Getting acquainted with

M i n i v a c 6 0 1

S C IE N T IF IC r. j |
D EV ELO P M EN T .. .3 T ir S t S te p

c o r p o r a t io n tow ard th e exciting world o f electronic

w— d ata processing via digital com puters

 I he M IN IV A C 601 computer and
this book are used together to demonstrate
the operation of digi+al computer components
and to teach p rinciples basic to their use.
As you become fa m ilia r w ith M IN IV A C 601
and understand its relationship to large
digital computers, you w ill be performing
the operations basic to a modern ele ctro n ic
data processing m achine.

This is the first in a series of books

using M IN IV A C 601 to demonstrate the
fundamentals of modern digital computers.
This book acquaints you w ith M IN IV A C 601 .
The second book in this series relates
the components of M IN IV A C 601 to
corresponding parts of a large commercial
digital computer. O ther books enable
you to use M IN IV A C 601 to discover
how these parts function to make
d ecisio ns, to do a rith m e tic, and
CONTENTS
to do p ractical work for men. A
book of games w hich may be
played w ith M IN IV A C 601
is also provided.
 P A G ES
I ) POW ER S U P P L Y ......................................................................................... 3
7.) P R O G R A M M IN G W IR ES....................................................................................................... 4
Closed C ircuit' P rin c ip le ........................................................ 4
Using the Programming W ire s............................................. 4
Experiment 1: Use of ProgrammingW ire s................... 5
W iring Shorthand......................................................................... 5
3fc) L IG H T S . ...................... 6
Experiment 2: Single Light O p e ra tio n ....................... 7
■
4 ) PU SH B U TTO N S...................................................................................................................... 8
Experiment 3: Pushbutton O p e ra tio n ........................... 9
Experiment 4: Series C ir c u it ........................................... | q
Experiment 5 : Parallel C ir c u it ............................................u
5) R E L A Y S ......................................................................................................................12
Experiment 6 : Manual Relay O p e ra tio n . . . ............ 1 3
E le c tric a l O peration of the R e la y ..................................... 14
Experiment 7: E le c tric a l Relay O p e ra tio n 15
Experiment 8: One Relay Controls A nother..........
Experiment 9: A Relay Controls Its e lf...........................17

__ ___ESP(i5Pi
f SLID E
S W IT C H E S .
.......................... * ............ . ......................... ............................ 18
Experiment 10: Slide Sw itch O p e ra tio n ........................19
'R O T A R Y S W IT C H ...................................................................................................................... 20
Experiment 11: Manual O peration of
Rotary Swi t c h ......................... 21
Motor D riv e .......................................................................................22
Resistors............................................................................................ 22
Decimal Inp ut-O utput........................................................... 22
Experiment 12: Basic Motor O p e ratio n .................... 23
Experiment 13: Motor Driven Rotary S w itch . • • • 23
Pulse G e n e ra to r......................................................................... 24
Programmer.................................................................................... 24
Experiment 14: Motor Reversing........................................ 24
8 ) M A T R IX ..................................................................................................................... 25
9 ) A U X IL IA R Y T E R M IN A L S ........................................................................................... 25
1 0 ) L O O K IN G A H EA D W ITH M IN IV A C 6 0 1 ................................................................................................................ 26
Summary: Book 1...................................................................... 26
Book II: What is a D igital C o m p u te r? .................... 26
Book III : How Computers
Make Logical D ecisions 26
Book IV : How Computers Do A rith m e tic ................. 27
Book V : How Computers Work for M an .................... 27
Book V I: M IN IV A C G am es................................................... 27
PROBLEM S FO R M IN IV A C 6 0 1 ........................... 28
to 32
 The M ini vac Manual was prepared and edited by the staff of

S c ie n tific Development Corporation

First Printing— August, 1961

EX LIBRIS ccapitalia.net
Copyright (c\ 1961 by S c ie n tific Development Corporation, 372 Main Street, W atertow n, Massachusetts
 POWER SU PPLY
he power supply converts the 110-volt A C power obtained
from a standard e le c tric outlet in your home to a safe working
level of 12 volts D C . This low voltage DC is used to
power M IN IV A C and is connected to a ll terminals +
and - on the power p a n e l. Most toy e le c tric trains
use the same low -voltage power as M IN IV A C and
therefore working w ith the M IN IV A C is as safe as
operating an e le c tric tra in .

The POWER SW ITCH is

used to turn power O N and O F F .
The power should be O F F when­
ever you are w iring the
computer. Following this
p ractice w ill establish good
working habits and w ill help
to prevent accid ental short-
c irc u its . The power must be
turned O FF when the computer
is not in use. The red power light
shows when the power is O N .

The red button above the power switch labeled C IR C U IT BREAKER*

is a safety switch w hich au to m atically turns o ff the power supply
when it is overloaded. An overloaded c irc u it or a sho rt-circu it
w ill cause the c irc u it breaker to "b lo w " just lik e a household
fuse. A short c irc u it occurs when + is wired d ire c tly to -
without a component in betw een. The power supply
may be restored to normal operation after a short-
c irc u it by changing the fa u lty w iring and then
pushing the red circu it-b re a ke r button back in .

C A U T /o n :
Do N O T hold the c irc u it-
breaker button in if it
continues to "b lo w ." Turn N o w .. . .plug the POWER C O R D into
power o ff and determine the a standard 110-volt A C (N O T DC)
cause for drawing excessive o u tle t, and you are ready to
current before proceeding G E T A C Q U A IN T E D
w ith the experim ents. W ITH M IN IV A C 6 0 1 ...........

3
w

2 PROGRAMMING W IRES

^Programming wires of different lengths are provided to make e le ctrica l

( programming) connections between various terminals on M IN IV A C 601 .
These wires establish closed e le c tric a l circ u its and route the current through
one or more components to create a desired computer c ir c u it .

Each component in the computer

is connected to nearby terminals
by permanent connections made
under the computer p a n e l.

The components may be interconnected by

inserting programming w ires into the term inals.
Each terminal has two holes side-by-side w hich
are connected together underneath the computer
panel . Programming w ires may be plugged into
either (or both) holes of the term in a l. A ll
programming w ires are color-coded by length.

C LO S E D C IR C U IT PR IN C IPLE

For any e le c tric a l d evice to operate, it is

necessary for e le c tric a l current to flow
TH R O U G H the d e v ic e . A closed e le ctrica l
A
Electrons our
c irc u it must be established to let the
electrons flow from the negative power
supply terminal ( - ) through the
component to the positive power supply
terminal ( + ) . Thus, a ll e le c tric a l
components on the M IN IV A C 601 require
at least 2 connections to operate.
ElectronsIN
U S IN G THE P R O G R A M M IN G WIRES

The programming w ires are used to connect

various components of M IN IV A C 601 . Closed
circu its are wired through the term inals provided
on the console for each component. Each
terminal is id entified by a letter printed below
i t , and by the number printed at the
top of the section in w hich it is lo cated .
This makes possible easy id e n tificatio n of
terminal points when w iring components
together in future experim ents. We are
now ready to use the programming w ire s.

4
EXPERIMENT |

OF PROGRAMMING WIRES
his experim ent demonstrates the use of programming
wires to make connections on M IN IV A C 601 .

l y Take two programming w ire s. G rasp one

end of the first w ire firm ly between your
thumb and forefing er.

Connect this end of the programming

w ire to terminal 1 + on the
computer panel by pushing
it into one of the two holes
just below the + in section 1
of the computer.

3 ) Take the other end of this program­

ming w ire and connect 1 + to
terminal 1 A by pushing it
into one of the two holes
just above the letter A in
section 1 of the computer.

^ / S im il a r ly , connect terminal 1 B to
terminal 1 - w ith the second programming
w ire . These connections cause lamp 1
to light when the Power Sw itch is turned o n .
This simple c irc u it is discussed
further in Experiment 2 .

Disconnect the programming w ire by pulling both ends from the

computer term inals. Do N O T disconnect the program­
ming wires from the console by pulling on the w ire
because the w ire may b re ak. A ll programming wires
should be disconnected after each experiment or problem.

W IR IN G SH O RTH AN D
A W IR IN G SH O RTH AN D is used to in d icate connections w hich are to be
made on the computer by simply w riting the two terminal designations
w ith a slash in between. In Experiment 1 , terminal 1 + was connected
to terminal 1 A . This is w ritten in shorthand-------- — ■, /
A series of such shorthand w iring instructions w ill be given to ' / 1/^
describe the w iring of a complete c ir c u it .

5
 < id entical lights Iabeled
"B in ary O utput" are used in
M IN IV A C 601 to show what is
happening in various parts of the
computer. A light w ill come on when
is connected in a closed c irc u it so
rrent flows through the lig h t.

This allow s the computer to tell an operator what

it is doing and to display the answer to a computer problem,

Each light is id entified by the section

number d ire c tly above i t .

In order for a light to turn O N (to lig h t),

current must flow TH R O U G H the lig h t.
The electrons must flow IN on one w ire
and O U T on another w ire . This is the
closed c irc u it p rinciple for lig h ts.

The terminals labeled A and B on either side of

each lig ht are used to supply the e le c tric a l
power to turn the light O N . A light comes
O N when terminal A is connected to one
power supply terminal ( + or - ) and terminal
B is connected to the other power supply terminal ( - or + ) .

l^o Co n n ectio n
When the circuit is
OPENED U §ht
off
fhe light is OFF

When the circuit ig

CLOSED >o )

the lig/ntcomesON
*\V '
* \
 —I
EXPERIMENT 2

SINGLE LIGHT OPERATION

his experiment demonstrates the operation of

a single binary output lig h t.

Using two short programming w ire s,

i;
make the connections 1 + / 1 A and 1 - / 1 B.

Turn the power switch O N (be sure

the main power cord is plugged into a
proper o u tle t). Light 1 comes O N ’
because current flows through the lig h t.

V Turn the power switch O F F and

disconnect the programming w ire s.
Power should be turned O FF and
a ll programming w ires should be
disconnected after each experiment
is com pleted.

V A ny of the lights may be operated by

1+ 1A connecting programming wires from + to A

and from - to B just as was done for light 1.

Make connections to two or more lights at once.

(The c irc u it diagram looks lik e this)

I - IB

The lights are used in later experiments to

represent numbers. The light O FF w ill represent Binary Number System
a zero ( 0 ) and the light O N w ill represent a one ( 1 ) .
'" • v
The numbers zero ( 0 ) and one ( 1 ) are the only numbers
used in the binary number system, and it is for this
S L ■
reason that the lights are labeled " Binary O u tp u t."
Binary numbers and their use in digital computers 0 FF= 0 0 N = |
Wl be taken up la te r.

7
 PUSHBUTTONS

he section of the computer panel labeled "B inary Input" contains

six identical pushbuttons w hich are used by the operator to supply
information to the computer. The pushbuttons may be used to tell the
computer what to do, or to in dicate the value of input numbers.
Pushbuttons may be used to control the other components of M IN IV A C 601
Each pushbutton is id entified by its section number on the computer p a n e l.

Each of the six binary input pushbuttons has three terminals located d ire c tly
above it on the computer p a n e l. Each terminal is connected under
the panel to a co n tact. The term inals are labeled X , Y and Z .

The three contacts for each of the six

pushbuttons are a ll connected to terminals in the same manner:

♦N
When the pushbutton is U P, an e le c tric a l
connection is made between terminals Y
and Z . When the pushbutton is D O W N ,
the e le c tric a l connection is switched to
terminals Y and X . Terminal Y is connected
to the "common" contact and is alw ays used when up= y/z Dowm= x/ y
the pushbutton is connected in a c irc u it .

When a pushbutton is shown in a c irc u it

diagram , the contacts are A LW A Y S shown
ir :l
w ith the pushbutton UP - in the "norm ally
closed" ( N . C . ) position. N o tice that
j !*2.
term inals X and Y are not connected when
the pushbutton is U P, and that terminals
Y and Z are not connected when the
pushbutton is D O W N .

The follow ing experiments demonstrate basic circ u its in w hich

one or two pushbuttons are used w ith a lig h t.
EXPERIMENT 3

PUSHBUTTON OPERATION
S t . .
I I his experiment uses a light to demonstrate the operation of a pushbutton
Both norm ally open and norm ally closed circu its are connected.

jl x
>) The normally open pushbutton-light 1+ -*—
circuit drawing looks like this:
Program this circuit by making
* ) the following connections: 1+/1Y, 1)(/1A, 1-/1B I -
Turn the power switch O N . Push DOWN on pushbutton 1. Light comes ON because the
3) circuit from + to - is closed through the pushbutton contacts and through light 1 .
Release the pushbutton and the light goes O FF. The light goes OFF because the circuit
from + to - is open (broken) by the pushbutton contacts. Turn the power switch O FF.
1Z
With a normally closed pushbutton-light added, ^ -
«)
the circuit drawing looks like this:
Program this by adding connections — 1Z/2A, 2B/2- ^
Turn the power switch O N . Light 2 comes on as soon as power IX
6) is turned O N . Light 2 lights because the circuit from + to -
is completed through the normally closed contact (Z) of pushbutton 1 .

7) Push down on pushbutton 1 . Light 2 goes OFF and light 1 comes O N .

Light 2 goes OFF because the circuit is opened by the switch and
1A 2A
electric current can no longer flow. Light 1 comes ON because the
circuit is closed through the switch.

s\ Push up and down slowly on pushbutton 1 and carefully watch the 2B

operation of the two lights. One light always goes OFF before the other IB
light comes O N . There is a position, with the pushbutton only partly
depressed, where both lights are O FF. Try to find this position. The
switch breaks one connection before it makes the other connection. The
pushbuttons are thus called "break-before-make" switches. This is important because it
I -
z-
shows that the normally open contact is never connected directly to the normally closed contact,
even for a very short time. This feature is necessary for the proper operation of some M INIVAC circuits.

This experiment demonstrates the two basic ways to use a pushbutton. The norm ally closed contact
( Y / Z ) may be used to complete a c irc u it when the pushbutton is U P , and the norm ally open
contact (X / Y ) may be used to complete a c irc u it when the pushbutton is D O W N .

The pushbuttons will be used in later experiments to

represent numbers. The pushbutton UP will represent a
Zero (0), and the pushbutton DOWN w ill represent a one (1).
Binary fdunnber Syste
The numbers, zero (0) and one (1) are the only numbers
used in the binary number system. and it is for this _ l
reason that the pushbuttons are labeled " Binary
Input. " Binary numbers, and their use in digital
computers, w ill be taken up later.
U P = 0" DOWN =*<■//
1
EXPERIMENT 4

SERIES CIRCUIT
I his experiment demonstrates the use of
two pushbuttons connected together in series
to form a c irc u it where both pushbuttons must be
pushed at the same time before the c irc u it is closed.

0 Program this c irc u it by making the

follow ing connections: 1+ /1Y 2> y2A
1 V 2Y 2 B /2 -

Turn the power switch O N and push down

pushbutton 1 . Watch light 2 . It does N O T
light because the connection from + to
the light is broken by pushbutton 2 .

3 / Release pushbutton 1 and push down push­

button 2 . Watch light 2 . It still does N O T
light because the connection from + to
the light is now broken by pushbutton 1 .

4) Push down both pushbutton 1 and

pushbutton 2 at the same time and see light
2 go O N . A complete c irc u it is now
made through both pushbuttons.

It is possible to connect more than two push­

buttons in series and make a c irc u it where aM the The c irc u it diagram looks lik e thi
pushbuttons connected must be pushed down before 1Z 2Z
the c irc u it is com pleted. Construct a series
c irc u it using three or more pushbuttons by your­ A - _________
l+ * iy ix vi
s e lf, to demonstrate this p rin c ip le . The c irc u it ,
w ith terminal numbers in d ica te d , should first be
drawn on paper. The c irc u it should then be
connected on the computer using programming w ire s.

A N "A N D " C IR C U IT — The series c ir c u it , where two or more

pushbuttons must A L L be closed to make £ complete c ir c u it , is ca lle d
an A N D c ir c u it , because both pushbutton 1 A N D pushbutton 2 must be closed
before the c irc u it is com plete. This is a basic c irc u it of computer
logic w hich w ill be used la te r.
EXPERIMENT 5 !

PARALLEL CIRCUIT
his experiment demonstrates the use of two
. I pushbuttons connected together in p arallel to
form a c irc u it where either pushbutton may be
pushed to complete the c ir c u it .

I) Program this c irc u it by making

the follow ing connections: 1+/1Y V Y /2 Y
1V IA 2V 1 X
1 B /1 -

Turn the power switch O N . The

light does not come on because there
is not a closed c ir c u it .

3) Push down pushbutton 1 . Light 1 comes

O N because a closed c irc u it from +
through the light to - is now made
through the norm ally open
contact ( I X ) of pushbutton 1 .

4) Release pushbutton 1 and push

down pushbutton 2 . The light again
comes O N because a closed c irc u it is
now made through the norm ally open
contact ( 2 X ) of pushbutton 2 .

5) Push down both pushbutton 1 and pushbutton 2 at

the same tim e. The c irc u it is closed and the light The c irc u it diagram looks lik e this:
comes O N if either pushbutton 1 is down O R if
12
pushbutton 2 is down OR if both pushbutton 1 and
pushbutton 2 are down at the same tim e. 1 ^ 0. 1Y IX
« ♦ 1A

0
6 ) More than two pushbuttons may be connected in
p arallel to make a c irc u it where any one of the
pushbuttons may be pushed down to complete
the c ir c u it . To demonstrate this p rin c ip le , construct 2Z •IB
a p arallel c irc u it using three or more pushbuttons.
Draw the c irc u it diagram , and then connect the c irc u it
on the computer using the programming w ire s. 2*

A N "O R " C IR C U IT — The p arallel c ir c u it , in w hich closing either

I -
of two or more contacts completes the c ir c u it , is ca lle d an OR
c ir c u it . Pushing down either pushbutton 1 O R pushbutton 2
completes the c ir c u it . This c irc u it is basic to much of the
computer lo gic w hich is used la te r.
11
 ^ ix identical relays labeled
"storage-processing" are used by M IN IV A C 601 to store or "remember"
and to process computer inform ation. Each re lay has two sets of contacts,
sim ilar to the pushbutton co ntacts, w hich open and close
together. The relays are operated by an e le c tric current
ch flows Through the re lay c o il, whereas the
pushbuttons are m anually operated.
The re lay is p a rticu la rly useful because it can
be connected so it w ill "remember" whether it is O N or
. O F F . Every computer has some d evice w hich is able
£ £ LA / : / X . "rem em ber." In M IN IV A C 601 this d evice
^ i s the re la y . The relays may be used
V sm C w beriro J1
X s-*. to control any of the components
device
used in M IN IV A C 601 .

Each of the six relays has six terminals

located d ire c tly below it on the computer panel
These terminals are connected under the computer
panel to contacts of the re lay sw itch . The
term inals are lab eled : G , H , J , & K , L , N .

RELAY OFF RELAY ON

When the relay is O FF, separate electrical con­ The two arms (H & L) are connected together.
nections are made between terminals H and J and Both contacts always close together.
terminals L and N . When the relay is O N ,
separate electrical connections are made between The relay switch contacts are connected to
terminals H and G and terminals L and K . terminals in the same pattern used with the
pushbutton contacts. The normal Iy-open (N .O.)
When relay contacts are shown in a circuit contact is connected to the left terminal, the
diagram, the contacts are ALWAYS shown with the relay arm (common contact) is connected to the
relay O FF. Notice that terminals H and Gand center terminal, and the normally-closed (N .C .)
terminals L and K are not connected when the relay contact is connected to the right terminal. The set
is O FF . Also, terminals H and J and terminals L of contacts closer to the operator is connected to
and N are not connected when the relay is O N . the set of terminals closer to the operator.
The M IN IVAC relays are Plasty Crossbar^) The two relay arms are connected together
mounted on the computer mechanically with a plastic crossbar. Both arms
panel with the contacts of j Arms on move together and make or break contact at the
norma/ly
the relay switch visible. The I same time. Observe the relay operation by
closed
relay arm is in the middle, contacts gently pushing on the plastic crossbar with your
between the other two contacts. hand. Notice how the arm leaves the normally-
Terminals are connected closed contact and swings over to touch the
to the contacts like this: normally open contact.
6 HJ 14 L N
^ T e rm in a ls /
EXPERIMENT £

MANUAL RELAY OPERATION

m is experiment uses two lights to demonstrate

the basic operation of a re la y sw itch .

> Connect the contacts of re la y 1

/ to turn light 2 O N when the re lay
is O F F , and to turn lig ht 1 O N when
the re la y is O N .

for this c irc u it is

1G / 1A
2 B /2-

Turn power O N .
Light 2 comes O N

A.) G e n tly push the re la y crossbar

to the le ft. Light 2 goes O FF and
light 1 comes O N . W atch the re lay
arm a lte rn a te ly touch one contact The c irc u it looks lik e this
and then the other as the relay
crossbar is moved back and fo rth.

1G IB
1 -
1A

13
 5 Electrical Operation of the Relay
covfO

gif U n lik e the pushbutton w hich must be operated m an u ally, the

re la y can be operated e le c t r ic a lly . Each re la y has a round
w hite c o il, located to the le ft of the arms. This is an electro ­
magnet w hich becomes magnetized when e le c tric current flows
through the c o il. When the co il becomes "e n e rg iz e d ," the e le ctro ­
magnet attracts the metal bar connected to the arms (by the plastic
crossbar) and pulls it to the le f t . Sending current through
the coil of the electro-m agnet operates the re la y s.
— Thus, the re la y may be operated e le c tric a lly without
touching it p h y sic a lly , but the pushbutton must
Terminals fv opera-he
be operated m anually.
i Relay
^ The re la y is turned O N by connecting the re la y coil in
a closed e le c tric c ir c u it . When the c irc u it is closed current
OO flows through the coil and the coil is "e n e rg ize d ."

Switches ———

jm m .
j /
USHT % !?
~Jethr\in2ls Connected
Three terminals are located d ire c tly above
~hSwitches “
_ each re lay on the computer p a n e l. These
OO oO \ . . 7 i i i i
H J terminals are connected under the panel to
the re la y c o il. Two terminals are connected
A . d ire c tly to the re lay c o il, and the third
OO J terminal is connected to a light wired in
t- N / series w ith the re la y coil as shown.
—— When the re la y co il is energized
by connecting power to C and F ,
the light d ire c tly below the re la y comes
O N because the light is wired in series
with the re la y c o il. If the re la y coil is
energized by supplying power through contacts
E and F , the light beneath the re la y w ill N O T
come on because there is no current flowing through
the lig h t. The light below the re la y coil may be used to
show when the re la y is O N .
The c irc u it drawing for a re la y switch looks lik e this: m*00
The follow ing experiments demonstrate re lay operation
with pushbuttons and lights in simple computer circu its Switches A lw a Y S
Rfi-tay
operate together
 EXPERIMENT
ELECTRICAL RELAY OPERATION
^ T h is experiment demonstrates the basic e le ctrica l
operation of a r e la y . Additions to the "manual
| re la y operation" c irc u it of Experiment 6 permit
-I the re la y to be operated e le c tric a lly
through a pushbutton.

0 The program for connecting the switch

contacts of re la y 1 as shown in
Experiment 6 is: 1+/1H 1 G /1 A
1 J/2 A 2 B /2-
1 B /1 -

2) The program for connecting the coil of

re la y 1 through the norm ally open
contact of pushbutton 1 is: 1+ /1Y 1 F/1 -
1 X /1E
Turn power O N . Light 2 comes O N .
3)

4) Press down pushbutton 1 and see the

re la y operate. The re la y makes a " c lic k "
as the coil is energ ized , and the re la y contacts
switch the lig h ts. Light 2 goes O FF and
light 1 comes O N . O perate the re la y several times
w ith pushbutton 1 and watch the movement of
the re la y co ntacts.

5 J Now, connect the other set of re lay contacts

(1 K , 1 L , I N ) to two unused lights so one lijght
comes O N when the re lay is O FF and the
other light comes O N when the re la y is
O N . Try this yourself.

£ ^ T o modify this program so that the coil indicator

lamp w ill light when the coil of re lay 1 is
energized: remove connection 1)^/1 E
add connection D y lC
The c irc u it for this modified part of the program is:

0
12 The dotted lin e in this c irc u it indicates that
1+ iy IX 1C I9
E 00000 I Ff I- contacts 1 G , 1 H, and 1J are part of the re lay
switch controlled by the coil connected
—• -
between IE and I F .

P LA TE The re la y is made w ith a small electro-m agnet c o il. When current flows through
the c o il, a magnetic force is produced w hich pulls on the small metal plate attached
to the p lastic crossbar to switch the co ntacts. In this experim ent, we have seen the
operation of the electro-m agnet and have operated the re la y to control a set of lig h ts.
We have also seen the re la y indicator lamp light when the re lay was turned O N .
EXPERIMENT 8

RELAY CONTROLS ANOTHER

■his experiment shows how a re la y may be used to

control another r e la y .

| ) The program for this c irc u it is:

14-/1V 1+/1H 2+/2H

1V1C 1 G /2 C 2 G /2 A
1F / l - 2 F /2 - 2 B /2 -

2*3 Turn power O N and m anually

operate re la y 2 by pushing the
p lastic crossbar to the le ft.
See light 2 come O N .

31 Release re la y 2 and m anually

operate re la y 1 . As you push
on the plastic crossbar
of re la y 1, re la y 2 w ill c lic k O N
A N D L IG H T 2 W ILL C O M E O N .
The switch contacts of re la y J_
supply power to re la y 2 by closing
the c irc u it to the coil of re lay 2_.

Release re lay 1 and press pushbutton 1 .

Relay 1 is turned O N by pushbutton 1,
re la y 2 is turned O N by the switch
contacts of re la y 1 , and light 2 is turned
O N by the switch contacts of The c irc u it looks lik e this:
relay 2 . Thus, the pushbutton 2J
controls relay 2 through the
contacts of relay 1 . 2+. 2H i 23 2A

5>1 It is possible to connect several

relays in a series, where each relay
controls another. Expand the
present c irc u it to use a ll six relays
w ith each re la y controlling the
coil of the n e x t. The last relay
in the series, re la y 6 , should
turn lig ht 6 O N .
 —I
EXPERIMENT

A RELAY CONTROLS ITSELF

I his experiment demonstrates how a re lay
may be used to control its own operation.
The re la y is connected to remain O N when it
IJ The program for this c irc u it is: is turned O N , and to remain O FF when it is
J IF / 1 - 2 Y / 2+ turned O F F . The re lay remains in one of the
1 C/1 X 2 Z /1 L two conditions (O N or O F F ) until signalled
1 Y /2 Y 1 K /1 C by one of the pushbuttons to change to the

Push pushbutton 1 and release. The

re la y indicator light comes O N and stays
O N . The re la y stays O N because an e le c tric
c irc u it is completed through its own contacts
to energize the c o il.

Push pushbutton 2 and release. This

momentarily interrupts current flow
through the coil and turns the re la y O F F .
The re la y indicator light turns O FF and
shows that the re la y is no longer energ ized .
The re la y c irc u it has now returned to its
original co nd ition.

O perate the c irc u it back and forth ^ '

a few times by first pushing push­
button 1 ; then pushbutton 2 .

The c irc u it used in this experiment is ca lle d

"b istable" because it is able to remain in
either of two stable conditions. Many
different types of "b istab le" circ u its are
used in modern digital computers.
t+
r
IN

2X
The two stable conditions of the re lay c irc u it
(O N and O F F ) are used by the computer in later
experiments to represent two numbers. When the 1Z
re lay is O F F , the number zero (0) is represented.
When the re la y is O N , the number one (1) is
0000 0
I-
represented. The re la y c irc u it thus remembers #
iy IX 1C 1£ IF
(stores) these binary numbers.

17
r

SLIDE SWITCHES
$ i x id e n tica l slide switches labeled "Secondary
Storage" are used by M IN IV A C 601 for manual
storage of data and for other computer switching
functions. Each switch has two sets of contacts
w hich open and close together.

The slide switch is set to one of two positions by the computer

operator, and it stays in that position until moved. The
switch thus "remembers" its position and "stores" this
inform ation. The slide switch also provides a convenient
means for changing circ u its w hich are programmed
on the computer, and it may be used for
other computer switching functions as w e ll.

Each of the six slide switches has six term inals

AO 0®
located d ire c tly above it on the computer
p anel. These term inals are connected
underneath the panel to the
slide switch co ntacts. The
term inals are labeled
R ,S , T , U , V , W , and
are connected in the same
manner for a ll six sw itches.
Each switch is identified
by its section number.

4 » *

When the switch is pushed to the le ft, When the switch is pushed to the rig ht,
separate e le c tric a l connections are made separate e le c tric a l connections are made
between terminals S and R and between between terminals S and T and between
term inals V and U : terminals V and W :

R S V w R S T W

18
 EXPERIMENT 10

/ ' J k
SLIDE SWITCH OPERATION
his experiment demonstrates the operation
of a slide switch controlling two lig hts.

Program this c irc u it by making

the follow ing connections:

1-/1B 2 -/2 B
1A /1R 2A / 1 W
1S/1+ 1V/1S

2 ) Move switch 1 to the LEFT

and turn the power O N .

V Light 1 comes O N but light 2

remains O F F . Trace the closed
c irc u it for light 1 and the open
c irc u it for light 2 on the
c irc u it diagram .

4 / Move switch 1 to the right.

Light I goes O FF and light 2
comes O N . Trace the open c irc u it The c irc u it for the experiment looks lik e this:
for light 1 and the closed c irc u it

19
 § h e rotary sw itch , labeled
"D ecim al Input-O utput,
may be m anually rotated
or motor d riv e n . Connected
d ire c tly with the pointer
knob there is a w iper arm
underneath the computer
p an el. As the w iper arm and
the pointer knob move around th
c irc le of co ntacts, the w iper arm
momentarily connects with each set of
contacts on the rotary switch d ial in turn

The w ip er arm rotates with the pointer knob and makes

an e le c tric a l connection w ith each set of contacts
in turn around the rotary switch d ial (see the
follow ing illu stra tio n ). The contacts
are designated by:

D = Decim al Input-Output
8 = terminal number (1 to 19)
. . . for example D8

The contact on the w iper arm is

referred to as "D 16. "

For exam ple, when the pointer

knob points at 1 , the w iper
arm makes an e le c tric a l
connection between
D l and D16.

V IE W UAIDER P A N E L

20
EXPERIMENT II
I

MANUAL OPERATION OF ROTARY SWITCH

i
| | his experiment demonstrates basic manual
-• operation of the rotary sw itch .

Connect the rotary switch to

operate lights 5 and 6 when the
pointer knob points to numbers
1 and 2 re sp e ctive ly.

I/ The program for this c irc u it is:

6+/D16 D1/5A
D 2/6A 5B/6B
6 - / 6B

2 / Turn power O N and rotate the

pointer to number 1.
l. Light
Lighi 5
comes O N and remains O N
as long as the pointer
points at number 1 .

Rotate the pointer to number 2 .

Light 5 goes O FF and light 6 comes
O N . N o tice that light 5 goes O FF before
light 6 comes O N . This shows that the
rotary switch is designed to disconnect
the arm from one contact before making
connection w ith the next co ntact.
The switch operation is described as
"b re a k-b e fo re -m ak e ," and thus is
sim ilar to the pushbutton sw itches.

The c irc u it looks like this:

4) Rotate the pointer knob
D16 ’
5A
back and forth a few
times between numbers 1 and 2 .
6+ VI
6A
vz
S) Connect several other lights
to the rotary switch terminals
©
to demonstrate that an e le c tric a l SB
connection is made from D16 to each 6-
term in al. More than one terminal
may be connected to each lig h t. When m anually operated, the rotary \
Try this yourself. switch is used to send information to the
computer from the operator. Thus, by
Connect the rotary switch to operate positioning the pointer knob, it is
one or more re la y s. Control a light possible to communicate information to
w ith each re la y w hich is energ ized . the computer. The rotary switch w ill be ,
used for this purpose in later experim ents. /
11
 7 m o t o r d r iv e
(corrf.)

The rotary switch may be positioned

either by hand or by the motor d riv e .
A DC motor turns the rotary switch
through a frictio n d riv e . The
rotary switch connections are
made seq uentially whether the pointer
is turned by hand or driven by the motor.

The rotary switch motor is an

e le c tric a l component w hich
requires two connections just like
a light or re la y . Current must
flow through the motor to make it
run and turn the rotary sw itch . The
c irc u it drawing of the motor Terminals 17, 18 and 19 are located on the
looks like the above. panel just to the left of the pointer knob
and rotary switch term inals. Power must be
supplied to terminals 17 and 18 (run) to operate
the motor d riv e . A short c irc u it may be
connected across term inals 18 and 19 (stop)
to halt the motor d riv e .

RESISTO R S
The e le c tric a l resistors used with the
motor drive lim it the current w hich
flows through the c irc u it in the same
way that the lights lim it the amount of
current w hich flows through a relay c o il.
The resistors make it possible to connect a
short c irc u it across the STO P term inals to stop
the motor without over-loading the power supply.

DECIMAL INPUT-OUTPUT
The motor driven rotary switch is ca lle d
a Decimal Input-O utput because it has
positions w hich are designated by ordinary
decimal numbers. "Input-O utput" means
that the rotary switch may be used by the
operator to m anually d e liv e r information to
the computer or that the computer may
au to m atically d e liv e r information to the
operator. The Decimal Input-Output is thus
used by the operator to communicate w ith
the computer using decimal numbers.
The follow ing experiments demonstrate the
rotary switch and motor drive operation.

22
experim en t
Ba s ic m o to r , o p e r a t io n
f his experiment demonstrates the basic motor drive
operation. The c irc u it looks like th is:

Sz / I Connect the motor to operate with pushbutton

5vJ/_ 5X pi7
control using the follow ing program:
6*+ -•-- 6+ /5Y
5 X /D 1 7
6 -/D 1 8
6 - / 6Y
6 X /D 19

6- D 18
—•-
Turn the power O N . Push pushbutton 5 .
This supplies power to the RUN terminals
(D17 and D18) and causes the pointer of the
rotary switch to rotate.

6z 3 y W hile holding pushbutton 5 down, push pushbutton 6 .

The motor stops when pushbutton 6 is pushed because
a short c irc u it is created across the STO P term inals
6Y 6X Pl9 (D18 and D 19). Current flows through pushbutton
6 instead of through the motor.
W hile holding pushbutton 5 down, release pushbutton 6 . The
_ _ _ _ _ | motor drive operates again because the short c irc u it has been removed.

e x p e r im e n t 1 3 ' $ Release pushbutton 5 . The motor drive again stops, this time because
power is no longer connected to the motor c ir c u it .

MOTOR DRIVEN ROTARY SWITCH

This experiment demonstrates how the motor driven rotary switch
may be used to generate e le c tric a l pulses for the computer and to
autom atically perform a series of operations in sequence.

J I Connect the motor to operate with Push pushbutton 5 . The motor rotates the
pushbutton control using the c irc u it of pointer knob as in Experiment 11. As
Experiment 12. the pointer knob turns, light 5 and light 6

^0 6
Connect lights 5 and to the rotary
flash O N in sequence. The rotary switch
is autom atically generating e le c tric a l
switch using the c irc u it of Experiment
pulses to flash the lig hts. F irst, light 5
11. The combined program from
flashes and then light 6 flashes, once each
Experiments 11 and 12 is:
revolution of the rotary sw itch .
6+/D16 D 1/5A 5 - /D 1 8
W hile holding pushbutton 5 down, push
D 2/6A 5B/6B 6 - / 6Y
pushbutton 6 . The pointer knob STOPS
6 - / 6B 6+ /5Y 6X/D 19
turning and the lights STO P flash in g .
5X/D17

23
 7
(con+.j
PULSE GENERATOR.
The electrical pulses which are used to flash light 5 (or the pulses which flash
light 6) continue as long as the motor continues to run. Pulses like these
are used to automatical Iv operate MINIVAC 601 in future experiments.
Most automatic digital computers have a mechanical or electronic device,
similar in function to the rotary switch, to generate a continuous series of
electrical pulses. This is done to make the computer operate continuously
and automatically at speeds much faster than a human could operate a switch.
Continuous pulses, generated to operate a digital computer, are called "clock"
pulses. The use of "clock" pulses in MINIVAC 601 and other digital computers
is explained and demonstrated in future experiments.

PROGRAMMER.
As the motor turns the rotary switch light 5, and then light 6 , flashes. This occurs in the same
order (sequence), once each revolution of the switch. Other lights could be flashed, or
relays could be energized, in any desired sequence by using other rotary switch terminals.
Thus, it is possible to use the rotary switch to carry out a whole series of operations on the
computer. The particular connections which are made to accomplish a series of computer
operations are called a "program." A ll digital computers, including MINIVAC 601 must be
"programmed" to solve a specific problem. The use of the rotary switch for "programming"
is explained and demonstrated in future experiments.

“ I
1 experim ent 1 4 !

MOTOR REVERSING
This experiment shows how the direction of rotation
I of the motor may be changed by reversing the direction
of current flow through the motor.
IJ Connect power to the
rotary switch motor through 6 +
two sets of slide switch contacts
so that the direction of current
flow through the motor may be
1
6w
reversed through the sw itch . The
c irc u it looks lik e this: D17
2. J The program for this c irc u it is:
6+/ 6 R 6T / 6 U 6V /D 1 7
6+/ 6W 6S/D 18
6 - / 6T
3) Trace the direction of current flow from
I The a b ility of the motor to reverse \
6 + through the motor to 6- w ith the switch
direction g reatly increases the value
in the right position and then again w ith the
of the rotary switch as a programmer.
switch in the left position. The current
Many future experiments make use of
flow through the motor is reversed when
the basic fact that the direction of
the switch position is changed.
rotation may be reversed by reversing
4-) Turn power O N and notice the direction \ the current flow through the motor, j
of rotation.

5J Chonge the position of switch 6 . The motor reverses

direction ond the pointer knob turns in the opposite d irectio n.
24
8 M ATRIX

1 he m atrix provides convenient

junction points for use when the computer
plays games or recognizes symbols.
The m atrix consists of a number
of term inals, located in the upper right corne
of the computer p an el. The term inals are used
as junction points, and there are no connections
to the term inals below the p an el.

The m atrix term inals are designated by:

M = m atrix
6 = terminal number (1 to 1 1 )
t = top or "b" for bottom
. . . for example M 6t
The lines between the term inals indicate
terminals w hich are connected together
to form a single co n tact.
The m atrix terminals are arranged in a
convenient pattern for playing games like
tic -ta c -to e or solving ch aracter recognition
problems. When the computer is programmed
to play t ic - ta c - to e , the various plays are
recorded by making connections on the m atrix
term inals. When the computer is programmed
to recognize ch aracters, the characters are
formed by making connections on the m a trix.
Games and ch aracter recognition problems
are presented in other books in this series.

9 AUXILIARY TERMINALS
A u x ilia ry term inals are provided
between the slide switches and the
relays for each of the six sections.

The terminals w ith four holes labeled

C O M M O N may be used when more
than two connections must be made to
a set of term inals on the computer
p a n e l. The four C O M M O N terminals
are connected together, but are not
connected to any other component.

The term inals marked "C A P A C IT O R "

are used when a u x ilia ry components are
added to the M IN IV A C 601 in advanced
programs. A t present these terminals
may be ignored.

25
 Book I - - "G ettin g Acquainted w ith M IN IV A C 601 " — has examined the operation
of each M IN IV A C 601 component. The symbols and shorthand of c irc u it drawings
have been introduced and basic computer programs were developed in the experim ents.

Now that you are com pletely fa m ilia r w ith the components of M IN IV A C 6 0 1 , you
are ready to explore the e xcitin g world of digital computers.

THERE ARE S IX B O O K S IN THE M IN IV A C S ER IES . THE O THER F IV E A RE:

Book II answers the question W HAT IS A D IG IT A L C O M PU TER ?

and introduces large scale e le ctro n ic data processing m achines.
You w ill discover how a digital computer:

— A ccepts input information representing basic data

to be processed and instructions describing the
ca lcu latio n to be performed.

--Stores (remembers) input and other information w hich it generates.

— Processes information and controls its operation, making logical

decisions and performing a rith m e tic.

— G enerates output information representing answers to a problem.

The components g en erally used in modern d ig ital computing equipment are discussed and
their operation is compared w ith the operation of analogous components in M IN IV A C 601 .

Book III provides an introduction to computer logic and uses

tS M IN IV A C 601 to demonstrate basic rules of lo g ic . The four
HOW fundamental logical operations— A N D , O R , N O T , EITHER BUT
N O T BO TH — are described and these operations are simulated
w ith computer c irc u its .
d e c is io n s
The p rinciples of Boolean Algebra are presented to provide a basis for the
solution of complex logical problems on M IN IV A C 601 . A series of experiments
and problems provides p ractical experience in converting logical problems to
c irc u it diagram s. C irc u it diagrams are programmed on M IN IV A C 601 and logical
problems and riddles are a c tu a lly solved on the computer.
 A d ig ital computer performs ad d itio n , subtraction, m u ltip lica tio n ,
d iv is io n , and other more com plex ca lcu latio n s in an arithm etic u n it.
Some computers use D EC IM A L numbers, but most high-speed d ig ital
machines do arithm etic w ith B IN A R Y numbers. A rithm etic is g en erally
done w ith simple addition or subtraction using e ith e r the "s e ria l" or
" p a ra lle l" method. Com plex ca lcu latio n s are made by doing a series
of additions or subtractions.

Book IV examines the Decimal Number System

and from this derives rules for forming any number
system. These rules are applied to the derivation
of the Binary Number System . The Relationships
and rules for converting between Binary and
Decimal Numbers are discussed.

A fte r the introduction to the number system g en erally used

in high-speed digital m achines, the Binary A rithm etic Operations
are explained and demonstrated w ith appropriate programs for M IN IV A C 601

Book V uses M IN IV A C 601 to demonstrate

how computer functions may be combined to
perform sp e cific jobs. Examples of computer
applications are examined under fiv e major headings:

— "Sim ple Com puter-like D evices Encountered Every Day"

--"Com puter A p p licatio ns in Business and Industry"

— "Computer A p plicatio ns in Science and the M ilita ry "

— "Computer A p p licatio ns in the Social Sciences"

— "Computer A p p licatio ns in Science Fictio n Film s"

Book V also discusses how computers "read " printed m aterial and understand "computer lang uag es."

Book V I contains games w hich may be played using

M IN IV A C 601 as an opponent, partner or re fe re e . These games
are designed for fun w ith M IN IV A C 6 0 1 , but they also
demonstrate important functions of modern high-speed
computers and provide examples of special kinds of problem so lvin g .
 I The following problems are presented to give you an opportunity to
II PROBLEMS I work out computer programs on your MINIVAC 601. When attacking
I a problem, it is often helpful to follow a step-by-step procedure:
3 ) Draw a schematic representation of the components you think will be used.
k ) Draw connections between the components to solve the first part of the problem.
4) Continue with each part of the problem in turn, adding or changing
connections to meet the new requirements.
3 ) From your completed drawing, program the solution on your MINIVAC 601.
PROBLEMS 1 THROUGH 6 USE PUSHBUTTONS, LIGHTS
1 } Design a circuit in which light 1 comes ON when for this problem. (Refer back to the relay experiments
pushbutton 1 is pushed and light 2 comes ON when if you have trouble with this problem.)
pushbutton 2 is pushed, and both come ON at the 10) Design a circuit in which relay 1 comes ON when
same time if both pushbuttons are pushed. pushbutton 1 is pushed if slide switch 1 is left; and
2) Design a circuit in which light 1 comes ON if 1
relay 2 comes ON when pushbutton is pushed if
either pushbutton 1 or pushbutton 2 is pushed, but slide switch 1 is right.
light 1 is OFF if both pushbutton 1 and pushbutton 10 Design an AND (series) circuit in which light 1
2 are pushed together. comes ON if and only if, slide switch 1 is left and
3 .) Design a circuit with 2 pushbuttons and a light in 1
slide switch 2 is right and relay is ON, and relay
which ci]_[ of the conditions indicated in the 2 is OFF. Control relay 1 with pushbutton 1 and
table below occur: relay 2 with pushbutton 2.

