in the two dimensional, fixed surface of a

painting, all the cues from parallax, convergence and
accommodation disappear. So the artist must rely on
other distance cues to create the illusion of three
dimensional space.

Some cues appear in the optical properties of

monocular (single eye) retinal images. In three
dimensional space, objects close to us appear larger
than those far away, so retinal image size is an
important distance cue, especially for objects we
recognize. Objects at our feet or just overhead
appear much lower or higher in relation to the
horizon than objects far away, so the vertical
position of objects in our visual field — compared to
each other or to the horizon — also serves as a
distance cue in natural environments.

All these distance cues seem related to our view of

detached objects. However, equally powerful depth
cues arise in the visual appearance of surfaces,
especially the textures and colors of the natural
world.
 distance cues in changing textures

The American photographer Ansel Adams had a

superb eye for perspective facts in visual design. His
photograph of an arid landscape contains not one
straight edge anywhere, and confronts the world
head on, making the landscape appear flat. Yet the
sense of depth in space is powerful and pervasive.

In the foreground, within our physical range of

motion, we usually distinguish separate objects, in
part by using the occlusion of one object outline by
another. The simple rule is, whatever covers is
closer, and this rule applies across any distance
(even when the sun sets behind a far mountain).
 distance cues in overlapping forms

This collection of circles illustrates that a complete

break in the outline of one form by another indicates
the unbroken form is closer (in front), opaque and
probably solid. If the covered outline is partly visible
(like the mountains through the shafts of light), we
infer the closer form is partly transparent. If two
objects meet in an outline that is irregular to both
(large circles at right), then the distance relationship
between them is ambiguous.

However, the main distance cue in the Adams photo

is the change in visual textures across space, called
a perspective gradient. The foreground rocks
appear large and extremely rough; with distance
they grow smoother, the spacing between them
becomes smaller, and the rocky surface appears
flatter, less irregular. Beyond the rocks, the
mountains and clouds have irregular outlines but
appear smoother than the rocky plain. And beyond
everything is the sky — the only perfectly textureless
"surface" in nature.

If the object or surface is far enough away, it is

"behind" a considerable distance of atmosphere,
which can obscure the object with suspended
particles of dust, smoke or molecules of water vapor.
The cumulative effect of these obscuring particles
creates aerial perspective in large objects visible
from a great distance, especially mountains,
buildings and desert or ocean horizons. Depending
on the time of day and strength of light, aerial
perspective can make distant objects appear less
distinct, less saturated and darker or lighter in value.
Smoke or dust shifts the hue of distant objects
warmer (toward red, yellow or yellowish white),
while water vapor shifts landscape hues toward blue.

We have to use the recognizable continuity of an

object's outline, or its "completeness of form," to see
occlusion, which is more difficult if objects are far
away or very small, dimly illuminated, or unfamiliar
to us. Look again at the Adams picture, and you'll
see that one rock clearly covers another at the
bottom of the image, but in the middle distance
these overlaps become harder to see. Instead,
everything merges into the average spacing
or spatial frequency of the rocks — that is, the
rocks do not separate themselves from the texture
as distinct forms. Wherever objects become too small
or complex to show occlusion clearly, texture takes
over.

This transition from form to texture means that

visual experience is a combination of objects
filled in by visual textures. Increasing distance in
space transforms the appearance of objects into
structurally or visually related textures. And at
extreme distances, texture itself dissolves into pure
color. So we have the following sequence that applies
to large vs. small or near vs. far visual elements:

pattern —> texture —> color

perspective transforms pattern into

texture and color

In this illustration, the band at the top of the image

is made of the same green and red squares as the
band at the bottom, but the squares are too small to
see individually: instead they mix visually to make
yellow or gray. There is a fusion threshholdfor
every texture, beyond which it is blended by the eye
(in visual fusion) into a single homogenous color.
Color TV screens, a distant mountain slope and a
sandy beach are all composed of tiny discrete forms
beyond the visual mixing threshhold.

Occlusion works because we can compare the

outlines we see with our idea of the objects we
look at: anything partly covered is a "broken" or
"altered" form of itself. So our knowledge and
expectations of the world are essential to create
effective distance cues. However, the boundary
between what we "see" with our eyes and what we
"know" with our memory and mind is not at all clear.
In fact, we can create the illusion of a recognizable
form entirely through the visual completioninduced
by forms around it.

Finally, these transitions from occluding objects to

patterns to textures to colors as distance increases
do not happen in the same way for all objects —
unlike the effect of aerial perspective or fog, which
causes all forms to fade equally from
view. Increasing distance creates characteristic
visual transitions in different objects, especially
in natural forms where there is a distinctive structure
at different scales of view. Trees are the classical
example, much studied by 19th century artists,
because different species of trees express a different
branching pattern that is recognizable from twigs up
to large branches; the tree's branching pattern, in
turn, determines the tree's overall form and the
clumped appearance of the trees in copses or
forests.

the unique sequence of patterns created

by perspective changes in oak trees

Many kinds of vegetation, rock formations, clouds

and water flows show similar interrelated patterns
across large changes in viewing distance. The point is
that the painting brushstrokes, color mixtures and
shading that artists use to represent the objects
must change with the object's distance: a distant
tree is not a miniature image of a tree nearby,
as crude perspective thinking might suggest. It has a
completely different visual character. The artist's
challenge is to find the right representation for the
object's appearance at the appropriate distance, not
just to paint larger or smaller versions of the same
thing. This can be done by understanding the
fundamental structure of the object, and how this
structure changes in apparent form, texture and
color across perspective space.

Linear perspective is space drawn as the geometrical

idea of itself. But we do not see the idea of space:
we see a world of light, colors, textures, objects and
opportunities for action. As we explore the artistic
uses of perspective, we will repeatedly grapple with
the fact that our visual experience of the world is
much richer and more complex than our idea of the
geometrical space in which it appears.

four perspective facts

Linear perspective simplifies the world in order to
create a coherent visual representation of the world.
It includes somefacts that determine our view of the
world (three dimensional space, light, surfaces) but
excludes others (movement, atmosphere, texture). It
includes some features of visual experience
(recession in space, convergence of parallel lines)
but not others (color, optical fusion, binocular
parallax). All these restrictions arise from the four
key facts on which perspective methods are based
(diagram, right).
1. Light travels in a straight line between any
two points in space. This is the foundation
of linear perspective: the behavior of light can be
described through traditional Euclidean geometry.

When light encounters the naturally dull and rough

surfaces of the physical world, it is reflected
or scattered in all directions. This means light is
always abundantly radiating in all directions on all
sides. But then the question arises: how can this
dense tangle of light create a perceptible image? the four perspective facts

(1) light travels in a straight line or "light ray"; (2) an image

2. An image is formed by light passing through is formed by light rays passing through a single viewpoint;
a single point. This is the viewpoint. The viewpoint (3) the viewpoint defines a visual cone centered on a
exactly matches the properties of a pinhole direction of view; (4) all visual rays appear "end view" as
points on an image plane
camera, which creates images by passing light
through a tiny hole in a screen. Because this pinhole
forms images, it can geometrically represent the
images formed by a lens, such as the artist's eye or
camera.

All light rays that intersect the viewpoint (pass

through the pinhole camera), and equivalently all
lines of sight emanating from the viewpoint, are
called visual rays. The only visual rays that matter
to our view of the world are those that converge on
the viewpoint: all other light rays are excluded. The
dense tangle of light becomes an image.

The eye is really a small sphere, and we normally see

with two eyes, so we have to simplify the facts of
sight somewhat, depending on what we mean
by looking at the world. If we mean a camera or
single immobile eye, the viewpoint is the nodal
point of the optics, which in the eye is located
slightly behind the center of the lens (because light
has already been refracted by the cornea). If we use
one eye but look in different directions, the viewpoint
shifts to the rotational center of the eye. If we
use both eyes, the binocular viewpoint is
approximately located between the two eyes.
3. Visual rays through the viewpoint define a
visual cone centered on a direction of view. We
can't see light through the back of our head, and
light does not enter a camera through both sides of
the pinhole screen. In nearly all optical systems,
images are created by light arriving from the "front
half" of the surrounding space (diagram, right).

The visual rays from the "front half" of space form a

cone, known as a visual cone or visual
pyramid, with the viewpoint at its point or apex. This
cone has a central axis, known as the optical
axis, which defines the center of a camera image or
our visual field. In linear perspective this optical axis
is called the direction of view (or sometimes
the central ray or principal visual ray).

The human visual field actually has a very complex

structure — crisp central vision and fuzzy peripheral
vision — but linear perspective assumes that any the visual cone and visual rays as points on an image
visual ray inside the visual cone contributes equally plane

to an image. This is a specific example of how linear

perspective does not represent what we
actually seewith our eyes, but rather what
we know about optics and the geometry of the
physical world.

4. Every image is a cross section through a

visual cone. An image is not formed at the
viewpoint, because a point has no dimension.
Instead the image is formed by making a
slice through the visual cone at some point other
than the viewpoint, either in front of or behind it.

This slice cuts across all the visual rays, so that we

only see visual rays "end on" within the visual cone.
As a result, all visual rays appear as points on an
image plane. The image is really a surface of
compacted points, each point signifying a visual ray
that has reached the viewpoint from a specific
location in physical space.

This description of light rays as straight lines,

arriving from objects in space to a viewpoint with a
specific direction of view, allows us to use
a geometrical method for describing the visible
world on a two dimensional surface from a single
point in space. This is called a central
projection. Geometry in turn gives us the procedures
necessary to construct these central projections
using the simplest tools: a pencil, a straight edge
and a compass.

creating the perspective view

Now let's apply the four perspective facts to create a
standard perspective setup, which will be the
mechanism that the the artist can use to construct
representational drawings. My explanation proceeds
in small steps so that you can see how the
mechanism actually works and understand the
assumptions that it is based on.
The Light Environment. We start with the light
environment as a viewer would experience it
naturally. The space around the viewer is filled with a
dense, rich scattering of light, coming at him from all
directions and distances, reflected from every surface
and even scattered by the atmosphere.

the light environment

The viewer is also alive and continually moving —

shifting his gaze, turning his head, leaning to one
side or another, stepping forward or backward,
walking or sitting or lying down. Before motion
picture cameras, there was no way to capture this
dynamic complexity.

The Stationary Viewer. The first step in

perspective is to exclude all the dynamic aspects of
visual experience and limit the problem to
a stationary viewer. The viewer takes in visual rays
only from a fixed location in space, in a fixed posture
(including both body position and orientation of the
head), and facing in a fixed direction (with a fixed
position of the eye or eyes). To my knowledge this is
not explicitly characterized in perspective texts, but a
stationary viewer is the fundamental premise of a
perspective drawing.

the stationary viewer

Once we freeze the viewer's location, posture and

gaze, we necessarily fix the viewer's visual
cone (what we would call the visual field in other
contexts). The light comprised by this fixed visual
cone represents a single place, a single view of the
world, experienced uniquely by a single viewer: no
one else can experience exactly the same stationary
view at the same time.
The fixed visual cone is defined by a fixed apex,
the viewpoint, and a fixed direction of view (also
called the central ray, axis of sight or principal visual
ray), which represents foveal ("in focus") vision at
the center of the visual field.

The third and final dimension is the width of the

visual cone. Any two visual rays within the visual
cone define a visual anglemeasured at the
viewpoint, which corresponds to the visual distance
between two points in the visual field. So what is the
visual angle of the visual cone? This was determined
by medieval optics to be 90° (one quarter of a full
circle), and later perspective practice adopted this
90° limit as a convenient standard (for
reasons explained below). This creates a circular
diameter to the visual cone, centered on the
direction of view, know as the 90° circle of view.

We know from our own visual experience that we see

clearly only in foveal vision, at the center of view: we
can't read unless we look directly at the words.
However this central clarity is not acknowledged in
linear perspective. Because of the fixed position and
viewpoint, motion parallax and binocular parallax are
excluded as well. We also cannot decide whether the
image represents a glance or a steady gaze, the view
of a moment or of eternity. These omissions give the
images created by linear perspective their surreal
clarity and static perfection. Clarity and perfection
are really cognitive, not perceptual, attributes:
perspective commits us to draw what we know, not
what we actually see.

The Ground Plane. The perspective act — the fixed

visual cone, viewpoint and direction of view — looks
out on abstract space. We have a point of view, but
nothing to look at. So the next step is to establish
a physical space that creates the visual rays
converging on the viewpoint. The simplest and most
elegant way to do this is simply to provide the viewer
with someplace to stand: the ground plane.

the ground plane

The ground plane is essentially the representation of

spatial extent: it goes off into the distance. By
convention, the ground plane is made as abstract as
possible: flat and perfectly level. In terms of visual
experience, it represents the spatially largest or
dominant level surface below the viewpoint. In this
location it symbolizes all architectural surfaces and
the great flat layers of geology — tilled fields, alluvial
meadows, dried lake beds, and large bodies of water.
By convention, the viewer is normally standing or
sitting, spine upright and head erect to balance the
downward pull of gravity, with eyes facing forward.
This puts the viewpoint at a fixed distance above the
ground plane: the viewing height.

As it extends outward in all directions, the ground

plane cuts the visual cone almost in half, blocking
the range of vision downward. This naturally orients
the direction of view straight ahead, which is fixed by
a second convention: the direction of view is
parallel to the ground plane. The viewer stands or
sits upright and perpendicular to the ground plane,
head upright, balanced against the downward pull of
gravity.

Although it is abstract, the ground plane is

extraordinarily rich with significance. It is the here
and now of the perspective act, and signifies that this
place is important to experience. It also characterizes
the perspective stance of the viewer — his
location, posture and focus of attention within a
specific physical setting.

Distance Measurement. The ground plane is our

reference for location in space, and therefore
the distance from the viewer to any objects in space.
To specify these concepts of location and distance,
the next perspective step is to define a metric grid
on the ground plane.

dividing the ground plane with a metric

grid

The most convenient approach is to partition the

ground plane by a grid of squares 1 meter on a side.
We can if desired create a second grid in a plane
perpendicular to the ground plane, so that we can
measure distance in three dimensions.

By convention, all lines in the grid are defined

either parallel or perpendicular to each other
and to the direction of view. This allows us to
measure distances in any direction in relation to the
viewpoint — 10 squares ahead, 2 squares to the left
— in the same way we would locate points on a
sheet of graph paper. The vertical grid allows us to
measure distances in height above (or below) the
ground plane or the direction of view. This metric
space allows us to extend or verify the facts of linear
perspective by means of geometrical proof.

This grid on the ground plane is one of the most

primitive conventional elements of linear perspective.
Early Renaissance artists actually included the
measurement grid in their finished paintings and
frescos, as a pavement of square tiles, often in a
strongly contrasted checkerboard pattern.

The Physical Geometry. The visual cone is filled

with an infinite number of visual rays, arriving to the
viewer from every visible object and surface in
physical space. However, thanks to the metric grid,
we can define the spatial location from which visual
rays originate. For example, we can limit our
attention to visual rays from intersections in the
metric grid, and ignore the rest. We assume
(correctly) that any insights we obtain from these
few visual rays will apply to any other visual rays in
the visual cone.
 visual rays in physical space

In the figure, five of these points are shown in

orange, and labeled d, c, b, a and x along one side
of the direction of view; a matching row of unlabeled
orange points is shown along the opposite side. Each
row of points lies on a single straight line, and the
two lines are parallel to the direction of view. At the
same time, the matching pairs of points define the
sideways or transverse lines in the metric
grid, perpendicular to the direction of view.

The visual rays from these points define the

geometry of visual rays in physical space. And they
allow us to address two fundamental questions
about recession, or changes in object appearance
with object distance:

• What happens visually at different distances to

objects arranged along a straight line parallel to the
direction of view (as defined by the line dx and the
matching line on the opposite side)?

• What happens visually at different distances to

objects arranged in equally spaced
rows perpendicular to the direction of
view (represented by the transverse lines ending at
each labeled point)?

The next perspective steps clarify the answers to

these questions.

The Perspective Geometry. By limiting the

perspective view to a handful of visual rays from the
intersections of the metric grid, we have started to
simplify or abstract the viewing situation, reducing it
to its geometric essentials. Let's complete that
process.

the basic perspective geometry

First, we excuse the human viewer and retain only
the fixed location of his viewpoint.

The viewpoint has a specific location in relation to

the ground plane directly underneath it. This is called
the station point. A line between the station point
and viewpoint is perpendicular or "square" to the
ground plane, signified by the small square at the
base of the line.

(Traditional perspective tutorials refer to the

viewpoint as the station point, but I feel it is very
useful to have a separate term for the viewpoint and
its ground plane location.)

The distance between the viewpoint and station point

is the viewing height above the ground plane. As
we've seen, this depends on the viewer's physical
height and location in relation to the ground plane
(sitting, standing on the ground, or standing at the
top of a tower).

The viewpoint is at the tip or apex of the visual cone,

and the origin of the direction of view. We have
already conventionally decided that the direction of
view is parallel to the ground plane. So we can
define a median line on the ground plane,
extending from the station point and parallel to the
direction of view, which divides the ground plane into
symmetrical left and right halves.
Finally, we can specify a object distance between
the viewpoint (or station point) and any object within
the visual cone.

At this point linear perspective becomes a precise

measurement system. All the distance
measurements within the metric grid, and the visual
angles of visual rays from the grid to the fixed
viewpoint and direction of view are defined by basic
trigonometry. In fact, as linear perspective
developed during the Renaissance, it was closely
associated with developments in surveying,
mapmaking, navigation and astronomical
observation. The tools and procedures for measuring
the physical world and for making perspective
images were often explained in the same book.

Making the viewing situation geometrically abstract

imparts a similar abstraction to the identity of the
viewer of the image. Paintings that create an identity
or presence for the viewer as an individual
recognized by persons in the painting, as in
Velázquez's Las Meninas, are rare in the perspective
tradition, especially in academic or history
paintings. More often the perspective viewpoint
implies a timeless or universal witness, an abstract
vantage that can be filled equally well by any
anonymous passerby.

The Image Plane. Next we turn to making a

perspective image. To do this, we insert an image
plane through the visual cone. This corresponds to
the fourth perspective fact described above: an
image is a cross section through the visual cone. It is
the "window" of perspective imagery.

an image plane in the basic perspective

geometry

To keep the geometry simple, and to mimic the

vertical viewing position of a vertically hung painting
or wall fresco, the image plane is conventionally
a flat surface perpendicular to the direction of
view and to the ground plane. (Linear perspective
works just as well if there are no right angles in the
setup, or if the image plane is curved rather than
flat, but these situations are geometrically more
complex and were not clearly analyzed until the early
18th century.)

The image plane does not have fixed dimensions —

its limits are only determined by the size of the
visual cone or by the size of the support that we
make the image on.

However, the image plane does have a fixed

location: the ground line directly underneath it. This
is equivalent to the base of a vertical wall on which
the painting or fresco is displayed.

Finally, the ground line is at a fixed distance from the

station point: this is the viewing distance.

The Perspective Image. The image plane is

commonly described as a window looking onto the
world. This means all visual rays pass through the
image plane on their way to the viewpoint.

perspective image on the image plane

The final step is to identify where each visual ray

passes through the window — the point where it
intersects with the image plane. This point is
its perspective image. In the figure, point a' is the
intersection between the image plane and the visual
ray from point a on the ground plane to the
viewpoint; that is, a' is the image in perspective
space of the point a in physical space. Point b' is the
image of real world point b, c' is the image of c...
and so on for all the points on those two parallel
lines of points we decided to study.

Physical lines — edges, tracks, borders, wires — can

also be projected onto the image plane. The
collection of all projected image points and image
lines is the perspective image of the corresponding
points and lines in physical space.

We have defined a way to map three dimensional

space onto a two dimensional surface, by locating
the image points for every detail of the visual cone.
For an optical image, these points are plotted for us
by rays of light. In artistic practice, perspective
constructions are typically made by plotting visual
rays point by point. To simplfiy this task, the
emphasis is on significant points, especially vanishing
points and corners or edges that can be connected
by straight lines or freehand curved lines.
Perspective drawing does not proceed by
mechanically connecting dots with lines, but by
choosing the dots that locate all essential elements in
the perspective image.

The image plane is conventionally divided by two

representations of the viewer's perspective stance.
The horizon line corresponds to the visual limit of
the ground plane if it extended infinitely far. This
divides the image plane horizontally. The median
line corresponds to the median line on the ground; it
extends vertically upward from the ground line and is
perpendicular to the horizon line. The two lines
intersect at the principal point, which locates the
direction of view as it passes through the image
plane.

Because we have defined all the relevant elements as

parallel or perpendicular to one another, the principal
point anchors two basic dimensions of the
perspective image. The distance between the
principal point and the ground line
is viewing height to the image plane; the distance
between the principal point and the circumference
formed where the 90° circle of view intersects the
image plane is the viewing distance.

Two important details: the horizon line is not

necessarily defined by the visible horizon on the
surface of the earth; we conventionally assume this.
As we'll see later, it can alternately be defined as
the pupil line (visual horizontal) of the artist's head.
(There would be a horizon line in outer space.)
Similarly, the image plane does not have to be
perpendicular to the ground plane, or even to the
direction of view, but defining it that way makes it
easier to work out practical perspective problems.
The Viewing Geometry. Once the artist has used
these principles to transfer the three dimensional
world onto a two dimensional canvas, and from the
significant points and lines developed a completed
perspective painting, a second situation arises that
must be governed by perspective principles:
the cultural encounter between the perspective
image and a human viewer.

If we whittle this encounter down to its essentials,

we are left with the vertical (wall hanging)
orientation of a faceless museum or
gallery painting, and the ghostly center of
projection, perpendicular to a line from the center
of the painting, which is the viewpoint implied by the
perspective facts in the image.

the visitor and the artwork

If the museum visitor stands so that his

viewpoint from one eye (or through a peephole)
exactly coincides with the center of projection of a
perfectly consistent perspective painting, all the
visual rays from the surface of the painting will
recreate the visual rays from the original scene,
within the limits of accuracy of the painter's
representation.

This is the illusionistic use of perspective, and it

is only effective when (1) the drawing is in strict
perspective, (2) the drawing contains the kinds of
receding lines and planes that make the strict
perspective construction visible, and (3) the drawing
is viewed through a peephole or eyepiece at the
center of projection. (Most photographs, excluding
most wide angle photographs and including all
telescopic photographs, are also in linear
perspective.)
It should be said that most paintings from most
historical periods contain perspective inconsistencies,
such that they define several similar centers of
projection; and indeed most images (the Adams
landscape photograph above, or a photograph of the
Museo Guggenheim in Bilbao; photo, right) do not
clearly define a center of projection because they
lack the edges, corners and distance cues that most images do not define a distinct center of
identify clear vanishing points or vanishing lines. In projection
these cases, viewers by default assume a viewing
position that centers the image in their field of vision,
perpendicular to their direction of view, at a distance
that brings the whole image into a comfortable circle
of view. This intuitive "center of projection" is
simply eye level, facing, and comfortably far
away.

Imagination does the rest. Because the viewers of

paintings and frescos rarely choose (or are able) to
stand at exactly the center of projection,
the symbolic or informal use of perspective — to
convey the idea of being in a certain place at a
certain time — plays a much greater role in the
stance viewers typically take toward a painting or
photograph. Paradoxically, the linear perspective in
the image must be reasonably consistent and
accurate to be acceptable, but the linear perspective
of the painting in the viewer's eyes does not. We see
the painting surface as an object, not a window.

Even when a painting is viewed from exactly the

center of projection, a perfect perspective drawing
may create apparent perspective distortions that
become intrusive or objectionable when the painting
is viewed with both eyes or from different points of
view. As this is how people normally look at
paintings, artists spent three centuries attempting to
understand and minimize these effects. Eventually,
in the process, they learned to use the distortions for
expressive purposes.

the perspective setup

We have progressed by logical steps from the four
perspective facts to a basic geometrical framework
for mapping objects in space onto a two dimensional
image plane. Now it is time to step into the viewpoint
and examine this framework as the viewer sees it.

To do this, I will create the perspective image of my

running example, the point intersections in a metric
grid. The most primitive and explicit way to do this,
which was the standard method in early
Renaissance paintings, was to define the
perspective geometry in paired horizontal and
vertical diagrams of the entire viewing situation —
the viewpoint, image plane and every object to be
drawn on the image plane.

These diagrams are known as

the elevation and plan, as shown in the figures
below and at right.
 the perspective framework in elevation
(above) and plan (right)

The plan (view from above) is based on an image

plane parallel to the ground plane, with all points in
physical space projected onto it by parallel vertical
lines perpendicular to its surface. The elevation (view
from the side) is always perpendicular to the plan
and ground plane (like a wall), again with points
behind it projected onto its surface by parallel
horizontal lines. (There is no convergence to a
viewpoint in a plan or elevation.) Conventionally the
elevation is parallel to the sides of the architectural
form it portrays, but for our purposes it is parallel to
the direction of view.

If we make a plan and elevation at actual size, and

very accurately specify the locations of the viewpoint
and the image plane, then we can draw visual rays
from objects to the viewpoint, measure where they
intersect the image plane in these views, then
transfer these horizontal and vertical measurements
to the painting format.
The figures (above) show this done twice: horizontal
green lines for measurements done on the elevation
(which gives the distance of the points above the
ground line), and vertical green lines for
measurements from the plan (which gives the
distance of the points to the left or right of the
median line).

If we make and measure these schematic drawings

carefully, then connect the dots to construct
the perspective image of our metric grid on the
image plane, we discover that the receding rows of
points appear as converging lines of image points,
as shown below.
 perspective image of the metric grid on
the ground plane

Terms introduced in the discussions of perspective

geometry, the image plane and the perspective
image are shown in plain italics; if any are
unfamiliar or unclear, please review those sections
carefully.

Now we see that the image plane roughly fills the

visual field; it slices through the visual cone to create
the 90° circle of view(or any other size circle of
view we want to define), centered on the principal
point — the intersection of the direction of view with
the image plane. The principal point and horizon line
also show the viewing height.

Now let's examine the perspective image of the

metric grid. First of all, we find that it still consists of
straight lines (in red). Connecting pairs of metric
points parallel to the direction of viewhas created the
image orthogonals (the mathematical term for
"perpendicular," which reminds us that the
orthogonals are perpendicular to the image plane).

Now we immediately see that image lines parallel

to the direction of view converge at the
principal point, which is therefore their vanishing
point (abbreviated vp) — the term coined by the
English mathematician Brook Taylor in 1715.
Because this vanishing point is identical to the
principal point (the direction of view), it controls
recession in space toward the focus of attention.
(Perspective drawings based only on the principal
point are in central perspective, as discussed on
the next page.)

Despite what we see, we know the lines in

the metric grid on the ground plane are constructed
parallel to the direction of view and are equally
spaced (we can confirm this in the plan view, above).
So we can conclude that the orthogonals define an
interval of constant width in perspective space.

Connecting pairs of metric points parallel to the

image planecreates the image transversals, which
are parallel to the ground line. Again we
immediately see that transversals become more
closely spaced as they approach the horizon line. Yet
because we know they represent equally spaced lines
in physical space, we can conclude that
the transversals define intervals of equal depth
in perspective space, from the ground line toward
the horizon.

We can hardly appreciate today the extraordinary

sense of discovery that early Renaissance
artists experienced as the first perspective drawings
took shape under their hands, and the paradoxical
relationship between see and know came into view.
We sense their delight and awe in their manuscript
attempts to solve more and more intricate
perspective problems, and in the reverent accuracy
with which they transformed these drawings into
finished works of art.

A basic principle was recognized early: the spacing

between transversals narrows more quickly with
distance than the spacing between orthogonals (the
vertically elongated squares of the metric grid at the
ground line become horizontally elongated rectangles
in perspective distance). Artists had unlocked the
fundamental proportions of foreshortening, which
is the compression of the visual angle of a
dimension or distance as the dimension
becomes more parallel to the direction of
view. Indeed, the earliest illustrations of artists
studying perspective problems usually show them
studying the effects of foreshortening — for example,
in Dürer's illustrations that show how to draw, point
by point, a foreshortened lute or a human figure.

Several decades later, artists also realized that that

the two diagonals within the squares created by the
orthogonals and transversals must also be parallel
lines (like the parallel diagonals of a chessboard) and
therefore must also converge to vanishing points on
the horizon line on either side the principal point.
These are the diagonal vanishing
points (abbreviated dvp), first described by the
French cleric and diplomat Jean Pélerin in 1505.

Pélerin described how contemporary artists used

the dvp's to find equal intervals of depth (the
transversals) from orthogonals of equal width
measured along the ground line. For this reason
the dvp's were traditionally called distance points,
because in central perspective they are used to
transform a measure of physical distance along the
ground line into an image recession in perspective
space. (They are also called distance points because
the distance on the image plane between a dvpand
the principal point is exactly equal to the viewing
distancefrom the viewpoint to the image plane. This
means the diagonals can be used to reconstruct
the center of projection implicit in a perspective
painting.)

The Circle of View Framework. The final step is to

standardize or abstract the insights we have drawn
from the perspective image of the metric grid, and
formulate them as a perspective machine. This is
the circle of view framework.

The key element is that the viewing distance (x, the

distance of the viewpoint from the image plane), the
viewing height (the distance of the viewpoint from
the ground plane or plane of orthogonals) and the
radius of the circle of view are all equal. We also
require, as a simplification of the perspective
problems we want to analyze, that the direction of
view is parallel to the ground plane and the image
plane is perpendicular to both the ground plane and
the direction of view. This creates the physical
arrangement illustrated and labeled in the diagram
(below).
 the circle of view framework: basic terms

the 90° visual cone with viewing distance set equal to

viewing height

We choose the 90° circle of view as the framework

for perspective operations because this circle has a
radius of 45° visual angle around the principal point,
so it contains all possible diagonal vanishing points.
In addition, 90° is the visual angle accepted since
the Renaissance as the outer limit of images
projected onto a plane, so we have no use for a
larger visual span.

To create the 90° circle of view, we simply define

the viewing distance as equal to the viewing
height, which aligns the ground line with the base of
the circle of view. Then the framework proportions
integrate the diagonal vanishing points, the
viewpoint, the viewing distance to the image plane,
the viewing height and the ground line around
the powerful central recession toward the
principal point that is created by the direction of
view.

the circle of view framework

the 90° circle of view as it appears from the viewpoint

The central vanishing point (vp) defines recession

along all lines parallel to the direction of view — the
convergence of all orthogonals. The horizon
line and median line intersect at the principal point,
dividing the circle of view into quadrants. Two pairs
of diagonal vanishing points lie on the horizon and
median lines on opposite sides of the circle of view.
And because the viewing distance is equal to the
viewing height, the ground line, median line and
circle of view all intersect at a single point, the
bottom dvp.

If we need to be precise in how the perspective view

is implemented, then the specific measurements
depend on the stature or vantage of the viewer.
However, as a general rule, an average size adult
has a viewing height of about 1.6 meters (63
inches), so the circle of view at the image plane will
be about 3.2 meters (10.5 feet) wide. (Note that the
viewing height is always measured from a
viewer's eye level, not the top of her head.)

The 90° circle of view is a very convenient

framework for working out perspective problems, but
drawings that completely fill the circle are subject
to perspective distortions that most artists find
objectionable. For that reason, the actual image area
typically is fitted into a much smaller circle of view,
such as the 60° or 40° circles shown in the diagram.
For example, a watercolor full sheet (22"x30")
would appear as shown in the diagram — nicely
contained within a 30° circle of view. Even the
massive emperor sheet (40"x60") only fills a 50°
circle of view at a 3.2 meter viewing distance.

Because the 90° circle of view framework explicitly

links together the principal point, viewing distance,
viewing height, ground line and all diagonal
vanishing points, it can be applied to solve any
perspective problem. It does not just provide a
system for copying nature point by point in order
to make a painting. We've actually invented
a system of perspective construction which can
be used to create new images at our pleasure and
imagined worlds from any viewpoint.

basic rules of perspective

At this point you should have a clear understanding
of how linear perspective connects the three
dimensional physical world to a two dimensional
perspective image. So this is the appropriate point to
review some of the basic and always trustworthy
perspective rules that can guide you in making a
perspective drawing. The rules can be pounded out
by geometrical deduction, but I will simply state
them in a logical order.

A Perspective Glossary. First, a summary of the

key terms. (1) Physical space refers to the three
dimensional, real world; (2) the ground plane is an
idealized flat, level surface representing the
pedestrian surface of architectural forms (lawn,
pavement, floor), or the average of flat natural
terrain (desert, salt flat, surface of a lake or ocean);
(3) the viewpoint is the unique location in physical
space of the nodal point of the observing eye or
camera, the convergence point or center of
projection for light; (4) the station point is the
point on the ground plane directly underneath the
viewpoint; (5) the direction of view is the optical
axis of a camera or the line of sight of a viewer
located at the viewpoint, typically aligned so that it is
parallel to the ground plane; (6) the image plane is
a two dimensional, flat surface, aligned so that it is
perpendicular both to the ground plane and to the
direction of view, on which the perspective image is
projected; (7) a visual ray is any line that intersects
(passes through) both the viewpoint and the image
plane; (8) the visual cone is a cone, with apex at
the viewpoint, axis along the direction of view, and a
base diameter on the image plane just large enough
to comprise all the visual rays contributing to an
image.

(9) An image point is the intersection of a visual ray

with the image plane; an image line is a line drawn
on the image plane between two image points, or the
line formed by the intersection with the image plane
of a plane in physical space; (10) the principal
point is the intersection (image) of the direction of
view with the image plane; (11) the ground line is
the intersection of the ground plane with the image
plane; (12) the median line is a line on the ground
plane directly underneath the direction of view, and
also the image of this line as a line perpendicular to
the ground line and through the principal point; (13)
the horizon line is an image line through the
principal point, parallel to the ground line and
coincident with the horizon in physical space of a
"flat" surface such as the ocean.

For an image plane perpendicular to the ground

plane and to the direction of view, (14) the viewing
distance is the distance between the viewpoint and
image plane and/or between the station point and
ground line; and (15) the viewing height is the
distance between the viewpoint and the station point
(in physical space) and/or between the ground line
and the principal point (on the image plane); (16)
the circle of view is the intersection of the visual
cone with the image plane, measured as a visual
angle from the viewpoint or as a radius from the
principal point on the image plane.

(17) A perspective image is the projection of

physical space onto the image plane by visual rays
converging at a viewpoint; (18) a plan is the
projection of physical space onto a horizontal image
plane (e.g., the ground plane) by parallel vertical
lines; and (19) an elevation is the projection of
physical space onto a vertical image plane by parallel
horizontal lines.

Again, as simplifying assumptions, (a) the image

plane is perpendicular to the direction of view; (b)
the image plane is perpendicular to the ground
plane; (c) the direction of view is parallel to the
ground plane; (d) the viewer is standing or sitting
upright on the ground plane; and (e) the viewing
distance and viewing height are equal. These
assumptions define the 90° circle of view framework
and make the perspective rules easier to understand
and apply.

The Basic Rules of Perspective

1. The image of a visual ray is a point on the

image plane. A visual ray is any line that intersects
the viewpoint and passes through the image plane.
The intersection of a line and a plane defines a point.
This corresponds to the fact that when we look
straight down any line or edge in physical space, its
image is only a point in our visual field. Thus, the
direction of view only appears as the principal point,
the origin of any visual rayappears as a point, and
any number of separate points on a visual ray all
appear as a single point on the image plane.
 figure 1

In figure 1, the visual ray (the line from the

viewpoint V) intersects the image plane at a single
point; in this case, because the visual ray is the
direction of view, this point is the principal point
(pp). Points a and b are located on the same visual
ray, therefore their point images are identical
with pp.

Figure 1 also shows that any feature of physical

space can be projected downward as a plan in
the ground plane. The point g is the image of the
principal point projected into the ground plan (as
shown by the dotted line); the station point (S) is
the image of the viewpoint in the ground plan, and
the median line is the image of the direction of view.
 figure 2

2. Any straight line in physical space that is not

contained in a visual ray projects a straight line
on the image plane. That is, a straight line or edge
in physical space always appears as a straight line in
the perspective image, no matter which way the line
is turned to the direction of view. (The sole exception
is when the physical line is contained in a visual ray,
when according to rule 1 it appears as a point.)

In figure 2, the line AB in physical space does not

intersect the viewpoint V, and therefore it is not a
visual ray. The visual rays AV and BV do intersect
the viewpoint, and therefore they also intersect the
image plane at X and Y. All the points
between Aand B can be projected in the same way,
and these create the image line XY (green line) on
the image plane.

The image of line AB in the plan is the line ab. Note

that when the points a and b are connected to the
station point S by lines in the plan, they intersect the
ground line at x and y, the plan image of the
points X and Y. Note than the plan image is
constructed by parallel lines perpendicular to the
ground plane (as shown by the dotted lines).

3. Any two points on a straight line, projected

onto the image plane, define that line on the
image plane. Thus, a straight line drawn between
the two points X and Y creates the image
line XY in figure 2.

Note that if the line has infinite length, then any two
distant points will serve; but if the line has a fixed
length (a line segment), then the two end
points are necessary to define its length. This leads
to the most economical method of perspective
construction: we project only the end points of a line
onto the image plane, then connect them by a
straight line. For example, we can define the edges
of a cube by projecting only its significant
points or defining elements — the six corner points
— onto the image plane, and then connecting the
appropriate corner points with straight lines to
construct the edges.
 figure 3

4. The image of an extended line must end in

two points: its intersection with the image
plane and its vanishing point. If we have drawn a
cube in perspective, what would happen if we
extended an edge of the cube to make an infinitely
long line in physical space? Would that make the
image line infinitely long as well? The answer is no:
the image line must end in two points: its
intersection with the image plane and its vanishing
point.

The only exceptions to this rule are lines parallel to

the image plane (they never intersect the image
plane, and they do not converge to a vanishing
point), and visual rays, for which the intersection and
vanishing point are the same (see rule 1).
In figure 3, the infinitely long line AB in physical
space intersects the image plane at B and recedes
toward the virtual point A, which is not a physical
point (and therefore is shown in blue) because
vanishing points are only points on the image plane,
not points in physical space. The vanishing point is
also the intersection of visual ray AV with the image
plane. The vanishing point projects into the plan
as x, and x lies on the line AS, the plan image of the
visual ray AV.

This rule, which the English perspective theorist

Brook Taylor called "the principal foundation of all
the practice of perspective," has important
consequences that we will explore in the next page.

5. The vanishing point of a line is the

intersection of the parallel visual ray with the
image plane. If our direction of view is exactly
parallel to any line, then we are looking directly at
the vanishing point for that line; and given a fixed
viewpoint, there is only one vanishing point for any
physical line and therefore only one visual ray
parallel to that line. (These fundamental principles of
recession were first proved geometrically by the
Italian mathematician and astronomer Guidobaldo
del Monte in 1600.)

In the metric grid perspective example used above,

the elevation and plan show that the direction of
view is parallel to the gridline of points abc, so those
two lines never actually meet in the real world. Even
so, the visual angle between the direction of view
and any point on the gridline becomes smaller as the
point moves farther away from the viewer — the
visual angle between point d and the principal
point p is much smaller than the visual angle
between p and a. When the points are very distant
from the viewpoint, the visual angle between the
points and p becomes imperceptibly small and the
points merge with the principal point, as we see in
the converging orthogonals of the metric grid.

In figure 3 (above), the visual ray AV passes

through the vanishing point for image line AB, as
does its image AS in the ground plan;
therefore AB, AV and AS are parallel.

figure 4

6. All parallel lines in physical space converge

to the same (single) vanishing point. If any two
lines are parallel to a third line, then they are parallel
to each other, which generalizes rule 5 to any
number of lines. Note again that vanishing points
only exist on the image plane, they have no location
in physical space.

An important corollary: any visual ray defines the

vanishing point for all physical lines parallel to
that ray. This allows us to work backwards, from the
perspective image to physical space. Thus, in figure
4, if we pick any arbitrary point C on the image
plane, and draw the image line Cvp (green line),
then this is the image of the line AC in physical
space, and we can deduce that
lines AB, AC, AV, AS, Ab and Ac are all parallel.
7. Lines parallel to the direction of view appear
to converge at the principal point. This is only a
specific case of rule 6, but it is very useful. We
concluded in the previous section that orthogonals
define a constant width across the receding
transversals in an image. The principal point, and its
associated orthogonal lines, define the primary
dimension of depth or recession in any perspective
image.
parallel railroad tracks converge toward the horizon
This is a fact of everyday vision as well. Straight
railroad tracks on level ground (right) are the most
striking example. (Here a camera lens, rather than
the eye, creates the perspective viewpoint.) Sunlight
provides another case — the sun is so far away that
its light "rays" are essentially parallel at the earth's
surface, and therefore seem to converge when
broken into shafts.

parallel sunbeams converge toward the sun

figure 5

8. Lines through the image plane that intersect

the extended line VS create image lines
perpendicular to the horizon line. The exceptions
are visual rays, which pass through the
viewpoint V and therefore appear as image points on
the image plane (perspective rule 1).

The key is that line VS, the viewing height from the
ground plane to the viewpoint V, is perpendicular to
the ground plane at the station point S. Extended
without limit, this line is equivalent to the head
midline of a standing viewer. If a line intersects VS,
its plan line will intersect S, and the two lines will lie
in the same plane as VS, which is perpendicular to
the ground plane. Therefore the line image of the
plane (perspective rule 10) will also be
perpendicular to the ground plane, the vanishing line
of the ground plane (horizon line) and its intersection
with the image plane (ground line).

Three examples in figure 5 demonstrate that the

direction of the line or its intersection point
with VS do not affect the vertical orientation of the
image line. Line AB intersects the line VS at B, and
creates the plan line ab which intersects the station
point S; BS is perpendicular to the ground plane; the
image line is vertical from the intersection ip1 to the
vanishing point vp. Line CS in the ground plane is its
own plan line; it forms a triangle with the visual
ray CV and its image line is perpendicular
from xto vp. Finally line DE intersects the ground
plane at E and forms the plan line ES; its image line
extends from ip2 to the image point e'.
In each case, the plane figures
(ABSa, VCS and DES) contain some part of the
line VS, which is perpendicular to the ground plane,
so the plane figures and their intersections with the
image plane are also perpendicular to the ground
plane.

Rule 8 explains why reflections from bodies of water

always form a vertical smear directly under the light
source (image, right): the reflections all lie along a reflections appear vertical in all directions
line from the base of the light source to the station
point, and therefore form a vertical line on the image
plane.

As a corollary of this rule: for an extended line VX

through V and perpendicular to a plane at X,
lines that intersect VX will form image lines
perpendicular to the vanishing line of the plane.
This means reflections on a wetted wall will be
horizontal, perpendicular to the vanishing line for the
wall, and reflections on a sloping wet street will
appear at an angle from vertical perpendicular to the
sideways slope of the pavement in the direction it is
viewed.

The rules developed for lines can also be applied to

planes. By knowing the location and orientation of a
plane, we also partially define the location and
orientation of any lines it contains. In a perspective
construction, points are used to define line edges,
and edges define the planes that contain their lines.

9. A plane that contains a visual ray intersects

the image plane as a line. This matches rule 1 for
lines. If a plane contains a visual ray then its surface
disappears, like a playing card viewed edge on, and
all we see is its straight line intersection with the
image plane.

A useful corollary: any straight line through a point

in perspective space is the intersection with the
image plane of the plane that contains the visual ray
passing through that point.

10. The perspective image of any two lines,

that either are parallel or intersect in physical
space, defines the image of the plane
containing those lines. This is the matching
principle to rule 3 for lines.

In figure 4 (above), the two image

lines Bvp and Avp are parallel, because vp is a
vanishing point; therefore the perspective image of
lines ABC defines the plane ABC. The same would be
true if vp were a point where the lines intersected in
physical space.

figure 6
11. The image of an extended plane must end
in two lines: its intersection with the image
plane and its vanishing line. This is the matching
principle to rule 4 for lines, and similarly the only
exceptions are planes parallel to the image plane and
planes that contain a visual ray — for these the
intersection line and vanishing line are the same.
In figure 6, a plane (magenta area) intersects the
image plane at ABC (green line). All lines in this
plane that are not parallel to the image plane recede
to its vanishing line XYz. I have drawn this plane so
that it is tilted to intersect the ground plane also; this
intersection is the line CK in physical space. The
vanishing point for CK is Y, the point where the
image vanishing line intersects the image horizon
line; and the line YC is the perspective image of the
line CK. Note as before that y lies on the plan
line KS.

As a important corollary, the intersection line and

vanishing line of a plane are always parallel on
the image plane. Thus, in the figure above, the
lines ABC and XYz are parallel. In figure 4 (above),
the parallel lines AB and AC define the image of the
plane ABC as the image lines Bvp and Cvp; the
intersection of this plane with the image plane is the
straight line passing through B and C (rule 3); and
the vanishing line for the plane is the line that passes
through vp parallel to BC (rule 9). Similarly, the
ground plane defines the ground line (its intersection
with the image plane) and the horizon line (its
vanishing line), and these two lines are always
parallel to each other in a perspective image.

12. The vanishing line for any plane is the

parallel plane containing a visual ray, or the
line connecting the vanishing points for any
two lines parallel to the plane. A plane that
contains a visual ray intersects the viewpoint V,
which means the plane is seen "edge on" as a line on
the image plane (rule 9). This matches rule 5 for
lines.

13. All parallel planes converge to the same

(single) vanishing line. This matches rule 6 for
lines. In the standard perspective setup, the horizon
line is the vanishing line for the ground plane and all
planes parallel to it, such as floors, ceilings, water
surfaces and cloud layers.

14. The vanishing line of a plane contains the

vanishing points for all lines in the plane and all
lines parallel to the plane. This is an extremely
powerful rule, because it makes the vanishing line of
an important plane the "attractor" for all lines
parallel to it. Thus, the horizon line, which is the
vanishing line for the ground plane, contains the
vanishing points for all lines constructed level to the
ground — that is, the horizontal edges found in
nearly all buildings and their diagonals — even when
the building walls are not parallel to the image plane.
15. The vanishing line for any plane parallel to
the direction of view intersects the principal
point. This matches rule 7 for lines.

All planes parallel to the ground plane any distance

above or below it must converge to the vanishing
line for the ground plane, which is the horizon line
(rule 13). In the vertical dimension, all vertical
planes parallel to the direction of view on either side
of it must converge to the vanishing line for the
median plane, which is the median line. Finally, any
plane tilted at an angle to the ground plane but
parallel to the direction of view will create a similarly
tilted vanishing line, which again will pass through
the principal point on the image plane.

16. Any plane that contains both a line and the

plan image of the line is perpendicular to the
ground plane, and defines a perpendicular
intersection line and vanishing line on the
image plane. This matches rule 8 for lines. Rule 16
is useful for the construction of inclined lines, and for
defining the light plane of shadows. Reciprocally, if
the vanishing line of a plane is not perpendicular to
the horizon line, then none of the lines contained in
that plane will be perpendicular to the ground plane.

Similarly, any plane that contains both a line and the

elevation image of the line is horizontal; therefore it
is parallel to the ground plane and its vanishing line
is the horizon line (rule 13).
 •

17. Finally, although they are not rules per se, it is

important to memorize the criteria for the four
different types of perspective drawings (discussed
in later pages):

• in one point perspective (or central perspective)

there is only one vanishing point, which is identical
to the principal point located on the horizon
line and the median line. Central perspective or 1PP
requires all six faces of all square solids to be either
parallel or perpendicular to the image plane and
direction of view.

• in two point perspective (2PP) there

are two vanishing points, neither of which is the
principal point, that define a single vanishing line,
usually (but not necessarily) the horizon line. 2PP
requires that two faces of all square solids must be
perpendicular (not parallel) to the image plane and
parallel (not perpendicular) to the direction of view.

• in three point perspective (3PP) there

are three vanishing points, none of them the
principal point, that define three vanishing lines,
none or any one of which may be coincident with the
median line, the horizon line or any other line on the
image plane. 3PP requires that no face of any square
solid is perpendicular or parallel to the image plane
or to the direction of view.
The most common type of drawing requries mixed
perspective, in which some objects appear in one
type of perspective and some objects in another. In
this case each object or group of similarly arranged
objects must be treated as a separate perspective
problem; they are combined as a single image
because they share a common circle of view.

image plane, viewpoint & direction of

view
Now it's appropriate to come back to the specific
viewpoint and direction of view that are the core of
any perspective image, and consider how these
relate to the image plane and to the features of the
scene or landscape.

Image Plane Orientation. First, let's revisit the

point mentioned earlier that the image plane is not
necessarily perpendicular to the ground plane (for
example, in a 3PP image), but is always considered
to be a flat surface, perpendicular to and centered on
the direction of view.

In terms of projective geometry, we can just as

easily and accurately record the optical facts of the
world on an image plane that is not perpendicular to
the direction of view (or to anything else). And we
can use a curved surface just as effectively as a flat
one, as was commonly done with the ceiling frescos
created for the domes and barrel vaults of European
Baroque churches and palaces and, more recently, is
used as the image plane in curvilinear
perspective.

In other words, the flatness and perpendicular

orientation of the image plane are
essentially conventional. The convention arises from
the way we typically (conventionally) make and show
art. We assume the image plane is perpendicular to
the ground plane because we expect the finished
image will be hung for viewing on a vertical gallery
or museum wall. We assume the image plane is flat
because stretched canvas and drawing paper are flat.
We might think of these as display
conventionscontained in the perspective geometry.

We display images the way we conventially do

because that makes them easy to view for people
who adopt a convenient posture and position: that is,
standing in front of the image surface. We might call
these the viewing conventionscontained in the
idea of the image plane, because "the right way to
hang the painting" depends on our assumptions
about "the right way to look at the painting." We can
specify these in terms of the the orientation of the
viewer's head in relation to the image plane, as
shown below.
 viewing conventions toward the image
plane

The human sense of visual orientation ("up" and

"down") depends on the head, not the body. The
head orientation is defined in three dimensions:
a pupil line drawn through the pupils of both eyes,
a direction of view perpendicular to the pupil line,
and a head midline perpendicular to both the pupil
line and the direction of view and usually parallel to
the erect spine. (This is the posture for binocular
vision. If the image plane represents a "peep show"
view from one eye, then the direction of view is the
optical axis of that eye.)
By convention the standard rectangular format of the
painting or photograph are aligned so that (1) the
direction of view is roughly through the center of the
format and perpendicular to its surface; (2) the pupil
line is parallel to the top and bottom edges of the
format or the horizon line within the image; and (3)
the head midline is perpendicular to the floor and
parallel to the image plane. All these conditions are
met if the viewer is standing squarely in front of the
painting with head erect, and the painting is hung at
eye height and level to the floor — display and
viewing conventions that are summarized as eye
level, facing, and comfortably far away. Note that
despite these ideal viewing conventions, paintings
are routinely displayed at heights or in locations that
make that impractical.

Finally, there is a third kind of structure folded into

the image plane, which is the projection
assumption that defines the artist's view of things.
The convention here is simply that the "artist's view"
(or camera view) at the time the image was created
explains the appearance of the world in the image.
The projection assumption governs the
interchangeable use of "we" or "the artist" in art
critical narratives ("in this painting, we are looking
down into Niagra Falls" or "in this painting, the artist
is/was looking down into Niagra Falls"). We expect,
for example, that if the horizon line is parallel to the
top and bottom of the image plane, then the artist's
pupil line was parallel to the horizon, even though
the artist may have been leaning or
crouching while working. We experience the book
reproductions of Michaelangelo's Sistine Chapel
paintings as vertical and flat, even when they are
located on curved walls or over the viewer's head —
and in execution required the painter to lean
backward or lie on his back.
The crux is that the display convention, viewing
convention and projection assumption fuse the
artist's view, the painting image and viewer's stance
within a common, conceptual visual framework.
The "right view" of a visual image and our
interpretation of it is anchored in spatial orientation:
we cannot recognize faces, or correctly judge the
relationships among objects, when they are "turned
the wrong way" (images, right). The "right
orientation" is embedded in our head axes, and these
must align with the image contents and its format
images are uninterpretable in the "wrong" orientation
borders to produce an acceptable image display.

These conventions are so powerful, and so basic to

visual experience, that we enforce them even for
paintings by Jackson Pollock or Bridget Riley, where
they mean nothing to the visual texture of the work;
or in the "conceptual" wall drawings of Sol Lewitt,
where echoes of the display or viewing conventions
belie the claim that the drawing instructions only
respond to the limitations of the drawing site.

Paintings gain visual drama or impact when there is

an obvious difference between the projection
assumption and viewing convention — for example,
when the artist's direction of view was downward or
upward in relation to the ground plane.
These elevation differences are acceptable because
they still imply a shared upright stance ("balance") in
both the artist and viewer. In contrast, we are
usually intolerant of image tilt in which the artist's
pupil line or horizontal camera frame are not parallel
to the ground plane (as if the artist's head was tilted
toward one shoulder, or the camera was askew when
the picture was taken).

Object Orientation to the Direction of View.

Dramatic changes in the image occur by changing
the angle between the direction of view (or camera
sightline) and the surfaces of a primary form. That is,
image perspective changes with the direction of
view, even when the viewpoint stays the same.

Two photographs (below) show a Roman arch in two

separate views from exactly the same
viewpoint, made with a pinhole camera — a camera
that focuses light through a tiny hole instead of a
lens. This exactly reproduces on film the perspective
optics from a single center of projection.
 effect of changing only the direction of
view

the viewpoint is fixed and the direction of view remains

parallel to the ground plane; from M.H. Pirenne, Optics,
Photography and Painting (1970)

The only difference between the two photographs is

in the direction of view, and therefore in the
orientation of the image plane in relation to the
frontal planes of the arch — the pinhole was kept in
exactly the same location. The image at left shows a
direction of view perpendicular to the face of the
arch; the horizontals appear parallel to each other
and to the horizon line. When the direction of view is
shifted 25° to the left, the horizontals now appear to
converge, and only those at the horizon are parallel
to the horizon. That is, simply by changing the
direction of view, we've transformed a central
perspective view into a two point
perspective view.

If the camera were instead rotated up or down, so

that the direction of view was no longer parallel to
the ground plane, the image would morph into a two
point perspective image with the vanishing points on
the median line; if it were rotated both horizontally
and vertically, the image would shift into the even
more complicated three point perspective view.
Linear perspective is not just about a viewpoint or
about a direction of view: it is defined by a specific
viewpoint and a specific direction of view.

The crux is that the design of a perspective image

does not consist simply in the choice of viewpoint
onto a primary form such as a building, but the
direction of view (location of the principal point) as
well. The guidelines for adjusting or choosing the
viewpoint and direction of view are somewhat
subjective, and depend heavily on the intended
impact of the image, but some suggestions are
provided in the section on drawing from blueprints
or plans.

Horizon Line and Viewpoint. An important and

useful fact of perspective is that all objects at the
same height as the viewpoint are intersected
by the true horizon line. This rule holds regardless
of how far above level ground the viewpoint may be,
and even when the direction of view is not parallel to
the ground plane.
 horizon line and viewpoint in landscape
perspective

from J.T. Thibault, "Application of Linear Perspective in

the Graphic Arts" (c.1860)

The French artist J.T. Thibault created a compact

illustration (above). The top, middle and bottom
views correspond to the sitting, standing or elevated
standing viewpoint of the blue figure at left, who
represents the viewing height of the artist in each
image. (Blue man's standing height is indicated by
the brown line fixed in front of the stairs.)

All the perspective relationships between other

figures or objects in the image and the true horizon
line (the orange line, not the apparent horizon line
defined by the hills) change with the viewing
height. When the viewer is sitting, the horizon line
passes through his head and therefore appears to
cross the waist of standing figures around him. When
he is standing on level ground, the horizon line
passes through his head and through the heads of all
standing figures as tall as he is — no matter how
near or far they are from the viewer. When the
viewpoint is from a raised platform, all figures on the
ground below appear below the horizon line. (Note
also the changing location of the horizon line against
the roadside pillar.)

This fact arises from rule 12: all parallel planes

converge to the same vanishing line. In this case,
the first plane is the ground plane, whose vanishing
line is the horizon. The viewing height, extended in
all directions, creates a second plane parallel to the
ground plane, like the surface of a large lake up to
the height of the viewpoint. This surface will also
converge to the horizon line. Regardless of the
direction of view, all objects lower than this plane will
be "under water" and therefore below the horizon
line. All objects above it will be "above water" and
above the horizon line.
In the photo of train tracks above, the horizon line
intersects the bottom edge of the red passenger car,
just above the wheels. This is somewhat lower than
the standing height of a man, so we can infer that
the photographer was crouching or sitting (or the
camera was on a low tripod) when the picture was
taken.
Many visual illusions of size depend on the position of
the object relative to the visible or presumed horizon
line, even when other perspective cues are removed.
The famous and delightful Ames room (right),
contrived by Aldebert Ames Jr. in the 1940's, is a
large trapezoidal enclosure that appears perfectly
square and level when viewed through a peephole
near one corner. Figures appear to grow or shrink in
opposite corners of the room because the "short"
corner on the left is substantially lower and farther
an ames room
away than the "tall" corner on the right, reducing
both the apparent size of the figure and her relative
position to the "horizon line" defined by the windows
and floor.

perspective distortions
The standard demonstration of linear perspective
— drawing on a sheet of glass the view from a
fixed location as seen through one eye — shows that
the geometry of linear perspective really works: what
you see is what you get!
Viewing Distortions. However, a perfect
perspective drawing or optically flat photograph
reproduces three dimensional space on the viewer's
retina only when we view it with a single eye, located
at the center of projection and looking along the
correct direction of view implied by the perspective
geometry.

And there's the catch. Even if the perspective

drawing accurately represents a specific
viewpoint, we typically don't look at the
perspective drawing from the "correct" center
of projection. The drawing may be done at a scale
that conveniently fits the space available within the
picture format, but creates a center of projection that
is too close to or too far from the picture surface; the
painting or fresco may be positioned too far above
the floor; or the painting may be viewed from
different distances or angles as it is hung in a room
or gallery; and, of course, we always look at it with
two eyes.

What happens if we look at a perspective drawing

from a different location? The following diagram
illustrates the crux of the problem.
 perspective geometry and viewing
distortion

We start by viewing from a distance of 5 feet (60") a

very large (40" x 60") painting of a rectangular office
building, conveniently drawn so that its vanishing
lines are at 45° to our direction of view. This places
the diagonal vanishing points of the drawing exactly
at the diagonal vanishing points of our 90° circle of
view, and the drawing perfectly recreates the illusion
of three dimensional space.

But this is a large painting, so we decide to step back

a few feet (to 90") and look at it again. Now
the drawing vanishing points no longer
correspond to our visual vanishing pointsas
defined by our 90° circle of view. As a result, the
edges and angles of the building seem to place the
vanishing points too close together, and the building
appears exaggerated in perspective proportions —
the front angle of the building seems more like 70°
than 90°.

Of course, linear perspective can produce compelling

illusions, but not easily — the image must be in exact
perspective, the edges of the image must be hidden,
and the image must be viewed with a single eye from
the center of projection, in what is called a "peep
show" or peephole arrangement. Binocular
photography and a special binocular apparatus that
presents each image to a separate eye can create
very vivid depth illusions, but even slight changes in
the point of view will destroy the effect.

Foreshortening Distortions. There is a second

problem caused by oblique (sideways, upwards or
downwards) angles of projection onto the image
plane. This is related to the perspective fact
of foreshortening, but a distinction between two
kinds of foreshortening is necessary to understand
what is going on.
 foreshortening and the triangular
proportions

(top) rotation foreshortening causes the object

surface XY to become oblique to the image
plane; (bottom) shift foreshorteningcauses the
object surface AB to remain parallel to the image
plane; both examples are an equal distance from the
direction of view and appear identically foreshortened
(by 25°) at the viewpoint

In shift foreshortening, a two dimensional surface

is shifted away from the direction of view (the
principal point) but remains parallel to the image
plane; the actual surface always appears
foreshortened because it is at an oblique angle to the
viewpoint.

In rotation foreshortening, the surface is rotated

so that it is no longer parallel to the image plane; the
actual surface may or may not appear foreshortened,
depending on whether it is at an oblique or
perpendicular angle to the viewpoint.

These different types of foreshortening have different

perspective effects.
 perspective image of flat forms

shift foreshortening has no effect on the perspective

image of a two dimensional surface parallel to the
image plane

The figure above shows the correct perspective

projection of an identical row of windows (center). In
the top row, the windows are kept parallel to the
image plane but become increasingly oblique to the
direction of view (shift foreshortening); in the bottom
row, the windows are rotated in place to remain
perpendicular to the viewpoint, which puts them at
an oblique angle to the image plane (rotation
foreshortening).

Surprisingly, even though it produces a

foreshortened view of the actual two dimensional
object, shift foreshortening has no effect on a
perspective image. A window shifted 45° to one
side is exactly the same size on the image plane as a
window centered on the direction of view. This occurs
because, at the location of the perspective image of
the window, the image plane is also
foreshortened by the same oblique angle of view,
and this "secondary" foreshortening matches the
foreshortening seen in the surface.

In contrast, rotation foreshortening always

alters the perspective image. The image becomes
"distorted" in the direction perpendicular to the axis
of rotation, regardless of whether the object is
central or peripheral in the circle of view and even
when the rotation eliminates any foreshortening in
the actual object! Remember: rotation foreshortening
is still a completely correct perspective view of the
rotated object, when viewed from the center of
projection; it just looks wrong when we view the
image from farther away.

The distorting effects of rotation are caused by the

recession that creates vanishing points. As explained
in the discussion of the orthogonals, an equal
physical displacement of the object from the
direction of view produces a smaller and smaller
perspective displacement from principal point as the
object is farther from the viewpoint. Rotation pushes
one half of the surface farther away from the image
plane, the other half closer to the image plane, which
makes the recession shift unequal on the two sides.

The objectionable perspective distortions occur in

the oblique view of a three dimensional object that
has only been shift foreshortened on the image
plane. In these cases, what "rotates" is not the plane
surface of a two dimensional object but our view of a
plane cross section through its three dimensional
form.
 perspective image of rounded forms

in a 90° circle of view; from M.H. Pirenne, Optics,

Painting and Photography (1970)

The diagram (above) shows a perfectly correct

perspective image of a regular row of cylindrical
columns with flat top surfaces supporting regular
spheres. If you could use one eye to examine this
figure from the true center of projection (directly in
front of the central sphere, at a distance equal to the
radius of the circle of view, roughly 5cm or 2" from
your computer screen), you would discover that all
the forms really are in perfect perspective.
But because we view the drawing from much farther
away (and with both eyes), the spheres and columns
appear grossly distorted. The columns give the
illusion of being viewed head on, when in fact those
near the circle of view are seen from one side, so
that the front and back of the forms define their
cross section. These are not the same distance from
the image plane, so they display unequal recession
toward the principal point, which elongates the form.

These distortions have distinctive features worth

memorizing:

• Radial thickening. The spheres and columns

displaced from the direction of view appear thicker
than those at the center of view; this thickening is
along a line from the object to the principal point.

• Displacement exaggeration. The amount of

thickening or distortion depends on the displacement
of the object from the principal point (the visual
angle between the object and the direction of view);
the distortion becomes more extreme toward the 90°
circle of view.

• Diagonal exaggeration. The distortions appear

most extreme in the diagonal directions, because
these combine the effects of the height and width
displacements.

• Radial tilting. Horizontal surfaces, such as the

orange flat tops of the columns, appear tilted along
the radial line of thickening rather than downward or
upward in relation to the viewer.

• Peripheral crowding. Equal intervals between

three dimensional objects (such as the spaces
between columns) close together as displacement
increases; eventually the spaces between the
columns disappear and the columns seem to overlap.

Cures for Perspective Distortions. If we keep in

mind that these rotation "distortions" are in fact
accurate perspective images when viewed from the
center of projection, then it is clear that the reason
they appear as distortions is because the image is
viewed from somewhere else. Managing the
distortions is therefore a concession to the
uncertain viewing geometrythat governs image
display.

The traditional diagnosis for perspective distortions is

that the width of the drawing is too large in
relation to the 90° circle of view. This is
equivalently expressed as "the vanishing points are
too close together", or "the distance points are too
close to the principal point", or "the viewing distance
is too close to the image plane." In effect,
the viewing distortions are more obtrusive when a
painting encompasses a large circle of view.

If the image vanishing points were much farther

apart (that is, if the image were enclosed by a
smaller circle of view), then the drawing would
represent objects as they appear from a
viewpoint much farther away, and changes in the
the viewing geometry would cause smaller
proportional changes in the image circle of view.

In effect, the viewing distance to the image is a

smaller proportion of the apparent distance to the
objects in the image, so the drawing can be
acceptably viewed from a wider range of viewing
distances. In addition, the rotation distortions and
crowding of serial forms that become exaggerated
toward the 90° circle of view are cropped out of the
image entirely.

The practical limit for an acceptable visual cone has

historically been a 60° circle of view — a
suggestion first made by Piero della Francesca in
c.1470 and repeated often since then. In fact,
depending on the geometry of the principal form and
the location of the vanishing points, a 40° circle of
view or less is much more typical.

Leonardo da Vinci devoted many pages in his

notebooks (c.1490) to the analysis of perspective
distortions, and he especially disliked the
exaggerated apparent size of the perspective grid as
it reached the ground line of the image plane (for
example, as in the ground squares of this image).
He recommended painting an object as it appears
from a distance of 3 to 10 times its actual
dimensions (e.g., a standing figure 1.75 meters tall
should be viewed from 5 to 18 meters). This is
equivalent to placing the figure within
a 19° to 6° circle of view. In fact, modern vision
research has found that most people say an object
"fills their field of view" once it occupies
approximately a 20° circle of view; the classical
French rule has been to contain the image within a
30° circle of view. I use a 25° circle of view as a
rule of thumb when designing or analyzing an image,
which corresponds to a viewing distanceto a
finished painting of about 2.5 times its height, width
or diagonal. (These issues are explored further in the
section on display geometry & image impact.)

So the restricted circle of view "cure" for perspective

distortions was well known to artists from the
beginning of perspective practice (even if the
necessary "dosage" was ambiguous). But these
artists also realized that some distortions are
more intrusive than others to a casual viewer.
Apparent distortions in rectangular forms are more
objectionable than distortions in curved forms;
distortions in the horizontal direction are more
obtrusive than distortions in the vertical direction (in
part because the format is usually wider than it is
high); distortions in unfamiliar objects are more
acceptable than distortions in familiar objects;
distortions in the apparent location of vanishing
points are more acceptable than distortions in the
outline of forms; distortions in a mixed perspective
drawing are more objectionable than those in a
rigorous perspective drawing; and so on.
As a result, if artists were working with a large fresco
or canvas format, or wanted a panoramic effect, they
adopted a radical practice guided by the context of
the painting: they would simply "correct" or
disguise perspective distortions wherever they
appeared objectionable. This was almost always
done for figures, rounded forms, the spacing
between columns of a facade, and so on. Often
several kinds of "corrections" were used at the same
time.

raphael's school of athens (1511) from an

elevated viewpoint

A fine example is Raphael's large fresco The School

of Athenswhich fills an almost 30 foot wide section of
Vatican wall. This huge format clearly imposes a
panoramic context on the image design, which
Raphael utilized in novel ways. He framed the
perspective construction within a relatively restricted
40° circle of view, which crops extreme distortions
from the image — although as a result the correct
perspective viewing point is not even in the room.
The perspective distortions are disguised through
numerous clever omissions from the picture space.
The vanishing point of the enormous central
passageway is hidden by the two approaching
figures. Most of the picture space is filled by walls
parallel to the picture plane. The pair of square
columns on each side are cropped at the top and
hidden at the bottom by standing figures, eliminating
the repeated sideways intervals or diagonal corners
that would accent perspective distortions. The
semicircular front arch of the barrel vault is cropped
at the top, because it would otherwise appear to be
elongated vertically. The floor tiles on either side of
the foreground are hidden by groups of figures. The
foreground stairs help to separate the figures
raphael's school of athens from a human viewpoint
vertically and interrupt the perspective continuity of
the tile floor.

Most important, all figures are drawn as if

centered on the direction of view — that is, with
no perspective distortion. This is easiest to see in the
two astronomers shown holding celestial globes (at
right). Both figures are located at the righthand edge
of the fresco, beyond the 30° circle of view. Rather
than draw the spheres with the correct but
elliptical perspective projections, Raphael simply
drew them perfectly round.
Thus, the architecture enclosing the figures is
cropped and oriented as a carefully edited and correction of perspective "distortions" in
arranged perspective speace, while each of the Raphael's School of Athens (1511)

figures is drawn in its own, "head on" perspective

space. Yet this hodgepodge of perspectives appears
coherent and harmonious.

The last piece of the puzzle is that the fresco is

normally viewed from a vantage too close to the
image plane and several feet below the center of
projection, which causes a distinct upward
convergence in the image verticals (image, below).
Yet in context the convergence lends a soaring
grandeur to the image, and by means of this esthetic
impact the overall perspective space appears
harmonious and convincing.
Expressive Uses of Distortion. As painters
developed dozens of similar tricks to exclude, hide or
counteract perspective distortions, thus minimizing
the effect of viewing a painting from "incorrect"
locations, they discovered that perspective
distortions could be used for expressive
effect or to counteract unfavorable display
conditions.

The earliest examples are image manipulations

necessary to produce the desired visual effect in
fresco images viewed from various locations on the
floor of a large building. Michelangelo's famous Last
Judgment demonstrates a dual compensation: the
"celestial" figures high on the wall are almost 50%
larger than the "damned" figures at its base (photos,
right). The sacred figures carry clearly even to the
back of the chapel; but viewed from the altar, the relative scale of figures high and low in Michelangelo's
higher figures are 50% farther from the viewer than "Last Judgment"

those at the base of the wall, so that the visual

differences combine as a balanced overall
composition.

The most extreme examples are anamorphic

images — especially popular in the 16th and 17th
centuries — which appear as unrecognizable smears
or blurs unless viewed from an extreme angle or with
a corrective mirror. These strange paintings suggest
how far artists were willing to play with the
geometrical implications of perspective in search of
new artistic resources.

In general, rendering a single three dimensional form

within a circle of view greater than 40° (that is, as
the form would appear to the naked eye from a close
distance) has four important effects on its visual
impact:

• principal forms become more dynamic —

buildings or figures seem to loom, surge and expand

• perspective space is enhanced — the

convergence among vanishing lines is more
emphatic, creating a vertiginous depth of space

• the front surfaces of the form dominate — the

sides of the form may disappear from view, or
appear smaller or highly foreshortened, and the side
surface textures are viewed at a more grazing angle

• vertical dimensions dominate — in particular,

the extreme corners of the form may appear to jut or
loom out of proportion with the rest of the figure.

Renaissance and Baroque artists who experimented

with these effects understood that perspective
paintings are effective even when they are not
viewed from the center of projection. This is
sometimes called Zeeman's paradox, but the
paradox is purely conceptual: it assumes we view a
perspective representation as a retinal simulation,
when in fact we view it as a two dimensional
painting. In other words, perspective
constructions create visual symbols, not visual
illusions. The key is that paintings lack the depth
of field cues created by binocular vision; we are
always aware a painting is flat rather than deep. And
that is how our mind interprets it, adjusting our
understanding of the painting to compensate for our
position.

Some famous problems are simply cases of incorrect

analysis. For example, artists from Leonardo down
to Flocon & Barrehave been vexed by the paradox
that long parallel edges (such as the top and bottom
of a straight wall or the sides of a cylindrical tower)
appear to taper away from the viewer; yet they are
drawn (in central perspective) as parallel straight
lines. This is because the same triangular
proportions that foreshorten the parallel edges of
the wall or column also foreshorten the parallel lines
in the image. Perspective changes the apparent
dimensions of the wall and the apparent dimensions
of the drawing of the wall: no "curvilinear correction"
is necessary.

As the illusionistic use of perspective was never a

serious goal in painting, artists were free to ignore
"exact" perspective projections and instead exploit
perspective for its representational, expressive
effects — mixing correct perspective buildings with
"incorrect" perspective figures, obeying perspective
recession but "bending" long foreground lines, and
always adjusting the circle of view and center of
projection to suit the subject, format, and installation
of the work.

The perspective foundations in ironclad geometry

and intricate drawing disguise how much exploration,
improvisation and creativity artists historically
allowed themselves when using perspective methods.
Raphael's figures and celestial spheres do not need
to be in correct perspective because they combine so
well as icons within an elegantly designed
symbolization of space. The rules of linear
perspective only help us to create the symbols, not
combine them into works of art.

N E X T : Central Perspective
 Last revised 07.I.2015 • © 2015 Bruce MacEvoy

central perspective technique

In this page I apply the circle of view framework to
create the simplest type of perspective drawing,
based on one point perspective (1PP) or central
perspective. Central perspective is governed by a
single vanishing point: its fundamental visual
characteristic is recession in depth or perspective
convergence.

This page addresses three general topics. The first is

the mechanical task of mapping objects from
physical space onto the image plane — actually using
the 90° circle of view framework to make a
picture. I explain a refinement known as the visual
ray method that makes it simple to find vanishing
points and to build perspective forms from drawings
or measurements placed along the ground line. Then the visual ray method
I introduce central perspective in terms of the 90°
circle of view framework, and in particular the use of one point perspective
orthogonals and diagonal vanishing lines to define
unit dimensions of width or perspective depth. orthogonals & central recession

The second general topic is the artist's control of this slanting & sloping planes
powerful depth illusion. I explain the triangular
geometry that governs the distance & perspective gradients
size relationships between object size, object
distance, drawing size and viewing distance; the distance & size
procedures for scaling the drawing that control the
size and arrangement of objects in perspective scaling the drawing
space; and the overall effect of display
geometry on the image impact. These topics are display geometry & image
conventionally obscured by cookbook perspective impact
construction techniques, but they offer many creative
resources for the design of compelling images. anamorphic images

The last general topic is the practical, step by step constructing a 1PP cube
method necessary to construct a perspective image.
I describe the early Renaissance methods, early renaissance methods
outlined by Alberti in 1435 and used by most 15th
century painters, as a point of contrast with
the modern approach.

I've written this page to emphasize the artistic

decisions that shape the geometrical construction of
a perspective image. Take the time to master the
concepts and methods on this page, and all other
perspective topics will come easily. (All the terms
used in this page are defined in the perspective
glossary.)

Before we start: why it is customary to illustrate

perspective constructions with those boring
rectangular solids? First, all edges are straight lines
and all corners contain 90° angles, so the forms are
easy to visualize from any point of view. Second,
cubes illustrate the basic perspective strategy of
plotting only the significant points: we only need
eight corner points to create the whole form. Third,
cubes and rectangles represent the basic volumes of
buildings and architectural spaces, which traditionally
have been the main application of linear perspective.
Last, a square grid or "cage" provides the framework
for the perspective projection of very complex curves
or three dimensional objects, as we'll explore in
a later section.

the visual ray method

The 90° circle of view framework and the basic
rules of perspective provide the mechanism for
creating perspective images of any real object on an
image plane. Now, we put it to use.

The challenge is to define an object geometry in

perspective space: locate it in relation to the
direction of view, scale it to the appropriate size, and
find the vanishing points for lines that define its
edges. The solution is called the visual ray
method of perspective construction. This was
repeatedly suggested in the 17th century, but was
first systematically applied to perspective problems
in 1715 by the English mathematician Brook Taylor.
Once mastered it is an elegant and powerful way to
solve any perspective problem.
The Double Fold. The visual ray method requires a
radical transformation in the basic perspective
geometry: a "double fold" around the image plane
(diagram, right).

• the viewpoint in front of the circle of view is folded

upward along the horizon line — just as we'd fold up
the tailgate on a truck — until it is lying in the image
plane above the horizon line.

• the ground plane behind the circle of view is folded

downward along the ground line — just as we'd bring
down a garage door — until it is lying in the image
plane below the ground line.

In this flattened position we see the viewpoint and

ground plane from below (as if lying on our back
looking upwards) at the same time that we
are looking through the circle of view from the
original viewpoint. This means the horizon line folding the ground plane and viewpoint into
also represents the image plane seen edge the image plane
on (from below).

Because the viewing distance is equal to the radius

of the 90° circle of view, the viewpoint lies at the
intersection of the circle of view with the median
line: it is directly above the principal point. At the
same time, the ground plane extends from the
ground line. Three different planes — the viewing
plane, the image plane and the ground plane — are
one.

The ground plane and viewpoint keep the same

relationship to each other — centered on the median
line — which preserves the triangular
proportions between the object, image area and
viewpoint. As a result, we can use a compass, ruler
and pencil to construct all the perspective
relationships, because they all lie in a single plane.

The folds have a specific significance: the ground line

is the intersection of the ground plane with the
image plane, and the horizon line is the vanishing
line for the ground plane. According to
perspective rule 11, planes are always bounded by
these two lines in the perspective image, and the two
lines are always parallel. The visual ray procedure
will work in any situation, even when the planes are
tilted to the horizon line, provided the folds are made
on the intersection and vanishing lines of the plane
that is to be scaled, and the viewpoint is folded in
relation to the vanishing line.

Visual Rays and Perspective Depth. First, let's

validate this new setup by repeating
the perspective problem introduced at the
beginning: constructing the perspective image of
a metric grid. This is done in two parts:

(1) The horizontal spacing of the orthogonals of the

grid is found from the intersection of the orthogonals
with the ground line; the intersections are projected
into the perspective image by connecting them with
their vanishing point (the principal point, vp).

(2) The vertical spacing of the transversals of the

image grid is found by drawing a visual ray from the
grid point in the ground plane to the viewpoint; the
transversal is at the intersection of this visual ray
with the orthogonal of the grid line in which the point
lies.

Thus, two lines are necessary to locate an

image point: (1) a visual ray from the physical
point to the viewpoint, and, in central perspective,
(2) an orthogonal from the ground line to the
orthogonal vanishing point (which is the principal
point in central perspective).
finding transversals with the visual ray
method
In the diagram, the points a through e mark
intersections in the metric grid on the ground plane.
Points a, c and e line on the orthogonal that
intersects the ground line at v; points b and dlie on
the orthogonal that intersects the ground line at y.

The first step is to connect the ground line

intersections (v to z) with the orthogonal vanishing
point (vp), to create five orthogonals in the
perspective image.

The second step is to connect each point by a visual

ray (blue lines) to the viewpoint.

Each visual ray intersects its corresponding

orthogonal at the image of the grid intersection in
the image plane: this defines points a' to e'.
Horizontal lines through these points define the
transversals, or the metric grid units of perspective
recession. (Compare this diagram with the earlier
diagram based on the elevation and plan.)

Why does this method work? I won't give the

geometrical proof, but the diagram (below)
demonstrates the geometrical basis in equivalent
elevation or plan views.
 why the visual ray method works

(top) the standard perspective setup, where diagram

represents both an elevation and a plan
view; (bottom) the visual ray setup, showing separate
locations of the image point (ip) and principal point (p)
when diagram is viewed as an elevation or a plan

In the original setup for the perspective image, the

visual ray from the physical grid point g to the
viewpoint intersects the image plane at the image
point (ip). This is located some distance b below the
principal point p (in the elevation) or shifted to one
side of the median line (in the plan). The image
point ip defines a matching distance a above the
ground line G(in elevation) or toward the median line
from intersection of the orthogonal with the ground
line (in plan).

Thanks to the triangular proportions

(explained below), the image distances a and b are
in the same proportion as the physical distances from
the physical point to the ground line (X) and from
the viewpoint to the image plane (Y). Therefore: a/b
= X/Y. As a result, we can add the corresponding
physical and image distances (X+a and Y+b) and
end up with the same proportional location of ip on
the visual ray (blue line). This means ip can be found
as before, at the intersection of the visual ray with
the image orthogonal drawn from the ground line
intersection to the principal point (red line in plan).

Visual Rays and Vanishing Points. The next

application of the visual ray method permits the
artist to find the vanishing point for any line in
physical space.

To locate the vanishing point(s), we require a plan

of the primary form. Recall that a plan is the
view straight down on the ground plane, which
shows the location, length and angles between walls
or sides of the form (for example, the floor plan of a
house). The primary form is the object we want to
draw in perspective — a skyscraper, a house, a
table, a chair — or a single representative form that
defines the perspective orientation of many other
forms, such as a single tile in a square tiled floor.
(A square plan is ideal, as it defines the diagonal
vanishing points as well.) Then we proceed in three
steps:
 plan of primary form located at the
viewpoint

the method used to find vanishing points on horizon

line; primary form in the orientation for central
perspective

1. Turn the plan to the same angle to the

direction of view as the actual object. In central
perspective, the four sides of a cube are
either parallel or perpendicular to the image plane
(ground line). So we rotate the square plan (orange)
in the visual ray setup so that its sides are either
parallel or perpendicular to the ground line and
horizon line.

2. Align one side exactly with the viewpoint.

Because all visual rays must pass through the
viewpoint, we can define a visual ray with any plan
line that is placed on the viewpoint.

3. Extend the plan line on the viewpoint to the

vanishing line of the plane that contains the
plan. For objects on the ground plane, this is the
horizon line. This intersection defines the vanishing
point for the plan line and all lines parallel to it in the
image.
This procedure works because any visual ray parallel
to a line intersects the line's vanishing point
(rule 5), and this point, because the line is parallel
to the ground plane, must lie in the vanishing line for
the ground plane (the horizon line, rule 14).

The central perspective diagonal vanishing point

(dvp) is found by extending a line from the
viewpoint through the diagonal corner, and
symmetrically on the other side. As shown above,
the vanishing point is at the principal point (direction
of view), and the diagonal vanishing points are at the
intersection of the 90° circle of view with the horizon
line.

In most situations it is necessary to work with

a much reduced plan in order to manipulate it
conveniently on the perspective layout. But so long
as the plan is turned so that its sides have exactly
the same angle to the direction of view as the
original object, and is large enough to produce an
accurate extended line to the horizon line, then its
size does not matter. Indeed, if convenient, we can
simply use a protractor, centered on the viewpoint,
to measure the visual angle that matches the angle
of the plan side to the direction of view.

Visual Rays and the "Principal Foundation". The

mention of a "much reduced" plan brings us to a key
practical problem. What if the form we want to draw
is a considerable distance away from the viewpoint?
Do we need a ground plan 100 meters long to work
out the perspective image of an object 100 meters
away?

The solution to this problem involves several steps,

but it starts with Taylor's "principal foundation of the
practice of perspective" — the fact that any image
line that is not parallel to the image plane must end
in an intersection with the image plane and in a
vanishing point (rule 4):

1. The intersection with the image plane of any line

contained in the ground plane is a point on the
ground line.

2. A line of sight parallel to the line in the ground

plan intersects the image plane at the vanishing
point for both lines (rule 5).
3. Because the lines are parallel, their vanishing
point is on the vanishing line of the plane parallel to
them both (rule 14), which for the ground plane is
the horizon line.

4. If we can define the intersection and vanishing

point for a line not parallel to the image plane, then
we can construct the image of that line by connecting
the two points (rule 3).

Let's demonstrate with a simple case: the line

images for two parallel line segments ac and bd.
 constructing two parallel lines with the
visual ray "end points" method

parallel line from viewpoint is used to find the vanishing

point; intersections with ground line are used to define
image lines from vanishing point and to construct
orthogonals that locate the end points (a' to d')

The lines ab and cd are not parallel to the image

plane, so they must define an intersection and
vanishing point on the image plane (rule 4).
To find the ground line intersections, we extend the
lines toward the image plane until they intersect the
ground line, creating the two intersection
points (orange), A and B.

To find the vanishing point, we construct a line from

the viewpoint that is parallel to ac and bd. Because a
visual ray is its own vanishing point (rule 1), and all
parallel lines converge to the same vanishing point
(rule 6), the intersection of this visual ray with the
horizon line defines the vanishing point for acand bd.
We then connect the vanishing point vp to the
intersection points A and B to produce two extended
image lines.

How do we find the endpoints of the two line

segments? As before, we could just draw the visual
rays from the endpoints of each line to the
viewpoint: the intersection of these rays with the
corresponding line images we just constructed would
define the endpoints in the image (a', b', c' and d').
However, this method becomes impractical when the
object is located very far from the ground line or
very far to one side of the median line: we'd need a
huge ground plan to locate it.

Instead, we use the ground line as a ruler. If we

know how far the point we want to find is to the left
or right of the median line (arrow to point d), we can
simply measure this distance directly at the ground
line. This is equivalent to extending an orthogonal on
the ground plan from point d to the ground line.
Using either method defines the measure
points w, x, y and z.

As all orthogonals are parallel to the direction of

view, the principal point is their vanishing point
(rule 7). So we connect the ground line measure
points to their vanishing point to construct
four measure lines (which are always image
orthogonals in central perspective) that extend the
ground line measure points (distances from the
median line) into perspective space. Then the line
segment endpoints (a', b', c'and d') are located at
the intersections of these measure lines with the
extended line images drawn before. Thus, the
triangle Awa' is the perspective image of the
triangle Awa in the ground plan.

The visual ray method has been applied in three

different ways: (1) to define recession in perspective
space (a metric grid), (2) to locate the vanishing
points for the sides of a plan, and (3) to construct
line images based on a vanishing point, a ground line
intersection, and orthogonals from ground line
measure points. All three methods show
the fundamental importance of a vanishing
point and ground line distances (from the median
line to the intersection or measure points) in
constructing perspective images.

Importantly, the ground line is always the

"actual size" ruler of object dimensions on the
ground plane and all planes parallel to it, that are
measured perpendicular to median line. If point d is
shifted one mile to the side of the median line in
physical space, then its measure point is located one
mile to the side of the median line along the
perspective drawing ground line.

Obviously the crucial step is still missing: how do we

work with objects that are very far from the median
line and/or ground line? These create ground line
measure points and/or intersection points that
cannot conveniently be located on the perspective
drawing ground line. This challenge requires the
artist to reduce the scale of the ground line
ruler as a substitute for working with unreasonably
large dimensions in actual size. This reduced ruler is
called a measure bar, and its use is explained
below.

one point perspective

In central perspective or one point perspective there
is only onevanishing point (vp), which is located
straight in front of the viewer at the principal point —
the vanishing point for the direction of view.

Defining Features of Central Perspective. The

characteristics of central perspective can be
displayed by placing a cube in the center of the circle
of view with its sides in the standard central
perspective orientation: either parallel or
perpendicular to the image plane and direction of
view. We additionally (and optionally) orient the cube
so that its sides are parallel or perpendicular to the
ground plane. In this view the cube is not tilted or
turned in any direction.

basic geometry of one point or central

perspective

There is one vanishing point, defined by the direction

of view. Its location is identified by the bold blue line
extending from the viewpoint, as explained above.
There are two diagonal vanishing points (dvp),
sometimes called distance points, at the
intersections of the horizon line and the 90° circle of
view. There are two vertical dvp's (not labeled)
where the median line intersects the circle of view.
All diagonal vanishing lines (shown as the two blue
lines extending at 45° from the viewpoint) end at
these diagonal vanishing points.

Because of the parallel alignment of the form to the

image plane and direction of view, some art texts
refer to this setup as parallel perspective. This is
incorrect: as explained later, all types of parallel
perspective, including the elevation and plan, are
constructed with lines that are actually parallel to
each other, rather than lines that converge to a
vanishing point.

Each perspective type is determined solely by the

orientation of objects to the direction of view, not by
the direction of view itself. However, the direction of
view does have its own vanishing point and its own
recession toward that point. Central perspective is
unique in that the lines of recession defined by
objects coincide exactly with the lines of recession
defined by the direction of view: two separate
recession geometries are laid one over the other.
This produces the most powerful illusion of depth
possible in a perspective construction.

orthogonals & central recession

The unifying framework for this central perspective
recession is formed by the orthogonals, which
are any lines ending at the principal point. A pair of
orthogonals defines a constant interval of visual
width (height, diameter, or distance parallel to the
image plane) across any perspective distance. Two
orthogonals automatically scale this dimension across
all depth locations in the image.

To illustrate, consider a single vertical dimension,

the unit dimension, drawn at any arbitrary location
on the image plane.

orthogonals from an arbitrary unit

dimension
By constructing two orthogonals from the
endpoints a and b of the unit dimension back to the
principal point (pp, the vanishing point vp for the
orthogonals per rule 7), the orthogonals scale the
dimension image to any distance in front of or behind
the unit line. They project a constant
dimension forward or backward in perspective
space — they show the visual width of an object at
any point in space. (In traditional texts, a pair of
orthogonals is called a vanishing scale.)

Any line drawn on the image plane parallel to the

original unit dimension, and ending at the two
orthogonals, defines the same dimension at other
locations in perspective space. The green lines
(below) are all assuredly the same height in physical
space (assuming they are standing on the ground
plane), because their images are bounded by the
same two orthogonals.
 projecting the arbitrary dimension in
depth

Diagonals and the Unit Depth. But where in space

are the lines located, exactly? We lack a measure
of perspective depth— a unit dimension for the
physical distance of the object's location that
corresponds to the unit dimension of visual width or
height we have just constructed.

This problem is solved by projecting the unit

dimension into perspective space, so that it no
longer represents a height or width dimension
parallel to the image plane, but a depth dimension
toward or away from the image plane.

The logic of this method is based on the square

diagonal. Using a square plan means that the width
and depth dimensions are defined as equal, e.g. the
ratio of width to depth is 1:1. As the Renaissance
artists quickly realized, the square is the
fundamental geometry for controlling perspective
depth, and is essential to the perspective
construction of complex forms.

The circle of view method for finding the diagonal

vanishing points confirms that the dvp's — of the
direction of view or the viewer's central recession —
are always located on the 90° circle of view.
Therefore, a line drawn from either unit dimension
endpoint (a or b, diagram above) to its
opposite dvp (so that the diagonal vanishing line
crosses the opposite orthogonal) intersects the
orthogonal at a point x which defines a line
segment bx that is equal in length to the unit
dimension ab. But now it is the unit depth, the
image of the unit dimension measured in physical
space along a line perpendicular to the image plane.
(This procedure is in part why the diagonal vanishing
points are traditionally called distance points: they
can be used to define units of perspective depth or
distance from the viewer.)

Using vanishing lines from the dvp, we can either

project the image of the unit dimension backward in
perspective space (to x) or forward (to y). This
allows us to find the horizontal or vertical dimensions
of objects at unit depth in front of or behind the
location in perspective space defined by the original
unit dimension (at b).

Because the unit dimension, orthogonal line and

diagonal vanishing line define a triangle, and a
triangle defines a plane, the plane's vanishing line
defines the diagonal vanishing points at its
intersection with the circle of view. This vanishing
line is the median line for vertical dimensions and the
horizon line for horizontal dimensions. If the unit
dimension is rotated, its diagonal vanishing points
are defined by the vanishing line through the
principal point that is parallel to the unit dimension.
 constructing equally spaced transversals
or verticals in perspective space

(left) projecting a vertical dimension ab into

perspective space, using the bottom dvp; (right)
projecting an arbitrary unit width into perspective
space, using a side dvp

Now we have a mechanical procedure to construct a

"depth ruler" that defines equal units of perspective
depth (diagram, above):
1. Define the unit dimension, either as an arbitrary
or scaled line segment ab in the image plane (above
left), or as a ground line "actual size" length (above
right), and draw the orthogonals from its end points
to the principal point (vp).

2. Connect one end point to the opposite diagonal

vanishing point (dvp) in the vanishing line of the
parallel median or horizon line, to define the
intersection x with the opposite orthogonal.

3. Construct a vertical line from x to the opposite

orthogonal to define the matching intersection y.

4. Connect point y to the dvp as before.

5. Repeat steps 3 and 4, as necessary, to construct

more transversals deeper into perspective space.

The spacing is reversible: for a vertical depth scale,

transversals from the base of all vertical line
segments (e.g., b, x and all similar points) creates a
horizontal depth scale.

In the diagram (above), the diagonal ray

from dvp through bdoes not intersect the upper
orthogonal within the circle of view. Because the
lower (ground) orthogonal intersects the ground line
within the square formed by the circle of view, we
know the transversal in front of b is closer to the
viewer than the image plane.
Unequal Spacing in Depth. Among the most
important reasons for projecting unit dimensions is
the measurement of unequal intervals in depth. For
example, suppose you want to project in depth the
unequal proportions in a building facade or unequal
spacings among a group of trees. You simply use
a measure bar with these unequal intervals marked
on it.

measure bar for unequal distances in

depth
 shown in a 60° circle of view; dvp's lie on the 90° circle
of view

Diagonal lines from each point to the mp transfer

these measures into perspective depth at the points
where the lines intersect the vanishing line. Draw
new verticals at each point up to the other vanishing
line.

constructed drawing of unequally spaced

verticals
 shown in a 60° circle of view; dvp's lie on the 90° circle
of view

If these intervals are repeated in depth, the solution

is simple. Rule a horizontal line across the horizontal
vanishing lines at the point where the front edge of
the form is located in depth, then use horizontal
vanishing lines (to vp) to project the measure bar
into space, as shown above. Recreate the measure
lines (to dvp) from there, and draw the new verticals
up to the other vertical vanishing line.

You can, of course, use the measure bar in a vertical

orientation to define distances projected into space.
Just connect these bars to measure points (dvp) at
the top or bottom of the circle of view, not those at
the side.

Shifting the Unit Dimension. Finally, as explained

on the previous page, shift foreshortening has no
effect on a two dimensional perspective
image parallel to the image plane: so it has no effect
on a one dimensional unit length parallel to the
image plane.

We may freely move the unit dimension anywhere

we need it on the image plane, or rotate it to define
a unit length horizontally or at any diagonal angle,
with the understanding that in its new location or
orientation, it defines the unit dimension at
the same perspective depth from the image
plane. If we require the unit dimension at a different
perspective depth, we must use the orthogonals to
project it backward or forward in perspective space,
either before or after we make the shift.

moving a unit dimension to other locations

in perspective space

In the diagram, rotating the unit dimension ab so

that it is horizontal, as bc, defines the same
dimension in physical space. Shifting the farther unit
dimension at x to a new location at zdefines the
same unit dimension at that perspective distance.
Unit depth dimensions may not be shifted or
rotated, because the amount of foreshortening
depends on the distance of the unit dimension from
the principal point. Thus, although the line
segments zy and xv are images of the same unit
depth, the image lines themselves are not of equal
length: xv, because it is closer to the principal point,
experiences a greater amount of foreshortening.

slanting & sloping planes

Because all objects are viewed frontally in central
perspective, the vanishing point and level orientation
for objects and the direction of view are the same.
But this is not true in situations where a plane is
slanting left to right across the field of view, or
sloping up or down along the direction of view. The
final problem is to show these surfaces in central
perspective as well.

Slanting Planes. For surfaces that are slanting

sideways across the circle of view, the solution is
to rotate the circle of view around the principal
point (just as we'd turn a steering wheel) to create
the required orientation.
 a slanting plane in central perspective

This new plane might be a sloping roof, or the

banked curve of a NASCAR racing track, or a
receding bank of gentle hillside. In each case the
receding surface can be described by a plane that is
not the ground plane. Because the tilted plane is
parallel to the direction of view, its vanishing line still
passes through the principal point (rule 15).

The essential step is to draw a new horizontal

vanishing linethrough the principal point (pp), with
the same side to side slope as the physical surface.
Then draw a new median line
through pp perpendicular to this vanishing line.

Where these lines intersect the 90° circle of view

they define four new diagonal vanishing points
(dvp), and these can be used to project unit
dimensions parallel or perpendicular to the slanting
plane into perspective depth.

All unit dimensions or orthogonals can be freely

placed and freely shifted, as before, and the
procedures for constructing transversals still apply.

Sloping Planes. The more difficult problem is a

surface sloping upwards or downwards in relation to
the direction of view. The classic example is a
staircase rising before us.

The staircase can be simplified as two parallel

planes, which represent all the front edges or all the
back edges of the steps. The flat surfaces of the
steps are formed by a stack of equally spaced,
parallel planes in central perspective, cut front and
back by the sloping parallel planes. Thus, the object
we want to represent (the staircase) is in central
perspective because all its surfaces are either parallel
or perpendicular to the image plane. What is missing
is the sloping limit of its defining edges.

Because the two sloping planes are parallel, both will

converge to the same vanishing line (rule 14). But
because the planes are not parallel to the direction of
view, their vanishing line will not intersect the
principal point: it will lie some distance above it.

The solution is straightforward. We need to know the

slope of the staircase, which is determined by the
depth of each step and the height of each riser.
These create two sides of a right triangle, whose
hypoteneuse is parallel to the sloping planes. In the
example, we want something grand, so the steps are
25" deep and the risers 7.5" high.
 sloping planes in central perspective: the
staircase problem

The visual ray method permits us to fold the

viewpoint into the image plane — except that we
fold it to one side so that it coincides with a
horizontal diagonal vanishing point. In this
orientation the slope of a visual ray above or below
the direction of view is equal to its angle above or
below the horizon line. We find this angle by placing
an elevation drawing of a single step so that the tip
of the hypoteneuse is on the viewpoint and its flat
surface is on the horizon line. Then a line from the
viewpoint extending this hypoteneuse intersects the
median line at a vanishing point in the vanishing line
of the sloping planes. This is the slope vp, and we
draw the slope vanishing line through it, parallel to
the horizon line.

The next step is to scale and position the drawing,

which means constructing a unit
dimension ab (magenta) that indicates the location
of the base of the first step, the apparent width of
the staircase at its bottom step, and the scale of the
riser (height of the first step above the ground plane)
at each side (ac and bd). A line across the top (cd)
creates the perspective image of the first riser.

From this point it is simply a matter of plotting the

significant points. The key is to remember what goes
where:
• The risers are parallel to the image plane, so they
have no vanishing point; their sides are defined by
parallel vertical lines, the front and back edges of the
steps by parallel horizontal lines.

• The steps are horizontal and perpendicular to the

image plane, so the vanishing point for
their side edges is the principal point (rule 15)
regardless of their height in relation to the viewpoint.

• The vanishing point for the recession of separate

steps and the width of the staircase in perspective
(assuming the staircase is an equal width from
bottom to top) is the slope vp, which also controls
the location of the front and back edges of each step.

Starting at the base, we construct two lines from the

ends of the unit dimension (a and b) to the slope
vp, and a matching pair of lines from the riser
(c and d). The side edges of the steps are then
defined by orthogonals from c and d to the principal
point. These intersect the lower slope line at e and f,
which locates the bottom edge of the riser. Vertical
lines from these points define the sides of the riser,
which intersect the upper slope line at gand h ... and
the drawing continues by repeating these steps until
the top of the staircase has been constructed.

As a check, a line from the dvp through f should

intersect the front step edge at x, where xd is equal
in length (along the unit dimension and as a
proportion of the riser) to the depth of each step.

finished staircase drawing

As the steps recede their image spacing becomes

smaller and smaller, until they may become so small
that the method of constructing lines to the
vanishing points may become inaccurate. In that
case they can be drawn as horizontal lines spaced by
eye or using a ruler.
The side bannister of the staircase can be added
after the more complicated step construction is
complete and all guidelines have been erased from
the drawing.

perspective gradients
The problems of drawing unit transversals and a
receding staircase have produced drawings with line
spacings that become smaller with distance. This
effect of linear perspective appears powerfully in
the perception of textures. As textures are usually
too small to have vanishing points of their own, their
appearance is controlled by the powerful central
convergence around the principal point. This is the
perspective we always see no matter which direction
we look.
 a perspective gradient

shown in a 60° circle of view; dvp's lie on the 90° circle

of view

If we extend any series of regularly spaced lines or

small objects until they cover most of the circle of
view, we have constructed a perspective gradient.
This is the texture of all objects and surfaces in
space, even when they do not define specific
geometrical forms. The necessary emergence of
perspective gradients from the regular spacing of
objects in depth is demonstrated by adding parallel
orthogonals and diagonals (in blue) converging
toward the principal point and diagonal vanishing
point.
A perspective gradient has a distinctive appearance,
regardless of context (diagram, right). We look
downward or into a surface texture when it is at our
feet, but view surfaces from the sidewhen we look
into the distance. Three visual constants result.

First is foreshortening: object dimensions are

compressedalong the direction of view — the floor
tile that appears square under our feet looks like a
thin rectangle in the distance; The apparent depth of
texture elements — the front to back size of floor perspective gradient in abstract art
tiles, for example — decreases more quickly than the Torben Giehler's "Circling Overland" (2002)
apparent width (the decrease in size due to distance
alone).

Second, texture elements parallel to the

direction of view are visible at greater
distances: at the point where the spaces between
the black lines disappear, the blue orthogonals are
still widely separated. As we travel in a car or train
past a vegetable field, the rows of plants are much
more visible than their spacing in depth; the
horizontal seams in a brick wall are visible after the
vertical seams disappear.

Third, we tend to underestimate the distance of

objects or landforms in the middle to far distance,
especially if the true horizon is not visible. In
contrast, visual angles to the left or right of the
direction of view make our perception of direction
quite accurate.

perspective gradient and dominant

interval

perspective gradient intervals spaced at 3x viewing

distance (green) or 1/3 viewing distance (orange); half
the horizon height on the image plane equals double
the perspective depth

As this figure shows, the apparent slant or

"steepness" of the gradient is due primarily to the
proportional relationship between the viewing
distance (to the image plane) and the primary
texture or basic intervals on the surface being
viewed. As these intervals grow larger in relation to
the viewing distance, we seem to take a more
slanted view of the surface, and there is a more rapid
recession of the texture. This is why a plain lawn
looks "steeper" or more expansive than a complexly
textured wildflower meadow, and why the widely
spaced columns of a Gothic church apparently define
a steeper gradient than the tile pavement under foot.

A very useful constant in perspective space,

regardless of the viewing distance (radius of the
circle of view), is that half the horizon height is
always twice the viewing distance in perspective
depth, where distance is measured along the
direction of view from the ground line. In the figure
above, if we take the ground line as the distance to
the vertical image plane (1.5 meters along the
direction of view), then the third orange bar from the
bottom is 1.5 meters in perspective depth (each bar
represents 1/2 meter).

Perspective gradients are based on a general

perspective rule, described below: the size of an
image, including the size of the image area between
an object and the horizon line, is inversely
proportional to its distance from the viewer. In this
sense, perspective gradients, whether steeply or
gently spaced, are the raw material from which all
visual images are constructed.
Eventually textures dissolve into colors with
distance. You cannot keep drawing thinner and
thinner gradient lines all the way back to the
horizon; at some point you must change from
drawing lines to painting a gray color that is the
average value of the imperceptible lines and the
imperceptible white spaces between them. If you
want to see how the surface under your feet will
appear at a distance, you must put your head near
the ground and view it from the side. Close by, the
color of a long brick facade combines the surface of
the red bricks and the white mortar between them;
at a distance, only the surface of the bricks is visible,
and the apparent surface color darkens. Typically
ground color also darkens with distance (and the
ocean darkens near the horizon) because textures
scatter much less light parallel to their surface (along
that slanting view) than they do back toward the
light source.
Finally, forms dissolve into textures with
distance. Rendering this transition gracefully is one
of the higher skills in perspective drawing. The
essence is to simplify by extracting the essentials at
each perspective step. Draw the foreground house
with a detailed brick facade, then a sketchy brick
texture for the house behind it; draw the farther
houses with only schematic doors and windows, and
the farthest houses in outline only. The artist gets a
head start by not putting too much detail in the
foreground, and designing forms that are easy to
simplify: then the transition to less detail can be
managed gracefully.

distance & size

Know that a painted thing can never appear truthful
where there is not a definite distance for seeing it.

—Leon Battista Alberti [1425]

The discussion of unit dimensions, equally spaced

transversals and perspective gradients has touched
on the issue of spatial distance and perspective
scale. The circle of view framework provides
complete control of the relationship between
the distance from the viewpoint of physical objects
and the image plane, and the effect this has on the
apparent size of the physical objects and their
perspective images.

The Geometric Elements. The leverage on these

problems comes from the principle of triangular
proportionality, the geometry at the heart of all
perspective and optical images:

Given two triangles of unequal size, if the three

interior angles of one triangle are equal to the three
interior angles of the other, then the length of all
three sides of the smaller triangle will be in a
constant proportion to the length of the
corresponding sides of the larger triangle.

The proof appears in Euclid's Elements, Book 6,

Proposition 2; and the the optical implications were
developed in Euclid's Optics, written c.300 CE.

Let's start with distance. Given the perspective setup

and the fixed proportions of the circle of view
framework, a physical point located beyond the
image plane will create an image point at a specific
location on the image plane. But where exactly?

distance and constant triangular

proportions

the three interior angles of triangles XYZ and xyz are

the same, so each side of the smaller triangle makes
the same ratio with the matching side of the
larger: x/X = y/Y = z/Z
We first have the visual ray PV defined by the
direction of view; this intersects the image plane at
point p, the principal point, and is perpendicular to
the image plane (as indicated by the small square).
The viewing distance between the viewpoint and the
image plane (the line segment pV) is the viewing
distance (x).

We then choose any point A in physical space (in this

example located on the median line), which creates
the visual ray AV to the viewpoint V, and the
point a where this ray intersects the image plane.
This is the image of visual ray AV in perspective
space (rule 1). We want to know the image
size (z), which is the length of the line
segment pa on the image plane.

The point A is at some distance from the viewpoint in

physical space, and this distance is measured either
along the median line (from point A to the station
point S), or along the direction of view (from the
point P perpendicular to A to the viewpoint); this is
the object distance (X) measured from the
viewpoint. And finally point A is displaced from the
direction of view by distance PA in physical space. I
call this the object size (Z), although it may also be
the space between two objects, or the dimension of a
group of objects, or any arbitrary unit of
measurement in physical space.

We have created two right triangles, the

triangle Vap to the image plane, whose sides are of
lengths x, y and z; and the triangle VAP to the point
in physical space, whose matching sides are of
lengths X, Y and Z.

Because the two triangles share a common point (V)

defined by the same two lines (the visual
rays PV and AV), the interior
angles PVA and pVa (at 1) are the same. The two
angles apVand APV are also equal (they are both
90° or right angles), so by subtraction the remaining
angle must be equal as well.

All three angles within triangle Vap are equal to the

three angles in triangle VAP, so the triangular
proportionality holds. This means there is also
a constant ratio or proportionalitybetween the
lengths of the matching sides of the two triangles:

x/X = y/Y = z/Z.

However, we are rarely interested in the line of sight

distances defined by the diagonals (y or Y), so these
constant ratios are best summarized as:

(1) z/x = Z/X, or

If Z (object size) represents the radius of the 90°

circle of view (or generally, the image distance
between the ground line and the horizon line), then
this formula is the basis of all perspective
gradients: multiply the object distance by X, and
the image size is reduced by 1/X.

Key Triangular Proportions. Let's put this key

formula into more practical forms. Thanks to the
circle of view framework, we already know or can
arbitrarily define x (the viewing distance, for
example 1.6 meters). Then, if we can specify the
object distance X (either arbitrarily or by measuring
it in physical space), we can solve the ratio x/X and
with this derive the image size z from the known
object size Z:

(2) image size (z) = Z*(x/X).

Conversely, if we don't know the actual distance of

an object shown in a perspective drawing, but we
know its image size z, its actual size Z and the
viewing distance x used to construct the perspective
view, then the object distance X is:

(3) object distance (X) = x*(Z/z).

Finally, we can determine the viewing distance; this

formula is helpful when the artist wants to discover
the effects of varying the image size when our
distance to the object is fixed (for example, by a
constrained choice of viewing locations):

(4) viewing distance (x) = z*(X/Z).

It is also sometimes useful to determine the image
scale, the relationship between a unit of
measurement on the image plane (painting surface),
and the units of measurement that describe the
object's size:

(5) image scale = (z/Z).

This formula is used to determine the appropriate

scale for measure bars in perspective space.
However, note that the drawing scale in a
perspective drawing is dependent on the object
distance, unlike the scale used in street maps,
elevations or plans where all objects are flattened by
parallel projection into the image plane and therefore
are represented at the same scale.

The crucial assumption when using the triangular

proportions is that the measurements of both object
size Z and image size zare perpendicular to the
direction of view or (equivalently) parallel to the
image plane. When this is not the case the image
size is altered by foreshortening.

The table (below) presents illustrative measurements

for the display geometry of three objects: a
contemporary single story home (20' from ground to
roofline) to represent architecture or landscape, a
man of average height (5'9") to represent figures,
and a basketball (9.5") to represent still life objects,
as they would be drawn in paintings viewed from a
fixed distance of 1.6 meters (63").

distance, object size and image size

image size3 (cm)
distance from
image one
image adult
scale2 story basketball
plane1 (meters) male
house
-0.80 200% 1220 350 50
0 100% 610 175 25
1 62% 375 108 15
2 44% 271 78 11
3 35% 212 61 8.7
4 29% 174 50 7.1
5 24% 148 42 6.1
10 14% 84 24 3.4
20 7.4% 45 13 1.9
50 3.1% 19 5.4 0.8
100 1.6% 9.6 2.8 0.4
200 0.8% 4.8 1.4 0.2
500 0.32% 1.9 0.6 0.1
1000 0.16% 1.0 0.3 ~0.0
1distance from image plane: Distance (in meters) from
the image plane out to the object viewed; negative
distance means the object is in front of the image plane.
The viewing distance (from the viewer to the image
plane) is 1.6 meters (63 inches); the object distance is
therefore the distance from the image plane plus 1.6
meters.
2image scale: The image size of the object at the

distance from image plane (given in column 1) divided by

 the image size of the object at 1.6 meters (distance from
image plane of 0 meters).
3image size: The dimensions of the image of the object

on the image plane, in centimeters, when the image plane

is at a viewing distance of 1.6 meters.

These figures provide an overview of the effects of

distance on image size. You should be able to derive
the same figures for yourself, using the formulas
given above.
Circle of View Formulas. Linear perspective is object distance
visual angle image scale
(size = 1)
based on methods of construction, not calculation.
1/2 90° 170%
However, the trigonometric basis for the circle of
1 53° 100%
view may be useful. If you know the distance of an
2 28° 53%
object and its physical dimensions measured parallel
3 19° 36%
to the image plane, and you have a calculator with
4 14° 26%
the arctangent function, then the object's visual
5 11.5° 22%
angle or angular size when the object is centered on
6 9.5° 18%
the direction of viewis:
7 8.2° 15%
8 7.2° 14%
9 6.4° 12%
10 5.7° 11%
The term angular size is used when the visual angle 15 3.8° 7%
describes a single object — that is, the circle of view 20 2.9° 5%
that will just enclose it. Thus the visual angle 30 1.9° 4%
subtended by a standing man 1.6 meters tall viewed 50 1.1° 2%
from 50 meters is: 100 0.6° 1%

distance/size ratio with corresponding

centered visual angle and image scale
Conversely, if you want to know how far away an
object must be in order to fit within a visual angle
(circle of view) of a certain width in degrees (CoV°)
on the image plane:

For the standing man 1.6 meters tall to fit into a 36°
visual angle (circle of view):

The table (right) gives illustrative values for the

visual angle and image scale resulting from
various object distance/object size(X/Z) ratios. (The
image scale, applied to the object size, gives the
image size.)

When using formulas (6) or (7) in situations where

the object size is not measured parallel to the image
plane (that is, the object is viewed at an angle), then
you must first identify the angle between the object
dimension and the image plane, then apply a cosine
correction for foreshortening to the object size.
Then continue as before.
 scaling the drawing
The orthogonals in central perspective specify the
changing image size of an object at different
distances in perspective space. But where the forms
are located and their apparent sizein the image has
to be specified by the artist, by scaling the
drawing to correspond to a specific object location,
object size and object distance in physical space.

To do that we put the triangular proportions to

practical use, by choosing an image format and
then locating key objects within the image by use of
an anchor point and anchor line.

Recession and Format Dimensions. Let's

approach the choice of image format with a concrete
illustration. The diagram (below) shows the 90° circle
of view standardized on a 1.5 meter viewing
distance and viewing height.

To measure the central recession, three orthogonals

are drawn to the median line (b) and the ground line
diagonal points (aand c). These define two unit
dimensions, ab and bc, whose "actual size" at the
ground line is 1.5 meters.
 projecting a unit dimension in central
perspective

depth is indicated by a ground plane grid of squares 1.5

meters on a side; the emperor paper format sheet,
centered on the principal point, creates
a 102x154 cm "window" into the circle of view

Using the central perspective method for projecting

a unit dimension in depth, the diagonal vanishing
line from the ground line
intersection b to dvp intersects the opposite
orthogonal at d, which locates the transversal 1.5
meters beyond the image plane, or 3 meters from
the viewpoint (3 meters at 3 meters). Repeating this
procedure extends the transversals to a distance of
15 meters, where a 3 meter square is located for
scale.

The dotted outline shows the emperor sheet

(40x60" or 102x154 cm), among the largest
watercolor sheets available, centered on the principal
point as a "window" into the perspective space.
Anything outside this format cannot be shown within
an image it contains — specifically, anything on the
ground plane that is closer than 4.5 meters (brown
transversal). Thus, even a very large painting format
crops out the most exaggerated perspective
distortions, which lie outside the 60° circle of view.

The format "window" is always located on the

image plane, but it looks out onto a world of
objects reduced in size by the viewer's central
recession, by an amount that depends on its distance
from the viewpoint. So we cannot determine the
image scale from the format alone. However, we can
determine the reduction in the image area caused by
the format "window" in comparison to the round
"window" created by the 90° circle of view. This is
the format scale:

A rectangular format has three plausible dimensions

— height, width and diagonal (which is how TV and
computer screens are measured). The format
dimension to rely on is generally the dimension
that most constrains the primary form, or the
dimension that crops the ground plane. In the
example, the ground plane is cropped by the vertical
dimension of the format. The half vertical dimension
is 51 cm and the viewing distance is 150 cm, so the
format scale is 34% (51/150 = 0.34). (Along the
185 cm diagonal it is 62%.)

Stepping back to view the sheet from 255 cm (2.5

times its vertical dimension) reduces it in the 90°
circle of view to the smaller dotted outline, where it
spans approximately a 30° circle of view and a
vertical format scale of 20%. Now the transversals
(which still project the viewing distance in depth)
represent a unit distance of 255 cm, so the format
"window" crops the ground plane at 12.75 meters.
Simply by increasing the viewing distance, we have
reduced the format scale and cropped out more of
the foreground.
 the unit dimension within a smaller circle
of view

shown in a 60° circle of view; dvp's lie on the 90° circle

of view; the full sheet paper format creates a 56x76 cm
"window" into the circle of view

In this second diagram we have returned to the 1.5

meter viewing distance, shown for detail so that the
60° circle of view fills the image. The gradient of 1.5
meter transversals is continued to 18 meters, where
the 3 meter square is now located; a human figure
stands next to it for scale.
The yellow outline shows the full sheet paper format
(22x30" or 56x76 cm) centered on the principal
point. This full sheet "window" encompasses
approximately a 25° circle of view or a vertical
format scale of 18.6%. This view shows a tighter
spacing in the foreground transversals because the
transversals visible in the image are farther away,
and this tighter spacing produces a more
homogenous or gradual sense of perspective
recession: even moderate distortions in the ground
grid have been cropped out.

Now, if we assume that the format is not a window in

the image plane but actually a rectangular arch or
tunnel opening in perspective space, then its bottom
edge would rest on the ground plane, forming a new,
"virtual" ground line. This image baseline crops the
ground plane at a perspective distance of 8.04
meters (on the image plane, 28 cm below horizon
line). Half its width defines a unit dimension in
perspective space of 2.04 meters (38 cm on the
image plane).

How do we derive these depth measurements? By

applying the triangular proportions to the format
dimensions and viewing distance, as shown in the
diagram, below.
 triangular proportions in the image
dimensions

The three critical measurements are the horizon

height (the distance between the image horizon and
the image baseline, 28 cm in the example), the
format or image half width (for a full sheet,
38 cm), and the viewing height (distance from
viewpoint to station point). Then the baseline half
width, image half height, baseline image distance (in
perspective space) all follow. (For reference, the
format proportions b/a or a/b are listed in the table
of watercolor paper formats.)
These examples are intended to show the importance
of the viewing distance in relation to the format
dimensions as a means to crop the central
recession to a restricted area around the principal
point — that is, to reduce the circle of view. The
format size, viewing distance and viewing height do
not affect the viewer's recession, the spacing of the
transversals or the perspective gradient. They
affect how much of this recession is visible in the
image, and also the depth spacing attributed to the
same perspective gradient (either 1.5 meters or 2.5
meters in the example above). The amount of
cropping increases with a smaller image format, a
longer viewing distance to the image plane, a greater
viewing height, or a horizon line located below the
middle of the image format.

Despite these cropping effects, the relationship

between viewing distance and horizon height in
the perspective gradient remains constant. In
particular, the half horizon height rule applies to
the perspective depth of the image baseline
(diagram, below). The viewer's perspective
gradient or central recession remains constant in
relation to perspective viewing distance, regardless
of image size. The location of the image baseline in
perspective space only determines the viewing
distance that will be doubled.
 the "half height rule" and image baseline

regardless of image size or horizon height, half the

distance from the image baseline to the horzion line
equals double the viewing distance in perspective depth

In landscape images, a narrower view around the

principal point edits out the extreme foreground
distortions at the bottom of the image plane,
moderates the amount of distortion visible in objects
near the left and right format borders, and controls
the visual evidence for flatness, depth or wide angle
spatial volume presented by the image.
These issues are less significant in figure, portrait or
botanical paintings, where the subject is usually
vignetted much closer to the image plane and the
viewer's central recession is obscured by a flattened
or distant background. Even in these situations,
however, drama is produced specifically by the angle
of view (downward, level or upward) and the
recession in front of and behind the figure; portraits
from J.S. Sargent to Lucien Freud provide many
provocative examples in the use of perspective depth
and viewing angles.

Anchor Line & Anchor Point. The choice of format,

viewing distance and horizon height constrain the
central recession defined by the viewpoint. But the
perspective of principal forms depicted in the image
depends on the recession they display. And object
recession is a function of image scale and
vanishing lines.

So the next step is to determine the location and size

of the primary forms. This is done with two
references:

• The anchor line is a unit dimension that

determines the image size of the primary form, and
therefore the actual image scale. Most often this is
the front edge of a square, cube or rectangular solid,
the diameter of a circle or sphere, or the height of a
standing figure.
• The anchor point is a landmark on which the
primary form can be constructed. Most often this is
the front corner of a square, cube, pyramid or
rectangular solid, the center of a circle or sphere, or
the standing location of a figure. This point locates
the object in relation to the direction of view — left
or right, high or low.

There are two alternatives to this basic design step:

either start with the format dimensions and work out
the placement and image size of primary forms, or
start with the arrangement and relative size of
primary forms and then choose the best format
dimensions to display them. In practice, this means
you either (1) outline at a much reduced scale the
format proportions you intend to use, then make a
freehand perspective sketch of the object and
horizon line in the desired size and placement within
the format; or you (2) draw the primary form(s)
freehand at whatever scale is convenient, then "crop"
the image by adding a format outline around it. From
this sketch you derive the proportional
measurements for the anchor point, anchor line and
horizon line from the center point or edge of the
actual support.

Either way, you use exclusively artistic or

compositional criteriato design the image. For the
moment you ignore the perspective problems implied
by the sketch, because these are resolved by the
construction steps.
Scaling Steps. To motivate this example, assume
that we want to build the image around a primary
form 209 cm in height. Object size is
the only uncontrollable aspect of perspective design:
we don't change the size of objects, real or imagined,
only our view of them (and, with the choice of
viewing distance and format dimensions, our view of
the view).

1. Design the image. Using a sketch, locate the

primary form(s) within an appropriae or esthetically
pleasing format outline (longer for landscapes, taller
for portraits).

2. Choose the format dimensions. Scale the

height/width proportions of the format outline in the
sketch to the desired overall dimensions of the
finished image; this gives the format dimensions for
the finished work. Display considerations (the size of
the room where the image will be displayed and the
appropriate viewing distance at that location) and
practical format sizes are also factors.

Let's assume the artist has decided the image width

should be about 1.4 times the image height. Using
the table of paper formats she finds she could use
the quarter, elephant, super royal, full or double
elephant sheets, or any sheet in the A or Bseries of
metric formats; she decides to use the full sheet.
Other dimensions can be produced by trimming
sheets or cutting pieces from a roll.
 scaling the object size and distance in
perspective

based on a 150cm (59") viewing distance (viewing

height)

3. Determine the viewing distance and format

scale. I explain below my recommendation to make
the viewing distance approximately 2.5 times the
controlling (form constraining) dimension of the
format. This is 140 cm in the full sheet (56 cm*2.5);
but for continuity with the previous example I will
keep a viewing distance of 150 cm. This crops the
viewer's central recession to a 21° circle of view or
a format scale of 18.6%. Any form that is located in
physical space behind the baseline image distance of
about 8 meters and that completely fits within the
image dimensions, must be at an image scale of
19% or less. Note that the format scale
depends only on the (1) format dimensions and (2)
viewing distance.

4. Specify the image size of the primary form. "Look

and feel" design considerations and the layout in the
preliminary sketch suggest the primary form should
have an image height within the full sheet format of
32 cm. This produces an image scale of 32/209 =
15.3%. Because this is less than the format scale of
19%, the object is on the far side of the image
baseline.

Using formula 3, we find that this image scale

corresponds to an object distance of 9.8 meters
(about 32 feet). This is only a judgmental check on
the physical distance we are portraying in the image,
the sense of intimacy or distance in the image of the
primary form, which becomes salient when the
physical size of the form is familiar to the viewer (as
with portraits or human figures, buildings, most still
life objects).

5. Construct the principal point and horizon line. This

step determines the location of the format within the
circle of view and the horizon height that
determines the baseline distance, calculated in the
diagram above as (A*x)/a. Implicitly, the horizon
line in relation to the format indicates the angle of
view in relation to the ground plane: a horizon line
above or below the midline of the sheet indicates a
downward or upward view into the physical scene. To
keep the example simple, I have centered the format
on the principal point, which puts the horizon line
28 cm above the lower edge and the median line
38 cm from the side edge.

6. Locate the primary form(s) in relation to the

horizon line. This step is crucial. Because objects at
the same physical height as the viewpoint (e.g., the
camera lens or viewer's eyes) are always intersected
by the true horizon, the viewing height is always
reflected in the image by the intersection of the
horizon line with primary forms.
 horizon height and intersection with
primary forms

location of horizon indicates upward or downward angle

of view; intersection of horizon with primary forms
indicates viewing height

In the example, the primary form is 209 cm; high, so

(assuming both the viewer and the form are standing
on the ground plane, and the viewing height is equal
to a viewer's "eye height" of 150 cm), the horizon
line intersection will be located 150 cm from the base
of the primary form. That ratio (150/209 = 0.72) is
the proportion of the 32 cm image size that must be
located below the horizon line in the image, so its
base (the bottom of the anchor line) will be
0.72*32 cm = 23 cm below the horizon line. This is
the anchor point or base of the primary form. By
subtraction, the top of the anchor line will be 32–23
= 9 cm above the horizon line.

If the primary form is shorter than the viewing

height, its top will be below the horizon line by a
proportional amount. Thus the anchor point a
primary form 120 cm high will be located 150/120 =
1.25 times its image height below the horizon line.

7. Locate the primary form left or right of the median

line. The last positioning step is simply to shift the
anchor line left or right in the image to locate the
primary form within the format.

Because shift foreshortening does not affect

perspective image size, the anchor line can be
moved anywhere to the left or right within the
image; where you choose to place it depends on
which part of the primary form the anchor line
represents (left edge, right edge, center height), and
where you want the primary form located within the
image (right, left or centered). Let's assume, based
on the sketches and design considerations, that it
should be located 15 cm to the left of the median
line.

Given the object at an image scale of 15.3%, this

15 cm interval is 15.3% of the physical distance
between the object and the median line on the
ground plane. So the physical distance is the
reciprocal of the image scale times the image size:
 object size = (1/image scale)*image size

or, in the example, (1/0.153)*15 = 98 cm.

8. Determine the baseline units. The dimensions and

scale of the primary form have been worked out, but
it is usually essential to transfer the implied
perspective units to the format image baseline. The
baseline then can be used as a virtual ground line,
and dimensions can be projected in depth using
orthogonals from the baseline.