Pushbutton 1 1 2 ) Design an OR (parallel) circuit in which light 2

Pushbutton 2 Light '1
comes ON if either slide switch 1 is right, or slide
Up Up On switch 2 is left, or relay 1 is OFF, or relay 2 is
Down Up Off ON. Control relay 1 with pushbutton 1 and relay
Up Down Off 2 with pushbutton 2.
Down Down On
PROBLEMS 13 THROUGH 19 USE THE ROTARY SWITCH
4) Design a circuit with 2 pushbuttons and 2 lights
13) Design a circuit in which the rotary switch turns and
in which al_[ of the conditions indicated in the
light 6 comes ON when pushbutton 6 is pushed.
table below occur:
Pushbutton 1 Pushbutton 2 Light 1 Light 2
14) Design a circuit in which the rotary switch turns
Only when pushbutton 6 is down. The rotary switch
Up Up Off Off should NOT turn when pushbutton 6 is up, or when
Down Up Off On both pushbutton 6 and pushbutton 5 are down.
Up Down On Off
Down Down Off On
15) Design a circuit to meet the requirements of
problem 14 using terminal 5X.
5 ) Design an OR (parallel) circuit in which light 1
comes ON if either pushbutton 1 is down, or if
)
16 Design a circuit in which the rotary switch turns
when pushbutton 6 is pushed but turns in the opposite
slide switch 1 is in the left position, or if
direction when pushbutton 5 is pushed. The rotary
pushbutton 2 is down.
switch should not turn if qo pushbuttons are down,
&~) Design an AND (series) circuit in which light 1 or if both pushbuttons are down.
comes ON if and only if, pushbutton 1 is down and
1 7 ; Design a circuit to automatically turn the rotary
slide switch 1 is left and pushbutton 2 is down and
switch back and forth between numbers 1 and 6.
slide switch 2 is right.
Use only 11 programming wires.
PROBLEMS 7 THROUGH 12 USE RELAYS 1 8 ) Design a circuit to turn the rotary switch when
7) Design a circuit in which light 1 comes ON pushbutton 6 is down, and to automatically stop the
when pushbutton 1 is pushed, and relay 1 comes ON rotary switch when it reaches DO. The rotary switch
when pushbutton 2 is pushed. Relay 1 should come should start again and make another revolution when
ON only if light 1 is ON. pushbutton 5 is pushed and then released.
8 ) Design a circuit in which light 1 and relay 2 are 19) Design a circuit to automatically flash lights 1
ON if slide switch 1 is left, and light 2 and relay 1 through 6 in sequence when pushbutton 6 is pushed.
are ON if slide switch 1 is right. Use separate contacts
to control each component. 2 0 ) (MORE DIFFICULT) Design a circuit to alternately
turn relay 1 ON and OFF as pushbutton 1 is pushed
9 ) Design a circuit in which relay 1 comes ON and momentarily and released. Relay 1 should come ON
remains ON when pushbutton 1 is pushed and released. the first time pushbutton 1 is pushed and it should go
Relay 1 goes OFF and remains OFF when pushbutton 2 is OFF the second time the pushbutton is pushed. The
pushed and released. Terminal 2Z may not be used rotary switch is not used for this problem.
28
kn
 1A IX 12 2A 2X 2Z
ANSWER. 1
6 ( G) (
1+/1Y
1 V IA
2+/2Y
2V2A 781 'l ./ y v-
£8. i <i
1B/l - 2B/2-
Jl l + 2 -J 1
ANSWER 2 1Z 2Z

1 +/1 Y 2Y / 1 A
1Z/2X 1B/l -
1V 2 Z

ANSWER 3
—•
1 +/1 Y 1 V 2Z ZX zz 2X ZZ
ZY
1+
1 V 2X 2Y / 1 A ZY
1B/l - ZB 1-
1A —•

ANSWER 4 -• 2 Z

1+/1Y 1Z/2X
1X/2A 2Y/1A
2B/2- 1B/l - 1-
18

ANSWER 5
1+ /1S 1 V 'R
1S/1 Y 1 V IA
1Y/2Y 1B/l -
2X/1X

ANSWER 6
1+/1Y 2V2T
1V l R 2S/1A
1S/2Y 1-/1B

29
 ANSWER. 7 1A
IX / 1Z
~L
2X 2Z

1-/1B 1X/2X >y.

1A/1X
I+/1Y
2Y/1C
1F/2-
IB
l+< 2y 1C 0- |£
JULSJUU
IF a-
1+

• 2- t l-

ANSW ER 8 1-
2F IF
3

0
1-/1B 1B IE
o2B 2 e <i
1V1R
1+/1S
1S/IV ©2A © 0
1A 2C l6<
1W/1C
1F/1-
lR lT W 1W
2-/2 B /
2A/1T
2-/2F lS
2C/1U IV
1*
1J

ANSW ER 9
1+/1G
1G/1X
1Y/1H
1H/1C
1F/1-
1F/2Y
1E/2X
2Z (not used)

1C IE IF
, toeoc ^ 1-
ANSW ER 10 0 «

1 + /1 Y
l/ lS
1R/l C
1R /
"IS
IT
2C
0
1T/2C
1F/1- / 1*2-
IX
2F/2- ,iy
1

ANSW ER U
1+/1H 1G/1R
1 H/l Y 1S/2S
1Y/2Y 2T/2J
1X/1C 2H/1A
IF / 1 - 1 B /2 -
2X/2C
2F/2-

30
ANSW ER 12 2A
IT 2 P. / lT US /~ 1J 2.G 2J
2-/2B
2A/1T
2+/2Y
2Y/1 Y (3) is
2S 1H 2H
1T/2R 1V i c 2B 12
2R/1J 1F/l —
1J/2G 2V2C iy
1+/1S 2F/2- QQ-*JUUUU>-
1S/2S 2-1 IX
1C jE IF
1-
2S/1 H .22
1H/2H
2+ • — 2 iV
2X a
2C
JtM i
2B 2F 2 -

ANSWER 13
6Y
6+/6Y
6 )/6 A 64-
6B
6B/6-
6V D 17
Dll
p is
D18/M- M-
6m
P19

ANSWER 14 5X 5Z 6X
6y
5Y
5+/5Y P17
5Z/6X D18
6Y/D17 5+ M -
M-/D18

D19
•----

a n sw er . 15 6*
6+/6X
6Y/D17
M-/D18
D18/5X
5Y/D19

ANSWER 16
5+/5Y
5Z/6X
6Y/6J
62/5 X
6Z/6C
6F/6-
6J/ 6K
6-/6G
6G/6N
6L/D18
6H/D17
 ANSWER 17
6+/6J 6C/6L
6J/6K 6L/D18
6K/D6 6H/D17
6E/D16 6-/6N
6F/6- 6N/6G
6G/D1

ANSWER 18 6Z P17

6+/6Y 6x 5Z P IS
6X/DU 6Y
M-/D18
5Z/D18 6+ M-
5Y/DO
D16/D19 P lfe P l9
5Y po

M+ PJ.7

ANSWER 19 T 6^ 7 7b X / b ;Z
D1/1A 6+/6Y •------ •-> 1— •-------------------- -• D ie
* ,bY M
D2/2A 6>/D17
D3/3A M-/D18 M-
26 1 %& OgL
D4/4A M+/D16 v l/
D5/5A
6* 0*9
3“ 33 \ 3. A P3
D6/6A
1B/l —
2B/2- ( T ) I 4A P4
3B/3- 5“ 5B,
4B/4- i 5,A V*
O 1
5B/5-
6B/6- 6; 6B, 6A Pt>
a

ANSW ER 2 0
1 -/1 B 1 Y /1 +
1- / 1C 1V 1J
1A/1 E 1J/2H
1E/2G 2J/ 2E
1F/2F 2E/2A
1 H/1 F 2B/2-
1G /1 Y 2-/2C

3z
M IN IV A C 601 A N D THE M IN IV A C M A N U A L

ARE PRO D UCTS O F :

i l k

S C IE N T IF IC D EV ELO P M EN T C O R P O R A T IO N

372 Main Street, W atertow n, Massachusetts

BOOKS...
I I .I I I .I V

SCIENTIFIC
DEVELOPMENT W hat is a Digital Computer?
CORPORATION How Computers M ake Logical Decisions
Watertown How Computers do Arithm etic
Mass.
 MINIVAC 601

BOOKS ll- lll- IV

S C IE N T IF IC D EV E LO P M E N T CO RP.
W A TE R TO W N , MASS.
 T h e M in iva c M a n u a l was prepared and edited by the staff of

Scientific Developm ent Corporation

First Printing— A ugust, 1961

EX LIBRIS ccapitalia.net

C opyright ( c ) 1961 by S cie n tific Developm ent C orporation, 372 M a in Street, W atertow n, Massachusetts
 CONTENTS

BOOK II: WHAT IS A DIGITAL COMPUTER?

Preface 1
1. IN TR O D U C TIO N 1
2. BASIC COMPUTER FUNCTIO NS A N D M IN IV A C 601 5
The Basic In p u t Function 5
The Basic Storage Function 7
The Basic Processing Function 12
The Basic O u tp u t Function 15
3. EXPANSION OF THE BASIC FUNCTIO NS 16
In p u t M edia and Codes 16
Storage M edia and Codes 21
Processing Techniques 24
O u tp u t M edia and Codes 25
4. C O M M E R C IA L COMPUTER EQUIPM ENT 27
In p u t U nits 27
Storage U nits 30
O u tp u t U nits 29
5. COMPUTERS OF TO M O RRO W 30
APPEN D IX: D ig ita l and A nalogue Com puters 32

BOOK III: HOW COMPUTERS MAKE LOGICAL DECISIONS

Preface 35
1. BASIC OPERATIONS 35
The O peration " A N D " 35
The O peration "O R " 38
The O peration " N O T " 39
The O peration "EITH ER BUT N O T B O T H " 41
2. RELAYS FOR MORE FLEXIBLE T H IN K IN G 42
The Relay A N D c irc u it 43
T he Relay OR c irc u it 43
The Relay N O T c irc u it 44
The Relay EITHER BUT N O T BOTH c irc u it 44
3. T H IN K IN G A N D D E C IS IO N -M A K IN G 44
Boolean A lge bra 44
D ecision-M aking w ith In s u ffic ie n t In fo rm a tio n 47
S im ulation 48
Thoughts A b o u t T h in k in g 49
4. SOME COMPUTER PROBLEMS 50
A M in d Reading Program 50
Q u a n tity Recognition 54
A Problem Involving Three G irls 57
The Farmer, the Goose, the Corn and the W o lf 60
The Television Problem 63

i
 BOOK IV: HOW COMPUTERS DO ARITHMETIC
1. THE B IN A R Y NUM BER SYSTEM 67
How Num bers A re Represented in the B inary System 67
B uilding a Single Inp ut Flip-Flop w ith C arry 68
Experim ent 1: A T h ree-B it B inary Counter 71
Experim ent 2: C ounter A rith m e tic 73
Experim ent 3: Universal Counter A rith m e tic 73
2. B IN A R Y A D D IT IO N 74
Rules fo r B inary A d d itio n 74
Experim ent 4: A H a lf-A d d e r w ith C arry 75
Experim ent 5: A Full A dder 76
Experim ent 6: A T h ree-B it A dder 77
3. H O W COMPUTERS SUBTRACT 78
Tw o's C om plem ent A rith m e tic 78
Experim ent 7: A T hree-B it Subtractor 78
4. COMPUTER M U L T IP L IC A T IO N 79
B inary M u ltip lic a tio n 79
Experim ent 8: The S h iftin g O peration 80
M u ltip lic a tio n by Num bers O ther T han Powers o f Two 81
Experim ent 9: The A ccu m u la to r 81
5. D IV IS IO N ON A COMPUTER 82
B inary D ivision 82
Experim ent 10: D ivision 83
6. CONVERSIONS 84
Experim ent 11: Decim al to Binary Converter 84
Experim ent 12: Binary to Decimal Converter 85
APPEN D IX: A u to m a tic S h ift Register 86
T w o-B it A dder w ith A u to m a tic Decim al Conversion 87
 BOOK II

What is a Digital Computer?

PREFACE
This is the second in a series of books using M IN IV A C 601 to explore the w orld o f "e le c ­
tro n ic b ra in s". In w ritin g th is book, the authors have assumed th a t the reader is fa m ilia r w ith
the in fo rm a tio n contained in the firs t book o f th is series and understands the operation o f the
components o f M IN IV A C 601.
The basic question which th is book was w ritte n to answer is " W h a t is a d ig ita l com puter?"
In order to answer th is question it is necessary to exam ine the functions and form s o f modern
high-speed d ig ita l com puter systems. This book describes the m ajor characteristics o f modern
com puter systems and compares the functions perform ed by the components o f the M IN IV A C
601 w ith those perform ed by s im ila r parts in a large scale d ig ita l com puter.

I. IN TR O D U C TIO N
In order to fu n c tio n as a d ig ita l com puter a m achine or com bination o f machines m ust be
able to handle in fo rm a tio n or " d a ta " in an orderly manner. It m ust be able to receive in fo rm a ­
tio n as " in p u t " fro m the outside world. Once received, th is in fo rm a tio n m ust be "processed" by
the com puter, and the result o f the processing must be "rem em bered" or "s to re d " fo r fu tu re
use. A fte r an answer has been obtained, it m ust be com m unicated back to the outside world as
" o u tp u t." Thus a general purpose d ig ita l com puter is made up o f fo u r basic units:

The in p u t u n it
The processing u n it
The storage u n it
The o u tp u t unit.

The basic units are connected together like this:

FLOW C H A R T OF COMPUTER OPERATION

Input Unit
In fo rm a tio n about a p a rtic u la r problem m ust be given to the com puter before any opera­
tio n can be perform ed. Inpu t in fo rm a tio n m ay be o f two kinds.
 (1) Data: The numbers or coded in fo rm a tio n to be used in calculation are called input data.
These numbers m ay represent physical measurements, m athem atical relationships, or
conditions o f a " lo g ic a l" decision-m aking problem.
(2) Instructions: The com puter must be instructed to perform specific operations in a d e f­
in ite sequence. Inp ut in fo rm a tio n which directs the com puter to perform certain oper­
ations and to handle the data in a specified way is called the instructions.
In p u t in fo rm a tio n is supplied to M IN IV A C 601 through the binary input pushbuttons and
the decim al in p u t-o u tp u t rotary switch. Instructions are com m unicated to M IN IV A C 601 by w ir­
ing on the com puter console a "p ro g ra m " w hich instructs the com puter to perform certain oper­
ations. The b in a ry in p u t pushbuttons are designed to com m unicate zeros and ones to the com­
p u te r and the decim al in p u t-o u tp u t rotary switch is designed to com m unicate decim al numbers
to the com puter.
A large scale d ig ita l com puter, such as the IBM 7090 illustrated later in this section, may
receive in p u t in fo rm a tio n d ire c tly through pushbuttons s im ila r to those used on M IN IV A C 601
or it may receive in p u t in fo rm a tio n through punched cards or m agnetic tape. Some com put­
ers receive inp ut in fo rm a tio n through punched paper tape; others receive direct in p u t in fo rm a tio n
through a typ e w rite r-like u n it called a "fle x o w rite r".

Later in this book, each type o f input u n it found on a large electronic data processing m a­
chine w ill be discussed and compared w ith the input devices o f M IN IV A C 601.

Processing Unit
The processing u n it o f a d ig ita l com puter perform s fo u r m ajor functions:
(1) C ontrol: the processing u n it controls the operations o f the com puter system and in te r­
connects the input, o u tp u t and storage units. A ll calculations, operations, and in fo rm a tio n trans­
fe r are accom plished under " c o n tro l" o f the processing unit.
(2) D ecision-M aking: the processing u n it is able to perform comparisons which are the basis
o f a ll com puter decision-m aking. By com paring two numbers or symbols w ith each other and de­
te rm in in g w hether or not they are equal, the com puter decides upon a course o f action.
(3) A rith m e tic : in the processing u n it o f a d ig ita l com puter a ll norm al a rith m e tic fu n c ­
tions are perform ed. These operations are actu a lly done in a p a rt o f a processing u n it known as
the "a rith m e tic u n it" which is designed to perform addition, subtraction, m u ltip lic a tio n and
division.
(4) Logic: the processing u n it o f most high-speed d ig ita l com puters is equipped to per­
form various " lo g ic a l" operations through which conclusions o f a non-arithm etic type m ay be
reached. Just as a rith m e tic operations provide the steps by which the solution to a m athem atical
problem is reached, the logical operations provide the steps in a "re a s o n in g " process.
The th ird book in this series, How Computers Make Logical Decisions, describes the nature of
logical operations and decision-m aking functions through dem onstration on M IN IV A C 601. The
nature o f a rith m e tic operations perform ed in the processing u n it o f large com puters is demon­
strated in d e tail in the Book IV: How Computers Do Arithmetic.
The processing u n it o f M IN IV A C 601 is made up o f six relays and the rotary switch. The
relays and the rotary switch are used to provide control, make decisions, and perform basic a rith ­
m etic and logical operations.
The processing units o f most large scale d ig ita l computers use advanced electronic com­
ponents to perform the functions dem onstrated by the relays and rotary switch on the M IN IV A C
601. The circu its o f the processing u n it in these machines use germ anium or silicon transistors
and diodes w hich are designed to perform m illio n s o f operations in one second.
In this book the im p o rta n t features and operating characteristics o f the processing u n it o f a
modern electronic data processing m achine are described. The processing u n it o f the M IN IV A C
601 is exam ined as a basic illu s tra tio n o f the functions o f the processing u n it o f a modern d ig ita l
com puter.

2
The Storage Unit
In order to fu n c tio n e ffic ie n tly , a d ig ita l com puter m ust be able to " s to re " or "re m e m b e r"
data fo r use in processing and com putation. Input in fo rm a tio n is " re a d " into storage under con­
tro l o f the processing u n it and called from storage as it is required fo r use by the processing unit.
In the course o f a norm al high-speed d ig ita l com puter program , the processing u n it o f the
com puter follow s a series o f instructions which are stored in the storage u n it using data which is also
stored in the storage u n it. In fo rm a tio n obtained d uring calculations perform ed is stored tem pora­
rily in the storage u n it fo r use a t a later tim e.
A com puter uses a storage u n it in m uch the same way th a t we use a piece o f paper when solv­
ing a long division problem . The p a rtia l answers to the long division problem are te m p o ra rily
stored (w ritten) on the paper u n til the com plete answer is obtained. S im ilarly, the p a rtia l answers
to the com puter's problem are held in storage u n til the com plete answer has been obtained. The
answer which is to be com m unicated by the com puter to the outside w orld as " o u tp u t" m ay also
be stored in the storage u n it u n til it is sent by the processing u n it fro m storage to the o utput
device.
Several d iffe re n t methods o f storage are used in modern com puter systems to enable the
m achine to "re m e m b e r" instructions, data, p a rtia l and fin a l results o f calculations, and o u tp u t in ­
fo rm a tion . In sm aller machines and the M IN I V A C 601, the processing u n it is also used fo r stor­
age. The relays o f M IN IV A C 601 supply the m ajor source o f operating storage. These "m em ory
u n its " o f the M IN IV A C 601, although sm aller than those found in com m ercial computers, demon­
strate the way in w hich in fo rm a tio n is stored in a d ig ita l computer.
The decim al in p u t-o u tp u t u n it also serves as a storage device d u rin g the operation o f some
programs. The rotary switch remains in a p a rtic u la r position u n til it is moved and thus enables
M IN IV A C 601 to "re m e m b e r" a decim al number.
Large scale d ig ita l com puters m ay use relays fo r storage and, in th is case, be identical to the
M IN IV A C 601. A m ong the most popular storage devices in use today are the coincident-current
m agnetic core, m agnetic drum , and m agnetic disc storage units. These storage systems are all
used fo r "o p e ra tin g sto ra ge "— tem porary data storage within the com puter w hile a program is be­
ing " r u n " . O ther storage media are used to store data and instructions fo r longer periods of tim e
and to save data and instructions outside the computer. These "p e rm a n e n t" storage media include
m agnetic tape, paper tape and punched cards. Each o f these storage methods w ill be discussed
later in th is book and the relays and rotary switch o f the M IN IV A C 601 w ill be used to illu stra te
the theory and operation o f each type o f modern storage device.

Output Unit
O u tp u t in fo rm a tio n generated by a d ig ita l com puter is com m unicated to the outside w orld to
present the answer to a problem or to describe the operations o f the com puter. O utput media used
in modern d ig ita l com puter systems include m agnetic and paper tape, and punched cards. Some
a u x ilia ry ou tp u t units convert num erical o u tp u t obtained from the com puter to charts, graphs,
photographic displays, and numbers or letters on a printed page. The great diversity o f o u tp u t de­
vices w hich have been developed fo r use w ith the modern high-speed d ig ita l com puter makes it
possible fo r in fo rm a tio n to be presented by the com puter in alm ost any fo rm in w hich it is desired.
The o u tp u t devices o f the M IN IV A C 601 are the binary ou tp u t lights and the decim al input-
o u tp u t rotary switch. W hen operated w ith the m otor, the rotary switch becomes an ou tp u t device,
since the m otor m ay be controlled by the processing u n it o f the M IN IV A C 601 and caused to stop
w ith the pointer ind ica tin g the num ber w hich is to be com m unicated as o u tp u t in fo rm a tio n . W hen
the b inary o u tp u t lights are used, o u tp u t in fo rm a tio n is com m unicated in a special b ina ry code
discussed in d e ta il in Book IV. M odern electronic data processing machines also have binary o u t­
p u t lights fo r d ire c t com m unication o f in fo rm a tio n from the com puter using the same binary code
employed by M l N I V A C 601.
Large scale d ig ita l com puters em ploy o u tp u t units using media previously discussed in con­
ju n ctio n w ith in p u t units. M edia employed fo r both in p u t and o u tp u t purposes include m agnetic
tape, paper tape, punched cards and data transm ission links.
V arious o u tp u t units are discussed in d e tail later in this book and the ou tp u t devices o f the
M IN IV A C 601 are used to illu s tra te the nature o f each type o f o u tp u t unit.

3
 I N P U T -O U T P U T STORAGE IN P U T -O U T P U T
AND AND AND
SECONDARY STORAGE PROCESSING SECONDARY STORAGE

OUTPUT

IN P U T IN P U T
AND
OUTPUT

IBM 7090 D A T A PROCESSING SYSTEM

STORAGE
O U T PU T AND
PROCESSING
SECONDARY
STORAGE

IN P U T
IN P U T -O U T P U T

M IN IV A C 601

4
 2. BASIC COMPUTER FUNCTIONS A N D M IN IV A C 601

The Basic Input Function

Human Input Functions

The in p ut fu n c tio n can be considered in term s o f the a ctivities o f a hum an being who, having
no previous tra in in g in m athem atics, is asked to perform a division problem . Consider this person
s ittin g at a table w ith pencil and paper in hand. Input information is a ll in fo rm a tio n which must be
com m unicated to the person before he can perform the division problem.
First, he m ust be given two numbers— the data. He m ust be given the num ber which is to be
divided (the dividend) and the num ber by which the dividend is to be divided (the divisor). In ad d i­
tio n to th is data input the person, since he knows nothing about a rith m e tic , m ust be to ld how to
proceed w ith the numbers which he has received as data in order to obtain the answer. In short, the
person m ust be given instructions on how to proceed to solve the problem. The instructions m ust be
very detailed. They must te ll him how to handle each num ber and how to proceed through each
step o f the process o f division.
A basic com puter which has not been equipped w ith any operating circu its is in much the
same position as the man who has never heard o f arith m e tic. Under such circum stances the com ­
puter, like the man, m ust be given very detailed in fo rm a tio n about how to proceed through the
problem . Fortunately, a d ig ita l com puter can be equipped w ith circuits which give it a s u ffic ie n t
"k n o w le d g e " o f the rules o f a rith m e tic so th a t given the divisor and the dividend, it can be told
sim ply " d iv id e " and it w ill proceed to obtain an answer.

Forms of Input

A s has already been noted, in put in fo rm a tio n m ay be o f two kinds: instructions and data.
Both kinds o f in fo rm a tio n may be com m unicated to the com puter in d iffe re n t ways. M ost large
com puters are equipped to receive in p u t in fo rm a tio n from several d iffe re n t media. For example,
the IBM 7090 com puter system is able to receive in fo rm a tio n from punched cards and from m ag­
netic tape.
In fo rm a tio n presented in each o f the media m ay be com m unicated in several d iffe re n t codes.
D ata may be presented as decim al in fo rm a tio n using the characters 0, 1 ,2 , 3, 4, 5, 6, 7, 8, and 9.
In fo rm a tio n in th is form is com m unicated to M IN IV A C 601 w ith the rotary switch.
Both data and instructions m ay be com m unicated to the com puter using a " b in a r y " code
based on the two-valued (zero and one) code referred to in Book I. The bin a ry num ber system is
discussed in detail in Book IV. For the purposes o f discussion in this book we w ill only need remem­
ber th a t bin ary codes involve only two characters, zero and one, w hile decim al codes use the ten
characters noted above.

M IN IV A C 601 Input Units

In p u t in fo rm a tio n is com m unicated to M IN IV A C 601 through the two input devices located
on the console.
Binary in p u t is com m unicated using the six binary input pushbuttons according to the con­
vention:

B utton up = zero (0)

Button down = one (1)

For exam ple, the bin ary num ber " 1 0 1 " (5) is com m unicated to M IN IV A C by pushing down push­
buttons 4 and 6 w hile leaving pushbuttons 1, 2, 3, and 5 up. The largest binary num ber w hich can
be com m unicated to M IN IV A C 601 is 111111 (63).
Decim al in fo rm a tio n is com m unicated using the decim al input d ial and the pointer o f the
rotary switch. To give M IN IV A C a decim al num ber as input, the pointer is turned so th a t it points
a t the desired d ig it. The decim al input capabilitie s o f the M IN IV A C 601 are lim ite d to the numbers
fro m zero through fifte e n .

5
 The fo llo w in g two experim ents illu s tra te com m unication to M IN IV A C in Binary and Deci­
m al. Each experim ent uses the relays to store (remember) the in fo rm a tio n com m unicated to the
com puter.

EXPERIMENT t | B IN A R Y IN PU T
This experim ent uses M IN IV A C 601 to dem onstrate how a d ig ita l com puter receives binary
in p u t data and instructions. The bin a ry num ber " o n e " is given to the com puter by pushing a push­
button down. The bin ary num ber " z e ro " is given to the com puter by leaving a pushbutton up. The
instruction to fo rg e t a ll previous data (to " c le a r " the memory) is given to the com puter by pushing
pushbutton 6.
The program and c irc u it draw ing fo r this experim ent are:

1- 2- 3- 4- 5-

1 X /2 X 6 Z /6 + 2 H /1 H 2 G /2C 4 Y /4 G
2 X /3 X 6 Y /5 H 1Y /1G 2 F /2 — 4 G /4C
3 X /4 X 5 H /4 H 1G /1 C 3 Y /3 G 4 F /4 —
4 X /5 X 4 H /3 H 1F / 1— 3 G /3C 5 Y /5 G
5 X /6 Y 3 H /2 H 2 Y /2 G 3F /3 — 5G /5C

1. T u rn power ON. Push pushbutton l and release. This transm its a "o n e " to section 1 o f M IN I­
VAC . The "o n e " is remembered by relay 1. Relay lig h t 1 comes ON to indicate th a t a "o n e " is
being remembered (stored). Data has been communicated from the operator to M IN IV A C 601
by pushing B IN A R Y IN P U T pushbutton 1.
2. Do N O T push pushbutton 2. T his leaves a " z e ro " in section 2 o f M IN IV A C 601. The "z e ro "
continues to be remembered by relay 2. Relay lig h t 2 remains OFF to indicate th a t a " z e ro " is
being remembered (stored).
3. Do N O T push pushbutton 3. T his leaves a " z e ro " in section 3. Relay lig h t 3 remains OFF to
indicate th a t a " z e ro " is being remembered.
4. Push pushbutton 4 to tra n s m it a " o n e " to section 4. Relay lig h t 4 comes ON to indicate th a t
a "o n e " is being remembered by section 4.
5. Push pushbutton 5 to tra n s m it a " o n e " to section 5. Relay lig h t 5 comes ON to indicate th a t
a "o n e " is being remembered by section 5.
The bina ry num ber 10011 (19) has been com m unicated to M IN IV A C 601 by using the B IN A R Y
IN P U T pushbuttons. This in p u t data is now being remembered (stored) in the firs t five sections o f
the com puter.
6. Push pushbutton 6 and release. T his action instructs the com puter to fo rg e t all previous data.
A ll relay lights go OFF, and the previous num ber is forgotten ("c le a re d " fro m the memory).

6
7. The com puter is now ready to receive another binary num ber from the operator. M ake up an­
other num ber yourself and com m unicate it to the com puter by using the B IN A R Y IN PU T
pushbuttons.

EXPERIMENT 2: D E C IM A L IN PU T
T his experim ent demonstrates how M IN IV A C 601 may receive decim al in p u t data and in ­
structions. A decim al num ber is com m unicated to the com puter by tu rn in g the D E C IM A L IN PU T-
O U TPU T knob to the desired num ber and pushing pushbutton 5 to in struct the com puter to re­
member the selected num ber. A n other in struction— to forget a ll previous data— is given to the
com puter by pushing pushbutton 6.
The program and c irc u it diagram fo r this experim ent are:

1- 2- 3- 4- 5-

1 H /2 H 6 Z /6 + IF /1 — 3G /3C D 5/5G
2 H /3 H 6 Y /5 X D 2 /2 G 3 F /3 - 5G /5C
3 H /4 H 5 Y /D 1 6 2 G /2C D 4/4G 5 F /5 —
4 H /5 H D 1/1G 2F /2 — 4 G /4C
5H /6 Y 1 G /1 C D 3/3G 4 F /4 —

1. T u rn power ON. T urn the D E C IM A L IN PU T-O U TPU T knob to num ber 4 and push pushbutton
5. T his tra nsm its the num ber " f o u r " to the com puter. Relay lig h t 4 comes ON to indicate th a t
a " f o u r " is being stored. Data has been com m unicated from the operator to M IN IV A C 601 by
tu rn in g the D E C IM A L IN PU T-O U TPU T knob to the desired num ber and by in stru ctin g the
com puter to remember the selected num ber by pushing pushbutton 5.
W hen decimal data is remembered by the com puter, the relay lights have a d iffe re n t meaning
tha n when binary data is being remembered. W ith binary data, the lig h t ON represents a data
" o n e " and the lig h t OFF represents a data "z e ro ". W ith decim al data, the num ber being remem­
bered corresponds w ith the section (1-6) which has a relay lig h t ON.
2. Push pushbutton 6 and release. T his action instructs the com puter to fo rg e t all previous data,
and relay lig h t 4 goes OFF. The com puter is now ready to receive another decim al num ber
(1-5) fro m the operator. Select another num ber yourself (between 1 and 5) and com m unicate
it to the com puter using the D E C IM A L INPU T-O UTPU T.

The Basic Storage Function

Three types o f storage are used in most high-speed data processing machines. These are:
Interna l storage
Secondary storage
External storage.

7
Internal storage, w hich is fo r our purposes the most im p o rta n t, is storage available in a storage
u n it connected d ire c tly to the processing u n it o f the com puter in such a way th a t the processing
u n it has "im m e d ia te access" to the in fo rm a tio n . Secondary storage refers to storage units in
which the in fo rm a tio n is available to the processing u n it but in which, due to the nature o f the
storage u n it, it is available only a fte r some delay. The difference between internal and secondary
storage in thus the diffe re nce between im m ediate and delayed access to inform ation.
External storage refers to storage in a media outside the com puter system. External storage
is thus accomplished by com m unicating the in fo rm a tio n to be stored out o f the com puter as " o u t­
p u t" and then saving this in fo rm a tio n in "e x te rn a l storage" on the o u tp u t media.

Human Storage Functions

The storage fu n c tio n can be cle arly seen in the analogy o f the hum an being w ith pencil and
paper. In this case, the paper is the external storage system, and the pencil is the output device
through w hich in fo rm a tio n is com m unicated from the hum an being to the external storage me­
dia. Internal and secondary storage are d iffic u lt to distinguish in the hum an case since we as
hum an beings have only one storage system available fo r our use— the hum an brain. In form ation
retained in the brain w hile the problem is being worked is probably best th o u g h t o f as held in in ­
ternal storage.

M IN IV A C 601 Storage
M IN IV A C 601 is equipped w ith internal storage in the form o f the six relays which also
serve as a p a rt o f the processing u n it o f the com puter. The processing u n it o f M IN IV A C does not
have direct access to external storage. The hum an operator can supply external storage by w rit­
ing in fo rm a tio n down on a sheet o f paper when it is com m unicated to him through the ou tp u t de­
vices o f M IN IV A C and later return th is in fo rm a tio n to the com puter through the in p u t devices as
it is required.
Before exam ining the storage units o f M IN IV A C 601 and larger d ig ita l computers in more
d e ta il, it is necessary to consider the general form in which in fo rm a tio n is stored in a d ig ita l
com puter.

The Binary Nature of Storage

Just as in fo rm a tio n is com m unicated to the com puter and processed by it in a binary form ,
it m ust be stored in binary form .
The reason fo r th is use o f the sim ple tw o-character (0 and 1) rather than the more com ­
plicated ten-character (0-9) decim al system in the com puter can be realized by exam ining the
decim al in p u t device. Using the rotary switch mechanism involves m any d iffe re n t physical posi­
tions w hich m ust be com m unicated by the sw itching system to the com puter. The mechanism used
to do th is involves physical m otion and several moving parts. In addition, the rotary switch sys­
tem requires more th an a second to go from zero to nine.
In contrast to the decim al system, the bin a ry in p u t button involves only one m oving p a rt and,
since it has only two physical positions, is sim pler in construction and can be moved from one po­
sition to another in a fra c tio n o f a second.
Thus, fo r the reasons o f sim p licity, e fficie n cy and speed a p tly dem onstrated by the com ­
parison between the decim al inp ut switch and the binary in p u t button, large com puters handle all
processing and storage using a binary system. Thus the basic elem ent o f storage in a com puter
is a bin ary " b i t " w hich m ay have the value o f 0 or 1.

The Structure of Storage

In any storage system, the bits are arranged into words consisting o f a specified num ber o f
bits. Storage w ith in the com puter is handled in words. Each storage location— called a "re g is te r"
contains one word o f data and has a location num ber associated w ith it so th a t it is possible to
id e n tify a register and store in fo rm a tio n in or c a ll in fo rm a tio n from th a t register.
The storage system o f a large com puter m ig h t be th o u g h t o f as a large num ber o f boxes w ith

8
each box id e n tifie d by a num ber in d ica tin g its location in the series o f boxes, w ith each box just
the rig h t size to hold a specified num ber o f bits o f data. The actual form o f these storage registers
varies w ith the machine.

Secondary Storage: Slide Switches

The six slide switches o f M IN IV A C 601 are used to supply secondary storage. These switches
are operated m a n u a lly and are used to store b in a ry in fo rm a tio n according to the convention:
switch rig h t = 0
switch le ft = 1
The follo w ing experim ent illustrates secondary storage in M IN IV A C .

EXPERIMENT 3: SECONDARY STORAGE

This experim ent uses M IN IV A C 601 to dem onstrate how a d ig ita l com puter remembers in ­
fo rm a tio n which is not required im m ediately by storing the in fo rm a tio n in SECONDARY STOR­
AGE. The slide switches are used as secondary storage fo r M IN IV A C 601. The operator transfers
data to the SECONDARY STORAGE by m anually operating the slide switches.

1S 1T

1. T u rn power ON. M ove slide switch to the left. L ig h t l comes ON to indicate th a t a "o n e "
(binary data) is being stored in SECONDARY STORAGE, Section 1. The SECONDARY STOR­
AGE slide switch continues to remember "o n e " u n til it is instructed to remember a " z e ro " by
the operator.
2. M ove slide switch 1 to the right. L ig h t 1 goes OFF to indicate th a t a "z e ro " (binary data) is
now being stored in SECONDARY STORAGE, Section 1.
M ost large d ig ita l com puters a u to m a tica lly transfer data a t very high speeds to secondary stor­
age. M odern com puters m ay use punched cards, punched paper tape, m agnetic tape, and other
special devices fo r secondary storage.
Secondary and external storage are required by d ig ita l computers because there is not enough ca­
pacity in the storage-processing u n it to store a ll the in fo rm a tio n required by some problems. Ex­
cess in fo rm a tio n which is not required im m ediately is transferred to secondary or external stor­
age. External storage fo r a d ig ita l com puter is s im ila r to the file cabinet which some people keep
in the basement. Some papers m ay be throw n away, but there is not room enough in the desk up­
stairs to keep a ll items which m ust be file d . " A c tiv e " info rm a tio n , which is used frequently, is
kept close at hand in the desk and "d e a d " in fo rm a tio n , which is used in frequen tly, is kept in a
file cabinet in the basement. Secondary storage in th is case would be in fo rm a tio n required only oc­
casionally, but im p o rta n t enough to be kept close by. This in fo rm a tio n m ig h t be kept in the back
o f a desk drawer, or perhaps in a cabinet several steps from the desk.

Internal Storage— Relays as Memory

Before going on to the more com plicated form s o f storage used in large computers, we w ill
tu rn again to M IN IV A C 601 and exam ine the storage system used to provide m em ory fo r this
sm all com puter. Storage or "m e m o ry " fo r M IN IV A C is supplied by the 6 relays. These relays can
be used to store or "re m e m b e r" six binary bits. As indicated above, a large com puter may have
storage capacity fo r 36 bina ry bits in each o f its storage registers. Thus the storage capacity o f
M IN IV A C 601 w ith its 6 -b it register is one-sixth th a t o f a register in the 3 6 -b it machine.
As noted above, in a large com puter each storage register is id e n tifie d by a location number.

9
In the case o f M IN IV A C 601 it w ill not be necessary to consider this problem o f m em ory location
id e n tific a tio n , since we are dealing w ith only one storage register.
W ith the exception o f the distin ctions noted above, the basic fu n c tio n in g o f the storage
system in M IN IV A C 601 exactly duplicates the fu n c tio n in g o f the storage system in a larger
com puter. Thus, the larger com puter m ay be th o u g h t o f as sim ply an extension o f m any M IN I­
V A C 6 0 1's lined up in a row and interconnected. To d uplicate the storage capacity o f the IBM
7090 com puter system, fo r instance, w ould require 192,000 M IN IV A C 601 Computers in com ­
bination.
In the discussion o f the operation o f the relay in Book I, the relay was used as a switch op­
erated by a pushbutton. It was noted th a t the relay could be used to indicate the two-valued binary
code by considering the relay OFF to be storing or rem em bering a ' 0 ' w hile the relay ON was th ough t
o f as rem em bering '1 '. The relay lig h t could be used to indicate when the relay was storing a M '
(lig h t on) or a '0 ' (lig h t o ff). T his sim ple c irc u it m ay be th o u g h t o f as a m anual memory: as long
as the pushbutton is being held D O W N , the relay remembers a one. W hen the pushbutton is re­
leased, the relay forgets the one and starts rem em bering a zero. The relay in th is c irc u it is a
memory element controlled by the pushbutton.

1 + /1 Y
1 + /1 H
1 X /1 C
1— / I F
1 G /1A
1 B/T —

M A N U A L RELAY M EM O RY C IR C U IT

A more e ffic ie n t m em ory u n it would be achieved if it were possible to m ake the relay remain
in the D O W N or 1 position once it was signaled to go to th is position by the pushbutton. Such a
c irc u it would enable the relay to remember a 1 once it had been signaled by pushing the pushbut­
ton to " fo rg e t 0 and s ta rt rem em bering 1".
The m anual m em ory relay c irc u it above can be easily m odified to achieve this by using one
o f the switches o f the relay to continue to supply current to the coil o f the relay a fte r the pushbutton
has been released. M a k in g use o f this relay switch, the pushbutton w ill in itia lly supply current to
the relay. Once c urre nt is supplied to the relay coil, the relay w ill close and a second path fo r the
c u rre n t w ill be supplied through the norm ally open switch o f the relay. W ith this c irc u it wired, the
relay becomes "s e lf-lo c k in g " in the 1 position.

7- f - / l V
74-/1H
1 X /1 C
1— / I F
1G /1A
IB /1 —
1 C /1A

SELF-LOCKING RELAY M EM O R Y C IR C U IT

10
 The evident d iffic u lty w ith th is self-locking relay m em ory c irc u it is th a t, although the relay
w ill now remember a 1 once it is told to remember 1, it cannot forget 1 unless the current to the
com puter is turned o ff. The next step in bu ild in g a usable m emory u n it is obviously to m o d ify the
c irc u it so th a t the relay can fo rg e t 1 and sta rt rem em bering 0 again.
As an in itia l step in the direction o f program m ing such a c irc u it, a second pushbutton m ay be
used to supply a means o f disconnecting power from the relay c irc u it. Using th is c irc u it, push­
b utton 1 w ill be used as in the self-locking m emory c irc u it to supply cu rre n t to the relay coil and to
cause the relay to stop rem em bering 0 and sta rt rem em bering 1.
Pushbutton 2 w ill be used to cause the relay to stop rem em bering 1 and begin rem embering
0. As indicated below, this program uses the norm ally closed contacts o f pushbutton 2 and the
n o rm a lly open contacts o f pushbutton 1. The self-locking connection used in the previous c irc u it is
retained.

T W O -B U TTO N RELAY M EM O R Y C IR C U IT

The tw o-button or "tw o in p u t" m em ory c irc u it provides a workable m em ory elem ent which
satisfies the condition th a t it be able to remember a 1 or a 0 on signal. The weakness in this c irc u it
is th a t a d iffe re n t signal is required to te ll the m em ory c irc u it to forget 0 and to begin rem ember­
ing 1 than is used to te ll it to fo rg et rem em bering 1 and recommence rem em bering 0. Pushbutton 1
serves as the "fo rg e t 0 remember 1 " signal and pushbutton 2 provides the "fo rg e t 1 remember 0 "
signal.
It w ould be p a rtic u la rly desirable if we could obtain a memory c irc u it which would respond to
a single signal such th a t, if the c irc u it were rem em bering 1, it would forget 1 and s ta rt rem ember­
ing 0. In short, it would be desirable to have a "s in g le -in p u t memory c irc u it" .

The Single Input "Flip-Flop"

Before program m ing M IN IV A C 601 fo r a single-input memory c irc u it a com m ent on
term inology is in order. A n exam ination o f the operation o f the tw o-button relay m emory c irc u it
w ill show th a t as the contents o f the m em ory are changed, the relay goes firs t ON and then OFF,
changing fro m one position to another. The sw itching m otion o f the relay has caused engineers to
refer to th is type o f c irc u it as a " flip - f lo p " and the tw o-button relay m em ory c irc u it above is
known as a "tw o -in p u t flip -flo p " .
So, the c irc u it which we are now seeking is a "s in g le in p u t flip - flo p ." The c irc u it and program
below are fo r a single in p u t flip -flo p . Once this c irc u it has been program m ed on M IN IV A C 601,
pushing one pushbutton w ill signal the com puter to remember a 1 or a 0. If the relays are re­
m em bering a 1, pushing pushbutton 1 w ill signal them to "fo rg e t 1 and s ta rt rem em bering 0 " . If
the relays are rem em bering a 0, pushing pushbutton 1 w ill signal them to "sto p rem em bering 0
and s ta rt rem em bering 1 ". L ig h t 1 w ill be ON when the flip -flo p is rem em bering a 1, and w ill be
OFF when the flip -flo p is rem em bering a 0.
 1—/IB
1— / 1 C
1A/1E
1 E /2 G
1F/2F
1H /1 F
1 G /1Y
1Y /1 +
I X / 1J
1J/2H
2J /2E
2E /2A
2 B /2 —
2— /2C

In Book IV, the single in p u t flip -flo p w ill be used to provide the m em ory necessary to enable
the com puter to count and to perform various m athem atical operations. For the purposes o f this
discussion o f com puter storage, it is necessary only to note th a t this single in p u t flip -flo p has the
characteristics o f the basic elem ent o f a com puter m emory system. It is able to remember a 1 or a
0 and to “ change its m in d " when signaled to do so.