The baseline units are derived from the format

scale — which is 19% in the example. This scale is
the reduction in a unit dimension along the ground
line, as measured on the image baseline. (Rounding
is acceptable to make the construction of the scale
easier.) An image scale of 19% (0.19) means that
10 cm along the ground line are represented by
1.9 cm along the image baseline.

Starting at the median line, measure left and right in

units of 1.9 cm (10*0.19) to create a 10 cm ruler
along the baseline. Do the same along one side of
the image format (as shown above). Use these rulers
to construct the orthogonals necessary to project
distances in depth. Assuming you have fixed the
image support to a drafting table or watercolor
board, it is best to mark out this baseline ruler on a
piece of masking tape applied along (but not on) the
bottom edge of the sheet. Markings are made on the
tape, not on the support, so there is nothing to
erase.

9. Locate the diagonal vanishing points. The last step

is to locate the diagonal vanishing points in relation
to the image format, so that we can project
dimensions in depth and construct the vanishing
lines to establish the diagonal edges of forms.

These vanishing points are located on the horizon

line to the left and right of the principal point, at the
viewing distance of 150 cm. These can be
established with push pins or dots marked on a piece
of masking tape.

10. Construct the anchor elements in the image. We

have everything necessary to lay the foundation for
the perspective drawing:

• Place the horizon line at the horizon height, then

locate the principal point on the horizon line.

• Draw a 32 cm line 15 cm to the left of the principal

point (median line), from 9 cm above the horizon line
to 23 cm below it.

• Construct the image baseline in 1.9 cm units

(image size) = 10 cm (object size).

• Locate the dvp's 150 cm to the left and right of the

principal point.
By carefully scaling and locating the primary form
within the image, we have established the baseline
units necessary to draw the orthogonals and project
in depth the transversals necessary to add other
objects to the drawing and construct the central
perspective view. We have also established the
landmarks (horizon height and intersection of the
horizon line with primary forms) that signal the
viewing height and angle of view onto the physical
scene.

display geometry & image impact

The importance of the display geometry has been
recognized by painters since the Renaissance,
especially in the design of large area frescos, but to
my knowledge it has not received a systematic
modern discussion. John Ruskin, in his Elements of
Perspective (1859), avoids the issue by remarking
that "in all points the subject is one of great subtlety
and difficulty."However, I have already introduced
some of the problems in the section on perspective
distortions, and have laid the groundwork in the
previous discussions of distance & size and scaling
the drawing. This section applies by example the
principles developed there.

The Five Key Measurements. The discussion

of triangular proportions identified four key
measurements in the perspective encounter between
a viewer and a painting hung vertically at eye level:
(1) the object size, Z, the actual size of the person or
thing represented in an image; (2) the image sizez,
the actual size of the image of the object on the
image surface; (3) the viewing distance x, between
the viewer and the painting surface; and (4)
the object distance X, which is the distance between
the viewer and the actual object required to make
the actual object appear to the viewer exactly the
same size as the image of the object in the painting.

The discussion of scaling the drawing introduced a

fifth key measurement: (5) the format
dimension S, which is the height or width of the
total image surface, whichever is larger.

For convenience, the diagram below summarizes

these relationships.

triangular proportions in the display

geometry
 showing the object size (Z), the image size (z), the
object distance (X), the viewing distance (x), and the
format dimension (S); as a general rule, x = 2.5*S

Because the object size Z is a physical fact entirely

outside the artist's control, the artist usually
proceeds by first choosing format dimensions that
are convenient for the work — for reasons of
economy (smaller image areas require less labor),
process (the image dimension limits of a letterpress),
or esthetics (larger works have more impact) — then
designing the image size z to place the primary form
within the format area. This effectively ignores the
object distance X as imaginary or irrelevant, and
leaves the viewing distance x for the viewer to work
out individually, wherever or however the finished
image is eventually displayed.

However the underlying creative problem actually

creates the need for a balance among three
dimensions:

x <—> S <—> z
viewing distance <—> format size <—>
image size

so that the apparent object distance X represents a

visible solution among these competing constraints.

The result is a specific display geometry. The

primary form appears as it would be seen from a
specific and virtual (only apparent) physical distance,
and this representation is viewed from a situated
viewing distance. These factors contribute to
the image impact — the esthetic or emotional effect
of the image separate from any features of its design
or composition.

I propose that the painting encounter can be

analyzed as follows:

1. The most important choice — the painting

attribute that directly affects all the others — is the
artist's selection of the controlling format
dimension (S). This is typically the largest
dimension of height, width or diameter measured on
the image surface, but it can instead be the
dimension of the image that crops the object in the
image area — the width of the format in portrait
orientation, if the portrait crops out the sitter's
shoulders, or the height of the image in landscape
orientation, if the image crops out objects on the
ground plane (1, diagram below).

triangular proportions in the display

geometry
2. Hung vertically at eye level, this largest or object
cropping format dimension defines a visual width or
angular size that becomes visually larger or smaller
as the viewer approaches or moves away from it. By
adjusting his or her vantage until the format fills a
comfortable area of the visual field from a convenient
location in the room, the viewer settles on a personal
(but often gallery typical) viewing distance (2). For
many viewers in many situations, this usually makes
the angular size of the painting appear to be about
25° wide, which is equal to a viewing distance
around 2.5 times the format dimension. Thus, a
painting 4 feet wide is most comfortably viewed from
a distance of about 10 feet.

3. The viewing distance fixes the image visual

angle (3) or apparent size of the primary form in the
image. If the viewer can compare this image size to
the actual physical size of the primary form, then the
viewer achieves an awareness of a virtual object
distance — the distance in real space that would
produce an image of the physical object that
matches the object's image size in the artwork.

4. Finally, the viewer may use the shape, vanishing

lines and central recession visible in the image
(especially any perspective gradient in surfaces
roughly parallel to the direction of view) to judge
the diagonal vanishing point spacing in the
image. The viewer implicitly compares this to
the dvp spacing in his own (retinal) central
recession. These appear to coincide most strongly
when the viewpoint is at the center of
projection and the image is neither telephoto nor
wide angle. Note that this correspondence is
completely unrelated to the image size.

5. As a result, the viewing distance and format

dimensions determine the format scale or apparent
size of the artwork within the viewer's visual field.
The viewing distance is fixed, and observable by the
viewer as a real distance in real space. The apparent
size of the virtual object (its angular size in the visual
field), compared to the known actual size of the
object in real space (if the object is familiar, or
viewed in a recognizable context) creates an implied
viewer-object distance. For example, an image on a
postcard of the entire span of the Golden Gate Bridge
implies that the bridge is very far away (as the
photographer was when he made the image).

viewer dimensions of the display geometry

This leaves a balance of relative physical sizes to be
implicitly worked out by the viewer. The bridge on
the postcard does look far away, but it also creates
the covert illusion that the viewer is somewhat larger
than normal. Alternately, a photograph of a fly
enlarged to cover the entire wall of a natural history
museum, and because of its size viewed from across
a large room, has the effect of making the fly appear
huge but also, in partial compensation, making the
viewer feel smaller.

The interior cues to the viewing geometry — image

visual angle and dvp spacing — compete with the
physical format dimensions and the architecture of
the display venue in determining the viewer's choice
of viewing distance — they can either reinforce or
conflict with the balance of relative scales produced
by the other dimensions in play.

Thus the viewing distance is an esthetic

choice because it is defined by viewer preferences,
display conditions, format dimensions and image
content, and because it moderates the image impact
on the viewer.

Certainly, esthetic choices often take second place to

expediency. Most printing formats assume fairly
close viewing distances: art book illustrations are
normally viewed from a comfortable reading distance
(usually less than 2 feet) and, even as wall posters,
art images are typically reduced to compensate for
display in small domestic spaces. However, the
original painting or art object always produces a
characteristic esthetic relationship to each viewer
through its physical presence and his or her
individual preferences, and this is almost entirely lost
in any reproduction in any other medium.

Similarly, my experience and review of the visual

perception literature suggests that the optimal
viewing distance from a painting brings its largest
(controlling) dimension just inside a 25° circle of
view; at that distance, an encompassing view of the
whole work and details of surface, line and
brushwork have roughly an equal impact. This
implies a viewing distance of about 2.5 times the
controlling dimension — that is, the
ratio x/S (viewing distance divided by the painting
format dimension) is roughly equal to 2.5.

But my observation of viewers in many gallery

settings, from an oversold Van Gogh exhibition to a
sparsely attended weekday at the Berlin
Gemäldegalerie, suggests that the majority of gallery
visitors examine a work from much closer in,
sometimes in order to read the wall mounted
painting titles and annotations, sometimes because
of event crowding, nearsightedness or a clinical
interest in the painter's technique, but often just
because they can — that's where the barrier cable
along their shins implies they should stand.
Image Impact. Thus, the actual image
impact depends on three viewing factors:

• physical size of the format – to fill a 25° visual

angle, a large painting must be viewed from far away
and a small painting must be viewed from close by:
this is a physical experience within a real
architectural space that imparts a physical and
spatial sense of grandeur or intimacy to the viewing
experience.

• image size vs. actual size – the image object

appears to be larger or smaller than the actual object
would be if the viewing distance to the actual object
were the same as the viewing distance to the image;
a discrepancy between the two may have the effect
of changing the viewer's subjective size so that she
feels unnaturally smaller (if the image is larger than
actual size) or larger (if the image is smaller).

• viewpoint vs. center of projection – the shape

of objects, vanishing lines and central recession
(especially the perspective gradient) in the image
determines the correspondence between the image
diagonal vanishing points and the 90° visual
separation of the viewer's diagonal vanishing points
— in other words, the spatial alignment between
the viewpoint and center of projection on a line
perpendicular to the image plane. A discrepancy
between the two produces the sense of a telescopic
or wide angle image. (Different convergence effects
are produced if the center of projection is above or
below the viewpoint.)

The diagram below summarizes the five major

variations in these factors.
 five principal variations of the display
geometry

(top) key to symbols in standard central projection set

up; areproduction (actual size or life
size); b reduction; c enlargement; dtelescopic
view; e wide angle view; cp = center of projection

The geometries of extreme enlargement, telescopic

view and wide angle view are typical of optical
systems (cameras, telescopes, microscopes) or
images based directly on them. I do not analyze
those image issues here, but a superb general
discussion is available in The Camera by Ansel
Adams.

These "technical" images are characterized by

extreme differences between image and actual size,
and/or by large discrepancies between the viewpoint
and center of projection. These effects can be
amplified or minimized by the format dimensions or
image design — "wide angle" effects are less
noticeable when the image is cropped to the area
close around the optical axis, and "telescopic" effects
are less noticeable in images of planes parallel to the
image plane, by images of objects governed entirely
by one point perspective, or by images of objects
that do not occlude (stand in front of) other objects
at a large distance behind them.

Three Illustrations. Let's apply these principles to

three examples from the western canon, to clarify
the implications of reduction, reproduction and
enlargement in paintings, drawings and photographs.

Reduction. In Albert Bierstadt's Looking Up the

Yosemite Valley (c.1865), the dominant image
element (the glorious El Capitan formation) has an
image height of 64 cm. This implies a viewing
distance of about 1.6 meters from the painting. As
the actual rock is about 1000 meters high, a strict
perspective solution indicates an object distance of
about 2.5 kilometers.

reduction in Bierstadt's
Looking Up the Yosemite Valley

If we take up a map of Yosemite Valley and draw a

circle around El Capitan with a radius of 2.5
kilometers (to scale), we discover that the circle
passes through the western location known as Valley
View (diagram, right).
This is the historically popular vantage for enjoying
the valley panorama on first visit and it clearly was
the location from which Bierstadt made his painting.
It also demonstrates that Bierstadt carefully matched
the visual size of the cliff, as seen by the naked eye,
to format dimensions that invite the 1.6 meter
viewing distance necessary to recreate that image for
the viewer. (In academic parlance, he probably used
the sight size method of image construction.)
geography of bierstadt's yosemite painting
Even though this is a "window view" or accurately
sized image for a correctly positioned viewer, the
object shows extreme reduction in order to fit the
format. A similar reduction appears in late medieval
and early Renaissance bust portraits, altar pieces and
carved sculpture, in landscape and historical
paintings up through modern times, and in most
wildlife paintings since the 15th century. (John James
Audubon's lifesized wildlife portraits, reproduced in
double elephant format, are a stunning exception.)
Reduction is consistent with idealization and a
nostalgic or lyrical interpretation of the image, as if
objects became smaller in memory. It also unifies
the representation of atmosphere, character or
composition, and subordinates detail to mood.

Reduction also accents the center of projection (or

viewpoint): the intended image effect is only
recreated at the correct viewing distance. Stepping
back 20 meters drastically reduces the apparent size
of the El Capitan contained in Bierstadt's painting,
but does not affect the apparent size of the actual El
Capitan. Baroque Dutch landscape etchings appear
dull and dark if viewed from too far away, scratchy
and disheveled if viewed from too close; but within
about a 25° circle of view they blossom into poetry.

Reproduction. Sandro Botticelli's splendid Birth of

Venus(c.1485) as it hangs in the Uffizi (Florence)
creates an unforgettable impact the first time you
see it, utterly different from any book illustration —
wow, it's so big! The dominant image element is
Venus herself, who is portrayed at about 145 cm
high; this implies a viewing distance of about 3.6
meters with eye level at the background sea level.
(At my encounter, this was perversely rather hard to
manage in the actual gallery space, as the painting
was hung too high and the central vantage was
blocked by a display case.)

reproduction in Botticelli's Birth of Venus

We can only guess the object size — the stature of

the model who posed for Venus, or the typical height
of women in renaissance Florence — but women then
were shorter than women today, with an average
height of perhaps 160 cm. In any case, it is clear
that Botticelli is using the strategy
of reproduction, creating an essentially life size
image in which the object virtual distance equal to
the viewing distance. Reproduction is common in
Baroque and modern portraiture (both bust and
figure), figure nudes in sculpture and painting
(Titian's Venus d'Urbino, Rembrandt's Bathsheba at
Her Bathand Velàzquez's Rokeby Venus are similarly
near reproduction scale) and, in smaller formats,
many still life and botanical paintings from the 17th
and 18th centuries.

The main impact of reproduction is a sense of

presence, a feeling that the object is actually
contained in the image or, in the case of Venus, that
we are carried into the actual scene portrayed. (The
location of the horizon, at Venus's waist, signals that
the viewer's eyes must be at the same level, as if her
presence has compelled us to kneel or stumble in
awe.) This is certainly the intended effect of the
grotesquely huge historical or mythological paintings
one sees in Europe (for example, in the Paris Musée
de l'Armée or the Rubenshuis in Antwerp), and it is
characteristic of much commemorative art, from the
figure portrait paintings of Rubens to
Picasso's Guernica. In many instances the canvases
are also physically huge, forcing the viewer back into
the gallery space and creating the illusion that a
historical moment has mythically erupted into
present time.

Because the image is very near life size, changing

the viewing distance produces the same effect as
changing the object distance. However images at
reproduction scale are not indifferent to the viewer's
location. Instead, the viewer seeks the center of
projection with the same intuitive sense of distance
that he or she would use to find a place within the
scene; the choice of the best viewing location occurs
with a tangible sense of "rightness".

It's my impression that specific esthetic effects are

produced when paintings or photographs are close
to, but perceptibly different from, object
reproduction. A small reduction (the ~10% reduction
in Botticelli's Venus or the ~20% reduction in
Manet's Olympia and in many figure paintings by
Lucian Freud) dilutes or distances the apparent
realism somewhat, causing it to merge into fable,
nostalgia, idealization or memory. A small
enlargement gives the image an exaggerated realism
and individuality and emphasizes its physical
presence.

Enlargement. Finally, many modern paintings go in

for enlargement, as exemplified in Chuck Close's Big
Self Portrait(1968). If we take the support as the
controlling dimension, then the viewing distance is
quite large — 5 to 7 meters. (The dominant image
element, the artist's face, implies a viewing distance
of 4 meters.) Yet the object (the artist) is implicitly
placed at a conversational distance, less than a
meter from the viewer. "So near, and yet so far!"
The effect is peculiarly unsettling, as we clearly see
facial details and flaws that polite social distance
serves to disguise. But it also has a miniaturizing
effect on the viewer, proportionately a mere fly
buzzing around the artist's head.

enlargement in Close's Big Self Portrait

This "viewer diminishing" effect of enlargement is

desirable in outdoor corporate advertising and
propaganda images (which strive to coerce), in
modern films (which strive to immerse and arouse),
and in commemorative or memorial images (which
symbolize superhuman virtues). The irony in Close's
portrait is the use of this display geometry for the
genre of self portraiture — and a sardonically
nonchalant and disheveled portrait at that.
Enlargement minimizes the effects of viewing
distance: the image expands into the available
space. Enlarged images change aspect less as we
move away from them and even seem to pursue us
across space; viewed from too close they dissolve
into a gibberish of detail. Other uses of enlargement
— one thinks of Monet's late lily paintings, abstract
expressionist canvases by Jackson Pollock or Morris
Louis, the Pop panoramas of James Rosenquist and
sculptures of Claus Oldenberg, the spiritualist
paintings of Joseph Raffael or Rudolph Stingel —
similarly imply modernistic disruptions of space and
uproot conventional frames of perception.

Conclusion. The commonplace painting procedure is

to choose a support that "seems right" for the
allowable time, available tools or intended display
impact, then to scale the image size to fit the
support. I argue that this procedure is not
necessarily wrong, just that it is unconscious and
habitual.

Georgia O'Keeffe thought differently about the

relationships between the viewer, the object, the
format, and the gallery encounter, and produced
some stunningly innovative botanical enlargements.
Rembrandt utilized the small format imposed by
intaglio mechanics to explore the poetics of memory
and mood in landscape imagery. These were choices
of esthetics, not convenience, and that difference is
partly why their works are so memorable.
Artists typically emphasize the "surface strategies" of
design and color as the keys to a memorable
painting, but I've argued that the display geometry is
an equally important ingredient in the image impact.
Give some thought to the display geometry — the
controlling effect of the physical format dimensions,
the correspondence between the image size and the
actual object size, and the correspondence between
the viewpont and the center of projection — and
you'll gain greater control of the image impact.

anamorphic images
Anamorphic images have been distorted so that they
appear flat or undistorted (veridical) when they are
(a) viewed from a direction that is not perpendicular
to the image plane; (b) viewed in a curved mirror or
other highly reflective object; or (c) painted on a
curved or faceted surface (i.e., the image plane is
not a plane).
As a simple example, the ellipse (right) is the
anamorphic image of a cross within a circle, if the
plane on which the ellipse appears is viewed from
above at an angle of about 45° to the computer
screen (more, if your screen is tilted). Foreshortening
causes the long dimension to appear visually smaller
in comparison to the width, compressing the ellipse
back to a circular appearance, and centers the cross
within the virtual circle.

Anamorphic images become a topic of interest early

in the 16th century, as an application of the
mathematical and geometrical study of projective
geometry beyond the restrictve assumptions of linear
perspective (a flat picture plane placed perpendicular anamorphic image of a circle
to the direction of view). Albrecht Dürer was so
adjust your angle of view from above until the
detailed in his study of geometrical projections ellipse appears as a circle
applied to human anatomy and complex curves,
published late in his life as the perspective
tutorial Instruction How to Measure with Compass
and Straight Edge (1525), that he is often cited as
the originator of projective geometry. (The basic
procedures illustrated by Dürer were systematically
developed as descriptive geometry by Gaspard
Monge in the 1760's.)

In the artist tradition, the popularity of anamorphic

projections peaked in the 17th century, especially as
applied in the design of frescoes on the curved
interor ceilings of European baroque church domes or
naves, so that these images would appear either to
be flat or to recede upward as a vision of Heaven.
This usage declined sharply by the end of the 18th
century. However anamorphic images have been
occasionally revived by contemporary artists, though
in the less saintly applications of visual puzzles and
street chalk drawings.

Geometrical Illustration. The procedures for

designing anamorphic images are mathematically
complex and vary with the viewing context. I
recommend later some software solutions to the
problem. Here I outline the basic geometrical
(constructive) soloutions to two basic applications: a
plane slanted to the direction of view, and a
cylindrical reflection.

Consider first the application of an anamorphic image

to an image painted or drawn on a sidewalk.

constructing a 1PP cube

To conclude, let's go through the steps necessary to
build a cubic or rectangular solid in central
perspective.
The primary form will be the Grande Arche located in
the Place de la Défense, Paris (right). The exterior
dimensions of this office building form a nearly
perfect cube, 110 meters on a side. Perspectivists
from around the world come here each year to press
their foreheads to the pavement, pray to the spirit of
Brunelleschi, and debate the exact street address of
the vanishing point.

This section will end up explaining

la grande arche de la défense, paris
the drawing procedures necessary to construct a
cube in one point perspective. But I'll use this
exercise to introduce and give examples of the
many artistic decisions or practical considerations
that go into planning and designing a perspective
drawing. You can't handle these decisions with
routine geometrical rules — they are matters of
drawing function, drawing design, artistic style,
display conditions and so forth.

Planning the Perspective Image. In the

discussion of display geometry, I explained that
good image composition represents a compromise
among the dimensions of image size, support size,
and viewing distance:

viewing distance <—> support size <—>

image size

Depending on circumstances, one of these three

constraints is usually more important. But a
reasonable procedure is as follows:
1. Specify the support size (image area). The
support size (paper format) is the most direct way to
define the size of the image and the impact of the
work, and it is typically the dimension constrained by
the architectural space in which the work will be
hung (or created, if it is a wall or ceiling mural). I'll
chose the large emperor format in landscape
orientation (40" x 60", or 102 cm x 152 cm).

2. Specify the approximate image size of the

dominant form. The dominant form is "what the
painting is about". This must fit into the support and
include an appropriate sense of space around it — to
put it in a physical context, a dramatic setting, a
decorative background, whatever the composition
requires. This is dependent on subject matter and
the artist's composition and style. I want the building
to show the surrounding sky and neighboring
buildings, so its image size should be about 1/2 the
support height, or 102 cm/2 = 51 cm.

3. Select an optimal viewing distance to the

support. This third step, in combination with the
previous two, determines the center of projection,
the object distance, the image plane circle of view,
and the image visual size. This dimension is sensitive
to the room in which the work will be hung, its height
on the wall, and so on. The 25° circle of view, as a
rule of thumb, suggests a viewing distance of
2.5*102 cm = ~2.5 meters judging from the support
dimensions, or 2.5*51cm = ~1.25 meters from the
image dimensions. However the building image is
proportionately small within the format, so I strike
the compromise 150 cm. Therefore:

•the radius of the circle of view is also 150 cm

•the image size is about 51 cm

•the visual angle created by the building image size

and viewing distance, according to the distance &
size formula 6, is 19°

•the object distance implied by the object size (110

meters), image size and viewing distance, according
to the distance and size formula 3, is about 323
meters

•the support diagonal (largest image dimension) is

183 cm; again using formula 5, this means the
image spans a 63° circle of view.

These calculations are simply checks on the choices

of support size and proportions, image size and
viewing distance. If anything seems not quite right,
now is the time to make adjustments. Everything
here seems fine — because I have already decided
that I want lots of space around the dominant form,
the "almost wide angle" compass of 63° will be
visually and dramatically effective — so I continue.

4. Choose the image point of view. Next the artist

will choose the type of perspective (angle of view)
toward the object. This is conceptually the same as
choosing the physical point of view toward view the
actual object. Explicitly it means that you choose
between one point, two point or three point
perspective; central perspective is the choice here.
Now you can specify the exact image size — rotating
the object in space will change its apparent width
and/or height. (Of course, if you knew at the outset
that you want the object to be viewed from an angle,
you would use the rotated dimensions as your
estimate of the image size in step 2.)

In central perspective the single vanishing point (the

principal point) has a very strong effect, a kind of
vortex for all the recession in the image. (In Baroque
painting, a central vanishing point was often used to
symbolize the immanence of God.) Because the
Grande Arche is itself a kind of tunnel, the tunnel
plus the vortex creates a very powerful effect. This is
fine if the dominant form fills the image area (as in
the photo, above right), but I want to include the
physical context, sky and so forth; and I don't want
this building, nice as it is, to symbolize God. So I
simply shift it to one side. This gives the building a
pleasing perspective shape, creates a dynamic visual
imbalance, and allows more of the surrounding urban
context into the image. (The point: these are design
decisions or artistic decisions: they are not
geometrical problems that you solve with cookbook
geometrical principles.)

5. Determine the vertical location of the horizon

line (in relation to the dominant object and the
support dimensions). The horizon line is basically the
image of the viewer's heightabove the ground
plane. So the placement of the dominant object's
image in relation to the line indicates our height in
relation to it. For the Grande Arche, if the horizon
line ran through the roof, we'd be viewing the
building from a height of 110 meters. If it ran
through the foundation, we'd be sitting on the
ground. Obviously, we can choose any placement we
want, and our artistic choice depends on the size of
the object and how we want it displayed.

In central perspective, extreme placements of the

horizon line have an exaggerated effect. Even so, I
choose to put the line somewhere near the base of
the Grande Arche, to give it an upward emphasis:
about 1/5th the height of the building from the
bottom edge, to give roughly a 21 meter viewing
height.

However, it is usually bad composition to divide the

image in half with a strong line (such as the horizon
line), and putting the horizon line in the middle of
the image would push the building comically up
against the top of the support. So I simply shift it up.
This is an artistic decision; but note that even though
this is a drawing in central perspective, the vanishing
point does not have to be "central" in the image!

6. Make a perspective sketch. At this point, make

a sketch of your perspective solutions — if you
haven't already! It's often easier, after step 1, to
outline the support in small scale, design the image
within this outline, and then work out the actual
dimensions from there; or use the support
proportions to crop or resize a photograph. Or simply
start with a freehand sketch of your concept, crop
the sketch to give the most pleasing composition,
then use the sketch dimensions to choose your
format dimensions.

At this point we shift emphasis from artistic design to

drawing procedures. The goal is a sketch that
contains all the information shown above, in a format
large enough to yield accurate measurements (12"
on the long size is usually sufficient). Then it is
necessary to locate the anchor line within the
drawing area or picture format. This step defines the
unit length, height and width of the form we want to
draw.
 the perspective sketch

shown in a 60° circle of view and emperor format (102

cm x 152 cm) for a 150 cm viewing distance

From the preliminary steps above, we already know

that the 110 meter Arche is going to be displayed at
an image size of 0.51 meters — the image is at a
scale of 0.51:110 or 1:216 (one centimeter equals
2.16 meters). This scale only applies to the Arche
and any objects in the drawing at an equal distance
(323 meters) from the viewpoint; objects closer or
farther away will appear at a different scale. But this
is enough for me to establish the scale of the anchor
line and measure bar necessary to construct a
detailed drawing of the building: 110 meters/2.16 =
51 cm.
Now refine the sketch — cleaning up the drawing,
adjusting the size or location of the primary form,
enlarging or reducing the format outline — until it
looks the way you want it.

Finally, using the location of the principal point and

the size of the circle of view, locate the vanishing
points and measure points, which in central
perspective are the same as the principal point and
the diagonal vanishing points. If these points are
outside the drawing, estimate their location from the
circle of view; choose a work area large enough to fix
them physically so that you can use a straight edge
(or a length of string or fishing line) to construct
perspective lines into the drawing.

6. Make the perspective layout. Now take these

measurements from the drawing (as shown above):

• vertical placement of pp, measured from the top

or bottom edge of the format. This is usually also the
vertical placement of the horizon line, unless your
direction of view is into the sky or the ground: 34
cm.

• horizontal placement of pp, measured from the

left or or right edge of the format: 76 cm.

• vertical placement of the anchor point,

measured from the pp or horizon line. A viewing
height of 21 meters is 1/5th the building height, so
this dimension is one fifth the image size: 51 * 0.2 =
10 cm below the horizon line (24 cm from the
bottom of the support).

• horizontal placement of the anchor point,

measured from the pp or median line: I choose 10
cm.

• length of anchor line, measured from the anchor

point: 51 cm.

Constructing the Perspective Image. Now you

have made all the preliminary decisions regarding
the perspective image. You can start the actual
perspective construction.
 frontal cube: the anchor line

shown in a 60° circle of view; dvp's lie on the 90° circle

of view

The first step in central perspective is always

to locate the anchor line within the drawing area.
This defines the unit length, the height or width of
the form we want to draw. The diagram shows a
vertical unit length, but a horizontal length is equally
good.
 frontal cube: completed front face

shown in a 60° circle of view; dvp's lie on the 90° circle

of view

Once we have the unit length, we can draw the front

face of the cube or rectangle. We draw all the corner
angles at 90° (right angles), because the face of the
cube is parallel to the image plane and there is no
perspective distortion in the apparent shape. If the
form is a square, we can use a ruler and a drafting
square, or standard construction methods, to
build the front face. If the face is rectangular, we
need a ruler to get the vertical and horizontal
proportions accurately.

frontal cube: recession and diagonal

vanishing lines

shown in a 60° circle of view; dvp's lie on the 90° circle

of view

Once we have the front face of the figure, we draw

lines from each of the four front corners back to the
vanishing point (vp). These lines are parallel to each
other in real space, which means they define the four
side edges of the cube, which are also parallel.
However, there is perspective foreshortening on the
length of these edges, so we do not know how far
back they should extend.

The face of the cube gives us the dimensions we

want: we just connect the front corners to the
opposite measure points (dvp). Where these lines
cross the vanishing lines defines the corner points of
the cube.

If the form is a rectangular solid, the length in depth

may not be the same as the length in height or width
on the image plane. In this case we use a measure
bar to project that length in depth.
 frontal cube: finished drawing

shown in a 60° circle of view; dvp's lie on the 90° circle

of view

Once we have the back corners, we can complete the

drawing and erase the vanishing lines we had to
draw to construct the figure. I've shown the cube as
an open box, to reveal some of the back lines.

Correcting Distortion. Unfortunately, as shown

below, there is typically something that doesn't look
right about a cube in central perspective. The front
face is perfectly square, which is a shape it can have
only when it is centered on and perpendicular to our
direction of view (dv), as shown above. Yet we can
see the outer side of the cube, which means our view
is oblique to the cube, not head on.

This distortion always arises when we look at a

perspective drawing farther away than the correct
viewing distanceimplied by the circle of view. In
this illustration, the radius of the circle of view is 4cm
on my computer, which means I should view it with
one eye 4cm from the central vp. That's not
comfortable!
 frontal cube: uncorrected (left) and
corrected drawings

This is a familiar problem in simple perspective

drawings, and it arises because artists commonly
overstate the size of the object within the circle of
view so that they can show more of its side surfaces
and depth. When this problem arises there are three
solutions. The first is to keep geometrical forms close
to or over the center of the circle of view, with all
parts inside a 30° circle of view. (The form appears
as if viewed from farther away.) This allows a greater
range of viewing distances with less noticeable errors
in the perspective facts.

The second solution is to devise the drawing so as to

minimize the areas of most obvious distortion. For
example, if the cube represents a room in your
house, you can crop the view so that the upper left
corner is outside the drawing.

Third, you can correct the drawing by hand to

make it look better. The correction is simple to do.
The worst problem is with the outer corners of the
cube, both in front and in back. In real space, we
would be looking into these at an oblique angle,
which would make a right (90°) angle appear to be
greater than 90° — the angle appears to flatten out a
little. To imitate this, we just have to move the
outside corners inward, straight toward the
direction of view, making this shift larger the
farther away they are from the dv. Then redraw all
the connecting lines to match.

The freehand adjustment is shown at right. It makes

the front face into a trapezoid or irregular four sided
figure, and makes the front edges no longer parallel.

Basically, you use the central perspective figure as a

first draft of your drawing, and by correcting it
freehand, using your eye, you've reasserted control
of the drawing and made whatever adjustments
seem most pleasing for the effect you want to create.
Or you may want to introduce distortions into a
perfectly accurate perspective drawing for expressive
effect. This is the theme that concluded the previous
page: don't let perspective tell you what to do if
you don't like the esthetic results.

early renaissance methods

It may be helpful to conclude with a description of
the perspective techniques used by early
Renaissance painters, who worked within a far
simpler geometrical framework. Their methods relied
on practical drawing methods and appear indifferent
to (or unaware of) the abstract perspective
geometry.
The Renaissance procedure, first described by Leon
Alberti in 1435, was anchored in the construction of a
grid of squares in the ground plane of
the perspective image. In the earliest paintings,
this grid was often carried into the finished image as
an ornamental pavement of contrasting Alberti
tiles, as shown in a painting by one of Alberti's
contemporaries (image, right).

Alberti's instructions for constructing this grid are

easy to follow (figure A, below): (1) define the Alberti tiles in the Feast of Herod by Fra
rectangular outline of the image area (on the canvas, Filippo Lippi (c.1430)
wood panel or wall), (2) insert the image of a
standing adult figure at the bottom edge of the
rectangle, to represent the appropriate vertical scale
of human figures at the very front of the image, (3)
draw the horizon line at the height of this figure's
head, (4) place the centric point on the horizon line
at or near the middle of the image area, (5) divide
the bottom edge of the image area into equal units
of scale (usually, into braccia, one third the height of
the adult figure inserted at step 2), and (6) connect
each of the braccia units along the bottom edge to
the centric point to define the
converging orthogonals.

renaissance method for constructing

central perspective

from the instructions in Alberti's De pictura (1435)

Lines that define the spacing of the tiles in depth

(the transversals) were constructed by duplication
the whole setup as a side view (figure B, above). The
steps were: (1) at one end of a long horizontal line,
draw a standing adult figure that corresponds to the
standing figure in the image rectangle, (2) divide the
horizontal distance in front of this figure
into bracciaunits proportional to the figure's height,
(3) at the desired apparent viewing distance (e.g.,
the location of the viewpoint within the world of the
painting, not necessarily the location of a viewer of
the painting or fresco in the real world), draw a
vertical line in front of the figure to represent the
image plane (canvas or wall), (4) connect with lines
each of the baseline braccia units to the eye of the
standing figure. The vertical spacing of the
transversals in the image is shown by the points
where the lines converging at the eye of the figure
cross the vertical line representing the image plane.
Then (5) transfer these vertical spacings back to the
wall or canvas (magenta lines) to locate the
transversals on the image plane. The accuracy of the
construction was confirmed if the diagonal corners of
the metric grid all fell on a straight line (the check
line, orange in the figure).

There are several interesting points about this

procedure. First, it does not describe actual
practice. We would expect that Alberti's book, as a
broad introduction to artistic methods, dealt with the
essentials rather than the refinements of perspective
construction. But the paintings by Lippi and others of
the time display a variety of artistic techniques that
have not been discussed in Alberti's text. Key among
these are the methods used to minimize or hide
the perspective distortions that occur when the
circle of view is too large (the viewpoint is too close
to the objects represented). The Alberti tiles and wall
cornices allow us to locate the principal point and
diagonal vanishing points, and these show that the
baseline of the painting crops the foreground at
approximately a 50° circle of view (diagram, below).
How, or why, was this chosen? In addition, the
skillful placement of side walls, staffage and the
upper decorative border have cropped the side and
top limits of the image within in an even smaller
circle of view. These refinements are not described in
Renaissance sources.

reconstructed central perspective in

Lippi's painting

Second, this method is inadequate to show very

large perspective distances behind the foreground
figures, because the transversal construction method
would require a physical baseline that was extended
out to the required distance. As a result, early
Renaissance paintings either represent a relatively
shallow, enclosed "shadow box" setting (like Lippi's
painting, above); or they show a central area of
perspective regularity carefully limited by staffage;
or they fill the background with a flat drape of
architectural facades or distant landscape. All three
devices are used in Botticelli's nativity images
(right).

More important, there's no geometrical rationale —

no explanation of the centric point as a vanishing
point, no understanding of how to locate the diagonal
vanishing points (distance points) before drawing
the transversals. Everything is developed step by
step from the image format and the scale of a human
figure within the image. This is surprising, given the
use of a distance point diagonal as a check line for
the perspective construction. Even more surprising,
the centric point was used to govern perspective in
parts of the image, the ground plan in particular, but
not in the whole. This is apparent in Lippi's painting,
because the vanishing lines for the wall cornices and
floor tiles do not converge to a single point.
Finally and perhaps most important, the perspective
of individual objects or figures is omitted from the
procedure. This is consistent with the lack of an
overall geometrical method. Artists could construct
the perspective grid that defines the stage and the
location on the stage of the actors and props, but
they did not explicitly develop the images of objects
(other than walls, tables, cornices, stairs and the
like) using strict perspective methods. With few
mixed perspective in "Adoration of the Magi"
exceptions (such as Mantegna, Correggio and by Sandro Botticelli (c.1500)
Tintoretto), painters throughout the early
Renaissance handled figure perspective much more
freely (or clumsily) than architectural perspective. In
Lippi's picture, for example, the front left figure is
huge in comparison to those standing just a few feet
behind, and the eyes of dancing Salome, in the white
dress at left, are at the same height as the seated
figures behind her. Even architectural features could
be represented with multiple vanishing points.
Sandro Botticelli seems sometimes to have done this
for dramatic effect, and even emphasized the
perspective disparities with strongly foreshortened
walls or platforms.

In its early stages, then, the emphasis in linear

perspective was on representation, not virtual
illusion. From the first, Renaissance artists were
indifferent to the peep showpossibilities of
perspective renderings: Leonardo mentioned them
only to dismiss them as impractical. In contrast,
examples abound of Renaissance frescos that place
the center of projection in an impossible viewing
location — below the ground, or behind a wall.
Leonardo's now cadaverous fresco of The Last
Supper puts the center of projection over 15 feet
above the floor!

In their simplicity and freedom of application, the

Renaissance methods show how much a desire for
coherence and consistency, both visual and
intellectual, was the impetus to later perspective
developments. Our modern understanding of
perspective is far more rigorous than the early
Renaissance approach. Although painting gained
thereby a heightened realism and virtuosity, it also
lost a uniquely naive, vernacular and lyrical portrayal
of human action in the world.

N E X T : Two Point Perspective

Last revised 07.I.2015 • © 2015 Bruce MacEvoy

 two point perspective technique
Two point perspective or 2PP is the most
commonly used construction method in artistic and two point perspective
architectural work, and it builds entirely on the
perspective techniques already introduced. rotating the vanishing points

The presentation of two dimensions of recession, and locating the measure points
the freedom we have to represent the primary form
from countless points of view, makes this framework constructing a 2PP cube
suitable for nearly all landscape, portrait and
architectural situations. The three vanishing points inclined lines & inclined planes
of 3PP perspective are usually necessary only for
downward or upward looking landscape views, or distance point projection
hovering views of buildings or still life objects.
the ground line framework
The unifying theme of this page is the correct
proportioning of objects in perspective through who has a 12 foot table?
methods for rotating vanishing points to a specific
angle of view and locating the measure points to VP spacing from an object
project in depth the dimensions of a measure drawing
barestablished to a specific image scale on the
image plane. where is the center of
projection?
Working with two vanishing points creates some
practical problems when the points are spaced very
far apart. I discuss some approaches to solving the
problem.

The problem must be solved rather than ignored,

because vanishing point spacing is critical to
the correct appearance of objects in space. The
final sections explain why, and describe methods to
identify the center of projection — the optimal
location for viewing the image — and the original
circle of view, using only clues within the image
itself. This can be done for any 1PP, 2PP or even 3PP
image where we can reconstruct diagonals or
vanishing points from the forms in the image.

two point perspective

Central or one point perspective represents a static
and symmetrical view of space and the objects within
it, with an emphasis on central recession in space
— the recession of perspective gradients and the
edges of forms exactly coincide.

Two point perspective represents a turning or moving

aside, a glance to the left or right, an approach to
the primary form that is more informal, idiosyncratic
and complex, composed as it is of two adjacent sides
of every object woven into a single perspective view.

This view is usually associated with the orientation of

the objects, which are simply turned at an angle to
the image plane. But we have already seen that 2PP
arises from a direction of view — a sideways
view — that is no longer perpendicular to the
primary form. These sideways, intimate or
idiosyncratic views greatly expand the range of
pictorial effects and emotional associations a
perspective image can create.

Defining Features of Two Point Perspective. The

diagram shows the simplest 2PP situation, in which a
cube is centered in view but rotated 45° to the image
plane.

two point perspective (2PP): the basic

geometry

In 2PP there are two vanishing points for the object

(vp1 and vp2) which form a right or 90° angle. Each
vanishing point also has its own angle to the
direction of view, shown in the diagram as the angle
between the thick blue lines at the top of the circle of
view and the median line (direction of view in
the visual raysetup). If one of these angles is x°,
then the other angle must be 90°–x° — the two
angles together must equal the 90° angle. All edges
and lines parallel to the horizontal edges of the cube
converge on one of these vanishing points.

There is no third vanishing point for the vertical

edges and lines parallel to them, because these are
parallel to the image plane. This is rarely a
disadvantage in the 2PP drawings of architectural
verticals, such as the walls of interior rooms or
exterior facades viewed from ground level. In these
limited situations the distortion in vertical proportions
is usually insignificant.

Now the perspective gradients created by recession

along the direction of view form one characterization
of physical depth, while the object's recession toward
the two vanishing points creates another. This
creates a palpable contrast between the viewer and
the object, highlighting the accidental or arbitrary
viewpoint of the viewer, or the distinctive physical
presence of the object. In traditional architectural
renderings, standard angles between the object and
the direction of view to the object — either 45°/45°
or 30°/60° — have been used to control this
contrast.
When the cube is turned at a 45° to the direction of
view, one of the diagonal vanishing points (dvp) is
now coincident with the principal point (dvp = dv)
and the other diagonal vanishing point disappears, as
its visual rays are parallel to the image plane, and
the contrast between viewer and object recession is
minimized or balanced.

No matter what orientation the object vanishing

points make take, the dv still defines the recession
in depth (changes in apparent size and textures)
produced by the viewer's central recession, and
the dv is still still used to project a unit dimension in
depth away from the viewer along the ground plane
and planes parallel to it.

However recession along the lines defined by the

object — changes with distance in the spacing of
object features, or in the spacing of a unit
dimension away from the object along the street plan
or geometrical grid defined by the object itself —
now create a separate system of recession in space.
Depth in this system is controlled by two measure
points (mp1 and mp2) which are used to project a
measure bar or unit dimension into the recession
created by each vanishing point.

rotating the vanishing points

In 2PP, the cube can be rotated around its vertical
axis to any position, as long as the vertical edges
remain parallel to the image plane. The side planes
of the cube, and all their horizontal edges, are no
longer parallel to the image plane. The first
perspective construction problem is to locate the two
vanishing points for these edges.

To review, in the visual ray method of perspective

construction, the radius of the circle of view is equal
to the distance of the viewer from the image plane.
So the geometry of the vanishing points is solved by
placing the viewpoint at the top of the circle of view.
In this position, we are looking down on the viewing
geometry. The vertical radius of the 90° circle of
view represents the distance from the viewpoint to
the image plane; the horizon line represents the
image plane viewed "edge on" from above, so it
shows the location on the image plane of all
vanishing points perceived from the viewpoint; and
the median line represents the direction of view.

In this setup, we locate the left and right vanishing

points through the angle of rotation between the
front face of the primary form and the image plane,
which is equal to the angle between a side face of
the primary form and the direction of view. The
"primary form" is simply the cubic or rectangular
solid that defines most of the perspective edges in
the drawing. For a room interior, it is the floor plan
that defines the visible floor and walls. For an
external architectural view, it is the visible sides of
the largest building in view.

With the viewpoint placed at the top of the circle of

view, we find the angle of rotation by drawing the
plan (floor plan or aerial view) of the primary form
within the top half of the circle of view, with the
corner nearest to the viewer placed on the location of
the viewpoint on the image plane. Then the two
vanishing points are found by extending the two
front sides of the primary form from the viewpoint to
the image plane or horizon line (diagram, below).

object geometry added to the circle of

view

primary form in the orientation for two point

perspective

This angle of rotation can be established exactly (by

using a protractor, or architect's triangles with their
standard 30°, 45° or 60° corner angles), or
approximately, by eye or feel. A useful way to
identify this angle from a photograph or physical
viewpoint is to mark on the plan of the primary form
the point where the front corner seems to cut the
back side of the form, creating the back edge cut.

Alternately, if the view is from a sufficiently far

distance, you can estimate the visual proportions
between the two front sides of the form (e.g., the
left side is visually 3 times wider than the right side),
then rotate the plan around the viewpoint until the
distance between the side corners and the median
line are in matching proportions.

If the circle of view is very large, the rotation can be

worked out on a reduced scale. With a compass,
draw a circle of view with a 7.5cm radius (15cm or 6"
wide), and mark the compass pinhole as dv. Scribe
the vertical median line over the pinhole, and locate
the viewpoint at the top. Then draw the horizon line
perpendicular to the median line and extending
beyond the circle of view on both sides. Next, lay a
sheet of paper over this circle with one corner at the
viewpoint, and rotate the sheet around this corner
until you have the approximate orientation of the two
sides you want. Make sure the corner is exactly
aligned with the viewpoint, then mark the two places
where the paper edges cross the horizon line.
Measure the distance (in centimeters) from these
two points to dv, and multiply by 20. You now have
the measurements of the two vanishing points
from dv in a 3 meter (~10 foot) circle of view.

Alternately, you can use a protractor centered at the

viewpoint on the circle of view to mark off the exact
angles of the two sides — just be sure the two angles
are exactly 90° apart. Extend these angles as lines
from the viewpoint to the horizon line, measure and
multiply by 20, as before.

The simplest (though nontraditional) method is to

derive the vanishing point locations as proportions of
the length of the circle of view radius at whatever
size you construct it. Any pocket calculator will
deliver these trigonometric ratiosdirectly, as
the tangent of the angle of any side in relation to
the direction of view. Landmark values are tabulated
below for reference.

VP spacing
for sides of a right angled object
(1.0 = radius of 90° circle of view)
angle to dv
left VP right VP
(degrees)
5 0.09 11.43
10 0.18 5.67
15 0.27 3.73
20 0.36 2.75
25 0.47 2.14
30 0.58 1.73
35 0.70 1.43
40 0.84 1.19
45 1.00 1.00
50 1.19 0.84
 55 1.43 0.70
60 1.73 0.58
65 2.14 0.47
70 2.75 0.36
75 3.73 0.27
80 5.67 0.18
85 11.43 0.09

In the example rotation diagram (above), the

lefthand face of the cube is at a 30° angle to the
direction of view, so vp1 is located 0.58 radius
lengths on the lefthand side of the horizon line,
measured from the direction of view dv. The
righthand face of the cube must be at a 90°–
x° (90°–30°) or 60° angle, which means vp2 is
located 1.73 radius lengths along the right side of
the horizon line from the direction of view.

The tangent derivation is very handy as it applies to

any surface or edge at any angle to the direction of
view — for example, the five sides of the USA's
Pentagon. All that is necessary is the angle of any
side to the direction of view; the angles of all other
sides can be derived by addition or subtraction, and
the tangent of these angles locates the vanishing
points as fractions of the circle of view radius along
the horizon line.
 locating the measure points
The next problem is: how do we determine
the spacing along vanishing lines as objects recede
from view? In central perspective the single
vanishing point (the principal point) defines recession
in space along the orthogonals (vanishing lines to the
principal point), and the diagonal vanishing points
project units of measurement from the image plane
onto the orthogonals.

In 2PP, the principal point still defines the viewer's

central recession and the perspective gradients that
the direction of view creates on the ground plane and
all planes parallel to the direction of view. But
the object vanishing points define their own
recession (vanishing lines) in two directions, and the
task is to establish units of measurement along these
vanishing lines.

We can still transfer a unit of depth established along

the vanishing lines toward one vanishing point onto
the vanishing lines toward the other vanishing point,
by means of the object'sdiagonal vanishing point
located within the 90° circle of view.

However we need a separate method for transferring

units of measurement from the image plane onto the
vanishing lines defined by either vanishing point.
This transfer is done with measure points, as first
explained by Brook Taylor in his New Principles of
Linear Perspective (1719).

Geometry of Measure Points. In the context of

the visual ray method of perspective construction,
the problem is to project a distance scale or unit
dimension, defined along the ground line, into
perpspective space along a vanishing line to one of
the object vanishing points.

the geometry of a measure bar

 image triangles Gaa' and GAA' are isosceles triangles
in physical space: the physical angles
at a, a', A and A' are equal and lines parallel
to aa' or AA' divide the physical sides aG and a'G into
line segments of equal length

defining an isosceles triangle — a triangle with

two equal sides and two equal interior angles.
One equal side of this triangle is in the image
plane, and the other side is in a vanishing line.
Every vanishing line ends in two points: its vanishing
point and its intersection with the image plane
(perspective rule 4). Thus the vanishing
line VG (diagram, above) must define an interior
angle VGA at its intersection with the image plane.
This is the image plane angle (x), and it is the
apex of the isosceles triangle we want to construct.

To find the base of the triangle, we construct an arc

in physical space from G with radius GA', the
distance along the vanishing line from the ground
line intersection to the front corner of the primary
form or line segment we want to measure in
perspective space (diagram, right). This arc
intersects the ground line at A. Then GA' = GA, the
interior angles (z) at A and A' are equal, and the
triangle A'GA is an isosceles triangle. Then the line plan view of geometry of measure bar
segment A'A, extended to the horizon line, defines
the measure point mp1 of the vanishing point vp1.

The purpose of this construction is that all lines

parallel to A'A define equal line segments in
both the ground line and in the vanishing line. Thus,
the lines aa' and bb', both parallel to AA', define a
line segment ab in the ground line that is exactly
equal in length to the line segment a'b' in physical
space.

Equivalently, the unit dimension BA in the ground

line is shifted into physical space as the measure
bar B'A', which represents dimensions along the
ground line as seen in perspective space at any
distance we require.

Because all parallel lines intersect the same

vanishing point (perspective rule 6), any pair of lines
through mp1 project a unit of measurement, defined
in the image plane, into perspective space along the
vanishing line VG. So we first find the measure point,
then draw a line from mp through the anchor
point A', align a measure bar with A', then rule off
the dimensions we require by connecting the
measure bar back to mp. Where these measure lines
cross the vanishing line, we have the measurements
in correct perspective along the vanishing line.

Locating the Measure Points. So how do we locate

the measure points? We cannot draw an arc
around G, as this is foreshortened into an ellipse on
the image plane. Instead, in the visual ray setup,
draw an arc from each vanishing point, from the
viewpoint to the horizon line (image plane) to find
the measure points, as shown below. This simply
defines the required isosceles triangle in the plan
view of the area between the viewpoint and the
image plane.

two point perspective: locating the

measure points

Thus, the arc from vp1 intersects the image plane

at B; the triangle VAB is an isosceles triangle in the
ground plan between the viewpoint and the image
plane (when folded along the horizon line into the
90° circle of view), and its base VB is the visual ray
that intersects the image plane at measure
point mp1. The arc from vp2 intersects the image
plane at Y; the triangle VXY is an isosceles triangle
and its base VY is the visual ray that intersects the
image plane at measure point mp2.

A little more care is necessary now when using a

measure bar than was needed in central perspective:
it matters which side of the figure the bar is placed
on, and which measure point you use, because we
have separate points for each side of the figure.

The guiding rule is that you are projecting

measurements onto vanishing lines, and the
vanishing lines recede to a vanishing point. So
you use the measure point defined by an arc
from the vanishing point. This is the measure
point opposite the vanishing point that controls the
vanishing line you want to measure.
two point perspective: projecting forward
or backward from a measure bar
The diagram (above) shows the four possible
combinations (in 2PP) for projecting from a measure
bar located on either side of the anchor point. In all
cases the correct measure point is the one that was
defined by an arc from the vanishing point that
controls the line that must be measured (as indicated
by the black dot). The measure point you use
is not determined by whether the measure bar is left
or right of the anchor point. Note also that a point
can be projected both backward and forward from a
measure bar; a visual ray from the dvp verifies
these forward projections.

As shown above, it is also always useful to identify

the diagonal vanishing point dvp (alias the principal
point in central perspective) when rotating the
vanishing points. This point can be used to project
unit dimensions in depth, for example to tesselate a
plane along the vanishing lines it bisects. The
location of the dvp also provides an informal
measure of the perspective emphasis in the image.
Perspective appears more dramatic if the dvp is
outside the 60° circle of view, and more classicizing
if the dvp is close to the direction of view.

If you are using a protractor centered on the

viewpoint, in order to define the rotation of the
vanishing points precisely, then the image plane
angle is the angle between the vanishing point and
a horizontal line drawn through the viewpoint
(labeled x, below).
 visual ray location of a measure point

any object side BC whose vanishing line intersects the

ground line at A defines the image plane angle x at the
ground line; then the vanishing point is at angle 90–
x from the direction of view and the corresponding
measure point is at angle x/2 from the direction of
view on the opposite side

Then the vanishing point for the object edge or side

(vp1) will be located 90-x degrees from the direction
of view (dv), and the measure point will be
located x/2 degrees from the dv, on the opposite
side. In the example, the front face of
object BC(shown in plan) is rotated 30° to the image
plane; therefore x is equal to 30°. This
means vp1 will be rotated 60° away from the dv on
one side, and its measure point will be
located mp1 or 15° to the right. These relationships
hold, no matter what orientation the object has to
the image plane.
For example, here is the geometry for central
perspective(diagram, right), in which a cube's faces
are either parallel or perpendicular to the image
plane. This means the angle of the faces at the
ground line is either x = 0° (there is no intercept
and therefore no angle) or x = 90° (edges or faces
are perpendicular to the image plane).

• For faces parallel to the image plane (x = 0°), the

angle of the vanishing point to the direction of view
is 90°–x = 90° (the vanishing point is in the image
plane, because the direction of view is perpendicular
to the image plane), and the measure point is x/2 =
0° (the measure point is 0° from the direction of
view, and is therefore the principal point).
geometry of vanishing and measure
points in central perspective
• For faces perpendicular to the image plane (x =
90°), the angle of the vanishing point to the
direction of view is 90°–x = 0° (the vanishing point
is the principal point), and the measure point is x/2
= 45° (the measure point is the diagonal vanishing
point).
The only finesse here is to distinguish between the
measure bar on the image plane and the measure
bar in perspective space.This is the difference
between a ruler we lay on the paper we draw on, and
the drawing we make of the ruler laid next to the
object in space. It is always more convenient to use
the ruler in perspective, because then we can align it
with the object by placing one end against the
anchor point. But this means we need to scale the
length of the ruler to match the scale of the object in
space. No problem: just apply the drawing scale of
the object to the ruler length, and use this scaled
length (measured from the image of the anchor point
on the paper) as your measure bar.

constructing a 2PP cube

Once you have located the two vanishing points and
two measure points, constructing a cube or
rectangular solid in 2PP follows the same steps as in
1PP, except that a measure bar is necessary to
define the two front sides.
 two point perspective: locating the anchor
line

As always, the first step is locating the anchor point

and its anchor line. This can be the back corner of a
room, the front corner of a building, or the nearest
tree on a leafy walk. This vertical defines the scale
and view of the whole drawing. If you're not sure
where to put it, work out the image composition with
a few perspective sketches.

Connect the top and bottom ends of the anchor line

to each of the vanishing points (blue lines). Define
the measure bars for the front sides of the primary
form, using the drawing scalemethods. The
measure bar will be equal to the vertical anchor line
if the form is cubic, and different if it is rectangular.
Remember, there is no perspective foreshortening of
the anchor line, so you can use it as your unit of
measurement by rotating it into the correct
orientation (as shown by the arc, below).

two point perspective: using the measure

bar

Now place one end of the measure bar at the anchor

point, and a draw line from the other end of the
measure bar to its corresponding measure point. The
measure bar is shown (diagram, above) in two
locations, to define the lengths on the two sets of
vanishing lines. In each case the line from the
measure bar is to the mp that was defined by the
controlling vanishing point.
 two point perspective: drawing the side
verticals

Construct the two side verticals, from the

intersection of the measure lines with the two bottom
vanishing lines to the vanishing lines above. The
vanishing lines define recession on each side of the
anchor line; the side edges of the primary form are
shorter than the front edge (anchor line).
 two point perspective: drawing the fourth
vertical

Draw vanishing lines from the ends of each side edge

to the vanishing point opposite it. These new lines
intersect at the top and bottom ends of the fourth
vertical edge. Construct it.
 two point perspective: the finished
drawing

Connect the ends of the verticals with horizontal

lines, showing only those lines that are visible from
the viewpoint. (The figure shows an open cube, to
reveal more of the perspective drawing.) When
finished, erase the vanishing lines, the vanishing and
measure points, and any other guide marks.

There are often many other forms in the drawing

besides the primary form. In that case, locate the
front and side verticals first. Locate any forms or
perspective details such as sidewalks, trees or
streets that are in front of or at either side of the
primary form. Finish these in outline, then go back
and finish the primary form. Last, fill in any
background forms or structures that are not occluded
by the objects already in view.

inclined lines & inclined planes

In most architectural and landscape applications of
perspective, inclined lines and planes are critical
components of the primary form. It is essential to
understand the geometry of inclined vanishing points
and vanishing lines, and to be able to construct them
in specific situations.

I use the 2PP framework as this is still the routine

basis for architectural purposes, but the underlying
geometry applies to 1PP and 3PP drawings as well.

For our purposes an inclined line is any line that is

not parallel or perpendicular to the ground
plane. The vanishing point of the line will
therefore not be in the horizon line, but displaced
some distance above or below it.

This vanishing point will be contained in a plane, and

this plane will have a vanishing line. There isn't a
unique vanishing line for a single inclined line,
because a line can be contained in an infinite number
of planes (all of them rotated around the line as an
axis). However two or more inclined lines that recede
to the same vanishing point do define a unique plane
that contains them both, and this plane has a unique
vanishing line.

Any inclined line defines a plan image in the ground

plane. The plane that contains both the inclined line
and its plan image will be perpendicular to the
ground plane, and its image line will be perpendicular
to the horizon line (perspective rule 16). In addition,
the plan line recedes to a vanishing point already
established for horizontals (either vp1 or vp1), and
this vanishing point will also lie on the vanishing line
for the plane that contains the inclined line and its
plan (perspective rule 14). Therefore the vanishing
line for the plane that contains the inclined vanishing
point (ivp) can be constructed as the vertical line
through a horizontal vanishing point (diagram
below).

This vertical plane is convenient because it

represents any vertical structure that contains or
defines the inclined line — the wall edge of a roof or
pediment, the side wall of a stairway, the
embankment or wall of a traffic ramp, a sloping
architrave above a vertical window, the slope of a
hillside, and so on.

Vanishing Point by Construction. The traditional

methods for locating the vanishing point for an
inclined line all construct the image of the inclined
line as the first step, then extend the image line to
the vanishing line of its plane. The methods only
vary in how the image line is constructed.

two point perspective: projected vanishing

points for inclined lines

The example (above) illustrates three different

methods:

• The upper end point of the inclined line (y), if it is

the peak of a gabled roof, is usually located on the
midline of the wall underneath; the midline passes
through the bisection point at the intersection of the
wall diagonals (x). Then the vertical distance ca is
projected to the horizontal vanishing point vp1from
a vertical measure bar to locate c; y is at the
intersection of the vertical midline with the vanishing
line from c. Then extend ay to the vertical vanishing
line (through vp1) to find the inclined vanishing
point ivp1.

• If the upper end point (y) of the inclined line

is not on the midline of the wall or vertical support
(as, for example, the side of a stairway), then it is
more convenient to project the horizontal (plan)
location of its upper end point (e) with
a horizontal measure bar to the appropriate measure
point, along with a second horizontal point (d, shown
in elevation) that locates the intersection of the
diagonal fy with the horizontal ab. Once these points
are located on ab, y is found at the intersection of
the vertical ex and the diagonal df, and the line ay is
constructed and extended as above.

• If the inclined line is extended, or lacks specific end

points, then it can be found from its pitch or slope
(as a slope horizontal/vertical ratio or an angle in
degrees) as follows. Draw an elevation (cross
section) of the slope as a right triangle, whose
altitude is equal to the image height of the lower end
of the inclined line above an anchor point in the
ground plane (e.g., the distance af). Align this plan
along the vertical af, and project its horizontal width
forward from the appropriate measure point to
intersect the plan vanishing line from the vanishing
point, at g. Then the line ga is the vanishing line for
the inclined line ay, and this also intersects the
vanishing point ivp1. The vertical that defines y must
still be defined using diagonal bisection or a
horizontal or vertical measure bar, as before.

In many cases, especially roof construction, what

goes up must come down. The vanishing
point ivp2 for the inclined line on the far side of the
roof will the same distance below the horizon line
that ivp1 is above it; this can be found by an arc
drawn from vp1 through ivp1 to the vanishing line
opposite. Or it can be located by extending the
line yb to the vertical vanishing line.

Inclined Lines Parallel to Image Plane. It can

happen that the inclined lines are parallel to the
image plane — for example, the front gables of row
houses viewed from the street.

In central perspective, the image of the front face of

a cube remains square no matter where it is locate
on the image plane; an inclined line drawn inside the
square (e.g., its diagonal) would therefore have a
constant angle on the image plane as well. So there
is no perspective adjustment for inclined lines
parallel to the image plane in central perspective.
 two point perspective: inclined lines
parallel to image plane

In 2PP an inclined line parallel to the image plane is

essentially an oblique diagonal inside a rectangular
solid that is oblique to the image plane; this solid is
constructed using the normal 2PP vanishing points,
and the inclined line then drawn within its envelope
(diagram, above).

It should be no surprise that, despite the changing

angle of view onto the rectangular armature and the
changing shape of its outline in 2PP projection, the
interior diagonal remains constant. We've defined it
as parallel to the image plane, and the shift
foreshortening of a two dimensional figure parallel
to the image plane does not alter the size or shape of
its perspective image.
 Vanishing Point by Rotation. The alternate,
analytical method starts by identifying the inclined
vanishing point and constructs all lines from there.
This is done by creating an auxiliary viewpoint, a
method that is explained in detail on the next page.

two point perspective: exact vanishing

points for inclined lines

The rotation is performed in the following steps:

1. Identify the vanishing line of the plane that

contains the inclined line(s). We have already
established that this is the vertical vanishing line
through vp1, the horizontal vanishing point.
2. Draw a line parallel to this vanishing line, through
the principal point. This line already exists, as the
median line.