The Basic Processing Function

The detailed operations of the processing u n it are discussed in Books III and IV. In this dis­
cussion, it is s u ffic ie n t to note th a t the processing u n it perform s two basic kinds o f operations.
(1) It controls the flo w o f in fo rm a tio n w ith in the com puter. The processing u n it "d ire cts
t r a f f ic " w ith in the com puter, sending in fo rm a tio n fro m the in p u t u n it to storage, fro m storage to
the various ca lc u la tin g units w ith in the processing u n it, and from processing to o u tp u t or storage.
In one sense, the processing u n it m ig h t be th o u g h t o f as the bookkeeper o f the com puter. It keeps
tra c k o f and handles the placem ent o f a ll data.
(2) It does a ll calculations. The processing u n it is the ca lculator o f the com puter. In the
processing u n it a rith m e tic data is m odified and "d e cisio n s" are made. Instructions to the com puter
d ire ctin g it to perform operations on data are carried out. A ll other units o f the com puter are used
to fa c ilita te transm ission and storage o f in fo rm a tio n . O nly the processing u n it actively m odifies or
uses data. The other units o f the com puter are passive w ith respect to the data.

Human Processing
The fu n c tio n o f the processing u n it can be visualized in term s o f the division problem dis­
cussed earlier. Once the hum an operator has the divisor and dividend w ritte n on paper and has
been given the list o f instructions fo r perform ing the division, his actions are analogous to those
o f a processing unit. He w ill fo llo w the instructions, a ctin g upon his data, u n til he reaches the
fin a l answer.
The processing u n it o f a com puter acts just like the hum an operator ju st m entioned after re­
ceiving both instructions and data. In a sense then, the processing u n it is the w ork center o f the
com puter.

M IN IV A C 601 Processing
Processing in M IN IV A C 601 is accomplished using logical circu its to d uplicate a rith m e tic
and decision functions. Relays are used as sw itching devices and th e ir operation is controlled by
instructions com m unicated to M IN IV A C by means o f the program wired on the com puter console.
D uring the "e x e c u tio n " o f a program , the processing section o f M IN IV A C 601 controls the opera­
tio n o f the com puter. The program w ired on the com puter console indicates to the processing sec­
tio n w hat operation it is to perform and the processing section controls the other units o f the com­

12
puter— o b ta inin g in fo rm a tio n when it is required from in p u t sources and com m unicating the fin a l
answer to the o u tp u t units.
The fo llo w in g experim ents illu s tra te some of the basic functions which a com puter can per­
form , using the processing u n it in various ways.

EXPERIMENT 4: CONTROL
T his experim ent uses M IN IV A C 601 to dem onstrate how the processing u n it controls the
com puter's operations. The rotary switch is controlled by the Processing U nit.

5 + /5 Y
5 X /5 C
5 F /5 -
5 + /5 H
5G /6C
6 F /6—
6 + /6 H
6G /D 17
D 1 8 /M -

T u rn power ON. Push pushbutton 5. T his energizes relay 5 (turns it O N) and causes current
to flow through the switch contacts o f relay 5 to relay 6. Pushbutton 5 controls relay 5. Relay 5 in
tu rn controls relay 6. Relay 6 controls the operation o f the motor.
The m otor is controlled by relay 6 w hich is p a rt o f the Processing U nit. Relay 6 is controlled by
relay 5 o f the Processing U nit. Relay 5 is controlled by Binary In p u t pushbutton 5. The operator
supplies the Input by pushing pushbutton 5. The program w ired on the com puter console gives the
instruction:
If pushbutton 5 is D O W N , tu rn relay 5 ON.
If relay 5 is O N , tu rn relay 6 ON.
If relay 6 is O N , tu rn m otor ON.
In a more com plicated program , the relay could be controlled by other relays or switches as
the result o f some ca lcu la tio n or series o f events.
This a b ility o f a d ig ita l com puter to control its own operations perm its the com puter to auto­
m a tic a lly m ake a whole series o f calculations a t high speed w ith o u t d ire ct operator command.

EXPERIMENT 5: DECISION M A K IN G
T his experim ent uses M IN IV A C 601 to dem onstrate how the Processing U n it o f a com puter
can m ake decisions based on rules given by the program m er. Two b inary num bers are compared
and the com puter decides w hether or not the numbers are equal.

1 + /1 H
1H /1 L
1L/1X
1 Y /1 C
IF /1 —
1 X /2 X
2 Y /2 C
2 F /2 —
1J/2H
2 J /2 K
1 K /2 L
2 K /3 C
3 F /3 —

13
1. T ra nsfer the decision rule to the com puter by w irin g the program onto the console. Pushbuttons
1 and 2 w ill represent the numbers to be compared. Relay 3 w ill indicate the com puter's decision.
If the numbers are equal, relay 3 w ill come ON. If the numbers are N O T equal, relay 3 w ill go OFF.
(N ote: a pushbutton UP represents zero; a pushbutton DO W N represents 1.)
2. T u rn power O N. Relay 3 comes ON because both numbers are zero.
3. Push pushbutton 1. Relay 3 goes OFF because the numbers are no longer equal. (The firs t
num ber is now 1; the second num ber is 0. )
4. Push pushbutton 2 w hile holding down pushbutton 1. Once again the numbers are equal (both
are 1) and relay 3 comes ON.
5. Release pushbutton 1 w hile holding down pushbutton 2. Relay 3 goes OFF because the num ­
bers are not equal.
This decision rule can be program m ed w ith o u t using relays 1 and 2, in which case the push­
button contacts perform the processing fu n ctio n . Since the relays have tw ice as m any contacts as
the pushbuttons and can be e le ctrica lly operated, they are considerably more versatile than the
pushbutton contacts. For this reason, relays are used to perform the processing fu n ctio n in a ll but
the sim plest programs.

EXPERIMENT 6: A R IT H M E T IC
This experim ent demonstrates how the Processing U n it o f a d ig ita l com puter m ay be used to
perform a rith m e tic calculations. Two numbers (0 or 1) are added together and the answer is in d i­
cated by the Binary O u tp u t lights.

1 + /1 H
1H /1 L
1L/1Y
1 Y /2 Y
2 Z /2 C
2 F /2 -
1X /1C
IF /1 —
1G /2 G
2 H /2 A
2 B /2 —
1N / 2 K
1 K /2 N
2 L /1 A
IB /1 —

1. T ran sfe r the rules o f a ddition to the com puter by w irin g the program onto the console. Push­
buttons 1 and 2 w ill represent the numbers to be added. L ig h t 1 ON w ill represent an answer o f 1;
lig h t 2 ON w ill represent an answer o f 2; NO lights on w ill represent an answer o f 0.
(N ote: a pushbutton UP represents 0; a pushbutton D O W N represents 1)
2. T u rn power O N. No lights come on because the in p u t numbers are both zero: 0 + 0 = 0
3. Push pushbutton 1. L ig h t 1 comes ON: 1 + 0 = 1
4. Push pushbutton 2 w hile holding pushbutton 1 down. L ig h t 2 comes ON: 1 + 1 = 2
5. Release pushbutton 1 w hile holding pushbutton 2 down. L ig h t 1 comes O N: 0 + 1 = 1.

14
 The Basic Output Function

The ou tp u t fu n c tio n is basically the com m unication of the processing un it's results to the
outside world. T hrough the o u tp u t u n it, in fo rm a tio n about the problem or the operation o f the
com puter is obtained from the com puter.

Human Output Functions

In term s o f the previously discussed hum an analogy, ou tp u t is obtained from the hum an
being when the answer is w ritte n on the paper (output media) by the pencil (output u n it) which is
controlled by the hum an being (processing unit). In the case o f the hum an being, the ou tp u t device
is used to com m unicate both p a rtia l in fo rm a tio n from the hum an processing u n it during com pu­
ta tio n o f the problem and to display fin a l answers upon com pletion.

M IN IV A C 601 Output Units

O u tp u t in fo rm a tio n is obtained fro m M IN IV A C 601 using two o u tp u t devices: the B IN A R Y
O U TPU T lights and the D E C IM A L O U TPU T mechanism. In form ation is " re a d " from the binary
o u tp u t lig h t by in te rp re ting a lig h t w hich is on as com m unicating a " 1 " and a lig h t w hich is o ff
as com m unicating a " 0 " .
O u tp u t in fo rm a tio n is "re a d " from the decim al ou tp u t mechanism o f M IN IV A C 601 by not­
ing the num ber a t w hich the pointer knob is pointing .
A n exam ple o f Binary O u tp u t is given in experim ent 6. The Binary O u tp u t lights were used
to indicate the answer to a sim ple a ddition problem.
The fo llo w in g experim ent illustrates the use o f the rotary switch fo r Decimal O utput.

EXPERIMENT 7: D E C IM AL OUTPUT
T his experim ent demonstrates how a com puter delivers decim al ou tp u t in fo rm a tio n to the
operator. The rotary switch (Decimal O utput) is controlled through the slide switches (Secondary
Storage) to com m unicate to the operator the contents o f Secondary Storage.

6Z

1R / D l 3 S /4 S 6 R /D 6 6 Y /6 +
1S/2S 4 R /D 4 6S /D 19
2 R /D 2 4S/5S D 18/D 16
2 S /3 S 5 R /D 5 D 1 6 /M —
3 R /D 3 5 S /6 S D 1 7 /6 X

1. Set all slide switches in the RIGHT position. The slide switches w ill represent the num ­
bers 1 through 6. T o place a num ber in secondary storage, the appropriate slide switch w ill be

15
moved to the LEFT position. Pushbutton 6 w ill be used to give the com puter the instruction:
Indicate num ber in Secondary Storage using Decimal O u tp u t rotary switch.
2. T u rn power ON. Move slide switch 4 to the LEFT position. This places the num ber 4 in
Secondary Storage.
3. Push pushbutton 6. This instructs the com puter to com m unicate the contents o f Sec­
ondary Storage to the operator. The rotary switch turns to 4 on the Decimal Input-O utp ut dial.
The num ber 4 is the Decim al O utput.
4. M ove slide switch 4 to the RIGHT position and select another num ber to be placed in
Secondary Storage. A g a in , push pushbutton 6 to instruct the com puter to com m unicate the con­
tents o f Secondary Storage through Decim al O utput.

3. EXPANSION OF THE BASIC COMPUTER FUNCTIONS

Input M edia and Codes

L im ite d am ounts o f in fo rm a tio n can be com m unicated d ire ctly to a large com puter through
its console. T his dire ct in pu t in fo rm a tio n may be either binary or decim al in form , depending
upon the operating characteristics o f the m achine, and is norm ally com m unicated through switches
on the m ain console o f the com puter. In some situations direct com m unication is obtained through
the use o f a " fle x o w rite r" which is a typewriter-1 ike device which, in a ddition to p rin tin g charac­
ters on a w ritte n page, transm its a coded representation o f the characters to the com puter fo r
interpretation.

Direct Console Input

In the photograph o f the 7090 console below the input buttons used to provide d ire ct binary
in p u t in fo rm a tio n to the IBM 7090 com puter are visible. The fu n ctio n in g o f these buttons is com ­
parable to the operation o f the bin ary input buttons on the M IN IV A C 601 Console. W hen a par­
tic u la r button is pressed down, it represents a one in th a t position and when the button is in the
norm al or up position, it represents a zero in th a t position.

IBM 7090 CONSOLE

In the photograph o f the IBM 650 C om puter console below, the input switches w hich com ­
m unicate in p u t in fo rm a tio n in decim al rather than bin a ry form are indicated. The operation of

16
these switches is s im ila r to the operation o f the decim al input switch on the M IN IV A C 601 Con­
sole. To com m unicate a decim al num ber to the IBM 650, the appropriate switches are turned so
th a t the desired decim al num ber is indicated by the switches.

IBM 650 CONSOLE

In ad d itio n to d ire ct console in p u t in the bin a ry and decim al fo rm using the same coding
system employed w ith the M IN IV A C 601, large com m ercial com puters have several other form s
o f input. Basic in p u t media in a d dition to d ire ct console in p u t are: punched cards, punched tape,
and m agnetic tape.

Punched Card Input

Punched cards are perhaps the most common m edium fo r com m unication w ith a com puter.
In fo rm a tio n is "re c o rd e d " on the cards by means o f a sm all hole punched in a p a rtic u la r location
on a card in accordance w ith a coding plan w hich the m achine has been program m ed to " u n ­
de rstan d". As the m achine "re a d s " a card, it obtains in fo rm a tio n by sensing the presence or ab­
sence o f holes in each o f a num ber o f locations. The in fo rm a tio n obtained fro m the reading o f the
card is then translated in to electronic in fo rm a tio n fo r processing and storage w ith in the com puter.

17
Alpha-Numeric Card Code
The standard alpha-num eric code is sum m arized in the photograph below. The numbers zero
through nine are coded as a single punch in a vertical colum n. The alphabetic characters and
symbols are represented by two punches in a single vertical colum n.
C om m unication using the standard alpha-num eric in p u t card code is lim ite d to 80 characters
(numbers, letters, or special characters) per card. The use o f " b in a ry " codes g reatly increases the
am ount o f in fo rm a tio n w hich m ay be com m unicated using a single card.
As noted earlier, although the punch its e lf is a binary variable— a variable w hich can have
only two values, either 0 or 1— the use of the bin a ry punch in conjunction w ith the alpha-num eric
code creates a situa tion where on ly one variable value m ay be com m unicated in each vertical
colum n. In a sense, this is s im ila r to the lim ita tio n imposed by the decim al in p u t d ial on the
M IN IV A C 601. A lth o u g h a t any location the pointer is e ither p o in tin g a t the num ber or is not
p oin ting a t th a t num ber (corresponding to the punch e ither being in a p a rtic u la r location or not
being in th a t location) m eaningfu l in fo rm a tio n can only be com m unicated by considering a ll 16
locations in w hich the pointer might be. T his corresponds to considering a ll 12 positions in a given
vertical colum n on a card in w hich a punch might be located.

PRINTED PRINTED
BY THE 2 6 — — 0123456769 ABCDEFGHIJKLMNOPQRSTUVUXYZ IBM DATA PROCESSING tilpS o d - BY THE 26
PRINTING PRINTING
CARD PUNCH II II I II I I CARD PUNCH
M

III I III I
•I 000 O'fl ooooooooooooooo| ooooooq| | oooooo|
« 11 Mtt aaaaaaaaanaaMaaaaaanaaMaanaaaTinniin
11 11 n111ii11111111111111niii 11111 n i i
222 2 22 2 2 222222222222| 222222222222||22222222

3 3 3 3 | jj 33333 33 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 31 3 3 3 3 31 3 3 3 3 3 3 3 3 3 3 3

44444| 4444441) 444 44 44 44 4 44 4 4 4 4 4 4 | 4 | 4 4 4 4 44 44 4 44 4 44 44 44 4

5 5 5 5 5 5| I 5555555 55 5 5 555 S5S5S5SS5S55555555555555|55S|5S55S5

6666666 66666666 98666 6666 66666S6666666666666666|666666666666

77777777 777777777 777777 7 77 77 77177717J7J777177777|J117J777|77777

• 8888888 311 8888888881 8868886 888888 lllllllllllllllllllllllllllllllllll
9 9 9 9 9 9 9 9 9 91 9 9 9 9 9 9 9 9 9 9 91 9 9 9 9 9 9 9 9 9999999 9 I I I I I I II I9 | I9 II II I9 I |I II II |9 9 II II I
>I I 1 I I I I I IR UMiiauunanaa aanaaana'aaaaaaa 4243 44454* 47« « l SI S H H H R D IR R III1 D H B N 17HU 7t 71 7373 74

A LP H A -N U M E R IC CARD SS

Binary Card Codes

As was noted ea rlie r, the b in ary num ber system is discussed in d e tail a t the beginning o f
Book IV and, fo r our purposes a t th is tim e, it is necessary only to rem ember th a t a bin a ry code is
one made up o f only two characters (0 and 1). Since the binary code is made up o f only zeros and
ones it is unnecessary to consider the location o f the punch in a ve rtica l colum n in order to know
the value o f the binary d igit com m unicated by the punch. The presence o f a punch com m unicates
a one and where there is no punch a zero is com m unicated. Since there can be only zeros and ones
in a b in a ry code a ll in fo rm a tio n w hich could be com m unicated about the value o f a p a rticu la r
b in ary d ig it is com m unicated sim ply by the presence or absence o f a hole in a p a rtic u la r location
on the card.
The length o f a b in a ry num ber or "w o rd " (i.e., the num ber o f b in a ry characters in the num ­
ber or word) can vary. For exam ple, the num ber labeled (A) below is a fo u r-d ig it bin a ry num ber or
word and the num ber labeled (B) is a te n -d ig it binary num ber or word.
(A) 1010
(B) 0 000001010
The card below is an in p u t card prepared using a binary code. On th is card bin a ry in fo rm a ­
tio n is represented in tw e n ty-fo u r 3 6 -d ig it words occupying positions in colum ns 1 through 36
and 37 through 72 respectively in 12 rows. T his is a "ro w b in a ry " card.

18
 * ‘+ + | + + + ' + + + H - + + l + + + ] + + + | + + 4 j + + + 4 - + + j + + + [ f | + ( i | | | i + + | + + + M - + - H + + + l + + + j + + + | + + + | + + + J f + + f * - + + i + | ' t j + + 4 | ^ + + + + + + +

I I
I
j_ _ _ I I
_ J ______
j— — r | - | | -
g g f e j+ f l
lo o 0 o o!g o o|o o o!o o oio o o 0 0 0 0 0 0,0 0 o'o 0 010 I 0|0 I I 9 0 0 0 0 0,0 0 0|0 0 0|0 0 o'o 0 0 0 0 0
I I I 1 i i 1
0 0 0|0 0 o'o 0 0 0 1 0| 0 0 DOOOOOOO
it: 2 3 4 5 6 ' 7 3 9*3 0 II 12:13 14 13.16 17 19 19 20 21 22 23 24.23 26 27)26 29 30|31 32 33 34 35 36 32 39 39 40 41 42.43 44 45,44 47 46,49 50 5 lls 2 S3 54 55 54 57 54 59 50,51 42 43l«4 65 84 67 64 6 9 |7 0 71 72 13 74 7 5 7 4 7 7 7 4 7 9 8 0
| l 1 1 1 i'll l [ l 1 1 1 1 1,1 1 1 1 1 1 1 1 1 , 1 1 1 ji 1 i n | i ; i 1 1 | l 1 1 1 1'1 1 1,1 1 1,1 1111 1 1 1 1 1 1 1 i , i i i h 1 1 h 1 1 n 11 11111111
I I 1 I I i i i
* 2 2 2 2 212 2 212 2 2 '2I 2 212 2 2 2 2 2 22 2l2 2 2l2 2 2,2 21 ||| | 2 2 2 2 2 2l2 2 2 2 2 2l2 2 2,2 2 2 2 2 2 2 22,2 22122 2121 2|2 22 2 2 2 2 2 2 2 2
I I 1 . I 1 I I I
§ 3 3 3 3 313 3 3 3 3 3|3 3 3)3 3 3 3 3 3 3 3)3 3 3j3 3 3| 3 3 | j | 3 1 3 3 3 3 3 3i3 3 3>3 3 3|3 3 3 !3 3 3 3 3 3 3 3 3 3 33i3 3 3.3 3JH I 3 3 3 3 3 3 3 3
I ’ 1 , I 1
| « 4 4 4 4 ^ 4 4 I4 4 4t4 4 4>4 4 4 4 4 4 4 4 4 ^ 4 4?4 4 4<4 4 | I 4 1 1 14 4 4 4 4,4 4 414 4 4,4 4 414 4 4 4 4 4 4 4 4 J4 4 4 J4 4 4!4 4 1 ^ 4 4 4 4 4 4 4 4 4 4
1 I I 1 I
i s 5 5 5 5 5 5 5j5 9 5 j5 5 5|5 5 5 5 5 5 5 5 5 J5 5 5 I 5 5 5|5 5 |I5 5 1 5 5 5 5 5 515 5 515 5 515 5 5,5J | J 5 5 5 5 5 5 5 5 J5 5 5*5 5 5|5 5 1 ’5 1 5 5 5 5 5 5 5 5 5
T
I I I I I
1st 6 6 5'6 6 g |e 6 8'o 6 6 6 S 6 66 6 6 6 6<666|S6 66 6 G ||| 9 6 6 8 6 6 6 6 6 1 6 6 6 1 6 6 8'8 6 6 6 6 6 6 6 6'6 6 6'S 6 6'6 6 1'6 6 6 66686666
I I I I I I
1:7 7 7 7 7'7 7 7 7 7 7*7 7 7l7 7 7 7 7 7 7 7 7|7 7 7,7 7 7I7 7 7 | 7 1 7 7 7 7 7 7]7 7 7|7 7 7,7 7 7|7 7 7 7 7 7 7 7 7|7 7 717 7 7'7 7 7 j | | 77777777
I ' l l
88 I I o| u a U|0 88 8 ■8 8 8 8 8 8 8 8 8 8
0 9 9 9 9 919 9 919 9 9 9 9 9'9 9 9 9 9 9 9 9 919 9 919 9 9>9 9 9'9 9 | 9 9 9 9 9 9|9 9 9l9 9 919 9 919 8 9 9 9 9 9 9 919 9 9<9 9 9'9 9 9191 9 9 9 9 9 9 9 9 9
s 4 9 4 | 7 I I p * II 1 2 1131)4 15 |1 l i t 11 13 74 79 74 77 74 79 40
22 23 24(25 26 27 |2 8 29 3 0 p i 32 3 3 |M 39 36 p ? 38 39[40 41 42 |4 3 44 4 5 |4 6 47 4 6 |4 6 90 91)92 S3 5* 55 96 97|5 6 96 6 0 |6 t 62 63 |6 4 65 88|6 7 16 6 6 |7 0 71 72jl-------------------------------

ROW B IN A R Y CARD

In place o f punching b in a ry in fo rm a tio n in h o rizontal rows as in the row binary card code, in ­
fo rm a tio n m ay be punched along the vertical colum ns o f the card. Twelve binary dig its can be
punched in each ve rtica l colum n. Thus, in the case o f a 3 6 -d ig it word length, three ve rtica l col­
umns are required using the colum n bin a ry code to com m unicate the in fo rm a tio n contained on
one above in row bin a ry form .

IBM L I I S 0 7 FO RM I 2 I - N - 2

C O LU M N B IN A R Y CARD

The Binary Nature of Input

It is im p o rta n t to note th a t a ll in p u t in fo rm a tio n is read fro m a card as 3 6 -d ig it b inary words.
Once the in fo rm a tio n has been read into the com puter, it can be interpreted by the com puter in ac­
cordance w ith any code th a t the com puter has been program m ed to understand. Thus, standard
alpha-num eric code cards, row bin ary cards and colum n binary cards are a ll read into the com puter
by the same process. However, once the in fo rm a tio n is in the com puter, it is interpreted by the pro­
gram prepared fo r the com puter so th a t the com puter can "u n d e rs ta n d " the in fo rm a tio n com m uni­
cated according to the appropriate code on the in p u t card.
The im p o rta n t p oint to remember in th is co n te xt is th a t all communication to the com puter is
a c tu a lly binary communication. A n y non-binary form o f in fo rm a tio n is read by the com puter as if it

19
were b in a ry in fo rm a tio n and then interpreted " in b in a ry " so th a t the com puter can "u n d e rs ta n d "
the in fo rm a tion .
Since M IN IV A C 601 operates in the same m anner as a large com puter, it must also be given
in fo rm a tio n in b in ary form . This is the reason fo r the six binary input buttons on the M IN IV A C
console. W hen non-binary in fo rm a tio n is to be supplied to the com puter as, fo r exam ple, when
using the decim al in p u t d ia l, M IN IV A C 601 w ill be program m ed to in te rp re t this non-binary in­
fo rm a tio n and to store and process the non-binary in fo rm a tio n in binary form .
In Book IV a program is given fo r conversion from decim al to binary. (See Decimal to Binary
Converter— Book IV.)

Paper Tape Input

In a m anner s im ila r to th a t in w hich in fo rm a tio n is com m unicated on a punched card, in fo r­
m ation on paper tape is com m unicated by punching holes in predeterm ined locations according to
various codes. The m achine reading a paper tape, as the m achine reading a punched card, in ­
terprets the presence or absence o f a hole in a p a rtic u la r location as bin a ry in fo rm a tio n . Using
paper tape it is possible to com m unicate in fo rm a tio n made up o f any num ber o f characters to the
m achine as one continuous record.
The photograph below illustrates the use o f the " 8 c h a n n e l" code, one o f several coding
systems used to com m unicate in fo rm a tio n on paper tape. This system is analogous to the alpha­
num eric code described above fo r use w ith the punched card. The nature o f this code can be easily
determ ined by exam ining the photograph. It should be noted th a t the sm aller dots appearing
along the center o f the paper tape are analogous to the sprocket holes on the edge o f movie film .
These sm all holes are used to move the paper tape past the reading p oint in the paper tape-reading
machine.

A B C D E F G H I JK LM N O PQ R S TU V W X YZO 2 3456789 ■ /& i0%♦H^5IScLQ-Q.Q-£ujt/>LLJ</H^ Jnmi

.
EI_- TTTTII M T • Mi l l
X- • • • • • • • • • • • • • • • • • • • • I • • I• I••• i i•• I••••
()- • • • • • • • • • • ••••••••• •• •• • • ••• • ••
CHECK—* • •• ••• • •• • • •• •• •• •••• • • • • •
8—* •• •• •• •• ••••••••••••••• •••••••
4 —* • • • • • • • • • • • • • • • • •••• •••• •• ••••
2—* • • •• •• •• •• •• •• •• ••• •••• •••••••
1—• • • • • • • • • • • • • • • • • • • • ••• • •••• • •

EIGHT C H A N N E L PAPER TAPE CODE

Magnetic Tape Input

M a g ne tic Tape in p u t systems d iffe r fro m those discussed above in th a t in fo rm a tio n is re­
corded on a plastic recording tape s im ila r to th a t used in home tape recorders. In the case o f mag­
netic tape, in fo rm a tio n is coded as m agnetic "m a rk s " on the tape rather than as holes in the tape.
A m ajor advantage o f m agnetic tape is th a t the m agnetic in fo rm a tio n recorded on the tape can be
erased and the tape can be re-recorded and used m any times.
A second advantage o f the m agnetic tape is th a t the m agnetic tape can be read a t a much
higher speed tha n the paper tape in a much sm aller physical space than is required using either
the punched card or the paper tape. S pecifically, the same am ount o f in fo rm a tio n (3,336 charac­
ters) can be recorded on fo rty-tw o punched cards or on one inch o f m agnetic tape (high density).
The photograph below illustra tes the use o f a seven-bit alpha-num eric code recorded on m ag­
netic tape. Using this code, a character is represented by the presence or absence o f m agnetic
m arks in specified positions across the w id th o f the tape.
B inary in fo rm a tio n m ay also be recorded on m agnetic tape using a 36 b it code s im ila r to
th a t discussed fo r use in connection w ith the punched card. The 3 6 -b it word on m agnetic tape is
recorded in six consecutive colum ns on the tape. A lth o u g h seven positions are available on the
tape, one position is reserved fo r inforrrtation used in checking the re lia b ility o f the reading and
w ritin g operation.

20
 0123456789 AtCDEFGHIJK LMNOfOR STUVWXYZ % i§

Zone

Numerical

STORAGE M E D IA A N D CODES
'Bi-Stable" Elements
In the firs t section o f th is book we exam ined the m em ory elem ent o f the M IN IV A C 601. Now
we are ready to exam ine the way in w hich the same fu n ctio n is perform ed by d iffe re n t elem ents in
larger computers. In every case, the fu n c tio n in g o f the m em ory elem ent in a large com puter is iden­
tic a l to the fu n c tio n in g o f a relay serving as a m em ory elem ent in the M IN IV A C 601. The m emory
elem ent is, in every case, capable o f m a in ta in in g one o f tw o positions u n til it is signaled to change
positions. Such a tw o-valued elem ent w hich can rem ain in e ither o f two positions u n til an action
outside the elem ent causes it to change to the other position is called a "b i-s ta b le " device. It is a
device w hich can stay (is stable) in e ither o f two positions (the p re fix " b i- " indicates two). The
photograph below illu stra te s several bi-stable devices discussed in this book.

"0" State ■I" State

IBM PUNCHED CARD
/L .......
0

MAGNETIC CORE

RELAY OR SWITCH

TUBE OR TRANSISTOR

ELECTRICAL PULSES

J 1

BI-STABLE ELEMENTS

21
Large Computer Storage
The actual form o f bi-stable elements which make up the storage register in large computers
varies w ith the machine. However, three basic types o f storage are presently in popular use. These
are:

m agnetic core storage

m agnetic drum storage
m agnetic disc storage.

Magnetic Core Storage

The elements o f m agnetic core storage are sm all, donut-shaped rings o f ferro-m agnetic m a­
te ria l. These sm all m em ory elements, one hundredth o f an inch in diam eter, can be m agnetized
in a few m illio n th s o f a second and w ill retain th e ir m agnetism in d e fin ite ly.
By passing cu rren t through a w ire going through the center o f the donut-shaped ring, it is
possible to m agnetize the m agnetic core m em ory element. The direction o f the m agnetic fie ld set
up w ith in the core is determ ined by the the direction o f current flow through the wire. The m ag­
netic core is thus a bi-stable elem ent having tw o states o f p o larization representing a 0 and a 1.
Just as in the M IN IV A C several relays m ay be used together to represent a series o f zeros
and ones, so in a com puter using m agnetic core memory, series o f m agnetic core elements may be
com bined to create a bina ry word. A 3 6 -b it bin a ry word is thus remembered by 36 separate m ag­
netic core elements, each of which m ay be m agnetized in either direction.
W hen using the M IN IV A C relay m emory elem ent, the " s ta te " or position o f the relay was
" re a d " on the lig h t attached to the relay switch. The lig h t ON indicated th a t the relay was
rem em bering a one and the lig h t OFF indicated th a t the relay was rem em bering a zero. The
lig h t was used to "se nse " the position o f the relay.

M A G N E TIC CORE PLANE

22
 Reading o f the contents o f a m agnetic core elem ent is accomplished by a "se n sin g " process.
One such sensing process works as follows:
The m agnetic core elem ent is forced into the zero direction by a current pulse. If the
elem ent is storing a one, the change in direction o f p o larization induces a pulse in a
"sensing w ire ." If the elem ent is storing a zero, there is no change in the direction of
p o larizatio n , so there is no pulse.
T his reading process, however, "d e stro ys" the in fo rm a tio n content o f the m agnetic core element.
If the in fo rm a tio n m ust be retained a fte r reading, a special device w hich w ill replace the in fo rm a ­
tio n as it is read must be used.
The preceding photograph shows a m a trix made up of many thousand m agnetic core elements
which together remember hundreds o f thousands o f binary bits in a large com puter.

Magnetic Drum Storage

A lth o u g h the m agnetic core provides a m uch faster means o f storing in fo rm a tio n , some com ­
puters use a m agnetic drum fo r storage. The m agnetic drum is a steel cylinder coated w ith ma­
te ria l s im ila r to th a t used on m agnetic recording tape. This m aterial can be m agnetized w ith a
num ber o f sm all m agnetic "s p o ts " in much the same m anner in which in fo rm a tio n is recorded on
m agnetic tape. The drum rotates a t a constant speed and in fo rm a tio n is "w ritte n o n " or "re a d
fro m " the m agnetic spots by a recording head in much the same m anner th a t in fo rm a tio n is re­
corded or played back from a tape recording.
In the m agnetic drum there is a problem o f id e n tify in g the location on the drum which is to be
read in order to obtain the in fo rm a tio n desired a t a p a rtic u la r point in tim e. Through a sequencing
system, each storage location on the drum is specified by a given address (so th a t a p a rticu la r
location m ay be determ ined) in order to obtain or store in fo rm a tio n a t th a t point on the m agnetic
drum .

W riting on a Reading from a

Drum Drum

23
 Because transm ission o f in fo rm a tio n can occur only when a p a rtic u la r location is passing
under the recording or play back head, the tim e required to obtain in fo rm a tio n from the m agnetic
drum or to record in fo rm a tio n in a p a rtic u la r location on the drum is greater than th a t required to
obtain in fo rm a tio n from the m agnetic core.

Magnetic Disc Storage

The m agnetic disc uses a process s im ila r to th a t used w ith the m agnetic drum . In fo rm a tio n is
recorded on discs w hich look much like standard phonograph records. The discs are stacked in
an a rray surrounding a central spindle in a mechanism w hich operates very much like an ordinary
ju ke box.
In fo rm a tio n is read into or out o f the discs by means of an arm which is able to enter into the
stack o f records and read fro m or w rite on either side o f a p a rtic u la r disc. It is im p o rta n t to re­
member th a t the recording on the disc is a magnetic recording. So th a t although these records
appear to be much like the o rdinary phonograph records, the recording process is one involving
m agnetic spots.

Other Forms of Storage

A lth o u g h the various form s o f storage described above are those most com m only encoun­
tered in com m ercial computers, other form s o f storage are used to a lim ite d degree.
B inary in fo rm a tio n is sometimes stored in an electric capacitor w ith the po la riza tio n o f the
capacitor determ ining w hether the b it stored is a one or a zero. In m agnetic recording, the pres­
ence o f a m agnetic spot was taken to indicate a 1and the absence a 0; in the capacitor the presence
o f extra electrons is taken to indicate a 1 and the absence o f electrons a 0.
Cathode ray tubes are also used fo r storage. W hen these devices are used, a point on the
screen o f the tube is used as the m emory element. A charged p o in t represents a 1 and the absence
o f a charge indicates a 0.
W hen acoustic delay lines are used as storage elements the binary state o f the elem ent is de­
term ined by the presence or absence o f an ultrasonic vib ra tio n in a flu id . The presence o f the
vib ra tio n indicates a 1, its absence a 0.

External Storage
The storage media discussed above are a ll used fo r internal and secondary storage. External
storage involves media associated w ith the o u tp u t units and w ill be discussed in the next section.
A t th is tim e it is s u ffic ie n t to note th a t three m ajor types o f external storage are encountered in
the modern d ig ita l com puter system. These are m agnetic tape, punched cards and paper tape.

PROCESSING TECHNIQUES

Large Computer Processing

Processing w ith in a high-speed d ig ita l com puter is accomplished using electronic circuits
analogous w ith those used in the M IN IV A C 601. W ith some large scale com puters the instructions
are com m unicated using a program board in w hich instructions are wired in a m anner s im ila r to the
wired program m ing o f the M IN IV A C . In most high-speed d ig ita l com puters, various operations
which can be perform ed by the processing u n it are permanently wired into the com puter and the
com puter is program m ed to perform a p a rtic u la r operation (to choose a p a rtic u la r w ired circ u it)
when a specific code is given as an instruction. A com puter, m ig h t fo r exam ple, be program m ed to
choose an a d ditio n c irc u it when it encounters the instruction " 0 1 " and to choose a subtraction
c irc u it when it encounters the instruction " 0 2 ."
W hen coded instructions are used in a large d ig ita l com puter, the numbers representing the
coded instruction m ay be stored in the storage u n it o f the com puter just as data is stored. W hen
th is is done, the processing u n it is directed to certain registers o f the storage u n it to obtain num ­
bers w hich are interpreted as instructions and to other registers to obtain numbers which are in te r­
preted as data. W hen this process is followed, the com puter is said to be operating under control

24
o f a "stored p ro g ra m ." In one sense then, the M IN IV A C 601 or any other w ire-program m ed com ­
puter m ay be said to be operating under control o f a stored program . In the case o f the nu­
m e rica lly coded instru ction m achine, the program is stored as numbers in the storage u n it w hile
in the case o f the w ire-program m ed m achine the instructions are stored in the connections
wired on the program m ing panel o f the com puter.
Electronic circu its used in the high-speed d ig ita l computers use components which are d iffe r ­
ent in fo rm fro m those used on the M IN IV A C 601. The functions w hich they perform , however, are
s im ila r to the operations o f the relays and rotary switch o f the M IN IV A C 601, and the M IN IV A C
components o ffe r the p a rtic u la r advantage o f being com pletely visible so th a t th e ir operation can
be easily viewed. It is easy to see the sw itching action o f a relay on the M IN IV A C 601 but impos­
sible to see the s im ila r fu n c tio n in g o f a transistor or m agnetic core elem ent.
The processing u n it o f the com puter is the central control instrum ent o f the com puter system.
The processing u n it is usually associated w ith the "m a in fra m e " o f the com puter and d ire ct com­
m unication w ith the processing u n it is provided fo r the operator through the switches and lights of
the com puter console.

The Binary Nature of Processing

The processing u n it o f most large com puters operates only in binary. Just as the relays used
fo r processing on the M IN IV A C are capable o f dealing only w ith zeros and ones, the processing
systems o f larger com puters w ork only in binary code.
Since in fo rm a tio n is com m unicated to the m achine in other than bin a ry fo rm , it m ust be
coded in b in a ry form according to a system which w ill pe rm it the m achine to recognize it as
num eric, alphabetic, or special character in fo rm a tio n . This binary coding o f non-binary in fo rm a ­
tio n is often accom plished using a Binary Coded Decimal or "B C D " code rather than using the
bina ry equivalents o f the decim al num ber (see firs t section o f Book IV). The fo llo w in g ch a rt pro­
vides a sum m ary o f the representation o f alphabetic, num eric and special characters according
to the Binary Coded Decimal System.

THE BINARY CODED DECIMAL SYSTEM USED FOR PROCESSING

Character BCD Code Char. BCD Code Char. BCD Code Char. BCD Code
blank 110 000 A 010 001 N 100 101 + 010 000
1 000 001 B 010 010 O 100 110 — 100 000
2 000 010 C 010 011 P 100 111 / 110 001
3 000 011 D 010 100 Q 101 000 = 001 011
4 000 100 E 010 101 R 101 001 - 001 100
5 000 101 F 010 110 S 110 010 011 011
6 000 110 G 010 111 T 110 011 ) 011 100
7 000 111 H 011 000 U 110 100 $ 101 011
1 011 001 V 110 101 * 101
8 001 000 100
9 001 001 J 100 001 w 110 110 111 011
0 000 000 K 100 010 X 110 111 ( 111 100
+0 011 010 L 100 011 Y 111 000
-0 101 010 M 100 100 z 111 001

In Book III o f th is series the decision-m aking fu n ctio n o f the processing u n it is discussed, and
examples o f the operation o f the processing u n it as a decision-m aking device are provided. Book
III also contains a detailed discussion o f the logical operations which enable the processing u n it
to fu n c tio n as a "R ea son in g " device. In Book III you w ill discover how M IN IV A C 601 is able to
d u plica te m any o f the functions o f hum an reason which are norm ally considered proof o f a hu­
man being's a b ility to " t h in k . "

O U TPUT M E D IA A N D CODES

Direct Output
Just as bin a ry in fo rm a tio n may be given dire ctly to the com puter using in p u t buttons on the
com puter console, b in a ry o u tp u t m ay be obtained d ire c tly from the com puter through lights
s im ila r to the bin a ry ou tp u t lights o f the M IN IV A C 601. W hen binary o u tp u t lights are used, the

25
code already established fo r use w ith the M IN IV A C 601 is employed. A lig h t ON indicates a " 1 " ;
a lig h t OFF indicates a " 0 . "
A photograph below shows the direct o u tp u t lights on the console o f the IBM 7090 com puter.
You w ill notice th a t there are 37 lamps in the row o f o u tp u t lights o f the 7090 com puter. These
are used to display the contents o f the 36 binary " b its " in a 7090 storage register. The fa r le ft
lig h t indicates the sign o f the num ber when the lights are displaying a binary number. The "sign
in d ic a to r" lig h t ON represents m inus; the lig h t OFF represents plus.

DATA PROCESSING SYSTEM |

Large Computer Output Units

O u tp u t in fo rm a tio n m ay be obtained fro m a high-speed d ig ita l com puter using lights and
physical indicators s im ila r to the bin a ry ou tp u t lights and decim al o u tp u t d ial o f the M IN IV A C .
W hen o u tp u t devices o f this type are employed, they are generally used only to obtain lim ited op­
era ting data from the com puter and are usually located on the com puter console.
M ore extensive o u tp u t in fo rm a tio n is norm ally obtained using a " p r in te r " u n it in which a
typ e w rite r-lik e device produces typew ritten pages o f numbers or other w ritte n in fo rm a tio n . The
p rin te r m ay be connected d ire c tly to the processing u n it o f the com puter in which case in fo rm a ­
tio n is com m unicated d ire c tly from the processing u n it to the p rin te r. The p rin te r m ay also be
connected to a u n it which is able to "re a d " one o f the other ou tp u t media (fo r exam ple m agnetic
tape or punched cards). In th is second case, output in fo rm a tio n fo r p rin tin g is a c tu a lly trans­
m itte d from the com puter on another media through another ou tp u t u n it and then transform ed
into prin ted m aterial away from the m ain com puter.
The discussion o f coding and media undertaken w hile discussing in p u t units is also applicable
to the o u tp u t situation. In essence the o u tp u t device is sim ply a m achine which reverses the
process o f the in pu t device. In the case o f m agnetic tape, the input and ou tp u t devices are actu a lly
in the same physical unit. The same equipm ent is used to record and to play back the in fo rm a tio n
on m agnetic tape.
The u n it used to reverse the process o f the card reader is the card punch u n it and the paper
tape reader has a com plem entary paper punch u n it. W ith both the paper tape and card punch,

26
a series o f dies are used to punch holes in the tape or card in the location corresponding to coded
in fo rm a tio n sent to the punch u n it from the processing u n it o f the com puter.
The cathode ray tube o utpu t u n it which is used to display o utput in a form which has no count­
erpart in the inp u t sequence should be noted. T his television-like display u n it provides an "a n a lo g "
in the form o f a p ic to ria l representation o f inform ation. A photograph of a Cathode Ray Tube
ou tp u t u n it appears in the next section o f this book.

4. CO M M ERCIAL COMPUTER EQUIPMENT

This section o f the book contains photographs o f equipm ent m anufactured by the In te r­
nationa l Business M achine C orporation to perform each o f the fo u r m ajor com puter
functions. These photographs are included to enable you to become fa m ilia r w ith the
appearance o f units p erform ing the various functions which we have discussed.

Card Input Equipment

Inpu t in fo rm a tio n fo r use by a com puter system may be prepared on a card punch o f the type
illustrate d below. The m achine in the photograph is the IBM 026 Card Punch which punches cards
using the IBM A lpha-nu m e ric Inp ut Card Code.

The cards to be punched are placed in the hopper on the rig h t hand side of the 026 and fed
through the m achine to the punching station. The operator, using the keyboard of the machine,
punches the card in much the same way th a t he would type a letter using a typew riter. The space
bar a t the bottom o f the keyboard is used to move the card under the punch dies w ith o u t punching.
W hen a key is pressed, a hole is punched in the card in the p a rtic u la r colum n under the punch dies
a t the tim e the key is pushed. The 026 also prints above the colum n which has been punched the

27
 I

num ber or le tte r represented by the punches. A n exam ple o f a card prepared by an 026 card
punch is shown below.

/ hiBCDEFGHi JKLMNOPQRST'JVWXYZ 1 2 3 4567390 = » $.-< *> + MINIVAC 601

/ lllllllll I I II I II
lllllllll I I I I
oooooaooooooaoooooooo||||||||ooooooooooooa|ooooo|ooo|oooaooooooo|oooo|oooooooooo
1 1 3 4 5 I 7 I I 70 II 1113 M W I 1 17 I I I I 30 21 31 23 34 35 38 37 78 29 30 31 31 33 34 35 3G37 % 394041 474344 45 45 47 4141 50 31 52 515(55 31 57 31 39 40 614743 M IS 14 IT 48 49 70 71 7773 74 78 76 77 71 71 M

t1t|l111111l|l111111111111111111l|l11111111111111111111111111111l|l11l|l11111111
Z222|222ZZ222|2222222|222222Z22222|222222222222222222222222222222222222222222222

33333|33333333|3333333|333333333333|33333333333||||333333333333333|3333333333333

444444|44444444|4444444|444444444444|44444444444444||||44444|4444444444444444444

S555555|5S55555S|55S5555|555S55555S55|555555555S55S55S5555S555|5|555555S5555555S

66S66S6G|66666666|66GG66G|6G6666666666|6666G66666666666666666666G666|66666666GG6

777777777|77777777|7777777|777777777777|7777777777777777/77777777777777777777777

8888888888|88888888|8888888|8888B8888888|B88888||||||||888888888S8B888888888B888

99999999999|99999999|9999999|999999999999|.9 999999999999999999|9|9999999999999999
I 2 3 4 8 • 7 I t t 0 1 1 1 2 1 3 1 4 1 S tlt7 U 1 l20 21 22 232 48 2l 27 2l2l3031323334333l 37 3 in 4 0 4t 42 43 44 4S 46 47 4l4S90 91 9293MSSSB57MS9 80 61 82 63 64 65 K I76 l8 B T 0 71 72 73 74 7S7l 77 7l7SIO
________________ I»M8081

IN P U T CARD PREPARED ON THE 026 CARD PUNCH

Once the in p u t in fo rm a tio n has been punched onto the card, the card can be read by the card
reader o f the com puter. The photograph below illustrates the IBM 7500 card reader used in con­
ju n c tio n w ith a high-speed d ig ita l com puter. The m echanism o f the card reader includes two sets
o f "re a d in g " brushes. In fo rm a tio n fro m the card is read beginning w ith the firs t 36 locations in
the " n in e " or bottom row o f the card. A fte r the firs t 36 locations o f the " n in e " row have been
read, the next 36 locations are read. The last e ight locations (columns 73-80) are not norm ally
used fo r in p u t in fo rm a tio n but are reserved fo r sequencing in fo rm a tio n used to id e n tify a p a rtic u ­
lar card.
For a 72 colum n read cycle, each corresponding row sta rtin g w ith the " n in e " (bottom ) row
is divided into tw o 36 colum n "w o rd s ". Thus the m achine reads tw e n ty-fo u r 3 6 -d ig it bin a ry words
in the order indicated in the illu s tra tio n below.

Unused
C o lu m n s

/
/ © © 12-Row

© m 11-Row

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0© 0 0 0 0 0 0 0 0 0 0 0 0 0 0 D0 G0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 @ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 i i i i i i i i

2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 <Ts) 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 0 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

33333333333333333033333333333333333 33333333333333333033333333333333333 33333333

44444444444444444044444444444444444 44444444444444444044444444444444444 44444444

5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5® 555 5555 5 585 55 555 55 5 5 5 5 5 5 5 5 5 55 5 5 5 5 5 0 5 5 5 55 5 5 5 5 5 55 5 5 5 5 5 55555555

5 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 0 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 0 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 66666666

7777777777 77 77 777 0 777 7777 77 777 77777 7777 7 777 77 7777 7770 7777777 77 7777 77 7777777777

88888888888668888086888888888888888 88888888888888888088888888888888888 88888888

9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 0 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 0 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 99999999
1 2 3 4 5 6 7 8 9 10 31 12 13 14 15 16 17 18 19 20 21 2 2 2 3 2 4 25 2 6 2 7 2 8 2 9 30 31 32 33 3 4 3 5 3 6 3 7 38 3 9 4 0 41 4 2 4 3 4 4 4 5 4 6 4 7 4 8 4 9 50 51 5 ? 53 54 55 5 6 57 58 59 60 61 62 6 3 6 4 6 5 6 6 6 7 68 69 70 71 72 71 74 75 7 6 77 76 79 80

THE SEQUENCE FOLLOWED W H E N READING 72 CO LUM NS

OF IN F O R M A T IO N FROM A N IN P U T CARD

28
Input-Output Units
The photographs below illu s tra te IBM card, tape, and p rin te r units.
W hen cards are read by the card reader, th e y are placed in the hopper at the fro n t o f the
reader. D uring the read cycle they pass under the read brushes and out into the hopper on top of
the unit.
The m agnetic tape u n it both reads and w rites on m agnetic tape. This u n it is thus a combined
in p u t-o u tp u t device.
The equipm ent used to read paper tape operates, as does the card reading equipm ent, by
in te rp re tin g the holes in the paper tape as b in a ry in fo rm a tio n . Once th is in fo rm a tio n is read it is
interpreted according to a program previously supplied to the machine. As in the punched cards, a
hole is read as a one and the absence o f a hole as a zero.

IBM 780 C R T Display

IBM 962 Tape Punch IBM 729 IV Magnetic Tape Unit IBM 1403 Prihter

IBM 382 Paper Tape Reader IBM 7500 Card Reader IBM 7550 Card Punch

29
Storage Units
U nits em ploying m agnetic core, m agnetic drum and m agnetic disc storage media are illu s ­
tra ted below.

CORE DRUM DISC

STORAGE U N IT STORAGE U N IT STORAGE U N IT

5. COMPUTERS OF TOMORROW

A lth o u g h com puter technology has already reached levels o f perform ance beyond im agina­
tio n only a decade ago, technological progress in the com ing years promises to open new horizons
fo r these am azing machines.
The capabilitie s o f these machines o f tom orrow can be forecast from the capacities o f the
most advanced com puter systems being operated today. A n example is the IB M "S tre tch System"
illu stra te d below. The " S tre tc h '" has an internal m em ory capacity o f more than 16 m illio n b i­
nary bits. The speed o f this system is such th a t it requires only 2 micro-seconds fo r "S tre tc h " to
add two numbers and store the results.
The research laboratories o f com puter m anufacturers throughout the country are w orking
now on com puter systems w ith greater capacity, greater speed and more e ffic ie n t design. Every
day, the men and women involved in this new and e xciting fie ld put com puters to w ork on new
problems, in d iffe re n t areas. The developm ent o f the com puter is m aking possible the e ffic ie n t
h andling o f vast am ounts o f data and the solution o f problems in science, engineering and
business.

30
IBM 7030 STRETCH SYSTEM
 APPENDIX
D IG ITA L A N D ANALOGUE COMPUTERS
T hrougho ut th is m anual, a ll discussion o f computers refers to digital computers. There are
in use, however, m any com puters whose basis of operation is com pletely d iffe re n t. These are ana­
logue computers. Both types o f com puters are capable o f perform ing sim ila r types o f operations
and can serve s im ila r functions.
The differen ce between digital and analogue com puter systems is basically a difference in the
m ethod the system uses to handle in fo rm a tio n . The analogue com puter handles in fo rm a tio n as a
continuously varying signal w hile the d ig ita l com puter handles in fo rm a tio n in discreet form .
The follow in g sketch illustrates gra p h ica lly the difference between continuous and discreet
representation:

t
The solid line represents the value o f a fu n ctio n as it changes over tim e. This is a continuous
fu n c tio n ; th a t is, there is a value fo r the fu n ctio n at every p oint in tim e. This representation can
be handled by an analogue machine.
The vertical bars also represent the value of the fu n ctio n as it changes over tim e. However,
the fu n c tio n exists only a t the d is tin c t points in tim e represented by the bars; the fu n ctio n is not
defined between these points. As the points in tim e are chosen closer and closer together, a line
connecting the ends o f the bars w ill approach the continuous curve.
The reason fo r this d ifference in fo rm o f in fo rm a tio n is essentially because o f a difference
in the type o f components which go into the systems. The analogue com puter uses m echanical or
electrical components to represent relationships. A n analogue com puter m ight, fo r example use
a ro ta ting s h a ft w ith various sha ft positions corresponding to values o f a variable. Or, an ana­
logue com puter could use an electric capacitor— which is capable o f storing electrons— w ith v a ri­
ous levels o f charge on the capacitor corresponding to values o f a variable.
The d ig ita l com puter, on the other hand, uses various components to represent specific va l­
ues o f a variable. The d ig ita l com puter can, fo r exam ple, use the two positions o f a relay (on
or o ff) to represent two d is tin c t values o f the variable.
As an exam ple o f how the two systems d iffe r in operation, consider the problem o f adding
tw o pounds o f salt plus three pounds o f salt:
T o solve this problem as an analogue m achine would, we would do the follow ing:

1. Set a scale to read zero

2. Pour 2 pounds o f salt onto the scale (the scale indicator would clim b as the salt was be­
ing poured onto it).
3. Pour 3 more pounds o f salt onto the scale (once again, the scale indicator would clim b
as the salt was being poured onto it).
4. For the answer, we would read the fin a l result on the scale: 2 + 3 = 5.

32
If we wished to do this on an analogue m achine, we would let amperes o f current represent pounds
o f salt. T h a t is, we would use numbers o f amps o f current as analogues o f the numbers o f pounds
o f salt. W e would then do the follow ing:
1. Set an am m eter to zero
2. Send 2 amps o f cu rren t through the am m eter
3. Send 3 more amps o f c u rrent through the am m eter
4. Read the fin a l result on the am m eter: 2 + 3 = 5
To solve the same problem as a d ig ita l m achine would, we would firs t have our salt in one-
pound blocks. W e would select 2 blocks, then 3 more blocks. Then we would observe th a t there
were a to ta l o f 5 blocks.
To solve th is problem on a d ig ita l machine, we would program the m achine to add and give
the m achine the numbers 2 and 3. The m achine would then give the answer 5.
The basic diffe re nce between the m achine can be sum m arized as follows:
A n analogue m achine accepts quantities
A d ig ita l m achine accepts numbers

33
 BOOK III

How Computers Make Logical Decisions

PREFACE
In Books I and II the general operating characteristics o f the M IN IV A C 601 and large com ­
m ercial com puters were examined. Emphasis was placed on the nature and fu n ctio n o f com po­
nents w hich are combined to m ake up a com puter system. This book assumes a basic knowledge
o f the functions o f the components o f d ig ita l com puter systems such as would be obtained from
reading Books I and II.
In this book a tte n tio n is focused on the process and techniques by which com puters are able
to perform operations w hich we m ig h t describe as th in k in g . There is a great deal o f question as to
w hether com puters a c tu a lly " t h in k " . Discussion o f this question is deferred u n til later in this book
a fte r you have done some o f the experiments. M ost o f w hat we speak o f as " th in k in g " is a process
o f searching m em ory or the environm ent fo r facts and m aking decisions based on those facts. In
this book you w ill learn how com puters make decisions based on such facts. The concepts associ­
ated w ith com puter logic and the techniques w hich com puters use to solve problems and m ake de­
cisions w ill be explored in some detail.

1. BASIC OPERATIONS
As you have undoubtedly come to realize, the m odern-day com puter perform s p ra c tic a lly all
o f its operations using the zeros and ones o f a binary code. In Books I and II the reason fo r this
use o f b ina ry coding was explained w ith reference to the "b i-s ta b le " elem ents used fo r control,
processing and storage in large scale com puters and the M IN IV A C 601.
In the experim ents w hich you have already perform ed, the lights were interpreted as a 1
when they were ON and a 0 when they were OFF; the pushbuttons indicated a 0 when they were
UP and a 1 when D O W N ; the relays stored a 1 when they were ON and a 0 when OFF. Through
your experience w ith the lights, pushbuttons and relays, you should be well acquainted w ith the
o n -o ff (0-1) nature o f the d ig ita l com puter.
As indicated above, th is book explores the logical operations used by a d ig ita l com puter
and examines how com puters " t h in k " and make decisions. Logical operations, as a ll other com ­
p u te r functions, are perform ed using the 0's and 1's o f the binary code. The nature o f this " th in k ­
in g " process can best be dem onstrated w ith some sim ple examples.

The Operation " A N D "

In beginning to exam ine com puter th o u g h t processes, it m ay be h e lp fu l to use examples in or­
der to develop some basic logical elements. These can then be used to solve more com plicated and
sophisticated problems.
As an exam ple o f a logical operation, consider the fo llo w in g series o f statements:
A. If I go outdoors
and
B. If it is raining
then
C. I w ill get wet.
Here we have three ideas w hich com bine in a " lo g ic a l" way to produce a conclusion w hich,
according to the rules w hich we use fo r th in k in g , is " tr u e " . In order to convey the basic charac­
teristics o f th is com bination o f ideas to the com puter, we must fin d a way to represent the basic
relationships w hich lin k the three sentences in the exam ple above using the 0-1 binary code w ith
w hich the com puter is able to work.

35
 Let us look a t the sentences again. There are three statem ents related in such a way th a t the
th ird statem ent can be arrived a t as a va lid conclusion, given the in fo rm a tio n supplied by the
firs t two sentences. To p u t it another way, if the firs t two statem ents are " tr u e " , then the rela­
tionships between the firs t two sentences and the th ird sentence are such th a t the th ird sentence
must also be true. On the other hand, neither the firs t sentence nor the second sentence alone
leads to a logical conclusion.
In order to make this point obvious, we m ight restate the three statem ents above in the fo l­
lowing form :
A. If it is tru e th a t
I go outdoors
AND
B. If it is tru e th a t
it is raining
TH EN
C. It w ill be tru e th a t
I w ill get wet.
If we now ignore the content o f the three sentences and look only a t the relationships which
exist between A, B, and C we can state the general condition th a t
If A is true
AND
If B is true
TH EN
C is true.
The relationship lin k in g statem ents A and B w ith statem ent C establishes an exclusive set
o f conditions which w ill be satisfied only when statem ent A and statem ent B are both true. It fo l­
lows from the above th a t:
If A is false
AND
If B is true
TH EN
C is false
or,
If A is true
AND
If B is false
TH EN
C is false.
To relate the contents o f these logical statem ents to the binary coding system used by the
com puter, the fo llo w ing relationships may be defined.
1 = true
0 = false (not true)
Using this coding system and rem em bering th a t (a) pushing button DO W N indicates as 1,
leaving it UP indicates a 0, and th a t (b) a lig h t O N indicates a 1, a lig h t OFF indicates a 0, we can
program the com puter so th a t the tru e or false conditions fo r statements A and B can be given to
the com puter as inp ut and the com puter w ill give us the condition o f the th ird statem ent as true
or false output:

1 + /1 Y
1X /2Y
2 X /3 A
3 B /3 —

COMPUTER REPRESENTATION OF THREE STATEMENTS

36
 Using the c irc u it above, pushbutton 1 w ill be used to com m unicate the true or false condition
o f statem ent A and pushbutton 2 w ill be used to com m unicate the condition o f statem ent B. If
pushbutton 1 is D O W N and pushbutton 2 is UP, statem ent A w ill be true and statem ent B w ill
be false. W hen both pushbuttons are held down, both statem ents A and B w ill be indicated as
true. In sum m ary:

PUSHBUTTON 1 PUSHBUTTON 2 LIG H T 3

Up or O ff Statem ent A Statem ent B Statem ent C

is false is false is false

Down or On Statem ent A Statem ent B Statem ent C

is true is true is true

Pushbutton— Light Equivalents

Using the c irc u it and representation indicated above, pushing both pushbutton 1 and push­
button 2 indicates to the com puter th a t both statem ents A and B are tru e — it is true th a t I am go­
ing outside and it is true th a t it is raining. The com puter then indicates th a t, given the in fo rm a ­
tio n w hich we have provided as input, it can in d ica te th a t statem ent C must also be tru e — it is
tru e th a t I w ill get wet.

This c irc u it is an example o f a basic relationship w hich w ill be found in com puter handling
o f m any logical problems. C ertain in fo rm a tio n about the relationships between the elements
(statements) in the logical problem is supplied to the com puter by a program . W e supplied this in ­
fo rm a tio n about the relationship between statem ent A, statem ent B, and statem ent C by w irin g the
c irc u it indicated onto the console o f the M IN IV A C . Once the com puter has this in fo rm a tio n —
once the com puter is "p ro g ra m m e d "— it is ready to use the in fo rm a tio n about statem ents A and
B to reach a conclusion about statem ent C. The program tells the com puter th a t both state­
ments A and B m ust be true fo r statem ent C to be true. Therefore both pushbutton 1 and pushbut­
ton 2 m ust be pushed before lig h t 3 comes ON.
T his simple com bination o f three statem ents demonstrates one o f the basic operations which
a com puter m ust be able to perform if it is to solve logical problems. This operation is called
" A N D " . Statements A and B m ust both be true fo r statem ent C to be true.
Just as symbols are used as a shorthand to represent the ideas in a rith m e tic ( " + " repre­
sents the idea o f a ddition ) symbols are used to represent the operations involved in solving logi­
cal problems. In the case o f the concept A N D the symbol X is used as a symbol:

X = AND

Thus, the relationships between statem ents A , B and C m ay be stated as:

AX B= C

Using th is notation , in com bination w ith the d e fin itio n s o f 1 and 0 as true and false (not
true) respectively, the conditions expressing the relationships lin k in g the three statem ents
above can be expressed as:

1X1 = 1
0 X 1= 0
1X0 = 0
0X0 = 0

The equals sign ( = ) in these equations m ay be read as "T H E N " rather than as "e q u a ls ".
The relationships expressed by each o f the fo u r logical statem ents above are:

37
 A. If 1 go out
AND
B. If it is raining (1 X 1 = 1)
C. 1w ill get wet.

A. If 1 do not go out
AND
B. If it is raining (O X 1 = 0)
THEN
C. 1w ill not get wet

A. If 1 go out
AND
B. If it is not raining ( 1 X 0 = 0)
TH EN
C. 1w ill not get wet.

A. If 1do not go out

AND
B. If it is not raining ( 0 X 0 = 0)
TH EN
C. 1w ill not get wet.

In the program fo r the com puter representation o f the three statem ents, the condition " I go
o u t" is represented by pushbutton 1 and the statem ent " I t is ra in in g " is represented by pushbut­
ton 2. The lig h t represents the statem ent " I w ill get w e t". Remembering th a t a 1 means th a t a
pushbutton is D O W N or a lig h t is ON and a 0 indicates th a t the pushbutton is in UP and th a t the
lig h t is OFF, we developed a com puter program based on the fo u r equations above.

The Operation "O R "

Just as a rith m e tic uses several d iffe re n t kinds o f operations (addition, subtraction, division,
m u ltip lic a tio n ) to solve a rith m e tic problems, the solution o f logical problems s im ila rly involves
the use o f several d iffe re n t logical operations. W e are now fa m ilia r w ith one o f these operations—
the operation " A N D " represented by the symbol X.
A second operation o f basic im portance is the "O R " operation. Just as the " A N D " oper­
ation represents the idea o f two things both being true, the "O R " operation represents the idea
o f either o f two or more things being true.
Using an exam ple s im ila r to our previous three statements, we can consider the follow ing:

A. If I stand in the rain

OR
B. If I stand in the shower
TH EN
C. I w ill get wet

Logical representation o f the relationships between these three statem ents may be sum m ar­
ized using the symbol + to mean "O R ".

+ = OR

A nd the relationships between the statem ents may be stated as:

A + B= C

These relationships can be represented to the com puter using the fo llo w in g program :

38
 1Z

1 + /1 Y
1 Y /2 Y
2 X /1 X
2 X /3 A
3 B /3 —

COMPUTER REPRESENTATION OF THREE "O R " STATEMENTS

T his program establishes the condition o f the three statem ents w ith pushbutton l represent­
ing statem ent A , pushbutton 2 representing statem ent B and lig h t 3 representing statem ent C.
W hen pushbutton 1 is pushed, statem ent A is indicated as true. W hen pushbutton 2 is pushed,
statem ent B is indicated as true. W hen lig h t 3 is O N, statem ent C is indicated as true.

" A N D " and "O R "

The concepts o f " A N D " and "O R " can, o f course, be combined to express somewhat more
com plicated conditions fo r a logical problem. You may wish to experim ent w ith some circu its rep­
resenting the conditions o f various com binations o f " A N D " and "O R " relationships. As an ex­
am ple o f one such com bination, consider a com bination o f the two sets o f statem ents used sep­
a rately fo r the " A N D " and "O R " examples thus far:
A. If I go outside
AND
B. If it is raining
OR
C. If I stand under the shower
TH EN
D. I w ill get wet.
This com bination o f relationships may be expressed sym bolically as:
A X B + C = D
The program and c irc u it representing th is set o f relationships in indicated below. Pushbut­
ton 1 is used to represent statem ent A , pushbutton 2 to represent statem ent B, pushbutton 3 to
represent statem ent C and lig h t 4 to represent statem ent D.
1Z 2Z

1 + /1 Y
1 Y /3 Y
IX /2 Y
2 X /3 X
3 X /4 A
4B /4 —

COMPUTER REPRESENTATION OF " A N D " A N D "O R " STATEMENTS

The Operation " N O T "

The logical operations defined thus fa r have each represented a concept which was basic
to the development o f logical reasoning. The " A N D " operation represents the concept o f com bi­
nation. The "O R " operation represents the concept o f alternative. The " N O T " operation repre­
sents an equally im p orta n t, but somewhat d iffe re n t concept.

39
 In the case o f " A N D " , we were concerned w ith things happening together. In the case of
" N O T " , we are concerned w ith the reverse o f the " A N D " situation. The " N O T " operation is used
to deal w ith the concept th a t som ething w ill happen if som ething else does not happen.
Returning to the fa m ilia r exam ple concerning the wetness o f rain, consider the follow ing
series o f statements;
A. If is it raining
AND
B. If I am N O T under cover
TH EN
C. I wi 11 get wet
The program and c irc u it representing th is set o f relationships is given below. Pushbutton 1 is used
to represent statem ent A , pushbutton 2 to represent statem ent B, and lig h t 3 to represent state­
m ent C:
1 + /1 Y 1Z 2Z 3B 3-
IX /2 Y
2 Z /3 A
1+ 1Y 1X 2Y 2X
3 B /3 —

COMPUTER REPRESENTATION OF THREE " N O T " STATEMENTS

Using the n o rm a lly closed contacts o f pushbutton 2, we are able to represent the concept
" N O T " . L ig h t 3 comes ON if pushbutton is pushed and pushbutton 2 is N O T pushed.
Sym bolically, the concept N O T is represented by a line over a symbol. For example,
Z = 1
Thus, the relationship between the statem ents above may be stated as:
A X "B = C

" A N D " , "O R " and " N O T "

The examples used to dem onstrate the three operations discussed thus fa r can be combined
fu rth e r to create a program representing the fo llo w in g series of related statements:
A. If it is raining
AND
B. I am outside
AND
C. If I am not under cover
OR
D. If I am in the shower
TH EN
E. I w ill get wet
In the fo llow in g program , statem ent A is represented by pushbutton 1, statem ent B is repre­
sented by pushbutton 2, statem ent C is represented by bushbutton 3, statem ent D is represented
by pushbutton 4, and statem ent E is represented by lig h t 5.

1—
|—/1 Y
1X /2 Y
2 X /3 Y
3 Z /5 A
1Y /4 Y
4 X /5 A
5 B /5 —

40
Sym bolically, the set o f relationships m ay be expressed as:

A X B X C + D = E

Note th a t the n orm a lly open contacts o f pushbuttons 1, 2 and 4 are used, w hile pushbutton
3 is wired using the norm ally closed contacts.

The Operation "EITHER BUT N O T BOTH"

As we begin dealing w ith more com plicated circuits, it becomes desirable to express con­
cepts which are somewhat more involved than the sim ple com binations o f " A N D " , "O R " and
" N O T " considered above. The "E ITH ER BUT N O T B O T H " concept provides an exam ple o f such
a case. The idea to be expressed in this concept can be illustrated w ith reference to our ever-pres­
ent wetness problem :

A. If I tu rn on the hot w ater

OR
B. If I tu rn on the cold water
TH EN
C. I wi 11 have to step out o f the shower.

C onverting this series o f statem ents into a program fo r the M IN IV A C , we have the follow ing
program and c irc u it:

1 + /1 Y
1 Z /2 X
IX /2 Z
2 Y /3 A
3 B /3 —

Pushbutton 1 is used to represent statem ent A, pushbutton 2 to represent statem ent B, and
lig h t 3 to represent statem ent C. The program uses the contacts so th a t if either pushbutton 1 or
pushbutton 2 is pushed, lig h t 3 w ill come ON. L ig h t 3 w ill NO T come on if both pushbuttons 1
and 2 are pushed.
Expressing these relationships sym bolically requires the use o f two statements, each express­
ing a condition which m ig h t exist:
1. Statem ent A , but N O T statem ent B, may be true:

AXB
2. Statem ent B, but N O T statem ent A , may be true:

AX B

Since either o f the above is permissible, the entire set o f relationships may be stated as:

( A X B ) + ( A X B) = C

A Short Summary

It m ay be h e lp fu l to review b rie fly the fo u r logical operations which have been demonstrated
thus fa r using the pushbuttons and lights:

41
 The concept "AND"

A B c

The concept "O R"

A + B= C

The concept "N O T "

~A- B

The concept "EITHER BUT

NOT BOTH"

(A X B ) + (A X B ) = C

2. RELAYS FOR MORE FLEXIBLE THINKING

In Book II we discovered th a t the relays could be used to supply an e ffective binary storage
system fo r the M IN IV A C 601. A t th a t tim e it was also indicated th a t the relays would play an
im p o rta n t p a rt in the processing operations o f the com puter. As you program more com plicated
problems on the M IN IV A C , you w ill fin d th a t the relay is an extrem ely versatile component. In
more com plicated circuits, the relays w ill be used to perform a ll the basic logical functions dem ­
onstrated using the pushbuttons and lights in the e arlier section. In fa ct, a single relay w ill, in
some situations, be used to represent tw o or more logical operations a t one tim e.
The fo llo w in g program s dem onstrate the use o f the relays to perform the basic " A N D " ,
"O R ", " N O T " , and "E ITH E R BUT N O T B O T H " functions already developed using the pushbut­
tons and lights.

42
The Relay A N D Circuit
The fo llow ing c irc u it and program represent the condition th a t if relay 1 is ON ( = 1 ) A N D
relay 2 is ON ( = 1 ) then lig h t 3 w ill be ON (— 1). Sym bolically, A X B = C:

1 + /1 Y
1X /1C
1 F /1 -
2 + /2 Y
2 X /2 C
2 F /2 —
1 + /1 H
2 + /2 H
2 G /3 A
3 B /3 —

THE RELAY A N D C IR C U IT

In this c irc u it, pushbutton 1 is used to control relay 1 and pushbutton 2 to control relay 2.
Pushing pushbutton 1 causes relay 1 to close by providing current to the coil o f relay 1. In a
s im ila r manner, pushbutton 2 controls the flo w o f current to the coil o f relay 2. The norm ally
open contacts o f the two relays are used to e ffe c t the " A N D " condition. L ig h t 3 indicates when
the " A N D " condition is met— both relay 1 and relay 2 are closed (ON).

The Relay OR Circuit

The program and c irc u it below use two relays to establish the conditions fo r the "O R " rela­
tionship. Using th is program , if either relay 1, controlled by pushbutton 1, or relay 2, controlled by
pushbutton 2, are O N, then lig h t 3 w ill come ON indicating th a t the OR condition is met.
Sym bolically, the condition is represented by the expression:
A + B= C

1 + /1 Y
1 X /1 C
IF /1 —
2 + /2 Y
2 X /2 C
2 F /2 —
1 + /1 H
1G /3A
2 + /2 H
2 G /3 A
3 B /3 —

THE RELAY OR C IR C U IT

The Relay N O T Circuit

The c irc u it and program below represent the NO T condition expressed in the form th a t if
relay 1 is not energized (is not O N) lig h t 2 w ill come ON. Sym bolically, this may be stated:

A = B

43
 1-4-/1V
ix/ic
1 F /1 -
1 + /1 H
1J/2A
2 B /2 —

THE RELAY N O T C IR C U IT

The Relay EITHER BUT NOT BOTH Circuit

The c irc u it and program below represent the EITHER BUT NO T BOTH condition using relays
1 and 2 controlled by pushbuttons 1 and 2 respectively. The symbolic representation fo r this con­
d itio n is:

(A X B) + (A X B) = C

1 + /1 Y
1 X /1 C
1 F /1 -
1 -4-/1H
1 J /2 G
1 G /2J
2 + /2 Y
2 X /2 C
2 F /2 —
2 H /3 A
3 B /3 —

3. THINKING AND DECISION-MAKING

Boolean Algebra
The operations w ith which you are now fa m ilia r are the basic operations o f Boolean Algebra.
Just as ad dition and subtraction are basic to o rdinary algebra, A N D and OR are basic to Boolean
Algebra.
Boolean A lgebra was introduced in 1847 by George Boole, an English m athem atician and
logician. Boole's system was developed to provide a "s h o rth a n d " fo r the system o f logic o rig in ­
a lly set fo rth by A risto tle . A ris to tle 's system d ealt w ith statem ents w hich were either tru e or false.
O nly recently has Boolean A lgebra become an im p o rta n t m athem atical fie ld . Because o f
its e ffic ie n c y fo r handling any single-valued fu n ctio n w ith only two possible states, Boolean A lg e ­
bra has become the basic language used to deal w ith the sw itching circu its which are basic to
modern computers. The app licatio n o f the operations and techniques o f Boolean A lgebra to
sw itching circuits was firs t suggested in 1938 by Dr. Claude Shannon o f the Massachusetts In­
stitu te o f Technology. It was Dr. Shannon's original development w ork which led to the M IN I­
V A C 601.
The value o f Boolean A lgebra becomes p a rtic u la rly apparent when com plex logical problems
must be organized fo r presentation to a com puter, or when it is desirable to s im p lify an involved

44
c irc u it. For the purposes o f this m anual, the follow ing b rie f introduction to Boolean algebra is
s u ffic ie n t:

Basic concepts
The two-state system:
Boolean A lgebra requires th a t any variable have only two possible states. Thus:
a statem ent m ay be TRUE or FALSE,
a c irc u it m ay be CLOSED or OPEN
a switch m ay be ON or OFF.
Representation and N otation:
Let: 0 = false, OFF, OPEN
1 = true, O N, CLOSED
O n -o ff electrical circu its provide a graphic picture o f the basic concepts o f Boolean algebra.
Consider the fo llow ing c irc u it diagram :

Switch A is shown in the OPEN (no cu rre n t flow ing) position. In notational form then,
A = 0
means: switch A is OPEN. S im ilarly, B = 1 means "sw itch B is CLOSED" (th a t is, current is flo w ­
ing through switch B).

Basic Operations:
AND;
consider two switches— A and B— connected in series:

A B

For the entire system to be O N, cu rre n t must flo w through both switch A and switch B. Using
the no ta tion th a t:
X means A N D
and Z represents the entire system then,
AX B= Z
There are fo u r possible states fo r the system; these possible states m ay be listed in a truth table:

A B z

0 0 0
1 0 0
0 1 0
1 1 1

45
(it m ay be h e lp fu l a t this p oin t to refer back to the o rig in a l discussion o f the A N D c irc u it to see
a verbal representation o f the tru th table).
OR:
Consider two switches (— A and B— ) connected in parallel:

For the system to be O N, curre nt may flow through either switch A or switch B. Using the no­
ta tio n th a t:
+ means OR
then,
A + B= Z
The tru th table fo r th is condition:

A B z

0 0 0
0 1 1
1 0 1
1 1 1

NOT:
This operation stands fo r the opposite state or condition. The notation is a line over the sym­
bol. Thus:

A means " n o t A "

or, in the b in a ry system:

T =0
0 = 1
The two negation relationships are known as the complementarity laws:

1) AX A = 0
A + A - I
g raphically:

a a

1) AXA = 0

46
 2) A + A = 1

M athem atical laws:

the com m utative law:
1) A + B= B+ A
2) A X B = BX A
the associative law
1) A + (B + C) = (A + B) + C
2) A X (B X C) = (A X B) X C
the d is trib u tiv e law:
(A X B) + (A X C) = A X ( B + C)
g raphically:

(A X B) + (A X C) = A X (B + C)

(The d istrib u tiv e law is p a rtic u la rly useful fo r c irc u it sim p lifica tio n ).

Decision-Making with Insufficient Information

In the preceding section o f this book, the basic elements o f com puter logic were discussed. Us­
ing these basic operations (the operations o f Boolean A lgebra) com plex logical problems can be
analyzed and solved in much the same way th a t com plex m athem atical problems are solved us­
ing the basic a rith m e tic tools— addition , subtraction, m u ltip lic a tio n and division.
The fin a l section o f this book is devoted to a series o f logical, decision-m aking, problems.
Each problem w ill be considered in some detail and each step in the solution o f the problem w ill
be discussed. The basic operations developed in the preceding sections w ill be applied to each
problem to obtain a com puter program fo r the solution o f the problem.
M IN IV A C 601, ju st like a large d ig ita l com puter, can be used to solve any logical problem
if the follo w in g two requirem ents are met:
1) if the com puter has s u ffic ie n t capacity fo r a ll conditions of the problem.
2) if the com puter is given s u ffic ie n t in fo rm a tio n fo r the solution o f the problem.

47
 The firs t requirem ent— s u ffic ie n t capacity— can generally be met by using a com puter o f a
size appropriate to the problem. T h eo retically, a t least, any logical problem can eventually be
solved on a com puter if the com puter is large enough.
The second requirem ent— s u ffic ie n t in fo rm a tio n — must be met if the com puter is to produce
a solution to a logical problem . A com puter cannot guess. If a ll in fo rm a tio n about a p a rtic u la r
problem is not available, assumptions about the missing in fo rm a tio n m ust be made and pre­
sented to the com puter. The com puter w ill then be able to solve the problem , basing its solution
on the facts and assumptions it has been given. W ith o u t s u ffic ie n t in fo rm a tio n — either in the form
o f fa cts or assumptions— a com puter cannot reach a solution to a problem. But then, neither can
a person.
For an exam ple o f a problem to be solved w ith in s u ffic ie n t in fo rm a tio n , let us return to the
problem o f ra inin g and wetness. Given only the fa c t th a t it is raining outside, we cannot reach a
conclusion as to w hether or not we are wet; nor can a com puter be program m ed to determ ine
w hether or not we are wet. The in fo rm a tio n is in su fficie n t.
However, if we assume th a t A) we are outside and B) we are not under cover, then we can con­
clude th a t we are wet. On the other hand, if we assume th a t A ) we are inside or B) we are under
cover, then we can conclude th a t we are not wet. Therefore, if we do not have s u ffic ie n t in fo rm a ­
tio n , we can m ake assumptions about our problem and arrive a t a series o f conclusions. S im ilarly,
we can program a com puter w ith both the in fo rm a tio n and the assumption to yield conditional
answers.
Fact: It is ra ining outside.

Assum ptions Conclusions

1 am outside
AND 1am wet
1 am not under cover

1am inside
OR 1 am not wet
1am under cover

W hen analyzing a problem fo r com puter solution, the distin ctio n between fa c t and assump­
tio n m ust always be kept in m ind. Otherwise, th e solution w ill be incorrect a n d /o r incomplete.

Simulation
Com plex problems fo r which a tria l-a n d -e rro r solution would be too costly or tim e-consum ­
ing are often simulated on a com puter. W hen a com puter sim ulation is used, the conditions of
the problem are presented to the com puter and a lte rn a tive solutions are tried. W hen the desired
sim ulated result is achieved, the m ethod o f solution or the actual sim ulated solution can be ap­
plied to the real w orld situation.
For exam ple, com puter sim ulation is used e ffe c tiv e ly to test the strength o f a rocket design
under the conditions o f outer space— w ith o u t a ctu a lly b u ild in g the design or sending it into space.
The rocket design is program m ed onto a com puter as a series o f com plex m athem atical re la tio n ­
ships in term s o f geom etrical shape, resistance to fric tio n , reaction to pressure, and m any other
im p o rta n t details. Once th is in fo rm a tio n has been given to the com puter, the com puter is con­
sidered to be a "s im u la te d rocke t".
To test the design, the engineers and scientists com m unicate varying conditions to the "s im ­
ulated ro c k e t", and the com puter com m unicates back the results o f the conditions. For instance,
the "s im u la te d ro c k e t" m ay be given a series o f accelerations; the results o f these accelerations
w ill be a series o f data which, when properly interpreted, w ill show w hat happened to the various
parts o f the rocket. S im ila rly, various m aterials can be tested under outer space conditions.
The result o f a sim u la tio n as involved as a rocket design is a vast am ount o f in fo rm a tio n
which m ust be c a re fu lly interpreted and analyzed before the fin a l answer can be achieved.

48
 For a sim ple exam ple o f some o f the advantages and problems o f a com puter sim ulation, let
us return to the rainstorm . A nd, let us now assume th a t we are w earing a new suit. The problem
co n fro n tin g us is, o f course, to stay dry. W e could easily enough stay inside, but th a t would be im ­
practical. On the other hand, we could step outside— which would be fin e unless it is raining. Ob­
viously, we do not w a nt to stay inside; but a t the same tim e, we do not w ant to risk ge ttin g our
new su it wet.
Using sim u la tio n , we can try various solutions to our problem w ith o u t risking ge ttin g wet.
Then we can analyze the results o f the sim ulation and fin d the best plan o f action.
Problem: I m ust not get wet
Possible Solutions:
A. I can stay inside
B. I can go outdoors
C. I can wear a raincoat
D. I can carry an um brella
W e m ust now fin d the results o f each possible solution:

Possible Solutions Results

A. Stay inside I w ill not get wet
B. Go outdoors I w ill not get wet if it is not raining or if I am un­
der cover.
C. W ea r a raincoat I w ill not get wet
D. C arry an um brella I w ill not get wet
T his is the form in w hich com puter results o f a sim ulation generally appear. It is le ft to the
persons involved to select the best possible solution. To choose the best possible solution, both the
solution and the result m ust be considered. For instance, these are our alternatives and w ith them
are other factors w hich w ill influence our fin a l decision:

Alternatives Results Other Factors

A. Stay inside. I w ill not get wet. have an appointm ent I m ust keep,
B. Go outdoors. I may not get wet. cannot a ffo rd a new suit,
C. W ea r a raincoat. I w ill not get wet. do not have a raincoat,
D. C arry an um brella. I w ill not get wet. feel s illy carrying an um brella.

A t this point, it is up to each o f us to decide on a best possible solution. A lte rn a tiv e B is not
the best possible solution, nor is a lte rn a tive C— each w ill probably cost us money. This leaves a
choice between A and D. If the a ppointm ent must be kept, then the only a lte rn a tive le ft is D. A l­
though we w ill feel s illy carrying an um brella, th is is the best possible solution to our problem — a
solution w hich we have achieved w ith o u t ruining a suit in the process.
Real problems are, o f course, much more com plex than our sim ple exam ple. However, our
exam ple had the same advantage as a large-scale sim ulation: we tested various situations before
a c tu a lly try in g them out, and thus saved ourselves the expense o f replacing the item under test if
one o f our solutions was incorrect. T h a t is, we now know th a t we w ill not get the new suit wet when
we go outdoors because we w ill be carrying an um brella.
O f course, we had to have the possible solutions in m ind before we could try them out. T his is
one o f the d iffic u ltie s o f a sim u lation technique: to fin d the best possible solution, a ll possible so­
lutions must be trie d . It is up to the persons involved in the problem to determ ine the alternatives.
The com puter w ill yield the results o f each a lte rn a tive , and then it is once again up to the per­
sons involved to decide upon the best possible a lternative.

Thoughts About Thinking

Before using M IN IV A C 601 to solve problems w hich require reasoning and 'th o u g h t" fo r
solution, we m ig h t pause fo r a m om ent to th in k about thinking.

49
 Does a com puter really " th in k " ? If by " t h in k " we mean does the com puter fo llo w a series
o f steps w hich are lo gica lly related and which lead the com puter to the solution o f a problem , our
answer m ust be yes. In this sense, the com puter d e fin ite ly does " th in k " .
If, on the other hand, we consider th in k in g in the sense o f creative hum an though t in which
ideas or thoughts are conceived, the com puter does not " th in k " . The com puter does not "k n o w "
w hat it is th in k in g . It has no consciousness. A com puter m ust (at least in itia lly ) be program m ed
to perform each step in the solution o f a problem . It can do only w hat the com puter operator or
program m er tells it to do. A lth o u g h we m ay not know the solution to a problem , we can decide
on the steps w hich should be taken to reach the solution and can com m unicate these steps to
the com puter.
The reason th a t the com puter can solve the problem which we m ay not be able to is th a t the
com puter remembers each step in d e ta il and does not become "co n fu se d ". W h ile we are ta lk in g
about the fo u rth step in a problem , we m ay have fo rgotten some detail o f the firs t step which is
necessary in order to arrive a t the correct solution. The com puter w ill not forget.
The p oint o f th is com m ent is th a t the com puter m ust be program m ed w ith absolute accu­
racy. The com puter does not know w hat it is to do; it has no way o f ju d g in g when it has made a
m istake. It w ill sim ply fo llo w the instructions it is given (by program m ing) and arrive a t the con­
clusion w hich is the result o f fo llo w in g these steps.
You m ay have noted the com m ent th a t the com puter must, at least initially, be to ld each
operation w hich it is to do. The word " in it ia lly " m ust be used because a com puter can be pro­
gram m ed to program itself. T h a t is, a com puter can be given instructions fo r fo llo w in g a series
o f steps or operations w hich w ill enable it to generate more instructions which it can then per­
form . A com puter, once it has been given such a series o f instructions and has been provided
w ith a d d itio n a l instructions which enable it to "e v a lu a te " the conclusions which it reaches
— to determ ine w hether the solutions reached meet previously supplied c rite ria — can begin to
" le a r n " concepts and relationships w hich the person o rig in a lly program m ing it did not know or,
a t least, did not com m unicate to it.
In m any ways, the process by w hich a com puter is educated— or ta u g h t to th in k — is
s im ila r to th a t used to teach a child. The com puter and ch ild must both be ta u g h t to gain in fo r­
m ation fro m th e ir environm ent, to in te rp re t it, and to use the in fo rm a tio n in association w ith data
drawn fro m past experience to arrive a t conclusions about a given situ a tio n or the solution to
a p a rtic u la r problem .
As you consider the program s w hich m ust be given to M IN IV A C 601 to enable it to solve the
problems and riddles in the fo llo w in g section, consider the in fo rm a tio n you would have to give a
person who had never before solved a problem to enable him to duplicate the perform ance o f the
com puter.
A s hum an beings, we can do m any things; b u t we sometimes fo rg e t th a t m any people and
much experience have gone in to the teaching process through which we have received and stored
in our memories the data and instructions w hich enable us to deal w ith our environm ent.

4. SOME COMPUTER PROBLEMS

A M ind Reading Program

As an in itia l exam ple o f the use o f logical operation to reach a decision, consider the fo llo w ­
ing situ atio n:
I f a person th in ks o f a num ber between 0 and 7, is it possible, by asking three "ye s-n o " ques­
tions, to determ ine the num ber o f w hich he is th in kin g ?
T his problem can, in fa c t, be solved w ith o u t resorting to m ind reading. Through a careful
choice o f questions we can elim in a te a ll b u t the correct answer through a series o f successive
logical operations. The three questions which provide the key to the problem are:
A. Is the num ber greater th an 3?
B. W hen the num ber is divided by 4, is the remainder greater than 1? (For example, if
6 is divided by 4 the answer is 1 plus a rem ainder o f 2; sim ila rly, 1 divided by 4
gives an answer o f 0 plus a rem ainder o f 1.)