3. Identify the intersection of this line with the 90°

circle of view (point x).

4. From vp1, draw an arc through x, the vertical

vanishing line, and the horizon line outside the 90°
circle of view. The intersection of this arc with the
horizon line identifies a new point A, the auxiliary
viewpoint.

5. Rotate around A the visual rays that define the

slope of the desired inclined lines. In the example, a
slope of 45° is demonstrated. The intersection of
these rotated visual rays with the vertical vanishing
line defines the two inclined vanishing
points, ivp1 and ivp2.

Vanishing lines from these vanishing points through

the vertical end points b and c intersect at a, the
peak of the roof; the vertical of the peak does not
have to be located.

These vanishing points control all parallel inclined

lines. Thus the edges of the roof at the opposite end
of the structure recede to the matching vanishing
points for the front edges: no to the same vanishing
point as ab, and np to the same vanishing point
as ac. In other words, triangle nop is parallel to
triangle abc, and all parallel planes converge to the
same vanishing line (perspective rule 14).

Vanishing Line for Inclined Plane. Two parallel or

intersecting lines are necessary to define the surface
of a plane (perspective rule 10). In most
architectural problems, the second line is a horizontal
or vertical element joining the inclined line. In the
example (diagram, below), the second line is
provided by the horizontals an and bo in the
trapezoid abno, which intersect the parallel inclined
lines ab and no.

two point perspective: vanishing line for

inclined plane
These horizontal lines are controlled by the vanishing
point vp2. Since the vanishing line of a plane
contains the vanishing points of all lines in the plane
(perspective rule 14), the vanishing line of
plane abno is necessarily the line
containing ivp1 and vp2; these two points define the
vanishing line (perspective rule 3). Similarly, the
vanishing line of the plane acnp is the line
containing ivp2 and vp2.

The vanishing point vp2 is, of course, the vanishing

point for all lines in the plane abno that are
horizontal (parallel to the ground plane).

two point perspective: vanishing line for

inclined plane
Lines to vp2 are gradient lines that indicate all points
on the plane at an equal height above or below the
ground plane. Lines to ivp1 are fall
lines perpendicular to the gradient lines that
represent the lines of steepest descent or the
direction of gravitational pull across the plane — the
direction in which water would flow or round objects
would roll down a perfectly smooth, flat surface.

All 2PP lines perpendicular to any inclined plane will

be perpendicular to the vanishing line of the plane, in
the same way that 1PP or 2PP verticals (lines
perpendicular to the ground plane) are perpendicular
to the horizon line.

Finally, regardless of its orientation to the viewer,

the vanishing line for a plane is always a line
(perspective rule 9), not a curve as maintained by
the theorists of curvilinear perspective.

Note the similarities and differences between the

location of vanishing lines for planes that are
slanting, sloping or inclined:

• The vanishing lines for slanting planes pass

through the principal point, but are not parallel to the
horizon line; they correspond to roll in airplane
terminology, and are rotated by turning the horizon
and median lines, with their diagonal vanishing
points, around the principal point.
• The vanishing lines for sloping planes are parallel
to the horizon line but do not pass through the
principal point; they correspond to pitch in airplane
terminology, and are located by rotating a visual ray
from the left or right diagonal vanishing points.

• The inclined planes discussed here may combine

roll, pitch and yaw or a left/right turning. These
planes are found first by establishing the 2PP
vanishing points; one vp is the horizontal vanishing
point for the gradient of the plane, and the other is
the vp for the vertical vanishing line that contains
the vanishing point for the fall line (the inclined line)
in the plane.

The point is to understand the differences among the

various oblique plane representations, and how to
construct them in a perspective drawing.

distance point projection

Using the 90° circle of view to rotate vanishing points
and locate measure points, either by construction or
using the trigonometric ratios, is the most accurate
method of creating a 2PP perspective space.
However it requires a large working area, and can be
cumbersome when objects that have a unique
orientation to the image plane require separate
vanishing point systems to be constructed.
Fortunately, the distance point projection method,
devised in the 16th century, permits the construction
of any 2PP object in any orientation using only the
principal point (pp), a diagonal vanishing point
(dvp), and a elevation and plan of the object. The
image object itself is then used to find the vanishing
points and measure points, and these can be used to
construct incidental objects in the same recession.
As an example I will use a pentagon (plan, right),
which actually requires five vanishing points. Note
that this is an irregular pentagon (the sides are not
of equal length) and none of the sides is
perpendicular or horizontal.

The plan is decomposed into two perpendicular

measure bars, Pand D, each bar locating a defining
corner of the form. The Pbar represents the view of
the pentagon perpendicular to the direction of view
(parallel to the image plane), and the D bar
represents the view in depth, parallel to the direction
of view. plan view of a pentagon

Divided Distance Scaling. The first steps, of

course, are to locate the object on the image plane
and determine its image scale. Let's arbitrarily decide
that this pentagon will be 3 meters wide, at an object
distance of 11.5 meters from its front corner (x).

We can find the location of the anchor point using

the methods described under scaling the drawing,
but let me show you the traditional approach. It is
based on the divided distance principle:

The point intersection of a principal point orthogonal

and a diagonal vanishing line from a unit
dimension X is also the intersection of the principal
point orthogonal and a vanishing line drawn from the
1/X unit dimension to a 1/X dvp vanishing point.

This principle is again based on the triangular

proportions and is obvious in a diagram.
 divided distance method of projecting a
unit dimension

Thus, one end of the unit dimension d in the ground

line, in this case 1.5 meters long, is placed against
the median line; this uses the median line as the
principal point orthogonal. Then the diagonal
vanishing line is constructed from the other end of
the unit dimension. The point where it intersects the
median line (d') is the image location of 1.5 meters
in perspective depth (as measured from the ground
line, not the viewpoint or station point). But, as
shown, the same point is identified by a line drawn
from 1/4 the unit dimension length to a vanishing
point that is 1/4 the viewing distance from the
principal point.

Once we have established the anchor point, the

constancy of shift foreshortening allows us to move it
left or right along its transversal, and it will still be
located 10 meters from the ground line, as measured
perpendicular to the ground line (e.g., not diagonally
from the median line).
 locating an anchor point at 10 meters

The benefit of this method is not that we must work

with smaller and smaller unit dimensions in the
ground line, but that we can project larger distances
into perspective space. In the example below, we
identify the location of the anchor point at 11.5
meters object distance by taking the full width of the
circle of view (3 meters) as the unit dimension. Then
we subtract the viewing distance, to get the ground
line distance: 10 meters — from the ground line, not
from the viewpoint or station point. The 3 meter unit
dimension is 30% of this 10 meter distance, so we
draw the principal point orthogonal from one end of
the unit dimension, and the diagonal vanishing line
from the other, but to a vanishing point that is 30%
of the distance from the principal point to the
diagonal vanishing point. This identifies the 10 meter
perspective depth from the image plane.

If we require even larger distances than that, a

second trick is to shift the ground line unit dimension
upwards, along the median line, by whatever
multiple of its dimension we require. Thus, shifting
the line upwards by half the circle of view radius (as
shown in the diagram, above) makes it an image line
effectively twice as long in perspective space, so that
the same vanishing line construction defines a
transversal at twice the depth (20 meters). Shifting it
up to 1/4 the radius creates an image unit dimension
of 12 meters, which when regressed on the
1/4 dvpdefines a perspective depth of 40 meters.

Alternately, you can just divide the distance between

the transversal and the horizon line by whatever
fractional depth you require, as explained in the
section on perspective gradients.

Distance Point Projection. Next we want to project

on the image plane the pentagon plan. First we
rescale the plan so that its width dimension (P) on
the image plane is equal to (formula 4) 3/11.5
meters or 78 cm wide, which is about 50% of the
radius of the 90° circle of view. Then we place the
front corner x at the anchor point. (Note that we
have to invert the plan when we do this, so that x is
at the top, otherwise the figure will appear in
perspective space as the mirror image of the plan.)

Next we want to locate the corner points of the plan

in perspective space. This is done by relying on
the distance point principle:

Any point is located in perspective space on its

anchor line orthogonal, at a perspective distance
equal to the plan distance from the anchor line.

That is, we first construct orthogonals for all points,

then measure the plan distance of each point from
the anchor line, then project this distance into
perspective depth using a diagonal vanishing line.
 projecting the plan: plan orthogonals

shown within the 60° circle of view; note that the plan
must be reversed left to right, as in a mirror, to project
the correct image

The first step in the distance point method is to carry

all the important object points up to the anchor line
with perpendicular lines. Thus, a defines the (new)
point x and b defines point y. Then the plan
orthogonals are constructed from these points back
to the principal point, as shown above (blue lines).
 projecting the plan: plan diagonal
vanishing lines

shown within the 60° circle of view

The next step is, for each point, to carry its plan
distance from the anchor line back to the anchor
line with by an arc centered on its orthogonal
intersection (pink lines, above), the method
of rabattement or plan rotation first attested in the
17th century.

Thus the distance ax is transferred to the anchor line

by an arc centered on x, defining the point a' on the
anchor line and the anchor line
distance a'x (where a'x = ax). Similarly the
distance by is transferred to the anchor line by an
arc centered on y, defining the point b' and the
anchor line distance b'y(where b'y = by); and so on
for all the other points.

Finally, we project all these distances into

perspective space, using vanishing lines drawn from
these new anchor line points to the diagonal
vanishing point on the opposite side (that is, each
diagonal vanishing lines must cross its corresponding
orthogonal). This is typically done by using one
diagonal vanishing point and carrying all the anchor
line arcs in the opposite direction.

Each intersection between an orthogonal and its

corresponding diagonal vanishing line locates the
plan point in perspective space. Thus, the
intersection of orthogonal y and diagonal vanishing
line b' locates the point b in perspective space.

All that remains is to connect the points in the image

plane. This constructs the perspective image of the
pentagon plan at an 11.5 meter distance — all
without constructing any additional vanishing points,
or using a work surface larger than the 90° circle of
view radius.
 reconstructed vanishing points

In the academic tradition, the perspective plan is

used to locate the actual vanishing points for the
plan, by extending the line segments of each side
back to the horizon line. I've done this for the three
vanishing points within the 90° circle of view and, as
a check, identified the same points by projecting the
plan sides from the viewpoint. As you see, the
correspondence is not exact. This is largely due to
the fact that the manipulations were made within a
small digital image and because the plan is
extensively flattened by foreshortening. But as a rule
the viewpoint rotation method is always more
accurate, especially when the projected plan is deep
in perspective space.

This may be a minor concern, since once the

vanishing points are established on the horizon line,
by whatever method we choose, then the points are
equally accurate for constructing any new lines —
details such as doors and windows in the primary
form, or new forms within the same recession
system. For objects that are defined by many
vanishing points, or that are relatively close to the
image plane, or that are defined by vanishing points
outside the circle of view, the distance point method
is reliable and remarkably accurate.

the ground line framework

It will be useful to conclude the discussion of 2PP
with a brief look at the perspective setup,
standardized in the 19th century, that is most
commonly taught in perspective tutorials:
the ground line framework. For a fuller
introduction and detailed instructions, see
the references by Robert W. Gill or Michael E.
Helms.
The ground line framework is economically defined
by the principal point, the station point (distance
point), and three horizontal lines (diagram, below).

the ground line framework

The utility of this framework is that the important

attributes of the image can be adjusted directly, by
the relative location of these two elements:

• The location of the ground line defines the level,

adjacent to the primary form, at which the viewer is
standing; it also defines the image plane on which all
object dimensions are measured.

• The horizon line defines the viewing height as a

proportion of the height of the primary form; it also
defines the angle of view in relation to the true
horizon.
Ground line drawings usually start with a blueprint or
set of plans for the primary form (diagram, right).
The plan image is critical for scaling the width and
depth of the perspective image. The elevations are
critical for exterior details and contrasting surface
features of the front and side faces.

architectural blueprints

deriving measurements in the ground line

method

The first step is the projection of vanishing points

and primary form measurements onto the horizontal
and vertical measurement lines (diagram, above).

The station point is established in relation to a plan

of the primary form, using the plan scale to
determine the object distance. For example, if the
scale of the plan is 1/4" = 1', and the form will be
viewed from 50 feet, then the station point is located
12.5" from the near edge of the plan outline.

This object distance and orientation is usually judged

by drafting a large isosceles triangle with an internal
apex angle equal to the desired circle of view; the
object is then moved toward the apex of this triangle
until it just touches both sides, while turning the
object so that the plan orientation gives the desired
direction of view.

The horizontal measurement line is drawn

perpendicular to the line of sight (center of the cone
of vision), typically so that the measurement line
passes through a front, prominent feature of the
plan. Divergent sightlines are drawn from the station
point through the principal features of the plan, and
their intersections with the horizontal measurement
line are marked and labeled, along with the location
of the 2PP horizontal vanishing points and the line of
sight (the "sight point").

The corresponding vertical measurement line is

projected by parallel lines from the elevations of the
primary form, including its intersection with the
ground line.
 deriving measurements in the ground line
method

Next, the horizon line and ground line are located in

the image area to produce the desired image
composition, specifically the height of the viewpoint
to the primary form, its distance from the viewer,
and the angle of view in relation to the ground plane
(diagram, above).

The horizontal measure line is located below the

image area, parallel to the horizon and ground lines,
with the sight point vertically aligned with the center
of the image area. Vertical lines are used to project
the horizontal plan and vanishing point
measurements onto the ground line.

The vertical measurement line is located on the

ground line. Its measure points were not projected
from the station point, therefore they must be
projected from the horizontal vanishing points
forward or backward from the ground line, and these
projected measurements carried by horizontal lines
into the image of the primary form. Note that the
vertical measurement line can be freely shifted along
the ground line, if convenient; the horizontal
measurement line, because it was scaled to a specific
location of the station point in relation to the primary
form, must be exactly aligned with the sight point
and cannot be moved.

There are many interesting methods of scaling and

projection with the ground line method that I omit in
this brief outline. To conclude, I mention four
principal shortcomings to the ground line method:

• procedural geometry: The ground line framework

is basically a scaffold built expressly to deploy
specific procedures for the construction of an image;
it has a weak connection to perceptual geometry. In
particular, it is more difficult to deduce the
construction principles for a new geometry, because
important basic concepts (such as measure
points or vanishing point rotation) are typically
excluded. The better ground line tutorials discuss
what to do if you discover that the perspective view
is not optimally oriented, or is incorrectly scaled,
after the drawing has been started; those kinds of
problems result from the narrowly focused,
piecemeal nature of the ground line method.

• errorful, inflexible scaling method: The ground

line method of scaling an image, which requires the
plan be oriented to scale and then projected by
diverging lines onto a horizontal measure line, is only
practicable for large objects seen from a relatively
close vantage point; otherwise the method becomes
inaccurate. It also requires greater precision, as
drawing errors are magnified by the expanding lines.
If the decision is made to change the angle of view
on the primary form, the entire scaling operation
must be repeated. In the 90° circle of view
framework, scaling to measure points compresses
measurement errors, and a new rotation of the
vanishing points allows the same measure bar to be
used from different perspective views.

• complex line construction: All the ground line

demonstrations that I have seen require a large
number of construction lines — at a minimum, the
projection onto the measurement lines, the
projection of the measurements across the image
area, and the joining of vanishing lines.

• adapted to drafting tools: The ground line

framework is adapted in various ways to technical
drafting tools, and when used with those tools it is
very efficient. But this means the framework is less
effective as a general, artist oriented perspective
model.

For these reasons, and in particular when linear

perspective is taught to photographers and painters,
I strongly prefer 90° circle of view method.

who has a 12 foot table?

Unfortunately it is fairly common to start with the
primary form in an orientation that puts the
two vp's inconveniently far apart. In the previous
cube construction example, assuming a 10 foot circle
of view, the cube is oriented so that the
two vp's would about 11 feet apart — one 3.2 feet to
the left of the dv, and the other 7.7 feet to the right.
This isn't very convenient for a drafting table.

If you have a 12 foot table, push pins and lots of

string (or the specialized drafting equipiment that
rescales vanishing point locations within a small work
area), you can work out the geometry of a cube at
any size, no problem. If you're lacking the table, you
can lay the support on any large bare surface, for
example a clean kitchen floor or concrete patio, and
work there — using tape instead of pins to hold the
string.

If those alternatives don't appeal to you, then you

can rescale the drawing. The basic geometry of
the vp's works exactly the same no matter how big
or small the circle of view is assumed to be. So just
get a large sheet of paper, draw the 90° circle of
view to a conveniently small size (20cm works well),
work out the vp's and perspective drawing in that
format, make a careful outline drawing in
perspective, then transfer the drawing to the
painting support, enlarging it as you make the
transfer. You can control the enlargement
by squaring the diagram or by using a surface
projector, adjusted so that the size of the image
matches the length and location of a reference
vertical (front vertical edge) marked in the right
place on the support.

When one or both vp's are really far from the

drawing surface, it's possible to calculate the relative
sizes of edges and angles in a drawing without ever
anchoring the vp's with string or ruler: you just need
to work out a few key measurements on a calculator,
and you need to know the exact distance of the two
vanishing points from the principal point (dv), which
is found either with a careful rotation around the
viewpoint, or by multiplying the radius of the 90°
circle of view by the tangent of the vanishing point
angle to the direction of view.
 method for scaling new lines without
vanishing points

The diagram shows all the points required for these

calculations. The procedure is simple once you
understand it: go through the instructions slowly and
carefully, and you should have no trouble.
(Caution: use a metric or engineering ruler for
these tasks.)

There are two situations: the anchor line is either

entirely above (or below) the horizon line, or it
straddles the horizon line.

Start with the straddled line (right side of the

diagram). The key measurements you must know in
advance are: (1) the length from dv (the direction of
view) to c (the vanishing point), (2) the length of
the anchor line above (A1 to A2) and below
(A2to A3) the horizon line, and (3) the distance from
the direction of view to the anchor line (d to a).
Once again, the triangular proportions provides
the frame of reference. First, by subtracting the
distance dv-a from the distance dv-c, you determine
the distance ac from the anchor line to the vanishing
point. (If the anchor line is on the opposite
side of dv from the vanishing point, you add the two
distances to get ac.) Now you want to insert a new
line N1 to N3 which has the same height in depth as
the anchor line.

First, determine how far to the left or right of the

anchor line the new line should be placed: this
defines ab, which gives bc when subtracted from ac.
Then the ratio bc/ac tells you the length of the new
line N2-N3 in relation to the length of line A2-A3, and
the length of the new line N1-N2 in relation to the
length of line A1-A2. For example, if bc/ac equals
0.80, and the upper part of the anchor line A1-A2 is
2cm long, then the upper part of the new line N1-
N2 will be 0.80 * 2.0 = 1.6cm long. Repeat for the
line segment below the horizon line (N2-N3).

If the new line is closer to the direction of

view d than the anchor line, or on the opposite side
of d from the anchor line, then you
would add ba to ac. In that case the ratio bc/ac will
be greater than 1.0, and the new line will be
correspondingly larger.

If the line is entirely above (or below) the horizon

line (left side of the diagram), then the
ratio bc/ac is applied to the length A1-A3 to get the
top end of the new line, and to the length A2-A3 to
get the bottom.

How do you define the crucial distance dc (from the

direction of view to a vanishing point) in the first
place? The easiest method is to use my vanishing
point calculator to get the measurements of
the vp's and mp's, and adjust the viewing distance
to the object and your angle of view until you get the
proportions that seem desirable. Or, as described
above, you can reduce the circle of view to a
workable size, use the method for rotating the
vanishing points to determine the locations
of vp1 and vp2, measure the distance from these
to dv on the diagram, then scale those distances
back to life size.

Unfortunately this method, even after you get the

hang of it, still forces you into a lot of poking of a
pocket calculator, and is hopelessly tedious and
prone to error if many lines must be inserted in your
drawing. The ultimate solution is to generate
a recession grid for the distant vanishing point, and
use this grid to determine the perspective reduction
for any verticals in the drawing.
 using a recession grid for distant
vanishing points

First work out the angles and distances of your point

of view within a reduced (20cm) circle of view drawn
on a large sheet of paper. Carefully measure with a
metric ruler the distances from dv to the two
vanishing points vp, the diagonal vanishing
point dvp and the two measure points mp, then
multiply these by 15 to get them in the same scale
as the 3 meter circle of view (the scale of the
perspective drawing). Locate the horizon line, dvand
the two measure points on your support.

Now draw a vertical line on the lefthand side of the

drawing, anywhere that is convenient — the line
should be farther to the left than any major form in
the drawing area. Mark off increments on this line
above and below the horizon line using any
convenient interval of measurement.

Next, draw a similar line on the righthand side of the

drawing, again putting it far enough to the right so
that it won't obstruct any major forms in the
drawing.

Now you want to find a reduced scale of

measurement for this righthand line to represent
perspective recession from the lefthand line toward
the vanishing point. You already know how to do
this: treat the lefthand line as the anchor line, figure
out the distance from this line to the vanishing point
(ac), then the distance from the righthand line to the
vanishing point (bc), and finally the ratio bc/ac.
This is the reduction in the scale of measurement
required for the righthand line. For example, if the
intervals on the lefthand line are in inches, and the
ratio bc/acis 0.80, then the intervals on the
righthand line are in 0.80 inches.

When you have intervals marked on both vertical

lines, connect the corresponding points to make a
recession grid of converging lines (parallel lines in
perspective). These lines show you the slope of any
horizontal converging to the distant vanishing point.
You can either draw the horizontals along an
inscribed recession line (as in the base of the
building in the diagram), or draw horizontals
between and roughly parallel to any two lines (as in
the top of the tower of the building in the diagram).
This grid is especially convenient if you must work
out the perspective recession for many repetitive or
similar lines, for example the windows, columns,
cornices and ledges on the front of a building.

vp spacing from an object drawing

Why not just say ... heck with it, I'll just draw the
cube at whatever size fits the drawing, at whatever
angles look good to me, and let the vanishing points
fall where they may?

You can do this, especially in a freehand drawing of

the object from life. In that situation the principles of
linear perspective guide you to look at the edges and
faces and proportional sizes of the parts, and to draw
these elements more accurately in relation to their
fixed vanishing points. This "imagined" perspective
context is useful because you can introduce
expressive distortions to the perspective facts,
controlling by eye how obvious or subtle they
appear.

However, if you are drawing an imaginary or

remembered form from scratch, such as that cute
little cottage you saw on yesterday's hike, then your
placement of the vanishing points can go badly
astray without the visual example in front of you.
And once you have drawn the primary form, you still
have to draw everything else to match its vanishing
points, direction of view and horizon line.

The most common drawing fault is placing the

vanishing points too close together. The informal
recommendation is simply to put the vanishing points
very far apart ... no, farther than that ... keep going
... — with the idea that inaccuracies in widely spaced
vanishing points are harder to see.

Let's start with a perspective constant: the distance

between 2PP vanishing points depends on the
viewing distance to the object. The closer an
object is to our view, the closer together its two
perspective points will be in relation to the object
size. This has a very powerful effect on the
perspective view, as is apparent in these four cubes
of exactly the same vertical size drawn as they would
appear at increasing viewing distances.

2PP cube seen from four different

distances

a cubic box 2 meters high seen from (left to right) 3

meters, 6 meters, 12 meters and 24 meters

The shape of the cube alone tells us a lot about its

distance from us. The flattening in the "far" cube (24
meters, at right) is what we'd expect to see in
binoculars or a telescopic lens, while the bulging in
the "near" cube (3 meters, at left) mimics a wide
angle lens. This "near" cube resembles many badly
done perspective drawings, because the cube is too
large relative to the vp's.

So the perspective problem is to find a vanishing

point separation that matches the apparent
distance to the object we want to represent. And
this is a problem that the circle of view
framework is designed to solve. Fortunately, if we
start with an acceptable 2PP drawing of the front
sides of the primary form, we can reconstruct the
90° circle of view from the object drawing using
the semicircle of Thales construction. The circle of
view then can be used to locate the vanishing points.
 using a semicircle of Thales to find the 90°
circle of view

Let's take as our starting point a rough, freehand

drawing of a cubic box 2 meters high viewed from
about 6 meters. We made this drawing on the back
of an envelope in the field, and now we want to build
a more finished drawing upon it.

First, extend the front edges of the primary form on

either side until they meet in two vanishing
points, vp1 and vp2. Connect these points with a
straight line, which is the vanishing line for the
primary form; if the form is level and upright to the
ground, such as a building, then this is also
the horizon line.

This is the point to make any esthetic corrections.

For example, if this is the horizon line, it should be
level. If it is not, redraw it level and relocate the
vanishing points on it by moving them vertically up
or down. Then redraw the vanishing lines from these
points back to the object drawing.

Next, find the midpoint M on the horizon line

between the two vanishing points, using a ruler or
the line bisection method. Then draw a semicircle
around the midpoint M from one vanishing point to
the other. This is the semicircle of Thales.

The useful geometrical fact is that the 90° angle of a

right triangle must lie on a semicircle, if the diameter
of the semicircle is also the hypoteneuse of the right
triangle. This right angle corner is of course the
viewpoint that we use to rotate the 2PP vanishing
points. This viewpoint must lie on the semicircle of
Thales.

But where? To find it, we have to locate the direction

of view. This is a somewhat arbitrary decision, but
usually the dv is located on the horizon line
somewhere around the front edge or center of the
form. From the dv, extend a line perpendicular to the
horizon line up to the semicircle of Thales, which
locates the folded viewpoint. The distance
from dv to the rotated viewpoint is the radius of the
90° circle of view.

I claimed that the cube in this example was cube 2

meters high viewed from 6 meters. Let's check. The
direction of view (dv) is located about 3/4ths up the
front edge, so the viewing height is about 1.5 meters
above the ground. As the horizon line is always at
the same level as the viewpoint, this corresponds
to our standing height on level ground when viewing
the cube.

By definition, this 1.5 meters is also the radius of the

90° circle of view, and is also the implied viewing
distance to the finished image.

In the drawing, the vertical of the cube is 16% of the

diameter of the circle of view, or 48cm; this is
the drawing size. So we have the viewing distance
(150cm), drawing size (48cm) and object size
(200cm). With formula 3, we find that the object
distance from the viewpoint must be 3.2 times the
object size, or 6.3 meters.

Thus, from a rough but accurate perspective

drawing, we have reconstructed the location of the
vanishing points and the circle of view. We now have
the framework to insert accurately details of the
primary object, and to add objects around it in the
same perspective space. The spacing of the
vanishing points in relation to the drawing size is not
merely "good enough," but represents the spatial
relationships we intend to portray.

where is the center of projection?

The methods just described can also be used to
locate all the perspective elements implied by a
finished painting. This is a problem of more interest
to art historians than to artists, but I will describe
the methods here for both 1PP and 2PP paintings.

Here again are the perspective elements that we

need to identify in the approximate order we locate
them:

• Median Line. This is parallel to the side edges of a

rectangular image format, or perpendicular to the
floor when the painting is correctly hung. The median
line is nearly always through the center of the image
format.

• Horizon Line. This is usually parallel to the bottom

or top edge of a rectangular or square image format,
or parallel to the floor when the painting is correctly
hung, at the height of the eyes of standing
figures (for a standing artist and viewer), or in
similar proportion to windows, tables, walls between
floors and ceilings, and so forth. The horizon line is
rarely through the center of the image format.
• Direction of View. This is at the intersection of
the horizon and median lines.

• Distance Points. In 1PP perspective these can be

found as the diagonals of any square element
receding to the dv, commonly the floor tiles of
Renaissance paintings or frescos.

• Vanishing Points. In a 2PP painting these are

located from the edges of any suitable rectalinear
(right angled) object in the image, of convenient
clarity and size.

• Circle of View. The radius of the circle of view is

determined by the distance points in central
perspective, or by the method of the semicircle of
Thales in two point perspective.

• Center of Projection. The implied correct

perspective location for viewing the painting (the
perspective center of projection) is at a distance
equal to the radius of the circle of view along a line
perpendicular to the direction of view dv.

1PP Construction. A straightforward example is

provided by Raphael's Philosophy, his first fresco
decoration on a large wall of the Vatican chambers.
The image I can provide is drastically small (the
original is a 27 feet wide), but a large format
reproduction is available in most Renaissance art
textbooks.
 1PP reconstruction of the center of
projection

By examination we conclude that the fresco is done

in central perspective, which means we are looking
for the direction of view (dv) and the distance points
or diagonal vanishing points (dvp's). These give us
the center and radius of the circle of view and the
implied center of projection.

The orthogonals necessary to find dv are found in

the receding barrel vaulted passageway at the center
of the image. I chose the square edged capitals of
the columns along both sides, which define two
orthogonals (red lines) intersecting between the
figures of Plato and Aristotle.

Since the composition is in central perspective I

know the horizon line is level and the median line
perpendicular to it, so I go ahead and draw these in
(blue lines) from the dv, extending the horizon line
far off the picture to one side.
The obvious place to prospect for diagonals is in the
tiled floor at the foot of the fresco. However, the
visual angle on these floor tiles is rather small,
making them hard to see clearly; also, I may be
wrong to think they are true squares. So I take a
second diagonal from the diagonal corners of the
capital of the front rectilinear column, which
conventionally would be square in classical
architecture. These diagonals (orange) intersect the
horizon line in close agreement (thank you,
Raphael!), so I conclude I really have found the
diagonal vanishing point.

A circle with center at dv through the dvp defines

the 90° circle of view, so the distance
between dv and dvp is also the viewing distance to
the painting. The width of the painting is 27 feet, so
by calculation the center of projection should be
about 31.5 feet directly in front of dv. I haven't been
to the Vatican chamber where this fresco is located,
but the photographs I have examined suggest the
center of projection is not in a practicable viewing
position. (It is approximately 10 feet above the floor
and several feet on the other side of the opposite
wall!)

The horizon line also lets me locate the implicit

location of the painter in terms of the room
represented in the fresco, which is roughly at the
height of the white robed figure to the right of
the dv (if the artist is assumed to be standing), or at
the height of the central figures at the top level (if
the artist is assumed to be sitting).

From the dvp I can also determine the circle of view

of the painting. Drawing a straight line from dvp to
the top of the fresco defines a 20° angle. So the
barrel vault boundary of the fresco represents a 40°
circle of view.

2PP Construction. In the case of two point

perspective, the necessary elements are the same,
except that you must start by finding a rectangular
form or forms that will reveal at least twovanishing
lines for each of the two vanishing points. This object
does not have to be square, but it must contain a
right angle at the intersection of two sides.

These vanishing points in turn determine all the

other perspective elements. The intersection of each
pair of lines defines a vanishing point, and the two
vanishing points define the horizon line. Orthogonals,
if visible, will point to the direction of view on the
horizon line; if there are no orthogonals, then the
median line of the format can be used to locate
the dv.

The final step is to reconstruct the right

triangle whose hypotenuse is the horizon line
between the two vp's and whose right angle lies on
the median line drawn from dv perpendicular to the
horizon line. This is easiest to find simply by
dragging the right angle of a drafting triangle, or the
corner of a large sheet of paper, up the median line
until both sides can be aligned over the two
vanishing points: the right angle corner is then on
the folded vanishing point on the circle of view. Or,
we can use the semicircle of Thales method.

2PP reconstruction of the center of

projection

In this painting by Edward Hopper, I ignore the

suggestion of 3PP in the slightly upward flaring sides
of the house and tilt of the telephone pole (we'll
come back to this later).

The diagonals from the eaves and base trim of the

windows (orange) are a little sloppy, but my best
guess puts their intersections (and the horizon line)
at the bottom of the picture. (This construction
fundamentally determines everything else, so it
should be done as carefully as possible, using as
many vanishing lines as you can find.)

The median line and direction of view (dv) are

arbitrarily located on the midline of the painting.

To find the folded viewpoint on the circle of view,

move the right angle corner of a drafting triangle up
the median line until the two edges lie on both
vanishing points: the right corner is then on the
circle of view; or use the semicircle of Thales method
by bisecting the distance between the two vanishing
points.

We discover several things from this construction:

• Hopper's eye level (the horizon line) is level with

the sidewalk in front of the house, which implies that
the house is at the top of a hill and the artist was
downhill from the house when painting — how far
downhill depends on whether he was seated or
standing.

• The radius of the circle of view is approximately 1.8

times the width of this 50cm wide painting, which
locates the center of projection (viewing distance)
about 35 inches from the painting surface; for best
effect the painting should also be hung slightly high,
with the bottom edge at eye level of an average
sized viewer.
• By taking the largest circle around dv to the edge
of the image, then measuring the angle at the folded
viewpoint (25°) defined by the radius of this circle,
we determine that the image is enclosed by a 50°
circle of view, which creates the slightly bulging
appearance of the front angle of the building.

• However, the outward flaring in the sides of the

building is directly contrary to the perspective
geometry: viewed from below, a building's sides (and
edges parallel to them, such as the telephone pole)
should seem to converge toward a third vanishing
point far above the horizon line, not below it. The
fact that they flare outward as the rise above the
viewer is a perspective distortion explicitly
introduced for its dramatic effect, as it gives the old
pile a characterful dynamism.

The Renaissance artist and his modernist colleague

have kept their art well within the commonly
recommended 60° circle of view but have allowed
some perspective distortion for dramatic or esthetic
impact. Creatively "adjusting" perspective distortion
has been one of the subtleties of painting for almost
six centuries.

3PP Construction. Finally, it is even possible to

identify the center of projection and circle of view in
a 3PP perspective drawing, provided that all three
vanishing points can be established from edges or
lines within the image. The method for doing this is
complex, but is explained on the next page as
the perspective sketch method of 3PP
construction.

N E X T : Three Point Perspective

Last revised 07.I.2015 • © 2015 Bruce MacEvoy

three point perspective technique

At this point it's customary to explore the capabilities
of 2PP in a variety of specific drawing problems. I
want to keep the momentum and look at three
point perspective, which allows you to construct a
form in any orientation (from any viewpoint).

Three point perspective is often illustrated with aerial

views of Manhattan, looking down on a skyline
bristling with skyscrapers. But artists will find 3PP
equally useful in still life or figure paintings — where
the view downward onto a table of objects or a piece three point perspective
of furniture can be just as steep — and in landscape
views up toward soaring cliffs or a stand of tall trees. the perspective sketch
construction method
The 3PP perspective problems and construction
methods are complex, and it may seem we lose more constructing a 3PP cube (sketch
in clarity than we gain in drawing power. Many method)
artists have come to the same conclusion, and avoid
3PP for simpler approaches, including freehand the horizon line construction
modification of drawings blocked out in 2PP, or the method
expedient of tracing photos.
constructing a 3PP drawing
I won't disagree with those solutions; they can be (horizon line method)
convenient and effective. They fall short, however, if
you must add new forms around the primary form —
for example, if you have traced the photograph of an
existing building, and want to insert new or different
buildings around it — or if you want to show the
building from a different point of view, or require
more precision than freehand perspective can
provide. For these common situations, 3PP is
invaluable.

three point perspective

As we add vanishing points, we remove aspects of
perspective that we can take for granted. In 1PP or
central perspective, the relationship of the vanishing
points and horizon line to the direction of view are
taken for granted. In 3PP both the vanishing point
locations and the relationship between in the
direction of view and the ground plane (horizon line)
must be specified.

Defining Features of Three Point Perspective. The

diagram shows the simplest 3PP situation: a cube
centered in view but first rotated 45° to one side and
then downward until all front faces appear of equal
size. In all three point perspective views there
are no faces or edges parallel with the picture plane.

In particular, because the direction of view is still

assumed to be perpendicular to the image plane, the
direction of view is no longer parallel to the
ground plane when the primary forms are
constructed as buildings are, with walls perpendicular
to the ground.

The canonical view places the three front edges of

the cube in a 54.7° angle to the direction of view, so
that all three vanishing points are outside the circle
of view. The planes of the three front faces are at a
35.3° angle to the direction of view, with vanishing
lines defined by the triangle of three vanishing
points.
 three point perspective: the basic
geometry

The three vanishing points (vp1, vp2 and vp3)

control the recession of all lines parallel to the edges
of the cube. This means the outline of each face is
determined by two vanishing points, rather than one
as in 2PP.

Connecting the vanishing points are three vanishing

lines, which control recession of all planes parallel
with each front and matching back face of the cube
and all planes parallel to them. Each vanishing line
also contains the vanishing points for all lines parallel
to their respective planes, including the diagonal
vanishing points (dvp1, dvp2 and dvp3) for the
planes.

A vanishing line perpendicular to the viewer's vertical

orientation (parallel to the ground plane) is typically
the horizon line in architectural or landscape uses
of perspective. It is the vanishing line for all planes
parallel to the ground plane, and contains all
vanishing points for lines parallel to the ground plane
(perspective rules 13 and 14).

Each vanishing line is connected to the vanishing

point opposite to it by an auxiliary horizon
line (shown in orange in the figure). These are
the vanishing lines for measure points for each of the
three dimensions of the cube. In 2PP, the horizon
line was a vanishing line for both the vanishing
points and measure points, but in 3PP these
functions can be separated.

The auxiliary horizon lines always intersect at the

direction of view (the principal point) — that is,
they link the vanishing points of the object to the
vanishing point of the viewer's central recession
(perspective gradient). Therefore the principal
point is always inside the vp triangle formed by
the three vanishing lines: if it is not, then the
primary form does not define right angled vanishing
points (it is a pyramid or a lopsided cube).

The measure points become significantly more

complex in the 3PP orientation: two vanishing
points define the edges of each face, and each
edge requires its own measure point. So we have in
all six measure points (mp1 to mp6) — two for
each vanishing point in relation to the two faces it
governs.

Finally, with the visual ray method we had a simple

way to rotate the vanishing points in 2PP, but this
also becomes significantly more complex in 3PP. In
2PP we just had to rotate two faces joined
in one right angle, which we could easily diagram in
two dimensions as two lines joined in one angle. In
3PP we must rotate three faces joined in three right
angles, and that complicates the visual ray approach
to a perspective solution.

Direction of View & Horizon Line. A 3PP

construction allows the direction of view to be
oblique to the ground plane, so that we
are looking down or up on objects rather than
looking at them directly from one side. Consequently
in 3PP it is necessary to distinguish between (1) the
object geometry (the vanishing points defined by the
edges of the primary form), (2) the central recession
defined by the direction of view, and (3) recession on
the ground plane, for example in the visual texture of
forests, grassy plains, deserts or bodies of water.

For example, we can redraw the cube illustrated

above in two point perspective so that it has
exactly the same angular size in the field of view
(using a measure bar), and is positioned below the
direction of view so that we look down on its upper
face at a 35° angle. This locates the top front corner
on the 71° circle of view and the bottom front corner
just in front of the ground line (diagram, below).

the 3PP canonical view in two point

perspective

Because both the angular size of the cube and the

angle of its faces to the direction of view are
identical, we are viewing it from exactly the same
location in physical space. All we have done
is shift our gaze from the object itself to the
horizon line behind it. This keeps the same visual
angle between the front corner of the cube and the
horizon line. But changing the direction of view in
3PP means that:
(1) the horizon line no longer must intersect the
principal point, and in fact may no longer be within
90° circle of view; and

(2) the geometrical relationship between any two

vanishing points (the size and shape of the triangle
the vanishing points define on the image plane)
depends on the location of the third vanishing point
and the location of the direction of view (the
orientation of the image plane to the perspective
problem).

the perspective sketch construction

method
The solution is basically to draw the form first, so
you can locate the vanishing points and measure
points which will produce that perspective view. You
then use these to reconstruct the primary form in
accurate perspective, and to add objects around the
primary form within the same perspective space.
Why not draw the primary form by freehand
perspective alone? Because, as we've already seen in
2PP, inaccurate placement of vanishing
points results in a distorted perspective view; even
small distortions can be obvious in a finished
drawing. There is a better way.
You start with a freehand perspective sketch or
scaled down perspective drawing at the center of a
fairly large piece of paper (a 3' section from a roll of
wrapping paper or white butcher paper is ideal).

three point perspective: perspective

sketch of the primary form

Your drawing or photograph of the primary form

should be small enough to fit all perspective points
on the sheet of paper, yet large enough to work with
accurately — usually a drawing about 10cm or 4
inches on its longest side is practical.

Take your time with the freehand drawing, and try to

capture the relative proportions of the dominant
edge angles and faces as accurately as you can.
Don't worry about extraneous features (such as
doors, windows or domes): you want to capture the
basic perspective shape as it recedes in three
directions. Be sure to define the edges and corner
points clearly.

You can also start with a drawing or photograph of a

building or monument that presents clear vanishing
lines in its edges or surfaces, in the perspective
orientation you want to duplicate. This photograph is
only used to specify the approximate perspective
view of the primary form in the drawing, so it does
not have to look anything like the primary form you
actually want to draw.

Once the drawing is finished to your satisfaction, or

you have taped your photograph to the sheet of
paper, you draw prospecting lines from the edges
of the front planes to find the three vanishing points.
Using a ruler or yardstick, extend the outer
edges of the form until these prospecting lines
intersect at three separate points. In a cube you
have three edges tending to each vanishing point;
use these in combination to reconcile discrepancies
and find the point that gives all three the best
definition.

The Vanishing Line Triangle. Next, connect these

three vanishing points with three vanishing lines. You
have defined the vanishing line triangle that will
define (and usually contain) the primary form.
 three point perspective: the vanishing line
triangle

This is the point to look at the overall placement of

the vanishing points in relation to the primary form
and the space around it that will appear in the
finished drawing. You can block in the format outline,
or sketch other large forms around the primary form,
to make sure you will get the effect you want.

Constructing Auxiliary Horizon Lines. Next, draw the

three auxiliary horizon lines through each
vanishing point and perpendicular to the opposite
vanishing line. There are two ways to do this. The
quicker is to use a large carpenter's square, laying
one side against each vanishing line and sliding it
back and forth along the line until the other arm is
exactly on the vanishing point. Then draw the line.
 three point perspective: constructing the
auxiliary horizon lines

A more accurate method in large drawings is

to construct the perpendiculars using a long piece
of fishing line, hemp (not stretchable cotton) string
or strip of cardboard as a compass measure. With
your thumb, a tack or a piece of tape, fix one end of
the measure at the vanishing point, and with the
other end scribe a wide pencil arc across the opposite
vanishing line. (Put the tip of the pencil through a
loop in the string or a small hole in the cardboard
strip.) The arc must intersect the vanishing line at
two widely spaced points. Then either scribe two
intersecting arcs centered on each of these new
points, or measure with a ruler 1/2 the distance
between them.
In the figure, the two arcs have been scribed
around vp1 and vp2, and through vp3, to define the
new points X and Y. Intersecting arcs drawn
from X, Y and vp3 create the new points P1 and P2;
lines to these points from the corresponding
vanishing points create two auxiliary horizon lines.
The direction of view (dv) is always at the
intersection of all three auxiliary horizon lines, so the
third line can simply be drawn from vp3through dv to
the opposite vanishing line. You end up with
a vanishing line triangle similar to the one shown
above.

Didn't I say elsewhere that the freehand

placement of vanishing points leads to distortions?
No: it's the clumsy scalingof drawing size in
relation to the distance between the vanishing points
that introduces distortions. If your three auxiliary
horizon lines are at right angles (perpendicular) to
their vanishing lines, if they meet in a single point
(dv), and if this point is inside the vanishing line
triangle, then the triangle defines a valid (physically
possible) perspective space for a rectilinear solid.

Constructing the Circle of View. Now we insert

the 90° circle of view. This requires you to (1) find
the midpoint of any of the three vanishing lines
(connecting two vanishing points), (2) draw
a semicircle of Thales over the vanishing line, (3)
extend to the semicircle the auxiliary horizon line
that intersects the vanishing line, (4) construct a line
parallel to the vanishing line, and finally (5) draw a
second arc back to this parallel line. The intersection
of this arc with the parallel line defines the radius of
the 90° circle of view around dv.

three point perspective: constructing the

circle of view

In the traditional solution, the artist uses either a

ruler or the method of intersecting arcs to find the
midpoint M on the vanishing lines between two
vanishing points. In the diagram, I've chosen the
vanishing line between vp2 and vp3. When arcs of
equal radius are inscribed across the vanishing line
from the two vanishing points, they intersect at two
points, x and y. (1) A line through these points
defines the midpoint M of the vanishing line.

(2) From point M the artist constructs a semicircle of

Thalesbetween the two vanishing points, then (3)
extends to the semicircle the auxiliary horizon line
that intersects the inscribed vanishing line at P. This
defines a new point C. (For visual clarity, the
semicircle is shown outside the perspective triangle,
but to save space it can just as well be drawn to
intersect the interior auxiliary horizon line.)

(4) Next, the artist constructs a line through dv that

is parallel to the vanishing line.

(5) Finally, the artist inscribes an arc from P with

radius equal to PC, the extended segment of the
auxiliary horizon line. This intersects the line parallel
to the vanishing line at either H1 or H2, depending
on where it is more convenient to construct the arc.

(6) The line segments dv-H1 or dv-H2 are

equivalently the radius of the 90° circle of view. The
artist draws this circle from H1 (H2) with dv as its
center.

It is often useful to include the 60° circle of view,

which is a second circle with a radius equal to 0.58
(58%) of the radius of the 90° circle of view. This
completes the perspective space.
Locating Measure Points. The last step is locating the
measure points. Six are required if they are marked
along the vanishing lines, but only three if you locate
them on the auxiliary horizon lines.

Auxiliary Horizon Line Measure Points. To find

the measure points on the auxiliary horizon lines, use
a protractor or architect's triangle (or the traditional
method for constructing a perpendicular) to
construct finish perpendiculars on each auxiliary
horizon line, from dv to the circle of view: the
intersection with the circle of view defines three new
points, C1, C2 and C3. Draw arcs from each of
these C points back to the auxiliary horizon line
perpendicular to it, using the vanishing point on that
auxiliary horizon line as the center of the arc.
 three point perspective: finding the
measure points

This completes the perpective space at a reduced

scale. I find that this entire procedure, starting with
a blank sheet of paper and ending with the finished
perspective space, requires about 20 minutes to
complete. Once you understand how to do it, the
work goes quickly and smoothly.

You must carefully make seven measurements on

this drawing (using a metric ruler) to rescale it to full
size: (1) the longest distance between any two
vanishing points (in the example, vp3to vp2), (2) the
distance from one of these vanishing points to the
intersection with the auxiliary horizon line (vp3 to h),
(3) the length of this auxiliary horizon line
(h to vp1), (4) the length to the direction of view
(h to dv), and finally (5-7) the distance from dv to
each of the three measure points.

Divide the radius of the circle of view you want in the

full sized drawing (say, 160cm) by the radius of the
circle of view in your perspective sketch: multiply all
the measurements by this number. This gives you
the full scale perspective space. Your perspective
work surface needs to be at least as long as the
longest vanishing line and as wide as the 90° circle
of view. In the example drawing, assuming a 3m
circle of view, this would be roughly 5m by 3m.

On a surface large enough to accommodate these

distances (a very large table, or a clean hardwood or
linoleum floor, or a clean, flat patio, garage floor or
driveway), measure out the longest vanishing line (in
the figure, vp2 to vp1), and the auxiliary horizon line
to vp3. Connect the three vanishing points to define
the vanishing line triangle. Measure the distance
from the vanishing line to dv, and draw the
remaining two auxiliary horizon lines from the
vanishing points through dv. Finally, mark the
three mp's on each auxiliary horizon line, measured
from dv.

Use the drawing scale shown in the distance to

size table to compute the drawing scale — the
percentage of the actual object size (for a given
viewing distance) that the drawing of the primary
form should have. On a piece of paper, make a rough
sketch of the primary form at this size, and lay the
sketch on the format (size of support) you intend to
use, to make sure the proportions work.

Vanishing Line Measure Points. The 3PP method

of using three measure points is convenient, but it
fails when the anchor point for measurements is
close to the direction of view (dv). In this case, you
may want to use the vanishing line points instead.

The construction of the circle of view required a

semicircle of Thales drawn around one of the
vanishing lines, centered on Mand intersecting the
vanishing points at either end of the vanishing line,
then extended the auxiliary horizon line to intersect
the semicircle in a point h'. This is all you need to
define the measure points on that vanishing line.
(Note that you can save steps and work space by
intersecting the auxiliary horizon line inside the
perspective triangle, to define interior h', and
construct the measure points from there.)

three point perspective: alternate method

to define measure points

The point h' will always define a 90° angle with the
two vanishing points on the vanishing line. That is, it
is equivalent to the viewpoint in a 2PP rotation of
vanishing points. So you can draw two arcs from this
point back to the vanishing line, using each vanishing
point as the center of an arc, to define the measure
points for the vanishing line — just as you would
in two point perspective.
Confusion about the choice of vanishing line measure
points is usually dispelled by the following two
criteria:

• The controlling vanishing point is the vanishing

point for the convergence of the edges that are
being sized by the measure bar. Thus, edges
converging to the right side vanishing point (vp2) are
controlled by that vanishing point.

• The measure point to use was defined by an arc

from the controlling vanishing point.
Thus, mp4 was defined by an arc centered on vp2,
so mp4 is the measure point to use when sizing
edges that recede to that vanishing point. The height
dimension is controlled by the vertical vanishing
point (vp3), which was the center of the arc used to
define mp3.

Measure bars to the vanishing line measure

points always must be parallel to the vanishing
line containing the measure point being used, not to
any auxiliary horizon line as before. Note that two
measure points are always available for each
dimension. In the example, mp6 can be used to size
the vertical edges receding to vp3, if for some
reason mp3 is inconvenient to use — but in that
case, the measure bar must be parallel to the
vanishing line containing mp6.

The measure bars in the illustration are the same

length as those used previously, and as you can see,
they define the same reduction in perspective depth.
You do not need to rescale or recompute the
measure bars you already have; just align them
parallel with the appropriate vanishing line.

Because the semicircle on M is part of the circle of

view procedure, and any vanishing line can be used
to define the circle of view, you should consider the
location of your anchor points in the perspective
space, and place the semicircle of Thales around the
vanishing line where measure points will be most
convenient.

For example: I had originally put the anchor point at

the front bottom corner of the cube; in that
location mp3 worked fine, but the other two points
created badly slanting measure lines that would
introduce inaccuracies. The best alternative points
would be found on the top vanishing line
(between vp1 and vp2), so I should have started
building the circle of view by putting the first
semicircle on that side.

constructing a 3PP cube

(perspective sketch method)
Once you have constructed the 3PP space, you can
begin construction of the cube or primary form. This
explanation excludes the procedures necessary to
scale the drawing, which are developed below.

three point perspective: locating the

primary form

Measure out the perspective space on the

perspective drawing surface (floor, driveway, patio),
and tape or tack the support to the surface,
oriented with the top edge parallel to one of the
vanishing lines (or to none, if the perspective view is
tilted), and the dv in the correct location within the
drawing. If you do not want to work directly on your
watercolor paper, reconstruct the drawing on a large
sheet of butcher paper or wrapping paper, and then
trace or square the drawing to the format when
you are done.
Mark the dv and draw the auxiliary horizon lines, the
measure points, the anchor point, and the base
vanishing lines through the anchor point.

The diagram shows this done on an emperor sheet,

40" x 60", located with the dv near the bottom. (If
you are going to all this trouble, you may as well
make the painting spectacular!) For clarity, the
support outline is omitted in the next several
illustrations, although it is assumed you are working
with the support in place.

three point perspective: constructing

measure bars

The last preparatory step is constructing the measure

bars. Do this from the center of the space (dv),
because each measure bar must be parallel with its
corresponding auxiliary horizon line. This is easiest to
do by simply drawing the measure bar on a separate
sheet of paper, directly over the auxiliary horizon
lines.

Draw the measure bars to the perspective length

they have in space, so that you can line them up
with one end against the anchor point. The length of
the measure bars determines the drawing size of the
primary form, so you want these to be accurate. For
example, if the dimensions of a building are 150 feet
long, 75 feet wide and 36 feet high, and you used
the length of the building to scale the drawing size,
then the proportions between the measure bars are
1.00 to 0.50 and 1.00 to 0.24. Since we are drawing
a cube, all three measure bars will be of equal
length, so we define them by drawing a circle
around dv (shown above).
 three point perspective: constructing front
vertical

The rest is a piece of cake. First, align the vertical

measure bar parallel to the vertical auxiliary horizon,
with the bottom end on the anchor point. Draw a line
from mp3 through the top end of the measure bar to
the vertical vanishing line (that is, the line parallel to
the measure bar you are using). This defines the
front height of the cube.

Using a yardstick, string or cardboard strip aligned

with the bottom vanishing point, draw a line from the
anchor point to the line from the vertical measure
bar to mp3. This is the front vertical. Use the
yardstick, string or cardboard strip to connect the
ends of this vertical to the two side vanishing points,
and draw the front top and bottom edges of the
form.
 three point perspective: constructing left
side

Next, use the second measure bar to define the

depth dimension on one side (to mp1, the measure
point on the auxiliary horizon line parallel to the
measure bar you are using). When one end of the
measure bar is aligned with the anchor point, the
back corner of the cube is located where the line
from the other end of the measure bar
to mp1 crosses the bottom left edge of the figure.
Mark this point.

Again aligning your straight edge with vp3, draw a

line from this point to the top left edge line: this is
the back vertical of the cube. Connect the ends of
this vertical along vanishing lines to vp2. These lines
define the back left upper and lower edges of the
figure.

three point perspective: constructing right

side

With the third measure bar, construct the opposite

side, define the corners, and connect to the
vanishing points as before.

Clean up the drawing as much as necessary to

visually confirm the final perspective outline meets
your expectations. Then go on to add any other
objects in the environment around the primary form,
or perspective details on its surface (doors, windows,
etc.).
 three point perspective: perspective
distortions

Familiarity with the 3PP mechanism will help you

understand how to use it effectively. The diagram
(above) gives some clues about the scale, placement
and cropping of forms:

• In general, distortions toward the side vanishing

points are much more objectionable than those
toward the bottom vanishing point: choose a
vertical or square formatwhenever feasible.

• Forms can be placed below the 90° border — a

90° angle placed to intersect the two side vanishing
points (red line) — to emphasize height or vertical
scale, but forms should not be placed near the
border on either side. (All possible locations of the
right angled corner of this border are defined by
a circle of Thales constructed below the horizon
line.)

• The same circle of view rules apply in order to

reduce perspective distortions, but the circle can
be displaced downward from the direction of view, as
if pulled away from the horizon line by the vertical
depth. It is better to think in terms of a column of
view centered on the principal point and extending
from below the 90° border to above the horizon line
(where cloud layers in perspective can enhance
distance depth to balance the vertical depth). Any
format that fits within a 40° to 60° column will
produce a handsome image.

When you have finished with the perspective

elements, carefully release the drawing surface from
the table, floor or patio, and lay it out on your
painting surface to erase the guidelines, measure
points, and other extraneous elements, or to transfer
the perspective outline to the actual painting surface.
When the drawing is fully cleaned, add by freehand
any additional outlines or guidelines necessary before
you begin to paint.
 three point perspective: finished drawing

The diagram shows the finished perspective form,

once again within the monumental 40"x60" emperor
format. In this reduced diagram, the primary form
appears to be little changed from the original
perspective sketch. But in practice, despite all the
work invested, you will be quite pleased with the
increased perspective accuracy and "weight" of the
finished drawing in comparison to anything you could
manage by freehand methods alone.

the horizon line construction method

Two significant problems with the perspective sketch
method are that it establishes the angles of the the
primary form to the viewpoint approximately,
through a sketch, and that it cuts the 3PP methods
loose from the procedures for scaling the
drawing within the circle of view. The actual
perspective angles and scale of the circle of view are
derived from the drawing, rather than given at the
start. An alternative method is to start with the circle
of view, and from there construct the vanishing
points. This method starts by specifying the location
of the horizon line (a horizontal vanishing line above
or below the direction of view), so I refer to it as
the horizon line method of 3PP construction,
though the circle of view method is also apt.
A discussion of the 3PP geometry will clarify how this
method works. Because all parallel lines converge to
the same (single) vanishing point
(perspective rule 6), and the 3PP vanishing points
define visual rays at right angles to each other, the
3PP vanishing points are equivalently defined by the
three right angled edges of a cube that can be turned
or rotated around a front corner fixed on the
direction of view (diagram, right).

These edges converge to the three right angle

vanishing points at the vanishing lines for the three
planes defined by the three front faces of the cube
(perspective rule 14). Therefore the vanishing lines
between the pairs of vanishing points will be parallel the 3PP vanishing points defined by
three edges of a cube
to the line intersections of the three front faces of
this cube with the image plane (green, corollary to
perspective rule 11). As a result, we have reduced
the geometry of the 3PP vanishing points to the
geometry of a three sided pyramid thrust through
the image plane in any arbitrary angle and rotation.

As explained earlier, the circle of view framework

provides a method to specify exactly the location of
any vanishing point as a line rotated to the required
angle around the viewpoint folded into the image
plane. What we require is a way to perform this
folding for elements of the 3PP "pyramid".

This is done by moving the fixed corner of the

cube forward until it coincides with the
viewpoint. In that position its three edges define
three visual rays to the vanishing points (magenta
lines, diagram above right). More important: the
altitude of the pyramid is now equal to the viewing
distance and therefore to the radius of the 90° circle
of view (diagram, below).
 folding a pyramid right triangle into the
image plane

Two kinds of folding operations are possible in this

3PP geometry. First are the auxiliary line folds that
define the interior angle between a pyramid edge, or
the pyramid face perpendicular to it, and the
direction of view. These are found by folding into the
image plane a vertical section of the pyramid defined
by an auxiliary horizon line, for example the interior
triangle PVC defined by the auxiliary horizon
line PC in the diagram above. This triangle contains
the two triangles VdvCand VdvP, each containing a
right angle at dv. The fold brings line Vdv into the
image plane as x'dv. Because the edge Cdv is
continuous with edge Pdv, the right angle at dv is
preserved. And the image
edges Cx' = CV and Px' = PV. Therefore, by
triangular equalities, the image angle 1' equals the
interior angle 1, the angle between the direction of
view and the face ABV.

This fold also identifies (at Cx'dv) the angle between

the vanishing point C and the direction of view, so
this folding down of an interior section of the
perspective pyramid is geometrically identical to the
folding of the viewpoint into the circle of view that is
used to rotate vanishing points to the direction of
view.

The second kind of folding operations are

the vanishing line folds that define an exterior
angle of one face of the perspective pyramid
(angle 2) as a "plan view" of the angle in the image
plane (angle 2'). This is the angle, on the face of the
3PP pyramid, between the edge of triangle ABV and
its altitude PV. The fold is achieved by constructing a
line (ab) that intersects the direction of view
parallel to the vanishing line (AB). This line
intersects the circle of view at x'.
Because Vdv equals x'dv, the line Px' equals
line PV, the altitude of ABV. Therefore an arc
constructed on P with radius Px' intersects the
auxiliary horizon line at x, and Px = PV. Therefore
the right triangle ABxis the perpendicular view of the
foreshortened triangle ABV, and x is the auxiliary
viewpoint for the horizon line AB.
 three right triangles folded out of the 3PP
pyramid

The diagram (above) shows the three possible

vanishing line folds and auxiliary viewpoints
(x, y and z) constructed from a 3PP vanishing line
triangle. Study this diagram carefully until you
understand how each fold has been done.

The geometry of triangles is efficient: defining any

one side with its two adjacent angles, or any two
sides with their common angle, defines the rest of
the triangle. Therefore only two folding
operations are necessary to define the image of a
3PP vanishing line triangle: one auxiliary horizon line
fold and one vanishing line fold. This is sufficient to
define the location of all three vanishing points and
vanishing lines in relation to the direction of view and
circle of view.

Finally, the 3PP construction releases the direction

of view from its parallel position to the ground plane,
and this creates several novel features in the
perspective geometry which affect in particular the
scaling of the 3PP drawing. For now I only want to
describe this geometry and define a few new terms
(diagram, below).

elevation view of 3PP geometry

In this example we assume the perspective view
is downward in relation to the ground plane: it can
just as well be upward (as the top of a skyscraper
viewed from the ground) or tilted (as a city viewed
from a turning airplane), a problem I leave for the
reader. In the downward view case:

• The image plane is oblique to the ground

plane, as is the direction of view. As a result the
direction of view does not terminate in a vanishing
point, but in a fixation point, some physical point
on the ground. This fixation distance is typically
different from the object distance from the station
point to the primary form.

• The station point S is still directly under the

viewpoint, but now the station point appears on the
image plane, where it is the image s equivalent to
the vertical vanishing point (vp1).

• The horizon line is now located above the direction

of view in the circle of view, which means the
principal point, the vanishing point for the viewer's
central recession (at p), is no longer the same as the
orthogonal vanishing point (at h), the vanishing
point of ground plane recession.

• The primary form appears in rotation

foreshortening — the vertical and horizontal
dimensions are in a different scale. Foreshortening is
corrected by using the measure points; measure
bars parallel to the image plane may be
rotatedin the image plane to any other angle.
However, it is sometimes useful to estimate the
amount of vertical foreshortening, for example when
planning the image layout. This is found by a cosine
correction for foreshortening:

where θ is the horizon angle. Because the angle of

view to the ground plane is only equal to the horizon
angle at the fixation point, a measure bar established
at any other point must be calculated with the
correct angle of view to that point on the ground
plane.

• An object's angular size or image size is

determined by the sight line distance from the
viewpoint, which is simply the hypotenuse of the
right triangle formed by the object distance and
viewing height.

These points need to be understood in order to apply

the correct distance & size calculations
when scaling the drawing in a 3PP construction.

constructing a 3PP drawing

(horizon line method)
The horizon line method builds on the assumption
that most three dimensional perspective problems
concern a viewer whose line of sight is not parallel to
the ground plane. Either the viewer is looking
upward, toward the top of a tower, building,
mountain or cliff; or the viewer is looking downward,
from a vantage at the top of a tower, building,
mountain or cliff.

Approximate Horizon Line Method. In this approach

the artist places the horizon line and vanishing points
by judgment or whim, but uses the pyramid
folds to make these landmarks consistent with each
other.