50
 C. Is the num ber odd?
The process follow ed in solving th is problem can be analyzed using a flow chart. T his ch a rt
indicates each step in the decision process and shows the sequence which leads to the fin a l result.
A flo w ch a rt fo r the solution o f th is problem appears below:

The flo w ch a rt shows th a t question A — is the num ber greater than 3— separates the possi­
ble numbers in to two groups: those greater th a n 3 (4, 5, 6, 7) and those not greater than 3 (0,
1, 2, 3). Question B then separates each o f these groups into two groups, and question Cdeter­
mines w hich num ber in the appropriate set is not excluded— in other words, determ ines the num ­
ber o f w hich the person was th in k in g .
The conditions fo r each o f the num bers can be expressed as logical operations as follows:
Let:

A = the num ber is greater than 3

A = the num ber is not greater than 3
B= the rem ainder is greater than 1
B= the rem ainder is not greater than 1
C= the num ber is odd
C= the num ber is not odd.

Using these d e fin itio n s, we can express the conditions fo r each o f the possible numbers as follows:

A X B XC_= 7 A X B X C = 3
A X BXC~= 6 A X B X C = 2
A X B X C = 5 A X B X C = 1
A X B X 'C 'r r 4 A X B X CT= 0

T o program the problem fo r solution on a com puter, each relationship is connected into a c ir­
c u it representing the equation. Then these c irc u it representations o f the equations m ust be com­
bined fo r a com plete program .
The circu its fo r the ind ivid ua l equations are sim ply com binations o f the A N D and N O T c ir­
cuits. For example, the c irc u it representation fo r an answer o f 5 would be:

51
 A X B X C = 5

C om bining the eight equations above into a single c irc u it w ill give the follow ing:

N otice th a t the c irc u it has the same fan-shaped appearance as did the flow chart. In essence,
th is c irc u it diagram is sim ply a c irc u it representation o f the flo w chart. Each " g a te " (switch) rep­
resents an answer to one o f the questions.
However, if we exam ine th is c irc u it representation, we see th a t there are fo u r contacts to the
gate (switch) representing statem ent C, and th e re are two contacts to the gate representing state­
m ent B. If we were to program this c irc u it onto a com puter then, we would have to have one gate
which was capable o f accepting fo u r separate inputs. This can be done on the M IN IV A C , and the
c irc u it could be program m ed d ire c tly from the above c irc u it representation.
It is possible, though, th a t we would w ant to program this problem onto a com puter which
had only three inp ut contacts fo r each gate. W h a t would we do in this case? Or sim ila rly, w hat
would we do if we wanted to keep a contact free fo r additions to the circu it?
The com m utative law o f Boolean A lgebra referred to in the previous section can be applied
to th is p a rtic u la r problem . The com m utative law states that:
A X B is the same as B X A
So, we can restate fo u r o f our equations as follow s:

A X B X C j= 3 becomes A X CX B= 3
A X J3X C T= 2 becomes A X ~C X J3 = 2
AXBXC =1 becomes AX C XB = 1
T x^
a x o becomes A X 'C X T r z O
The c irc u it representation fo r the eight equations now becomes:

52
 T his is a sim ple exam ple o f the m a n ipulatio n o f circu its which professional program m ers do
to f i t a logical solution to the capa bilitie s o f a p a rtic u la r com puter. Some com plex logical prob­
lems can be program m ed fo r com puter solution o n ly by m aking m axim um use o f the com puter's
capacity in this fashion.
W e shall use a program based on this la tte r c irc u it representation fo r solving the problem on
M IN IV A C 601. The slide switches w ill be used fo r data in p u t and data storage; the relays w ill be
used fo r processing and operating storage; the rotary switch w ill be used fo r output. A pushbutton
w ill be used fo r the instru ction : move fin a l answer from operating storage (relays) to ou tp u t (ro­
tary switch).
In sum m ary then, the rules fo r using the " M in d Reading P rogram " are:
Let:

Slide Switch 4 represent the answer to question A

Slide Switch 5 represent the answer to question B
Slide Switch 6 represent the answer to question C.

A slide switch LEFT indicates a YES answer; a slide switch RIGHT indicates a NO answer. The
questions are:

A (slide switch 4): Is the num ber greater than 3?

B (slide switch 5): W hen the num ber is divided by 4, is the rem ainder greater than 1?
C (slide switch 6): Is the num ber odd?

A fte r the answer to each question has been com m unicated to the com puter through the slide
switches, pushing pushbutton 6 w ill in struct the com puter to indicate the fin a l answer on the deci­
mal in p u t-o u tp u t dial.

A program fo r the solution of this problem follows:

53
5 V /4 S 5 U /5 C 6 T /5 L D6/5K
4 R /4 A 5 F /6 F M + /6 Y D7/5G
4 B /4 - 6 F /6 - 6 X /D 1 7
5 C /5 A M —/D 1 8 D 1 6 /D 1 9
5 B /5 - D 1 8 /4 V D 0 /6 N
6 C /6 A 4 U /6 S D 1 /6 K
6 B /6 - 4 W /5 S D 2 /6 J
6 X /6 V 5 R /6 H D 3 /6 G
6 V /5 V 5 T /6 L D 4 /5 N
6 U /6 C 6 R /5 H D 5 /5 J

To use the program :

1. A sk a frie n d to th in k o f any num ber between 0 and 7.
2. A sk this person to answer "y e s " or " n o " to the follow ing questions about the number:
A. Is the num ber greater th a n 3?
B. W hen the num ber is divided by 4 , is the remainder greater than 1?
C. Is the num ber odd? ©
3. Indicate the answers to the questions as follows:
If the answer to question A is YES, move slide switch 4 to the LEFT.
If the answer to question A is NO, move slide switch 4 to the RIGHT.
If the answer to question B is YES, move slide switch 5 to the LEFT.
If the answer to question B is NO, move slide switch 5 to the RIGHT.
If the answer to question C is YES, move slide switch 6 to the LEFT.
If the answer to question C is NO, move slide switch 6 to the RIGHT.
4. Push pushbutton 6. The po in ter knob o f the rotary switch w ill tu rn to the num ber your
frie n d was th in k in g of.

Book V I— M IN IV A C GAMES— includes a v a ria tio n on this " M in d Reading P rogram " which
uses names in place o f numbers.

Quantity Recognition
As a fu rth e r exam ple o f the use o f logical operations, let us consider how a com puter could
be program m ed to recognize quantities. The problem w ill be to instruct the com puter to deter­
m ine the num ber o f slide switches w hich are O N (in the LEFT position) regardless o f which slide
switches are used.
Q u a n tity recognition circu its have m any a p plicatio ns in com puter logic. They can be used,
fo r instance, to perform additions, to indicate the results o f a game or a vote (win, lose, draw) or
to give d ire ct count fo r a q u a lity control system. Q u a n tity recognition circuits of th is type are
based on w hat m athem aticians call a "sym m e tric fu n ctio n la ttic e ". For our purposes, we may
consider the c irc u it as a series o f "y e s -n o " decisions.
Slide switches 2 through 6 w ill be used to provide input: a slide switch LEFT is O N , a slide
switch RIGHT is OFF. Lights 1 through 6 w ill be used fo r o u tp u t according to the follow ing conven­
tion:

L ig h t 1 ON indicates th a t 1 switch is ON
L ig h t 2 ON indicates th a t 2 switches are ON
L ig h t 3 ON indicates th a t 3 switches are ON
L ig h t 4 ON indicates th a t 4 switches are ON
L ig h t 5 ON indicates th a t 5 switches are ON
L ig h t 6 ON indicates th a t No switches are ON.

To program the com puter fo r q u a n tity recognition, we w ill firs t describe the possible condi­
tions as a series o f logical expressions:

54
 Let:
A = the condition th a t slide switch 2 is ON
B = the condition th a t slide switch 3 is ON
C = the condition th a t slide switch 4 is ON
D = the condition th a t slide switch 5 is ON
E = the condition th a t slide switch 6 is ON

Then, A = the condition th a t slide switch 2 is OFF, and so on.

Each o f the various com binations o f switches w hich yield a p a rtic u la r answer must now be
stated and then converted to a c irc u it and a program . For example, the case in which lig h t 6
comes O N — in d ica tin g th a t no switches are on— can only be expressed as:

A X B X C X D X f = light 6 ON
However, the possibilities fo r lig h t 1 com ing ON— in d ica tin g th a t one switch is O N — m ust be
stated as five separate expressions:

A X B X C X D X T = light 1 ON
A X B ^ X C " X D X F = light 1 ON
A X fX C ^ X D X E _ = light 1 ON
A X F X £ X D X F = light 1 ON
A X B X C X D X E —: light 1 ON
S im ila rly, there w ill be ten separate expressions fo r the possible conditions fo r lig h t 2 to come
O N — in d ica tin g th a t two switches are O N, ten expressions fo r lig h t 3, five expressions fo r lig h t 4
and the single expression

A X B X C X D X E = lig h t 5 ON

The program and c irc u it fo r the com plete solution must f i t a ll o f the conditions o f the prob­
lem as stated in the 32 logical expressions o f a ll possible cases. The fu ll c irc u it m ay be repre­
sented as follows:

55
 A

You w ill notice th a t this representational c irc u it has a lattice shape. M oreover, it is ele c tric­
a lly symmetrical about the low er-left to upper-right diagonal. Hence the m athem atical phrase
"sym m e tric fu n c tio n la ttic e " referred to earlier. This c irc u it representation shows the conditions
A and A , etc., as separate switches. In the a ctu a l program , o f course, both conditions fo r a given
switch w ill be wired onto a single switch using the norm ally-open and norm ally-closed contacts fo r
the A and A etc., conditions respectively.

A n exam ination o f the c irc u it representation shows th a t fo r conditions C, D, and E (corres­

ponding to slide switches 4, 5 and 6) more contacts are required than are available on M IN IV A C
601. Howover, by using the relays and co n tro llin g them by the slide switches, we w ill be able to
meet the requirem ents o f the problem.
The fin a l program fo r the solution o f the q u a n tity recognition problem appears below. This
program is based on the c irc u it representation above and uses the relays to supply the required
a d d itio n a l contacts.

56
 Program for Quantity Recognition:
1B / l — 6 F /6 - 5 A /2 G 5 R /6 H 5 K /4 N
2 B /2 — 6 + /6 S 4 A /2 K 5 W /6 L 5 G /4J
3 B /3 — 6 S /4 V 3 A /2 N 4 T /6 K 4 N /3 H
4 B /4 — 4 V /3 V 2 A /2 T 4R /6 J 4 J /2 L
5 B /5 — 3 V /2 V 1 A /3 J 6 N /5 L 4 G /2 H
6 B /6 — 2 U /2 C 6 A /3 N 6 J /5 H 4 K /2 S
2 F /3 F 2 C /3 C 6 T /5 V 6 G /4 H 3 J /3 K
3 F /4 F 3 U /4 C 6R /5S 6 K /4 L 3 G /2 T
4 F /5 F 4 C /5 C 5 U /5 T 5 J /4 K 2 J /2 K
5 F /6 F 4 U /6 C 5 U /4 S 5 N /3 L 2 N /2 R

To use this program :

1. Set slide switches 2 through 6 to the RIG HT (the slide switches are now OFF).
2. T u rn power ON. L ig h t 6 w ill come ON in d ica tin g th a t NO switches are ON.
3. Set any two o f the slide switches 2 through 6 to the LEFT. L ig h t 2 w ill come ON in d ica tin g th a t
two switches are ON.
4. Set a ll switches to the RIGHT again. Once again, lig h t 6 w ill come ON in d ica tin g th a t NO
switches are ON.
5. Select various com binations o f switches and note th a t in each case the o u tp u t lights in d i­
cate how m any switches are ON (to the LEFT). There are 120 possible com binations o f switches,
and in each case M IN IV A C 601 w ill recognize how m any switches are O N, regardless o f which
switches are used.

A Problem Involving Three Girls

Logical riddles provide apt illu stra tio n s o f problem s which can be program m ed fo r com puter
solution. For m any riddles, there are tw o basically d iffe re n t methods fo r handling the com puter
solution: the com puter can be program m ed w ith the conditions o f the problem so th a t various
solutions m ay be trie d and the com puter w ill indicate whether or not the attem pted solutions are
correct; or, the com puter can be program m ed to give the fin a l solution to the problem im m ediately.
As an exam ple o f a logical riddle, consider th e follow ing in fo rm a tio n about three girls named
C am ille, Sara and W anda:
If Sara shouldn 't, then W anda would. It is impossible th a t the statem ents "S ara should"
and "C a m ille could n o t" can both be true. I f W anda would, then Sara should and C am ille
could. Given th is info rm a tio n , the problem is:
Is the conclusion "C a m ille C o u ld " valid? M ore sim ply, we are asked, "C a n C a m ille "?
As we have m entioned, there are two methods o f handling this problem . W e can now follow
either one o f the two possible methods to solve th e problem.
1. W e can represent the conditions o f the problem as logical expressions, convert these
expressions into the appropriate c irc u it and program , and program the com puter w ith the
conditions so th a t by try in g various solutions th e com puter w ill indicate to us w hether or not our
solutions are correct. In this case, tru th or error are expressed in Boolean algebra statem ents which
are d ire c tly transferred to the com puter's contacts. The com puter w ill then test each o f our a t­
tem pted solutions against the logical conditions and indicate whether or not the solution is cor­
rect. Because o f the d ire ct Boolean nature o f th is method, we shall refer to this method as the
" Boolean" m ethod o f solution.
2. W e can represent the conditions o f the problem as logical expressions, convert these expres­
sions into the appropriate c irc u it and program , and program the com puter w ith the conditions so
th a t the com puter w ill a u to m a tic a lly give us the fin a l solution to the problem . In this case, the
logical expressions are transferred to contact networks w hich w ill force the relays to com ply w ith
the conditions o f the problem . W hen this program is turned on, the com puter w ill a u to m a tica lly
seek the correct answer. Because o f the problem conditions which have been program m ed onto

57
the com puter, only the correct result w ill be perm itted and this result w ill be com m unicated to
the operator. W e shall refer to this method as the "autom atic" method o f solution.
The basic diffe re nce between the two methods can be compared w ith the d ifference between
a therm om eter and a therm ostat. A therm om eter (the analogy o f the "B o o le a n " solution) w ill de­
term ine the tem perature and indicate w hat it is. A therm ostat (the analogy o f the "a u to m a tic "
solution) w ill not only determ ine the tem perature, b u t w ill also a tte m p t to "c o rre c t" the tem per­
ature to m atch the tem perature w hich has been "p ro g ra m m e d " into it. For instance, if the tem ­
perature were 65°F and we had set a therm ostat fo r 6 8 °, the fo llo w in g would occur fo r a th e r­
m om eter and a therm ostat:

1. The therm om eter w ould indicate th a t the tem perature was 65°F. It would be up to us to do
som ething about the fa c t th a t it was 3 ° cooler th a n w hat we wanted.
2. The therm ostat would determ ine th a t the tem perature was 65°F and would a u to m a tica lly
tu rn up the furnace u n til the tem perature reached 68°F.
Essentially then, the "B o o le a n " m ethod o f solution indicates tru th or error; the " a u to m a tic " m eth­
od o f solution determines the correct solution and indicates a fin a l result.
R eturning now to the problem o f the three g irls, we w ill firs t represent the problem conditions
as logical expressions, since both methods require this step. W e w ill then develop program s fo r
solution by each o f the two methods.

Logical Expressions fo r the Problem Conditions

From the conditions o f the problem , we can see th a t there are only two possible situations
fo r each g irl:

C a m ille could or could not

Sara should or should not
W anda would or would not

W hatever other conditions m ay exist, we can be sure th a t there are only two possible conditions fo r
each g irl.

W e can therefore represent the possible conditions fo r each g irl by the fo llo w in g convention:

C _ = C a m ille could
c =C am ille could not
s_=Sara should
s =Sara should not
w =W anda would
W - W anda w ould not

There are two ways to set up the logical expressions fo r each problem condition. W e can set
up e ithe r a " tr u th s ta te m e n t" or an "e rro r sta te m e n t". A " tr u th s ta te m e n t" describes the per­
missible conditions; an "e rro r s ta te m e n t" describes the conditions which are not permissible.
The two statem ents are, o f course, com plem entary.
The problem conditions can be stated as:

If Sara shouldn 't, then W anda would:

S X W = error or S + W = tru th
It is impossible th a t the statem ents "S ara sh o u ld " and "C a m ille could n o t" can both be true:

C X S = error or C + S = tru th
If W anda w ould ,then Sara should and C a m ille could:

W X (S + C) = error or W - f (C X S) = tru th

58
"Boolean solution"
Because it is sim pler in this case, we w ill set up a program fo r the "B o o le a n " solution fro m the
three "erro r" expressions above. Slide switches 1, 2 and 3 w ill be used to provide the contacts fo r
C am illa, Sara and W anda respectively. L ig h t 3 w ill be used as the error indicator. Since there are
three error conditions and only two contacts fo r lig h t 3, we w ill use the "C o m m o n " term inals to
provide the necessary extra contacts. (The notation fo r a connection to a common te rm in a l is
"1 co m ", " 2 c o m ," etc.)
The c irc u it diagram and program fo r the "B o o le a n " solution o f the problem appear below.
Each o f the logical expressions fo r the problem conditions can easily be traced on the c irc u it d ia­
gram.
Slide switch 1 represents C am ille
Slide switch 2 represents Sara
Slide switch 3 represents W anda
A slide switch LEFT indicates the positive condition: C am ille could, Sara should, W anda would.
A slide switch RIG HT indicates the negative condition : C am ille could not, Sara should not, W a n ­
da would not.
L ig h t 3 ON represents an error.

1 + /2 S
2 S /3 V
2T /3 S
3 T / 3 com
2R /1S
I T / 3 com
3 U /2 V
2 V /1 V
1W/2W
2 W / 3 com
3 A / 3 com
3 B /3 —

PROGRAM A N D C IR C U IT— BOOLEAN SOLUTION

To use the program fo r the "B o o le a n " solution:

T u rn power ON. Using slide switches 1, 2 and 3, try various com binations o f "S ara should",
"C a m ille could n o t", etc. u n til the error lig h t (lig h t 3) goes OFF. If the power is ON and lig h t 3
is OFF, slide switches 1, 2 and 3 are in d ica tin g a correct solution to the problem. (Note: fo r some
logical riddles there is more than one com plete solution)
The problem to be solved is "C a n C a m ille ? " To discover whether or not "C a m ille c a n ", the con­
d itions o f the other girls m ust also be considered.

"A utom atic" Solution:

The program and c irc u it diagram fo r the " a u to m a tic " solution are based on the truth expres­
sions stated above. The relays are program m ed so th a t they are forced to fo llo w a ll o f the problem
conditions to yield the fin a l result. The relay in d ica to r lights w ill be used to indicate the solution:
Relay lig h t 1 w ill indicate the condition of C am ille
Relay lig h t 2 w ill indicate the condition o f Sara
Relay lig h t 3 w ill indicate the conditon o f W anda
A relay lig h t ON w ill indicate the positive condition ; a relay lig h t OFF w ill indicate the negative
condition.

59
 Since there is a possibility o f more than one correct solution, pushbuttons 1, 2 and 3 are
w ired in so th a t a d d itio n a l in fo rm a tio n may be entered.
Pushbutton 1 represents C am ille
Pushbutton 2 represents Sara
Pushbutton 3 represents W anda
A pushbutton UP transm its no additio n al in fo rm a tio n to the com puter. A pushbutton DO W N,
however, supplies the positive condition (C am ille could, Sara should, W anda would).
W hen the power is turned O N , a correct solution w ill im m ediately appear in the relay indica­
to r lights. By pushing a pushbutton, we m ay enter a d d ition al positive in fo rm a tio n about a girl.
The relay in dica to r lights w ill then indicate the correct solution to the problem w ith the added
inform a tion .

14 - / 1 V
1Y /2Y
2 Y /3 Y
1 X /1 C
2X /2C
3X /3C
2 K /1 X
2L /2 +
2 N /3 C
3 G /1 H
1H /1 C
1G/2C
3 H /3 +
1F/2F
2F /3F
3 F /3 -

To use the program fo r the "a u to m a tic solution:

T u rn power ON. The solution to the problem appears in the relay indicator lights. A relay
in dica tor lig h t ON represents the positive condition: C am ille could, Sara should, W anda would.
A relay in d ica to r lig h t OFF represents the negative condition.
To add positive in fo rm a tio n about a g irl, push the appropriate pushbutton. The relay in d i­
cator lights w ill indicate the new solution, given this a d d ition al inform ation.

The Farmer, the Goose, the Corn and the W o lf

A t one tim e or another you have probably come acrossthe problem of the fa rm e r w ith a row
boat who has to cross a river in order to reach his home on the opposite shore. The fa rm e r has
w ith him a goose, some corn, and a w olf. The fa rm e r wants to bring both anim als and the corn
home w ith him , but the boat is sm all and w ill hold only one o f the three objects, in a ddition to
the fa rm e r, a t any one tim e. The real problem facing the fa rm e r is th a t if he leaves the goose
and the w o lf alone, the w o lf w ill eat the goose. A nd, if he leaves the corn and the goose alone, the
goose w ill eat the corn.
The fa rm e r's problem provides an excellent exam ple o f the advantages o f sim ulation solu­
tion. The fa rm e r wants to get across the river w ith a ll three o f his possessions, and solution o f his
problem by tria l-a n d -e rro r in the real world could be expensive fo r him . Should he m ake a mis­
take, he could lose both the goose and the corn.
However, if the fa rm e r has access to a com puter, hecan simulate his problem on the com ­
puter and try out various solutions u n til he reaches the rig h t one. In this way, the fa rm e r can use
a tria l-a n d -e rro r method w ith o u t any danger o f a c tu a lly losing either his goose or his corn.

60
 To solve the farm e r's problem on the M IN I VA C , we w ill state the problem as a series o f logi­
cal expressions. These relationships w ill then be converted to c irc u it representations and the
conditions of the problem w ill be program m ed onto the com puter fo r solution.
The Basic C onditions:
The farm er's problem can be stated as three basic conditions:
A. The goose and the w o lf cannot be le ft alone w ith o u t the farm er.
B. The goose and the corn cannot be le ft alone w ith o u t the farm er.
C. The boat w ill hold only one object in ad d itio n to the farm er.
The firs t two conditions— A and B— can be expressed as logical statem ents as follows:
Let:
F represent the presence o f the farm er
F~ represent the absence o f the farm er
G represent the presence o f the goose
G represent the absence o f the goose
W represent the presence o f the w olf
W represent the absence o f the w olf
C represent the presence o f the corn
C represent the absence o f the corn
E represent an error (loss o f one o f the farm er's possessions)
Then:
C ondition A (the goose and the w o lf cannot be le ft alone) can be expressed as:
GXWXF + GXWXF = E
C ondition B (the goose and the corn cannot be le ft alone) can be expressed as:
GX CX F+ GX CX F= E
C ondition C (the size lim ita tio n o f the boat) need not be stated as a logical expression. The
fin a l program w ill include appropriate c irc u itry so th a t if the boat is overloaded, only one object
w ill be recognized as being in the boat w ith the farm er.
P rogram m ing the S im ulation:

The two problem conditions— A and B— can be combined as follows: F X G X (W -f- C) + F

XGX(W + G = E
This is the error condition which must be program m ed onto the computer. W e w ill program the
com puter so th a t if either error condition is met, an " a la r m " lig h t (or an " e rr o r" light) w ill come
on.
I f we had only to program the error condition, the program would be a sim ple one, and this

F X G X ( W + C) + F X G X ( W + C) = E
C IR C U IT REPRESENTATION

61
 However, th is c irc u it represents only the processing and ou tp u t portions o f this program . W e
m ust also provide means o f in p u t and storage fo r the various solutions which we w ill a ttem pt. By
w irin g in the slide switches and pushbuttons fo r in p u t and storage, and using the relays fo r proc­
essing and the lights fo r ou tp u t, we w ill be able to program the entire problem fo r com puter solu­
tion.
W e shall use the fo llo w ing conventions in the program :
A n ou tp u t lig h t ON represents an object at home.
A relay ind icato r lig h t ON represents an object across the river.
A slide switch LEFT represents an object in the boat.
A slide switch RIG HT represents an object not in the boat.
L ig h t 5 ON represents the condition: " W o lf eats goose".
L ig h t 6 O N represents the condition: "Goose eats co rn ".
Pushbutton 6 represents m oving the boat towards home.
Pushbutton 1 represents m oving the boat from home.

The fo llo w in g components w ill be used to represent the objects in the problem :

O u tpu t L ight Relay Slide Switch

Farmer 1 1 1

Goose 2 2 2

W o lf 3 3 3

Corn 4 4 4

Program fo r the solution o f the Farm er's problem :

1A /1 J 1 U /2 V 2 N /3 N 3 N /4 N 5 B /5 -
1B / l — 1 V /1 X 2 W /3 V 3 W /4 V 6 B /6 —
1C /1G 1 X /6 X 3 A /3 J 4 A /4 J 6 Y /6 —
1E/1 U 1 Y /1 + 3 B /3 - 4 B /4 -
1F /2F 2 A /2 J 3C /3 G 4 C /4 G
1 H /2 H 2 B /2 - 3E /3U 4 E /4 U
1 K /2 K 2 E /2 U 3 F /4 F 4 F /4 -
1K /1 + 2 C /2 G 3 H /4 H 4 H /4 +
1 L /2 L 2 F /3 F 3 K /4 K 4 L /6 A
1 N /3 K 2 H /3 H 3 L /5 A

To use the program :

1. T u rn power ON. M a n u a lly push relays 1 through 4 to the LEFT. The relay indicator lights
w ill come ON to indicate th a t a ll objects are across the river. Move slide switches 1 through
4 to the RIGHT.
2. Place the fa rm e r in the boat by m oving slide switch 1 to the LEFT. (The fa rm e r m ust be
in the boat or the boat w ill not move)
3. Choose another object and place it in th e boat by pushing the appropriate slide switch
to the LEFT.
4. M ove the boat across the river by pushing pushbutton 6. If lights 5 and 6 rem ain OFF,
no error has been made.
5. C ontinue m oving the boat back and fo rth across the river w ith the d iffe re n t objects u n til
a ll fo u r objects are a t home (output lights 1 through 4 are O N .) Do not fo rg e t to remove
objects from the boat when they have reached the other side (by m oving the slide switch
to the RIGHT)

62
 6. If an error is made, reset the com puter by setting slide switches 1 through 4 to the RIGHT
and m a n u a lly setting relays 1 through 4 to the LEFT.

The Television Problem

As a fin a l exam ple o f a problem w hich can be program m ed fo r com puter solution, consider the
follow ing:
Barry, his w ife C lara, and th e ir children— D avid, Edward and Francis— are at home. It is 8:00
in the evening.

1. If Barry is w atching television, so is his wife.

2. Either Edward or Francis, or both, are w atching television.
3. Either C lara or David, but not both, is w atching television.
4. Edward and David are e ither both w atching or both not w atching television.
5. If Francis is w atching television, then Barry and Edward are also w atching television.

W h o is w atching television?
T his problem provides an excellent example o f a problem which is com plicated enough to be
confusing to a person a tte m p tin g to solve it. T h e problem conditions and the details o f the rela­
tionships are d iffic u lt to remember. Because o f its confusion-proof memory, the com puter can
easily handle a problem o f th is type.
To solve th is problem , we w ill state the problem conditions as logical expressions and devel­
op the appropriate c irc u it fo r a "B o o le a n " solution using a " t r u t h " lig h t to te ll us when we have
arrived a t the correct solution. This w ill, o f course, be a sim ulation technique since we w ill not
have to a c tu a lly drop in on the fa m ily to fin d out who is w atching television.

Logical expressions fo r the Problem Conditions:

Let:
B = the condition th a t Barry is w atching television
C = the condition th a t C lara is w atching television
D §g the condition th a t David is w atching television
E = the condition th a t Edward is w atching television
F m the condition th a t Francis is w atching television
Then:
B = the condition th a t Barry is N O T w atching television, etc.
A nd: T = tru th (permissible condition)

1. If Barry is w atching television, so is his w ife:

"B + C = T

2. Either Edward or Francis, or both, are w atching television:

E+ F= T

3. Either C lara or David, b u t not both is w atching television:

C X D + C X D = T

4. Edward and David are either both w atching or both not w atching television:

E X D + IX D = T

5. If Francis is w atching television, then Barry and Edward are also w atching television:

T + ( BXE) = T

Each o f these expressions can be represented as logical circuits in the follow ing manner:

63
 1. B+ C= T

T ---------------------•

2. E+ F= T

3. CXD + CXD = T

4. EXD + EXD = T

5. F + ( BXE) = T

64
The Program fo r a "B o o le a n " Solution:
T o convert the representational circu its to a program , we w ill use the fo llo w in g convention:
Let:
Slide switch 1 represent Barry
Slide switch 2 represent C lara
Slide switch 3 represent David
Slide switch 4 represent Edward
Slide switch 5 represent Francis
A slide switch LEFT indicates th a t the person is w atching television; a slide switch RIG HT indicates
th a t the person is N O T w atching television.
L ig h t 4 w ill indicate the correct solution: lig h t 4 O N indicates th a t the solution represented by the
slide switches is CORRECT; lig h t 4 OFF indicates th a t the solution is N O T CORRECT.
The com plete program fo r the "B o o le a n " Solution is:

1U /1S 2 W /3 U 4 B /4 -
1S/2S 3 R /4 R 4 U /5 V
1 T /2 R 3 S /3 V 4 U /1 V
2 R /2 V 3 T /4 T 4 V /4 +
2 U /3 W 4 A /4 S 5 W /1 U

To use the program :

Set slide switches 1 through 5 to the LEFT (th is indicates th a t no one is w atching television).
T u rn power O N. By m oving slide switches 1 th ro u g h 5, try various solutions to the problem . L ig h t
4 w ill come ON only when the correct solution has been indicated.

65
 BOOK IV

How Computers Do Arithmetic

1. THE B INARY NUMBER SYSTEM

In the previous books, we have fre q u e n tly made use o f the two-valued binary code. W e have
seen th a t this is a convenient representation fo r the o n -o ff nature o f the com puter's components.
Let us now consider the binary number system which is basic to the fu n ctio n in g o f a com puter system.

How Numbers Are Represented in the Binary System

A ll o f us have been ta u g h t to th in k in the decimal number system. The decim al system has ten
d iffe re n t symbols— zero through nine— which can be used alone and in com bination to represent
num erical concepts. The binary system, however, uses only two d iffe re n t symbols— zero and one
— to represent the same num erical concepts.
Let us consider a th re e -d ig it decim al number: fo r example, 547. W e know th a t th is a ctu a lly
means 500 plus 40 plus 7— or 5 tim es 100 plus 4 tim es 10 plus 7 tim es 1. In other words, we
know th a t the right-m ost d ig it represents the num ber of l's (units) in a num ber, the next d ig it to
the le ft represents the num ber o f 10's in the number, the next d ig it to the le ft represents the num ­
ber o f 100's in the num ber, and so on.
Now let us see why this is so. Suppose we s ta rt counting from 0. W e go 0, 1, 2, 3, 4, 5, 6, 7,
8, 9. Now we have used up a ll o f the symbols available in the decim al system. To continue, we
m ust invent a m ethod fo r handling numbers greater than 9. W e do th is by placing a 1 to the le ft
and a 0 in the fa r rig h t position (called the "lo w order p o s itio n ") so th a t we now have " 1 0 " which
represents one ten and no ones. S im ilarly, if we reach 99— which represents 9 tens and 9 ones—
and then add one more we m ust invent a m ethod fo r handling numbers larger than 99. W e do
this by placing a 1 to the le ft and 0's in the two low order positions, giving 100— 1 tim es 100 plus
0 tim es 10 plus 0 tim es 1.
So we have a sim ple rule fo r fo rm in g any num ber system: when we run out of symbols in the
low order position, we place a 1 in the next higher position and return the low order position to 0.
N otice w hat this means: if we have 70 symbols, the low order position represents l's ; the next
high order position represents 10's; the next high order position represents 100's; and so on. Each
tim e we move one position to the le ft the "w e ig h tin g " o f th a t position is ten tim es as great as the
" w e ig h tin g " o f the position to the right.
Now notice th a t if we had only 5 symbols— 0, 1, 2, 3, 4,— to w rite the next higher num ber
(5) we w ould have to place a 1 in the le ft position and a 0 in the rig h t to represent one 5 and
no l's .
The bina ry system is a c tu a lly sim pler than the decim al system. W e count 0, 1 and then we
have run out o f symbols. So the next symbol (to represent 2) m ust be 10 representing one 2 and no
l's . Three is w ritte n 11, representing one tim es 2 plus one tim es 1. Once again, we have run out
o f symbols. To w rite 4 we must w rite 100, representing 1 tim es 4 plus 0 tim es 2 plus 0 tim es 1.
Thus the rule fo r fo rm in g numbers in the binary system is: the low-order d ig it represents l's ,
the next le ft d ig it represents 2's, the next le ft d ig it represents 4's, and so fo rth . Let us now count
fro m 0 to 10 in bina ry numbers:

67
 :im a l B inary
0 0
1 1
2 10
3 11
4 100
5 101
6 110
7 111
8 1000
9 1001

THE FIRST TEN B IN A R Y NUMBERS A N D TH E IR D EC IM AL EQUIVALENTS

N otice th a t each tim e we add 1 to a b inary num ber— every tim e we count from one num ber
to the next— the fa r rig h t position changes, either from a 0 to a 1, or fro m a 1 to a 0. A nd, if the
right-m ost position is a 1 and 1 is added, the right-m ost position changes from a 1 to a 0 and the
next le ft position also changes— e ither from a 1 to a 0 or fro m a 0 to a 1.
Since a com puter works in the bin a ry system, it m ust then be capable o f changing— either
from a 1 to a 0, or fro m a 0 to a 1, i f it is to perform a rith m e tic. W e have already become ac­
quainted w ith a program w hich perm its the com puter to do just this— the single-inpu t flip -flo p
described in Book II.

Building a Single-Input Flip-Flop W ith Carry

However, to perform a rith m e tic the com puter m ust also be able to " c a rry " . T h a t is, i f the
com puter is to add 1 and 1 in binary, it must be able to indicate th a t the solution is: "1 and 1 are 0
and carry I ” . As a basic c irc u it to p erform a rith m e tic then, we w ill need a "s in g le -in p u t flip -flo p
w ith c a rry ".
In developing the sin gle -in p u t flip -flo p in Book II, we firs t b u ilt a "tw o -in p u t flip - flo p " (see
"T w o -In p u t M em ory Relay C irc u it" — Book II.) This tw o -in p u t flip -flo p used only one relay which
remembered e ithe r a zero or a one, depending upon which it had last been instructed to remem­
ber. T o m ake this tw o -in p u t flip -flo p "change its m in d ," we had only to instruct it to remember
the other num ber by pushing the appropriate pushbutton. However, to m ake the tw o -in p u t flip -flo p
change, we have to know a t w hat state it is in a t this m om ent in order to m ake it change.
The sing le -in pu t flip -flo p , however, does the changing a u to m a tic a lly by using a second relay
to rem em ber the state the flip -flo p is c u rre n tly in just long enough to allow the firs t relay (which is
actually rem em bering the 1 or the 0) to change. By using a second relay, then, we no longer need
to know w hat state the flip -flo p is in a t this m om ent in order to make it change. W e m erely in­
s tru ct the com puter to change, and it does so a u to m a tica lly because o f the single-input flip -flo p
which is program m ed onto it.
Let us go over again the functions o f the two relays o f the single-inpu t flip -flo p :
W hen the pushbutton is pushed, relay A instructs relay B to go to the same state relay A is
in a t th is m om ent. W hen the pushbutton is released, relay B instructs relay A to move to the oppo­
site state. Relay A does the actual rem em bering; relay B assists relay A when it is necessary fo r
relay A to change.
T o b u ild a single-input flip -flo p on the M IN IV A C , we w ill begin w ith two relays wired to be
self holding. (T hat is, each relay w ill rem ain in position once it is set.) The program fo r this is:

5 + /5 H 6G /6 F
5 H /6 H 5 C /6 C
5 G /5 F 6 C /6 -

W e w ill now add to the program so th a t when pushbutton 6 is pushed, relay 6 w ill go to the
same state th a t relay 5 is in. T his w ill require the fo llo w in g addition al connections:

68
 6H /6Y 5K/6F
6 X /5 L 5N/6E

A g a in we w ill add to the program so th a t when pushbutton 6 is released, relay 5 w ill go to the
opposite state from relay 6:

6 Z /6 L
6 K /5 E
6 N /5 F

A n d fin a lly , we w ill add an in d ica to r lig h t to show w hat state the flip -flo p is in:

6 G /6 A 6 B /6 -

W e now have a s in g le -in p u t-flip -flo p program m ed on M IN IV A C . The c irc u it diagram fo r this
is:

However, our sin gle-inp u t flip -flo p s till cannot be used fo r a rith m e tic : it does not provide a
" c a rry s ig n a l". T h a t is, it does not n o tify us when it has changed from 1 to zero. A nd, since all
contacts on both relays are used, we cannot arrange fo r a carry signal by m erely adding connec­
tions.
The solution to th is d iffic u lty is to b u ild a d iffe re n t version of the single-inpu t flip -flo p — one
which leaves contacts free fo r the program m ing of a carry signal. The version o f the flip -flo p which
we w ill use d iffe rs fro m the previous one in another respect as well. The stable states o f the flip -
flo p when the pushbutton is up are either both relays ON or both relays OFF. For this program , lig h t
5 is used to indicate the state o f the flip -flo p . L ig h t 6 is used only to provide a path fo r current when
relay 6 is "s h o rte d -o u t" (OFF); lig h t 6 is not used as an in d ica to r in this program . The program
and c irc u it diagram fo r th is flip -flo p are:

69
 S © -g .5 - 6, » © a — .s -

Program:
5 + /5 G 6J/6E
5 H /5 F 6E /6A
5 F/6F 6 B /6 —
5 C /6C 5 H /5 A
6 C /6 — 5B /5 —
5 J /6H 6 + /6 Y
6G /5E 6 X /6 H

SIN G LE-IN PU T FLIP-FLOP— SECOND VERSION

W e now have a single-inpu t flip -flo p w ith contacts free fo r a carry signal. W ith the power
OFF, program this flip -flo p on your M IN IV A C 601. Then:
1. T u rn power ON. N otice th a t both relays 5 and 6 are OFF.
2. Push and release pushbutton 6. N otice th a t both relays 5 and 6 are now ON.
3. Push pushbutton 6, but do not release. N otice th a t relay 5 is OFF; relay 6 is ON. W e w ant a
carry signal to occur N O W .
To add the carry signal, we w ill add the connections:
5 G /5 N 6 K /4 A
5 L /6 L 4 B /4 -
L ig h t 4 w ill now indicate th a t a carry is ta k in g place.
W e now have a basic com puter a rith m e tic device— the single-input flip -flo p w ith carry. The
c irc u it diagram looks like this:

-•6 -

6 + *~ 6Y 6X

C IR C U IT FOR SIN G LE-IN PU T FLIP-FLOP W IT H CARRY

70
 W e can use single-input flip -flo p s as "b u ild in g blo cks" fo r more com plex a rith m e tic devices.
For exam ple, three flip -flo p s can be wired together on your M IN IV A C to produce a counter which
operates as follows:

Light 4 Light 5 Ligh 6

I 1 J pushbutton
6
FLIP-FLOP carry FLIP-FLOP carry FLIP- F .OP
•
A B to A B C to B C

As pushbutton 6 is pushed and released again and again, this device w ill store in binary code the
num ber o f tim es the pushbutton has been pushed. The follow ing ch a rt shows w hat happens to
each stage o f the counter as the pushbutton is pushed and released, pushed and released: (the
ch a rt is read from rig h t to left).

Flip-Flop Carry Flip-Flop Carry Flip-Flop Pushbutton Times

A B to A B C to B C carry Pushed

0 NO 0 NO 0 UP 0
0 NO 0 NO 1 DOWN

0 NO 0 NO 1 UP 1
0 NO 1 YES 1 DOW N

0 NO 1 NO UP 2
0 NO 1 NO 1 DOWN

0 NO 1 NO 1 UP 3
1 YES 1 YES 1 DOWN

1 NO 0 NO UP 4
1 NO 0 NO 1 DOW N

1 NO 1 NO 1 UP 5
1 NO 1 YES 1 DOWN

1 NO 1 NO UP 6
1 NO 1 NO 1 DOWN

1 NO 1 NO 1 UP 7
1 YES 1 YES 1 DOW N

0 NO 0 NO 0 UP 0 or 8

B IN A R Y COUNTER— SEQUENCE OF OPERATIONS

T his counter, shown in the preceding block diagram , can be b u ilt on the M IN IV A C as the
fo llo w ing experim ent demonstrates:

Experiment 1: A Three-Bit Binary Counter

In th is Experim ent we w ill b uild a counter and use it to see how the binary num ber system
works.
A counter can be made up o f any num ber o f single-inpu t flip -flo p s w ith carry. One flip -flo p
is required fo r each b it to be handled by the counter. Because the M IN IV A C 601 has six relays, we

71
w ill be able to b u ild a th re e -b it counter (using tw o relays fo r each flip -flo p ). T his counter w ill allow
us to count fro m 0 to 8.
The c irc u it diagram w hich follow s is fo r the m iddle b it (or " 2 " b it) o f the counter. The "h ig h
o rd e r" and "lo w o rd e r" bits (th a t is, the " A " b it and the " 1 " bit) are identical except th a t no "c a rry
o u t" is required fo r the high order b it and the "c a rry in " to the low order b it is supplied through
pushbutton 6. The program fo r the entire counter is given below w ith the c irc u it diagram fo r the
m iddle stage:

Full program fo r B inary C ounter:

1A /2E 2 C /3 C 3 J /6K 5F /5 H
IB /1 — 2E/2J 3L/4 L 5 G /5 +
1C /2C 2 F /4 A 3N /4 H 5 J /6H
1 E/2 G 3A /6E 4 B /4 — 5 J /6 X
1F/2F 3B /3 — 4 C /5 C 5 L /6 L
1F/1H 3 C /4 C 4 E /4 J 5 N /6 H
1G /1 + 3E /4G 4F /5 A 6 A /6 F
1J/2H 3 F /3 H 5 B /5 — 6 B /6 —
1J/4K 3 F /4F 5 C /6 C 6 C /6 —
2 A /4 E 3G /3 + 5E /6G 6 E /6 J
2B /2 — 3 J /4 H 5F/6 F 6 Y /6 +

Procedure:
1. T u rn power ON. Push pushbutton 6 and release. B inary ou tp u t lights 4, 5 and 6 w ill read OFF,
OFF, ON or 001. The B inary C ounter is displaying a 1.
2. Push pushbutton 6 again and release. Binary o u tp u t lights 4, 5 and 6 w ill now read OFF, O N,
OFF or 010. The B inary C ounter is now displaying a 2.
3. Push and release pushbutton 6 several times. Each tim e a fte r releasing the pushbutton, note
the b in a ry num ber displayed in lights 4, 5 and 6. Compare your num bers w ith the table o f binary
numbers on page 66.
Y ou w ill notice th a t the counter repeats a fte r it has reached 7. T his is a result o f the fa c t th a t
the counter is made up o f 3 flip -flo p s . Each flip -flo p has 2 stable states (0 or 1), so the com bination
o f 3 flip -flo p s has 23 or 8 stable states. These 8 stable states represent the binary numbers 0
through 7.
As you push pushbutton 6 several tim es, notice th a t the low order flip -flo p (represented by
lig h t 6) changes each tim e the pushbutton is pushed and released. The low order flip -flo p repre­
sents 1's (units). The next flip -flo p to the le ft (represented by lig h t 5) changes every other tim e the
pushbutton is pushed and released. T his flip -flo p represents 2's. A nd o f course, the high order

72
flip -flo p (represented by lig h t 4) changes every fo u rth tim e the pushbutton is pushed and released;
the high order flip -flo p thus represents 4's.
You w ill notice th a t the m iddle flip -flo p changes on a regular basis; it changes every tim e the
low order flip -flo p changes fro m 1 to 0. The high order flip -flo p behaves s im ila rly— it changes
every tim e the m iddle flip -flo p changes from 1 to 0.
R eferring again to the table o f bin a ry numbers, we can see th a t this is a characteristic o f
bina ry numbers: adding 1 to a b inary num ber changes the low order b it— either fro m 0 to 1 or
from 1 to 0. A ll other, higher order, bits change i f and only if the b it to the rig h t is changing from
1 to 0.
Leave the B inary C ounter on the M IN IV A C fo r Experim ent 2.