The first step is the placement of the horizon line in

relation to the principal point: either above (for a
downward direction of view) or below (for an upward
direction of view). Then, using a drafting triangle, the
artist finds the 90° angle at one of the diagonal
vanishing points, and extends this line until it meets
the median line below the circle of view: this is the
vertical vanishing point (vp1).
 three point perspective: rotating the
horizon line

It is useful to bisect this angle to find the diagonal

view (45° from either the horizon line or vp1 visual
rays), as this projects in depth the viewing height
above the ground plane.

Next the second vanishing point (vp2) is located on

the horizon line somewhere to the left of the median
line. A ruler laid from dv to this point will show the
angle to view of a cubic form in perspective space at
the direction of view.

three point perspective: approximately

placing vp2

Once the location of the point is completed, draw

the vanishing line between the two vanishing points.
Then you must construct a perpendicular line from
this vanishing line through the direction of view (dv),
as described here.

The steps are: (1) draw a circular arc around dv that

intersects the vanishing line at two widely spaced
points, a and b; (2) draw an arc from each point
with a radius greater than half the segment length
between them; (3) draw a line through the double
intersection of the arcs to define the normal point c;
(4) draw a line from c through dv until it intersects
the horizon line on the opposite side of the circle of
view. This is the auxiliary horizon line for the
constructed vanishing line; it locates vp3.
 three point perspective: completed
"approximate" perspective triangle

Construct the third vanishing line and its auxiliary

horizon line from vp2 through dv.

Find the internal or external altitude points on the

auxiliary horizon lines, and from these locate the six
measure points on the vanishing lines. (The diagram
above shows one internal altitude point, h' and the
two measure points constructed from it.) This
completes the three point perspective triangle.

Exact Horizon Line Method. In some cases

(illustrated below) it is desirable to locate the three
vanishing points precisely. In this case the pyramid
folds are precisely defined with a protractor or using
the tangent ratio for the required angle, applied to
the radius length of the circle of view that is
perpendicular to the viewpoint.

Required is one auxiliary horizon line fold along the

vertical auxiliary horizon line (median line) to
establish the tilt of the horizon line and the location
of the vertical vanishing point (vp1), and
one vanishing line fold along the horizon line to
establish the left/right placement of vp2 and vp2.

The diagram (below) shows these operations to

provide an exact 25° downward angle of view to the
ground plane (upward horizon angle of 25°), and a
placement of vp2 55° to the left of the median line.
This places vp3 35° to the right of the median line.
 three point perspective: exact rotation of
vanishing points

The vanishing lines are added as before; the

auxiliary vanishing lines can be drawn directly, as
lines from the vanishing points through dv to the
opposite vanishing line, because the vanishing points
have already been precisely located.

Measure points have been added using the

"alternative" horizon line method described above.

three point perspective: completed

"exact" perspective triangle

Although this 3PP triangle is very similar to the one

constructed from an approximate judgment of the
correct angles, here all the perspective landmarks
are exactly placed from given values established in
advance. This is especially important when the goal
is a 3PP view of a specific primary form from a
specific location — as is typical in architectural
renderings or historical reconstructions — or when a
certain arrangement of key forms within the image is
required.

Diagonal Vanishing Points. It is usually very useful to

take the extra step and establish the diagonal
vanishing points on the horizon line. Once this is
done a unit dimension on the station line can be
projected across the ground plane, using the method
of projecting a unit dimension in depth from the
diagonal vanishing points.
 three point perspective: locating the
central dvp's

In fact, no separate rotation is required to define the

diagonal vanishing points from the auxiliary
viewpoint: they are already located at the
intersection of the arc used to define the auxiliary
viewpoint with the horizon line (diagram, above).
The method of rotating the diagonal vanishing points
around the auxiliary viewpoint A, so that a 90° angle
is bisected by the vertical auxiliary horizon line
(median line), is shown simply to confirm this.

Once these diagonal vanishing points have been

established, ovp serves as the orthogonal vanishing
point, the convergence for recession in depth parallel
to the ground plane (the ground plane central
recession); but dv remains the principal point, the
convergence for recession parallel to the direction of
view (the viewer's central recession). The depth of
transversals across orthogonals to ovp are found by
diagonals to dvp1 and dvp2; the depth of
transversals across orthogonals to dv are found by
vanishing lines to points on the circle of view.

Scaling the 3PP Drawing. This task is more complex

than it is in one or two point perspective, but I
outline it here because I have not seen it discussed
in any other source. A minimal reliance on
trigonometry is required, both to validate the basic
principles and to provide calculation shortcuts or
remedies to complex construction problems.

Construction Methods. Three drawing scale guides

are already available: (1) the circle of view and the
many visual anglesthat can be computed within it;
(2) the viewing height in depth, added when the
horizon line was rotated; and (3) a ground line
scale, which is used in combination with the
orthogonal vanishing point (ovp) to project a unit
dimension in depth to approximately locate objects
in depth and scale their image size. Provided image
scale and perspective accuracy are not critically
important, these are almost always adequate to scale
the 3PP drawing.

three point perspective: ground plane

recession
Two scaling approaches can be used. In the first
example (above), an arbitrary unit dimension of
50 cm measured along the station line image is
projected into perspective space by orthogonals
drawn to the orthogonal vanishing point (ovp).
These indicate that the viewing height in depth is
about 7.2 times the viewing height (measured from
the station line) and that the fixation point is about
16 units away.

Dividing the viewing height by the depth units yields

the ground plane width of the unit dimension. If the
viewing height is 300 meters, then the unit
dimension represents about 300/7.2 = 42 meters,
and the fixation point is about 672 meters from the
station line — all distances measured on the ground
planerather than along the line of sight. If the
viewing height is 3 feet, then the unit dimension
represents 36/7.2 = 5 inches and the fixation point is
80 inches away.

The orthogonals define this unit dimension at any

depth; a transversal established at the base of the
primary form creates a measure bar in unit
dimensions at that distance. In the diagram (above),
a measure bar is shown at the fixation point that is 3
unit dimensions long. If the viewing height is
300 meters, then the measure bar defines a width of
126 meters at a distance of 672 meters. This image
bar is used to measure out the size of the primary
form image along and above the base transversal.
 three point perspective: key scaling
dimensions

The second method is to establish an exact unit

dimension. This is done by (1) drawing a line from
one of the dvp's through the viewing height in
depth (vhd) and extending this line until it intersects
the station line image, then (2) dividing the distance
from this intersection to the station point (vp1) by an
appropriate number of units. In the example this
procedure yields a station line length of 354 cm,
which is conveniently divided into six 59 cm units. If
the viewing height is 270 meters, then this unit
dimension represents exactly 45 meters on the
ground plane.

As explained in the discussion of scaling the

drawing in the 1PP context, the location of the
format requires the artist to decide the appropriate
size and location of the primary form image. The
principal scaling restriction is the horizon line rule:
the horizon line intersects all forms at a height above
their intersection with the ground plane that is equal
to the viewing height, or causes the forms to appear
below the horizon line by an equivalent added
proportion of their total height. Following the
procedure explained for central perspective, the
artist finds the ratio between the viewing height
and object height, then places the object so that the
horizon line divides or stands above the object image
by this ratio. This rule holds regardless of the angle
of the direction of view to the ground plane.

In the example I will develop, I want to render a

primary form that is 300 meters high and
125 meters wide. I want the upper portion of the
form to cut the horizon line, so that its top portion is
silhouetted against the sky. Therefore I plan on
approximately 30 meters of the form appearing
above the horizon line and the remaining 270 meters
below the horizon line. This means the viewing
height will also be 270 meters (according to
the horizon height rule), and 30/300 or 10% of the
primary form image will be above the horizon line.

To minimize perspective distortion and create a "long

view" image of the primary form, I decide to use the
fixation point as the anchor point. The 13th unit
transversal is just behind the fixation line (orange),
which indicates an object distance (on the ground
plane) of less than 585 meters. Now
the horizontal unit dimension can be derived directly
from the orthogonals along that transversal, then
rotated 90° to provide the vertical image dimensions.
Then the approximate image area of the primary
form can be defined within the circle of view (green
rectangle).

Finally, the format dimensions are positioned around

the primary form area, horizon line, direction of
view, or any other important composition elements.
This can be done first, and the primary form fitted
within the format, or done after the primary form is
located within the circle of view. (Either way, the
format dimensions are established as a proportion of
the circle of view radius, as explained here.) Given
my angle of view and the monumental size of the
primary form, I decide on a large format. The
example shown below is the 29"x42" (74 cm x
107 cm) double elephant (USA) format, in "portrait"
orientation and positioned to accommodate the
primary form above and below the horizon line
(yellow rectangle).
Comment: if you compare the perspective gradients
in the previous two diagrams to the perspective
gradient in central perspective, the recession
appears more gentle in 3PP — the squares at the
base of the circle of view are still vertically elongated
in 3PP, but are horizontally elongated in 1PP. This is
because the ground plane is viewed at a more
oblique angle and is therefore less foreshortened (at
the station point or vp1, the view is perpendicular to
the ground plane, as indicated by the shoe prints).
But in the visual area above the location of the
viewing height in depth, the two gradients become
equivalent.

Calculation Methods. The alternative scaling

method uses calculation rather than approximate
construction. This method is more precise and
robust. The diagram below identifies the key scaling
terms in relation to the elevation view of the image
plane and visual rays from the viewpoint, and the
image plane as it appears in the perspective
diagrams.
 three point perspective: key scaling
dimensions

vp = viewing distance; vs = station line

distance; hvp = horizon angle; h = orthogonal
vanishing point (horizon line); p = principal point
(fixation line); z = viewing height in depth; s = image
station point (station line image); hp = horizon
height; hs = ground image height; hz = image depth

The only preparation necessary for these calculations

is specification of (1) the viewing height, (2) viewing
distance perpendicular to the image plane, and (3)
horizon angle. Continuing the example above, I set
the viewing height at 270 meters, the viewing
distance at 1.5 m (150 cm) and the horizon angle at
25°.

In addition, you will need a pocket calculator that

can provide the three trigonometric
functions (sine, cosine and tangent) for
any horizon angle. In the example, the horizon
angle is 25°, therefore:

sine(25°) = 0.423
cosine(25°) = 0.906
tangent(25°) = 0.466

Format Dimensions. The procedure for establishing

the format dimensions is explained here. It is
useful to do this first, if an appropriate format size
can be decided in advance, as this provides a frame
of reference for other scaling decisions. I will
continue with the double elephant example
illustrated above.

Ground Scale. The second step is to establish

the ground scale, the scale of the station line S on
image plane at s (refer to the diagram above):

• The "level line" role of the horizon line is taken by

a fixation line through the direction of view and
parallel to the horizon line above or below it. This
defines the vanishing line for all planes parallel to the
direction of view and to the horizon line
(perspective rule 15), and the "actual size" image
scale (at a viewing distance of 150 cm, the circle of
view radius along the fixation line is 150 cm).

• The horizon height, the distance of the horizon

line above the fixation line as measured on the
image plane, is equal to the viewing
distance multiplied by the tangent of the horizon
angle θ. The tangent of a 25° horizon angle is 0.466,
so the horizon line is 150 cm*0.466 = 70 cm above
the direction of view.

• The station line image is exactly below the

viewpoint v, at point s, which defines the right
triangle vps (because the direction of view is
perpendicular to the image plane). Therefore
the station line distance (the distance of image
plane below the viewpoint, or vs) is the hypotenuse
of the right triangle vps. This is equal to the viewing
distance divided by the sine of the horizon angle, or
150 cm/0.423 = 355 cm.

• The ground image height is the extent of the

image plane between the horizon line and station line
image, or the hypotenuse of the right triangle hvs.
This is found as the station line distance divided
by the cosine of the horizon angle: 355 cm/0.906 =
392 cm.

• The station line image is below the fixation line at a

distance on the image plane equal to ps, or the
ground image height (hs) minus the horizon height
(hp): 392 cm – 70 cm = 322 cm.

• Finally, formula 5 provides the image ratio for the

station line scale. If the viewing height is 270 meters
and the station line distance is 355 cm, then the
station line image scale is 3.55 m/270cm = 1.31%;
or equivalently, a 1 centimeter unit dimension in the
station line equals 270/3.55 = 76 cm on the ground
plane. To obtain the 45 meter unit dimension in the
ground plane as units of the image plane at the
station line: 45*0.0131 m = 59 cm unit dimension in
the station line. This is the ground scale, as
summarized in the following formulas:

where the ground unit is whatever measurement unit

is convenient for mapping objects on the ground
plane (e.g., 1 meter or 100 meters), and the viewing
distance, station line distance and viewing height are
all measured in the same units.

• Finally, the diagonal vanishing points can be used

to project the ground scale unit dimension in depth.
The diagram (above) shows that the 270 meter line
established by the projecting the 45 meter unit
dimension in depth exactly coincides with the
location of the viewing height projected in depth by
the horizon plane rotation.

Fixation Line Scale. The third and last step is to

determine the image depth z on the image plane for
an object on the ground plane at distance X from the
station line, or the object distance X on the ground
plane of a particular image depth z on the image
plane. These relationships are defined in the
following formulas:
 three point perspective: anchor point and
anchor line

The image plane extends from the image station line

(s) to the horizon line (h), defining the horizon
height (sh). This image plane can be duplicated by
a secondary image plane Sx, some distance in front
of and parallel to it. The secondary plane intersects
the horizon line at x, removed from the
viewpoint Vby the offset distance O, and intersects
the ground plane at the station line (S), which is a
zero horizontal distance from the viewpoint. These
intersections are identical with the image
points h and s, so the horizon height sh is the image
of the physical distance Sx when the object distance
(on the ground plane) is zero.

Moving the secondary plane forward by the object

distance Xrelocates the ground intersection to S' and
the horizon line intersection at x', so that the image
of S'x' is now zh — the image point z is at an image
depth (zh) below the horizon line. In this new
arrangement, the triangular proportions define the
proportional equality:

zh/sh = xV/x'V = OD/(X+OD)

which solves either for (16) the image depth (zh) or

(17) the ground plane object distance (X) by
algebraic rearrangement.
Using these formulas, we establish (for a horizon
angle of 25°, a viewing height of 270 m and a
viewing distance of 1.5 m) that:

(8) format scale = 25%(W), 36%(H)

(11) horizon height = 70 cm
(11) station line distance = 355 cm
(13) ground scale: 66 cm = 50 meters
(14) offset distance = 126 m
(15) ground image height = 392 cm
(16) image depth (of fixation line) = 70 cm
(16) image depth (of viewing distance in depth) =
125 cm
(16) image depth (of format baseline @ 500 m) =
79 cm
(17) object distance (at fixation line) = 579 m.

Anchor Line and Anchor Point. Given the fixation

point as the anchor point, the ground plane object
distance is 579 meters at a viewing height of
270 meters, which is a diagonal object distance of
639 meters to the base of the primary form. So
the fixation line scale (derived from formula 5) is:

(18) image scale (at fixation line) =

1.5/[5792+2702]1/2
= 1.5/639 m
= 0.00235 (0.235%)
To go from image plane units to ground plane units
(at the fixation line), you divide by the image scale
factor:

1 cm = 1 cm/0.00235 = 4.26 m.

To go from ground plane units to image plane units,

you multiply by the image scale factor. Thus, an
object 125 meters wide and 301 meters tall, oriented
parallel to the image plane, creates the image
dimensions

125 m*0.00235 = 0.294 m = 29.4 cm

301 m*0.00235 = 0.707 m = 70.7 cm.

These are the measure bar dimensions for the

image at the anchor point (established as the fixation
point), as shown in the diagram (below).
 three point perspective: anchor point and
anchor line

Confusion about the choice of vanishing line measure

points and the orientation of measure bars (for cubic
forms) is usually dispelled by the following three
criteria:

• The measure point to use for any edge is the

measure point defined by an arc from the
controlling vanishing point. Thus, the height
dimension is controlled by the vertical vanishing
point (vp1), which was the center of the arc used to
define mp2 and mp5.

• The controlling vanishing point is the vanishing

point for the convergence of the edges that are
being sized by the measure bar. Thus, the left side
edges of a plan in the ground plane are defined by
the lefthand horizon vanishing point.

• The measure bar is always oriented parallel to

the vanishing line containing the measure point.
With a plan and elevation of the primary form
(diagram, right), the artist is ready to construct the
perspective drawing.

elevation of the primary form

 three point perspective: constructing the
primary form

diagram enlarged to 60° circle of view for clarity

It is usually convenient to establish the plan or

"street map" dimensions of the drawing first, as the
plan outlines do not intersect one another and clearly
establishe the front to back ordering of large forms.
Here the drawing is being constructed from the
elevation only, as the base is square. Only the 60°
circle of view is shown for clarity.

Using the measure point guidelines (above), the

controlling vanishing point for the right side of the
base of the tower is vp3; and this vanishing point
defined the arc for mp3 (refer to the diagram
above). The controlling vanishing point for the
height of the tower is the vertical vanishing
point, vp1; and this vanishing point defined the arc
for mp2.

The fixation line measure bar is used to establish the

base width of 125 meters, and this dimension is
projected in depth by the lines to the opposite
measure points mp3 and mp4. The measure bar is
parallel to the vanishing line containing the measure
points, Then vanishing lines from the anchor point
to vp2 and vp3 establish the sides of the plan.

The tower is symmetrical on its four sides, but the

sides are not vertical: they define an exponential
function designed to maximize the tower's strength
against strong winds. To facilitate the perspective
construction we have to find the central axis, which
is simply the intersection of the diagonals of the
plan.

three point perspective: finished drawing

diagram enlarged to 60° circle of view for clarity

The major stages of the tower are marked off on a

vertical measure bar, this bar is rotated to be parallel
with the vanishing line of the appropriate measure
point, and the tower stages are projected onto a
vertical axis.
Two strategies are available. The existing elevation
drawing can be used to create the measure bar: this
drawing is in the image scale defined by the anchor
point on the fixation line, 579 meters from the
viewpoint. Elevation points are projected onto a
vertical line constructed from the anchor point. These
elevation points are the front corners of new
squares, of equal size as the base of the tower,
containing the tower platforms. These are
constructed "scaffold style", by vertical lines from the
four corners of the plan. At each level the scaffold
squares are recessed to the side vanishing points
from the front corner, and the diagonal found as
before. Then the plan of the tower platform is
constructed within this square, its four corners along
the diagonals.

The alternative method is to project the elevation

points onto the central axis, and project the tower
platforms out from these central points. This method
is also shown in the diagram: the measure bar must
be anchored at the base of the central axis. Note
however that this point is over 80 meters farther
away from the viewpoint than the front corner (as
shown by the 50 meter distance transversals in the
ground plan), therefore the measure bar must be
sized, using formula 18, to the new image distance.
First the added distance is derived from the whole
diagonal, which is then aligned to the direction of
view by the cosine correction:
base diagonal = [1252+1252]1/2 = 177 m
half diagonal = 177 m/2 = 88.5 m
ground plane distance = 88.5 m * cosine(10°) =
80.2

and the new ground plane distance (579+80 = 659)

is used to compute the new image scale:

(18) image scale (at central axis) =

1.5/[6592+2702]1/2
= 1.5/712 m
= 0.00211 (0.211%)

(Note that the central axis distance could be

estimated from its position just beyond the 650 m
transversal distance line.) Once the major external
points of the tower profile are established, the
outside curves of the tower can be drawn with a
French curve or freehand, and details of the tower
filled in as appropriate.
 three point perspective: finished drawing

diagram enlarged to 60° circle of view for clarity

And here is the finished drawing. The point of using

the exact rotation method is that the Arc de
Triomphe could be precisely positioned behind the
Tour Eiffel, and both positioned in relation to the
direction of view and horizon line, to produce a
specific effect.

The plan of the distant streets is taken from a

Michelin map of Paris, projected onto the ground
plane using the foreshortened and recessed
orthogonal squares and plotting the major streets,
square by square, as far back as useful.
 N E X T : Advanced Perspective Techniques

Last revised 07.I.2015 • © 2015 Bruce MacEvoy

advanced perspective techniques technique

The previous pages developed the methods of linear
perspective using the cube (or other rectangular
objects) as the primary form. This was convenient
because the edges and right angles of these objects
simplify perspective constructions.

When tackling the perspective of complex or irregular

shapes, the basic strategy is to fit these shapes
inside a regular geometric figure or solid, like a
valentine inside an envelope or a vase inside a box,
then use this rectangular solid to "mail" the form into
perspective space. This page gives several examples
of how to do this for plane figures and solid perspective of complex plane
forms. figures

Next, I provide a tutorial on projecting a building, perspective of complex solid

which is just another type of complex volume, from forms
architectural elevations and plan. The same
methods apply to any object for which elevations and buildings from blueprints or
plan are available. plans

The methods of paraline perspective, based on the paraline perspectives

geometry of parallel projection, were developed in
the 18th century (at the rise of the Industrial curvilinear perspectives
Revolution) for a variety of engineering and
manufacturing applications. I present a few of the
common forms and discuss how they relate to the
perspective of central projection.

I conclude with a brief discussion of curvilinear

perspectives, a modern and dogmatic answer to the
"distortions" in traditional linear perspective. I show
how to make a basic curvilinear template and explain
why the usual justifications for curvilinear
perspective are fallacious.

perspective of complex plane figures

One of the most common perspective problems is
rendering in perspective a curved or irregular figure
that is not a composite of squares or rectangles. The
most common example is a circle in perspective, as
the rim or contour of a disk, drinking glass, bowl or
cylinder.

The solution in each case is the same: to use the

square (or a metric grid within a square) as the
projection framework or projection square. The
rationale is that it is that it is simple to project a
square in perspective, and once this is done the
square or grid can be used to transfer descriptive
points from a plan or elevation view of the figure.

The general procedure is: (1) enclose the complex

figure within a regular rectangular form (square or
rectangle); (2) divide the rectangular area with a
regular grid and/or a major diagonal; (3) identify the
point intersections of the figure with the sides of the
square, the grid or the diagonal; and (4) transfer
these landmark points into the image plane, where
they are used to reconstruct the figure image.

Projecting A Circle. Let's start with the simplest

case, projecting a circle in perspective. I know of
several different methods to do this, but provide here
two that are among the easiest and most effective.

Circle Without a Plan. There is a very useful

method to construct the circle entirely from the
geometry of the square. No plan is required, because
the points are defined entirely within the image
plane; the diagram (below) shows a plan view only
to clarify how the method works.
 projecting a circle without a plan

Begin with any square established in perspective at

the appropriate scale, location, and angle of view.
The circle defines a plane, and the vanishing line for
this plane must be available as the principal point
(orthogonal vanishing point) and diagonal vanishing
point, or the controlling vanishing point(s) and
measure points. A diagonal vanishing point or
measure point is only necessary to define the square
in depth, but the principal point or primary vanishing
point of recession is required. Then:
1. Construct the full diagonals, ab and matching.

2. From the full diagonal intersection (center of the

square), construct the half transversal to c and the
half orthogonal from the principal point through d.
Mark the four intersections with the sides of the
square, c and d and the points opposite (black).

3. Construct two quarter diagonals, c to d and

matching on the opposite side. Construct the two
quarter orthogonals from the principal point through
the intersection e and its twin.

4. From the intersections of the quarter orthogonals

with the full diagonals, such as e, construct the
quarter transversals. The projection square is now
divided into sixteen smaller squares.

5. Construct the two rectangular diagonals from each

corner of the square to the intersection of the
nearest quarter line with the opposite side of the
square: that is, a to g opposite, h to fopposite, and
similarly for the other six rectangular diagonals.

6. Near each corner, mark the intersection of the

rectangular diagonals from that corner with the
nearest quarter orthogonal or quarter transversal:
that is, x and y for corner a, and similarly for the
other three corners (black).

7. Finally, join the twelve points, freehand or with a

French curve, to produce the circle.

This method is "nearly accurate", because

points x and y stand slightly outside a perfect circle,
as is visible in the diagram. However this is
inconsequential when working small or with very
foreshortened figures; or the circle can be drawn to
miss slightly the pair of corner guide points.

Circle With a Plan. Using a plan results in a slightly

more accurate set of guide points, and additionally
requires fewer guidelines to define. The diagram
below shows the procedure.
 projecting a circle with a plan

Begin with any square established in perspective at

the appropriate scale, location, and angle of view.
The circle defines a plane, and the vanishing line for
this plane must be available, either as the principal
point and diagonal vanishing point or controlling
vanishing point(s). A diagonal vanishing point is only
necessary to define the square in depth. Then:

1. In the plan (projection square) and image square,

define the full diagonals, ab and matching.

2. In the plan and image square, divide the square

half by a perpendicular line (plan) or orthogonal to
the principal point (image) through the intersection
of the full diagonals. Divide again by a perpendicular
horizontal line (plan) or transversal (image). Mark
the four points where these lines intersect the
square, c and d and the points opposite (black).

3. Construct two quarter diagonals, c to d and

matching on the opposite side. Construct the two
quarter orthogonals from the principal point through
the intersection e and its twin.

4. From the intersection of quarter and full diagonals

(e and matching on the opposite side), construct a
horizontal line to the side of the square. This
intersects the circle at f and matching on the
opposite side.

5. Mark the intersection of the circle with the full

diagonal, at gand on the opposite side.

6. With eight vertical lines, carry the eight

points d, b, a, g, f, eand matching up to the
projection line, then project to the principal point
with orthogonals. Using the intersections of
orthogonals with the image square diagonals,
identify the points within the image circle; then use
orthogonals and transversals to reproduce matching
points at other corners.

7. Finally, join the twelve points, freehand or with a

French curve, to produce the circle.

Note that a and b are already defined in the image

square; dcan be located with an orthogonal from the
principal point throught the image diagonal
intersection; and the projection line from e (and the
matching point) can be found in the image square by
a transversal from the diagonal intersection of the
projection line from f, through the circle to the
opposite diagonal. Consequently only four projection
lines are necessary — from f and g and the matching
points on the other side — as shown by the pink lines
in the diagram.

Circle with Trigonometric Ratios. Finally, an even

more accurate method for projecting a circle was first
used by Paolo Uccello, as described below). It is
based on a close partitioning of the circle's
circumference into 32 equal segments, which
however makes the projection task more efficient.
 uccello's method of projecting the circle

1. Divide the circle plan with 16 equally spaced

"spokes"; these are found by dividing the circle into
quarters by a horizontal and vertical line intersecting
at the circle center, then bisecting the two upper 90°
angles 3 times.

2. Carry the 17 intersection points to the projection

line, and project by orthogonals to the principal
point.

3. Mark the intersection of each orthogonal with a

major diagonal of the image square (magenta
points). From these points construct 15 transversals
across the image square.

4. Identify the point intersections of each transversal

with the matching orthogonals on both sides of the
square, and the two point intersections of the central
orthogonal with the front and back of the square.
Connect these 32 points to form the circle.

The elegance of the original plan bisection is that

each projection point stands for both a horizontal and
vertical location of the circumference points; the
artist simply locates the intersection of the projection
orthogonals with a major diagonal of the image
square and the transversals duplicate these locations
on opposite sides of the circle, creating the 32 points
that define the circumference.

The location of these points along the width of a

projection line, of unit length 1.0, is derived from the
cosine of the angle of each "spoke" to the direction of
view. The sequence is tabulated below for reference.
 unit ratio locations
for 16 spoke division of circle circumference
(0.5 = midpoint of unit distance)
proportion of
spoke no. angle to DOV
unit dimension
1 -90° 0.000
2 -78.75° 0.009
3 -67.5° 0.038
4 -56.25° 0.084
5 -45° 0.146
6 -33.75° 0.222
7 -22.5° 0.309
8 -11.25° 0.403
9 0° 0.500
10 11.25° 0.597
11 22.5° 0.691
12 33.75° 0.778
13 45° 0.854
14 56.25° 0.916
15 67.5° 0.962
16 78.75° 0.991
17 (=1) 90° 1.000

To scale this series, just multiply by the unit length

and measure from one end of the dimension. Thus,
to project a circle from a projection line that is 20 cm
wide, multiply by 20: then the -45° angle (spoke #5)
is located 2.9 cm from the dimension end.
Ellipse Construction. Now an important constant in
perspective construction is that:

A circle in physical space always appears as an

ellipse on the image plane, except when it is viewed
edge on.

This means we can circumvent the whole rigamarole

of partitioning a plan and projecting it into
perspective space: let's just construct the ellipse
directly on the image!

Every ellipse can be described by its height and

width dimensions, known as its major axis (widest
dimension) and minor axis (perpendicular to the
major axis). This leads to two simple methods for
ellipse construction and also a calculation to estimate
the foreshortening of a circle.
The diagram (right) shows how to construct an
ellipse from fixed height and width dimensions. In
the first method (A), the height and width define a
rectangle, which is then divided into equal quadrants
by two lines. Then one interior horizontal line
segment and one exterior vertical line segment are
divided into proportionately equal parts, creating
proportionately spaced points. (The points do not
have to be equally spaced, only equal in their
proportional spacing on the two lines.) Lines are
drawn from the two midline points, a and b, through
the respective points, as shown. The intersection of
matching lines defines a point on the ellipse in one
quadrant. The landmark points are joined by a
freehand curve or segments of a French curve, then
traced or copied into the other three quadrants.

The alternative and more efficient trammel

method (B) is to define the ellipse rectangle, then
transfer the length of the major and minor axes,
aligned at one end, to a strip of cardboard or heavy
paper (diagram, right). Because the major and minor
axes are unequal in length, there is an interval
between their end points at the other end (magenta
line). These end points are aligned with the minor
and major axes of the ellipse rectangle, and the
circumference of the ellipse marked off from the
aligned end points at the other end of the card. This
method is quick, although it becomes much less
accurate as the major and minor axes become equal
(the ellipse approaches a circular shape). three methods for constructing an
ellipse of fixed height and width
The third method (C) uses two concentric circles,
centered on point a; the two circles are divided into
quarters by perpendicular lines defining the major
and minor axes of the ellipse. An arbitrary number of
lines are drawn radially through both circles from
point a, creating pairs of points at the intersection of
the line with the inner and outer circles. Then lines
are extended from these points, parallel with either
the major or minor axes of the ellipse; their
intersections define points on the ellipse in one
quadrant. An advantage of this method is that, by
extending the "spokes" and the horizontal and
vertical construction lines completely across the
larger circle, the entire circumference can be
identified.

However, there is a problem. The circle

construction diagrams above show that the center
of the ellipse is not coincident with the center of the
image square (the intersection of the image
diagonals), because recession causes the back half of
the square to appear somewhat smaller than the
front half. Thus, the black cross identifying the
center of the ellipse is not located at the diagonal but
somewhat below (in front of) it.

This is the same difficulty that produces the visual

discrepancy between the visible circumference of a
sphere (equal to the image width of the major axis of
the ellipse) and the visual angle of its diameter
(equal to the image width of the perspective square
across its center). This problem is examined in the
section on projecting a sphere. Unfortunately,
there is no simple way to scale the width of the
ellipse, other than making a scale drawing in plan, as
the major axis is not coincident with the midline
transversal of the square, and the points where the
ellipse touches the square envelope are typically not
on the major axis of the ellipse. But for perspective
circles within a 20° circle of view, the discrepancy is
so tiny that it can be ignored.

This is a principal reason why architects traditionally

used ellipse templates and now rely on computer
drafting programs. The templates contain a very
large number of ellipse cutouts, each slightly larger
than the last, all scaled to a standard angle of
view onto the plane surface containing the circle. The
artist simply chooses the template angle that
corresponds best to the proportions of the major and
minor axes of the ellipse required, then chooses the
cutout that is closest to the right image size.

The method for estimating the foreshortening of a

circle (the ellipse template angle) derives from
the trigonometric tangent within the circle of view
geometry:
Given a perspective square located near the median
line, draw the vertical line A from a front corner, and
the horizontal line Bfrom the opposite back corner;
these lines intersect to form a right triangle. Using a
ruler, measure the lengths of A and B and find the
arctangent of their ratio. This is the angle of view
onto the plane surface of the square at point x. This
angle is used to identify the most suitable ellipse
template.

In the illustration (counting in pixels), A = 101

and B = 187, so the angle at x is approximately the
arctangent of 101/187, or 28.4°. The formulas
quoted in the section on distance & sizecalculations
allow you to use this angle to infer the radius of the
circle of view that contains the square, and the
object distance (X) of the center of the square from
the viewpoint. For ellipses turned at an angle due to
perspective distortion (see below), the ellipse should
be enclosed by a rectangle tilted to the same angle;
the tangent is found from the height to width ratio.
Architects do not bother with any of this: they just
try one or another template until they visually
discover the best match in angle and size.

Photoshop note: Because a circle is an ellipse whose

major and minor axes are equal, any ellipse is just a
circle image compressed along one dimension.
The Ellipse Marquee Tool can be used to
approximately define the ellipse outline, and once
this is colored in it can be adjusted to an exact fit by
horizontal and/or vertical compression with the Free
Transform tool.

Perspective Distortions (Reprise). It should not

be surprising, if perspective "distortions" are in
fact accurate perspective images and the circle
construction methods create accurate image circles,
that constructed circles will display perspective
distortions. So we have to reprise the issue of
distortions and how to deal with them.

The example below is an extreme case, but if you

compare it to the appearance of spherical
forms similarly displaced from the direction of view,
you will see it is no worse than expected.
 perspective distortion in a 1PP
constructed circle

This is not just an artifact of the 1PP perspective: if

we use a 2PP geometry the shape of the square
containing the circle is improved noticeably, but the
circle is still strongly tilted. (The reduction in
elongation and size is due to the fact that the image
square is closer to the principal point dvp and to the
horizon line.)
 perspective distortion in a 2PP
constructed circle

Some perspective tutorials advocate the radical

solution of drawing an image circle in all situations as
an ellipse with major axis parallel to the horizon line.
Robert W. Gill, in an otherwise sensible perspective
tutorial, claims that the normal perspective
distortions of circles "are contrary to the laws of
perspective" — which is flatly incorrect. The
problem is that perspective distortions are
sometimes contrary to a pleasing image.

"Pleasing" is a practical rather than geometrical

issue, so the practical question is (for example) how
an ellipse, with major axis parallel to the horizon
line, can be fitted into the geometrically correct 2PP
image square shown above. This is only possible if
the ellipse does not touch the square on one or two
sides at the same time that the ellipse is grossly
flattened. Gill evades this difficulty by standing the
columns in his illustrations flat on the ground plane:
but most architectural columns rest on a square base
or plinth, or are surrounded by square tiled floors, so
the proportions and shape of the cylindrical column
and square base must correspond. To accommodate
an acceptably rounded ellipse shape that touches the
four sides of an image square base, the perspective
shape of the square base must also be "adjusted" by
reducing its visual width. But now the column and
base are no longer in scale to the architectural
elements around or behind them, so these too must
be adjusted ...

In effect, all these adjustments are incremental steps

toward shifting the diagonal vanishing points, and
with them the 2PP vanishing points, farther apart. So
the appropriate solution for this problem is
the classical remedy for perspective distortions:
reduce the circle of view contained within the
image format or (equivalently) increase the distance
between the principal point and diagonal vanishing
points, or (equivalently) increase the object
distance in perspective space.

If you are using an ellipse template, the major axis

of the ellipse should be aligned either with a line to
the opposite diagonal vanishing point (in 1PP) or at a
slightly less tilted angle than a line to the opposite
vanishing point (in 2PP). I find an arc drawn from the
opposite vanishing point, from the center of the
ellipse to the horizon line, reasonably locates the
direction in which the minor axis should be oriented.

Projecting Complex Plane Figures. A wide range

of more complex plane figures can be handled by the
square or rectangular projection, and the method
of distance point projection is the foundation
method in these cases.

However, the distance point procedure of drawing

arcs to identify the diagonal projections for every
point in the figure quickly becomes cluttered, or
requires a large working area. The more compact
solution is to use the diagonal contained within the
projection square as the depth projection
mechanism, and project everything from the plan
using only orthogonals to the principal point (or, for
2PP or 3PP drawings, the appropriate two measure
points).
 projecting a pentagonal plan into
perspective space

The method of diagonal depth

projection or rabattement is elegant and simple:
draw a square around the form you want to project;
draw a diagonal across this boundary; establish the
diagonal twin for all key points; carry each point and
its diagonal twin forward to the measure bar; project
the points from the measure bar into a square in
perspective space with a diagonal carried to the
diagonal vanishing point (dvp); establish the key
points in perspective by construction.

In the example above, we want to project the plan of

an irregular pentagon into perspective. We first scale
and rotate the plan, as described below, to the
correct orientation and dimensions. Then we enclose
it in a square, and draw a diagonal through the
square. (Note that we don't have to exactly center
the form within the square, and in fact it is
the diagonal, and not the enclosing square, is the
essential component of the method. However, it is
handy to put required points on the sides of the
square, if possible, to eliminate one or more of the
projecting orthogonals.)

For each key point needed to construct the form, we

first carry a horizontal line over to the diagonal, then
two vertical lines, from (1) the original point and (2)
its intersection with the diagonal line, up to
the projection line. Thus, starting with point a, we
carry a horizontal to the diagonal at x, then verticals
from a and x to the projection line.

From the projection line, we project all the points

back to the principal point (pp). We also project
to pp the width of the square. Then, using the
diagonal vanishing point, we construct the image
square and, within the square its major diagonal.

Finally, for every point intersection with the plan

diagonal, we construct a transversal from its
intersection with the image diagonal. Thus the
orthogonal for the plan diagonal point xintersects the
image diagonal at point x', which gives us the
recession. A transversal from x' intersects the
orthogonal from aat the perspective location of
point a'. The same is be repeated for each key point,
except that orthogonals from points on the front or
back of the square (such as b) require no
transversal, and points on the sides of the square
(such as c) require no separate orthogonal.

Provided that the vanishing points have

been accurately rotated in relation to the 90° circle
of view (the principal point and dvp's), these
procedures work exactly the same in 2PP, and the
2PP vanishing points are not at all required to project
the figure in perspective. In fact, any number of
forms can be projected into the same perspective
space using the same diagonal depth method of
projection, and their vanishing lines relative to each
other will harmonize exactly.

Finally, and most usefully, once a plane figure has

been projected into perspective space, a line
extended from any of its sides to the vanishing line
for the plane that contains it (e.g., the horizon line
for figures in the ground plane) identifies the
vanishing point for that side and all physical lines
parallel to it (diagram, above).
 projecting a street map into perspective
space

North Tribeca historic district, from New York Historical

Society map

For example, in the irregular street alignments

typical of many premodern city plans, a detailed
street map of the area can be projected onto the
ground plan, and this image street layout used to
define the 2PP vanishing points for the horizontals of
the various buildings (pink lines, above).

Alternately, a metric grid can be projected onto the

image plane at the appropriate spacing and
perspective depth (blue lines, above), and the map
copied into the grid square by square, with diagonal
depth projection used to trace out the contours or
locations of difficult problems, such as the traffic loop
in the right foreground.

perspective of complex solid forms

The strategy for projecting complex solids is
essentially the three dimensional extension of the
strategy for projecting two dimensional figures. The
complex form is enclosed in or reduced to cubes or
rectangles, and/or the grids or diagonals they define,
and the object is reconstructed from the landmark
points defined.
I think most artists remember their astonishment on
first encountering the "wire frame" perspective
drawing of a chalice by Paolo Uccello (right). In this
case the complex construction was achieved by the
painstaking accumulation of simple perspective
tasks, and in tribute I summarize them here:

1. The plan of a square for the base of the cup was

constructed in perspective space. From the ellipse
ratios evident at the top and bottom of the chalice, I
find that Uccello used a distance point (viewing
distance) of approximately 8 times the height of the
cup (e.g., the chalice is contained within a 7.2° circle
of view). Thus, if the drawing is actual size (29 cm
high), the viewing distance would be about
2.3 meters; and the base of the cup about 58 cm
below eye level.
perspective drawing of a chalice

2. Separately, the plan of a circle was bisected, then Piero Uccello (c.1450)
quartered, and then each segment bisected again
three times, resulting in 32 equal divisions of a circle.

3. The intersection points were brought by vertical

lines to the projection line, then projected by
orthogonals into the plan of a square in perspective
space. Note that the bisection method has produced
intersection points that are mirror symmetrical both
horizontally and vertically, so all that is required to
reconstruct the square is the intersection of each
orthogonal with the diagonals of the square
(see diagram, above).
4. This square projection was repeated over sixty
times, each time at a slightly different scale and
vertical location, to form the principal circumferences
of the cup. The vertical spacing of the squares was
accomplished with an elevation drawing of the cup,
or equivalently by physical measurement; and the
horizontal spacing by measurement.

5. The landmark points were connected horizontally

to define the circumference edges, and vertically to
the matching points in the circles above and below to
define the cup surface.

Small misalignments and changes in line weight

suggest the finished cup drawing was assembled
from two or three component drawings; this implies
that the drawing we have was transferred, by pin
pricks, from other drawings, or is a scaled down
version of drawings done in a larger format for
accuracy. The whole project must have taken weeks
to complete.

In that context, it is interesting to hear Giorgio

Vasari's comments on Uccello's consuming
perspective studies:

"Paolo Uccello would have been the most gracious

and fanciful genius that was ever devoted to the art
of painting, from Giotto's day to our own, if he had
labored as much at figures and animals as he labored
and lost time over the details of perspective; for
although these are ingenious and beautiful, yet if a
man pursues them beyond measure he does nothing
but waste his time, exhausts his powers, fills his
mind with difficulties, and often transforms its
fertility and readiness into sterility and constraint,
and renders his manner, by attending more to these
details than to figures, dry and angular, which all
comes from a wish to examine things too minutely;
not to mention that he very often becomes solitary,
eccentric, melancholy and poor, as did Paolo Uccello.
This man, endowed by nature with a penetrating and
subtle mind, knew no other delight than to
investigate certain difficult, nay impossible problems
of perspective, which, although they were fanciful
and beautiful, yet hindered him so greatly in the
painting of figures, that the older he grew the worse
he did them. ... For the sake of these investigations
he kept himself in seclusion and almost a hermit,
having little intercourse with anyone, and staying
weeks and months in his house without showing
himself." [Lives of the Painters, Sculptors and
Architects, 1550; "Paolo Uccello, Painter of
Florence"]

A cautionary word across the centures for the many

modern perspective dabblers, including digital
rendering engineers, who spend days or weeks on a
single texture map or illumination model. (Those
solitary melancholics who spend months porting it all
onto an obscure web site, well ... they are exempt
from caution.)
Projecting a Sphere. The sphere and its related
geometrical forms the cone and cylinder present a
subtle difficulty. On the one hand, they are all
circular in cross section and therefore, in most cases,
can be represented by an elliptical outline along the
front edge or circumference. On the other hand, they
are solids rather than plane figures, which produces
specific problems of image scale and foreshortening.

Sphere Image Scale. The scaling problem is that a

sphere relatively close to the viewpoint presents a
visual angle that is larger than the visual angle of its
physical diameter. That is, using a measure bar or
unit distance to project the diameter of a sphere in
perspective space will underestimate its actual
apparent size. The diagram shows why: the angle of
view to the limb of the sphere is in front of the
diameter, at its visible circumference.
 discrepancy between the visible
circumference and angular diameter of a
sphere

sphere shown at an object distance 1.4 times its

diameter

It's odd that this problem gets extensive treatment in

some perspective handbooks without asking the
question: does the discrepancy matter? For a sphere
at an object distance (ground plane distance) from
its center to the viewpoint that is 2.5 times its
diameter, the angular diameter of the sphere is
22.6° but its visible circumference is 23.07°. This is a
discrepancy of about 0.5° or the visual width of the
full moon (1 centimeter at 1.15 meters). That
probably matters.
For a sphere at an object distance 5 times its
diameter, the visible discrepancy is about 0.05°, or 1
centimeter at 10 meters; for a sphere at 10 times its
diameter, the object distance recommended by
Leonardo to reduce perspective distortions, the
discrepancy is 0.007°, which is equivalent to 1
centimeter at 80 meters and is below the optical
resolution of the human eye or an equivalent
camera. Thus, I'd suggest that the problem can be
ignored for a sphere, cylinder or cone at an object
distance more than 5 times its diameter: the visual
discrepancy is then less than 0.5% (e.g., 1/2 mm in
10 cm), which is smaller than the random variations
introduced by drawing inaccuracy.

For anything closer, the scaling problem can be

handled by (1) constructing the measure bar for the
diameter of the sphere from a plan drawing (such as
the drawing above) that reproduces the sphere
diameter/object distance proportions; or (2)
rescaling a measure bar that is based on the image
size of the sphere to reflect the visible circumference,
using the proportions in the table below.

measure bar correction

for spherical/circular image width
Angular
OD*/Sphere Diameter VC*/AD*
Diameter (AD*)
[VC greater than
0.5 90°
2.0]
1.0 53.1° 1.129
 1.5 36.9° 1.056
2.0 28.1° 1.031
2.5 22.6° 1.020
3.0 18.9° 1.014
3.5 16.3° 1.010
4.0 14.3° 1.008
4.5 12.7° 1.006
5.0 11.4° 1.005
Note: OD = object distance, viewpoint to center; AD =
angular diameter; VC = visible circumference.

Sphere Image Foreshortening. The second

problem is the foreshortening of the circle/ellipse
used to represent the sphere. Because the sphere is
visible in depth, its apparent diameter
undergoes rotation foreshortening, which causes
its circular outline to appear elliptical when the
sphere lies to one side of the direction of view. The
axis of maximum elongation is always approximately
radial from the principal point, so the sphere may be
vertically, horizontally or diagonally elongated,
depending on its position in relation to the direction
of view.

Yet there is nearly universal consent that spheres

should be represented as circles in a perspective
drawing; Robert W. Gill provides the most detailed
defense of this solution but it is common practice. In
fact, I have never found a perspective text that
explains the correct way to draw the true central
projection of a sphere.

Reasoning from the basic rules of perspective lets

us develop a correct procedure. Start with the fact
that a sphere can always be enclosed by a cube,
whose width is equal to the diameter of the sphere,
so that the sphere is touching the center of each face
of the cube. This perspective cube can be viewed
from any angle or orientation in physical space; the
sides of the corresponding image cube will recede to
their 1PP, 2PP or 3PP vanishing points as the angle of
view toward the cube may require.

As a demonstration example, the 2PP circle will be

used as the plan of the sphere we want to construct.
It is necessary first to construct the perspective
cube, as diagonals across the interior of this cube
locate the center of the sphere we want to construct,
and diagonals across the base locate the point where
the sphere touches the ground plane, the perspective
image of its object distance. The measure bar for the
front face of this cube is also the diameter of the
sphere we want to construct.
 perspective drawing of a sphere

construction of the perspective cube used to locate the

sphere center and ground plane distance; also shown is
a guesstimate of the circular profile of the sphere it
contains

Now the sphere inside the perspective cube, in

physical space, appears to have an unchanging
circular profile regardless of the angle of view to the
perspective cube. To justify this unchanging
appearance, imagine a projection plane that (1)
passes through the center of the sphere and (2) is
always perpendicular to the line of sight from the
viewpoint to the center of the sphere. On this plane,
the outline of the sphere will always appear as a
perfect circle, and will always be enclosed in
a projection squarewhose bottom edge can
(arbitrarily) be made parallel to the ground plane.
However this projection square is not a cross section
through the perspective cube, because that might be
a rectangle, trapezoid or irregular hexagon,
depending on the orientation of the perspective cube
to the viewpoint. Instead, the projection square is
the central cross section of a projection cube that
has the same dimensions as the perspective cube but
is oriented so that its front face is perpendicular to
the line of sight to the center of the sphere and its
horizontal edges are parallel to the ground plane
(diagram, right). This projection cube and its cross
section, the projection square, will have vanishing
points different from the perspective cube and its
plan. sphere inside a perspective cube
and a projection cube
 true perspective drawing of a sphere

locating the vanishing points for the projection cube

The vanishing points of the projection cube (and the

projection square) are found with the 3PP methods
for exact rotation of vanishing points, and by
deduction from the given orientation of the
projection cube (diagram, above):

• The vanishing point for the horizontal top and

bottom edges is found by rotating a visual ray from
the viewpoint, folded to a vertical diagonal vanishing
point, to the horizontal (left or right) displacement of
the center of the sphere from the principal point; this
is the intersection of a vertical line
(perspective rule 8) from the center of the sphere to
the horizon line. Then the vanishing point we want
(vp1) is on the horizon line, 90° to this visual ray.
• The receding side edges of the cube are parallel to
the line of sight to the center of the cube (because
the front face of the cube is perpendicular to the line
of sight), so their vanishing point is the center of the
sphere (perspective rule 5).

• The upright sides of the projection cube are parallel

to a plane that contains the line of sight and is
perpendicular to the ground plane. Therefore their
vanishing point is in the vanishing line for this plane,
which is the vertical line from the center of the
sphere. The vanishing point is located as described
here, and a visual ray is rotated to the vertical (up
or down) displacement of the center of the sphere
from the horizon line; the vanishing point (vp2) is on
the sphere centerline 90° to this visual ray.
 true perspective drawing of a sphere

scaling the measure bar for the sphere diameter, and

projecting to measure points

The dimensions of the projection cube are found

from the measure bar used to define the sides of the
perspective cube (diagram, above):
• Orthogonals are used to project the original
measure bar (magenta line) to the image depth of
the center of the sphere (green line).

• The measure bar is centered on the center of the

sphere (blue line).

• The length of the measure bar is rotated to parallel

with the front face of the projection cube by
vanishing lines to the measure point for the
horizontal vanishing point. The measure bar is
projected onto a line from this vanishing point
through the center of the sphere. Note that the
projection is backward and forward in perspective
space, depending on the horizontal tilt of the
projection cube to the image plane.

• The measure bar is rotated 90°, and its vertical

dimensions are projected to the measure point for
the vertical vanishing point, to correct for the vertical
tilt of the projection cube to the image plane.

• The measure bar has defined four points: these are

the four sides of the projection square that are
tangent to the enclosed circumference of the sphere.
These dimensions can be rescaled, if necessary, to
account for the larger visible circumference of the
sphere. The measure bar (the diameter of the
sphere) in the example problem is 1.2 meters long,
and (based on the image height of the point where
the sphere rests on the ground plane) the center of
the sphere is 3 meters from the viewpoint. So, using
the table above, the dimensions can be increased
by 3%.

• Vanishing lines from the two vanishing points,

through the four side points, are used to complete
this square. This is the perspective image of the
projection square.

• Vanishing lines are used to perform the "planless"

squareconstruction method or another more exact
method if necessary.

• The projected points are connected as an ellipse to

form the visible circumference (or visual diameter) of
the perspective sphere. The diagram (below) shows
the finished drawing.
 true perspective drawing of a sphere

constructing the sphere profile from the "planless"

square projection method

I've pursued this digression for four reasons. First,

I've verified by example that the correct perspective
image of a sphere is not an ellipse.

However, the amount of elliptical distortion, even for

a very large, closely placed sphere far to the side of
the direction of view, appears much smaller than it
does in a ground plane circle at the same location.
This (and the complexity of drawing a sphere the
"right" way) provides justification for the practice of
using a circular outline to represent a sphere, as has
been customary and wholly acceptable perspective
solution since the Renaissance.

Third, my perspective solution suggests why it is that

circles can be acceptable images of spheres in
perspective images. In effect, spheres define no
vanishing points in visual experience: they only
reflect the viewer's central recession in their image
size. We artificially introduce vanishing points by
constructing a projection cube around the sphere,
and this cube is always in 3PP, whatever may be the
vanishing points of the perspective cube.

Many attributes of the sphere — the lack of linear

elements on the sphere's surface, the unvaryingly
equal dimensions of width and depth from every
point of view, the typically small size of physical
spheres in everyday experience, and the optical
equality of the paired images in binocular vision —
are quite unlike the linear edges, large physical
extent and binocular disparity that define many
"linear" perspective examples. As a result,
our habitual visual concept of spheres is different
from the recession and depth convergence we
associate with railroad tracks, fences, streets,
buildings and other typical perspective themes. The
point is that perspective involves drawing what we
know (or what we think we see) rather than what is
a geometrically correct projection onto an image
plane: this problem is at the heart of all perspective
"distortions".

Finally, I've demonstrated the power of the the basic

rules of perspective, combined with the 90° circle of
view and the explicit rotation of vanishing points and
measure points, to solve novel and complex
perspective problems.

Projecting a Cylinder. In most perspective

constructions, cylinders are columns, and columns do
not present unusual foreshortening problems
because the circular base of the column is defined by
the enclosing square, and the column is
perpendicular to the ground plane.
But if the column tips over, or seems about to — like
the Tower of Pisa (right) — then we have to find the
angle of its base to direction of view, and from that
construct the circle foreshortening, in this case to
find the circumference of each level of the tower.

the tower of pisa

 perspective drawing of the tower of pisa

showing rotations for image scale and vertical angle,

and two circular constructions

This drawing is made by first establishing, from a

photograph or accurate plan and elevation, the
necessary measurements. If the angle of the tilt is
perpendicular to the direction of view, then the tilt is
at an angle of 5.5° from vertical. Assuming an object
distance of about 75 meters, the 56 meter high
tower would span a vertical visual angle of 36°.
(Other tower dimensions, such as diameter, are not
considered here.)

To model the tilt, the median line and horizon line

are rotated around the principal point by a 5.5°
angle, to produce a new horizon line (magenta) and
a new median line, which is now the axis of the
tower cylinder.

Next, a 36° angle is rotated from one of the side

diagonal vanishing points to locate the vertical
dimension of the tower image. I have done this from
the original horizon line to the original median line,
assuming that the tower height measurement was
true vertical. If the measurement were along the axis
of the tower, the rotation would be done from the
"tilted" dvp to the tower axis.

A measure bar is used to find the vertical location of

each tower level along the axis; two examples are
shown for the top platform (a) and a middle level
(b). If the points are scaled at the distance of the
front side of the tower, they will be on the front face
of the perspective square; if they are scaled to the
distance of the center of the tower, they will lie on
the tower axis and be at the diagonal center of the
perspective square.

In either case, the perspective square is constructed

from the height point, using diagonals to the
tilted dvp's (blue lines). Thus, diagonals
from b define the front half diagonals of a
perspective square. A measure bar for the tower
width (tilted perpendicular to the tower axis and
centered on b) defines the front corners of the
perspective square (n and o); orthogonals from
these points to the principal point (dv) define the
square sides. A second diagonal from the intersection
of these orthogonals with the original diagonals
to b define side midpoints (e.g., at r); diagonals
from these points intersect at the back side of the
square (at s). (Alternately, diagonals
from nand o intersect the orthogonals at the back
corners of the square.) A line through s and parallel
to no defines a perspective square section. Finally,
the front circumference of a circle is projected into
this square using any of the methods described
above; given the number of diagonals already
constructed, the circle without a plan method
might be most efficient.

Projecting a Spiral Staircase. The Tower of Pisa

example tackles the tilt of a cylinder but left out the
vertical scaling of the tower levles, which is done
with a measure bar or elevation (side view) of the
tower. Spiral staircases, although they almost never
appear in a drawing, are hoary perspective clichés
and a good example of how elevation and plan are
combined to project a complex object in three
dimensions.
perspective drawing of a spiral staircase

using the Uccello method of circle projection, and

transversals to locate the stairs in depth on an
elevation
There is really little to explain. The plan view is
simply the Uccello format for projecting a circle,
which represents the outer edge of the stairs. The
elevation is constructed by carrying the stair
locations at each level to the side with transversals,
then projecting these in depth with orthogonals to
the principal point (or, in 2PP, to the controlling
vanishing point).

Projecting a Cone. Finally, I demonstrate the

procedure with a cone, whose axis can equivalently
be the axis of a cylinder.
The easy problem is when the cone stands with its
base on or parallel to the ground plane (diagram,
right). In architecture this occurs, for example, in the
roof of a circular tower, silo or minaret. Then the
base is defined by the square parallel to the ground
plane enclosing its circle; the apex is on the
perpendicular axis drawn from the diagonal center of
the square.

The most complex case is when the axis of the cone

is at an angle both to the ground plane and the
direction of view. In the example, the cone has a
a cone with base parallel to the
base diameter of 1 meter and a height of 3 meters, ground plane
is lying on its side in the ground plane, with the axis
at a 30° angle to the direction of view, at an object
distance of 4 meters. The completed construction is
shown below.
 perspective drawing of a reclining cone

using the method of vanishing point rotation, horizon

line rotation and measure points

The first step is to establish the vanishing point

framework, since this is necessary to scale the image
size.

Two angles are involved. If the base of the cone

were exactly parallel to the direction of view while its
axis were parallel to the ground plane, the base of
the cone would be perpendicular to the ground plane.
If the cone is lying on its side, the base would define
a visual tilt of 9.5° (1/2 the interior angle at the apex
of the cone). This would simply require a
corresponding tilt in the horizon line and median line
around the principal point (as for the tower example,
above). However, the base is actually at a 60° angle
to the direction of view, so the 9.5° angle is
foreshortened by this angle.

This is solved in two steps: (1) rotate the vanishing

points around the viewpoint (a vertical dvp) to
obtain the 2PP framework for the base, and then
(2) rotate the vp's around the principal point, to
obtain the tilt caused by the cone lying on its side.
(The steps can be performed in reverse order if
desired.)

Alternately, the 9.5° angle can be marked from the

base to the top of a rectangular solid, and the cube
projected into 2PP perspective space with the
required vanishing point rotation (see next section).

Next, the measure points are defined in the usual

way, as arcs drawn around the two vanishing points
to the rotated horizon line.

Third, the measure bars for the cone height and base
width are defined using the procedures for scaling
the drawingdescribed earlier.
Using the measure points, a rectangular solid that is
3 unit dimensions high and 1 unit dimension square
is projected into the perspective space.

Diagonal lines are used to find the center of the far

square face, which is the apex of the cone.

Finally, the "planless" method is used to construct

the elliptical base of the cone within the square base
of the rectangle at the opposite end. If an ellipse
template is used, the major axis of the ellipse is
usually aligned perpendicular to the axis of the cone.

The same method is used to construct a cylinder at

an oblique or acute angle to the image plane, ground
plane and/or direction of view. The only difference is
that a circular or elliptical circumference is
constructed at both ends of the rectangular solid.

Projecting Complex Solids at a Compound

Angle. I use as my examples two of the Platonic
solids, which were among the first perspective
challenges taken on by Renaissance draftsmen.

We've already been working with one of the Platonic

solids — the hexahedron or cube — and the cube (or
a rectangular solid) can be used to project complex
solid forms in the same way a square is used to
project complex plane figures.

Octahedron and Diagonal Centering.

The octahedron is a regular polygon with eight faces
and six vertices (corners). The eight faces are
equilateral triangles joined at an angle of 109.5°,
which is inconvenient to measure through multiple
vanishing point rotations. In all these situations, the
projection cube/rectangle comes to the rescue.

perspective drawing of an octahedron

using the method of diagonal centering, in the 60°

circle of view

The example is straightforward. A cube is projected

into 2PP space, using a measure bar taken at full
length for the height of the cube and projected to the
appropriate measure point to define the
foreshortened faces of the cube.

Diagonal lines are used to define the perspective

center of the image squares. These locate the
vertices of the octahedron; the points are simply
joined for all front faces of the form.

Dodecahedron and Layered Projection. The

dodecahedron is a regular polygon with 20 vertices
and 12 pentagonal faces, each at an angle of about
116.5° to the five adjacent faces. Althougth the
vertices all intersect the surface of a sphere, they do
not have any simple connection to the geometry of a
cube. Nevertheless, a projection cube can be used to
construct the perspective image; although for very
complex forms and drawings at modest scale, the
method requires professional drafting equipment to
be reliable.

The cube functions in two parts, a series of (in this

case) horizontal layers through the cube, each
showing a section of the form in plan at specific
intervals, and a vertical measure barthat defines the
separation between layers. The cube can just as
easily be divided into a series of sections or
elevations, registered with a horizontal measure bar;
the best strategy depends on the characteristics of
the primary form.

The plan is constructed first, as separate layers, and

the layers must be inspected to ensure they define
all the necessary significant points. If possible, the
primary form should be tightly enclosed by the
projection cube so that faces or corners of the form
are coincident with faces and/or corners of the cube;
this reduces the projection work. When several
layers or plans are used, each layer must be
enclosed by the same registration marks or cube
outline, so that layers will aligned exactly with each
other during the projection steps.

perspective drawing of a dodecahedron

constructing the vertical measure bar from the

dodecahedron elevation

The vertical measure bar is constructed from an

elevation of the primary form, which is cut through
at the levels containing the significant
points necessary to reconstruct the outlines, corners,
edges etc. of the form.
In this case, just as we have been doing with
perspective cubes, the significant points are the
vertices (corners), which define all the edges and,
with the edges, the faces of the form.

The vertices divide the cube into four

layers, a, b, c, d(diagram, above), with an added
interval x to indicate the distance between the base
of the dodecahedron and the base of the cube.

Note that the dodecahedron is oriented

symmetrically or regularly with the sides of the cube;
this should always be done, if convenient, with any
complex form, so that its orientation can be
manipulated entirely through the vanishing points for
the projection cube.
 perspective drawing of a dodecahedron

constructing the projection cube, with measure points

and diagonal vanishing point in the 60° circle of view

Next, the projection cube is constructed in

perspective space in the location, orientation and
scale desired for the dodecahedron object (diagram,
above). The procedure for constructing a 2PP image
cube is described here.
 perspective drawing of a dodecahedron

projecting the vertices in layer "a"

Now the projection of the separate plan layers begins

(diagram, above). The following steps are used for
each layer:

• The vertical measure bar is aligned with the anchor

point, and the level location (a in the diagram) is
marked off. Usually the best procedure is to work
from the layer closest to the viewpoint to the layer
farthest away, so that significant points that are
occluded or hidden by the front part of the form can
be omitted as work progresses.

• The level lines (green) are drawn from this point

to the vanishing points; these define the edges,
along the faces of the cube, of the layer to be
projected. The layer diagonals are drawn from
opposite edges of the cube where they are
intersected by the level lines.

• The projection bar is aligned level with the

location level (a).

• The appropriate plan level is aligned with the

projection bar (in the example, a square outline and
a centering "+" are used for the registration), and
the points to be projected — five vertices and three
diagonal depth points — are carried up to it with
vertical lines, where they define the projection
points. The accurate location and alignment of the
level location, level lines, projection bar and plan
outline are critical; in particular, the top face of the
plan square must be exactly parallel with the
projection bar, and the projection bar must be level
(for horizontal layers).

• The projection points (intersections of the vertical

lines with the projection bar) are projected onto the
level line (green) by lines to the appropriate measure
point (as the projection is onto the cube face whose
recession is defined by vp2, the correct measure
point is mp2). These lines intersect the level line at
the image points for their edge locations.

• The edge locations of the image points are

regressed to the appropriate vanishing point (vp1 in
the example) by vanishing lines (blue for vertices,
pink for diagonal depth points).

• Where the diagonal depth vanishing lines intersect

the level diagonal, those intersections are regressed
to the opposite vanishing point (vp2) by vanishing
lines.

• The corresponding intersections of vanishing lines

are used to locate the image vertices (orange
points).

It is evident from the diagram that each layer of a

complex form may require dozens of vanishing lines.
To eliminate erasure and clutter, it is useful to draw
each plan layer on a large sheet of drafting vellum or
tracing paper, oriented so that the projection cube
area is also covered. Then the entire sheet is laid
over the work area and taped taut in place; then the
level lines, projection lines and vanishing lines are
drawn upon it. When the significant points for that
layer are located they are marked with a pin prick
through the paper onto the drawing paper below.

The location of the points is confirmed with small

pencil points before the layer sheet is removed; then
the sheet is taken off and the additions to the
drawing are cleaned up, connected as edges, etc.
before proceeding to the next layer.

perspective drawing of a dodecahedron

projecting the vertices in layer "b"

The vertical measure bar is used to locate the next

layer position (b) and the projection bar is moved up
to be exactly level with it. Then the plan is aligned
below it and the projection steps described above are
repeated.

A significant drawing problem arises when the

projection layer is oriented in perspective space so
that it is seen nearly edge on: in the example,
level c is nearly on the horizon line. In those
situations the location of the points is defined by
vanishing lines that intersect at a very small angle,
introducing potentially large inaccuracies.

The solution is to create a second projection layer at

a distance far on either side — in the example, in the
base of the projection cube or even below it — and
locate the points horizontally by vertical lines from
their perspective location in this second projection
layer. These replace the vanishing lines to one of the
two vanishing points, and the diagonal depth points
and their vanishing lines can now be omitted, which
substantially reduces clutter. The vanishing lines to
one vanishing point and the vertical lines from the
second projection layer intersect at nearly right
angles, so that both the horizontal and vertical
locations of the points are accurately and clearly
defined.

This technique requires the image points to be

constructed twice, once in the second projection
layer and then again in the final image layer, and
this repeated projection is also a source of
inaccuracy.
 perspective drawing of a dodecahedron

the finished drawing in the 60° circle of view

After all the layers have been projected into the

image, any remaining construction lines are erased,
the points are connected, and the drawing finished
off. The image shows the projection cube in place, to
facilitate comparison with the octahedron drawing
above.
Projecting the Human Figure. Hands down, the
most difficult perspective problem artists have
tackled has been the human figure. It was also one
of the first to be tackled. A complex but precise
method is illustrated in Piero della Francesca's De
Prospectiva pingendi (c.1474), and rather crude but
efficient methods are explained in Albrecht
Dürer's Vier Bücher von menschlicher
Proportion (1528). Things really heated up during the
16th century, when all those ceiling frescos of saints
and angels soaring to Heaven required careful
analysis of human foreshortening (and the soles of
human feet). By the 17th century this stuff was
school study trailing in the wake of Tintoretto's
career.

The simplest method for transferring the figure into

perspective is to make a drawing from life, or trace a
photograph, that shows the figure in the correct pose
and from the correct point of view to match its
orientation in the master drawing. This figure study
is then scaled to the appropriate size and traced into
location.

perspective drawing using a viewing grid

the figure is copied square by square from the viewing

grid to a smaller grid on the paper, and this drawing is
then scaled to fit the master painting; from Dürer's Vier
Bücher (1528)
The more anal, rigorous method is to recreate the
figure by the three dimensional mapping of points
into perspective space. To my knowledge there are
basically three approaches in this tradition: (1)
sectional projection, (2) volumetric projection, and
(3) armature projection.

Piero used a sectional projection: he divided the

human head into parallel sagittal planes, projected
the key points for each section much as we've
projected the octagonal plan above, but spaced each
square vertically to match the anatomical separation
of the sections in space. This creates a "cage" of
points and the face is reconstructed simply by
translating the points back into facial features.

The second method, volumetric projection, first

analyzes the human figure into so many
interconnected eggs, cylinders, boxes or pyramids,
then projects the major corners or axes of these
simple forms in perspective, then reconstructs the
figure around them. This approach was popular in
the Baroque and even dribbles like a late party guest
into 20th century figure drawing and perspective
texts. I dislike it very much because it completely
destroys the tensile, articulated and rounded
strength of the human form. I feel an active schedule
of live figure drawing is a better solution to learning
the shape and heft of the body from various points of
view.
If you do have this basic understanding,
then armature projection is a very efficient method
to get the human proportions in perspective
projection. All you really need is one of those wooden
anatomical manikins sold in every art store.

projecting the human figure from an art

school mannekin

the 12 inch long Dick Blick hardwood manikin

The illustration shows the basics of the approach.

Arrange the manikin in the desired anatomical
position, then set it on a glass table top or projection
stand. Cast a shadow from the manikin onto a stiff
white card below the figure, using a ceiling light or
spot light placed as far above the set up as possible.
Mark the major joints on the card, using the shadow
as a guide.

Now place a spot light or desk lamp to one side of

the figure, at the same height as the figure, at right
angles to the major axis of the figure, and at the
same distance from the figure as the ceiling or
overhead spot. Firmy support a second stiff white
card behind the figure, at the same distance as the
previous card was below it. Mark the joints in the
same way.

Choose the card with the better spacing of the joints

as your primary face, and either trace the points
onto a sheet of graph paper or take measurements
directly from the card, from each point to one long
edge and to one end edge. Take a single set of
measurements from the second card to one long
edge. These measurements can be scaled, rotated
and transfered to a measure bar using the methods
described above, and from there projected into
perspective space. The foreshortened figure is then
reconstructed freehand around the joints.

I've explained this approach with a manikin, but it

really excels if you can take two perpendicular views
of a figure pose from exactly the same distance.
Measurements can be taken directly from the
photographs, using each one as the "card" on which
the image is projected. With computer image
processing software, such as Adobe Photoshop, you
can even distort and scale the images to match the
outlines of a predrawn rectangular solid in
perspective, then connect matching features in the
two photographs directly, without any measurement.

This armature approach is implicit in the series of

photographs of human and animal movement
created by Eadwaerd Muybridge. In many motion
series there is a side and front view, or two
complementary front and back diagonal views, taken
with exactly right angled directions of view. These
permit the two dimensional measurement from the
photograph of major joints and body dimensions
along the two sides of a square or rectangle (the
third dimension of vertical distances are the same in
both photographs, or can be scaled so). These points
can be projected into space within a rectilinear solid,
either using Piero's sagittal sectional method or the
armature method, and the body then can even be
viewed from any angle simply by rotating the
vanishing and measure points for the enclosing
rectilinear solid.

Of course, this whole discussion is moot. Artists now

can use software such as Poser to create male or
female "digital mannekins" in any pose, clothed or
unclothed, and render drawings or art from that
foundation; and a whole series of VirtualPose discs
are available that rotate static figure poses in two
dimensions. Programs for major animals are sure to
follow.

buildings from blueprints or plans

The most common application of linear perspective
starts with the elevation and plan of a building or
object, and transforms these into a three
dimensional perspective view.

Depending on the shape of the object or building,

one, two or sometimes three separate views are
necessary to construct it. Sometimes the plan view
(view from above) is necessary; a single side view by
itself is sufficient. If more than one view is used, the
views must be taken at right angles to each other.

I will use the blueprints for a detached commercial

greenhouse (shown below). I render this building
in two point perspective, which is both the most
common architectural convention and a relatively
straightforward method to use.
 blueprint used for perspective plan

The first steps are always to establish the scale and

fundamental proportions of the drawing. As
described in previous sections, this means (1)
choosing the image format or dimensions of the
drawing to best display the important shapes in the
image; (2) choosing the best viewing angle (frontal,
oblique) to show the important features and
proportions of the building; (3) adjusting the
apparent distance to the building along the angle of
view to create pleasing shape proportions within the
circle of view and format; (4) moving the viewpoint
up or down to establish an effective anchor point and
horizon line; and finally (5) locating the necessary
vanishing points and measure points to start the
perspective construction.

Image Format and Viewing Distance. I decide for

presentation purposes that I want a moderate scale
for the drawing or painting, and therefore choose to
work within a quarter sheet (11" x 15"). Given the
size of the sheet, we're assuming the viewing
distance to the drawing will be about two feet (24"),
which is slightly more than the normal distance for
reading a book (18") but much less than the normal
distance for viewing a painting (60"). That is, we
intend the drawing for close inspection rather than
grand effect. Other format sizes and proportions
would be more appropriate for other presentation
aims, display settings, media, etc.

Scale and Viewing Angle. From the blueprints, I

determine that the finished greenhouse will be built
to be 27 feet long, 25 feet wide and 14 feet high.
From these specifications I can define a scaling
form — a rectangle in the same proportions as the
plan outlines of the building (in fact, it can be a
tracing or full size photocopy of the blueprint plan
itself), approximately as large as the image format
with a diagonal line included.

The plan proportions are 27/25 or 1:1.08, and the

image format is 11" high (in landscape orientation),
so the scaling form is drawn to be 11.9" x 11" with a
diagonal included. This represents the plan
proportions of the building, and is shown as the
magenta rectangle in the diagram.
 constructing from a plan: dimensions and
layout

shown in a 60° circle of view

Next I turn or rotate this scaling form until the angle

of the sides in relation to the median line matches
the angle of view on the building that you want. In
other words, I twist the magenta rectangle left or
right until I get the desired visual proportions in the
front face and side of the structure.

Once I have the angle of view to my satisfaction, I

choose a point on the diagonal of the scaling form
that defines the visual width for the structure that
best fits into the format proportions. In the figure,
I've chosen a point exactly 66% of the original
diagonal length of the scaling form. This width
provides room to include a rendering of the setting
around the greenhouse — paths, trees, sky, etc.
Now, by extending two new sides parallel with the
sides of the scaling form from this diagonal point, I
have a scaled plan (gray rectangle in the figure) at
the exact proportions to fit the format.

I move this scaled plan left or right until its

horizontal position in the format is where I want it,
then drop two vertical lines from the opposite
corners of the scaled plan to define the visual width
of the building. I drop a third line from the corner of
the scaled plan that represents the closest corner of
the building as it will be viewed in the final drawing.

As a check, I now determine the scale of the

drawing. I do this by measuring any side of the
scaled plan, then dividing that length by the actual
length of the building to be constructed. In this
example, the width (short side) of the scaled plan
turns out to be 7.26". The basic size and distance
proportions dictate that this drawing size divided by
the viewing distance to the drawing (18") equals the
actual width of the greenhouse (300") divided by the
viewing distance to the greenhouse. Doing the math
shows the greenhouse will be drawn as it would
appear from a distance of roughly 62 feet, in a
reduction of roughly 2.4% from actual size.

It is also useful to establish the scale of the drawing

in relation to the scale of the blueprints, so that any
measurements taken from the blueprints can be
directly converted into drawing measurements. (If
you just use a full size photocopy of the blueprint
plan as your scaling form, you've already done this
step when you created the scaled plan.) In this case
the blueprints are in the scale 1/2" = 1 foot, so the
width of greenhouse in the blueprint plan is 12.5".
The corresponding width of the scaled plan is 7.26",
which is a reduction of 58%. Now, for example, if I
measure a window width in the plan of 1", I can
immediately transfer this to the drawing as a window
width in reduced scale of 0.58".

Finally, I can establish the length of the anchor line:

it's 2.4% of the building height of 14 feet, or 58% of
the blueprint elevation height of 7", or roughly 4.0"
high. I then determine the point where the horizon
will intercept the anchor line based on the implied
height of the point of view. For example, if the
greenhouse is viewed as it would appear to an adult
standing on level ground, this height (the height of
the observer, or 68") is equal to a drawing size of
1.63" (2.4% of 68"), so the horizon line would
intersect the anchor line 1.63" from its bottom end.
Instead, I decide to take a slightly higher vantage of
about 8 feet (96"), as if the greenhouse were viewed
from a raised patio or shallow slope. That puts the
anchor point (the bottom end of the anchor line)
about 2.3" (2.4% of 96") below the horizon line.

The last step is to locate the horizon line in relation

to the top or bottom of the format. Start with the
horizon line through the middle of the format, and
diverge from that location for visual effect. Normally
an upward view (viewpoint close to the ground
plane) implies a low horizon line, as the direction of
view is toward the sky; a raised viewpoint implies a
high horizon line, as the direction of view is
downward.

In this case, even though I've chosen a slightly

elevated viewpoint, I also choose a horizon line that
is slightly below the horizontal midline of the image
format, to provide a view of the setting behind the
greenhouse and off into the distance, which gives a
feeling of open space and the outdoors. The point
where this horizon intersects the median line —
placed down the center of the format — is the
direction of view.

Circle of View and Drawing Impact. Because the

viewing distance to the drawing is 18", I have started
with the assuumption that 18" is also the radius of
the 90° circle of view at the image plane (the plane
of the drawing): so the circle of view is 36" or 3 feet
wide. As I have already established the median and
horizon lines, anchor point and anchor line, I could
proceed from here to draw the circle of view around
the principal point, use the scaled plan to rotate the
vanishing pointsaround the intersection of the
circle of view and median line, and from these
vanishing points establish the measure points,
and start the drawing.

What kind of visual impact does that circle of view

create? To find out, I divide the 62' object distance
by the 27' object size to get the ratio 2.3. Reference
to the circle of view tableindicates that this
distance/size ratio roughly a 25° minimum circle of
view for that object size at that distance. This is well
within the 40° maximum circle of view that keeps
extreme perspective distortions out of the drawing.

However, simply by enlarging or reducing the circle

of view from its appropriate 18" radius, I can
increase or decrease the visual impact created
by perspective distortion effects.

A smaller circle of view increases perspective

distortions, which will make the building or principal
object appear more dynamic, will enhance
perspective space and the volume of the object, and
will emphasize the front surfaces or vertical
dimensions of the form.

A larger circle of view minimizes these effects, which

will make the form appear less dynamic and more
"abstract" or idealized, will flatten the perspective
space, will make the object appear less three
dimensional (as in a telescopic view), and will tend
to emphasize all parts of the object equally.

In this case, I know the owner of the greenhouse

values her peace and tranquillity ... no looming or
soaring shapes for her. I also judge that the basic
design of the greenhouse balances height and floor
plan well, so there is nothing to gain by emphasizing
the vertical dimension. And I also know that the
greenhouse is designed to merge well with its
setting. With these considerations in mind, I expand
the drawing circle of view by 25% (from 36" to 45"),
to produce a flatter, more idealized conception of the
finished building, and to push the building visually
into its background setting by flattening the
perspective space, much as it would appear within an
18° circle of view.

Now all the layout considerations — format, viewing

distance, object orientation, drawing size, scale of
view, anchor point, anchor line, horizon line, median
line, direction of view, object circle of view and
drawing circle of view — have all been carefully
thought through and specified in relation to each
other and to the design goals of the image. Now I
can inscribe the drawing circle of view, rotate the
vanishing points and establish the measure points,
as shown in the figure above.

Measuring the Front Projections. With the

important design and layout decisions established,
the next steps are straightforward and mechanical.
The front plan is taken first, and scaled to the same
size as the drawing. The actual blueprint or object
elevation can be enlarged or reduced using a zoom
photocopying machine, or the dimensions can be
measured off the original and scaled with a hand
calculator, or the dimensions can be scaled by
construction.
Either way, the amount of reduction required
depends on the scale of the orignal. If the blueprint
is in a standard architect's scale of 1/4" = 1 foot, for
example, then it is already at a 2.1% reduction in
relation to the actual structure. In this case, I've
already determined that my drawing is at a 56%
reduction of the blueprint scale, so I can use a zoom
photocopier to produce plan and elevation at that
scale, or rescale the key measurements from
construction.

constructing from a plan: front projections

shown in a 60° circle of view

As shown in the figure, I place the scaled front

elevation below the anchor point, with the right edge
of the building exactly underneath the anchor point. I
extend a horizontal line to the left of the anchor point
as the measure bar (thick magenta line). (If I am
working directly on the drawing, I find it is clean and
convenient to create the measure by with a single
piece of drafting tape; when I'm finished these
measure points and the measure bar can be removed
by simply peeling the tape away.)

Next, connect the anchor point to the lefthand

vanishing point (vp1), drawing the line with a light
graphite or eraseable blue pencil.

Now I carry the important horizontal intervals in the

elevation — the sides of the door, the width of the
entryway, the peak of the roof, the width of the
building — straight up to the measure bar. I mark
these as precisely as I can.

Finally, I use a long straight edge to connect each

mark on the measure bar with the appropriate
measure point (mp2). The intervals in the front
elevation in perspective recession are located where
these measure lines intersect the vanishing line
to vp1.

I mark each intersection point carefully, then a draw

vertical line upward from each point using either very
light graphite pencil or an eraseable blue pencil.
Finally, I remove the front elevation and erase or
peel away the measure bar.
 constructing from a plan: side projections

shown in a 60° circle of view

Measuring the Side Projections. Next I repeat

these procedures with the side elevation, this time
drawing the vanishing line from the anchor point
to vp2, placing the measure bar on the right of the
anchor point, taking the measure marks from the
righthand measure bar to mp1, and drawing the
verticals at the point where each line intersects the
vanishing line from the anchor point to vp2.
 constructing from a plan: vertical
projections

shown in a 60° circle of view

Measuring the Vertical Projections. The side and

front projections will share a common vertical line,
the anchor line, extending upward from the anchor
point. This line also serves as the measure bar for
vertical projections — the top and bottom of the door
and entry steps, the covered entry way, and the
eaves and peak of the roof.

If there are distinctly different features on the front

and side of the building — as there are, for example,
in the facade and sides of a Gothic cathedral — then
you must take the vertical measurements separately
for the two sides. Again, masking tape makes an
excellent removable measure bar.

For this greenhouse drawing, the major features

along the side are defined by the roof eave, which
appears on the front elevation. So only that elevation
is used. First I have to align the base or
foundation of the building so that it lies exactly on a
horizontal line from the anchor point. Then I carry
the important elevation heights to the measure bar
with horizontal lines.

Finishing the Drawing. The vertical projections are

carried back to both vanishing points (vp1 and vp2),
and the vertical lines marking the important
horizontal intervals on the front and side are
trimmed off at the appropriate heights.

This drawing has a peaked roof, which requires a

little finesse, as shown in the figure below. The peak
of the roof in recession is indicated by the line
extending upward from b, which was specified from
the front elevation. However its height is indicated on
the front measure bar at a, on a vanishing line
carried back to vp1. The front peak of the roof is
located where this line and b intersect, at point x.

constructing from a plan: finishing the

drawing

shown in a 60° circle of view

The eaves are located where their elevation

(indicated by the vanishing line from point c on the
vertical measure bar) crosses the vertical lines
marking the front and left corners of the greenhouse,
the lines extending upward from d and e. Connecting
these eave elevations to x defines the front and back
pitch of the roof in perspective.

Connecting x to vp2 defines the peak of the roof

along the length of the greenhouse. But how to find
the peak at the back of the greenhouse? I
connect a to vp2, and locate the point where this line
intersects the line extending upwards from the back
corner of the greenhouse (f). Then I find the line
from this point to vp1: then the point x' is located at
the intersection of this line with the roof peak.
Finally, I locate c' by connecting c to to vp2, and
finish the back pitch of the roof with a line
from x'to c'.

Now I can close up the exterior surfaces, erase

guidelines and hidden lines, and finish the drawing
with as much detail, shading and backdrop as I want.
The approximate layout of the finished drawing
within the 11"x15" format is shown in the figure.

paraline perspectives
In the introduction to these perspective materials I
stated that the point of view, not objects in space,
is the fundamental perspective theme. One
consequence of this is that the objects in space are
not always clearly defined. The side of a building may
recede along the direction of view, obscuring its
length, openings or surface details of the side, or the
building may extend outside the 60° circle of view,
causing the form to appear distorted.

As commerical manufacture, military surveying and

industrial engineering extended their scope in the
18th century, technical drawing methods were
developed for civilian and military applications that
showed the three dimensional form of objects or
sites while accurately recording their physical
dimensions. These techniques were first published by
Christian Rieger and Johann Heinrich Lambert in the
1750's and were extended by the Rev. William Farish
in 1820.
 difference between central and parallel
projections

The innovation common to these

nonconvergent, paraline projections (a contraction
of parallel line projections) is that the physical form
is projected onto to image plane by means of parallel
projection lines. This gives paraline images three
unique features (diagram, above):

• there is no viewpoint or convergence point for the

projection lines (technical sources state this as "there
is infinite distance between the image plane and
viewpoint")

• as a consequence of this projection method, all

parallel lines in space are parallel in the image (in
other words, there are no vanishing points in any
direction) — so the name "parallel projection" is apt
in a second sense

• the image size is independent of projection

distance; paraline projections cannot represent
recession in space.

In contrast, central projections

or perspectives project the physical form onto the
image plane with convergent projection lines to the
viewpoint. This may cause parallel lines in space to
appear convergent, depending whether the
projection is 1PP, 2PP or 3PP and whether the lines
are horizontal or vertical; in any case, the focus is on
the relative location of the vanishing points. The
image size is now dependent on the object distance
from the image plane or viewpoint and its orientation
to the image plane; depth dimensions are
foreshortened and recession is represented.

Types of Paraline Projection. The differences

among paraline images can be defined using one of
two conventions (diagram above):

• analytical: definitions are in terms of (1)

the projection anglesbetween the edges or faces of a
cube and the image plane, and (2) the projection
angle between the parallel projection lines and the
image plane

• graphical: definitions are in terms of the

(1) relative scale of the three dimensions of a cube in
the image, and (2) the three arbitrary image
angles between the edges of a cube in the image.

The analytical definition derives from projective

geometry, has a mathematical basis, and defines
images in terms of the projection angles, which
typically produce irrational numbers for the
dimension foreshortening or face angles of a cube;
there is no explicit reference to the image attributes.
At the end of the 20th century this tradition was
adapted to the computational methods of computer
graphics, especially game programming.

The graphical tradition derives from technical

drawing methods and drawing conventions: only
standardized (template) angles and dimension scales
are used, one dimension is always verticalin the
drawing, and the scale of any foreshortened
dimension is usually a simple fraction (e.g., 1/2) of
the other dimension(s); there is no explicit reference
to the projection geometry.

Mischief occurs when these traditions are confused or

interbred. It is pointless to define projection angles
for noncomputational, graphical applications. Many
online sources to the contrary, the graphical
definition of paraline projections must state both
relative scale of the horizontal, vertical and depth
dimensions and the graphical angles between them.
Relative scale is defined as isometric (all three sides
of a cube are drawn in equal scale), dimetric (two
sides of a cube, usually the horizontal and vertical,
are drawn in equal scale), and trimetric (all sides of
the cube are drawn in different scales).

The diagram (below) presents five illustrative

paraline projections using the same "barracks, wall
and watchtower" example: multiview orthographic,
30°/30° isometric, 60°/30° isometric, 42°/7°
dimetric, and military (45°/45° isometric).
 paraline or parallel projections

Multiview Orthographic Projections. The first

parallel projections were the elevation and plan of
a building. In the analytical literature these are
termed orthogonal projectionsbecause the projection
rays are at right angles to the image plane, and in
the graphical literature are termed orthographic
views because right angles in a cube appear as right
angles in the image.

To achieve this, two dimensions of the primary form

are oriented parallel to the image plane. The third
dimension is not simply foreshortened — it is
eliminated from view.

This is the chief disadvantage of orthographic

renderings: each two dimensional projection entirely
suppresses the third dimension, which forces the
reader mentally to combine two or more different
drawings to understand a three dimensional
conception. In the example, it is not possible to
identify the shape of the barracks or tower roof from
the plan, and only the y/z elevation shows that the
barracks roof has a gable. Hence, multiple
orthographic views are necessary to completely
understand the physical shape of the form.

Axonometric Projection. In axonometric

projections, all three dimensions are represented as
a two dimensional image; the third or depth
dimension is brought into the image by drawing all
three dimensions at an explicit relative scale and
interior angle.

The generic method for developing an axonometric

drawing is as follows: (1) start with the plan drawn
to scale, oriented to produce the optimal paraline
image; the horizontal dimension is denoted x and the
vertical dimension z; then select the corresponding
elevation at the same scale and with the same
horizontal dimension x and the vertical dimension y;
(2) construct all verticals y as parallel vertical lines in
the drawing, at either 1:1 scale or a reduced scale to
the scale of the elevation; (3) draw all x dimensions
at a constant angle to horizontal, either to the left or
right of the end points of verticals, at either 1:1 scale
or a reduced scale to the scale of the elevation; and
(4) draw all z dimensions at a constant angle and
scale to horizontal, on the opposite side of verticals
from the x dimensions.

Within this generic mapping recipe, the only

graphical variations in paraline projections are the
relative scale of the three dimensions and the angles
of the two other dimensions to the horizontal.

30°/30° Isometric Projection. This is among the

most common paraline projections used today, so
much so that "isometric" has become synonymous
with a paraline projection. Analytically, this
projection is produced when all three front edges of a
cube are at an equal (~35.3°) angle to the image
plane. Graphically all three corner angles are
represented by equal (120°) interior angles (e.g., the
two nonvertical dimensions are at 30° to a horizontal
line), and all three dimensions are drawn in equal
scale (1:1:1).

Graphically, a 30°/30° isometric drawing is defined

as follows: (1) vertical (y) dimensions are drawn
vertical; (2) the x and zdimensions are drawn at a
30° angle to the horizontal, and (3) all dimensions
are drawn to the same scale. As a result, both
horizontal and vertical circles are shown as ellipses.

Standard isometric paraline drawings are facilitated

by the use of a preprinted isometric grid (which
consists of parallel vertical lines cut by parallel lines
at 60° and 30° angles to vertical) laid under the
working surface, or similarly preruled sheets of
architect's vellum, or the standard 30°/60°/90°
drafting triangles. In computational graphics (for
example, in Sim City), because of the limitations of
pixel representation, the graphical angles are
actually 26.6°, so that oblique lines can be
represented in ascending or descending two pixel
segments.

60°/30° Isometric Projection. Although it is

visually pleasing and approximates very well a
similarly oriented 2PP perspective drawing, there are
two problems with the standard isometric format: (1)
the plan is not reproduced, but appears compressed
in a lozenge form, and (2) this compression affects
the appearance of many irregular forms (such as
circles and spheres, or rectangular forms at odd
angles to the three primary dimensions; cf. the
"tower" in the example drawing).

Both these problems are remedied by raising one of

the oblique dimensions to a 60° angle to the
horizontal; now the drawing is planometric (the plan
is reproduced exactly in the image), and as a result
irregular forms are easier to interpret.

However, this introduces a new problem: the vertical

dimension now appears exaggerated or elongated.
This can be partly remedied by drawing the vertical
dimension at 3/4 or 2/3 scale.

45°/45° (Military) Isometric Projection. This is

also an isometric projection because all three sides
are in the same scale. This is sometimes called
a plan oblique projection or planometric projection in
the graphical literature because the plan angles and
dimensions are reproduced exactly. (Note that the
military "projection" is not orthogonal but is not
oblique either: analytically, it is not possible to
produce the same image through parallel lines at an
oblique angle to the image plane.)

The primary objection to military projection, as with

the 60°/30° isometric, is that it appears to
exaggerate the "depth" of the drawing. As a result
the z axis (for sideways views) or the y axis (for
downward views, as in the example) is sometimes
shortened by 1/2, creating a true dimetric drawing:
this is called a cabinet projection.

42°/7° Dimetric Projection. Several proposals

have been made for paraline projections that more
closely approximate central perspective; these
appear on casual inspection to be perspective
drawings (at "telephoto" or very large object
distances) although each dimension is in an exact,
constant scale to the corresponding elevations or
plan. One proposal orients the two oblique
dimensions at 7° and 42° to the horizontal; the
length of the dimension at 42° is also reduced to a
1/2 scale.

Although these formats perform very well for cubic or

rectangular forms, they are less successful for
irregular forms, as appears in the "tower" of the
42°/7° dimetric drawing.

Artistic Importance of Paraline Projections. Why

should artists bother with the rigid methods of
technical drafting? Because perspective in all its
aspects is a remarkably clear "laboratory" in which
we can study many of the deep or complex problems
of artistic representation.
In terms of art history, paraline projections
characterize classical Chinese and Japanese scroll
paintings. In Europe, paraline representations
became an important style feature of early 20th
century painting; Cubist paintings abound with
examples of essentially planometric or isometric
designs, and Charles Sheeler produced a delightful
painting (image, right) that combines planometric,
isometric and dimetric images of rectagular or square
forms to suggest spatial volume while contradicting
spatial recession; note the precise way that the table
legs intersect the square rug pattern behind. Sheeler
was obviously familiar with the conventions of
technical drawing, and could deploy them to good
artistic effect.

At a deeper level, both the orthographic and paraline

projections represent different solutions to two
Charles Sheeler's Interior (1926)
fundamental and related artistic
problems: simplification and schematization. All
artists, whatever their style or artistic goals, grapple
continually with these two problems.

Simplification is the decision to throw away

information — details, complexities, dimensions or
features — in a way that makes other information
clearer. In orthographic projection the third
dimension is eliminated. In both the orthographic
and isometric projections, the object appears as if
viewed from infinitely far away, which discards
information about the physical location of the
viewpoint in relation to the objects in the image; the
viewpoint is genuinely imaginary.

Yet all projection drawings, paraline and perspective

alike, retain the direction of view as the average or
parallel direction of the projection lines. The back
side of the object is not represented and all angles
are shown in a specific and consistent relationship,
either to each other or to a vanishing point. This
leads us to the insight that simplification can quite
often be paradoxical — how can there be a direction
of view if there is no viewpoint?! — and that by
throwing away information we actually create
contradictions or puzzles, in that sense making the
image or visual idea more complex conceptually
if lesscomplex perceptually. Try looking for similar
paradoxes in other kinds of artistic simplification
(cubism, fauvism, expressionism, abstract
expressionism).

Schematization involves smoothing out these

conceptual paradoxes by making them consistent or
equivalent wherever they appear, or by creating a
hierarchy or system of dominance among them.
Although schematic choices are often arbitrary,and
may depend on subjective esthetic criteria such as
clarity, harmony, emphasis, or contrast, the
schematic criteria can be chosen because there is a
primary external audience or purpose for the drawing
which the schematic is adapted to serve. Thus, the
objection to some paraline drawings is that they
appear to exaggerate one dimension over another;
the paraline schematization must be made more
complex (for example, by using a dimetric or
trimetric format instead of a "simple" isometric
format) in order for the resulting image to be easier
(simpler) to interpret.

In general, "artistic style" refers to the strategies

used to simplify and complicate an image in ways
that produce a visually or esthetically more desirable
artifact. The beauty of paraline projections, and fine
paintings, lies in the way they use simplification and
schematization to create a more legible and
impactful image of the world.

curvilinear perspectives
One of the most elusive but apparently inspiring
goals of perspective studies since the 19th century
has been curvilinear perspective, which involves the
representation of space using vanishing curves rather
than vanishing lines. As these curves seem to
converge at both ends, the horizontal and vertical
transversals create two vanishing points each with a
fifth created by the orthogonals parallel to the
direction of view. Hence the name five point
perspective or spherical perspectivefor some of these
projection systems.
The appearance differences between linear and
curvilinear perspective are shown in the exaggerated
example at right. The linear projection seems to push
distant objects farther away, and to make nearby
objects loom too close, appear out of scale and
exhibit gross distortions at the extreme ends.
Curvilinear perspective crowds the side views toward
the center of the image yet implicitly strengthens the
sense of personal presence through the rapidly
square columns and tile floor
increasing divergence in the approaching orthogonals drawn in linear perspective
(lines in the floor).

Curvilinear perspective has often been justified as

part of a critique of linear perspective. Many
objections arise from the well known perspective
"distortions". The standard (and completely
effective) remedy for these representational conflicts
was to take a view of the subject from a large
distance, so that it fits within a reduced circle of
view, or to take an oblique view so that the
tapering of the horizontal or vertical elements was
consistent with the effect of one of the vanishing same setting as drawn in
points. However these evasions are impractical in curvilinear perspective
particular for the representation of architectural (from Ulrich Graf, 1940)
interiors, such as the nave of a cathedral, where a
restricted circle of view excludes an adequate view of
the architecture.

I present here a simplified method for making your

own curvilinear drawings, and a longer review of the
historical justification for curvilinear methods, with
citations to books where you can find more
information. Unfortunately, the best of these are
either out of print or untranslated, but a good
university library may help you find them.

Constructing Curvilinear Drawings. The

perspective framework for making curvilinear
projections is tedious to set up but not difficult to
work out. Your goal is to make a template that
represents a rectilinear (right angled) grid of infinite
height and width, parallel to the picture plane, as
imaged in the curvilinear perspective system of your
choice. You then draw this curvilinear grid within a
circle of view, as shown for the spherical template
below.
 template for curvilinear transversals in
central (1PP) perspective

shown with 5° transversals and the normal 60° circle of

view, which equals a 70° circle of view in the spherical
projection

To use the grid to map a normal (linear) perspective

drawing or photographic image into the new
perspective space, you must first square the
drawing in the normal way. However, you must be
careful to make the angular size of the squaring in
the linear image equal to the angular interval of the
transversals in the curvilinear grid, or the image will
appear in exaggerated distortion. (The template
above uses a 5° interval, which according to
the distance/size table is roughly the same as a 1
foot width seen from 11 feet away.) The final step is
to copy the image into the curvilinear grid, square by
square, then clean up any inaccuracies as you
progress to the final state of the drawing or painting.

Constructing a similar grid in two point perspective is

much more complex, as the vanishing lines are
foreshortened front to back. But good results are
possible by constructing the 2PP drawing in the
normal way, squaring the drawing, and then
projecting the drawing onto the curvilinear surface
using the grid above. (If you know of a more explicit
method for constructing a 2PP template to any
random rotation of the 2PP vanishing points, please
email me.)

Historical Uses of Curvilinear Perspective. From

the 16th to the 20th centuries, perspective theorists
explored the problem of anamorphic or geometrically
distorted images, which can be rectified back to a
normal perspective image when viewed using an
appropriately curved mirror. These studies often
overlapped with the problems of projective
distortions in two dimensional images, specifically
the difference in perspective view straight ahead and
the view obliquely to either side.
Curvilinear perspective was proposed at least as
early as 1624, in a pamphlet on meteors by the
Tübigen mathematician Wilhelm Schickhardt (as
quoted in Erwin Panofsky's Perspective as Symbolic
Form, 1924):

Schickhardt's "proof" of optical curves

(1624)

"I say that all lines, even the straightest, which do

not stand directly in front of the eye, or go through
its axis, necessarily appear somewhat
bent." Schickhardt's argument is that the
lines ab or cd appear large when close to the viewer
(at V) but necessarily and visibly grow smaller as
they recede toward x or y; therefore "the sides
become narrower and necessarily curved; not like a
roof, to be sure, so as to produce a sharp angle at
points o and p, but rather gently and gradually,
indeed unnoticeably, something like a belly, as is
appropriate for such an arc."

Thus, if we stand at the base of a large tower, the

masonry at eye level appears in central perspective;
if we look up, we see the sides near the top in
converging three point perspective. The same occurs
in horizontal lines when we stand facing a long wall,
then look toward its end on either side.

However, the argument here is flawed, as was

pointed out by Schickhardt's contemporary, the
German astronomer Johannes Kepler. This becomes
obvious if we express the "proof" as it appears in
linear perspective:

the logical fallacy in optical curves

In this diagram, the break in the vanishing lines

indicates that we are physically unable to see the
image of the vanishing points x and y while directing
our vision straight ahead; to see them we must alter
the direction of view, and thereby
completely change the perspective geometry. If
we turn our heads to one side while looking at an
infinitely long wall, then the convergence to x or y is
produced by foreshortening of the image surface,
now viewed at an angle, and this convergence
steadily increases as the direction of view becomes
parallel to the surface of the wall.

These changes in our perspective view must be

accounted for in the diagram by "broken" vanishing
lines, to indicate that different directions of view
apply between x and y. Schickhardt's curves are the
average of an unaccounted number of different
perspective projections produced as the head and/or
the image plane is slowly turned from one direction
of view to another. In fact, his argument originated
in the observation of a bolide, which traversed a
large part of the night sky and carried observers'
rapt eyes along with it.

But why stop there? One can also create a

panoramaic photograph with a 360° view, which is
optically impossible because the eye would then have
to be lens in all directions at the same time that it
would be retina on all sides. At some point, we
accept that composite images do not make valid
perceptual or perspective arguments and we part
company with Schickhardt.

Another 17th century argument was that the eye is

an internally convex surface, and this must cause the
curvature in lines projected onto it, an argument
refuted empirically by M.H. Pirenne in his Optics,
Painting and Photography (1970). A more
contemporary argument is based on the appearance
of wide angle or fisheye photographs, which show
curved lines projected into a flat photograph and
therefore seem to validate the curviness of visual
space. But in these photographs the distance points
in the image are compressed in the field of view,
which shifts the virtual center of projection in front
of the viewpoint.

In any case, it wasn't until the 19th century that

curved vanishing lines were offered as better
representations of extended horizontal or vertical
recessions. The eccentric painter Arthur Parsey (in
1836) and the amateur artist and astronomer William
Herdman (in 1853) published systems of perspective
that replaced parallel transversals in central or two
point perspective with slanting or curved lines. These
systems culminated in the subjective
perspective developed by the German
mathematician Guido Hauck in 1879. Long, lofty
church interiors were especially popular set pieces
for early curvilinear perspective representations, as
in this drawing by Herdman, which is surely intended
as a contrast to the many 18th century Dutch
paintings of church interiors in perfect linear
perspective.
 architectural drawing in curvilinear
perspective

Interior of Rosslyn Chapel by William Herdman

(c.1850)

Curvilinear perspectives had a fitful history after the

late 19th century, then saw a resurgence late in the
20th century. Seminal in this context were Erwin
Panofsky's Perspective as Symbolic Form (1924),
based on Hauck's mathematics; La perspective
curviligne (1967, published in English as Curvilinear
Perspective in 1987) by the French theorists Albert
Flocon and André Barre, the "hyperbolic" system
proposed by the artist Robert Hansen in 1973, and
the "fisheye" or wide angle perspective developed by
artist Michael Moose in 1986.

Several cultural or technological factors have been

used to explain the development of these new
systems, including 19th century advances in
nonlinear geometries and compound (wide angle)
optics, perceptual experiments in subjective
curvature by Hermann Helmholtz (in particular, the
demonstration that a row of widely spaced point
lights, moved perpendicularly away from a viewer in
complete darkness, appear to follow a curved rather
than straight line convergence), the discovery of
subtle curvatures in ancient Greek architecture,
renewed interest in linear perspective distortions, the
study of new types of mapmaking projections, the
unparalleled "wide angle" vistas made possible by
modern iron towers, skyscrapers and air travel ... the
list is a long one.

I think the primary issue is much simpler. One of the

founding ideas of the Renaissance, advocated by
artists and scientists alike, was the fundamental
unity between seeing and knowing.In this tradition,
linear perspective was not so much a representation
of seeing as an area where seeing and knowing
overlapped. Throughout the 16th century,
mathematics and projective geometry, the
procedures of perspective drawing, and the tools and
methods of surveying, navigation and astronomy
were treated as different aspects of the same
fundamental discipline, and several of these topics
were often discussed together in a single book.

At the same time, artists from Leonardo to Turner

were well aware of, and troubled by, the many ways
linear perspective did not seem to accurately
record all of visual experience. By the 18th century,
European culture began to grapple seriously with
problems of color perception and visual illusion that
demonstrated seeing was a psychological and
subjective process, very different from knowing and
with its own quirks and powers. This realization
created a fundamental divide that has expanded and
ramified in artistic practice ever since the late 19th
century.

Some artists pursued the representation of visual

experience or "visual facts" separate from the
"knowing" that comes from perception. This was the
point of departure for many 19th century "seeing"
artists (from Constable to Monet to Bonnard), who
described their work as copying whatever was
available on their retinas; and for artists such as
Manet, Seurat or J.S. Sargent, who analyzed the
process of vision by creating images from painterly
visual deceptions, showing that what we see (or how
we see) does not represent what is "actually there"
on close inspection.

In reaction, other artists rejected the visual facts in

favor of the insight or "knowing" that seems to be
the experiential fruit of perception. That is, they
found ways to represent "higher realities" as a kind
of visual experience that has no explicit referent in
optical or static images — in particular,
representations of spirituality and emotion. This
highly diverse tradition emerges in "spiritual,"
"constructivist", "cubist", "antiretinal," "conceptual",
"nonrepresentational" or "expressionist" artists as
diverse as Matisse, Kandinsky, Duchamp, Picasso,
Pollock, Riley, Rosenquist and Martin.

Against that background, curvilinear perspective

seems to be a conservative reaction, an attempt to
recreate the union between seeing and knowing by
altering rules of seeing to correspond to the
intuitions of three dimensional knowing. Certainly,
altering the rules is explicit in the "argument for
optical curves" advanced centuries ago.

The consensus after more than a century of debate is

that Panofsky and other critics of linear perspective
are factually wrong: no other two dimensional
projection is superior to standard methods of linear
perspective when the perspective drawing is viewed
with a single eye from the perspective viewpoint
(center of projection). Under those conditions, a
perspective drawing really does capture exactly the
visual angles of the original scene — as
demonstrated by M.H. Pirenne.
Apparent perspective distortions arise because the
image is not viewed from the correct center of
projection and direction of view, or the perspective
geometry is changed, or different geometries are
fused into a single image.

For me, that's the point: curvilinear perspective

represents the state of looking at the same time in
many directions. In contrast, I have repeatedly
stressed that linear perspective is the image of a
specific viewpoint and direction of view, and once
that restriction is relaxed or abandoned, images can
easily bend, flow or warp into unpredictable and
highly expressive new geometries.

place furstenberg, Paris, August 7,8,9

by david hockney

Curvilinear perspective in effect averages or

summarizes the many possible views from a single
viewpoint, much as David Hockney assembles an
image from dozens or hundreds of localized,
narrowly cropped photographs. In that context,
curvilinear methods can be justified as visually
syncretic and philosophically "postmodern".

Leonardo and many others after him identified

"flaws" in linear perspective only because they
considered the same perspective situation from two
or more directions of view. Culturally we are no
longer predisposed to see multiple perspectives or
multiple points of view as disruptive affronts to
orthodoxy.

N E X T : Shadows, Reflections & Atmosphere

Last revised 07.I.2015 • © 2015 Bruce MacEvoy

shadows, reflections & atmosphere technique

The geometry of natural (solar) shadows is one of
the most complex perspective constructions because
of the many possible arrangements of light source,
shadow casting edge and shadow receiving surface in
relation to the viewpoint. The basic rules of
perspective shadows explains the consistent
features — the light plane and surface plane — that
define all shadows.
The shadows from solar light can be classified into basic rules of perspective
nine types, depending on the orientation of the shadows
shadow casting edge and shadow receiving plane.
These shadow types provide the foundation for shadows from solar light
constructing shadows from local lights, such as
outdoor and interior lighting fixtures. shadows from local lights

Reflections create complex perspective problems, perspective of reflections

especially if the reflecting surface is curved. I explain
plane reflections — as from a mirror, or the surface aerial perspective
of a large body of water — and briefly describe the
reflections from rippling water and curved reflecting rainbows
surfaces such as a metallic globe.

One of the earliest criticisms of linear perspective,

found in Leonardo da Vinci's notebooks, was of its
inability to account for the atmospheric effects of
clouds, haze and smoke. I clear up the geometry
of aerial perspective and contrast the visual effects
of water vapor and of suspended particles.

Finally, rainbows were among the great

preoccupations of Romantic era landscape painters,
especially in Germany and England. As this
fascination came at the end of the topographical
painting and academic perspective traditions, a
review of rainbow geometry makes a fine conclusion
to perspective studies.
 basic rules of perspective shadows
Whenever you start the perspective analysis of a
shadow, it is always helpful to consider the six
factors that will determine the shadow's shape and
appearance:

• the location of the light source in relation to the

shadow casting object, the viewpoint, and the
direction of view (in front of or behind the viewer,
above or below the viewer, left or right of the
viewer)

• The object geometry, the shape and location (in

relation to the viewer and light source) of the
material form that is casting the shadow (a vertical
pole, a horizontal bridge, a slanted roof, or a curving
dome).

• The surface geometry on which the shadow is

cast (a vertical wall, the horizontal ground, a slanted
ramp, a curved surface such as a column or dome, or
an irregular surface such as sand dunes or a rutted
road).

• The visual angle of the light source as seen from

the shadow casting edge; measured in radians, this
is its projection ratio, which determines the
darkness and sharpness of the shadow edge (visually
small or distant light sources produce sharper
shadows).

• The shadow distance between the shadow

casting edge and the shadow receiving surface (a
smaller shadow distance produces sharper shadows).

• The difference in luminance and color between

the light source creating the shadow and the light
source (if any) shining into the shadow; a large
luminosity contrast darkens the shadow and
decreases the tinting caused by illumination into the
shadow.

In most landscape situations the last three factors

have a fixed effect: the angular size of the sun is
relatively small (about 0.5°) and constant, most
shadow casting objects are close to the ground, and
the sun is an extremely bright light source. (The
exceptions occur under cloud layers, which weaken
and diffuse the solar light.) This reduces the
construction of perspective shadows to the geometry
of light source, viewer, object and surface, which
have been traditionally illustrated using the straight
corners and curved arches of architectural forms.

Once solar shadows are mastered, the geometry of

shadows from artificial or "local" light sources, as
from streetlights or indoor ceiling fixtures, can be
analyzed using the same basic rules of shadow
perspective. These are listed below.
 The Basic Rules of Shadow Perspective

1. Light appears to emanate as straight line

"rays" from the surface area of a light source.
Light appears to follow a straight line path from its
origin at the light source. (This corresponds to
perspective fact 1.) The origin is physical — it has a
measurable surface area — and light radiates from
all parts of the surface.

2. Light appears to radiate from the physical

center of the surface or volume of the light
source. When constructing perspective edges, light
always appears to emanate from the point at the
center of the volume of the light source (if it radiates
in all directions) or from the center of its surface
image (if it is a directed light or reflecting surface).

As a corollary, all shadow edges can be

constructed as light rays from a point light
sources. That is, the shape and edges of a shadow
can be constructed on the image plane as if all light
emanated from a single point on the image plane.

This principle does not adequately describe actual

shadows, because these typically have a diffuse or
fuzzy border of partial shadow, called
the penumbra. The penumbra appears to straddle
or balance on the shadow edge, reducing the darkest
area of complete shadow while adding a boundary of
partial shadow, as explained below.

However, a traditional rule of thumb is: solar

shadows are indistinguishable from the
shadows created by a point light source; the
penumbra of solar shadows can be ignored in a
perspective image. This rule actually applies only
when the shadows are viewed from a distance — the
object and its shadow are within a 60° circle of view.

angular size of the moon (and sun)

as shown in this 60° circle of view, the full moon has

such a small visual size (compared to a floor lamp) that
it is equivalent to the vanishing point for parallel light
rays
The sun's distance from earth is about 108 times its
diameter, which means its visual size is about 0.5°.
The moon coincidentally is almost exactly the same
visual size, so the image size of the sun can be
represented by the full moon surrounded by familiar,
local objects, for example as it appears through a
window. Then it does indeed resemble a point source
(photo, above).

Because the sun is so far away, compared to the

height of objects on the earth's surface, light from
the sun displays the same visual convergence as
parallel light rays, and its disc is visually small
enough to approximate the vanishing point for
parallel visual rays (perspective rule 6). So the
traditional perspective rule for solar shadows is
actually: the sun is a point light source of
parallel light "rays".

3. Parallel light rays define two vanishing

points on opposite sides of the viewer's visual
sphere. Only one of the two points can appear on
the image plane, when it is the light vanishing
point (lvp). Light appears to emanate from one
vanishing point (the light point) and to converge at
the other vanishing point (the antilight point). The
only exception is when the light rays are parallel to
the image plane.

This applies to any visually small light source, but

specifically refers to the location of the sun in the sky
(called the solar point) and the point exactly
opposite from the sun on the viewer's visual sphere
(called the antisolar point; diagram, below).

The antisolar point is always located in the shadow

of the viewpoint on the image plane, if it is
visible. In other words, if you can see your shadow
on the ground, then the shadow of your head is
centered on the antisolar point. (The same principle
applies to the antilight point of artificial light
sources.)
 location of the solar and antisolar points in
"about face" orientations of the 90° circle
of view

either point acts as the light vanishing point (lvp) in a

perspective drawing

4. The light vanishing point is located on the image

plane by two angles: its altitude and the azimuth.
The altitude is the visual angle (in degrees) of the
light above or below the horizon line; the azimuth is
the visual angle of the light to the left or right of the
median line (principal point).

The altitude of the solar/antisolar point above or

below the horizon line is the the angle of its altitude,
as rotated from the horizon line at the left or
right dvp. This altitude, when projected onto the
median line, defines a horizontal line on the image
plane. The distance of the solar/antisolar point to
the left or rightof the median line is the the angle of
its azimuth, as rotated from the median line at
the top or bottom dvp. This azimuth, when projected
onto the horizon line, defines a vertical line on the
image plane. The image of the sun or the antisolar
point is located at the intersection of the two lines
(diagram, above).

If the viewer makes a 180° turn ("about face"), then

the antisolar point (violet) is in the same location as
would be the reflection of the sun in a body of water:
at the same altitude rotated below the horizon line,
and at the same azimuth on the same side (left or
right) of the median line.

5. Shadow edges form at the surface of light

obstructing objects tangent to the direction of
light rays. Shadows typically have edges even if
objects don't (because the objects are round or
cylindrical, for example). This is because the surfaces
have turned parallel to the direction of the light rays,
creating a shadow edge.

This edge is actually two edges. One edge completely

encircles the shadow casting object along the surface
boundary between lighted and shadowed surface:
this edge is called the shadow terminator. The
terminator forms on all surfaces of the object
tangent (parallel) to the direction of the light.

The other edge encloses the shadowed area on the

shadow receiving surface, as a projection of the
terminator or a silhouette of the object as seen from
the light source: this is the shadow edge.

The right angled corner edge of a cubic or

rectangular solid defines a terminator that remains at
the corner of the object even as the light source
moves in space, so long as one face on either side of
the edge is illuminated and the other is in shadow.
This is unlike the shadow casting edge on a
cylindrical or irregularly shaped object, which moves
around the surface of the object, and may reveal
changing contours, as the location of the light source
changes.

6. A shadow edge is visible where it intersects

a translucent medium or opaque surface plane.
Shadows normally are not visible in air or water, and
if they are they appear as shafts of darkness. A
recognizable object shadow only appears when the
light shining around an object strikes a surface
behind it. As a simplification, this surface is defined
as a plane, even though many natural and artificial
surfaces are irregular or complex (hillsides are
eroded, building walls contain windows, sidewalks
are sloped, etc.).

The surface plane has its own vanishing line,

the surface plane vanishing line (spvl). In most
perspective problems, the surface plane is either the
ground plane, whose vanishing line is the horizon
line, or the wall of a nearby building, whose
vanishing line is perpendicular to the horizon line.
Inclined planes also occur, as hillsides, sloping roofs
or the average rise of stairways. So surface planes
may be horizontal, vertical or inclined.

7. The light plane is defined by the light

vanishing point and the terminator edge of the
shadow casting object. For a straight edge
terminator, such as the terminator along the corner
of a rectangular solid such as a building, this plane
represents a sheet of parallel light rays that define a
straight line shadow edge on any flat (plane) shadow
receiving surface (diagram, below).

On the image plane, the length and orientation of a

straight shadow edge are defined by its two end
points (ep1 and ep2) (perspective rule 3). The line
segment between these end points, if extended,
defines the edge vanishing line, which terminates at
the edge vanishing point (evp).

This plane has a unique vanishing line, the light

plane vanishing line (lpvl). Obviously, because the
edge is contained in the light plane, the edge
vanishing point (evp) is contained in the light plane
vanishing line (lpvl) (perspective rule 14).

This produces an important corollary: the light

vanishing point and edge vanishing point define
the light plane vanishing
line (perspective rule 14). Once lvp and evp have
been established on the image plane, the light plane
vanishing line is simply the line that contains them
both.
 the light plane, surface plane and shadow
edge

8. A straight line shadow edge is contained in

the intersection of the light plane and surface
plane. The intersection of two planes is a line,
the shadow vanishing line,which has two vanishing
points on opposite sides of the visual sphere — one
toward the solar point and the other toward the
antisolar point. Only one of these is visible on the
image plane, as the shadow vanishing point (svp,
diagram above).
As a corollary: the shadow vanishing point is
located at the intersection of the light plane
vanishing line and the surface plane vanishing
line. The shadow edges are always directed toward
the shadow vanishing point, if the light vanishing
point is the antilight point, or away from it, if the
light vanishing point is the light source.

9. Two vanishing points and at least three

image points are necessary to define an edge
shadow on the surface plane. The two vanishing
points are (1) the shadow vanishing point svp and
(2) the light vanishing point lvp. Both points lie in
the light plane vanishing line (lpvl).

The svp for a straight line shadow requires at least

(1) one edge station point (sp) to define the location
and direction of a shadow vanishing line in the
shadow receiving plane, and (2) the lvp requires two
edge end points (ep1 and ep2) to define two shadow
termination lines that intersect the shadow vanishing
line at the two shadow end points (sep1 and sep2).

Outline of Shadow Constructions. All perspective

shadow problems require you to solve four basic
construction tasks: (1) define the vanishing point
geometry of the shadow edge, (2) define the
vanishing line of the light plane, (3) define the
vanishing line of the surface plane and the location of
the shadow vanishing point(s), and (4) construct the
shadow edge in the intersection of the light plane
and surface plane.

These four groups of construction steps vary in

complexity depending on whether the shadow edge
and shadow plane are parallel or oblique to the
image plane and parallel, oblique or
perpendicular to each other. This generic diagram
(below) summarizes the perspective shadow
elements and the abbreviations I use to identify
them.

basic elements of perspective shadows

 using vertical projection points for an inclined edge
over a horizontal surface plane

(1) Perspective of Shadow Edge. Very often we

already know the vanishing point for the shadow
edge because we've used it to draw the shadow
casting form. If not, we identify the shadow edge end
points (ep1 and ep2) and their station
points (sp1 and sp2) located vertically underneath
them on the surface plane or ground plane. These
can be used to find the edge vanishing line(evl) and
the edge vanishing point (evp), except in cases
where the edge is parallel to the image plane (and its
vanishing point is not in the image plane,
perspective rule 4).

(2) Perspective of Light Plane. The light vanishing

point(lvp), which is either the solar point or antisolar
point, is found by rotating the angle of the altitude
and azimuth location of the point. The lvp will be
either above the horizon (solar point) or below it
(antisolar point). Then a line
from lvp through evpdefines the light plane
vanishing line for that edge (lpvl edge); a line
from lvp perpendicular to the horizon line defines a
similar plane for vertical edges (lpvl vertical).

(3) Perspective of Surface Plane. The surface

plane is usually the ground plane, in which case
the surface plane vanishing line (spvl) is the horizon
line. Otherwise the surface plane is part of a
constructed object, and its vanishing line can be
found from the edges of the form or their vanishing
points (perspective rule 10). Finally, the intersection
of lpvl verticaland spvl defines the shadow vanishing
point for vertical edges (svp vertical), and the
intersection of lpvl edge and spvldefines the shadow
vanishing point for the shadow casting edge
(svp edge).

(4) Construct Shadow Edge. A line from the

shadow vanishing point svp vertical through an edge
station point spdefines a shadow vanishing line. A
line from the light vanishing point lvp through the
matching edge end point ep defines a shadow
termination line. These lines intersect at the end
point of the shadow edge (sep) on the shadow
receiving surface.

These four groups of steps are used to organize the

instructions for drawing perpsective shadows in the
next section.

shadows from solar light

This section describes how to construct the
perspective solution for all combinations of vertical,
horizontal or oblique edges casting shadows onto
horizontal, vertical or inclined planes in 2PP or two
point perspective.

The defining attributes of solar light are: (1) the sun

has an overwhelming illuminant power, which means
there is only one shadow geometry among the
objects in view; (2) the sun has a insignificant
angular size (about 1/2° of visual angle) and a
relatively small penumbra that grows smaller as the
shadow casting object is closer to the shadow
receiving surface, which means the sun is effectively
a point source that generates parallel rather than
divergent rays of light. Solar shadow edges can
usually be described by a single line, as changes in
the sharpness of the shadow edge are negligible over
the distances defined by most architectural forms.

The principal alternative light source, architectural or

artificial lighting, is discussed in a later section.

Nine Types of Solar Shadow. Solar shadows can

classified into The table below, adapted from Michael
Helms' Perspective Drawing: A Step-by-Step
Approach, provides a summary of the nine variations
and their numerical order in this section.
The discussion proceeds from the simplest problems
to the most complex, and indicates where simplifying
solutions are possible for each combination. The
examples are written to stand alone as separate
reference units.

All abbreviations are introduced in the section

on basic rules of perspective shadows, but I
summarize them here in the order in which they
should be identified/constructed in a perspective
drawing.

elements of perspective shadows

(1) shadow edge: the perspective image of the physical
edge that casts a shadow.
• end point (ep): a point defining the end or corner of the
line segment defining a (straight) shadow casting edge.
• station point (sp): the point on the surface plane that is
directly underneath a shadow edge end point; a vertical
line from a station point passes through an end point.
• edge vanishing line: the vanishing line defined by the
two end points, ep1 and ep2.
• edge vanishing point (evp): the vanishing point of the
edge vanishing line; because the light plane contains the
shadow edge, the evp always defines the edge light plane
vanishing line lpvl edge.
• edge construction lines: any line used to construct the
image of the shadow casting form.
(2) light plane: the plane that contains the shadow
casting edge (e.g., the line defined by ep1 and ep2) and
the light vanishing point lvp.
 • light vanishing point (lvp): the vanishing point for
approximately parallel light rays, which is either the solar
point or antisolar point, depending on the orientation of
the viewer (image plane) to the sun.
• light plane vanishing line (lpvl): the vanishing line for
the light plane; there must be a separate lpvl for every
distinct orientation of shadow casting edge (that is,
one lpvl horizontal for horizontal edges, a
second lvpl vertical for vertical edges, and
other lpvls inclined for inclined edges).
(3) surface plane: the flat surface on which the shadow
is cast.
• surface plane vanishing line (spvl): the vanishing line
for the plane on which the shadow is cast. For shadows
cast on the ground plane and all horizontal planes parallel
to it, the spvl horizontal is the horizon line; otherwise it is
the vanishing line(s) for one or more constructed vertical
planes (spvl vertical) or inclined planes (spvl inclined).
• inclined plane vanishing line (ipvl): the vanishing line
for the inclined plane that casts or receives a shadow.
• shadow vanishing point (svp): the vanishing point for
shadows cast by the shadow edge onto the surface plane;
the svp is always at the intersection of a surface plane
vanishing line spvl and a light plane vanishing line lpvl.
(4) shadow edge: the perspective image of the shadow
on the surface plane.
• shadow vanishing line: a vanishing line from the shadow
vanishing point svp through an edge station point sp; this
line is always also the intersection between an lpvl and
an spvl.
• shadow termination line: any line from the lvp through
an edge ep.
• shadow end point (sep): the intersection of a shadow
vanishing line with a matching shadow termination line.

A common shortcut for rendering shadows, especially

in architectural drawings, is the mixed method. This
arbitrarily locates the solar light source so that light
falls at a 45° angle to all horizontal and vertical
edges. This permits all shadows to be constructed
using the diagonal vanishing points (dvp's) in
the circle of view framework.

1. Vertical Edge Shadow on Horizontal Surface.

This and the next type of shadow are the simplest
perspective problems and the ones easiest to
visualize from common shadow examples such as
exterior building walls, telephone poles or fence
posts, large signs or the sides of window openings,
all casting shadows on the ground, pavement or
interior floors.
Shadow type 1 demonstrates the basic procedure for
identifying a shadow edge end point (sep): at the
intersection of (1) a line from svp through an sp and
(2) a line from lvp through the matching ep. If the
antisolar point is used, the shadow lies between
the svp and the sp, if the solar point is used, the
shadow lies on the opposite side.
Construction Procedures. Elements given: edge
line segment, edge end points, edge station point,
surface plane, horizon line. Key perspective elements
and their abbreviations are summarized here.
 (1) vertical edge shadow on horizontal
surface

compare with shadow type 2 and shadow type 4

(1) Perspective of Shadow Edge. The shadow casting

edge is vertical (parallel to the picture plane), so
there is only one station point (sp), and no edge
vanishing point evp (there is no vanishing point for
lines parallel to the image plane). If the vertical edge
ends in the ground plane, the second end point (ep2)
is equal to the station point.

(2) Perspective of Light Plane. First, locate

the lvp (solar or antisolar point) above or below the
horizon line, to the left or right of the median line.
The shadow casting edge is vertical, therefore the
light plane is vertical for all locations of lvp (solar or
antisolar point). Draw the light plane vanishing
line lpvlvertical as a vertical line through the lvp to
the horizon line. If the sun is near zenith (outside the
90° circle of view), then draw lpvl vertical as a
vertical line that intersects the horizon line at
the azimuthal angle of the lvp left or right of the
median line.

(3) Perspective of Surface Plane. The surface plane is

horizontal, so the surface plane vanishing line spvl is
the same as the horizon line (parallel planes
converge to the same vanishing line,
perspective rule 13). Finally, mark svp vertical at
the intersection of lpvl vertical and the horizon line
(spvl).