Experiment 2: Counter Arithmetic

It is possible to perform additions very sim p ly on a B inary Counter. W e have already seen
th a t it is possible to add 1 to the num ber in the counter m erely by pushing and releasing push­
button 6. T his technique can easily be extended to any num ber w hich can be represented on
the counter.
For exam ple, to add 2 and 3: push and release pushbutton 6 twice; the counter w ill dis­
play the bin a ry num ber 2. Now push the pushbutton 3 more tim es and the counter w ill display
the answer— the b ina ry num ber 5. T his is e xa ctly w hat would have happened had you just pushed
the pushbutton 5 times.
T his is one way in w hich an ad d itio n can be perform ed in a com puter and is, in fa c t, a com ­
mon m ethod fo r handlin g certain kinds o f a d d itio n problems.
A t th is p o int your counter should be displaying bin a ry 5. Now push button 6 five more
times. The pushbutton has now been pushed ten tim es, and the counter should display binary equiv­
a len t o f decim al 10. However, bin a ry 10 is a fo u r-b it num ber (1010). Since we have only three
bits in our counter, we would expect to see only th e low order three bits o f the num ber displayed.
The b it pattern fo r the bina ry equivalent o f decim al 10 is 1010. The counter w ill display 010.

Experiment 3: Universal Counter Arithmetic

It is possible to w ire the Binary C ounter so th a t it w ill count down as well as up. In this way,
both additions and subtractions can be perform ed. The program below gives the necessary m o d ifi­
cations to the Binary C ounter fro m experim ents 1 and 2:

M o d ify the program from Experim ent 1 as follows:

Remove the connections:

6 K /3 J
4 K /1 J

Add the connections:

3 K /6 W 5 N /6 N
3 N /4 N 5 K /6 T
6 R /6 K 6S/3J
4 K /6 U 6 V /1 J

Procedure:
1. B uild the counter or, if you have the counter from experim ents 1 and 2 program m ed, make
the necessary m odifications. If slide switch 6 is LEFT, the counter w ill count UP; if slide switch 6
is RIGHT, the counter w ill count DOW N.
2. Subtract 4 fro m 7 as follows:
a. M ove slide switch 6 to the LEFT.
b. Enter the larger num ber (7) by pushing pushbutton 6 the correct num ber o f tim es. The
counter is now displaying the larger number.

73
 c. M ove slide switch 6 to the RIGHT.
d. Enter the sm aller num ber (4) by pushing pushbutton 6. The counter is now displaying the
answer: 7 — 4 = 3.
Note: a d dition is carried o u t exa ctly as in experim ent 2 when slide switch 6 is LEFT.

2. B IN A R Y A D D IT IO N

Rules for Binary Addition

Let us add together two bin a ry numbers: 11 010 (26) and 11100 (28):

11010
11100

sta rtin g from the low order (right-m ost) bit:

0 plus 0 equals 0.
1 plus 0 equals 1.
0 plus 1 equals 1.
1 plus 1 equals 0 and ca rry 1.
1 plus 1 plus carry 1 equals 1 and ca rry 1.
The answer is:
11010
11100
110110
As you can see, the a d d itio n tables fo r b in a ry a rith m e tic are very simple. They are:

0 0 1 1
+0 +_]_ +0 -M
0 1 1 10

The problem above illustrates how a b in a ry adder fo r a com puter must work. Each b it of
the adder m ust be able to accept three inputs:
a. the firs t num ber to be added
b. the second num ber to be added.
c. a carry fro m the b it to the righ t.
Each b it o f the adder m ust also be able to yield tw o outputs:
a. the sum
b. the carry.
Let us call the three inputs " A " , " B " , and "c a rry in " ; and let us call the two outputs "s u m "
and "c a rry o u t". N otice th a t the sum w ill equal 1 if and only if either one or three o f the inputs is
1. Carry out w ill equal 1 if and only if e ither two or three o f the inputs are 1.
If we restrict the in p u t to only A and B, we can b u ild the follow ing table showing inputs and
outputs o f an adder:

A B Carry Out Sum

00 0 0

0 1 0 1
1 0 0 1

1 1 1 0

74
 The c irc u it w hich a c tu a lly perform s this operation is called a "h a lf-a d d e r w ith c a rry ". A h a lf­
adder w ith ca rry is a device w ith two inputs and tw o outputs such th a t the sum w ill equal 1 if and
only if one and only one o f the inputs is 1. In other words, the sum c irc u it is an EITHER BUT N O T
BOTH c irc u it. The fo llo w in g experim ent illustrates the operation of a half-adder w ith carry.
Experiment 4: A Half-Adder with Carry
In th is experim ent we w ill b u ild and test a sim ple h a lf-adde r w ith carry. The c irc u it and pro­
gram are:

Program:
5 A /6 U 6 R /6 Z
5 B /5 - 6 T /6 X
6A /6S 6 T /6 V
6 B /6 - 6 Y /6 +

The entire program requires one pushbutton, one slide switch and two lights. The slide switch rep­
resents in p u t A ; the pushbutton represents in p u t B. The lights represent the tw o outputs: lig h t 6
represents the sum; lig h t 5 represents carry out.
Procedure:
1. T u rn power O N. M ove slide switch 6 to the R IG H T : in p u t A is now 0. Donot push pushbutton
6: in p u t B is 0. L ig h t 5 and 6 are OFF: sum is 0 and carry out is 0.
2; Push pushbutton 6 and hold it down. In p u t B is now 1. L ig h t 6 comes O N , in d ica tin g th a t the
sum is I , T his is equivalent to the m athem atical statem ent: 0 + 1 = 1.
3. Release pushbutton 6 and move slide switch 6 to the LEFT. Input A is now 1; in p u t B is 0.
L ig h t 6 comes O N , in d ica tin g th a t the sum is 1:1 + 0 = 1 .
4. W ith the slide switch LEFT, push pushbutton 6. L ig h t 6 goesOFF and lig h t 5 comes ON.
This represents the m a them atical statem ent: 1 + 1 = 0 and ca rry 1.
I f we now have three inputs (A, B, and ca rry in) and two outputs, the table o f values looks
like this:
Carry In Carry Out Sum

75
 The c irc u it w hich a c tu a lly perform s th is operation is called a " f u ll adder w ith c a rry ", and is
made up o f two half-adders as follows:

SUM

Like the single-inpu t flip -flo p w ith carry, the half-adder w ith carry is a basic com puter c ir­
cu it. H alf-adders can be used as "b u ild in g b lo cks" fo r more com plex a rith m e tic devices— an ex­
am ple o f w hich is the fu ll adder.

Experiment 5: A Full Adder

In this experim ent we w ill m o d ify the c irc u it o f experim ent 4 to build a fu ll adder. A fu ll
adder is made up o f two half-adders and a sim ple OR c irc u it. The c irc u it and program fo r a fu ll
adder are:

5 A /6 U
5 B /5 —
5C/6S
5 F /5 —
5G/6J
5 H /5 +
5J/6G
6 A /6 H
6 B /6 —
6 J /6 L
6 K /6 U
6 R /6 Z
6 T /6X
6 T /6V
6 Y /6 +

76
Relay 6 represents a ca rry in from a previous stage. Relay 5 represents the sum o f inputs A and
B; this sum w ill be added to the carry in to give the sum fo r the fu ll adder. N otice th a t a carry
signal m ay come from either o f two places: either from the half-adder representing the slide
switch and pushbutton (inp ut A and input B), or from the half-adder representing the carry in (re­
lay 6) and the sum o f the inp u t (relay 5). Either half-adde r can produce a carry signal.

Procedure:

1. T u rn power ON. M ove slide switch 6 to the RIGHT; Input A is 0. Do not push pushbutton 6; in ­
put B is 0. Relay 6 is OFF; carry in is 0. Lights 5 and 6 are OFF. This is equivalent to the m athem at­
ical expression: 0 + 0 + 0 = 0.
2. M ove slide switch 6 to the LEFT; input A is now 1. L ig h t 6 comes O N: 1 + 0 + 0 = 1 .
3. M a n u a lly move relay 6 to the le ft to indicate th a t carry in is 1. L ig h t 6 goes OFF and lig h t 5
comes O N : 1 + 0 + 1 = 0 and ca rry 1.
4. T ry other com binations o f inputs by moving slide switch 6, pushing pushbutton 6, a n d /o r
m a nu a lly m oving relay 6.
W e have now seen how two half-adders can be combined to produce a fu ll adder. Full adders
can in tu rn be used as b u ildin g blocks fo r adders capable o f accepting greater inputs and yie ld ­
ing greater outputs. The follow in g experim ent illustrates the com bination o f three fu ll adders to
produce a "th re e -b it adder".

Experiment 6: A Three-Bit Adder

In th is experim ent we w ill b u ild a device w hich is capable o f adding together tw o 3 -b it b i­
nary numbers and displaying a result o f not more than 4 binary bits. The program fo r the 3 -b it
adder is:

2 C /3G 3L /4 X 5K /6 W
2 F/3F 4 A /4 V 5 L /6 A
2G /4S 4 B /4 — 5 N /6 U
2 H /4 Y 4E /5F 5 R /6 T
2 K /4 W 4F/5S 5T /6 R
2 L /2 + 4 G /5 U 5 T /5 V
2 N /4 U 4 H /4 N 5 Y /6 V
3 A /M 1 0 4 K /5 Z 6B /6—
3B /3 — 4 L /5 A 6E /6—
3 C /4G 4 N /5 X 6 G /6R
3F/4E 4 Y /5 Y 6 H /6 X
3 G /3 N 5 B /5 — 6J/6K
3 H /4 Z 5C /6K 6 L /6 Z
3J /4T 5F/6E 6 N /6 T
3K /4R 5G /6S 6 V /6 Y
3 K /M 1 0 5 H /5 + 6 Y /6 +

The three fu ll adders which m ake up th is c irc u it are very much like the single adder o f ex­
perim ent 5. However, the c irc u it has been m odified s lig h tly so th a t three adders use only five re­
lays. The high order stage o f the adder is wired so th a t the carry from it is displayed by lig h t 3.
T his allows us to display a 4 -b it result w ith a 3 -b it adder.

Procedure:
1. T u rn power ON. Enter a binary num ber in the adder using slide switches 4, 5 and 6. For ex­
am ple, to enter bin a ry 4 (100) set slide switch 4 LEFT, slide switch 5 RIG HT and slide switch 6
RIGHT.
2. Enter another b ina ry num ber using pushbuttons 4, 5 and 6. For exam ple, to enter binary 5
(101) push button 4 and 6; do not push pushbutton 5.
3. The answer w ill be displayed in bin a ry o u tp u t lights 3, 4, 5 and 6. For example, if lights 3
and 6 are ON and ligh ts 4 and 5 are OFF, the answer is 1001 (9).

77
 3. H O W COMPUTERS SUBTRACT

W e have now discussed two methods o f com puter addition. The firs t method was a counter
which was dependent upon the num ber o f tim es an in p u t button had been pushed. The second
m ethod used three fu ll adders w ith ca rry which allowed us to enter two 3 -b it numbers sim ultane­
ously.

Two's Complement Arithmetic

S im ila rly, there are two methods o f com puter subtraction. The sim pler method is the "dow n
c o u n te r" dem onstrated in experim ent 3. However, in spite o f its sim p licity, the down counter is
seldom used in a com puter. The more common method o f subtraction used in com puters is called
"tw o 's com plem ent a rith m e tic ". Two's com plem ent a rith m e tic in the binary system is the same as
ten's com plem ent a rith m e tic in the decim al system.
Two's com plem ent a rith m e tic in the binary system works as follows: the subtractor is "c o m ­
plem ented ". T his means th a t every 1 is replaced by a 0, and every 0 is replaced by a 1. The com ­
plem ented subtractor and the subtrahend are then added. If a carry occurs out o f the high order
position, it is added to the low order position. The resulting sum is the proper d ifference between
the two numbers.
As an exam ple, let us subtract b in a ry 4 from b inary 13. W e firs t w rite down 1101 (13) and
below it we w rite 1011— w hich is the complement o f 0100 (4):

13 1101 ^ 1101
-4 -0 1 0 0 1011

decim al bin ary com plem ented subtractor

W e now add in the usual bin ary m anner except th a t if a carry occurs o u t o f the high order b it,
we w ill add it to the low order b it:

1 1 0 1
1 0 1 1
J_ 0
1 0
1 0
© 1
1 0 0 1

A great advantage o f two's com plem ent a rith e m e tic is the fa c t th a t there is an autom atic
bookkeeping m ethod fo r in d ica tin g the sign o f the result. If a ca rry occurs out o f the high order
position, the subtraction has resulted in a positive number. On the other hand, if the ca rry does
not occur, the result is a negative number.
The experim ent w hich follows illu strates the operation o f two's com plem ent subtraction on
the M IN IV A C .

Experiment 7. A Three-Bit Subtractor

In th is experim ent we w ill program the M IN IV A C to perform subtraction by the two's com ­
plem ent m ethod. T his w ill be done by m o difying the th re e -b it adder o f experim ent 6 as follows:
The carry out fro m the high order stage (represented by the w ire connected to lig h t 3) w ill be
connected to the carry in o f the low order stage. In th is way, if a high order carry occurs it w ill be
read into the low order stage, just as was illu stra te d above.
The adder w ill also be m odified so th a t the subtractor is a u to m a tic a lly com plem ented. To
bu ild the th re e -b it subtractor, make the fo llo w in g changes to the program fro m experim ent 6:

78
REMOVE these connections:
3 A / M 10 4 K /5 Z
3 L /4 X 6 H /6 X
3 H /4 Z 6 L /6 Z
4 N /5 X

Now A D D these connections:

6 F /M 1 0 4 K /5 X
3 L /4 Z 6 C /6 F
3 H /4 X 6 H /6 Z
4 N /5 Z 6 L /6 X

To use the program :

The subtractor is used in the same m anner as was the adder.
A th re e -b it bina ry num ber (the subtrahend) is entered on slide switches 4, 5, and 6. The subtractor
is then entered on pushbutton 4, 5 and 6. The answer w ill be displayed in the binary ou tp u t lights.
If we a tte m p t to subtract a larger num ber fro m a sm aller number, we w ill have an incorrect
answer. This condition can be detected by observing the carry in relay o f the low order stage. If this
relay fa ils to p u ll in when the subtraction takes place, we have attem pted an incorrect subtrac­
tion.
A n exam ple:
To subtract bin a ry 2 (010) fro m bin ary 5(101):
1. Enter the subtrahend (101) by m oving slide switches 4 and 6 to the LEFT, leaving slide switch
5 RIGHT.
2. Enter the subtractor (010) by pushing pushbutton 5 D O W N and leaving pushbuttons 4 and 6
UP. W e need not enter the complem ented subtractor; the program a u to m a tica lly com plements the
subtractor.
3. The answer appears in the bin ary o u tp u t lights: lig h t 4 is OFF, and lights 5 and 6 are ON. T his
is read as 011, w hich is b in ary 3 : 5 — 2 = 3.
Had we attem pted to enter bin ary 7 (111) in the pushbuttons, relay 6— the carry in relay fo r the
low order stage— w ould not have pulled in and th e answer displayed in the o u tp u t lights w ould be
incorrect.

4. COMPUTER M U L T IP L IC A T IO N

Binary Multiplication
In the decim al num ber system we can m u ltip ly by ten m erely by placing a zero to the rig h t
o f the number. S im ila rly, in the b in a ry system we can m u ltip ly by two by placing a zero to the rig h t
o f the num ber.
To m u ltip ly in the bin ary system, we need o n ly rem ember th a t:

0 X 0 = 0 0 X 1 = 0
1X 0=0 1X1=1

and:

10 (2) X 1 = 1 0 (2 )
100 (4) X 1 = 100 (4)
1000 (8) X 1 = 1000 (8), etc.

In order to m u ltip ly in the b in a ry system on a com puter we w ill need a device known as a
shift register. A s h ift register shifts each b it o f a num ber one space to the le ft which, o f course, rep-

79
resents m u ltip lic a tio n by two. M u ltip lic a tio n by 4 is accomplished by s h iftin g le ft tw ice; m u lti­
p lic a tio n by 8 is accom plished by s h iftin g le ft three tim es, and so on.
The fo llo w in g experim ent dem onstrates how a com puter can m u ltip ly using a m anual shift
register.

Experiment 8: The Shifting Operation

In this experim ent we w ill m u ltip ly and divide numbers by powers of tw o by m anually s h iftin g
the numbers rig h t and left. The device which we w ill use fo r this operation is called an accumu­
lator. In th is experim ent, we w ill be concerned only w ith its s h iftin g abilities.
The program fo r the accum ulator is:

1A / I S 2 H /4 Y 3T /4 R 5L /6 A
1B / 1— 2K /4 W 4 A /4 V 5 N /6 U
1C /3U 2 L /2 + 4 B /4 — 5R /6 T
1 F /2 F 2L/2S 4E /5F 5 T /6 R
1G /2 W 2 N /4 U 4F/5S 5 T /5 V
1 H /2 A 2 S /2 V 4 G /5 U 5 Y /6 V
1J/2U 3A /3S 4 H /4 N 6 B /6 -
1K /1T 3 B /3 — 4 K /5 Z 6 C /M 1 0
1L/2R 3C /4G 4 L /5 A 6E /6—
1 N /1 R 3F /4E 4 N /5 X 6G /6R
1 N /2 T 3 G /3 N 4 Y /5 Y 6 H /6 X
1U / M 1 0 3 H /4 Z 5B /5 — 6J /6K
1V /1 T 3J /3R 5C /6K 6 L /6 Z
2 B /2 — 3 J /4T 5F/6E 6 N /6 T
2 C /3G 3 K /4 R 5G /6S 6 V /6 Y
2F/3 F 3L /4 X 5 H /5 + 6 Y /6 +
2G /4S 3 V /3 T 5K /6 W

To m u ltip ly by two, we w ill perform a le ft shift operation as follows:

1. Enter a bin a ry num ber by setting the slide switches appropriately. (A slide switch RIGHT
represents 0 ; a slide switch LEFT represents 1.) The num ber is now displayed in the binary
o u tp u t lights.
2. Beginning a t the le ft end o f the console, set slide switch 1 to whatever position slide switch
2 is in; set slide switch 2 to w hatever position slide switch 3 is in; set slide switch 3 to w h a t­
ever position slide switch 4 is in. C ontinue u n til slide switches 1 through 5 are in the positions
th a t slide switches 2 through 6 were previously in. Set slide switch 6 to zero. The result w ill be
displayed in the bin ary ou tp u t lights.
To m u ltip ly by fo u r, enter a bina ry num ber on the slide switches and perform two consecutive le ft
s h ift operations. To m u ltip ly by eight, enter a th re e -b it bin a ry num ber in slide switches 4, 5 and
6 and p erform three consecutive le ft s h ift operations.
To divide by two, we w ill perform a right shift operation as follows:
1. Enter a bin ary num ber by setting the slide switches appropriately. (N ote: it is usually
more convenient to move the binary point— w hich corresponds to the decimal point— to the le ft
fo r division. For instance, we m ig h t place the bin a ry p oint between slide switches 3 and 4, in
w hich case the dividend w ould be entered on slide switches, 1, 2 and 3.)
2. Beginning a t the right end o f the console, set slide switch 6 to w hatever position slide
switch 5 is in; set slide switch 5 to w hatever position slide switch 4 is in. C ontinue u n til slide
switches 2 through 6 are in the positions slide switches 1 through 5 were previously in. Set
slide switch 1 to zero. The result w ill be displayed in the bin a ry o u tp u t lights. A n y th in g to
the right o f the b in a ry p o int is remainder.
T o divide by fo u r, enter a bin ary num ber and perform two consecutive rig h t s h ift operations.
T o divide by eight, enter a bina ry num ber and perform three consecutive rig h t s h ift operations.
This m anual s h ift register dem onstrates m u ltip lic a tio n and division by powers o f tw o on an accu­
m ulator. Large d ig ita l com puters perform shifts exactly like these, b u t the s h iftin g is done auto­
m a tica lly. (See appendix to Book IV: A u to m a tic S h ift Register).
Leave th is program w ired on your M IN IV A C fo r experim ent 9.

80
Multiplication by Numbers other than Powers of Two

M u ltip lic a tio n in the bina ry system can be done exactly as it is in the decim al system by m u l­
tip ly in g by each b it o f the m u ltip lie r and adding. For exam ple:

1011
l_l
1011
1011
100001

This same operation can also be done in a s lig h tly d iffe re n t m anner. For instance, let us m u ltip ly
111 ( 7 ) X 110 (6)
Let:

1 1 1 = M u ltip lic a n d
110 = M u ltip lie r
Then: 111 X 110 = ?

First, we m u ltip ly the m u ltip lic a n d by the left-most b it o f the m u ltip lie r:

1 1 1 X 1 = 111

Now, we s h ift le ft one by adding a 0 to the right:

1110

To this, we add the product o f the m u ltip lic a n d and the second left-m ost b it o f the m u ltip lie r:

1110
1 1 1 X 1 = 111
10101

Once again, we s h ift le ft one by adding a 0 to the right:

101010
A g ain , we add the product o f the m u ltip lic a n d and the next left-m ost b it o f the m u ltip lie r:

101010
111X0= 000
101010

which is the fin a l result:

111 X 1 1 0 = 1 0 1 0 1 0 o r 7 X 6 = 42

M u ltip lic a tio n on a com puter is carried out in ju st th is fashion, using an accumulator. The accu­
m u la to r acts just like a piece o f paper on which we record the steps which were illu stra te d above.
The fo llo w in g experim ent demonstrates m u ltip lic a tio n on a com puter using an accum ulator.

Experiment 9: The Accumulator

W ith o u t changing the program fro m experim ent 8 we w ill now perform m u ltip lic a tio n by num ­
bers other th a n powers o f two. M u ltip lic a tio n s w ill be perform ed by successive additions and shifts.

81
The accumulator, w hich is wired on the console from experim ent 8, is a device which is capable o f
p erform in g both additions and shifts.

As an exam ple, le us m u ltip ly :

6 X 10 or in binary: 110 X 1010.

Let: 110 = M u ltip lic a n d
1010 = M u ltip lie r

The left-m ost b it o f the m u ltip lie r is 1, so we w ill enter into the accum ulator:

110 X 1 or 110.

W e w ill do this by pushing the appropriate pushbuttons (in this case, pushbuttons 4 and 5.)
The slide switches should a ll be RIGHT.
As we push the pushbuttons, the o u tp u t lights display the num ber— 000110. Record this
num ber on a piece o f paper and release the pushbuttons.
Enter the num ber w hich you recorded in th e slide switches and perform a le ft s h ift one b it
as described in experim ent 8. The o u tp u t lights w ill now read 001100.
The next left-m ost b it o f the m u ltip lie r is 0. W e w ill add nothing to the accum ulator, but
w ill s h ift le ft one b it. The o u tp u t lights now show 011000.
The next left-m ost b it o f the m u ltip lie r is 1, so we w ill add— through the pushbuttons:

1 1 0 X 1 or 110

The ou tp u t lights now show 011110. Record th is num ber on a piece o f paper and release
the pushbuttons. Enter the num ber which you just recorded in the slide switches and s h ift
le ft one bit. The o u tp u t lights now show 11100.
The low order b it o f the m u ltip lie r is 0, so it is not necessary to add anything more into the
accum ulator. The bina ry o u tp u t lights now show the fin a l result: 11100 or 60.

110 X 1010 = 111100 o r 6 X 10 = 60

Select other numbers and m u ltip ly them on the accum ulator. The lim it o f this accum ulator is
111111 (63).

5. D IV IS IO N ON A COMPUTER

Binary Division
D ivision in the bin ary system is perform ed exactly as in the decim al system. For example, to
divide 11011 (27) by 101 (5):

W
101 | 11011
101
111
101
10

11011 — 101 = 101 plus a rem ainder o f 10

or
27 -r- 5 = 5 plus a rem ainder o f 2.

82
 Perform ing th is sequence o f operations on a com puter can, however, be d iffic u lt. In fa ct,
m any com puters have no provision fo r a u to m a tic division. For such a com puter, a subroutine is
w ritte n so th a t the com puter can carry out the individual steps o f a division. T h a t is, the com­
puter is given a program of instructions which, when followed, pe rm it the com puter to divide.
The follo w in g experim ent demonstrates the steps involved in certain types o f com puter d iv i­
sion. The actual program required is a com bination s h ift register and subtractor.

Experiment 10: Division

This experim ent demonstrates division on the M IN IV A C w ith your help. The program and
c irc u it diagram fo r a com bination s h ift register and subtractor which w ill allow us to divide are:

1 A /1 V 2F/3F 3 G /3 N 4 H /4 N
1B / l — 2 L /2 + 3 K /M 1 0 4S /4 +
1R /3K 2R /3 T 3 R /6G 5 A /5 R
1S/2G 2S/4F 3S/5G 5 B /5 -
1 T /3 J 2T /3 R 3 T /6 N 5 C /6 K
1 U /2 N 2 T /2 V 3 U /5 N 5F/6E
1W /2K 2 U /4 G 3 V /3 Y 5 H /5 +
1 X /3 H 2 X /4 K 3 W /5 K 5 S /5 +
1 Y /2H 2 Y /6 S 3 X /6 L 6 A /6 R
1 Y /2 Y 2 Z /4 N 3 Y /3 + 6B /6 —
1 Z/3 L 3A /5 L 3 Z /6 H 6E /6—
2 A /4 L 3 B /3 — 4A /4 R 6 F /M 1 0
2 B /2 — 3 C /4G 4 B /4 — 6 J /6K
2C /3G 3F /4E 4E /5F 6 S /6 +

W e w ill now divide 011011 (27) by 101 (5) as follows:

W rite the dividend on a piece o f paper, w ith a vertical line between the th ird and fo u rth
digits. The ve rtica l line is always placed so th a t the divisor is greater than w hatever is to
the le ft o f the line. Now w rite the divisor under the le ft end o f the dividend:

011 O il

101
Place the dividend on the slide switches. The dividend is now displayed in the o u tp u t lights
as 011011.
The divisor is greater than the left-m ost three bits o f the dividend. Therefore, s h ift le ft one
bit.
W e now have:

110 110

101
Subtract the divisor from the dividend by entering the divisor on pushbuttons 1, 2 and 3. The
result w ill appear in the o u tp u t lights:

110 110

101

001 110
Place this result in the slide switches. Since the subtraction proceeded properly, we w ill add
1 to the low order b it and place this new result in the slide switches:

83
 001

To go on to the next step, s h ift le ft one b it and again w rite the divisor below the left-m ost
three bits:

O il 110

101
The divisor is greater th an the le ft-m ost three bits. Therefore, we w ill s h ift le ft one bit:

111 100

101
W e can now subtract by entering the divisor in pushbuttons 1, 2 and 3 and entering the re­
sult on the slide switches:

111 100

101

010 100
Since the subtraction proceeded properly, we w ill add 1 to the low order b it to yield:

010 101

The division has now been com pleted: There have been as m any shifts le ft as there were bits
to the rig h t o f the ve rtical line. The answer is to the rig h t o f the vertical line; the rem ainder is
to the le ft o f the ve rtical line:

rem ainder answer

01 0 101

11011 H- 101 = 101 plus a rem ainder o f 10.

or
27 5 = 5 plus a rem ainder o f 2.

6. CONVERSIONS

Since com puters w ork in the b in a ry system and we generally w ork in the decim al system,
an im p o rta n t fu n c tio n o f a com puter is conversion between the two systems. The fo llo w in g experi­
ments give the program s fo r conversions between the bin a ry and decim al systems.

Experiment 11. Decimal to Binary Converter

Conversion fro m decim al to b in a ry is a p a rtic u la rly im p o rta n t fu n ctio n fo r in p u t devices.
There are tw o methods fo r conversion com m only in use in large d ig ita l computers. The firs t method
is a counting scheme w hich is p a rtic u la rly suitable fo r conversion o f numbers on a punched card.
W ith th is m ethod the com puter generates a binary num ber on the basis o f the length o f tim e
which elapses before a decim al code is observed.
As an exam ple o f how the counting m ethod works, consider the follow ing:

84
 A card is fed through a card reader a t a constant rate o f speed. W hen the edge o f the card is
detected in the reader, the counter in the com puter comes on. W hen a punch in the card is de­
tected, the counter goes o ff and a b inary num ber corresponding to the decim al num ber
punched on the card is displayed in the counter. This method depends on the tim e lapse be­
tween the card edge and the punch.
It is possible to b uild a converter o f this type fo r the M IN IV A C , using the rotary switch to
actuate a counter. Since you are fa m ilia r w ith the rotary switch m echanism and a counter, you
w ill be able to b u ild such a device yourself.
The second method fo r conversion fro m decim al to binary is essentially a decoding scheme. In
this method, the com puter is program m ed to recognize a decim al num ber by the location o f the in­
p u t pulse; a series o f d ire ct contacts then convert th is pulse to the appropriate binary number.
The follow in g program mechanizes such a method. Setting the rotary switch to any num ber
from 0 through 10 w ill cause the equivalent binary num ber to appear in the ou tp u t lights.
Program fo r Decimal to Binary Converter:
1 E /D 1 0 3 B /3 — 4 H /4 L 6 B /6 —
1 F /2 F 3 E /D 7 4 H /5 H 6 E /D 3
1G /3com 3 F /4 F 4 K /4 c o m 6 F /6 —
1 H /1 L 3 G /6 E 5 A /5 c o m 6G /6co m
1 K /6 K 3 H /3 L 5 B /5 — 6 H /6 L
1 L /2 L 3 K /4 K 5 E /D 5 6 H /6 +
2 E /D 9 3 L /4 L 5 F /6 F 6 K /5 c o m
2 F /3 F 4 A /4 c o m 5 G /6 G D 1/ 6com
2 G /5 G 4 B /4 — 5 H /5 L D 2 /5 c o m
2 H /2 L 4 E /D 6 5 K /4 c o m D 4 /4 c o m
2 H /3 H 4 F /5 F 5 L /6 L D 8 /3 c o m
2 K /3 c o m 4 G /5 co m 6 A /6 c o m D 1 6 /M +
3 A /3 c o m

T o use the Decim al to Binary Converter:

Enter any decim al num ber from 0 through 10 by tu rn in g the rotary switch to the desired posi­
tion. The bin ary equivalent o f the decim al num ber w ill appear in the binary ou tp u t lights. (A lig h t
ON represents 1; a lig h t OFF represents 0.)

Experiment 12: Binary to Decimal Converter

Conversion from binary to decim al is generally handled by a " tr e e " c irc u it (For an exam ple
o f a " tr e e " c irc u it, see the M in d Reading Program o f Book III.)
W ith a " tr e e " c irc u it, each com bination o f switches energizes one and only one te rm in a l of
the rotary switch. The rotary switch is program m ed so th a t when it turns to the energized te rm in a l
it w ill stop. Thus, setting a binary num ber in the slide switches w ill result in the same num ber
being displayed on the rotary switch.

Program fo r Binary to Decimal Converter:

1 C /4 R 2 K /3 L 4 C /5 C 5 K /D 1
1 F /2 F 2 L /4 W 4 F /5 F 5 N /D 0
1 G /6 H 2 N /4 L 4 G /D 1 3 5 S /6 S
1 H /5 U 3 C /6 R 4 J /D 1 2 6 F /6 —
1 J /5 H 3 C /4 C 4 K /D 9 6 G /D 7
1 K /6 L 3 F /4 F 4 N /D 8 6 J /D 6
1 L /5 W 3 G /D 1 5 4 S /4 + 6 K /D 5
1 N /5 L 3 J /D 1 4 4 S /5 S 6 N /D 4
2 C /5 R 3 K /D 1 1 5 C /6 C 6 X /D 1 7
2 F /3 F 3 N /D 1 0 5 F /6 F 6 Y /6 +
2 G /3 H 3 U /4 V 5 G /D 3 D 1 6 /D 1 9
2 H /4 U 3 V /3 — 5 J /D 2 D 1 8 /M —
2 J /4 H 3 W /5 V

To use the B inary to Decim al Converter:

Enter any bin ary num ber fro m 0 through 1 5 on slide switches 3 through 6. Push pushbutton
6. The rota ry switch w ill indicate the decim al equivalent o f the binary number.

85
 A PPENDIX

Automatic Shift Register

The s h ift register w hich was previously discussed can be ''a u to m a te d '' so th a t the com puter
w ill ca rry out the instruction: SHIFT LEFT ONE BIT whenever the rotary switch d ial is turned. A l­
though this is a more convenient method o f sh ifting-as opposed to m a n u a lly operating the slide
switches— you can see th a t it uses m any more contacts.
The program fo r the e ntire six -b it a u to m a tic s h ift register follows, w ith a c irc u it diagram fo r
the m iddle two bits:

C IR C U IT D IA G R A M : M ID D LE T W O BITS OF A U T O M A T IC SHIFT REGISTER

Program fo r fu ll six -b it auto m atic s h ift register:

1 A /1 X 3 A /3 X 5 A /5 X
1B / l — 3B /3 — 5 B /5 —
1C /2C 3 C /4 C 5 C /6C
1E/2N 3 E /4N 5 E /6 N
1 F /1 G 3F/3G 5F/5G
1F/2K 3 F /4 K 5F/6K
1 G /1X 3 G /3 X 5 G /5 X
1 H /2 H 3 H /4 H 5 H /6 H
1L/D1 3L /D 3 5 L /D 5
1 Y /2 Y 3 Y /4 Y 5 Y /6 Y
2 A /2 X 4 A /4 X 6 A /6 X
2 B /2 — 4 B /4 — 6 B /6 —
2C /3C 4C /5C 6 C /6 —
2 E /3 N 4E /5N 6E /D 7
2 F/2G 4F/4G 6F/6G
2F /3 K 4 F /5 K 6 G /6 X
2 G /2 X 4 G /4 X 6 H /6 +
2 H /3 H 4 H /5 H 6L /D 6
2 L /D 2 4L /D 4 6Y /D 1 6
2 Y /3 Y 4 Y /5 Y D 1 6 /M +

86
To use the automatic shift register:

Enter a bina ry num ber in the s h ift register by pushing and releasing the appropriate pushbut­
tons. The num ber w ill appear in the o u tp u t lights.
S h ift LEFT one b it by slowly tu rn in g the ro ta ry switch d ial one com plete revolution. The shifted
b in a ry num ber w ill now appear in the o u tp u t lights.
To s h ift le ft more tha n one b it sim ply tu rn th e rotary switch dial the desired num ber o f revo­
lutions.

Two-Bit Adder with Automatic Decimal Conversion

The fo llow ing program is given as an example o f the com bination o f basic com puter a rith m e tic
circuits. A tw o -b it adder is com bined w ith binary-to-decim al converter so th a t when two 2 -b it b in ­
a ry numbers are added together the sum is shown on the decim al o u tp u t dial.
The program fo r the tw o -b it adder w ith a u to m a tic decim al conversion is:

1C /4H 3J/D 3 5 U /5 Z
1 F /2 F 3 K /D 6 5 W /5 X
1 G /3H 3 N /D 2 5 Y /6 V
1H /6G 4 A /5 R 6 A /6 C
1 J/2H 4 A /4 K 6B /6—
1K /3L 4B /4 — 6C /6 V
1 L/6J 4 C /5 V 6 F /6 —
1N /2L 4 F/5 F 6 H /5 —
2 C /4 K 4G /5J 6S /6 W
2C /3C 4 H /5 A 6 U /6 Z
2 F/3 F 4J/5G 6 W /6 X
2 G /D 5 4L/4J 6 Y /6 +
2 J /D 1 5 B /5 - D 16/D 19
2 K /D 4 5 C /6R D 1 7 /M +
2 N /D O 5F/6F D 1 8 /M —
3 F/4 F 5 H /5 +
3 G /D 7 5 S /5 W

To use the program :

Enter a 2 -b it b in a ry num ber in slide switches 5 and 6.
Enter a second 2 -b it bin ary num ber on pushbuttons 5 and 6.
The rotary switch d ia l w ill tu rn to the decim al sum o f the numbers. (Binary o u tp u t lights 4, 5
and 6 w ill display the sum in binary.)

87
M IN IV A C 601 A N D THE M IN IV A C M ANU AL

ARE PRODUCTS OF:

CORPORATION

372 Main Street, Watertown, Massachusetts

BOOKS...
V AND VI

Copyright (c ) 1961

by
S C IE N T IF IC
D EV ELO P M EN T
How Computers Work for Man
C O R P O R A T IO N

Watertown MIN IVAC Games

Mass.
 Ml N I V A C 601

BOOKS V-VI

SCIEN TIFIC DEVELOPMENT CORP.

W A T ER T O W N , MASS.
 T h e M in iva c M an u al was prepared and edited by the staff of

Scientific Developm ent Corporation

First Printing— A ugust, 1961

EX LIBRIS ccapitalia.net
Copyright ( c ) 1961 by Scientific Development Corporation, 372 M ain Street, Watertown, Massachusetts
 CONTENTS

BOOK V: HOW COMPUTERS WORK FOR MAN I

1. SIM PLE C O M PU TER -LIKE D EVICES EN C O U N TERED EV E R Y D A Y 1

Introduction 1
Experiment 1: The Light on the Stairs 2
Experiment 2: The Burgular A larm 2
Experiment 3. The Metronome 3
Experiment 4: The T ra ffic Light 4
Experiment 5: The T ra ffic Light with Pedestrian Control 5
Experiment 6: The Automobile Speed Tim er 6
Experiment 7: The T rain Gate Control 8
Experiment 8: The Two-Floor Elevator 10
Experiment 9: The Three-Floor Elevator 11
Experiment 10: The Automatic Toll Collector 12
Experiment 11: The Telephone Dialing System 13
Special-Purpose vs. General Computers 14

2. C O M PU TER A P P LIC A TIO N S IN BUSINESS A N D IN D U STR Y 16

The Computer System in the Business World 16
W hat Computers Can Do 16
W hat Computers Cannot Do 16
Truth vs. Output 17
W ill Computers M ake Management Obsolete? 17
Examples of Computer Handling of Business Problems 17
The Payroll Problem 17
The Inventory Problem 19
Check Processing: How Computers "R e a d " Printing 20
Experiment 12: Arabic Numeral Recognition 22
Examples of Computer Handling of Industrial Problems 23
Experiment 13: M ultiple Point Control 24
Experiment 14: Sequence Control with M anual Operation 24
Experiment 15: Automatic Sequence Control 25
W hat is the "B e st" Computer? 26

3. CO M PU TER A P P LIC A TIO N S IN SC IEN C E A N D T H E M IL IT A R Y 27

Introduction 27
Real-Time Problem Solving 28
Computer Handling of Scientific Problems 28
Examples of Computer Handling of M ilitary Problems 29
Signal Systems 29
Experiment 16: Automatic Message Transmission 29
Experiment 17: Automatic Name Transm ission 30
Experiment 18: Automatic Transmission with Differential
Spacing 31
Computerized Coding Systems 31
Experiment 19: Encoder for Morse Code 32
Experiment 20: Decoder for Morse Code 34
Experiment 21: "Search and T ra c k " Radar 35
Countdown Control 36

iii
4.C O M PU TER A P P LIC A T IO N S IN T H E SO C IA L A N D P O L IT IC A L SC IEN C ES 36
Vote Registering Machines 36
Computers and Election Predictions 36
Language Translation 37
Symbols and Meaning 37
Translation of Symbolic Combinations 37
Simple Translations between English and German 38
Job Selection 40
Mate Selection 42
Behavioral Simulations 44
Associative Memory 44
A Simulated M aze Solver 44

5.C O M PU TER A P P LIC A TIO N S IN SC IEN C E FIC TIO N FILM S 45

The Flashing Lights Circuit 45
The "Sup er" C ircuit 46

A P P E N D IX : Programming Languages 46

BOOK VI: MIN IVAC GAMES

Preface 51

T; M IN IV A C AS AN O PPO N EN T 51
The Secret Code 51
The Combination Lock 52
The Electronic Maze 53
The Match Game 54
Tic-Tac-Toe 55

2. M IN IV A C AS A REFEREE 56
The Philosophic Tug of W ar 56
The Mind Reading T ric k 57
The Fortune Teller 58
The Random Number Generator 59
Scissors, Paper or Stone 59
Reaction Tim e Tester 60

iv
 BOOK V

How Computers Work for Man

PREFACE

The previous books in this series have examined various functions performed by modern
high-speed digital computers. In this book, M IN IV A C 601 is used to demonstrate how these func­
tions are combined in various situations to enable the digital computer to do a specific job. Ex­
amples of computer applications are examined in five major sections.
The first section— Simple Computer-like Devices Encountered Every Day— presents a number
of fam iliar basic switching circuit devices. These applications involve relatively simple, special-
purpose machines which are not "com puters" in the general sense in which computers were de­
fined in Book II, although they perform computer-like functions.
The second section— Computer Applications in Business and Industry— contains a description of
some general applications which computer systems find in business and industry. Included in this
section is a discussion of how computers "re a d " printed material in processing bank checks. Indus­
trial systems used to provide automatic process control and checking of production are also in­
cluded in this section.
Section three— Computer Applications in Science and the Military— contains examples of prob­
lem-solving and control applications in which large computers are now being used. In addition to
illustration of the kinds of problems which computers are helping scientists to solve, this section
contains experiments demonstrating how computers are utilized in radar tracking, automatic
coding systems, and missile check-out procedures.
Section four— Computer Applications in the Social Sciences— examines the potentials of com­
puters in such areas of election prediction, language translation, and simulation of human learn­
ing functions.
Although the first four sections of this book provide a summary of the many areas in which
high-speed computers are finding applications, there is one computer application which cannot
reasonably be considered under any of the four headings discussed above. Although this fifth area
is probably the least important of all computer applications, it may well be the only application
with which your friends are really fam iliar. Section five— Computer Application in Science Fiction
Films— is included to insure that you will be able to use M IN IV A C 601 to demonstrate how com­
puters are used on the movie and T V screen.
Much work has recently been done in the development of "compiler programs" which permit
programmers to communicate instructions to computers using a language sim ilar to that used to
communicate mathematical problems to men. The compiler program converts the symbols of
mathematical and/or logical expressions into binary codes which can be processed by the com­
puter. In the appendix to this book— Computer Programming Languages— commonly used program­
ming languages are discussed.

1. Simple Computer-like Devices Encountered Every Day

Th is section contains examples of several fa m ilia r computer-like machines. These simple
devices perform specialized functions using basic computer or "switching circu it" functions.
It is important to remember that the machines used to perform these functions do not have

1
the capacity or flexib ility of the more generalized high-speed digital computers discussed in later
sections of this book. These simple machines, however, provide excellent examples of specialized
computer functions used to perform some basic jobs for man.

Experiment 1: The Lights on the Stairs

It seems appropriate to begin our examples of simple computer-like systems with the ordinary
light switch. In many homes a light in a particular location can be controlled from 2 points. Th is is
often true of the light in a stair well which can be turned on or off either at the top or the bottom
of the stairs. The problem presented in this case (i.e., how to turn the light on or off from either
the top or the bottom of the stairs) is solved using a basic computer function.
The function used in this circuit is a version of the A N D circuit. The circuit diagram and pro­
gram below summarize the situation. W ith this circuit in use, the light can be turned on or off
from either of 2 locations. Assume that slide switch 1 is the switch at the top of the stairs and
slide switch 2 is the switch at the bottom of the stairs. Regardless of which position the switch is
in, the switch at the other location can always turn the light either on or off.

Program for dual-control light:

1A/1 +
1B/1V
1U/2U
1W/2W
2 V /2 -

To use the program:

Light 1 can be turned either ON or OFF using slide switches 1 or 2.

For example, moving slide switch 1 to the L E F T will turn light 1 ON if slide switch 2 is
RIG H T. The light will go OFF if slide switch 2 is now moved to the LE FT .

Experiment 2: The Burglar Alarm

Th is next circuit provides an example of the basic NOT function in a device found in some
homes, most office buildings, and all banks: the burglar alarm . The essential element of this
application is that all of a set of contact points must be closed for the alarm not to sound. The con­
tact points may be on doors, windows or vault openings.
If any set of contacts is opened— by a door or window being opened— the circuit is broken
and the alarm sounds. Once the alarm sounds, the device can be re-set by closing the contacts and
throwing a re-set switch. In actual practice, the re-set switch is not readily accessible; in some
cases, in fact, the re-set switch may be at police headquarters.
The program which follows illustrates a simple alarm circuit. If any contact is broken by
pushing any of pushbuttons 2 through 6, the alarm (light 5) comes on. The system is re-set by
pushing pushbutton 1.

2
 C IR C U IT D IA G RA M — B U R G LA R A LA R M

Program for Burglar A larm :

1X/6F 3 Z/4Y 5 E/5 + 6 C /6 +
1Y/2Y 4 Z /5 Y 5F/5H 6F/6L
2 Y /2- 5 A /5 + 5J/6J 6H/6—
2 Z/3Y 5B/5J 5Z/6Y 6K/6Z

To use the program:

Turn power ON. The alarm (light 5) comes ON. Set the system by pushing the re-set
button (pushbutton 1). The burglar alarm is now set. If any contact is broken by pushing any
of pushbuttons 2 through 6, the alarm will come on. A fter an alarm has been given, the
system can be re-set by pushing pushbutton 1.

Experiment 3: The Metronome

Computers frequently use timing circuits to perform a series of operations or to repeat an
operation periodically. A fam iliar example of a tim ing device is the metronome, which is designed
to generate a signal at periodic intervals.
The program which follows demonstrates a basic timing circuit using the turning of the
rotary switch dial between two points to generate a periodic signal. Commercial metronomes can
produce signals with a wide range of periods; with this circuit, however, only four periods can be
produced using the positions of slide switches 4 and 5 to determine the period. Through an ap­
propriate circuit, the M IN IV A C is capable of generating 15 different periods.
The program and circuit diagram for a metronome are:

3
 C IR C U IT D IA G RA M — M ETRONO M E

Program for metronome:

4R/D3 4W/D9 6F/6— 6J/6K
4S/5R 5S/6K 6G/D1 6 J/6 +
4T/D5 6C/6L 6G/6N 6L/D18
4U/D7 6E/D16 6H/D17 6N/6—
4V/5T

To use the program:

Turn power on. The rotary switch dial will begin to turn back and forth, generating a periodic
signal. To change the period of the signal, move slide switches 4 and 5.

Experiment 4: The Traffic Light

Tim ing circuits are used in many devices. One of the most fam iliar is the tra ffic light which
follows a pre-programmed cycle from red to green to yellow and back to red. A continuous cycling
device is used in this case to produce the fam iliar sequence of light.
Using the rotary switch mechanism and the lights, M IN IV A C can be programmed to simu­
late the three-light sequence of a tra ffic light.

RED AMBER GREEN

C IR C U IT D IA G RA M — T R A F F IC LIG H T

4
Program for traffic light:
4 A /4 + D1/D2 D8/D9 D15/D0
4B/D11 D2/D3 D9/D10 D16/M—
5 A /5 + D3/D4 D1I/D12 D17/M—
5B/D7 D4/D5 D12/D13 D18/M +
6 A /6 + D5/D6 D13/D14
6B/D1 D7/D8 D14/D15

To use the tra ffic light program:

Turn power on. The rotary switch will turn, turning lights 4, 5 and 6 on and off in succession.
Light 4 represents "re d ", light 5 represents "am b er", and light 6 represents "green". The length
of time any given light remains on may be varied by changing the connections on the rotary
switch.

Experiment 5: The Traffic Light with Pedestrian Control

The tra ffic light provides a good opportunity to move from a simple timing circuit to a more
complicated circuit which involves sequence control. The tra ffic light in Experiment 4 moves
through a pre-determined sequence, and unless the device is re-programmed, it will continue to
fo llo w th e s e q u e n c e a s lo n g a s p o w e ris fe d to it.lt is often desirable, however, to interrupt the
sequence or control it in some manner.
Specifically in the case of the tra ffic light it is desirable to have an "over-ride" control which
allows a pedestrian to modify the normal pattern of the light. The program and circuit which fol­
low illustrate an "interrupt circu it" which permits an operator to temporarily change or start a
pre-programmed sequence.
RED AMBER 6REEN

Program for tra ffic light with pedestrian control:

4A/4 + 5F/5Z 6B/D7 D5/D6 D12/D13
4B/D1 5H/5— D1/D2 D7/D8 D14/D15
5 A /5 + 5J/D18 D2/D3 D8/D9 D15/D0
5B/D14 5Y/D6 D3/D4 D9/D10 D16/M —
5 E/5 + 6A/6 + D4/D5 D10/D11 D17/M +
D11/D12

5
To use the program:
Turn power on. The "re d " light (light 4) comes ON. To cross the street: push pushbutton
5 until the "re d " light (light 4) goes OFF and the "green light" (light 6) comes ON. A fte r the
light has turned green, it will go to amber (light 5) and return to red again.
In actual practice, this tra ffic light would be tied to a complementary light facing the flow
of tra ffic . Th is tra ffic light, facing the pedestrian, would normally be red, the complementary
light would normally be green. Since the two lights are connected, changing the pedestrian's light
by pushing a button would autom atically change the complementary light.
Experiment 6: The Automobile Speed Timer
A s an example of a computer-like device encountered, we hope, not every day, consider the
device used by police who post the warning: "Speed Electrically T im e d ." The mechanism used
to determine the speed of passing cars depends essentially on a device which measures the time
which lapses as a car passes between two given points.
The actual devices use a variety of mechanisms (radar beam, switches on the road, etc.) to
indicate when a car has passed the two selected points. However, we can simulate the action of
the timing mechanism by pushing pushbuttons to indicate when a car passes the check points.
The signal that a car is passing the first check point will be indicated by pushing pushbut­
ton 6; the signal that the car is passing the second check point will be indicated by pushing push­
button 5. Pushing pushbutton 6 starts the timing motor (in this case the rotary switch); pushing
pushbutton 5 stops the motor. As the rotary switch turns, M IN IV A C will count its revolutions.
A fter the motor has stopped, we will be able to calculate the speed of the passing car.

Program for the automobile speed timer:

1C/1G 2L/D 11 4H/5H 6C/6X
1C/2K 3C/3G 4L/D13 6F/6—
1F/1- 3C/4K 5C/5G 6H/6 +
1H/1 + 3F/3— 5C/D15 6H/6L
1H/2H 3H/4H 5F/5— 6K/D17
2C/2G 3L/D12 5L/D14 6 Y /6 +
2C/3K 4C/4G 5X/6E D16/M +
2F/2— 4C/5K 5Y /5 — D18/M—
2H/3H 4F/4— 6C/6G

6
 The simplest way to calculate automobile speeds with the above program is to make up a
dial plate to cover the regular input-output dial. To make such a dial plate, follow these steps:

1. Measure the distance between the two points to be used as check-points.

2. Correct this distance from whatever unit it was measured in to miles. For example, if the
distance was measured in yards, divide this distance by 1760 (the number of yards in a
mile) to get the distance in miles.
3. Count the number of complete revolutions which your rotary switch makes in one minute and
note this number. Since the program above indicates a revolution of the rotary switch when
the dial has completed 15/16 of a revolution, we must multiply the number of complete
revolutions in a minute by 16/15 to obtain a "revolutions per m inute" figure for the dial
plate.
4. M ultiply the "revolutions per m inute" figure which you have obtained by 60 to get the num­
ber of revolutions per hour.
5. M ultiply this last "revolutions per hour" figure by the distance between the two checkpoints
expressed in miles. This will give you the speed in miles per hour which a car would be trav­
eling if the rotary switch turned exactly one revolution while the car passed between the two
checkpoints.

For example: assume that the checkpoints are 50 yards apart, and that the rotary switch
turns 45 full revolutions per minute. Then:

50
50 yards = ------ = .029 miles
1760

16
45 X — = 48 revolutions per minute
15

revolutions per hour = 48 X 60 = 2880

And if a car were to pass from the first checkpoint to the second checkpoint while the rotary
switch dial moved from 0 to 15, its speed would be:

.029 X 2880 = 83.52 miles per hour

To find the car's speed if the dial turned through two revolutions— from 0 to 15 and then from
15 to 14— we will divide the last figure by 2:

83.52
= 41.76 miles per hour.

Sim ilarly, if the dial turns through three revolutions— from 0 to 15, then from 15 to 14, then
from 14 to 13— the speed of the car is:

83.52
---------- = 27.64 miles per hour.

The dial plate will thus have a sequence of speeds depending upon the number of revolu­
tions the rotary switch makes as the car passes between the check points. Since the program is
such that the relay indicator lights count the number of revolutions of the rotary switch, we will
know which speed to read by observing which relay lights are on.

The dial plate for the example above would look like this:

7
 READ SPEED IN MPH FROM THIS
BAND IF:

RELAY LIGHT 5 IS ON

RELAY LIGHT 4 IS ON

RELAY LIGHT 3 IS ON

RELAY LIGHT 2 IS ON

RELAY LIGHT 1 IS ON

D IA L P LA T E— A U TO M O BILE SPEED T IM E R

To time an automobile's speed:

Push pushbutton 6 as a car passes the first check point.
Push pushbutton 5 as the car passes the second check point.
Read the speed of the passing car from the dial plate: the speed is read from the band cor­
responding to the lowest numbered relay indicator light which is on.

Experiment 7: The Train Gate Control

The gate control at a train crossing provides a simple example of a function known as "en ­
vironmental sensing." A device with environmental sensing is one which is capable of receiving
information about its environment and then acting on the basis of the information. In a large-
scale computer, the actual "sensing" device would ordinarily be a specially-designed input unit
which would autom atically feed environmental information into a computer which was pre-pro­
grammed to handle it.
A n extremely complex version of environmental sensing is the control system used in an au­
tom atically controlled oil refinery process. Input in the form of fluid flow, temperature, pressure,
etc., is feed into a computer which processes this information and feeds its results into special­
ized output units which adjust valve settings, re-route fluid flow, or, in case of emergency, turn
on various safety devices.
In the case of the simple train gate control, the control device must be capable of sensing
the presence of a train and, when it does, must act to close the gate. The actual "input u nit" for
this type of system is generally a set of pressure switches under the rails. These switches remain on
as long as a train's weight is on them. In actual practice there are many sets of switches located
up and down the track, spaced less than a locomotive's length apart. When any of these switches
is activated, the gate closes and remains closed until none of the switches is on.

8
 For demonstration purposes, we can assume that there are only three switch locations: one
each on either side of the crossing, and one in the middle of the crossing. The switch in the mid­
dle of the crossing must be so designed that cars passing over it will not close the gate.
The circuit diagram and program for a train gate control system are:

Program:
4C/5C 5F/5K 6J/D17
4F/4X 5G/6F 6 L /6 +
4F/4K 5H /5+ 6L/6Y
4G/5G 5L/5X 6N/D18
4H/5H 5Y/6Y D0/M +
4L/5L 6C/6— D7/M—
4 Y /5Y 6G/6N D16/D19
5C/6C 6H/6—
5F/6X 6J/6K

To use the program:

Turn power on and set the pointer knob of the rotary switch at 0.
The rotary switch at 0 represents the gate up. To indicate that a train approaches from the
left and continues through the crossing, push pushbutton 4, 5,and 6 as follows:
1. Push pushbutton 4.
2. W hile holding pushbutton 4 down, push pushbutton 5.
3. Hold pushbutton 5 down and release pushbutton 4.
4. W hile holding pushbutton 5 down, push pushbutton 6.
5. Hold pushbutton 6 down and release pushbutton 5.
6. Release pushbutton 6.
The pattern for a train moving from left to right through the crossing may be expressed as
the following sequence of operations:

9
 Pushbutton 4 Pushbutton 5 Pushbutton 6
0 0 0
1 0 0
1 1 0
0 1 0
0 1 1
0 0 1
0 0 0
Notice that the gate closes (the pointer knob moves to the "gate closed" position— 7) as
soon as the first switch is activated and remains closed until none of the switches is activated
The same will be true for a train moving from left through the crossing, or for a train which en­
ters the crossing, stops and backs out again.
Push pushbutton 5. This represents a car passing over the center of the crossing. Notice that
the gate does not close.

Experiment 8: The Two-Floor Elevator

A number of fam iliar devices exhibit an important characteristic of a computer - memory.
One of the most common devices which has a memory built into it is the automatic elevator. The
complexity of the controlling mechanism for an automatic elevator depends prim arily on its size
— that is, the number of floors which it services.
The simplest possible automatic elevator control, of course, would be for a two-floor elevator.
The system would then be required to remember only at which of two floors it was stopped, and it
would not need to remember commands from various other floors. The program which follows
simulates this basic elevator control circuit.
The rotary switch dial at 1 indicates that the elevator is at the first floor; the dial at 7 indi­
cates that the elevator is at the second floor. Sim ilarly, binary output light 5 on indicates that
the elevator is at the first floor; binary output light 6 on indicates that the elevator is at the sec­
ond floor. These lights are analogous to the indicator above an elevator door which tells you
where the elevator is.
Pushbuttons 5 and 6 serve a dual purpose; they represent the " c a ll" buttons on the floors, as
well as the control buttons inside the elevator. T h at is, pushbutton 5 is used to direct the elevator
to the first floor— as would be done either by pushing a " c a ll" button on the first floor or by
pushing the "firs t floor" button inside the elevator. Sim ilarly, pushbutton 6 is used to direct the
elevator to the second floor.
The circuit diagram and program for the two-floor elevator control system are:

C IR C U IT D IAG RA M — TW O-FLO OR ELEV A TO R C O N TRO L SYSTEM

10
The program is:
5A/6C 5G/6E 6 F/6- 6N/D18
5B /5 - 5L/5X 6G/6N 6Y/D1
5B/6B 5N/6E 6H /6+ D1/M +
5C/5— 5Y/6Y 6J/6K D7/M—
5C/5H 6A/6K 6 L/6 - D16/D19
5F/6X 6C/6G 6J/D17

To use the program:

Set the "elevator" at the first floor by setting the pointer knob of the rotary switch at 1. To
call the "elevato r" to the second floor, push pushbutton 6. The "elevator" will immediately move
to the second floor (the pointer knob will point to 7). To return the "elevator" to the first floor,
push pushbutton 5.

Experiment 9: The Three-Floor Elevator

The requirements for the previous elevator control system were very simple. Since there
were only two possible locations for the elevator, the control system's memory had only to re­
member whether the elevator was up or down. If we add only one more floor to the system, how­
ever, the necessary circuitry becomes much more complicated. The system must be able to sense
where it is, remember its destination, and then must be able to move to its destination.
A s before, the pushbuttons will be used to represent both the call buttons on the various
floors and the control buttons inside the elevator. Whenever a pushbutton is pushed, the system
will go through the following steps:
1. The system will determine on which floor the elevator is presently located.
2. It will compare its present location with its destination (as indicated by which pushbut­
ton was pushed).
3. If the destination differs from present location, the system will store the destination and
proceed with step 4.
4. The system will select the direction in which it is to move (up or down).
5. The "elevato r" will begin to move in the selected direction.
6. A s the "elevato r" begins to move, the pushbuttons will interlock so that the system will
ignore further calls.
7. The "elevator" will continue to move to its stored destination.
8. When it reaches the stored destination, the "elevato r" will stop and clear the destination
from its memory.
9. A s the "elevato r" stops, the pushbutton interlock will clear and the system is ready for
a new command.
The program for the three-floor elevator control system is:
1A/1E 2B/2— 3F/3G 5F/6C
1B /l — 2C/3C 3G/3X 5G/5N
1C/2C 2E/D7 3J/4E 5J/5K
1E/D1 2F/2G 3Y/4J 5J/D17
1F/1G 2G/2X 4C/5F 5L/5—
1G/1X 2J/3H 4F/4 + 5N/D18
1H/1 + 2X/6H 4G/4H 6C/6—
1H/1L 2 Y /3 Y 4H/4 + 6F/D12
1J/2H 3A/3E 4L/D19 D0/D1
1K/5E 3B/3— 4N/D18 D6/D7
1Y/2Y 3C/4C 5C/5G D12/D13
2A/2E 3E/D13 5E/6G D16/M—

call button indicator light

Floor 1 pushbutton 1 output light 1 D0-D1

Floor 2 pushbutton 2 output light 2 D6-D7
Floor 3 pushbutton 3 output light 3 D12-D13

11
To use the program:
Set the "elevator” at the first floor by setting the pointer knob at DO. Turn power on. Out­
put light 1 will come on indicating the location of the "elevato r". To call the "elevato r" to the
third floor, push pushbutton 3. To direct the "elevator" to the second floor, push pushbutton 2.
Notice that, even with the relatively complex circuit above, the system is still not able to
accept more than one command at once. Much more capacity is required for a control system
which will remember successive commands and select the appropriate next move.

Experiment 10: The Automatic Toll Collector

Most of the specialized circuits discussed so far have been designed to perform various con­
trol functions. This next circuit, however, is designed to function as a decision-maker. The au­
tomatic toll collector has the responsibility of deciding whether or not a combination of coins
given to it is sufficient to pay the toll. Ignoring the very specialized problem of recognizing va ri­
ous coins, let us consider the decision-making problem which the automatic toll collector must
solve.
W e will assume that a 10^ toll is to be collected. Th is toll can be paid by depositing:
one dime
two nickels
one nickel and five pennies
ten pennies
The automatic toll collector must examine the input (the coins) and decide if it is adequate. If
the payment is sufficient, the system must indicate to the driver that he may proceed. However,
if a car passes through the toll gate without paying sufficient toll, the system must sound an
alarm so that the guards will be notified immediately.
The program for an automatic toll collector is:

1A/3K 2 H /2 + 4A/4E 5J/5X

1A/2E 3A/3E 4B/5B 5J/6H
IB / IE 3B/4B 4C/5C 5L/6G
1B/2B 3C/4C 4E/4J 5X/D5
1C/1J 3E/4G 4H/5N 5Y/6Y
1E/2C 3F/3H 4K/5N 6A/6E
1F/1H 3F/4F 4Y/5Y 6C /6-
1J/2K 3G/3L 5A/5E 6E/6J
2A/3N 3G /3+ 5B/6B 6X/D16
2B/3B 3J/4H 5C/6C 6X/D18
2C/3C 3J/4X 5E/6G D0/D19
2F/2G 3L/3Y 5F/5H D5/D10
2F/4N 3N/6Y 5F/6F D17/M—
2H/2L 3X/4L 5G /5+

Output light 1 ON represents "T h a n k You— Go Ahead"

Output light 2 ON represents "Stop— Pay Toll lOtf"
Relay 1 and relay indicator light 1 ON represents "A la rm "
Pushing pushbutton 3 represents "D rive A w ay"
Pushing pushbutton 4 represents depositing one dime
Pushing pushbutton 5 represents depositing one nickel
Pushing pushbutton 6 represents depositing pennies (number of pennies indicated on rotary
switch)
To use the program:
Turn power on. Output light 2 comes on indicating "Stop— Pay Toll 104."

12
 Deposit 10<? in toll:
to deposit one dime, push pushbutton 4
to deposit two nickels, push pushbutton 5 twice
to deposit one nickel and five pennies:
push pushbutton 5 once; then set the rotary switch dial to 5 and push pushbutton 6
to deposit ten pennies:
set the rotary switch dial to 10 and push pushbutton 6.
When 10P has been deposited, output light 1 will come on indicating "T h a n k You— Go
A head ."
Push pushbutton 3 to "D rive A w a y."
The toll collector will re-set.
If you try to drive away without depositing enough money, the alarm will sound. To turn off
the alarm , deposit 10P. If you deposit more than 1OP, the automatic toll collector will accept the
extra money and do nothing about it.
Experiment' 11: The Telephone Dialing System
The telephone dialing system is an excellent example of a highly specialized computer sys­
tem capable of handling a particular type of complex problem. The entire dialing system is far
too complicated for this discussion. However, we can examine one small segment of the prob­
lem: the conversion of a dialed number into a type of binary code which can be understood by
the circuits of the telephone exchange.
The telephone dialing system uses a special binary code known as the Gray Code. This code
is binary in nature; that is, it uses only the digits 0 and 1. However, it does not follow the usual
rules for the development of a number system. The system is so designed that when moving from
any number to the next highest number there is a change of only one bit.
For this experiment, we will use a three-bit Gray Code which will allow us to count from 1
through 7. W e will use the Gray Code equivalent of zero for 8 so that we can represent the num­
bers 1 through 8:
Decimal Number Three-Bit Gray Code
001
2 O il
3 010
4 110
5 111
6 101
7 100
8 000
The program which follows is in essence a conversion program from decimal to Gray Code.
The decimal number "d ialed " will be displayed in Gray Code on output lights 4, 5, and 6. This
represents only the first step in the handling of dialed numbers in a telephone system: the con­
version of the dialed number to a form which can be processed at the telephone exchange.
The program is:
IC / 1 - 2N/6G 4E/5E 6L/6W
1C/1H 3A/6E 4F/5F 6R/6U
IF/DO 3B /3- 4H/6N 6S/D1
1G/4B 3C/4C 5A/5K 6U/D18
2C/3C 3E/4J 5B/6B 6V/D2
2E/3E 3F/4G 5C/6C D1/D3
2F/3F 3G/4A 5F/5K D2/D4
2F/3G 3 H /3 + 5H/6T D3/D5
2G/2N 3K/4E 5L/6H D4/D6
2H/5G 3L/6K 6A/6F D5/D7
2J/2K 3N/4F 6C/6— D6/D8
2J/6E 4B/5B 6F/6G D16/M +
2L/5J 4C/5C 6H /6+ D17/M—

13
To use the program:
W ith power off, set slide switch 6 R IG H T. Set the rotary switch dial at 0. Turn power on.
Slowly turn the rotary switch dial to the number you wish to "d ia l” . (Be sure that the pointer
knob is pointing directly at the number.)
Set slide switch 6 LE FT .
The rotary switch will return to 0 and the number "d ialed " will appear in Gray Code on out­
put lights 4, 5, and 6.
(Note: output light 3 should be ignored.)
To " d ia l" another number, set slide switch 6 R IG H T and re-set the system by turning power
off and on again.

Special-Purpose vs. General Computers

In the preceding experiments, various devices which exhibit certain computer-like charac­
teristics have been examined. These devices share one common characteristic: they are designed
to perform a specialized and routine function. None is capable of performing all of the major
computer functions discussed in Book II. Moreover, none is capable of performing generalized
problem-solving functions.
In each case, though, a large-scale generalized computer could easily be programmed to
solve the specific problem. It is just not practical to have a full computer system to, for exam ­
ple, control the lights in a home. Therefore, simple computer-like devices are designed to per­
form these specific tasks. Technically, any of these simple, highly specialized devices could be
called a computer since it can receive input, can process and/or store the input, and can yield
output. More accurately though, these devices should be called "com puter-like" because of their
extremely limited capabilities.
The computer systems discussed in Book II were general computers. T h at is, their input and
output capabilities allow them to handle a wide range of communication, and their processing and
storage units are such that various types of problems can be handled. These general computers
are used prim arily in situations where there is likely to be a great deal of variety in the type of
problem to be solved. A research organization, for example, would require the versatility of the
general computer if it were using the computer for "one-shot" problems (problems which would
only be programmed onto the computer once).
For routine problem-solving— for example, making out a weekly payroll— an organization
may find a special-purpose computer most practical. A special-purpose computer is just what its
name implies: a computer designed to handle a lim ited number of problems in a particular fash­
ion.
Special-purpose computers are often made to order, in which case computer "building-
blocks" are frequently used. Computer building-blocks are simply generalized logical units which
perform particular tasks. A building-block might be a half-adder, or one bit of an accumulator,
or a single flip-flop. A ny number of building-blocks can be combined to produce a special-pur­
pose computer to handle a specific problem.
When a special-purpose computer is in the design stage, the phrase "b la ck box" is often
used. This is just a convenient, if amusing, method of expressing the particular problem at hand.
When dealing with a black box, the designers are not concerned with how the system will actually
go about solving the particular problem; they are concerned rather with what the system must
do. The black box is essentially the processing section of a particular part of the problem-solving
unit. The inputs and outputs are specified; later in the design stage, building-blocks will be used
to carry out the requirements of the black box.

W e might define a black box as follows:

a black box is a device with specified inputs and outputs; the outputs are expressed as func­
tions of the inputs. The inner workings of the black box are unspecified.
To illustrate both black boxes and building blocks let us consider the problem of designing a
special-purpose computer which will do the following.

14
 The system will have three inputs (A , B and C) and one output. Each of two inputs (A and
B) is a series of electrical pulses of different rates. The third input (C) is a re-set input. Inside
the computer are two counters: one counter counts the number of pulses from input A ; the sec­
ond counter counts the number of pulses from input B. If at any time the contents of counter A
equal the contents of counter B, but do not equal zero, both counters will stop and a pulse will
appear as output. Input C will then set counters A and B to zero and the system will begin count­
ing inputs A and B again.
The first step in the design of this special-purpose computer will be to "block diagram " the
problem; that is, we will represent the problem graphically so that we can see just what will be re­
quired. The block diagram of our problem looks like this:

In this diagram, the black boxes are represented by the rectangles; the building blocks are
represented by the circles. The building blocks a re simply basic logical units which perform a defi­
nite function. The black boxes, however, are specified only as far as input and output; the method
for arriving at the output as a function of the input is not defined. When we use a black box to
represent some part of our computer we are saying only "insert some circuitry which will do the
necessary job".
A t a later stage in the design of our computer, we can take each black box apart and select
the appropriate building blocks which are required. In actual practice, this job would probably be
done by the design department of the firm from which we were buying our special-purpose
computer.
A s an example of taking a black box to the building block stage, the "T est for Zero" black
box might look like this:

" T E S T FOR Z ER O " B L A C K BOX

15
 When the actual computer is built, the appropriate components will be wired together ex­
actly as indicated in our sketches. Notice that when completed, our special-purpose computer is
actually only a particular combination of the basic logical units which we examined in Book III.
A general purpose computer, on the other hand, is a large number of basic logical units com­
bined in such a way that particular combinations of them can be autom atically effected by in­
ternal programming.

2. COMPUTER APPLICATIONS IN BUSINESS AND INDUSTRY

The Computer System in the Business World
General purpose computers and electronic data processing machines have found extensive
application in the business world, performing a wide variety of tasks with great speed and accu­
racy. In this section, we will discuss the kinds of functions which computers can effectively handle
in a business operation.
A basic answer to the question, "W h a t can computers do in business?" might very well be
"an y repetitious ta sk". In general, the computer is best equipped to handle any situation in
which the same basic operations are to be performed many times using different data, but fol­
lowing identical, prescribed procedures.
Sim ilarly, a valid answer to the question, "W h a t is it unreasonable to expect computers to do
in a business situation?" would be, "to solve a unique or non-recurring problem." These answers
are not necessarily valid in a research or scientific problem-solving situation where the complex­
ity of a given problem may well justify the use of a high speed computer in order to obtain only
one answer at a particular time.
However, in most business situations the problems to be solved are such that computers can
be profitably employed in performing repetitive calculations which must be made in the course of
daily operations. In such situations it is usually well worth the time spent in programming a com­
puter to solve a particular kind of problem since the computer will solve the same problem many
times in the future using whatever data is applicable at the time. On the other hand, the exten­
sive programming time required to prepare a computer to solve a particular problem may make
it economically unwise to program the computer to solve a unique problem. The computer can
be a valuable tool for the businessman who knows how to use it and understands what it can and
cannot do. It can, though, be a very expensive and highly inefficient window-dressing in the of­
fice of a businessman who does not fu lly understand both its potentialities and its limitations.
What Computers Can Do
Books III and IV examined specific arithm etic and logical operations which computers can
perform. When these operations alone or in combination are used to solve a problem requiring
large amounts of data at once or using different data over a long per-iod of time, the right com­
puter can be an invaluable asset.
In general, an electronic data processing machine can do the job more effectively than a
human clerk if the job to be done involves:
Repetitive arithmetic calculations.
Repetitive decisions based on explicit decision rules.
Categorizing or cataloging large numbers of items.
Comparing large numbers of items.
Tabulating or summarizing quantitative information.

What Computers Cannot Do

In general, an electronic data processing machine is not worth using when small amounts of
data are to be analyzed in a unique manner at a single point in time. A few people with adding
machines or calculators can process several statistical tests on hundreds of pieces of data more
rapidly than a programmer can prepare a high speed computer to process that data and card
punch operators can prepare the data for input into the electronic data processing system.
The section in Book III dealing with sufficient information is particularly applicable in
business situations. In order to solve a problem a computer needs both data and instructions. The

16
importance of this simple fact is often overlooked, sometimes with disastrous results. The basic
requirement of sufficient information leads to a list of jobs which a computer should not be
called upon to perform:
Non-repetitive calculations on small amounts of data.
Problem-solving where inadequate data is available.
Problem-solving where the exact procedure to be followed can not be specified.
Decision-making in situations where qualitative factors must be considered.

Truth vs. Output

It is an unfortunate fact that some people consider information to be accurate simply because
it appears on a printed page. Since most businessmen are exposed to computer output only in the
form of printing prepared by an output device attached to the computer, there is a danger of ac­
cepting the results of the computer operation without first questioning (1) the data which the
computer was given and (2) the methods, procedures, or techniques which the programmer in­
structed the computer to use in order to arrive a t its conclusion.
The answer which a computer produces is only as good as the data which it was given and
the procedures which it was instructed to follow. The computer is a completely faithful servant.
It does everything it is told to do and, providing that the instructions which it is given are not in­
ternally inconsistent, it will not question the legitimacy of either the data or procedures.
Will Computers Make Management Obsolete?
Some businessmen view computers as antagonists ready to usurp power and reduce the man­
agement function to punching computer input cards. These businessmen, who seem to fear that
their conference room will be replaced by a panel of blinking lights, cannot appreciate the legiti­
mate potentialities of electronic data processing equipment because they fail to understand the
basic and important limitations of the potential.
Rather than detracting from the importance of management personnel, the computer, if
properly used, can add immeasurably to their ability to handle a wider range of problems more
effectively, to explore greater numbers of alternatives and to make decisions based on more in­
formation, more completely analyzed. The computer simply removes the tedium of repetitious
considerations and permits management personnel to focus their attention on policy-making and
effective utilization of more and better information in more competent problem-solving.

Examples of Computer Handling of Business Problems.

The Payroll Problem
Computer handling of a large payroll processing job provides a good example of an opera­
tion which computers are well-qualified to perfom in a business situation. Assume that the com­
puter is given the following information for each employee on punched cards or that it already
has this information in storage or on magnetic tape:
1. An identification number
2. The employee's guaranteed pay rate
3. The employee's regular pay rate
4. The employee's overtime pay rate
5. An itemization of all authorized deductions for that employee.
Assume also that the computer has been programmed so that storage locations are avail­
able in which the following information may be accumulated for each employee:
Attendance hours
Regular pay
Overtime pay.
W ith this information stored in the computer, work records from various departments are
fed into the machine. A s this information is received by the machine, identification numbers on
departmental job cards are matched with employee identification numbers stored in the com­
puter and the appropriate calculations are undertaken as follows:

17
 1) A comparison instruction is used to determine whether the man receives his guaranteed
rate or a job rate.
2) The appropriate rate is multiplied by the number of hours in regular and overtime ac­
tivity and the appropriate amounts are stored in the reserved location.
3) A record of total hours on each job is accumulated and this information is stored in the
appropriate register.
4) A fte r total earnings have been determined, deductions are processed and final pay rec­
ords and labor costs records are supplied as output. The following flow chart summar­
izes the operation of this program:

18
 T h is flow chart illustrates a relatively comprehensive job done by a computer using only
the simplest of operations. The only logical operation used to perform the job outlined above is a
"com parison" operation which is simply a logical A N D . In performing the comparison the com­
puter compares each bit of the register containing the identification number for a particular man
with each bit of the register containing the identification number on the job record. If each bit of
the man's identification number register AND each bit of the register containing identification
number from the job record are the same, the comparison is satisfied and the computer proceeds
with the program.
The flow chart shows graphically one of the advantages of a computerized payroll system—
the ability to perform more than one job at one time. Notice that, in the same operation, the pay­
roll is processed and the cost distribution is calculated. This is a valuable trait, particularly to a
large firm which has many operations, each of which must be correctly accounted for.

The Inventory Problem

M aintaining an appropriate inventory for several thousand parts used in an active produc­
tion line is a very real problem to businessmen who must minimize time delays, storage costs, and
purchase costs in lot sizes, to mention only a few factors. An appropriate computer system can
handle the routine data flow and standardized decision-making after the correct program has been
written.
It cannot be over-emphasized that a computer system is only as good as its input. This is
definitely true of an inventory control system in which the computer yields purchase decisions
as output. If the computer can be given precise decision rules for purchase, as well as accurate in­
formation regarding the inventory, it can easily produce routine purchase decisions.
A particular advantage of a computerized inventory control system is the speed with which
information can be processed, making possible daily control of large inventories which would
otherwise be practically impossible.
Various inventory control systems use punched cards, paper tape or magnetic tape— or com­
binations of these. As an example of a functioning system, let us consider one which uses mag­
netic tape plus punched cards.
A complete record of the inventory is kept on a master tape which is up-dated daily. Addi­
tions to and issues from inventory are recorded first on punched cards and then fed into the mas­
ter tape. Daily output from the processing is in the form of punched cards which give order in­
structions, notice of abnormal usage and account summaries, as well as a revised master tape.
The following chart shows the input, processing steps and output for such a system:

NEW
M ASTER

a
MASTER

o
TA P E TA P E

Dally Processing

1. Bring on-hand and on-order balances up to date.

2. If total balance for a part is below "order point", a to-be-ordered card is TO-BE-ORDERED
punched showing economical order quantity and expected delivery date. CARD
D A ILY
3. If on-hand balance for a part is below "minimum bank" figure, an ex­
ORDERS EXPED ITE
pedite card is punched.
CARD
D A ILY 4. For parts having a receipt, new average unit cost is computed and included
RECEIPTS in new master tape.
EXCEPTION
5. If actual daily usage of a part varies by more than ± 10% from planned CARD
D A ILY average daily usage, an exception card is punched.
ISSUES
6. Each issue is costed and extended (quantity x unit cost = total cost) and EXTEN DED ISSUE
an extended issue card is punched. CARD
7. A distribution of costs by account is built up from the extended issues.
A CCO UN T SUM M ARY
CARDS

19
 Notice that of the five output card types two are "w arning" cards. Both the expedite and
exception cards alert management to the fact that some part of the system requires attention. As
in the payroll problem, cost distributions are built up as the inventory processing is done.

Check Processing
The use of computers to facilitate the physical handling of printed m aterials is effectively
illustrated in the computerized processing of bank checks. The actual job to be done is a book­
keeping procedure and, as such, the instructions and procedures are not unlike those of the pay­
roll problem.
The distinguishing aspect of the bank check problem is that the "in p u t" is a printed paper
(a check) rather than a punched card. In order to handle the problem then, the computer must be
able to "re a d " the printing. (This ability of a computer to "re a d " printing also arises in the proc­
essing of credit billings.)
Previously, we have discussed how a computer "understands" information fed to it by recog­
nizing the presence or absence of holes in a card, magnetic marks, etc. at particular locations.
The problem in "reading" printed characters is basically the same: the computer must be pro­
grammed to recognize and interpret a "code" which identifies the input symbols. The "code" in
this case is the standard alphabet or standard arabic numbers.
The first step in communicating information to a computer as input in printed form is to
position the printing so that the computer can observe it. This is directly analogous to placing
punched cards in a reader, or loading magnetic tape on a magnetic tape reader. A variety of "read­
ing" systems for printed material are used, and most require placing the printed material in a
device which either holds it or moves it to a light source or magnetic sensing device.
Once the material is positioned in front of the sensing equipment, these specialized input de­
vices communicate to the computer the presence or absence of a magnetic or ink spot at a par­
ticular location in a m atrix of light or of magnetic sensing elements. The process by which the
computer recognizes the symbols positioned against the m atrix can be demonstrated on M IN I-
V A C 's game m atrix.
For sim plicity, consider a m atrix composed of four sensing points: M l, M 3, M5 and M7.
Using this m atrix, M IN IV A C can be programmed to recognize simple geometric shapes. The pos­
sible geometric shapes which M IN IV A C can recognize are:
POINTS
M1 M3
x X •

X X
M5 M7

L IN E S
M1 M3 M1 M3 M1 M3

M7 M7 M5 M5 M5 M7

T R IA N G LES
Ml M1 M3 M3 Ml M3

M7 M5 M5 M7 M5 M7

SQUARES

d
M1 M3

M7 M5

G EO M ETR IC SH APES— V A R IO U S FORMS

20
 These various forms would be communicated to a commercial computer as information re­
garding the presence or absence of spots of ink or magnetic material at the "sensing points" of
the m atrix. The "sensing points" of M IN IV A C 's m atrix respond to electricity rather than light
or magnetic ink. Therefore we will supply current to the various sensing points of the m atrix to
simulate the effect produced by the sensing elements in a printed character reader.

Light 1 on will indicate that M IN IV A C senses a point

Light 2 on will indicate that M IN IV A C senses a line
Light 3 on will indicate that M IN IV A C senses a triangle
Light 4 on will indicate that M IN IV A C senses a square

The program for the geometric shape recognition circuit is:

1A/2K 2A/2G 3H/5G 5H/6G

1B / 1— 2B/2— 3J/3K 5J/5K
1C/2C 2C/3C 3L/5K 5L/6J
1F/2F 2F/M7t 3N/4G 6C/6—
1G/4A 2H/3N 4B/4— 6F/M H
1H/3G 2J/2K 4C/5C 6 H /6+
1J/1K 2L/4J 4F/M5t M 10/M +
1K/3A 3B/3— 4H/5N
IL/3 K 3C/4C 5C/6C
1N/2G 3F/4F 5F/M3t

To use the program:

Indicate a geometric shape to M IN IV A C by making connections from M 10 to the top ter­

minals of the various points on the m atrix. For example, to indicate the triangle:

Make the connections:

M lO / M lt
M 10/M 3 t
M 10/M 5t M5

Light 3 will come on indicating that M IN IV A C senses a triangle.

This program for the recognition of geometric shapes is sim ilar in function to the Quantity
Recognition circuit discussed in Book III. A large computer would determine how many points in
its m atrix of sensing elements were being activated by the presence of ink or magnetic m aterial.
W ith the limited size of the m atrix used in the above example, a properly programmed computer
can recognize a particular geometric form.
To recognize printed characters requires an expanded m atrix— one large enough so that
each possible character can be uniquely defined. W e can uniquely define the numbers 0 through
9 using the full game m atrix on the M IN IV A C . The result will be ten forms, each with a unique
recognition code.

21
 FORM CODE FORM CODE

269 235 6 7 8 9

1 26 7 8 9 24689

2345679 24569

12379 23679

23459 2 46 8

C H A R A C T E R FORMS A N D CODES
Notice that not only the form, but also the position of the characters is important. For ex­
ample, the character 1 must be centered in the block of sensing elements.
The experiment which follows uses a more complex variation of the quantity recognition
circuit to recognize arabic numerals which are "draw n" on the game m atrix.
Experiment 12: Arabic Numeral Recognition
In this experiment we will program M IN IV A C to recognize arabic numerals which are
"d raw n " on the m atrix. The nine terminal locations of the m atrix will provide the sensing ele­
ments for the computer. The program will allow M IN IV A C to recognize numerals according to
the code represented above.
The program for the recognition of arabic numerals is:
1C/2C 3C/4C 4H/5J 6H/6—
1F/M9t 3F/M6t 4L/5K 6X/M10
lG/6com 3G/D6 5C/6C 6X/D17
1H/4K 3H/4N 5F/M4t 6 Y /6 +
1J/D0 3J/D9 5G/D5 6com/D4
2C/3C 3K/D8 5H/6G M10/M1 1
2F/M5t 3L/4J 5L/6J D16/D19
2G/D2 3N/D7 5N/D1 D18/M—
2H/4G 4C/5C 6 C /6 -
2J/D3 4F/M8t 6F/M7t

22
Now connect four short programming wires to M 10, leaving one end of each free. Also, connect
four short programming wires to M l 1, leaving one end of each free.
To use the program:
Turn power on. Select a number from 0 to 10. "D raw " this number on the top terminals of
the m atrix with the free wires from M l 0 and M l 1, using the arabic numeral code previously
presented. Push pushbutton 6. The pointer knob of the rotary switch will turn to the number
"d raw n " on the m atrix.
To try another number, be sure that all eight wires to M10 and M l 1 are free. Then "d raw "
the new number on the top terminals of the m atrix and proceed as above.
You have undoubtedly noticed that some of the numbers used in the above experiment are
peculiarly shaped. For instance, the " 8 " is a bit lop-sided. However, for the computer to recog­
nize the numerals, each character must be presented according to the code which the machine
is programmed to understand. This is true of any device designed to "rea d " printing: the device
will read only the particular type face which it is designed to understand.
Since each character must be unique, the type faces which are designed to be machine-read
do not look exactly like the numerals with which we are generally fam iliar. An example of a
specially-designed type face for use in a print-reader is that now used by many banks. In the
illustration below, the characters are printed with a special magnetic ink for sensing by the
reader:

1 -9 8 7
YOUR N A T IO N A L BANK 210

New York, N. Y.. *< 59.

PAY TO THE (/ 4 -S XV/ - . . 70
ORDER OF

.DOLLARS
vo \ 5 ^ — A . B . D e p o s it o r
k W M a r y F . D e p o s it o r

co z lO 'O s a ?■: z zo o a i. z t ?oir i O L Z,' ' 0 0 0 0 0 0 5& 7 0 /

CHECK ABA ACCOUNT PROCESS AMOUNT

R O U T IN G T R A N SIT NUM BER CONTROL.
SYM BOL. NUM BER

B A N K C H EC K SH O W IN G S P EC IA L T Y P E FA C E FOR M A C H IN E R EAD IN G

M achine recognition of printed characters is presently limited to specially designed type

faces. Once the printed information is recognized by the machine, modern computers can rapidly
process this information. Banks now use computerized check processing systems, and the large
credit card companies have found computerized handling of accounts to be a valuable tool. Re­
search is presently being carried out on equipment designed to read any type face, and it is hoped
that in a short time computers will be able to understand printed characters regardless of the type
face.

Examples of Computer Handling of Industrial Problems

Although it is artificial to separate business from industry, we are establishing this distinc­
tion because of the types of problems which are handled by computer systems. The "business"
applications discussed have involved prim arily computer handling of data-processing problems.