(4) Construct Shadow Edge. Locate the vertical edge

station point (sp), and draw a line
from svp vertical through sp verticalto define the
shadow vanishing line svl. Then draw two lines
from lvp through ep1 and ep2 to intersect svl at the
shadow end point(s) (sep). The length
of svl between the two points sep is the image of
the shadow edge.

Comments. For vertical edges that meet the ground

plane, such as wall edges or telephone poles, sp =
ep2 and only one line to lvp is necessary to define
the length (top end) of the edge shadow.
2. Horizontal Edge Shadow on Horizontal
Surface. This is a very common type of shadow,
encountered as the shadows cast on the ground,
pavement or interior floors by the horizontal (top)
edge of a wall, roof or sign, a horizontal suspension
such as a telephone wire or bridge, or the top or
bottom of window casements.

Shadow type 2 is the simplest problem that requires

two shadow vanishing points. One of the shadow
vanishing points is the vanishing point for the
horizontal edges of the primary
form: vphorizontal = svp horizontal; the svp vertical
is defined by the light plane. This creates two
different light plane vanishing
lines, lpvl horizontal and lpvl vertical for the
horizontal and vertical edges. Separate lpvls and
two svps are used in more complex shadow
problems, so the principles for identifying and
constructing them in this case should be thoroughly
memorized and reinforced by observation.

Construction Procedures. Elements already given:

edge line segment, edge end points, edge station
points, surface plane, horizon line, vanishing point
(vp) in horizon line used for perspective construction
of the horizontal edge. Key perspective elements and
their abbreviations are summarized here.
 (2) horizontal edge shadow on horizontal
surface

compare with shadow type 4

(1) Perspective of Shadow Edge. The shadow casting

edge is horizontal (parallel to the ground plane), so
the edge vanishing point evp lies in the horizon line
(perspective rule 14). The horizontal shadow casting
edge was drawn using a specific vanishing
point vp horizontal, which is therefore the vanishing
point for the horizontal
edges: vp horizontal = evp horizontal.

(2) Perspective of Light Plane. Insert

the lvp (solar/antisolar point) above or below the
horizon line by rotating the sun's altitude to the
horizon and azimuth to the direction of view.
Construct the lvp vertical as a vertical line
through lvp to the horizon line. (The light plane
vanishing line for the horizontal edge
(lpvl horizontal) is not needed, but shown for
clarity.)

(3) Perspective of Surface Plane. The surface plane is

horizontal, so the surface plane vanishing line spvl is
the horizon line (parallel planes converge to the
same vanishing line, perspective rule 13). Finally,
(A) relabel the vp horizontal as svphorizontal, or
extend a horizontal edge of the shadow casting
object as a line through the horizon line, to
locate svphorizontal; and (B) locate svp vertical at
the intersection of the vertical light plane
(lpvl vertical) with the horizon line (spvl).

(4) Construct Shadow Edge. Each shadow end point

(sep) is located at the intersection of a horizontal
and vertical shadow vanishing line. Each end
point ep has a matching station point sp in the
ground plane, which is found with a vertical line
from ep to the base of the wall or vertical support
containing the horizontal edge. Then for each station
point sp, draw a shadow vanishing line (svl vertical)
from svp vertical through the sp. Then for the
matching end point ep, draw a line
from lvpthrough ep to the svl vertical in the surface
plane; their intersection defines the location of the
shadow end point (sep). Finally, draw the horizontal
shadow edges by connecting the ends of the edge
shadows with a line to svp horizontal.

Comments. The diagram shows two separate light

plane vanishing lines
(lpvl vertical and lpvl horizontal) to demonstrate
that a lpvl always contains the light vanishing
point lvp and the edge casting the shadow, and that
in every case the shadow vanishing point svp is
located at the intersection of two planes — the
surface plane vanishing line (in this case, the horizon
line) and the light plane vanishing line for a specific
edge (horizontal or vertical).

The example above distinguishes the two end

points ep and their corresponding station
points sp and shadow end points sep, to make
explicit that the two shadow end points sep can be
constructed using the methods for shadow type 1.
The ground plane lines from svl vertical through
the sps are used to define the vertical edges of
shadows, and lines from lvp through the ep's are
used to define the vertical location of shadow ends
(corners). The svls horizontal are found by
connecting seps projected vertically from the same
horizontal edge. Using this method,
the svp horizontal is not needed in the construction.

Alternately, if the svp horizontal and the station

points sp are already known from the perspective
construction of the wall or support containing the
horizontal shadow edge, then only one shadow point
(e.g., sep1) need be constructed using the lvp; the
shadow outline can be constructed with shadow
vanishing lines from this point to the two svps
and/or from the svps through the sps. This is more
convenient when lvp is nearly overhead and far
outside the circle of view.

3. Inclined Edge Shadow on Horizontal Surface.

These shadows most commonly arise from the side
edges of a pitched roof or gable, a building access
ramp or inclined roadway, or from the stay cables
supporting a vertical pole (such as a mast or tent
pole).

All vertical and horizontal shadow elements are

constructed as in shadow type 1 and shadow type
2. As background to the geometry of inclined plane
vanishing lines (lpvl), see the discussion of inclined
lines & inclined planes.

Construction Procedures. Elements already given:

edge line segment, edge end points, edge station
points, surface plane, horizon line, vanishing points
in horizon line for perspective construction containing
the inclined edge. Key perspective elements and
their abbreviations are summarized here.

(3) inclined edge shadow on horizontal

surface

compare with shadow type 2

(1) Perspective of Shadow Edge. The inclined edge

vanishing point evp lies in the vanishing line for the
vertical plane of the wall or support under the
inclined edge (perspective rule 14).
The vp horizontal is found by extending a line
through the station points to the horizon line; the
edge vanishing line is a line drawn vertically through
this point. Then lines extended through the edge end
points intersect this line at their vanishing points.
The diagram shows two inclined edges in the same
vertical plane: ep1 and ep2 extend to intersect the
wall plane vanishing line at edge vanishing
point evp1; ep2 and ep3extend to intersect the
vanishing line at the edge vanishing point evp2.

(2) Perspective of Light Plane. Insert

the lvp (solar/antisolar point) above or below the
horizon line by rotating the sun's altitude to the
horizon and azimuth to the direction of view.
Construct the lvpl vertical as a vertical line
through lvp through the horizon line (for shadows
from the vertical edges of the supporting wall).

The diagram shows two inclined edges, therefore

there are two corresponding light
planes: lvpl1 inclined from lvp through evp1,
and lvpl2 inclined from lvp through evp2.

(3) Perspective of Surface Plane. The surface plane is

horizontal; the spvl is the horizon line. The vertical
shadow vanishing point (svp vertical) is at the
intersection of the horizon line with a vertical line
through lvp.

There are two inclined shadow vanishing

points svp inclined for the two inclined edges. Both
of these are at the intersection of the
respective lpvl with the surface plane vanishing
line spvl(the horizon line). Only svp2 inclined is
visible in the diagram; svp1 inclined is outside the
image very far to the right.

(4) Construct Shadow Edge. The elevation of the first

end point (ep1) is found as a vertical shadow
(shadow type 1) from sp1, which defines the
shadow end point a. The opposite shadow end
point c is found as the vertical shadow for ep3.
Finally, a line is drawn from a to svp1 inclined, and
from svp2 inclined through a: these intersect at b,
the shadow end point for the peak of the end wall.

Comments. I've contrived this example to show how

inclined edges can often produce edge vanishing
points or shadow vanishing points that are very far
out of the image area. However, it will be obvious
that the inclined edge shadow can be constructed
just as accurately by an indirect method — as the
vertical shadows from the three edge end points
(a, b and c), including the center station point sp2.
Then the three points are connected with straight
lines to define the inclined edge shadows.

This "indirect" method (using vertical shadows only)

is preferable when there are few inclined edges,
the lvp is within or not far outside the circle of view,
or the inclined edge shadow is nearly parallel to the
horizon line. The "direct" method (using
the svps inclined) is better if the lvp is far outside
the circle of view, or there are multiple inclined
edges at the same angle receding in perspective (for
example, a row of houses on level ground with
identical roof pitches); in this case a single evp can
be used to construct the edges for them all.

4. Vertical Edge Shadow on Vertical Surface.

This is a very common type of shadow produced by
commercial or residential concentrations of buildings
in close proximity to each other: the horizontal
elements of one building cast shadows on the vertical
surfaces of a neighbor building. It arises from any
vertical element near a wall, for example a telephone
pole located next to a building or the edge of one
building near the wall of a neighboring building. (See
also shadow type 5.)

This type of shadow is constructed as a combination

of shadow type 1 and shadow type 2. All edges
are vertical to the ground plane, and all vertical
surfaces are parallel to each other, so all shadows
recede to the svp vertical or svp horizontal.

Construction Procedures. Elements already given:

edge end points, edge line segment, edge station
points, horizon line, base in the ground plane for all
vertical walls or surfaces, and vanishing points in
horizon line for horizontals in perspective
constructions. Key perspective elements and their
abbreviations are summarized here.
 (4) vertical edge shadow on vertical
surface

(1) Perspective of Shadow Edge. The shadow casting

edge is vertical (parallel to the picture plane), so
there is only one station point (sp), always at least
one end point (ep1) and no edge vanishing
point evp (there is no vanishing point for lines
parallel to the image plane). If the vertical edge ends
in the ground plane, the second end point (ep2) is
equal to the station point. Next, mark the end station
points sp vertical for all vertical edge end
points ep casting shadows. The edge vanishing
point evp horizontal is the horizontal vanishing point
established for the perspective construction, or is
found by extending a line through the
appropriate sps to the horizon line.

(2) Perspective of Light Plane. Insert

the lvp (solar/antisolar point) above or below the
horizon line by rotating the sun's altitude to the
horizon and azimuth to the direction of view.
Construct the lvpl vertical (for shadows from the
vertical edges of the supporting wall) as a vertical
line through lvp through the horizon line.
The lpvl horizontal is a line
from lvp through evphorizontal.

(3) Perspective of Surface Plane. For shadows on

horizontal surfaces, the spvl horizontal is the horizon
line. For shadows on vertical surfaces,
the svpl vertical is a vertical line
through evphorizontal. Then svp vertical is at the
intersection of lpvlvertical and spvl horizontal (the
horizon line); svp horizontal is located at the
intersection of lpvl horizontal and svpl vertical: it
is evp horizontal.

(4) Construct Shadow Edge. In the most common

situation, a shadow extends across the ground to the
base of a wall, then extends up the wall to its end
point. Construct the shadow in the same way. First
draw the shadow vanishing line (svl)
from svpvertical through the sp vertical of the
shadow casting edge. Extend the shadow
from sp along the shadow vanishing line (svl) until it
reaches the base of the vertical surface (sp [wall]).
Draw a vertical line from sp [wall] up the side of the
vertical surface. Cut this line with a shadow
termination line from the vertical end
point ep to lvp to locate the vertical end point of the
shadow sep. Finish any horizontal shadow elements
using the procedures for shadow type 2 or shadow
type 5.

Comments. The color of the shadow projected onto

the vertical surface will be different than the shadow
projected onto the ground: it is typically a darker
shade of the vertical surface color, and is in most
situations lighter valued than the shadow on the
ground because it is illuminated by light reflected
from the ground or pavement. The ground shadow,
in turn, can be colored by reflected light from the
illuminated areas of vertical shadow receiving
surfaces.

5. Horizontal Edge Shadow on Vertical Surface.

This is a very common type of shadow produced by
commercial or residential concentrations of buildings
in close proximity to each other: the horizontal
elements of one building cast shadows on the vertical
surfaces of a neighbor building. The shadow casting
edge may be oriented (1) parallel to, (2) oblique to,
or (3) perpendicular to the shadow receiving vertical
surface. (See also shadow type 4.)

The key to this shadow type is that the orientation of

the shadow receiving surface to the shadow casting
edge is not important. In all cases, the shadow edge
is found by projecting the location of the shadow
edge (and shadow end points) onto the vertical edge
where the shadow receiving surface ends or changes
orientation to the shadow casting edge. Then the
shadow edge is defined by a line connecting these
surface edge points.

Construction Procedures. Elements already given:

edge end points, edge line segment, edge station
points, horizon line, all station points for the edges of
the vertical surfaces (where they change orientation
to the horizontal shadow casting edge). Key
perspective elements and their abbreviations
are summarized here.
 (5) horizontal edge shadow on vertical
surfaces

example shows vertical surfaces that are parallel,

oblique and pependicular to the horizontal shadow
casting edge

(1) Perspective of Shadow Edge. The example

presents the horizontal edge as a suspended beam
(for clarity) and the vertical surfaces as the sides of
an octagonal building. Extend the horizontal edge
line segment (ep1 to ep2) to the horizon line to
locate the horizontal vanishing point
(evp horizontal), or use the
existing vp horizontal for the wall or support
containing the horizontal edge. Find the horizontal
edge end station points (sp1and sp2) by vertical lines
from the edge end points, and draw the horizontal
edge station line through these points
to evphorizontal. Finally, locate all the station
points sp for edges in the vertical surfaces where
they change orientation to the shadow casting edge
— labeled a, b, c and d in the diagram.

(2) Perspective of Light Plane. Insert

the lvp (solar/antisolar point) above or below the
horizon line by rotating the sun's altitude to the
horizon and azimuth to the direction of view.
Construct the lvpl vertical (for shadows from the
vertical edges of the supporting wall) as a vertical
line through lvp through the horizon line. No
other lpvl is required.
(3) Perspective of Surface Plane. The horizontal
shadow receiving surface is the ground plane or
surfaces parallel to it; spvl horizontal is the horizon
line. The vertical shadow receiving surface is a
constructed wall or building at any arbitrary angle to
the image plane; the spvls for these surfaces are not
required for the construction. The shadow vanishing
points svp verticaland svp horizontal are located at
the intersection of the horizon line with
the lpvl vertical and spvl vertical.

(4) Construct Shadow Edge(s). For each vertical

edge of the shadow receiving vertical surface(s),
repeat the following steps: (1) Extend a shadow
vanishing line from svp vertical through the wall
station point for a wall edge to the horizontal edge
station line, to define a corresponding horizontal
edge station point. (2) Construct a vertical line from
this point to find the corresponding edge end point.
(3) Construct a shadow termination line from this
end point to lvp. (4) Identify the intersection of this
line with the wall edge. Thus, a shadow vanishing
line from svp vertical through a defines spa in the
horizontal edge station line; a vertical line
from spa defines epain the horizontal edge; and a
shadow termination line from epato lvp defines a' at
the surface edge; and so on to define b', c'and d' in
the diagram. Finally, connect all the wall edge points
(a' to d' in the example) to define the shadow edge
across the vertical surface(s).
Comments. The method for constructing horizontal
shadows on vertical surfaces is just a repetition of
the method to define vertical shadows on vertical
surfaces (shadow type 4), with all the relevant
edge station points defined in the horizontal edge
station line. It is typically more tedious than
complicated. There are three simplifying situations
when only one shadow edge end point must be
located:

• for vertical surfaces parallel to the horizontal

shadow casting edge: these shadows recede along a
shadow vanishing line from a single shadow edge
point (a' or b' in the diagram)
to svphorizontal across the parallel vertical surface.

• for vertical surfaces perpendicular to the horizontal

shadow casting edge: these shadows recede
toward lvp and are defined by a shadow termination
line from lvp through a single wall edge point
(c' or d' in the diagram) across the perpendicular
vertical surface.

• for vertical surfaces oblique to the horizontal

shadow casting edge: these shadows recede along a
vanishing line found as follows. (1) Extend the base
line (station line) of the oblique edge, to create a
new evl, until it intersects the horizontal edge
station line (X in the diagram). (2) Construct a
vertical from this point to the horizontal shadow
casting edge to define the edge point (Y). (3) Draw a
vanishing line from Y through the oblique edge
shadow point (b' or c' in the diagram) across the
oblique vertical surface.

6. Inclined Edge Shadow on Vertical Surface.

This is less frequent but not uncommon type of
shadow that can arise from the escalator ramp in a
commerce or office atrium, an exterior staircase or
fire escape attached to the wall of a building, the
edge of an inclined roof casting a shadow on the wall
of an adjacent building, etc. The inclined edge may
lie in a plane that is either parallel or oblique to the
vertical shadow receiving surface.

This shadow type requires you to construct the

vertical vanishing line for an inclined edge
(epvl inclined) and locate the edge vanishing
point evp on it. For a general introduction to this
geometry, see the discussion of inclined lines &
inclined planes.

Construction Procedures. Elements already given:

(inclined) edge end points, edge line segment, edge
station points, horizon line, vanishing points in
horizon line for perspective construction containing
the vertical surfaces, and vanishing points for
perspective construction containing the inclined
edge. Key perspective elements and their
abbreviations are summarized here.
 (6) inclined edge shadow on vertical
surface

compare with shadow type 8

The diagram shows the inclined edge as a ramp to

approximate a stairway or the end portion of an
inclined roof.

(1) Perspective of Shadow Edge. First identify the

vanishing point vp horizontal and surface plane
vanishing line spvl verticalfor the shadow receiving
vertical surface. Next, extend the station line under
the inclined edge (from sp1 through sp2) to the
horizon line to locate the inclined edge vanishing
point (vp1), then construct a vertical line through
this point: this is the inclined edge plane vanishing
line (epvl inclined). Finally, extend a line
from ep1 through ep2 to intersect epvl: this is the
vanishing point for the inclined edge (evp).

(2) Perspective of Light Plane. Insert

the lvp (solar/antisolar point) above or below the
horizon line by rotating the sun's altitude to the
horizon and azimuth to the direction of view.
Construct the lvpl vertical (for shadows from the
vertical edges of the supporting wall) as a vertical
line through lvp through the horizon line.
Finally, lpvl inclined is found as a line
through lvpand evp.

(3) Perspective of Surface Plane. For vertical

shadows on horizontal surfaces,
the spvl horizontal is the horizon line, and
the svp horizontal is the intersection
of spvl horizontal with lpvlvertical.

For horizontal shadows on vertical surfaces,

the svpl vertical is a vertical line
through vp horizontal, and svp vertical is the
intersection of spvl vertical with lpvl horizontal (not
shown in diagram), which is vp horizontal.

Finally, isvpV, the vanishing point for shadows of the

inclined edge on vertical surfaces, is at the
intersection
of spvl verticaland lpvl inclined; and isvpH, the
vanishing point for shadows of the inclined edge on
horizontal surfaces, is at the intersection
of spvl horizontal (the horizon line)
and lpvl inclined.

(4) Construct Shadow Edge. The shadow edges for

the inclined edge must recede either
to isvpV or isvpH, depending on the surface they are
cast on. The diagram shows a typical sequence: the
shadow vanishing line svl on the horizontal ground
plane extends from the inclined edge station point
(ep1) to isvpH until it intersects the station line of a
vertical wall (a); from that point it recedes
toward isvpV until it reaches the edge of the wall
(b). From there it is back in the horizontal plane, and
therefore in the svl to isvpH until it meets the
shadow termination line from lvpto ep2 at d. The
shadow edge of the wall is in the svl from the wall
station point c to the svp [verrtical].

Comments. Inclined edge shadows, like the

horizontal edge shadows in shadow type 5, depend
on the angle between the shadow edge and the
vertical surface. However, as with shadow type 5,
the shadow construction can be simplified to the
projection of the shadow edge points at the edges
where the vertical surface changes its orientation to
the inclined edge; the shadow edge is traced by
connecting these points.
It's worthwhile to study the figure until you can
visualize what happens to the shadow vanishing
points and the appearance of the inclined shadow
when the relative positions of the light point, inclined
shadow casting edge and the vertical shadow
receiving surface are changed. Use a pencil, a book
stood upright on its covers and a desk lamp to try
out the variations. Three general principles emerge:

• Shifting the location of the light point in relation to

a fixed inclined edge plane and fixed vertical plane
only changes the location of the shadow edge on the
vertical plane: the angle of the shadow edge is
always equal to the angle of the inclined edge to the
ground plane.

• Rotating the vertical wall plane, in relation to a

fixed inclined edge plane and light source, changes
the angle of the shadow edge to the ground plane:
but the angle can only vary between the angle of the
inclined edge to the ground plane and the angle of
the altitude of the light source.

• Rotating the inclined edge plane, in relation to a

fixed vertical plane and light source, can change the
angle of the shadow edge to the ground plane across
the entire range from vertical to horizontal,
depending on the altitude of the light source. If the
inclined edge plane is nearly parallel to the direction
of light, with the higher end point closer to the
vertical plane, the shadow edge will always be nearly
vertical; if the lower end point is closer to the wall,
the shadow edge may become horizontal or even
incline in the opposite direction, depending on the
altitude of the light source and the slope of the
inclined edge.

7. Vertical Edge Shadow on Inclined Surface.

This is a common shadow in landscape settings, for
example the shadows of tree trunks or farm buildings
on a sloping hillside. It occurs in urban settings as
the shadows falling on hilly streets or onto the broad
stairway entrances to public buildings.

This shadow type requires you to construct the

vanishing line for the inclined plane (ipvl) in order to
find the inclined shadow vanishing point for the
vertical shadow casting edge (isvpvertical). For a
general introduction to this geometry, see the
discussion of inclined lines & inclined planes.

Construction Procedures. Elements already given:

edge end points, edge line segment, edge station
points, horizon line, perspective construction for all
shadow casting edges, two line segments contained
in the inclined shadow receiving surface. Key
perspective elements and their abbreviations
are summarized here.
 (7) vertical edge shadow on inclined
surface

(1) Perspective of Shadow Edge. The shadow casting

edge is vertical (parallel to the picture plane), so
there is only one station point (sp), always at least
one end point (ep1) and no edge vanishing
point evp (there is no vanishing point for lines
parallel to the image plane). If the vertical edge ends
in the ground plane, the second end point (ep2) is
equal to the station point. In the diagram, three
vertical edges are shown: the foreground pole, and
the verticals on both sides of the inclined plane.

(2) Perspective of Light Plane. Insert

the lvp (solar/antisolar point) above or below the
horizon line by rotating the sun's altitude to the
horizon and azimuth to the direction of view.
Construct the lvpl vertical (the light plane defined by
shadows from vertical edges) as a vertical line
through lvp and the horizon line. Because the
shadow casting edge is vertical, an lpvl inclined is
not required.

(3) Perspective of Surface Plane. The shadow

receiving surface is inclined, therefore it is necessary
to construct the inclined plane vanishing line ipvl in
order to locate the shadow vanishing point. Two
vanishing points are needed to identify this line:

(a) Two parallel edges or lines in the inclined plane

can be extended, and their intersection defines one
edge vanishing point, evp1. Alternately, one inclined
edge (e.g., ad) can be extended as a line; its station
points is extended as a second line that meets the
horizon line (at vp1), and a vertical line from this
vanishing point intersects the edge vanishing line
at evp1.

(b) A horizontal line segment, in this case the base of

the ramp or stairway, is extended until it intersects
the horizon line; this intersection defines the
horizontal edge vanishing point, evp2.

The ipvl is then a line drawn between these

two evp's. The svpvertical is located at the
intersection of the lpvl vertical and the horizon line.
The inclined plane shadow vanishing point for vertical
edges (isvp vertical) is found at the intersection of
the light plane vanishing line for vertical edges
(lpvl vertical) and the inclined plane vanishing line
(ipvl).

(4) Construct Shadow Edge. As in previous

examples, the construction strategy is to begin at the
edge station point, and follow a shadow vanishing
line for the shadow receiving surface, until this line
intersects a shadow termination line from lvp to ep.
For the vertical pole in the diagram, the shadow
termination line, ep to lvp, is drawn first. The
shadow edge for the vertical pole is drawn from the
vertical edge station point (sp)
toward svp vertical, until it meets the edge of the
inclined surface (at x); then the line is
continued isvp vertical until it meets the edge with a
horizontal or vertical surface (y); then it is continued
as a vertical line until it meets the shadow
termination line (at z).

For the shadow of the vertical corner on the inclined

plane, the shadow termination line (b to lvp) is
drawn first. Then the shadow for the vertical
edge ab is constructed as a line from the edge
station point (a) to isvp vertical; this intersects the
shadow termination line at c, the shadow of the
corner point b. The horizontal shadow is constructed
as a line from the corner shadow point c to the
horizontal edge end point d. If this is not appropriate
to the problem at hand, then the isvp horizontalmust
be found to complete the line; see shadow type 8.

8. Horizontal Edge Shadow on Inclined Surface.

This shadow type is less common in landscape
settings than shadow type 7, but it is not unusual
in urban settings, for example as the shadow of a
horizontal roof edge on a lower, sloping roof.

As background to the geometry of inclined plane

vanishing lines (lpvl), see the discussion of inclined
lines & inclined planes.

Construction Procedures. Elements already given:

horizontal edge end points, horizontal edge station
points, horizon line, perspective construction for all
shadow casting objects and shadow receiving
surfaces, two line segments contained in the inclined
surface. Key perspective elements and their
abbreviations are summarized here.
 (8) horizontal edge shadow on inclined
surface

compare with shadow type 6

(1) Perspective of Shadow Edge. Because the shadow

casting edge is horizontal, the horizontal edge
vanishing point is contained in the horizon line
(evp horizontal = vp1 horizontal;see shadow type
2). This is already given in the perspective
construction.
(2) Perspective of Light Plane. Insert
the lvp (solar/antisolar point) above or below the
horizon line by rotating the sun's altitude to the
horizon and azimuth to the direction of view.
Construct the lpvl vertical (for shadows from the
vertical edges of the supporting wall) as a vertical
line through lvp through the horizon line. Construct
the lpvl horizontal as a line
from lvp to evp horizontal (vp1 horizontal). The
inclined edge shadows are onto the horizontal plane,
so lpvl inclined is not required.

(3) Perspective of Surface Plane. The surface

receiving the shadow is inclined, therefore it is
necessary to construct the ipvlfrom the vanishing
point for two lines in the inclined plane. One of these
is typically its intersection with a horizontal plane,
which extends to vp2 horizontal; the other is found
by extending a side edge, then extending the station
points for this edge to the horizon line (vp3) and
constructing a vertical line from this point to identify
the inclined evp. Then a line
through evp and vp2 identifies ipvl.

The vanishing point for horizontal shadows on the

inclined plane (isvp horizontal) is found at the
intersection between the horizontal light plane
vanishing line (lpvl horizontal) and the inclined plane
vanishing line (ipvl). The vanishing point for vertical
shadows across the inclined plane
(isvp vertical,necessary for the vertical supports for
the horizontal edge, or for the pole in the diagram) is
found at the intersection between the vertical light
plane vanishing line (lpvl vertical) and the inclined
plane vanishing line (ipvl). The vanishing point for
horizontal shadows on a horizontal plane
(svp horizontal) is the horizontal vanishing point
(vp1 or vp2) in the horizon line.

(4) Construct Shadow Edge. The shadow edge is

constructed by starting at one horizontal edge
shadow end point and working toward the other,
using the appropriate shadow vanishing point. In the
diagram, the corner point a is found as in shadow
type 1, and extended in the horizontal plane
toward svphorizontal until it intersects the inclined
plane (b); from there the shadow edge is continued
toward isvp horizontal until it reaches the edge of
the plane (at c). In the diagram, the end of the
horizontal edge is out of view, on the back vertical
side of the inclined structure.

The shadow of the vertical pole on the inclined plane

is defined by two hidden points: a shadow vanishing
line from svp verticalthrough the hidden pole station
point to the hidden station point in the vertical plane
containing the inclined edge (x), and from there by a
hidden vertical line to the inclined edge point y. Then
the shadow edge extends along the shadow
vanishing line from isvp vertical through y to the
shadow termination line from lvpthrough the
pole ep; these intersect at z.
9. Inclined Edge Shadow on Inclined Surface.
This is a rare shadow type in any situation, but one
example would be an inclined roof casting a shadow
onto a lower inclined roof in a different orientation.

As background to the geometry of inclined plane

vanishing lines (lpvl), see the discussion of inclined
lines & inclined planes.

Construction Procedures. Elements already given:

inclined edge end points, horizon line, perspective
construction for all shadowing objects and shadow
receiving surfaces, two line segments contained in
the inclined surface. Key perspective elements and
their abbreviations are summarized here.
 (9) inclined edge shadow on inclined
surface

As a study aid, the diagram shows the inclined plane

vanishing lines and light plane vanishing lines for
both planes (labeled 1and 2). However, in the
example only the vanishing line for the edge casting
the shadow from plane 1 (lpvl [inclined 1]) and the
vanishing line for the plane receiving the shadow on
plane 2 (ipvl2) are necessary to construct the
shadow.

(1) Perspective of Shadow Edge. The edge casting

the shadow is in the inclined plane 1. The station
points for the vertical sides of this plane recede to
the horizon line at vp1, and the inclined edges
intersect a vertical line from vp1 at evp1. The second
edge vanishing point (evp2) is located on the horizon
line at vp2.1. (To finish shadows cast by the second
inclined plane on horizontal surfaces, vp2.2 is
included.)

(2) Perspective of Light Plane. Insert

the lvp (solar/antisolar point) above or below the
horizon line by rotating the sun's altitude to the
horizon and azimuth to the direction of view.
Construct lvpl inclined 1 (for shadows cast by the
inclined edges in plane 1) as a line
through evp1 and lvp. (Compare with shadow type
8.)

(3) Perspective of Surface Plane. Construct the

vanishing line for the second inclined plane, ipvl2,
using the methods described under shadow type 7.
Locate the shadow vanishing point for inclined
shadows on the inclined surface, isvp12, at the
intersection of ipvl2 and lpvl inclined 1.

(4) Construct Shadow Edge. Construct the shadow in

sequence, starting from the inclined edge station
point a, and from each plane edge intersection
(b and c) toward the appropriate shadow vanishing
point.

Study the diagram, and note the construction of the

three light plane vanishing lines lpvl; note also the
relative locations of the shadow vanishing point for
the inclined edge of plane 1 on horizontal
surfaces, isvp1, which is found at the intersection of
the light plane vanishing line for inclined
plane 1, lpvl inclined 1,and the spvl horizontal (the
horizon line); and the location of the vanishing point
for the inclined edge of plane 2 on horizontal
surfaces, isvp2, which is found at the intersection of
the light plane vanishing line for the second
plane, lpvl inclined 2, and the horizon line.

Irregular Shadow Edges. In all these perspective

examples the shadow casting edge is a straight line
and the shadow receiving surface is a plane. What
happens in situations where the edge or the surface
or both are curving or irregular?

There are no universally valid rules to follow here,

but there is usually a simple procedure you can use
to make the perspective construction. The best
procedure is always defined by two things: (1) the
rectilinear shadow type (from the nine shadow types
above) that the irregular shadow most resembles,
and (2) a method for projecting the irregular edge
into perspective space using a rectilinear framework,
for example as was done to project a circle onto a
plane.

shadow of a curving edge in a vertical

plane on a horizontal surface

The example above shows a common architectural

problem: an arch or vaulted opening casting a
shadow on the floor or ground.

In this case, the basic problem most resembles

shadows from edges in vertical planes cast onto a
horizontal surface, which is either shadow type
1 or shadow type 2.

After looking at the problem in the diagram (above),

the best way to project the curve of this arch seems
to be: trace the curve of the arch with separate,
approximately equally spaced edge points (ep, green
dots), and use vertical lines to project these edge
points onto the ground plane as station points (sp) in
the edge station line for the arch. Then use shadow
type 1to project the eps onto the ground plane as
separate shadow end points (x), using separate
shadow vanishing lines and shadow termination lines
for each point. Finally, connect the points in the
ground plane to construct the shadow edge.

Most curving edges will be contained in regular forms

that are drawn first, using standard two point
perspective methods, which means the vanishing
points and measure points for the perspective image
are already known. In that case, it would also be
possible to project a grid of squares over the surface
of the arch opening, using the measure point and
vanishing point to determine the correct amount of
foreshortening and recession. Then shadow type 1
would be used to project this grid onto the ground.
Finally, the grid could be used to reconstruct the
shadow projection, again as was done in the
demonstration of projecting a circle onto a plane.
This method is more convenient and more accurate
when the shadow edge is intricate or very irregular.

If the shadow receiving surface is irregular, for

example convex like the exterior surface of a dome,
then you must make a slice through the surface
along the shadow vanishing line for the vertical or
horizontal edge falling across the surface, then
project onto this line the profile (elevation) of the
curving surface, using the shadow vanishing line as
the baseline or station line for the profile. Then all
shadow construction lines drawn from the edge
points to lvp terminate along this profile edge, rather
than along the shadow vanishing line (the line from
the sp to svp) at its base.

For complex edges cast on complex surfaces, the

simplest solution is to abstract both the edge and
surface into a more basic geometric problem, solve
the basic shadow forms in this context, then draw
freehand an approximation of the shadow deviations
from this simplified edge.

shadows from local lights

The perspective geometry of shadows from local
lights — all artificial light sources, including fires,
lamps, street lights and illumination through windows
or reflected from architectural structures — differs
from solar shadows in three specific ways:

• the shadow vanishing point is replaced by

the normal point(s) for each shadow receiving
surface

• light sources are usually visually large and close to

the shadow casting edge, which requires us to
calculate the projection ratio for the
shadow penumbra, and

• there is usually more than one light source, and

the relative difference in luminance between the
primary and secondary light sources is always much
smaller than the difference between sunlight and
skylight in landscape settings.

The Normal Point. The diagram below shows the

case of an elevated local light such as a street lamp.
Now the vertical shadow vanishing
point svp vertical is not a vanishing point on the
horizon line but a specific point within the local
perspective space — the normal point.

the geometry of the normal point

The normal point is the point on a plane that
is physically closest to the light source: that is, a
line from the normal point perpendicular to the
surface plane will intersect the physical center of the
light source. For horizontal planes, the perpendicular
line will be vertical, so the normal point is directly
underneath the light source (diagram, above). Thus,
the svpvertical in landscape settings is actually on a
great circle, from the viewpoint to the location on the
earth where the sun appears directly overhead.

The essentials of shadow construction remain the

same, however. The shadow vanishing line is still the
intersection of two planes, the light plane and the
surface plane, which is always drawn from the
normal point through an edge station point; and the
length of the shadow is still determined by a shadow
termination line from the center of the local light
source to the end of the edge, as described
for shadow type 1.

The major complication is that each interior wall,

floor, ceiling and furniture surface creates a
separate normal point for all shadows cast onto it
by each light source in the room. The diagram
(below) shows the simplest arrangement: a single
ceiling fixture within an empty room, but if we add
several light sources and pieces of furniture, the
problem can become quite complex.
 normal points for an indoor light source

For each architectural surface, the normal point acts

as the shadow vanishing point for shadows cast upon
it. The diagram (below) shows the case of two
separate shadow casting posts fixed into the floor
and a wall, and the shadow vanishing line (from the
normal point through the perpendicular point or
"station point" sp ) and the shadow termination line
(from the light source through the edge end
point ep) in each case.
 indoor cast shadows on a single surface

The diagram (below) shows the case of horizontal

edges that are either parallel, perpendicular or
oblique to the wall and/or floor. In this case the
oblique horizontal shadows can be constructed using
methods similar to those for shadow
type 5 or shadow type 3, or the points can be
plotted out at each surface edge (floor, ceiling or wall
corners) using principles for horizontal or vertical
shadows cast on horizontal or vertical surfaces
(shadow type 1, type 2 or type 4).
 indoor cast shadows on multiple surfaces

In the diagram, the horizontal bar ab is oblique to

both walls behind it. Drawing its shadow, shadow
vanishing lines are extended from the floor normal
point through the edge end station points
(sp1 and sp2) to the surface edge (baseboard). From
there vertical lines are constructed to meet the
shadow termination line from the light source
through the end points aand b: these intersect at the
shadow end points a' and b'.

Rather than construct the inclined shadow vanishing

point necessary to define the oblique horizontal
shadows, the simpler approach is to locate the
shadow transition at the corner between the two
walls. To do this, a line is extended from the normal
point to the corner point (x); this intersects the
horizontal edge station line at y; a vertical from this
point locates the horizontal edge point (c) that casts
its shadow into this corner. A vertical
from x intersects the shadow termination line (from
the light source through c) at c'; the shadow is
completed by drawing the two edge
segments a' to c' and c' to b'. Note that, because
the shadow distance increases, the shadow and
penumbra at the corner point c' are slightly larger
than they are at a' or b'.

The Penumbra. With very few exceptions, light

sources are not point sources — they have a
perceptible visual size — and therefore the shadows
cast by local lights (and by the sun or moon) have a
characteristic penumbra or diffuse edge, caused
when light from opposite sides of the light source
creates overlapping shadow edges.

These overlapping edges produce a continuous gray

scale, from the shadowed to sunlit value of the
shadow surface, that depends on the visual angle of
the source as seen from the shadow edge (diagram,
below).
 projection ratio and penumbra width

The visual size of the light source is essentially

the source diameter divided by the source
distance from the shadow casting edge. The exact
method for calculating the source visual size is given
here, but it is more convenient to express this angle
in radians, as the projection ratio of the source at
the shadow casting edge:

projection ratio = visual angle * (π/180)

The projection ratio becomes larger as the light

source diameter is greater or as the source distance
is less. But once the projection ratio is determined,
then the penumbra width is the projection ratio
times the distance between the shadow casting edge
and the shadow receiving surface, or the shadow
distance:

penumbra width = projection ratio *

shadow distance

The penumbra always straddles the shadow

boundary that would be produced by a point light
source at the physical center of the actual light
source (dotted line, diagram above). Half of the
penumbra width is inside this point source shadow
boundary, reducing the area of complete shadow;
half is outside this boundary as an area of partial
shadow.

The calculation for penumbra width applies to a

shadow receiving surface perpendicular to the
direction of light; if the surface is at an oblique angle
to the light, then the projection ratio must be divided
by the cosine of the amount of deviation from a
perpendicular angle (cosine(0°) = 1.0). Otherwise,
the projection ratio is constant for all edges under
solar illumination, which means the penumbra width
is dictated by the shadow distance. As the shadow
distance increases, the penumbra increases
proportionately (image, below).

the penumbra is proportional to the

shadow distance

in a cable railing system, the higher cables cast a

farther shadow, producing a larger penumbra

The term "projection ratio" emphasizes that the

penumbra is actually a kind of image — a gradient
image of the visual diameter of the light source. As
with any image, the size of the image depends on
how far it is projected. In the demonstration below, a
pinhole image of the sun is projected either a short
or long distance, and as the shadow distance
increases, both the diameter of the sun's image and
the penumbra width increase equally.

penumbra width and pinhole image of sun

This image quality of the penumbra causes it to

mimic the shape of the light source; this is easiest to
see in the shadows cast by an oblong window or a
single fluorescent light fixture (diagram, right).
Because the sun forms a round pinhole image, its
penumbra has a constant width regardless of the
orientation of the shadow edge; however the tonal
gradient of its penumbras forms an ogive or "S"
curve, steepest in the middle, which gives solar shadow of a square object cast
by a fluorescent light fixture
shadows a relatively distinct inner boundary
between dark and light, especially when the
penumbra is small (image, above).
The importance of the penumbra to a perspective
image depends on several factors — the visual
angle of the light source, the distance the shadow is
cast, and the distance of the viewer from the shadow
edge — and these change with the specific situation.
There is no hard and fast rule as to when a
penumbra significantly alters the appearance of a
shadow edge.

However, as explained above, the projection ratio

for solar shadows is nearly constant at 0.5°, or a
projection ratio of about 0.009, so the width of all
solar penumbras is roughly 1% of the shadow
distance. For very large shadow distances, for
example in the shadows cast by cliffs or multistory
buildings, this projection is substantial. The
observation deck of the Eiffel Tower is about 1000
meters above the ground, so the penumbra of its
noon solar shadow is more than 10 meters wide. But
when viewed from an airplane at an altitude of 3000
meters, the shadow appears clearly defined.
 viewing distance and penumbra visibility

solar shadow of a sculpture viewed from 1/2, 1 and 2

times the object height

Most objects illuminated by sunlight are closer to the

ground than we are to their shadows, and solar
penumbras form a relatively sharp border on their
interior edge, so the 1% penumbra is visually
insignificant when the viewing distance to the
shadow is at least twice the shadow
distance (photos, above). This rule applies to
all objects located on the ground and contained
within a 60° circle of view or less; the same
distance rules that eliminate perspective
distortions also disguise the penumbra in solar
shadows. At most viewing distances, the sun can be
treated as a point source.

The rule for local lights is more complex, but in

general the penumbra is significant for artificial
lights. This is because local lights (1) are physically
large, (2) are fitted with shades, reflectors or covers
that enlarge the visual size of the light source or
diffuse the light it emits, and (3) are contained in
rooms or passageways that constrain the viewing
distance to the shadow. These factors usually offset
the small source distances of local lights. The
distinctly fuzzy edges that result are more restful to
the eyes, but also give interior light shadows a
distinctive appearance (diagram, below).

differences between solar and local light

shadows

If diffusing shades are removed, local lights can have

moderate to small projection ratios. For example, a
standard frosted incandescent light bulb, 5 cm wide,
has a projection ratio of 5% at 1 meter and 1% at 5
meters. But local lights are not normally viewed in a
"naked" mode.

Much indoor lighting is created by partially cropping

exterior daylight (blue sky or reflected sunlight), or
by artificial light sources that are close enough to
present a significant angular size. In either case the
light source will cast a diffuse shadowwith a very
large penumbra. Most diffuse shadows are subtly
complex, and it is impractical to recreate them using
perspective construction. However, a knowledge of
perspective geometry can be used to define the outer
limits and dark core of a shadow, and can guide your
observation of the shadow so that you can reproduce
it freehand more accurately.

the geometry of indirect light

The example shows a daylight window opening. The

labels identify a single light point A in the window
and two points, a1and a2, at the top of the two
shadow casting sides of a cylinder. These points cast
two shadow points, s1 and s2, on the horizontal
surface. Their locations are found using the same
technique as for vertical solar shadows on a
horizontal surface (shadow type 1). Point A is the
light vanishing point, and point C, on the horizontal
plane directly below A, is the vertical shadow
vanishing point (svp vertical). Two lines from the
base of the cylinder through these points define the
edges of the cylinder shadow cast from all points
between A to B in the window.

The diagram also shows three additional mappings

from three other points in the window (D, E and F),
which produce three similar, overlapping shadows
behind the cylinder from different locations of the
shadow terminator on the cylinder surface.

If it were feasible to map all points across the

window in this way, the result would be a large
penumbra area of diffuse or overlapping shadows.
This composite shadow is lightest at the edges and at
the extreme end (farthest from the cylinder), which
receives some direct light from some parts of the
window (the penumbra), and darkest at the center
just behind the base of the cylinder (the umbra),
which receives no direct light from any part of the
window. All diffuse shadows have a penumbra, but
they only have an umbra if the shadow casting
object is close enough to the shadow receiving
surface to completely block all light from all parts of
the light source. The umbra is typically uniform in
darkness and, for rectangular sources such as a
window, the shadow lightens continuously between
the edge of the umbra and the edge of the
penumbra.

All diffuse cast shadows can be solved in a similar

way, first by plotting the extreme outside edges of
the shadow from the extreme top, side and bottom
points of the large light source, which will show the
boundaries of the penumbra and umbra, then
gradating the difference between the umbral and
penumbral values from the edge of the umbra out to
the edge of the penumbra.

Luminance Contrast. The amount of brightness or

lightness contrast between the diffuse shadow and
the illuminated part of the horizontal surface
depends on the total amount of light coming from
the diffuse light source (the window), the proportion
of light coming from other sources in the room,
including reflected light from the walls and ceilings,
and the light adaptation of the eye.

The appearance of artificial light shadows cannot

easily be presented as specific principles, but a few
rules of thumb are helpful:

• Interior shadows are not nearly as dark as solar

shadows, but in general shadows become darker
when there is only a single source of illumination and
the illuminance from that source is very low (for
example, candlelight). As rooms become brighter,
the light intensity pushes the eye into a larger
luminance response range, and interior surfaces
reflect more light back into the room.

• Interior shadows are more distinctly colored by

light reflected from other surfaces, and especially by
skylight entering the room from a nearby window

• By design, interior light fixtures diffuse light to

soften shadows, as this is more restful and pleasing
to the eye; as a result, interior shadows are
generally very diffuse, and the actual extent of the
shadowed area may be much smaller or of a different
shape than the perspective construction indicates it
should be. In other words, the more complex the
shadow, the more important observation becomes in
rendering it accurately.

perspective of reflections
Reflections are the linear perspective of an image
transformed by the shape of a surface. If the surface
is flat, like the surface of still water, then the image
is reversed — top to bottom, for water reflections,
and left to right for vertical mirrors. If the surface is
curved or perturbed, then the reflection is both
reversed and distorted.

Reflections in Mirrors or Windows. The rules for

plane (flat surface) reflections are
simple: everything is governed by lines
perpendicular to the mirror surface. That is,
identical points in the object and its reflection lie on a
single line perpendicular to the mirror surface, and
the reflected object appears at the same distance
behind the mirror as the actual object is in front of
the mirror.
The optics of this situation are shown at right, for an
observer at a viewpoint looking obliquely into a
reflecting mirror lying in the plane ax. (We look
edgewise at this plane, to see front and back in the
same way.) The visual rays are reflected at the
mirror surface, and in this reflection the size of the
visual angle 1before reflection and angle 2 after
reflection are on opposite sides of a line
perpendicular to the mirror surface and are exactly
the same size, so the object appears as if in a
straight line path through the mirror and into the
virtual space on the opposite side. This produces a
view of the object as if it were seen from the virtual
viewpoint, not the actual viewpoint, and with the geometry of mirror reflections
object reversed (left to right) in a direction that is
also perpendicular to the mirror surface.

This geometry also holds for objects that

are not reflected in the mirror, for example the
viewer. The virtual viewpoint is as far behind the
mirror plane as the actual viewpoint is in front of it,
and both are on a line perpendicular to the mirror
plane around the point x.

The geometry for vertical mirrors usually must be

worked out as part of a perspective drawing. In the
simplest problem, the mirror is hanging on a vertical
wall, or the reflection is in a window or glass wall.

the geometry of vertical (mirror or

window) reflections

This wall has its own vanishing point for the floor and
ceiling edges (vp2), which creates a second vanishing
point for lines perpendicular to the mirror (wall)
surface (vp1). If one or both vanishing points are not
already defined, they can be found using the
methods for circle of view reconstruction. It is
also necessary to find the mirror station line,
which is usually the base along the floor of the wall
supporting the mirror or window, or the ground line
of the exterior foundation supporting the reflecting
window or door.

To construct the reflection of an object in front of the

mirror (such as one of the four colored spheres), first
construct a line from the object's station point (for
example, a under the red sphere) back to the
perpendicular vanishing point vp1. This line is
perpendicular to the mirror plane and it defines an
intersection with the wall station line (b for the red
sphere).

Now you must find the same distance ab behind the

mirroralong the perpendicular line a-vp1. This is
found by taking a line from the appropriate measure
point (mp2) — which is always a measure point
defined by the mirror plane vanishing points and not
any reflected object vanishing points — through
point buntil it intersects a line through a parallel to
the image plane: this defines point c. The line
segment ca is the distance bawithout foreshortening.
Recess this length to the perspective depth of
point b by drawing a line from c to vp1 until it
intersects a line through b parallel to the image
plane at point d. Foreshorten this distance by a line
from d to mp2; the intersection e of this line with the
line a-vp1 is the virtual station point of the
reflected object. The rest of the image can be
constructed from this point.
The Virtual Viewpoint and Station Point. This
method can be repeated for other objects reflected in
the mirror or glass, or it can be done once only, to
identify a reflected image and its station point, which
then can be used to find the virtual viewpoint and
virtual station point. These are useful because they
define the location and size of every reflection in the
mirror:

• a line from the object station point to the virtual

station point (of the viewer) intersects the mirror
station line directly under the object's virtual
station point. In the diagram, the line from a to the
virtual station point intersects the mirror station line
at x, which defines a vertical line to e, the reflected
image of a. In the same way a line from 1 at the
base of the pole to the virtual station point intersects
the mirror station line at 2.

• lines from the top and bottom of the object to the

virtual viewpoint define the vertical size of the
reflected object. Thus, the two lines from the virtual
viewpoint to a and g define the location of the
points e and f in the reflection where the lines
intersect the vertical line from x; a line from the
base of the pole 1 to the virtual viewpoint defines
the base of the pole 3where the line intersects the
vertical line from 2.

In the diagram above, the red sphere defines its

station point aand its reflected virtual station point e.
A vertical line from e to the mirror station line
defines a new point x. Then a line
from athrough x will intersect a vertical line
from vp1 at the virtual station point. Notice how
point y determines the location of the pole reflection,
and the height of the pole reflection is found by two
lines from the ends of the pole to the virtual
viewpoint.

Note that the virtual viewpoint is the vanishing

point for mirror perpendiculars (pvp). This is
obvious when you think about it: because all the
mirror perpendiculars are all parallel, they all recede
to a single vanishing point (perspective rule 6). A
vanishing point is simply a visual ray viewed head on
(perspective rule 1), and the only mirror
perpendicular you can view head on is the reflection
of your own viewpoint (that is, the reflection of your
eyes). All mirror perpendiculars always recede to
that point, even when your reflection is not visible in
the mirror.
In addition, all lines that intersect the viewer's
station point are perpendicular to the horizon
line in the image plane. So in the diagram above,
the station point is found on a vertical line from vp1!
This may go against your intuitions, but it is another
perspective "distortion" that corresponds exactly to
the visual facts.

How far below the horizon line is the virtual station

point? You can find it by constructing the reflection of
a single object, as described above, or you can
construct it more accurately from the intersection of
the mirror station line with the vertical line from the
virtual viewpoint (point s in the diagram at right).
Simply draw a horizontal line from s to the median
line to find x. This creates a triangle with sx as its
far side, s/viewpoint as its hypotenuse
and x/viewpoint as its base. (Remember, the how to locate the virtual station
vertical line from pvp, although it seems to be point behind a vertical mirror or
window
parallel to the median line, is actually at a
converging angle to the median line, and both lines
intersect at the viewpoint.)

To find a triangle of exactly the same size and

proportions in perspective, you draw one line
from s to the principal point pp, and a second line
from x to the perpendicular vanishing point (virtual
viewpoint) pvp. These two lines intersect at z, which
is at the same distance from the image plane as the
virtual station point. So draw a second horizontal line
through z to intersect the vertical from pvp: this
intersection is the virtual station point, vsp.

What if you look directly into the mirror, so

that pvp (virtual viewpoint) = pp? In that case, the
point s is located at the intersection of a vertical line
from your left or right dvp down to the mirror station
line, and the point x is at the intersection of the
median line with the mirror station line.
Connect s to pp and x to dvp to find z, and from
there a horizontal line back to the median line
locates vsp.

Reflected Vanishing Points. I used spheres in the

first example to avoid the distracting issue of
reflected vanishing points. Now we can tackle these
using the example of a reflected rectangular box. In
the diagram below, the mirror (purple rectangle) lies
in a vertical plane which intersects the horizon line at
the mirror vanishing point (mvp, magenta). The
angle of the mirror plane across the direction of view
is shown by the visual ray from mvp to point v at
the top of the circle of view; this angle is close to
45°. At 90° to the mirror vanishing point is
the perpendicular vanishing point (pvp,
magenta), the vanishing point for all perpendiculars
to the mirror plane which, as we've seen, is also the
virtual viewpoint. on the median line.
 the geometry of reflected vanishing points

In real space the rectangular box recedes to

vanishing points vp1 and vp2 (blue); in virtual space
the reflected box recedes to the reflected vanishing
points (vp1) and (vp2) (yellow).

The virtual vanishing points are found by rotating

the vanishing points around v. The location
of vp1 is shifted to the right of the mirror plane by
the angle z shown at the top, so the virtual position
of (vp1) must be shifted to the left of the mirror
plane by the same amount.
Usually vp1, vp2 and mvp are already known by
construction in your perspective drawing. So to
construct the reflection, simply draw lines from these
points to the point v at the top of the circle of view,
measure the angle z with a protractor centered on v,
and make the rotation by measuring angle z on the
opposite side of the line to mvp. Finally, locate (vp2)
as a line from v at a right angle to the line from
(vp1).

To position the reflection, draw a line

from vp1 through a bottom corner of the box (x1,
magenta), and into the mirror station line (the line
going back to mvp) to define point R. Then draw the
reflection of this line from point R back to (vp1).
Next, draw a line from the same corner x1 to pvp;
this will intersect the first line at x2, the virtual
location of the bottom corner of the reflected box.
Draw the second reflected vanishing line from x2 to
(vp2). Finally, draw a line from the opposite
corner y1(green) to pvp; this line will cross the front
vanishing line of the reflected box at y2. You now
have the location and size of the front bottom width
of the reflected box.

Determine the height of the reflected box by a third

line from a top corner of the actual box to pvp, and
intersect this line with a vertical from x2. Determine
the depth of the reflected box using a line from a
back corner of the actual box to pvp; the corner will
be located at the intersection with the reflected
vanishing line to (vp1). Complete the front and sides
of the box with virtual vanishing lines from these
points to (vp1) and (vp2) as appropriate, using the
steps described for the 2PP example.
The vanishing points for edges that are parallel or
perpendicular to the mirror plane reflect to their
original locations, so they do not have to be rotated.

Finally, if the mirror plane is at an angle to the

direction of view and the mirror is tilted forwards or
backwards from that plane (for example, a mirror
leaning backward against a wall or hanging forward
from a loose length of picture wire), then the
perspective problems become more complex,
because vertically oriented objects will appear
reflected in three point perspective (3PP) and these
three points must be determined separately for all
objects in 2PP with unique vanishing points. These
problems are too complex to delve here.
Approximate them freehand, or trace a photograph
of a similar setup.

Reflections in Water. After dipping our toe in the

riptides of plane mirror reflections, water surface
reflections are easy swimming. (Isn't it annoying
when people write in multiple metaphors?)

Gravity ensures that the surface of still water is

always perfectly level. This means all the mirror
plane vanishing points are in the horizon line. The
perpendicular vanishing point (virtual viewpoint) is
directly under the viewer's feet, which means it is
never visible on the image plane. Vertical lines define
all reflection perpendiculars. Finally, most buildings
are constructed plumb and level, so the building
vanishing lines are either parallel or perpendicular to
the reflecting plane, which means they do not have
to be reflected (rotated). No pvp, no reflected
vanishing points ... what could be easier?

the geometry of horizontal (water)

reflections

Well, there are two nuances here and they both

involve the vertical height of the reflection in the
water.

First, the height is always judged from the reflection

plane and not the base of the object itself. In the
diagram, the end point ep of the pole is not reflected
around its station point sp but around its station
point in the plane of the water surface, spW.

For objects standing along the edge of a gently

sloping beach or marsh this is not a distraction,
because the ground level is roughly the same as the
water level. However the diagram shows the pole
standing on a raised embankment, and this distance
has to be included in the total length that is included
to find the reflected end point, epR.

The second nuance is that the object and its

reflection will have an identical height only if the
direction of view is nearly parallel to the water
surface. This is the case in those dramatic postcard
views of huge mountains seen reflected in the
surface of an enormous lake. In the example above,
however, the reflection of the pole is clearly viewed
at a downward angle, which will make it appear
foreshortened in comparison to the actual pole.

No problem: because the perpendiculars are parallel

to the image plane, their measure points are
the diagonal vanishing points or distance points
(dvp) at the top or bottom of the circle of view. Just
use these diagonals to determine the correct height
of the reflection, using the methods described
here for projecting a measure bar or unit dimension
into perspective space.

That is, (1) draw a line from spW to any convenient

point along the horizon line (the example uses vp1,
but any point will do), (2) project in depth the
vertical distance to be reflected with a line
from ep to a dvp to define point R1, (3) draw a
vertical line through R1, (4) get the location of R2 by
a diagonal from spW to the same dvp through the
vertical line, and (5) locate epR by a line from your
vanishing point (vp1) through R2 to a vertical line
under spW. Piece o' cake.

Reflections in Rippling Water. Water surfaces that

are disturbed by waves or ripples create very
complex reflection patterns. Here the issue is not
reconstructing them with geometry, but
understanding how to interpret the reflections
visually so that you can approximate their poetry
freehand.
 reflections from waves (side view)

This diagram shows the surface of water across two

wave fronts, viewed from the side. The parallel gray
lines approaching from the right represent lines of
sight to the viewer (which at most landscape
distances are nearly parallel for the small angular
size of most ripples or waves). The source of these
lines in the landscape are shown as blue lines. The
vertical bars at the right represent the appearance of
the waves or ripples in vertical section, with the
reflection from the crest of each wave (c) indicated
by a black dot.

The top example shows the appearance of the waves

when the angle of view is close to parallel to the
water surface — that is, the waves or ripples are
viewed from a great distance and/or from a
viewpoint along the shore of the water. In this
situation the crest of each wave sharply crops the
view of the wave behind it, so reflections skim the
tops of the waves in tight visual spacing, shown at
right as closely alternating bars of brown (far
landscape) and blue (low sky). At this angle the
inverted reflections of landscape features on the far
side of the water are recognizable, or the water takes
on an indistinct, darker color.

The middle example shows the wave viewed from a

slightly sharper downward angle. In this case more
of the wave front reflection is visible, and it is upside
down: the lower part of the wave reflects a higher
location in the sky, indicated by a darker blue bar.
This produces a sharp contrast between the crest of
one wave and the forward surface of the wave
behind it, and the reflections of far objects are lost in
the appearance of alternating bands of light and
dark.

The bottom example shows ripples or waves from an

angle where everything except the back side of each
undulation is visible. Now two things happen. The
numbered dots (1 to 12) show that the reflections
first cycle up into the sky, then angle back towards
the earth, so that each ripple reflection is an inverted
image of earth, low sky and high sky (brown, white
and blue bars at right) joined to an upright image of
sky, low sky and earth. However, at the point where
the reflections join, the angle of view into the wave
front is close to perpendicular to the water surface,
so the reflection is weakened and the line of sight
may actually penetrate the water (points a and b),
darkening the reflection and shifting its color towards
blue green.

reflections from waves (top view)

This diagram shows the waves or ripples viewed from

above, and in this view waves can take two forms. If
the undulations are parallel wave fronts and are not
perpendicular to the direction of view, which is the
typical case in still water disturbed by a prevailing
wind or passing water craft, then reflections appear
as repeated, evenly space wiggles. If the waves are
actually intersecting each other from many different
angles, which is the typical case in open water, they
create a pattern of roughly circular undulations of
different sizes. When viewed from an acute angle,
these produce complex reflection patterns of wiggles
within wiggles, repeated at irregular intervals. If the
reflection is from a point source, such as the moon or
a streetlight, the reflections seem to break apart and
form complex rings or orbits in the water.
These irregular ripples, the darling of marina
painters, are less common in landscape views.
Instead, ripples of all types (but especially parallel
waves or ripples created by a steady prevailing wind)
produce an effect called anisotropic reflectionsor
simply vertical smearing, which only affects
the reflected vertical location of horizontal
features, rather than reflected horizontal location of
vertical features. So the sides of the sun or the
vertical bridge pilings and arcade columns have the
same horizontal (left to right) location in the
reflection, making their edges relatively crisp, while
the horizontal top and bottom of the sun or
horizontal bridge span and arcade arches have
smeared out vertical locations in the reflection
(image at right).
In most cases the sky is pretty much one color and
the land silhouette a second color, which creates a common landscape artifacts
simple two tone pattern to the reflections. The areas from large ripples or waves

of sky are called skypoolsand the areas of land (top) vertical smearing of horizontal
reflections landpools (image at right). The landpools features, such as glitter from the
dominate at far distances close to the opposite sun or the span of a bridge;
(bottom) interweaving areas of
shore, and skypools dominate at near distances close
landpools (dark) and skypools(light
to the observer. The transition between the two orange)
areas appears as a complex interweaving or
(adapted from Lynch & Livingston,
overlapping of roughly circular areas of reflection,
2001)
which represent the troughs of the irregular circular
waves shown in the illustration above. These roughly
circular troughs appear to be stretched out because
we view them from the side, which causes extreme
foreshortening.

The optical effects of water reflections are a deep

and fascinating study. I strongly recommend you
consult the chapter "Water and light" in Color and
Light in Nature (Second Edition) by David Lynch
and William Livingston (Cambridge, 2001).

Reflections in Curved Surfaces. Finally, there is

the problem of reflections from curved surfaces such
as cylinders (pipes, steel columns), spheres, bubble
windows and the like.

I have made a diligent search of the literature, and

the solutions I have found for creating images of
convex or concave reflections consist either of
proprietary computer animation software or
incredibly tedious ray tracing procedures. The best
solution for an artist today is either to trace or copy a
photograph of the effect, or make a patient freehand
drawing. In that case, a general expectation of the
distortion can help guide your view.
M.C. Escher's Hand with reflecting sphere (1935,
right) beautifully demonstrates the general form of
spherical reflections, and is in a tradition of
anamorphic portrait studies going back to
Parmigianino's Self portrait in a convex
mirror(c.1524). This lithograph is precisely
informative about the optical distortions imposed on
the reflection of the surrounding room (his studio in
Rome) and displays the curvature of perspective
lines in the borders of the ceiling and walls and in the
bookshelf along two walls. (The globe is positioned
below the bookshelf against a wall, with the closest
room corner on the right and the entire room and
illuminating window behind the artist.) It also shows
how the spherical reflection magnifies objects very
close to it and diminishes objects farther away.

This complex image would be far easier to draw "Hand with Reflecting Sphere"
freehand, or by tracing a photograph, than through
by M.C. Escher (1935; 32cm x
laborious perspective construction. But as a 21cm)
construction stunt, the major landmarks of the room
(the ceiling corners, the window frame, the
bookshelves and the location of the chairs) could be
mapped by hand into the sphere outline using the
transversals of the spherical projection template.
The rest of the drawing could be filled in from those
reference points.
These distorted reflections can be "reverse
engineered" by anamorphic images, which are two
dimensional, unrecognizable drawings that become
recognizable if viewed from an extreme oblique angle
or in the reflection from a cylindrical or spherical
mirror. These were very common in the 17th and
18th centuries, both as perspective curiosities and as
examples of the projective distortions of interest to
mapmakers and geometers.

Scanned into a computer, Escher's image (for

example, the bookshelves and ceiling boundaries)
could be "reverse engineered" by animation software
into a precise map of the room reflected there, and I
assume that some aspiring student of art history will
do the exercise simply to find out whether the image
consists of a traced photograph or contains the
inevitable systematic distortions of a freehand
drawing. (From what I know of Escher's esthetic, I'm
betting on freehand.)

aerial perspective
Aerial perspective is not really perspective at all, but
is the effect of the atmosphere on the appearance of
distant landscape. Leonardo da Vinci originated the
term when he wrote "There is a kind of perspective
called aerial perspective which depends on the
thickness of the air," and he contributed important
observational studies of the phenomenon. Since
then, aerial perspective has often been cited as a
feature of distant views that linear perspective could
not explain. But that belief is incorrect: aerial
perspective is exactly governed by linear perspective
geometry in the same way that a perspective
gradient is.

Aerial perspective is partly caused by the Raleigh

scattering of light by individual air molecules, but
this effect is sometimes overwhelmed by the
scattering due to particles of smoke or dust, and
molecules of water, mixed in the atmosphere. The
scattering due to smoke or dust is highly variable,
and if extreme can affect the color of objects at a
distance of a few dozen meters; the Raleigh
scattering has more consistent perspective limits.

As the diagram shows, the Raleigh scattering

increases linearly with distance (roughly double the
effect at twice the distance), and is usually first
noticeable at a distance of about 750 meters or
more. In the standard 90° circle of view the viewing
distance (usually 1.5 meters) has a fixed
location on the image plane, just half (50%) of the
distance from the ground line to the horizon line.
This locates an object on the ground plane that is
750 meters from the viewer at 99.9% of the distance
from the ground line to the horizon line. So aerial
perspective is a horizon feature, and does not apply
to objects that appear visibly lower than the horizon
line. Any atmospheric effect that occurs below the
horizon can only be due to mist, fog or smoke.

the geometry of aerial perspective

The two effects traditionally attributed to aerial

perspective are (1) a shift toward blue color as
objects are farther away, and (2) a blurring of object
edges. Both statements are inaccurate — especially
the first, which was formulated in those faraway days
when smokestack manufacturing, coal generation
and millions of cars did not exist.
The atmospheric color shift depends on the
atmospheric contents: smoke, dust or smog shifts
the color toward white, yellow, brown or gray, while
water vapor in clean air shifts the color toward blue,
an effect called airlight (illustration at right). This
landscape contrast is especially evident between
winter and summer months in desert areas such as
Arizona or New Mexico.

The veiling due to particles of smoke, dust or mist

shows a significant increase in effect closer to the
ground, because these particles tend to sink in the
air. Thus, the top of a distant mountain will appear
darker (less veiled) than its base, and this contrast
increases with distance (right). The veiling due to
water vapor and air molecules is much more evenly
distributed from the ground upward, and there is
little or no effect change across distance. aerial perspective from smoke or
dust (top) and from water vapor
The veiling effect also depends on the angle of solar (bottom)

illumination; water molecules and dust or smoke

particles mostly scatter light forward or backward
along the path of the light ray, so the veiling appears
most intense looking into a low sun or with the sun
at your back. Air molecules scatter much more light
in directions oblique to the path of light, so the
Raleigh scattering is little affected by time of day;
this is also why the sky appears to be roughly the
same blue color in all directions.

Aerial perspective of any kind does not cause

perceptible blurring of the image — which is obvious
if you examine the daytime horizon with a telescope.
Veiling acts instead to obscurethe image, producing a
fading or reduction in lightness contrast so that
whites and blacks tend to fuse toward a middle gray.
This lack of contrast makes all edges less distinct and
increases the visual fusion of tiny details or textures
into a homogenous color.

Atmospheric blurring is almost always the result

of thermal turbulence, the same phenomenon that
causes stars to twinkle at night. Rising currents of
warm air create a mixture of tiny pockets or cells of
warmer (less dense) air and cooler (more dense) air,
and these different air densities refract or scatter air
at slightly different angles, acting like tiny lenses or
prisms to distort the image passing through them, in
the same way images are blurred by textured shower
doors or rippling water.

The takeaway is that aerial perspective arises from

several different causes. All of them increase in
effect with distance, but each of them produces
subtly different effects on the color, contrast, image
clarity and vertical consistency of the horizon
backdrop for landscape painters.

rainbows
To conclude these perspective studies, it's
appropriate to look at rainbows. Rainbows became a
significant theme in western art during the late 18th
and early 19th centuries, when academic perspective
rigor was displaced by a new interest in the
atmospheric effects of light and clouds. And so they
signal the end of our perspective labors.

The rainbow geometry is simple and everywhere the

same: the essentials were clearly described in
Aristotle's Meteorology about 2400 years ago. Once
you understand this geometry it's truly amazing how
many perspective howlers you can find in 18th and
19th century paintings — rainbows that are too wide,
too narrow, curve with an incorrect radius, don't
match the position of shadows or the sun, or are
distorted by bogus foreshortening. These distortions
often look odd even if the viewer cannot explain why,
and they usually do not produce a pleasing effect.
 the geometry of rainbows

shown within a 90° circle of view; the sun is about 20°

above the horizon (late afternoon)

The rainbow is a circular band of refracted light

centered on the antisolar point (asp), the point in
the sky directly opposite from the sun (diagram,
above). It results from the refraction and oneinternal
reflection of light inside a raindrop, as described in
the page on the causes of color.
The rainbow always has a radius (at its outer
circumference) of 42° visual angle, and the spectral
band is roughly 2° of visual angle wide (the width of
four full moons). Colors are arranged in spectral
order from red (at the outer edge) through yellow,
green, blue to violet (at the inner edge). A useful
mnemonic is Red on the Rim, Blue in the Bowl of the
Rainbow.

Sometimes a second, fainter rainbow appears outside

the first, the same width as the primary bow, with a
radius of 51° visual angle from the asp to the inner
edge. This secondary bow results from two internal
reflections of the light ray inside the raindrop. The
second reflection makes this second bow a mirror
image of the primary bow, so the color order is
reversed: red on the inner edge and blue to violet on
the outer circumference.

The perspective observer always appears to stand at

the center of a celestial sphere, so the antisolar
point below the horizon is always located at the
center of the shadow of the observer's head on the
ground.

The vertical shadow vanishing point (svp vertical) is

located on the horizon on a vertical line from the
antisolar point; the svp is always directly underneath
the highest point in the curve of the rainbow and on
level ground it is centered between the two rainbow
ends.
Raindrops between the apparent position of the
primary and secondary bows do not refract light back
to the viewer, so this area appears noticeably darker
than the rest of the sky, an effect
called Alexander's dark band (diagram, above).
The area contained within the primary bow almost
always appears significantly brighter (as well, to a
lesser extent, the area outside the secondary bow)
due to random white light scattering (image, right).
Because this internal light is spectrally neutral, it can
be tinted with any bias in the color of sunlight —
typically the reddening of light at sunset. Sunrise or
sunset rainbows also display significantly weakened
green and violet bands, as these wavelengths are
already filtered out of the light.

All foreground shadows will seem to recede into the

rainbow (a beautiful effect in a field of tall grass or a
desert landscape of low brush), as shown by the two
foreground posts (diagram, above). J.M.W. Turner,
an extraordinarily observant artist and professor of
perspective at the Royal Academy, was harshly
critical of Peter Paul Rubens's Rainbow Landscape (in
internal brightening visible
the Wallace Collection, London) because Rubens against a dark sunset sky
showed foreground shadows raking sideways in front
of a rainbow in the distance. However, as the far
right post in the illustration shows, distantobjects at
either end of the rainbow can be displaced so far to
one side of the svp vertical that their shadows do
seem to fall almost parallel with the horizon.
We can never step to one side of a rainbow. A
foreshortened rainbow, for example an arc from
beyond the horizon that lands in a pot of gold at your
feet, is optically impossible. The rainbow is
always centered on the shadow of the
observer's head, so the observer's apparent
position is always directly in front of a perfectly
circular rainbow arc. This is why rainbows appear to
travel alongside an observer in a moving car or train. the angle of a rainbow to the
For the same reason, the sun and a rainbow can ground depends on the height of
never appear in an image at the same time. Shafts of the sun in the sky
sunlight (for example, through the overhead clouds (left) rainbow near sunset, (right)
of a departing storm) will always appear to rainbow in early afternoon
converge into the antisolar point at the center of the
arc — another breathtaking atmospheric display.

Seen in nearby waterfall mists or the spray from a

garden hose, rainbows can appear nearly circular.
However, when the sun is above the horizon (as it
must be to light a landscape rainbow), the antisolar
point is below the horizon, so the ground plane
always cuts off half or more of the circular display.
Rainbows seem to straighten and fall closer to the
ground as the sun rises higher in the sky. At sunset
or sunrise the rainbow can approach a semicircle and
reach nearly halfway to the zenith (images, right).

Rainbows cannot appear if the sun is higher than 42°

in the sky, which gives them a subtle dependence on
latitude, geography and time of day. They appear
only during rainy seasons; during summer months
(when the sun is often high in the sky) they appear
only in the morning or afternoon, but during winter
months (when the sun is always low in the sky) they
may appear at any time of the day. The most
spectacular displays occur when heavy rainstorms
move eastward and break in the late afternoon. And
rainbows in the northern hemisphere can never arc
across the southern sky (because the sun must be at
the viewer's back). Thus, rainbows depend on the
same basic elements of viewpoint and direction of
view that are always at the core of perspective
studies; the beauty of rainbows is that they put
these basic elements on display by combining radiant
color with the facts of season, geography and time of
day.

Yet the spectral colors of the rainbow are not as pure

as the colors seen through a prism. The intensity is
reduced by spectral smearing or overlap between
similar wavelengths, caused by the 1/2° angular
diameter of the sun and by slightly divergent paths
of light refracted through the raindrop. This smearing
reduces the purity of "blue" spectral light by more
than 75%, and the purity of "red" light by about
40%. Red is therefore usually the dominant or most
vivid color in a rainbow, a bias that is enhanced
around sunrise or sunset as the rainbow arc rises in
the sky.

The intensity of rainbow colors also varies with the

amount of solar illumination and the darkness of
background clouds or sky, and especially with the
size of the raindrops. The largest drops (larger than
2mm) produce the brightest, clearest colors; as
drops get smaller the colors appear "smudged" or
whitened by overlapping spectral hues; the yellow
and green bands are the first to fade to white.
Smearing is extreme in mists where the drops are
very small and evenly dispersed, so that all colors
blend back into white. These
achromatic fogbows have the same radius and obey
the same perspective rules as a chromatic rainbow.

Because there are more raindrops along a line of

sight near parallel to the ground than upward toward
the zenith, and as raindrops tend to increase in size
by aggregation as they fall, these more abundant,
larger raindrops produce colors that are brighter near
the earth than at the top of the arc.

Back to Elements of Perspective

Last revised 07.I.2015 • © 2015 Bruce MacEvoy

 format proportions technique
As visual displays, paintings have the potential to be
three things at once: a representation that mimics the key format proportions
natural perception in a physical environment; a
pleasing geometrical or abstract arrangement of line, constructing the format
texture, symbol and color; and an artful arrangement proportions
of forms within the edges of a flat surface.
composing with the format
So there are three design systems operating in an proportions
image at the same time: (1) the proportions among
physical objects in relation to a projective other format proportion schemes
geometry that simulates the appearance of a three
dimensional space, taught as linear perspective; format proportions in western
(2) the arrangement of shapes, tones and colors in paintings
relation to each other that creates an attractive
visual pattern, usually taught as principles are the proportions "real"?
of composition and design; and (3) the
distribution of all image elements within the format use what works
proportions, the space defined by the image
boundaries.

Across a long pictorial tradition, the picture plane

was thought of as a window opening onto a three
dimensional picture space. Linear perspective
dictated the size and shape of objects in this space,
but left to the painter the problem of presenting
them to view, much as a director arranges the actors
and props on a stage. These problems of "staging
and set design" were governed by principles of
composition, primarily to make the grouping of
elements in the painting legible, pleasing and
meaningful.

This emphasis on perspective and composition

demoted the format edges to peripheral vision and,
until the early 20th century, paintings were almost
always framed (often ostentatiously in ornate and
gilded materials) in order to separate the image
limits from the image itself by making the image
limits appear to be imposed by a foreign object —
the frame. These conventions transformed the
format into an illusory, arbitrary restriction, as if
removing the frame would reveal an expansive,
unedited view of the picture space.

In contrast, the 20th century esthetics of

nonrepresentational and pop painting often made
explicit expressive use of the size and shape of the
image itself. Everything imagined to create the
image is available to view on its surface; the limits of
the painting were integral to its overall legibility and
impact, and there was no implied representation
outside the format. These paintings are typically
displayed unframed or, in paintings by Paul Gauguin,
John Marin, Howard Hodgkin and many others,
image elements were actually painted onto the
frame, or framing elements were painted into the
canvas, so that frame and image united as a single
object.
To govern these new styles, painters did a peculiar
thing: they attempted to justify in abstract or
"scientific" terms the compositional principles
developed in the representational (perspective)
tradition. This process of abstraction took off in the
19th century, for example in the writings of John
Ruskin or Auguste Laugel, and it flowered into a
largely subjective system of "design principles" that
is still taught to art students today. These dogmas
disregard format based rules of design, although
format principles seem clearly implied by
contemporary painting assumptions.

As a corrective to this oversight I have tried to

understand how the different format dimensions
create lines of proportional stress or emphasis
on the picture plane. In this page, I describe an
explicit proportional system based on half folds or
squares of the format dimensions, and suggest
some design principles based on them.

These format proportions apply to rectangular

formats with proportions between 1:1 and
2:1 (height:width or width:height). This includes the
majority of artworks — etchings, prints, paintings,
frescos — since the Renaissance, and all popular
watercolor paper formats. I believe other formats
— circular, oval, elongated landscape or standing
figure formats — can be controlled with similar
proportional schemes by enclosing the format within
a rectangle or square, though I don't pursue those
possibilities here.

In my experience these format proportions produce

balanced and satisfying images that can be
architectural in their stability and poise; compositions
judged with them are often very good, and never
"bad". They provide a reliable frame of reference —
simple, explicit, objective — within which the
perspective and formal composition of a painting can
be critiqued. They are apparently an intuitive solution
to compositional problems, as paintings that rely on
the format proportions can be found in any survey
of Western paintings from the 15th to 20th
centuries. In many cases an artwork does not
conform to the proportions exactly, but achieves
dynamism by approaching them nearly.

Analyzing a painting in terms of the format

proportions puts both representation and
"composition" on an equal footing, and is surprisingly
useful to uncover basic design decisions. In fact,
editing an "unruly" painting to improve its format
proportionsoften clarifies the effects that are
created by disregarding the proportional divisions.

Note added in 2005: Several years after this page was

written, a reader alerted me to Charles
Bouleau's Frameworks: The Secret Geometry of
Painters (1963), which analyzes paintings in terms of "the
armature of the rectangle" and describes a method
of rabattement des petits côtés that I present as the square
divisions and quarter folds of the rectangle. However, I was
disappointed to find that across more than 100 painting
examples Bouleau merely inserts whatever ad hoc
"framework" seems convincing to him in each case, without
any historical testimony or xray evidence to confirm the
constructions were actually used by the painters. Despite
Bouleau's priority, this page is my independent and far more
systematic contribution to the problem of design within
rectangular formats.

the key format proportions

The key format proportions are based on
symmetrical foldsacross the picture space
and squares constructed from these fold dimensions.
The folds divide the format into fundamental units,
while the squares are (as it were) the harmonics that
emerge from the fundamentals.

In rectangular paintings with an aspect ratio (the

ratio of height : width) between 2:1 (a "portrait"
rectangle) and 1:2 (a "landscape" rectangle), the
obvious format proportions are the dominant half
folds of the height or width, which divide the picture
plane into quarters and locate the format center; and
the weaker quarter folds that divide the sheet into
sixteenths. (These "folds" are imaginary; you don't
actually crease the paper.)
 folds in the format

As shown above, these folds create the

horizontal midline (M, blue in the figure), which
divides the picture into two equal horizontal bands,
and the vertical centerline (C, red in the figure),
which divides the picture into two equal vertical
sections. Each of these sections can be folded in half
again, creating two secondary midline folds (m) and
two secondary centerline folds (c), or six folds in all.

I've also shown, as an orange box, the

outermost eighth folds(cc and mm), which are
halfway between c or m and the nearest edge of the
sheet or frame. These establish an implicit content
boundary inside the image; as I explain later,
strong forms in the image usually should not extend
beyond this frame, unless the forms extend beyond
the image boundaries or represent flow or dynamic
movement.

The compositional importance of these divisions is

roughly C > M > c > m. This is revealed in the
esthetic effect of edges that match the folds. It is
visually more distracting if a vertical edge (the side
of a building) coincides with C across most of the
height of the picture, than it is when a horizontal
edge (the horizon) coincides with M across most of
the width of the picture; this implies M is a weaker
division. Placing a pictorial edge along C seems to
split or unbalance the image, while an edge
along M does not.

The second system of format proportions is based

on squaresconstructed from the smaller format
dimension — the height of the sheet in the landscape
format, the width in portrait format — and from the
"edges" created by the centerline and midline folds.
 squares in the format

The diagram (above) shows these three divisions:

the square defined by the height of the format
(which creates the verticaldivision h'), the square
defined by the width of the half sheet on either side
of the centerline (which creates
the horizontaldivision c'), and the square defined by
the height of the quarter sheet above or below the
midline (which creates the verticaldivision m'). These
square proportions are constructed from both the top
left and bottom right corners, producing six more
proportional divisions.
In general, the visual emphasis carried by these
divisions is h' > c' > m'.

In the portrait orientation the height of the image is

larger than its width, so a square based on the larger
format dimension falls outside the image. However,
when hanging a painting, this square is useful to
determine the minimum distance between the
painting and objects on either side of it (diagram,
below).

use the square of the vertical side to

separate a picture from objects around it

These various divisions rarely duplicate each other to

divide the format into precise subsections. Instead,
they indicate stronger and weaker lines of emphasis,
which suggest the effective placement within the
image of strong edges, important detail shapes, or
accenting changes of contour.

constructing the format proportions

Now I'll explain different methods for defining or
constructing the format proportions on your painting
surface or photographic image.

Construction by Measurement. The easiest

method for making the divisions is to use a
straightedge premarked with the divisions for a
specific format.

Watercolor blocks are glued at the edges and bound

with a paper cover: the format divisions can be
marked on the outside of the cover, and transferred
to the sheet from these marks. A yardstick can be
marked with a colored felt pen with the divisions for
a full or half sheet. These divisions would be
indicated very lightly with pencil on each new sheet
before outlining or painting the composition. These
prepared guide marks or measuring sticks are
especially useful when you habitually use the same
format to make many works.

Larger or unusually sized sheets require individual

measurement. The most accurate way to do this is
with a ruler or tape measure, as follows:

1. Measure the long and short dimensions of the

sheet.

2. Divide these two dimensions by 4.

3. Mark the centerline, midline and quarter folds
from these divisions. Measure on four sides of the
sheet and connect to create six lines.

4. Mark the shorter dimension of the sheet along the

longer edge to locate the square h'.

5. Mark the centerline and midline dimensions, along

the edges parallel to each fold, to define the
squares m' and c'.

Constructing With "Compass" and Straight

Edge. Classical methods avoided working with
numbers. In this approach, you can use a long piece
of string tied around the tip of a pencil, or a
cardboard strip fixed at one end with a push pin, as
both compass and ruler.
 marking format proportions with string
and pencil

1. Put the pencil point at the upper left corner of the

format (at 1 in the diagram, above), then gently
stretch the string to the lower left corner (at 4) and
anchor it firmly at the corner with your left thumb.

2. With your right hand, bring the pencil across and

down the sheet toward the bottom edge, keeping the
string taut. As you do, lightly draw a short arc at
three places: around the approximate centerline of
the sheet, around the approximate midline of the
sheet, and at the bottom edge (blue marks in the
diagram).

3. Repeat this procedure three more times, fixing the

string at each of the remaining three corners, and
keeping the string the same length. This will produce
matching edge marks for the h'divisions at the top
and bottom of the sheet, and four "x" marks within
the sheet where two arcs cross at the midline or
centerline.

4. Lay a straightedge or long ruler across the paper

between matching edge marks at the top and bottom
of the sheet, and make faint vertical line on the
sheet for the two square divisions h'. C is defined by
a vertical line through the two arc "x"'s on the
centerline; and M is indicated by a horizontal line
through the two arc "x"'s at the midline, as shown
above.
5. To find the quarter divisions, lay the straightedge
or ruler so that one end is lying on the midline or
centerline, and the other edge at an opposite corner.
Then draw a faint line where the ruler crosses the
centerline or midline (black dots in the diagram,
above). These indicate the placement of the
horizontal and vertical quarter folds.

6. Again fixing the string at each corner, place the

pencil on the midline M, then swing this distance up
or down to the top or bottom edge; these marks
define m' on both sides. Connect with two vertical
lines.

7. Finally, fixing the string at each corner, take the

distance to the centerline down to the side edges;
these marks define c' on both sides. Connect with
two horizontal lines.

Rabattement des petits côtés. Charles Bouleau

describes a method he calls rabattement des petits
côtés ("folding down of the smaller sides"), as shown
in the diagram (below).

1. Measure and mark the length of the shorter sides

of the rectangle along the longer sides, from the
corners at both ends, and connect the marks to
define the divisions h'.

2. Construct the two diagonals inside each of the

squares formed by h'.
3. These diagonals will form a small diamond or
rotated square in the center of the format. A vertical
line through the top and bottom corners of this
square defines C; a horizontal line through the left
and right corners of this square defines M; horizontal
lines through the top or bottom corners define the
two m' divisions; vertical lines through the two side
corners define the two c' divisions.

4. Using diagonal lines, as described in step 5 of the

previous method, construct the quarter folds on all
four sides.

Bouleau's method of rabattement in the

format

In the diagram, the rabattement of the short sides

locates the division h', point 1 determines the
midline and the vertical division m', and
point 2 defines the centerline and the horizontal
division c'. This construction method creates more
lines on the sheet than necessary, although the
diagonal bands may be useful as compositional
guides.

composing with the format

proportions
If you are working at any stage with digital
photographs, then the format proportions are most
easily implemented as an image overlay onto the
reference or original photograph, using graphics or
image editing software.

Computer Format Templates. You must first

determine the file image size (pixel height by pixel
width) appropriate for your working methods. This
means you have format templates for:

1. each of the paper formats you use most often,

2. sized to accommodate the projection or copying

method(s) you use to transfer the image outlines to
the paper.

The first issue arises because different brands of

paper define the standard formats differently (the
double elephant in Arches is not the same as the
double elephant in Saunders Waterford, for
example); some brands (Zerkall) use nonstandard
dimensions; and all handmade sheets can be quirky.

The second issue arises because optical projection

systems have different focal lengths or enlargement
ratios. If you want to project an image onto a full
sheet paper, for example, then the image you are
projecting should be large enough to produce a
large, bright image for copying, but not be so large
that it cannot be brought into crisp focus across the
entire format area.

The following images show the format proportion

templates for two common watercolor paper
formats and the classic golden rectangle. These also
illustrate what happens to the format proportions as
the rectangle becomes more elongated. (For
comparison, the horizontal and vertical divisions by
thirds are shown as small blue dots.)
the full sheet (aspect ratio 1:1.36)

the half sheet (aspect ratio 1:1.47)

 the golden rectangle (aspect ratio 1:1.62)

composition proportions of common

formats
(in landscape orientation)

If you open these templates in a new window, you

will find they are actually 450 pixels wide. I use that
standard width because when it is printed at 100%
size it is the largest image that will fit into the
window of my opaque projector without optical
distortion. You may need a larger or smaller
template for your particular equipment and methods.

Computer Image Editing. Once the format

proportions have been constructed or templated,
then fitting the image into the format and aligning
the image elements with the format proportions is a
matter of experience, artistic style and the specifics
of the image.

I use the format proportions to edit the composition

of a digital photograph, as described next. However
the logic of the method applies as well to composing
the image by eye in a freehand drawing or through a
camera viewfinder.

First do any major image editing in the full size

image before working with format proportions. For
example, in landscape photos, I typically correct for
foreground expansion caused by the camera
optics, so I do this before adjusting to format
proportions. I may also delete or reposition (cut and
paste) objects in the photo to improve the image
composition and legibility before using the format
proportions, or to bring everything into better
aligment with the folds and squares when the format
proportions are added.

Here, step by step, are the menu and keystroke

commands for Adobe Photoshop in the Mac OS:

1. Open as separate files the art image and the

format template for the support you intend to use for
the painting or drawing.

2. In the format template file, Select

All (Command+A) and Copy (Command+C) the
format template. Close this file.

3. Switch to the digital image file

and Paste (Command+V) the template as a new
layer on top of the digital image.
4. In the Layers window, set the transparency of the
format template to 50%. (If you do not see the
Layers window already open with your files, it is
found in the Windows > Layerspulldown menu.)

5. In the Layers window, click on the digital image

layer. Then Select All (Command+A) and choose
the Free Transformfunction (Command+T). The
image will be outlined by an animated dashed box,
with small squares at each corner and the center of
each side. Note that I resize the photo rather
than the template, because the template has been
presized to accommodate my printing and projection
tools.
 editing a photograph with a format
proportion template

(top) the image being resized and repositioned to fit

the outline and guidelines of the full sheet format
template; (bottom) the final composition in the full
sheet format
At this point your working image will appear similar
to the upper image above. You will see the digital
image and, floating in front of it, the semitransparent
format template.

6. Hold down your shift key and option keys

(Shift+Option) at the same time, then click on and
hold a corner box of the image with your mouse
cursor.

— Move this corner toward or away from the center

of the image to resize the image.

— Release the shift and option keys, then click and

hold anywhere inside the image to grab the image to
move it up, down, left or right. (You can also use
your keyboard arrows to make these repositioning
adjustments.)

Continue resizing and/or repositioning the image

until you get the image proportions you want within
the format outline.

7. Double click inside the image to resize it. The Free

Transformguides will disappear.

8. Using the Layers window, click on the template

layer.

9. Using the magic wand tool, click on the

area outside the template. Then Select
Inverse (Shift+Command+I) to select the template
itself.

10. Crop the image (from the pulldown Image >

Crop menu). Your file will now look like the bottom
image, above.

11. In the Layers window, grab the template layer

and drag it to the layer trash can. The template
disappears.

12. Mark the corners of the image with black dots if

they will not be easily visible when the image is
projected, so that the image can be correctly sized,
focused and registered over the support for tracing.

Note that the resizing and repositioning may produce

an empty border along one side of the image (as you
see along the lefthand edge in the example above).
This area is usually small enough so that it can be
completed freehand during the process of tracing or
painting. If it is much larger, you can use the image
editing program to fill the area by copying and
pasting elements of the image.

Format Design Principles. Step 6, the actual

resizing and repositioning, is a subjective and
exploratory process. It is important to play around
with the image, trying different sizes and different
positionings of a new size, before you find a solution
that "clicks" (and you double click on it).
I sometimes just observe the effects of resizing until
I notice a solution that seems to work, then I
reposition and resize to the template guidelines. If
the result does not seem satisfying, I return to the
original image size (Control+Z) to find another one.
Often, the size of the major form dictates the
approximate enlargement required, and it's just a
matter of moving the resized image around to
position it against the template guidelines.

A reliable method is to start with the most dominant

image element(s), and work from those to the less
important image elements. In the example image,
the cat, bookshelf and window created an obvious
vertical point of interest, so first I positioned the
image so that these fell against the
righthand h' and cdivisions. Then I resized the image
to find a good placement against the
lefthand h' and c, and finally moved the resized
image up or down to find a good placement against
the horizontal guidelines.

Here are some suggested formatting principles, with

reference to the demonstration photograph (lower
image, above):

Do not center important large forms. Avoid

centering any dominant object, either around the
midpoint of the format or on the centerline or
midline. The major exception is when the object is
isolated in the image and is the sole focus of
attention (for example, the animal skull in
the O'Keeffe painting below, or any portrait by
Chuck Close). In the example image, no object is
centered.

Do not place strong edges along the centerline

or midline. For example, in a landscape, do not
place the horizon or rooftops along the midline, or
the edge of a building along the centerline. In the
example, there is no strong edge along the centerline
or midline.

Emphasize the edges or dimensions of

important large forms, or the center of
important small forms, by placing them on or
next to the format folds or squares (other
than C and M). The goal is to place the dominant
forms and edges in the image so that they seem to
fit into or hang from the format guidelines. How you
approach this depends on your choice of important or
interesting forms and their spatial relationships
within the image. During digital editing you can
resize the image so that the dominant objects
correspond to the template guidelines, and use cut
and paste or drawing tools to move or edit less
important objects or edges into better alignment.

For partial balance, place two important but

contrasting forms along the two strong vertical
lines (h' and c) or (h' and m'). In the example,
these are the front left corner of the table, the
paintings propped on the table top, and the nearby
front right corner of the cardboard box (on the
lefthand h' and c), and the edge of the bookshelf,
window and the inquisitive cat (on the
righthand h' and c).

For asymmetry, place a single important form

or strong edge along only one strong vertical
line (h' and c) or (h' and m'). In the example, the
bookshelf and cat at the righthand h' and c, and the
illumination flowing into the frame from the window,
provide stronger visual interest and greater emphasis
than the corner and leg of the table at left. All the
forms are arrayed in relation to the compact, bright
window puncturing the shadowed wall, and the large,
diffuse area of illuminated wall and floor.

Place the center of attention off center but

inside the central rectangle. The central rectangle
is defined by the dominant vertical lines h' or c on
either side and by c' and/or mabove and below. Note
that these lines may be close together (as
are h' and c in the image above) or may be
separated (c'and m in the image above), which I feel
defines gradations in emphasis that are unique to
each format.

The central rectangle is typically the place in the

composition where the principal subjects (including
strong contrasts of value, color, texture or visual
detail) are located. In the full sheet format (used in
the example above), this area is large and well
structured, allowing for an expansive visual interest
— including the cat, the bookshelf, the chair and the
table.

Notice that the same image could be resized and

cropped in many other ways to emphasize the cat,
the table, the cat and the window, and so on. As is,
the image implies a broad view of the whole room.
(The photo was taken as my wife and I were
unpacking in our retirement home, and for me
evokes a moment of nostalgia, which is not a detail
state of mind.)

Enclose at least one visually interesting or

important form inside the central rectangle. The
quarter sections of the central rectangle are usually
the smallest inner rectangles formed by the format
divisions. These keystone rectangles often include a
local center of interest in the painting but do not
have to do so (as we'll see in a painting
by Caravaggio). In the example, the back of the
wooden chair, with its pattern of closely spaced
vertical posts, provides visual interest to balance the
cat, window and table.

If perspective or spatial depth is implied, space

distance markers along the format lines. In
landscape the horizontal lines are especially
important. Do not link the horizontal and vertical
lines in nested rectangles toward the center unless
you want to create a tunneling effect.
In the example, the nearest table leg falls on the
lower eighth fold (mm), the far front leg and the
hind feet of the cat on the lower m, the seat of the
chair and far table leg on c', the corner of the table
top on M, the window frame and top of propped
paintings approximately on the upper c', and the top
of the lamp and bookshelf on the upper mm.

Do not extend or place important static forms

outside the "frame" of eighth folds (cc and
mm) unless the forms are cropped by the image
borders. This is not a hard and fast rule, though it
usually contributes to a good result. In the example,
the tops of the torchere lamp and rowboat bookshelf
lie on the top eighth fold; the corner of the table and
cardboard box lie on the lefthand eighth fold; the leg
of the table and bottom of the box lie along the
bottom edge fold.

The window extends outside the righthand eighth

fold, but it (actually, the molding around it) is
cropped by the image boundary. Although the
window placement up against the format corner was
an unanticipated result of the resizing, I decided it
worked in context to "hang" the window around the
cone of light entering the room. And this suggests a
final principle:

Place important forms outside the "frame" of

eighth folds (cc and mm) to create a sense of
dynamic movement across the image or into
space outside the image. In the example, the
window symbolizes the flow of light into the room
and the presence of a landscape beyond its confines.
In other situations a strong form of that shape in
that location would probably look wrong, and this
would need to be addressed by editing (deleting or
moving) the offending object into a better position.

other format proportion schemes

Many other format proportion schemes are possible
and several deserve mention here.

The most common approach is the intuitive

arrangement of major forms or shapes in the
image. The criterion here is just "a good effect" for
the specific image, without relying on general
principles. In the simplest approach, the most
interesting part of the image is located on a sweet
spot — a point displaced to one side and above or
below the center of the image — and then the image
is resized or cropped to fill the rest of the format in a
pleasing way.

Another approach is the mise en scène strategy of

stage managing the forms within the image to
emphasize dramatic relationships, tell a story,
highlight main characters or character contrasts, and
so forth. A delightful example is Weegee's The
Critic, in itself a carefully staged "class encounter"
that was cropped in two versions to emphasize the
character contrast between attenuated socialites and
a disheveled street character.

There is nothing wrong with either approach; in fact,

by playing around with the resizing and positioning of
any image within a format template, you are simply
using "what works" or "what looks good" to find an
effective solution. But even with the two intuitive
methods, the format proportions operate in the
background, as can be used a reference and to
critique specific ways in which the composition does
or does not work.

A third compositional strategy is a geometrical

arrangement of dominant forms. It is common in
art history to hear a painting analyzed into a
simplified geometrical framework, such as the golden
rectangle, the square, two or more overlapping
circles, the main diagonals, a triangle or pyramid,
nested triangles or pyramids, and so on. I have
never had sympathy for this brutish and
simpleminded analysis and, excepting the analysis of
paintings from the Italian Renaissance, don't see
much relevance in it.
In his book, Bouleau contrived many complex
variations on the square divisions, including the use
of harmonic intervals across the format dimensions
— proportions that match the whole number
fractions of musical intervals: 1/2 (the musical
octave), 2/3 (the fifth), 3/4 (the fourth), 4/5 (the
third), supplemented by diagonals drawn from the
major divisions. I find the result (diagram right) to
be so cluttered as to be useless.

My dissent from this approach is that it creates diagonal (quarter) divisions

formalisms that have little relevance to the way we of the rectangle

look at the world or at art. In particular, most of the Bouleau's "armature du

geometric formalisms really define solutions to rectangle"
composition and design, not to the specific problem
of fitting the image within a format.

Our visual experience is not aimless, ad hoc or

mathematically prissy: it is consistently skilled,
motivated and pragmatic. It's my claim that the
format proportions represent a form of visual field,
and the structure of the visual field represents the
structure of our interests and aims in relation to the
world. We identify the most important features of our
world within our continuous stream of experience,
then apportion or balance our looking among these
features in a way that reflects how we believe
they relate to each other and to our immediate
priorities of enjoyment, understanding or action. We
also commonly do this from a distance that allows us
to see the important forms within a situating context.
I believe the whole point of compositional systems
based on the rectangle format is that they shape
our attention dividing visual strategies. Looking
directly at something emphasizes its importance.
Dividing attention often means we balance our
looking across competing things available to view,
which tends to center them in a wider view.
Disregarding or discounting means we push things
into our peripheral vision. This is basically what the
format proportions do — they divide up the support the format proportions as
into competing areas of unequal interest. A simple symbols of binocular
interpretation is that the h' squares represent the attention
overlapping retinal areas of a binocular field of view,
the outer eighth folds the widest area of view, and
the central rectangles the focus of attention (at right,
in the full sheet format). This dictates that more
elongated formats are appropriate for forms viewed
from a distance or as a series or progression, while
squarish formats are appropriate for forms viewed as
exemplars or individuals, separate from context and
movement.

The compositional guides are not an empty or

arbitrary geometry. They suggest how the viewer
should look within the image, as he or she were
actually inside the world of the painting or
photograph. They do this because they mimic our
habitual strategies of visual attention. By placing
the important parts of the image along or inside
strong divisions and keystone rectangles, the artist
implicitly directs the viewer's selective exploration
and appreciation of the total information in the
image. This is why format based composition turns
up in photographs or images — even cinematic or
advertising images — that were created by artists
who may have rejected the classical formalisms of
composition and design.

Why not just divide the format by thirds (the

musical interval of a fifth) and be done with it? As
you can see in the format templates above, formats
with an aspect ratio around 1:1.5 (the half sheet or
emperor) place the h' and m' vertical divisions at the
vertical 1/3 divisions. But in many compositions
based on the thirds, the central rectangle is crowded
into too small an area in the middle of the sheet, so
that the focus of attention is, as it were, viewed from
too far away. In most formats, two of three vertical
divisions — c, m' or h' — and both horizontal
proportions (m and c') are closer to the edge than
the divisions by thirds. Placing compositional
elements on these outer divisions seems more
pleasing.

The major exception is the square format or

formats that are nearly square. In a square the
square division h' disappears, c'= M and m' = C,
and the sheet is defined entirely by the half folds. In
that case the division by thirds can supplement the
half fold divisions.
 composition proportions in a near square
format

The illustration (a CD cover for the album Animals by

Pink Floyd) illustrates use of the third and half fold
proportions in a nearly square (1:1.15) format. I
offer this as an especially convincing example
because the flying pig is placed exactly at the point
where the upper and left third divisions cross. But
nearly all the format lines are aligned with or
centered on some important architectural edge or
form. The monumental impression created by the old
London power plant is emphasized by its
harmonious, stable placement within the image
dimensions.
 format proportions in western
paintings
At this point there will be two critical questions: Do
the format proportions really produce a better
design? And are the format proportions so complex
as to be arbitrary? The most direct way to answer
these questions is to make a review of images from
western painting.

I have done this with a large number of published

paintings over the years, and have been impressed
by how consistently the format proportions anchor
important image elements in paintings by a wide
variety of artists in a wide range of styles. I suggest
you test this for yourself: leaf through any art
historical book or photographic magazine, or freeze
frames from your favorite wide screen film, and pick
out images that seem to be effectively or badly
composed. Then use the image outline to construct
the format proportions in the image, and see
whether the pleasing composition can be explained
by the guidelines.
In this section I offer analysis of 15 paintings from
the western canon, both to show how they respect
the format divisions (whether intended or not by the
artist), and to illustrate how departures from
the formatting principles are interesting for the
effects or insights they can reveal.
In all images, white or gray lines are used to show
the six format half and quarter folds (C, M, c, m),
red lines the six square folds (h', c' and m'), and
orange the eighth fold frame (mm and cc). The
square folds h' and c' (w' and m' in portrait format)
are shown as bold lines.

Two cautions are necessary when doing this. First,

paintings may have been trimmed during relining,
stretching or restoration, or arbitrarily cropped to fit
page formats in books, and these proportions will
give a distorted idea of the format proportions of the
original work. Second, paintings are rich with detail,
which means only large or important forms should be
interpreted.
 format proportions in a painting by
Pollaiuolo (1.43:1)

The first example, a large oratory panel of The

Martyrdom of Saint Sebastian (1475) usually
attributed to Piero del Pollaiuolo, is famous for its
arbitrarily formal design devices — a triangular
arrangement of figures including six archers in three
poses viewed two ways. As you can see, the design
also builds on the format proportions. The heads of
the four central archers lie very close to the four
corners of the central rectangle (defined top and
bottom by the strong division at w'/c') and this
rectangle roughly defines the height of the back two
archers. However, the keystone rectangles are empty
of visual interest, which pushes the structural
emphasis outward, to the triangular lines of aim
taken by the four archers. The left standing archer is
as high as M and the balance of both corner archers
is along cc. The position of St. Sebastian does not
correspond well to the format divisions — the
placement would be stronger if his shoulders were
at mm — but as I will explain later, this
displacement appears to have an important
expressive purpose.

format proportions in a painting by

Botticelli (1:1.61)

The second example, Sandro Botticelli's Birth of

Venus (c.1485), has exactly the aspect ratio of a
golden rectangle, and it captures the "portal"
implications of that format in the arrangement of
attending immortals on either side of the newly
created figure of Venus. On the right, the two hands
and face of Hora (and the coast behind her) fall
on h', m' and crespectively, and Hora's figure leans
into but does not cross c. The centerline C, which
runs through the left side of Venus, her raised thigh
and her weight bearing foot, emphasizes the tilt to
her left that prepares her first step on land; most of
her torso falls within two of the four keystone
rectangles. (An interesting compositional detail is
that, in the golden rectangle format, the central
rectangle formed by h' and c' is also a golden
rectangle, as are the two rectangles outside
the h' divisions.) The location of the horizon, just
above the waist of the two standing figures, implies
that we viewers are kneeling on the shore. The shell
carrying Venus is not centered between c and m' (as
if pushed backwards and to the left by her step), but
the width of the shell exactly equals the distance
between Hora's face and the blowing mouth of
Zephyr. On the left, the major features in the figure
of Zephyr (blowing mouth, right shoulder) and his
companion (face, left shoulder, right knee) also lie
along the lefthand format lines. All three major
figures are arranged parallel to three upward
converging lines (in green) defined as diagonals from
edge intersections of the major format folds; by
tilting the figures in this way the emphasis on the
lefthand division is reduced and space is opened
below to receive the implied forward movement of
the Venus figure. The divisions m and c'are not
clearly accented above or below, which reduces the
spatial layering of the image. I think it's more
common for c' top and bottom to define the height of
standing figures in the 1:1.6 format, but here the
measure is the eighth fold frame — mmtop and
bottom — which seems to bring us and the figures
closer together. In addition, the feet or legs of the
companion gods extend outside this frame, which
creates a kind of spatial expansion away from Venus,
and this is supported by the beautiful flow of her hair
into Hora's upraised and windblown drape. Finally,
Venus's head pokes above the mm line at top, which
gives greater vertical emphasis to her already
elongated (more than 8 heads high) proportions.

format proportions in a painting by Titian

(1:1.40)
Another secular image of beauty, the Venus
d'Urbino (1538) by Titian, with an aspect ratio of
about 1:1.4, is interesting for the ways it "violates"
the design principles outlined in the previous section
— for example, by placing the edge of the drapery
along the centerline C. Figure nudes are difficult
generally because the landmarks are created by the
continuous curves of the figure. The points of
emphasis depend on the posture of the figure and
how it has been cropped. Usually the top of head,
eyes, shoulders, breasts (in females), elbows or
hands, hips, groin, knees, and ankles or feet are the
points of emphasis horizontally or vertically. The
point of greatest inflection in the spine, or the center
of gravity of a standing or sitting figure, can also be
marked. In this horizontal Venus (as in the vertical
Venus of Botticelli) the figure pushes outside the
eighth folds on both sides, though the girl's heel and
head mark them. Recession is marked by placing the
base of the back wall on M and the ledge above
on c'. The format lines clarify the balanced
assymetry in the composition: the heel of the lady's
foot and the head of the servant are at the top and
bottom mm at right, while her head and hand are
bracketed by the upper and lower c' at left.
 format proportions in a painting by
Caravaggio (1:1.36)

Caravaggio's paintings, for all their violent contrasts

and dramatic action, often have an overall visual
balance that seems painstakingly planned. In the The
Taking of Christ (1598) there is a general shoving of
humanity across the picture plane, but this dynamic
flow is controlled by the format placement. The key
event, the Judas kiss between two faces, is placed on
the lefthand strong lines, and the actors' eyes are
gathered along the upper c' and their shoulders
along M. Especially significant are the picture details
that lie outside the eighth folds (orange rectangle).
The shout and upraised hand of the disciple explode
to the left like water splashed from an impact,
balanced by the portrait profile of Caravaggio as the
disciple pushing in from the right. The keystone
rectangles are completely empty of interesting
content, so our attention is forced to the edges of the
format as if jostled back and forth by the action. The
format lines also help the viewer appreciate other
points of conceptual symmetry behind the visual
asymmetry that is characteristic of Caravaggio's
carefully designed pictures. I especially like the
variety of hands outside the frame — two hands
joined below, two hands separated above, one arm
stretched across the picture vs. two raised arms
above on either side, three heads geometrically
balanced on either side of the interlaced fingers, two
foreground heads facing one another matched by two
foreground heads facing in the same direction, and
so on. (All of Caravaggio's compositions merit close
study.)
 format proportions in a painting by
Poussin (1:1.39)

There is a huge number of classicizing landscapes

that I find difficult to tolerate, such as Nicholas
Poussin's Summer (Ruth and Boaz) (1664). Some
might admire the spatial staggering of trees and
fluffy clouds, or the contrived figure placements. It
seems obvious, however, that Poussin was not much
concerned with the format proportions: the
disjointed, stagey feel of the painting reveals it. I
especially dislike the placement of the foreground
figures along the bottom edge of the picture, well
below the lower eighth fold mm. These are esthetic
decisions consistent with Poussin's aim, which was to
present a familiar (Biblical) story in familiar dramatic
terms, not to dissect or recreate our perception of
the physical world. In fact, the flatness of everything
behind the three foreground figures looks remarkably
like the painted backdrop to a three actor morality
play: I wonder whether contemporary dramatic
conventions and perspective principles of set design
influenced Poussin's concept.
 format proportions in a painting by
Vermeer (1.19:1)

In contrast, Vermeer's Woman in Blue Reading a

Letter (1664) has a stillness and harmony that must
have been conceived in terms of the format
proportions. Too many features coincide to be
coincidental, and there are too few edges and forms
to create random correspondences. Broad,
rectangular forms are placed outside the eighth fold
frame and all are cropped by the format (except for
one chair, which is cropped by the table); the figure
is slightly off center, enclosed within a narrow
column defined by the midline square m' on the left
and the centerline fold c on the right. The woman's
head is topped by h', her eyes by m, her hands
by c' and her elbow by M. Even the lower centerline
square c', which doesn't seem to indicate anything
visible, runs directly through the hidden fetus —
corresponding to the c'above, through the hands
holding the letter.

format proportions in a painting by

Velázquez (1:1.46)

The Rokeby Venus (c.1650) of Diego Velázquez is

closely modeled on Titian's pose — but viewed from
the back! Like Titian, Velázquez places the major
accents at the knees, hips and shoulder, the head
against one cc, with the elbow and foot jutting
beyond the eighth fold frame. But he adds several
important innovations. The format itself is elongated
horizontally, reducing its visible headroom and
drawing both h'/m', and c'/m closer together. To
compensate, Venus's legs are slightly flexed (not
fully extended as in Titian), the curve of the buttocks
is accented, not the point of greatest weight at the
hips, and the torso is contracted upwards to bring
the left shoulder and right margin of the rib cage
(accented below by darkened folds in the gray
bedsheet) into h'; in this way the whole figure is
made larger within the format (closer to the viewer).
Velázquez also emphasizes the
opposing h' and m'format lines through the
relationship between Venus and her reflection and
Venus and the putto, whose heads are symmetrically
placed around C. Velázquez places the reflected
portrait of Venus within one of the keystone
rectangles, and the contrivance is revealed by the
fact that this is optically implausible (it allows us to
see Venus, but does not let her see herself). The
righthand dominant verticals are marked by the
putto's knee, crossed hands, the fabric dangling from
his wrist, the edge and corner of the mirror frame,
and the right foot peeking under Venus's crossed leg.
The upper boundary made by m/c' is marked by the
top edge of the glass, the amorino's elbow and wing,
and the top of Venus's head and reflected head. The
gray and crimson fabrics are separated by M. The
lower boundary is marked by her feet and rib cage;
the sag of her hip below this line, reinforced by the
broad curve in the gray bed cover, creates an erotic
focus on this part of her figure.
 format proportions in a painting by David
(1:1.28)

The Oath of the Horatii (1784) by Jacques-Louis

David utilizes the format proportions in an explicit
way that does not require detailed explanation. Note
that the aspect ratio of 1:1.28 (in watercolors,
typical of the medium or demy sheets) creates an
unusual spacing system: a strong vertical emphasis
at c/h'(where h' is actually closer to the edge
than c) linking the three warriors with their three
wives opposite; the contrasting horizontal division of
the image into nearly equal intervals (approximately
corresponding to the feet, knees, waist/heads,
shoulders, upraised swords, capitals and arch
centers); and the small, nearly square central area
created by the midline square m' and centerline
square c'. As in Poussin, the figures extend
below mm and again the effect (with the lighting)
suggests a dramatic stage. I like the way one
keystone rectangle contains the nexus of upraised
swords and hands, symbolizing the oath; the way
that the strong emphasis at h'/c unites the group of
warriors at left, but divides the group of grieving
women at right; and the bridging function of the
central archway and the feet of the foreground
figure, which connect the two m' lines and pull the
figures and background into a single, continuous
frieze.

format proportions in a painting by Manet

(1:1.26)

There are many Impressionist and later painters,

from Edouard Manet to Paul Cezanne to Pablo
Picasso to Barnett Newman, who do not seem much
aware of the format proportions. In every case the
question is how this disregard affects the impact of
the image. In Le Dejeuner sur l'Herbe (1863), Manet
has chosen a format that spaces out the format lines
into an almost even grid, but unlike David he uses
them haphazardly. There are a few points of obvious
correspondence — the capped man's head and hand
on h', his upraised hand and the bathing woman's
head on m', the line of feet, knee, hand and hip
along the lower m, etc. But look at the way the
upper format lines are lost in vague foliage, or the
way the basket and picnic covers spill out of the
frame at lower right, the random placement of tree
trunks. One of the things that always puzzles me
about this picture is that peculiar alley of lawn and
shade that recedes along the lefthand h', between
trees and bushes on that side. (If there is water back
there, its shadowed color disconnects it from the
water in which the woman is wading.) That recession
always has an unsettling effect on me, a tunnel of
obscurity contrasting with the field of light on the
opposite side. (I will show later that these oddities
have a good purpose.)
 format proportions in a painting by Degas
(1:1.48)

Format proportions can be wildly disregarded in

paintings by Edgar Degas, as in his The
Rehearsal (1873-78). But Degas was among the 19th
century artists whose design sense was strongly
influenced by photographic materials, and this shows
in his bold use of empty space, cropping, and
unusual points of view or strong perspective. Here,
except for the edge and central support of the large
window at right, which fall
along h'/c and mm respectively, the line of heads
along c', and perhaps the location of the two front
dancers at left, it's hard to see many format
correspondences. But I think this illustrates two basic
points about the format proportions: (1) they seem
to represent or symbolize the natural distribution of
human visual attention, and (2) they tend to produce
balanced, somewhat static images. In Degas's
conception the cropping of the windows and the
violinist at right seem to assert a flat, photographic
rather than human focus, and the composition is
dramatically off kilter (like the position of the
dancers). However, the warm, analogous color
scheme seems to place the emphasis on light itself —
flowing through the windows, across the floor, and
through the gossamer dresses of the dancers — and
it's possible that strong format markers would
disrupt this effect of gradation and flow, captured so
beautifully in the colors of light along the floor and
walls.

format proportions in a painting by Renoir

(1:1.34)

The eternally happy Danse au Moulin de la

Galette (1876) by Pierre-Auguste Renoir shows how
the format proportions can control an image that is
dense with competing details. This lovely format,
which is very close to the full sheet, places
the h'and c intervals close together, and
puts m' halfway between them and C. The closest
two female faces are placed like cameos within two
keystone rectangles, facing the male whose head is
against the righthand c/h'. On the opposite side, the
strong h'/c division is marked by a dancing couple
(actually, the back of the man, who seems to lean
against it); the m' division is marked by a second
dancing couple behind them. On the right, a man in a
blue coat leans against c/h' in the opposite
direction, and his angle carries up into the tree trunk
and down into the back of the seated man, lending
the composition a wonderfully lightheaded, lilting
effect. All the legible faces of the background figures
fall within the m/c' divisions at the top of the sheet;
everything above m is an indistinct crowd, and the
foot of the male dancer is on m below.
 format proportions in a painting by Signac
(1:1.44)

Another painter who seems to have consciously used

the format proportions was Paul Signac, as shown
in his Quai de Clichy(1913). Composed in a format
close to a half sheet, there are too many precise
alignments to be due to chance: the placement
of h' on the trees on opposite sides of the street,
of m' with the distant smokestack, c' with the top of
the roof at right and the base of the two
trees, M with the horizon and top of the fence at
right, C with the distant house and c with the side of
the fence at right and the row of trees at left. This
picture also gives a sense for the feeling of balance,
stability and spatial recession that is created when
the format proportions are explicitly accented by
major image elements.
 format proportions in a painting by J.S.
Sargent (1.29:1)

The format proportions work just as well in portrait

format, as this lovely painting of Lady Agnew of
Lochnaw (1893) by J.S. Sargent demonstrates. The
only difference in the portrait orientation are that the
largest square in this format (equivalent to a demy
or medium sheet) is formed by the width of the
canvas (w'), and what used to be the midline (the
horizon in landscape format) is now the centerline C.
Lady Agnew's eyes and the top of the chair are
perfectly aligned with the area of emphasis created
by m and w' at top, matched by the location of her
left hand below. Her nose, pendant and left eye lie
along C, her torso is bracketed by m' on both sides.
Her right forearm is flexed to lie with her crossed leg
along the lower c', while her neckline dips down to
the upper c'. The top of her head does not extend
above the top eighth fold (mm), while her left
forearm and hand lie exactly along the right eighth
fold (cc). The edge of the chair and the shadow
behind it push outside the eighth fold on the right,
just as the woman's left arm and the shadow behind
it push out on the right; this seems effective to
accent the Lady's turned position in the chair and
draws attention to the angular flow of her waist sash.
Her skirt flows outside the bottom mm, but this is
fine because it is cropped by the bottom of the
picture.
 format proportions in a painting by
O'Keeffe (1.22:1)

Typical of many modern paintings, Summer

Days (1946) by Georgia O'Keeffe shows less interest
in the format proportions, despite the strong
centering of the major form and the sprawl of the
antlers. These do roughly hang on the
upper m (usually a weaker line than w') to match
the location of the red blossom below, the lower
antler curve is along c' the skull eye orbits are within
two keystone rectangles, and the upper border of
blue sky is along the lower h', but the overall sense
of floating is created by the curvature of the horns
and the broad background of undifferentiated clouds.
In cases where the image does not seem strongly
formatted, it is instructive to digitally alter it to make
it so, then assess the differences between the two
images.

format proportions in a painting by Hopper

(1:1.43)

Finally, there is Bill Latham's House (1927)

by Edward Hopper, in a format close to a half
sheet. As we've seen, this format tends to create an
"opposition" emphasis, which Hopper exploits in the
strong placement of the tree on the left and the
house (and small fencepost) on the right. Hopper
places the visually most delicate and active details of
branches and foliage within the central rectangle;
each keystone subdivision contains a contrasting
visual pattern. The upper division at m/c' is
accented by the widest part of the dark foliage, the
lower division only weakly, by the lowest branches in
the tree. The midline M defines the eaves of the roof
and tops of the windows, and is positioned exactly
between the ocean horizon and the horizontally
extended branch extending left from the tree.

are the proportions "real"?

These "analyses" invite the obvious and legitimate
rebuttal that geometrical interpretations applied after
the fact can always be made to match some aspects
of a painting: that doesn't mean the proportions
actually highlight key aspects of the composition.

That's a fair challenge, and a fair test is to find a

composition that does not seem to correspond to the
format proportions in an obvious way, then change
the major elements to coincide with those
proportions and see what effect this has on the
image.

To illustrate the basic procedure, I will use a painting

by Edward Hopper: Portland Head Light (1923).
 a painting altered to accent the format
proportions

The original image is at top; the three lower images

above show the editing process. The second shows
the format proportions scribed over the original
image; the painting is in the half sheet proportions
(1:1.47). The composition seems based primarily on
the horizontal thirds; the vertical divisions do not
correspond to strong image elements.

To accent these proportions, the center of the

lighthouse was shifted left to fall on the closely
spaced h'/m' lines. The base of the tower and small
house were dropped to the lower m/c'lines, and the
top catwalk in the light tower was moved down to
the upper m/c' lines. The small house was moved
inward between c and m', and the main house was
shifted slightly to bring the join between large and
small parts of the house to h'and the porch
to cc along the right edge of the picture (third
image). The bottom image shows this altered
painting with the proportion lines removed.

To me, the revised composition has a less dramatic

and compact effect than the original, but it has a
greater feeling of openness and visual stability. There
is a distinct tension between the tower and house,
and a more interesting rhythm connecting the
buildings with the rocky beach. The weight of the
horizon is greater, partly because the height of the
tower has been reduced and the ocean's area within
the central rectangle has been increased. (In my
experience, emphasizing the format proportions
often creates a stronger sense of recession and a
greater feeling of space.)

I chose this example partly because the original

seems to have been painted (or sketched for
painting) in a open field. The altered image would
closely correspond to the view that Hopper could
have taken by standing about 50 feet to his left.
Although the site today has been heavily landscaped
and fenced, there is a sea cliff and steep walkway at
that location which would have made an inconvenient
perch even in Hopper's time. Therefore his design
was at least in part dictated by site specific
obstacles, yet the format analysis still identifies the
resulting compromises in his painting.
 the Pollaiuolo painting altered to improve
format proportions

As another example, here is the Pollaiuolo

panel analyzed above, with the saint lowered to
bring his shoulders onto mm, his hips to m, his groin
to w' and his feet to M. Although the changes are
very minor, they significantly alter the image. The
archers do not seem to be working as hard, nor the
saint suffering as much: tension has been lost. In the
original the saint's unseen heart is at the intersection
of mm and C and at the apex of the archers' aim; his
head pushes against the picture limits much as his
spirit strains to leave its body. In the revised version
he just seems to be having a bad day.

By making the archers correspond closely to the

format divisions, Pollaiuolo invested
the displacement of the saint's body from these
divisions with greater expressive significance: the
misalignment symbolizes the agony. More generally,
the effect of using the format proportions
consistently throughout a design is to produce a
more balanced and integrated image, which can also
cause the picture to seem static or tidy. For some
painting goals this will not be desirable!

the Manet painting altered to "improve"

format proportions
And here finally is the Manet painting I complained
about earlier. Here the changes, though clumsy, are
more substantial and involve almost all parts of the
image. In particular the sky has been opened behind
so that the arc of four heads leads into the bright
distance, and landscape layers have been aligned
with the divisions to accent this recession. The
basket has been moved up and in, the figures shifted
and edited, the near trees brought to the strong
picture divisions, and that strange alley of lawn
diverted to run off the side rather than into the
picture.

The composition has become more balanced and

stable, and in the process more domesticated. The
grouping of figures is tightened in a way that seals
them off from the uncanny woods around them,
there is more light and bourgeois interest in keeping
food nearby. One misses the unsettling effect in the
original of the strong tilt of the figures and lawn
downward to the left, which doesn't consciously
register until it has been removed. The revised
version makes it seem as if ladies go naked in the
park because men are docile and the lawn free of
insects. The insouciance of the avant garde has been
lost.

use what works

I conclude with a hazard that is obvious by now: that
one can "explain" any random distribution of forms in
a rectangle with some system of proprotional or
geometrical divisions, provided you can use as many
lines or curves as necessary. The problem is
especially vexing in Charles
Bouleau's Charpentes, because he applies circles,
ellipses, triangles, interval divisions, musical ratios,
diagonals and other constructions differently from
one painting to the next, with no systematic
justification and often imprecise results.

Reproduced below as one example is his

"explanation" of Vermeer's Allegory of
Painting, which requires a "framework" of 45 unique
lines!
 the arbitrary effect of explaining a
composition with too many lines

Most of these lines do not define any important edge

or form in the painting. Analytic clarity is lost in the
complexity, and we have no reason to believe
Vermeer ever considered such a network or that
these lines indicate how most viewers would look at
the painting. In other words, the Boileau principles
are not fundamentally analytical.

A more elegant explanation has been developed by

Robert DiCurcio as the Grail Geometry in Vermeer's
images. I can't do justice to his ideas here, but the
pictorial analyses that result are substantially simpler
than Boileau's; the flavor can be conveyed in
DiCurcio's anchoring of a hexagram in the image
geometry (image below).
 a symbolic geometry derived from image
elements

DiCurcio's point is not that we can derive patterns

from the image, but that Vermeer (said to be
represented by the painter in the image) arranged
pictorial elements as a covert code — in this case, of
his secret religious beliefs in a violently contentious
historical epoque. Again, however, Vermeer's
religious symbolism is probably not of interest to a
modern painter, and methods of symbol encryption
are not useful methods of painting design.

How does my modest system of format proportions

fare? The illustration (below) indicates several points
of convergence, although there is no guarantee that
the digital image dimensions match the original —
either as it was framed by Vermeer, or as it exists
today after centuries of curatorial trimming and
relining.
 the format proportions

the green rectangle encloses the area on the painter's

canvas

However my system catches alignments missed by

the others, and in particular nicely characterizes the
location of the female subject within
the c' and m' defined keystone area and the painter
between the h' and c' folds outside it.
However, it seems to me the assumption that the
painter in the image is actually Vermeer raises an
obvious question: what kind of compositional
framework is the painter using? Vermeer shows
enough of his painter's work for us to impose his
canvas formats onto the image (green rectangle).
And from this we can construct the format
proportions of the painter's composition (image,
right).

Not only do my format proportions locate several

significant features of the female image — her eyes,
the format proportions in
wrists, hat, book corners, etc. — within the painter's
the painter's painting
painting, the major horizontal
divisions h' and M correspond closely with the
horizontal divisions c' and M of Vermeer's painting!
If there is a visual code hidden in this painting, it
seems to be a key to the way Vermeer himself
constructed his images.

My approach has been to choose the proportional

constructions in advance, apply them consistently to
historical examples or my own works, and from this
study develop principles that describe the effects that
result when a design adheres to or ignores the
format proportions. This minimizes the arbitrariness
of the activity and increases my understanding of
design decisions.
In that spirit, I urge you to browse any book of art
reproductions, choose paintings that seem especially
well composed or satisfying to you, then establish
the format proportions in the image to see how well
these correspond to significant guidelines. Try them
in your own paintings, too.

Last revised 11.I.2015 • © 2015 Bruce MacEvoy