23
The "in d u strial" applications which follow deal prim arily with computerized control systems for
production problems.
Experiment 13: Multiple Point Control
A n expansion of the simple "Lig h t on the S ta irs" circuit (see Experiment 1) is used in indus­
trial situations which require control from a number of points. An example of such a situation
would be the problem of handling radioactive materials without approaching them. Intricate sys­
tems have been designed for this particular process; these systems consist of a great number of
complex mechanical devices which allow an operator to handle these dangerous m aterials from
behind safety shields.
V ital to such a remote control operation are, of course, a great number of safety devices—
for instance, emergency cut-off switches as well as emergency "o n " switches for certain safety
mechanisms. For added safety, duplicate emergency switches are generally placed at various
locations— at the actual equipment location, in the safety engineer's office, in the central
power plant, etc. Thus, in case of an emergency, safety devices can be turned on or dangerous
equipment can be turned off from the most accessible location and with the least possible danger
to the employees.
This experiment demonstrates a basic control circuit which permits a single light (which
could represent any safety device) to be turned on or off from twelve different "locations".
The program for a multiple point control system is:
1V/1Y 4A/6K 5E/6G 6A/6E
1W/2V 4B/4— 5F/6F 6 B /6 -
1Z/2Y 4W /5V 5F/5H 6C/6—
2W /3V 4Z/5Y 5G /5+ 6E/6J
2Z/3Y 5A/5E 5J/6H 6H/6W
3W/4V 5 B /5 - 5W/6V 6L/6Z
3Z/4Y 5C/6C 5Z/6Y 6 Z /6 +
To use the program:
Set all slide switches to the R IG H T. Turn power on. Light 4 may be turned on or off by push­
ing any pushbutton or by moving any slide switch to the L E F T and then back to the R IG H T. In ac­
tual practice, these twelve switches would be located at various control points, rather than being
grouped in one place. (Note: lights 5 and 6 are wired into the control circuit.)
Experiment 14: Sequence Control With Manual Operation
Another form of a control system used in an industrial process is one which insures that a pre­
determined sequence of operations is executed in the proper order. This type of system finds
application, for instance, in a situation which calls for the combination of chemical compounds in
a specified order— particularly when an error in the sequence would prove dangerous.
The program below provides constant control as would be required in the situation just out­
lined. The control circuit provides that the first step in the process be taken before the second can
be begun, and so on through twelve consecutive steps. This particular circuit will not permit an
error in sequence to be made. The system will not recognize a step taken out of order. Only the
correct steps given in the correct sequence will be accepted as input. When all twelve steps in the
operation have been completed, light 1 will come on indicating the end of the operation.
The program for a control system with manual operation is:
1A/1G 2F/2G 3H/4G 4com/5L 5S/5X
1B /1— 2F/3K 3N/4N 4R/3com 5T/6T
1C/2C 2H/3G 3com/4L 4S/4X 5Y/6Y
1F/1G 2N/3N 3R/4com 4T/5T 6A/6J
1F/2K 2com / 3 L 3S/3X 4 Y /5 Y 6B/6—
1H/2G 2R/5com 3T/4T 5C/6C 6C/6—
1com/2L 2S/2X 3Y/4Y 5F/5G 6F/6G
1R/6com 2T/3T 4C/5C 5F/6K 6F/6com
1S/1X 2 Y/3Y 4F/5G 5H/6G 6H/6 +
1T/2T 3C/4C 4F/5K 5N/6N 6 R / 1com
1Y/2Y 3F/3G 4H/5G 5com/6L 6S/6X
2C/3C 3F/4K 4N/5N 5R/2com 6Y/6 +

24
To use this program:
Set all six slide switches to the R IG H T and turn power on.
Carry out the twelve steps in the operation by setting the slide switches and pushing the
pushbuttons in this order:
Set slide switch 1 LE FT
Push pushbutton 1
Set slide switch 2 LE FT
Push pushbutton 2
Set slide switch 3 L E F T
Push pushbutton 3
and so on, through pushbutton 6.
W hen you have completed all twelve steps, light 1 will come on indicating "operation com­
plete."
If you attempt to carry out a step in the wrong order, nothing will happen. The system accepts as
input only the proper steps.
To re-set the system, set all slide switches to the R IG H T and turn power off, then on again.

Experiment 15: Automatic Sequence Control

The previous program provides basic control over an operation by requiring that a human
operator perform the required steps in the appropriate order. However, it leaves execution of each
of the steps to the operator.
In some complex industrial processes not only the sequence of operations, but also the tim ­
ing of each step, is crucial. Under such circumstances it is often desirable to use a completely
automated system to control the entire process. W ith the appropriate circuitry, each step in the
operation will be performed in the specified sequence, with the specified delay between steps.
In the following program, pushing pushbutton 6 will start the automatic sequence. Lights 1
through 6 will come on in order, indicating that each step in the sequence has been completed.
When the last light— light 6— comes on, pushing pushbutton 5 will turn the system off.

The program for automatic sequence control is:

1A/6G 2F/2— 4B/4— 5Y/M —

1B /l — 2G/5A 4C/4G 6B/6—
1C/1G 2H/3H 4C/5K 6C/6G
1C/2K 2L/D11 4F/4— 6C/6X
IF/1 — 3A/4G 4H/5H 6F/6—
IG/6A 3B/3— 4L/D13 6H /6+
1H/1 + 3C/3G 5B/5— 6H/6L
1H/2H 3C/4K 5C/5G 6K/D17
2A/5G 3 F/3 - 5C/D15 6 Y /6 +
2B/2— 3G/4A 5F/5— D16/M +
2C/2G 3H/4H 5L/D14 D18/M—
2C/3K 3L/D12 5X/6E

To use this program:

Turn power on. Start the system by pushing pushbutton 6. Lights 1 through 6 will come on
in order, representing the completion of six successive steps. W hen light 6 has come on, push
pushbutton 5 to stop the system.
Notice that the sequence can be stopped at any time by pushing pushbutton 5. T h is is an­
alogous to having an emergency shut-off switch wired into the control circuit.

Control With Feedback

The three previous control circuits dealt with situations in which the sequence of operations
was completely specified. Only the order of execution and the timing of each operation was con­
trolled by computer programming. M any industrial processes, however, require execution of cer­
tain operations which must be determined by conditions occurring during the actual processing.

25
This would be the case if, for example, a specific ingredient were to be added to a chemical com­
pound when the compound had reached a certain temperature.
Control in which information from the process determines execution of successive steps is
known as "control with feedback." Information from the process is "fed back" to the control
mechanism to modify successive operations, or to activate successive operations.
As in the sequence control systems, control systems with feedback can be completely auto­
mated or can be m anually operated. W e can demonstrate a manual control system with feed­
back using the "two-floor elevator control" circuit from Experiment 8. W e will assume that the
process to be controlled is dependent upon the continued oscillation of the rotary switch between
1 and 7.
To start the system, we will set the pointer knob at 1 and push pushbutton 6. This moves
the pointer knob to 7, and light 6 will come on to indicate that the system is now at 7. Light 6 is,
in this case, a "feedback indicator." A s soon as light 6 comes on we must push pushbutton 5
to return the pointer knob to 1. When the pointer knob is at 1, light 5— another "feedback in­
dicator"— comes on, telling us that pushbutton 6 must be pushed to turn the pointer knob to
7 again.
In an industrial situation, lights 5 and 6— the feedback indicators— would have special
meaning:
Light 6 on would indicate that the first of two steps had been executed and pushbut­
ton 5 must be pushed to continue the process.
Light 5 on would indicate that the second of two steps had been executed and push­
button 6 must be pushed to continue the process.
Notice that in this manual control system with feedback, the system indicates to the operator that
an operation has been executed. The operator must then carry out the next operation on the basis
of this information which has been "fed back" to him.
Industrial processes which require split-second reaction to a particular signal are often de­
signed so that the signal will autom atically set off the appropriate next operation. In the illustra­
tion above, this can be done by a program which will autom atically sense the position of the
pointer knob and immediately act to change its position. An example of such a program is the
metronome in experiment 3. The metronome circuit provides constant oscillation between two
points on the rotary switch dial. The operator can change the period of oscillation by moving the
slide switches, but the system autom atically reacts to its own operation.

What is the "Best" Computer?

A s more businesses turn to computerized operations, the question of the "b est" computer
arises more frequently. There can, of course, be no general answer to the question, since each
firm must decide what problems it can best solve with a computer's help and then select the best
system for its particular needs.
There are, however, three basic decisions to be made when the final selection of a computer
system is to be made:
1. How large a computer system is required?
2. W hat speed of operation is required?
3. W hat kind of input the computer must be equipped to handle.
W ith these factors specified, the choice of the best system will be made simpler.
The first factor— size— can be broken down into two considerations: the size of the com­
puter's memory and the size of numbers which the computer will be required to handle. In terms
of memory size, the basic question to be answered is, "How much information will the system
have to handle at any one tim e?" A firm with 1000 employees may want a system capable of
processing the total payroll in a single operation; but if management is willing to process the
payroll in two sections, a smaller system— one with less memory— could be used.
The size of numbers which the system will be required to handle raises the question of ac­
curacy. If the system is to be used in applications where ten significant figures are sufficient, it

26
would be unnecessary to pay for the additional size of a machine capable of producing figures ac­
curate to twenty significant digits.
In general, the larger the capacity of a computer system, the more it will cost. The capa­
city requirements must be accurately determined if the system is to be a practical one. And it
should always be kept in mind that either in sufficient or excess capacity will be expensive.
The second factor— speed— is one which must be considered from two aspects: actual proc­
essing speed and input-output speed. Computer systems can be designed with high speed input-
output units connected to high-speed processing units. For a firm which plans to enter large quan­
tities of input, perform a large number of calculations on its data input, and which will require
output immediately, a high-speed system would probably do the job most economically.
However, in an application where a great deal of calculation is to be done on a limited
amount of input, with a limited amount of output, the high-speed input-output units would not
be practical. To determine the required speed of the system— and hence the required speeds of
the individual units of the system— the following should be accurately determined:
1. How much work must be performed in a given amount of time?
2. How quickly must output be received?
3. W hat is the ratio of calculations to communications?
4. Is the limiting factor in the operation calculating speed or communication speed?
When these questions have been answered, the speeds of the system's functional parts will
be known and a complete system can be determined. W ith an efficient computer system, there
should not be delays while one segment of the system catches up with another segment— such de­
lays cost money.
The appropriate input-output devices can often determine the ultimate practicality of a com­
puter system. For a great number of operations, punched cards are completely adequate. How­
ever, operations which require feeding large amounts of information into a master file of some
type may find the added expense of magnetic tape well worth the cost. Certain specialized cases
— where information is to be obtained directly from an operating process, or communicated over
telephone lines from distant points— will require specially designed input-output devices which,
although expensive, will be worth their cost through time saved.
The form of the output is particularly important since it must present the results of the
computer's work to those who will make use of it. The permanency, accessibility and comprehen­
siveness of the output must be determined before final selection of the correct unit can be made.
Deciding upon the best system for a given operation— or set of operations— is just as im­
portant as the decision to computerize the operation in the first place. Whether the best system
is to be simply a few small units or a complex combination of many high-speed, specialized units
can only be determined on the basis of the job to be accomplished. When correctly selected, a
computer system can be a valuable asset to a business organization. When incorrectly selected, a
computer system can be expensive and time-wasting.

3. COMPUTER APPLICATIONS IN SCIENCE AND TH E M ILITA RY

Introduction
A s in business and industry, computers find wide applications in basic problem-solving opera­
tions for science and the m ilitary. And, the characteristics of computer problem-solving which
influence the decision to use or not to use a computer in the business situation are equally applic­
able in the m ilitary or scientific situation. T h at is, computer problem-solving is normally most ef­
fective when it can be applied to a repetitive process to obtain solutions of a consistent type
using relatively large amounts of data.
However, in the areas of science and the m ilitary computers find particularly valuable ap­
plication in the solution of unique problems. The simulation technique discussed in Book III is
frequently used, for instance, in the solution of single problems which would otherwise require
many months— even years— of tedious calculation by mathematicians and scientists. Because of
the specialized nature of the problems to be solved, computers are often used by scientists and

27
members of the m ilitary for projects which would be uneconomical in the business world. A prob­
lem involving a defense project, for example, cannot be evaluated in terms of cost alone; pro­
gramming a unique m ilitary problem for computer solution may be well worth the expense if the
results will affect the progress of the national defense effort.
A characteristic of computers which is particularly useful in scientific and m ilitary work is
their ability to perform extremely rapid calculations with absolute accuracy. This means that a
properly programmed computer can, in a matter of minutes, examine and analyze hundreds of
alternative solutions to a given problem. For situations in research where the only reasonable ap­
proach is a "trial-and error" method, the computer's capabilities are making possible the solution
of previously "im possible" problems.

Real-Time Problem Solving

In experiment 7 in this book we briefly examined "environmental sensing." An expansion

of this function is often applied with great success in the handling of scientific and m ilitary prob­
lems. In general, the situations which we have examined have involved feeding data into a com­
puter system— data in the form of mathematical and/or logical relationships, quantitative re­
sults from operating processes, etc. Real-time problem solving differs from these situations in
that the data is supplied directly to the computer from its environmental and the computer is
instructed to act upon this information immediately and provide recommendations for action
or analyses of the environmental situation.
In real-time problem solving, the computer is given the capacity to directly sense its en­
vironment without requiring a human intermediary to supply the environmental information.
For example, a computer can be used in conjunction with a radar tracking system so that as soon
as the tracking system identifies an object in its path the computer will immediately give an
alarm and begin plotting the location and movement of the object. A coordinated system such
as this provides constant watch on the "environm ent" (the scope of the radar tracking equip­
ment), yet does not require constant scrutiny by human operators.
Computer systems which are used for real-time problem solving in conjunction with spe­
cialized input equipment can greatly increase the value and utility of that equipment.

Computer Handling of Scientific Problems

Computers are rapidly becoming as fam iliar— and valuable— tools to the scientist as the
vacuum pumps, power supplies, intricate glass tubing and other equipment vital to the experi­
mental stage of a research project. It would be an impossible task to even list the scientific prob­
lems and projects which are presently being programmed for computer solution.
This does not mean, of course, that computers have all the answers to the problems. The
computer is simply a valuable tool which permits scientists to examine more information more
accurately, and with far greater speed than has ever before been possible.
Scientists use computers and computer techniques at every level of sophistication— from
simple tabulation and correlation of data to extremely complex simulation programs for com­
plicated analyses. For example, research teams have simulated the atom on computer systems in
attempts to predict the locations and actions of atomic particles. In other scientific fields, vast
amounts of data taken from thousands of observations— of the earth's structure, of the flow of
ocean currents, of the paths of molecular particles— are being analyzed in attempts to find the
basic concepts and relationships which govern their actions.
The field of engineering— which is becoming increasingly indistinguishable from what was
previously known as "pure science"— has put computers to work on its particular types of prob­
lems with great success. Engineers find the simulation technique described in Book III especially
useful: they are now able to test designs safely, and often without the cost of building a test model.
Technically speaking, the actual design of computer systems falls under electrical engi­
neering, while the design and development of the various computer techniques falls between
electrical engineering and pure mathematics. However, as is often the case in scientific and en­
gineering work, there is a considerable amount of tra ffic among the various specific fields. A phys­

28
icist may find it necessary to develop a special technique to handle a problem; and this technique
may well be one which can be effectively used by a chemist or a metallurgist. On the other hand,
an electrical engineer may develop a technique which proves valuable to a biologist.
A s a basic scientific instrument, computers have only begun to prove their value. A s scien­
tists and engineers use this new tool, they will make it even more valuable by expanding its ca­
pabilities and finding new uses for it.

Examples of Computer Handling of Military Problems

Computers are, of course, used by the m ilitary to handle operating problems, just as they
are used in the business world. Basic research projects carried out under m ilitary contract re­
quire the use of computer systems as indicated above. However, there are many computer appli­
cations which are unique to the m ilitary— particularly applications which concern national de­
fense. The experiments and discussion which follow illustrate only a few of the applications pres­
ently in use.

Signal Systems:

W e have mentioned the use of a computer to signal a human operator when an operation
has been completed, or to notify the operator of an unusual processing situation. In these cases,
though, the signal is only an indication that something has occurred: the signal is important
only in that it alerts the operators to the occurrence. In a m ilitary situation, however, the signal
itself becomes important. Systems are often required which w ill, upon receipt of a specific input,
produce a particular continuous signal.
The complexity of a computerized signal system is a function, basically, of the required in­
puts and outputs. The actual input unit may be as simple as a button to be pushed when the
system is to be started— as, for example, an automatic SOS transm itter. On the other hand, the
input unit can be a sophisticated radar system which produces a particular input when an object
passes into its scope. In any case, the actual processing unit of a signal system is designed to:

1. Recognize a particular input as a "s ta rt transm ission" command.

2. Search its memory for the appropriate signal.
3. Feed the signal into a designated transmitting device.

Once the signal system has begun its cycle, it will continue to transm it the signal until it
reaches "end transm ission" command— which can be pre-programmed into the system or can be
given by a human operator. The output from an automatic signal system can vary in complexity
from a simple blinking light to a coded message directed to distant points.
The three experiments which follow provide examples of the programming necessary to
processing units used in automatic signal systems.

Experiment 16: Automatic Message Transmission

Th is circuit uses the turning of the rotary switch to generate a sequence of dots and dashes
representing the amateur radio call "C Q " in Morse Code. Once turned on, this system will con­
tinue to "tra n sm it" the CQ call until it is turned off.

The program and circuit diagram for the automatic CQ transm itter are:

29
 C IR C U IT D IA G RA M — A U T O M A T IC CQ T R A N S M IT T E R

Program:
5A/D3 6B/6— 6H /6+ D6/D10
5B /5- 6C/6— 6H/6Y D8/D9
5X/6E 6F/6G D3/D5 D9/D11
5Y/6Y 6F/D17 D4/D6 D16/D17
6A/D4 6G/6X D5/D8 D18/M—

The Morse Code representation for CQ is:

C: — •— ■

Light 5 on represents a dash.

Light 6 on represents a dot.

To use the program:

Set the rotary switch at 0 and turn power on. Push pushbutton 6. This provides the input
command "start transm ission." As the rotary switch turns, lights 5 and 6 will flash the CQ signal.
To stop transmission, push pushbutton 5.

Experiment 17: Automatic Name Transmission

This program presents a slightly more sophisticated version of the transmission technique
used in the previous experiment. The name to be transmitted is "JO H N ", and it will be sent in
two parts: first " J O " , then "H N ".
To send the entire name requires a "selective sequencing" circuit. The computer must de­
termine whether it is in the first or second path of the 2-path cycle, and produce the output ap­
propriate to each path. Th at is, on its first revolution the rotary switch will instruct the com­
puter to transm it " J O " ; on its second revolution it will instruct the computer to transm it " H N ."
A s in the previous experiment, will be transm itting in Morse Code. The Morse Code repre­
sentation for JO H N is:

H : -----
N: — •

30
The program for this automatic transm itter is:
3C/4C 4F/5N 5L/D0 6H /6+
3F/4G 4H/5H 5X/6E 6H/6Y
3G/3K 4K/5F 5Y/6Y 6com/D2
3G/5com 4L/D15 5com/Dl 1 D3/D4
3H/D12 4N/5E 6A/6com D4/D5
3J/6com 5A/5com 6B/6— D5/D12
3L/D13 5B/6B 6C/6— D16/D17
4C/5C 5C/6C 6F/6G D18/M—
4E/5K 5F/5G 6F/6X
4F/4G 5 H /5 + 6G/D17

Light 5 on represents a dash.

Light 6 on represents a dot.
To use the program:
Set the rotary switch at 0 and turn power on. Push pushbutton 6. The system will immediately
begin transm itting "JO H N " in Morse Code. Notice that as the rotary switch turns the computer
alternately transmits " J O " and " H N ." The selective sequencing circuit causes the computer to
select the correct segment of the "m essage" for each part of the transmission.
To stop transmission, push pushbutton 5.
Different code words can be transmitted using this basic circuit by re-programming the con­
nections to the rotary switch.
Experiment 18: Automatic Transmission with Differential Spacing
The preceding experiments used different lights to indicate dots and dashes. It is possible,
however, to produce "differential spacing" so that dots and dashes may be distinguished by the
length of time a single light is on.
The program below illustrates a circuit with differential spacing by autom atically transm it­
ting the "SO S" distress signal in Morse Code:
3E/3H 5E/D19 6G/D4 D7/D8
3F/3 - 5F/6K 6H/D16 D8/D11
3J/6A 6A/D1 6L/D18 D11/D12
4C/5C 6B/6— D1/D2 D12/D13
4C/4F 6C/6— D2/D3 D16/D17
4E/5E 6E/D9 D3/D6 D17/M +
5C/5F 6F/6G D6/D7 D18/M—

The Morse Code representation of SOS is:

S; . . .
0 : -------
S; . . .
To use the prgoram:
Turn power on. The system will immediately begin transmitting "SO S ". Light 6 and relay
3 will both indicate dots and dashes: they will remain on longer for a dash than for a dot. The
computer will continue to transm it "S O S " until the power is turned off.
Computerized Code Systems:
A computer application of particular value to the m ilitary is the processing of messages
transmitted in code. Th is application has three distinct aspects:
1. The routine translation of messages transmitted in "know n" codes.
2. The more d ifficu lt translation of messages received in "unknow n" codes.
3. The generation of new codes.
Routine translations are handled in much the same fashion as was done in the binary-deci­
mal converters. A computer is simply programmed for direct conversion. If the system is handl­

31
ing more than one code, a technique known as "table-lookup" is often used. W ith "table-look-
u p", the computer holds the various codes in storage until directed to use a particular one. To
translate, it "looks up" the appropriate substitution for each letter. The advantage of a compu­
terized system for routine code translation is, of course, the speed with which the translations
are carried out.
Translation of "unknow n" codes— that is, "b reaking " codes— can often be effectively
handled by a computer system. If the "unknow n" code was generated by a computer, another
computer system will be required to break that code.
If the "unknow n" code was originally man-made, the computer attempting to break the
code will use a trial-and-error method based on the frequencies of appearance of the letters in the
coded message. One type of code which a computer cannot break is one which does not use sub­
stitution of letters, but depends instead on phrases with special meaning only to the receiver for
whom they were intended. Breaking codes of this type requires "inside inform ation" which a
computer cannot provide.
A n example of a man-made code which a computer could easily break would be a substitu­
tion code based on a simple three-letter shift. T h a t is, A is D, B is E, C is F, etc. An example of
a man-made code which a computer could not break would be the following set of messages which
would be known only to the transm itter and the receiver:

"Roses are red" will mean " A ll going as planned"

"V iolets are blue" will mean "There is a change in plans"

The ability of computer systems to generate codes requires that computer systems be used
in any attempts to break those codes. Codes generated by computer system are usually the result
of the computer's ability to generate pseudo-random numbers. Breaking a computer-generated
code is accomplished through sophisticated statistical techniques, using as a basis the fact that
no computer can actually generate completely random numbers.
The two experiments which follow provide examples of simple coding and decoding programs
for the international Morse Code.

Experiment 19: Encoder for Morse Code

This experiment will use M IN IV A C to provide automatic letter-by-letter translation into
Morse Code. The international Morse Code is:

A •— N— •
B — • •• O ------------
C — •— • P • ------- •
D — •• Q --------—
E • R •— •
F •• — • S •••
G • T —
H U ••—
I •• V • • •—
J • w • -----
K — •— X — •—
Y — • -------
M Z -------- • •

The program which follows will autom atically display the Morse Code representation of any
letter entered on the rotary switch. Output lights 3 through 6 will be the display lights.

32
The program for an Encoder is:
2C/3B 3G/D8 4com/5G 6F/D3
2C/3C 3H/3L 5A/5com 6G/6com
2F/D15 3L/4L 5B/6B 6H/6L
2G/4F 4A/4com 5C/6C 6H /6+
2H/2L 4B/5B 5F/D13 6com/D1
2H/3H 4C/5C 5H/5L D1/D9
2K/5F 4F/D14 5H/6L D2/D10
3A/3K 4G/4com 5K/6com D3/D11
3B/4B 4H/4L 5com/6K D4/D12
3C/4C 4H/5U 5com/D2 D5/D13
3E/D0 4K/5com 6A/6com D6/D14
3F/3G 4com/Dl 2 6C/6— D7/D15
D16/M +
To use this program most easily, the rotary switch dial should be fitted with a dial plate
made by tracing the illustration below on a piece of paper.

MORSE CODE: A L P H A B E T IC D IA L PLA T E

33
To use the program:

Set the rotary switch dial to 0 and turn power on. Turn the rotary switch clockwise until the
pointer knob is pointing at the letter you wish to code. Read the Morse Code representation of
the letter starting with output light 3. If the letter you wish to code is in band 1 of the dial plate,
read only light 3. If the letter is in band 2, read lights 3 and 4. If the letter is in band 3, read lights
3, 4, 5. If the letter is in band 4, read lights 3, 4, 5 and 6.

A light on represents a dash.

A light off represents a dot.

Experiment 20: Decoder for Morse Code

T his experiment uses the rotary switch dial plate described in the previous experiment. Let­
ters entered in Morse Code on the slide switches will be indicated on the rotary switch.

The program for a Decoder is:

1A/1T 1W/4A 3G/D15 5J/D3

1B/1C 2B/3B 3J/D7 5K/D10
1B/2B 2C/3C 3K/D14 5N/D2
1C/2C 2F/4R 3N/D6 5S/6R
1F/2F 2G/4H 3S/4S 5V/6T
1G/3H 2H/5T 4C/5C 6C /6-
1H/5R 2J/6H 4F/5F 6G/D9
1J/5H 2K/4L 4G/D13 6J/D1
1K/3L 2L/5W 4J/D5 6K/D8
1L/5U 2N/6L 4K/D12 6N/D0
1N/5L 2 V /2 + 4N/D4 6S/6—
1R/3A 3B/4B 4S/6X 6X/D17
1S/2U 3C/4C 5C/6C 6 Y /6 +
1U/2A 3F/3R 5F/6F D16/D19
1V/2W 3F/4F 5G/D11 D 1 8 /M -

To use the program:

W ith the dial plate from experiment 19 in place, turn power on. Set slide switches 1 and 2 to
indicate in binary the number of Morse Code characters (dots and dashes) in the letter you wish
to decode:

Number of Characters Slide Switch 1 Slid eS w ith 2

1 0 (right) 1(left)
2 1 (left) 0 (right)
3 1 (left) 1(left)
4 0 (right) 0 (right)

Starting with slide switch 3, enter the dots and dashes according to the convention:

a slide switch R IG H T represents a DOT.

a slide switch L E F T represents a DASH

Push pushbutton 6: this gives the computer the instruction "read out." The rotary switch will
turn to the correct spot on the dial. Read the alphabetic letter from the dial plate as follows.

Read the letter in Band 1 if output light 1 is on.

Read the letter in Band 2 if output light 2 is on.
Read the letter in Band 3 if output light 3 is on.
Read the letter in Band 4 if output light 4 is on.

34
Experiment 21: "Search and Track" Radar
In the previous section of this book the industrial uses of feedback were discussed. The im­
portance of feedback in enabling a machine to deal directly with its environment makes it es­
pecially useful in m ilitary applications. An excellent example of the use of feedback in a m ilitary
situation is provided in the basic operation of a "search and tra c k " radar system used to detect
and follow missiles and satellites.
A "search and tra c k " radar system operates in two distinct modes: the "sea rch " mode and
the "tra c k " mode. When operating in the search mode, the system constantly scans its environ­
ment, signalling the operator when it senses an object in the path of its scan. When an object
is sensed, the system can be set in the track mode. Operating in the track mode, the system will
find the object and continue to follow it if it moves.
W hile in the search mode, the radar system is a simple signalling system. Once in the
track mode, the system utilizes feedback control to keep the desired object in "vie w ." The actual
operation of the system while in the track mode is as follows:

The system scans the horizon until it picks up the signal which tells it there is an object in
its path. It sweeps past the "object sig nal" until it has lost it and then reverses its direction,
returning to the object signal. It again sweeps past, reverses and continues in this fashion
until given other instructions or until the object signal is out of range.

The feedback in this case is information about the presence or absence of the object signal. The
control action initiated by a computer system is to reverse the direction of scan immediately
when the object signal is lost.
The program below illustrates a system which operates in two modes: a search mode and a
track mode. The command to change mode must be given m anually with this program; real
search-and-track systems are often equipped to change modes autom atically.

The program for a search-and-track system is:

1A/2E 2J/3H 3E/3J 5F/6G

IB/1— 2K/3N 3 K /3 + 5H/6H
2A/3E 2K /2— 3L/D18 5L/6J
2B/2— 2L/D17 4C/5C 6A/6V
2C/3C 2N/3K 4F/5K 6B /6 -
2E/3G 3A/5E 4G/5E 6E/6—
2F/3F 3H/4F 4H/5H 6F/6U
2F/2H 3B /3- 5C/6E 6H /6+
2G /2+ 3C/4C 5F/5G 6V/D16
M + /D1, 2, etc.

* Connection can be made between M + and any point on the rotary switch dial.

To use the program:

Set slide switch 6 R IG H T and turn power on. The system is now operating in the search
mode. Light 3 will flash on whenever the system detects an object signal (a connection on the
rotary switch dial). You may move the object signal by changing the connection on the rotary
switch dial.

35
 Set slide switch 6 LE FT . The system is now operating in the track mode. Notice that the
rotary switch oscillates about the object signal. If you now move the object signal, the system
will continue to search until it finds the new location and will then resume tracking it.
Countdown Control:
One of the most highly publicized applications of specialized computer circuitry is the count­
down used so dram atically on rocket test flights. The countdown is carried out not for effect, but
because of the complexity of the equipment under test— as well as the complexity of the total
test operation.
A n actual countdown begins many months before "zero hour", and it is only at the final
stages that specialized computer circuits are used for control. A computerized countdown con­
trol system is in essence a sequence control circuit, (see experiment 14) The inputs are the
results of final checks; the output is either an "O K — step X " signal or an "E R R O R " signal which
immediately alerts the operators to the error and may, if further steps depend directly on the
step in error, stop the countdown. The firing button is so programmed that it will not send the
command " fir e " until every step is correct.
Totally computerized check-out systems are built in to real quick-fire missiles. A s soon as
the " fir e " command is given, the computer system begins to check out the individual parts of
the missile, matching the check results against the pre-programmed requirements. If the com­
puterized control system finds an error in the missile system, it autom atically stops the firing pro­
cedure and signals an alert to the operators informing them of the location and nature of the
error. The speed with which such control systems function makes it possible to reduce the count­
down time on operational missiles to a few minutes.
The basis of any control system of this type is the simple A N D circuit. Only after each in­
dividual switch is closed, through receipt of the correct input, will the " fire " command be executed.

4. COMPUTER APPLICATIONS IN THE SO CIA L AND POLITICAL SCIENCES

Vote Registering Machines

A computer-like device now fam iliar to many is the vote registering machine being used by
some state legislatures. This device autom atically registers the total vote on an issue without
revealing the identity of the individual voter. A s such, it is used in place of the "voice vote."
The basic circuit used in this vote registering machine is the Quantity Recognition circuit
discussed in Book III. The inputs are supplied by the individual members of the legislative body,
each of whom has a set of buttons with which he can indicate "y e s," "n o " or "a b sta in ." Only the
totals appear as output— usually displayed on a light board in the front of the room.
Although they "count votes," these vote registering machines are not functionally sim ilar
to the voting machines used in public elections. The vote registering machines use a basic com­
puter circuit; the voting machines simply tabulate successive inputs.

Computers and Election Predictions

The national elections have spotlighted the capabilities of computers as predictors. Election
prediction programs take advantage of the computer's ability to handle large amounts of data and
to perform extremely rapid calculations. M any specialized techniques are used to extrapolate
from early election return to the final outcome. However, the basic system is essentially a process
of comparing early returns with historical data and performing various statistical operations to
predict the final result.
W hen a computer is programmed for election prediction, it is fed information concerning
the historical pattern of returns in certain precincts. These precincts are chosen on the basis of
the relationship between returns in these precincts and returns in key national areas. A s the re­
turns from these precincts come in on election night, the computer compares the current ratio
of returns to historical ratios. An extrapolation is then made on the basis of these comparisons.
The results of such a prediction scheme are stated as estimated conclusions with a speci­
fied degree of "confidence"— that is, a specified probability of occurrence of the predicted out­

36
come. When a computer produces a predicted result of this type, it is always very important to
note the degree of confidence associated with that result. As the computer receives more and
more information, its predicted results will have increasing degrees of confidence. The 100%
degree of confidence is reached as a lim it when the computer is given all the information— that is,
when all returns are in.
The accuracy of the computer's predictions is a direct function of the program used to arrive
at that prediction. The programs to predict the outcome of national elections are extremely in­
volved mathematical statements designed to take into account as many factors as possible. Built
into these programs is a vast amount of information about the social, economic and political
status of samples of the key precincts and sim ilar information about the country as a whole. In­
formation about the candidates and their parties is also part of the total program, as well as
information from previous elections concerning the vote-attracting characteristics of candidates
and parties in various sections of the country.
Since a computer will accept only quantitative information, the hundreds of qualitative fac­
tors influencing a national election must be correctly stated in quantitative terms for inclusion in
the total program. The analysis of these qualitative factors is in itself valuable work which is fre­
quently made easier and/or speeded up by the use of computers. The knowledge gained from an
attempt at one election prediction is not limited solely to knowledge of whether or not the pro­
gram gave the correct result. Knowledge of indicative factors, changes in voting habits, shifts in
attitudes towards national issues— all are valuable parts of the results of an election prediction
program.

Language Translation
W e have already examined a particular type of translation by a computer— the translation
of messages into code. In terms of the circuitry involved in translation with a pre-determined
code, this is merely a Ietter-by-letter substitution scheme. Translation between languages, how­
ever, requires that the computer be programmed to perform symbol-to-symbol substitution and,
in addition, be capable of performing shifts in word order whenever necessary.

Symbols and Meaning

Before taking up the computer's problems in symbol-to-symbol substitution, let us briefly
consider the nature of communication processes. In a normal day we encounter thousands of d if­
ferent symbols, each of which conveys a p articular message. The red light of a tra ffic signal con­
veys the message "stop ." The telephone bell indicates that someone is calling. Reading the word
"w ater" brings a mental picture of a clear liquid made up of two hydrogen molecules for each
oxygen molecule. The skull and cross-bones on a bottle warn that the contents are poisonous.
In each situation, we autom atically translate a symbol into a message with particular mean­
ing. W e translate a color, a sound, a group of letters, or a picture into a meaningful idea. To a
baby, though, these symbols have little meaning. The baby has not yet learned to associate a
specific meaning or concept with these symbols; thus, the baby cannot "tra n slate" the symbols.
Teaching a baby to associate meanings with symbols is analogous to developing a computer
program which will translate languages.
W e must "te a ch " the computer to associate the correct meaning with a given symbol—
and we must, just as in teaching the baby, teach each symbol in its proper environment. A baby
will learn that a ringing bell does not always mean that someone is calling. A computer must be
taught that certain symbols have different meanings depending upon the situation, or it will not
be able to perform adequate translations.

Translations of Symbolic Combinations

Of the three successive stages of machine translation, we have already examined the first
two:
1. Letter-by-letter translation as performed through the circuitry of the Morse Code de­
coder and encoder.

37
 2. Translation of a pictorial symbol as performed through the circuitry of the A rabic
Numeral Recognition program.
The third and most complex form of machine translation is that required for language trans­
lation: translation of one set of symbolic combinations into another. A t this stage, the combination
of symbols becomes important. The basic problem of interpretation between two symbolic sets
is complicated by the "environm ent" of the symbols. T h a t is, a given symbol may have different
meanings depending upon the other symbols associated with it.
Perhaps the easiest way to see some of the problems inherent in machine language transla­
tion is to develop a few simple translation programs. W e will do this using English and German.

Simple Translations between English and German

A s a first step in developing a computer program for language translation, we will require a
"d ictio n ary." This will be ours:

EN G LISH GERM AN
Pete Piet
and und
Eve Eva
play spielen
plays spielt
with mit
his seinem
her ihrem
their ihrem
ball Ball

Each English word or symbol has a German word or symbol corresponding to it and indicat­
ing the same idea or concept. In the simplest case, translation would involve merely a series of
direct substitutions. W e could program this elementary case by wiring pushbuttons to lights and
letting the pushbuttons represent English words and the corresponding lights represent the cor­
responding German word. W e could then translate from English to German by entering English
input (pushing pushbuttons) and reading the German output from the lights.
However, we could not simply reverse the process to translate from German to English.
Notice that the German "ih rem " translates as either"h e r" or"th e ir." The simple program out­
lined above will not produce an adequate translation; we must add to the program a rule for
selecting the appropriate English translation of "ih re m ."
The basic rule which we must teach the computer is: a plural subject requires the use of
"th e ir" as an adjective; a singular feminine subject requires the use of "h e r" as an adjective.
T his can be stated logically as:
P X E = their and E X P = her
If we now let the pushbuttons and lights represent words as follows, we will be able to pro­
gram the rule for proper selection of an adjective, given the subject of the sentence.

Output light 1 on represents "th e ir"

Output light 2 on represents "h e r"
Output light 3 on represents "P ete"
Output light 4 on represents "E v e "
Pushbutton 1 represents "ih rem "
Pushbutton 2 represents "P ie t"
Pushbutton 3 represents "E v a "

The program and circuit diagram are:

38
 " t h e ir " "h er" " P e te " "E ve"

C IR C U IT D IA G RA M — SELEC TIO N R U LE FOR T R A N S LA T IO N OF "IH R E M "

Program:
1A/3G 2A/3J 3A/3C
IB/1 — 2B/2— 3 B /3 -
1X/3H 2X/3C 3B/3F
1Y/3X 2 Y /3 Y 4B/4—
1Y/4A 2Y/2 +

To use this program:

Select a singular or plural subject by pushing pushbuttons 2 and/or 3. Enter the adjective
"ih rem " by pushing pushbutton 1. The English Translation of the German input will appear in the
output lights.

To further complicate matters, consider the expansion of our limited vocabulary to include
the past perfect as well as the present tense of the verb. Direct translation of the past perfect
tense is:

has played— hat gespielt

have played— haben gespielt
However, in German the word order changes when the past perfect tense is used. The auxiliary
verb is placed as usual within the sentence, but the associated past participle must be placed at
the end of the sentence. For instance:

Eve and Pete have played with their ball.

Eva und Piet haben mit ihrem Ball gespielt.
Thus, a translation program must be able to recognize the existence of the past perfect tense and
modify the word order accordingly.
By limiting the number of words permissible in a sentence, we can develop a program to im­
plement this requirement. Different slide switches will be used to communicate the English nouns
and verbs. In actual translating programs, the gramm atical identification of each word is sup­
plied through an au xiliary coding scheme. Positioning will be accomplished by programming the
German sequence so that "h a t" and "h aben" are required to appear after the subject and before
the predicate if and when they occur.
The program which follows incorporates the word order requirements of the past perfect
tense with the selection rule for the appropriate adjective. The slide switches will be used to pro­
vide information as to the players and whether they are playing now or played at some time in the
past. The correct German expression of the English idea expressed will be read from the output
lights and the relay indicator lights.

39
The translation program is:
1A/2V 1S/2S 2S/3S 3F/6A
IB / IE 1T/2V 2T/3L 3F/3R
IB/2B 2A/3G 2W/4A 3S/3 +
1C/2U 2B/3B 3A/3N 4B/5B
1C/4C 2C/3K 3B/4B 5A/5 +
1E/2E 2E/3E 3C/3J 5B/6B
1R/2U 2R/3H 3E/4E 6B/6—

Slide switch 1 R IG H T indicates "P ete"

Slide switch 1 L E F T indicates "E v e "
Slide switch 2 R IG H T indicates "Pete or Eve"
Slide switch 2 L E F T indicates "Pete and Eve"
Slide switch 3 R IG H T indicates present tense
Slide switch 3 L E F T indicates past perfect tense

The German sentence will be read in this sequence as follows:

Output light 1 on represents "P ie t"
Relay light 1 on represents "E v a "
Output light 2 on represents "haben m it"
Relay light 2 on represents "h a t m it"
Output light 3 on represents "spielt m it"
Relay light 3 on represents "spielen m it"
Output light 4 on represents "seinem "
Relay light 4 on represents "ih rem "
Output light 5 on represents " B a ll"
Output light 6 on represents "gespielt"

To use the program:

Enter an English idea by moving slide switches 1, 2 and 3 appropriately. Read the German
sentence from the lights as indicated above. The German sentence will be in the proper
order and will be complete except for the conjunction which must be supplied if Pete and
Eve are indicated as input: supply "u n d " for "a n d "; "oder" for "o r."
From these simple examples you can see the problems encountered in a large-scale trans­
lation program. In addition to the grammatical problems which can be solved by converting the
rules of grammar to logical statements for programming, a large-scale translation program must
also be capable of recognizing the subtle shades of meaning conveyed by various words and
phrases.
Present work on the development of language translation programs is by no means complete.
Various research teams are at work on different languages, attempting to perfect present pro­
grams and developing new and better programs for the rapid translation of languages.

Job Selection
Computers' ability to perform high-speed comparisons has led to their application in per­
sonnel work, matching the qualifications of individuals with the requirements of specific jobs.
In applications of this type, the results of computer comparisons are generally used to assist
the personnel staffs of very large firm s who find it necessary to process many job applications
in order to select people to fill specific jobs. Seldom, if ever, are the computer's results taken as
the final word on who is to be hired. The true value of the computer in this application is to point
out those best qualified for a particular job— in essence, a narrowing-down procedure.
A computerized selection program requires two inputs for comparison:
1. The requirements of the specific job
2. The qualifications of the individual applicant

40
The job requirements are supplied by the personnel department. The applicant's qualifications
are provided by asking the applicant to fill out a specially designed questionnaire. Such a
questionnaire might look like this.

PERSO N AL Q U A L IF IC A T IO N FORM
Check only one
1. W hich would you rather do: for each question
a. Add a column of numbers a.
b. Count the number of items in a row b.
2. W hich would you rather do:
a. W rite a short story a.
b. Paint a bookcase b.
3. W hich would you rather do:
a. Be responsible for selling 100 items by yourself a.
b. Be responsible for selling 500 items with four other
people b.
4. W hich would you rather do:
a. Set type for a printing press by hand a.
b. Correct and rewrite articles for a newspaper b.
5. W hich would you rather do:
a. M ake final decisions and be responsible for the conse
quences a.
b. Make recommendations to the person above you b.
6. W hich would you rather do:
a. Be responsible for hiring and firing close business as
sociates a.
b. Be responsible for preparing the payroll for a company b.

The job requirements information is generally in standardized form, with certain key fa c­
tors specified. Some organizations use a standard job evaluation form, an example of which
appears below:

JOB EV A LU A T IO N FORM
check only one
1. Does the job require more: for each question
a. Accuracy and attention to detail a.
b. Speed with minimum accuracy b. ---------
2. Does the job require more:
a. Creativity a.
b. Routine performance b.
3. Is job performance more dependent upon:
a. M anual dexterity a.
b. Working cooperatively with others b.
4. Does the job require more:
a. Direct supervisory activity a.
b. Mental effort b. ---------
5. Does the job require:
a. M ajor responsibility for policy decisions a.
b. Minor responsibility for policy decisions b.
6. Does the job require:
a. Direct supervisory activity a.
b. Little or no supervisory activity b.

41
 Using the two forms above, we can write a comparison program designed to match individual
qualifications with job requirements. The questionnaires above are, of course, much simpler than
any in actual use.
The program makes a direct comparison between questions on the two forms. When an ap­
plicant's answer is consistent with a particular job's requirements, a light will come on. The best-
qualified applicants for the job will be those for whom the most lights come on.

The program for section 1 is:

1A/1S IY/1 + 1T/1Z

IB /1 — 1R/1X

Repeat this program for sections 2 through 6 on M IN IV A C .

To use the program:

From a completed Job Evaluation form, enter the job requirements as follows:

Set the slide switch corresponding to the question number

L E F T if the answer was " a "
R IG H T if the answer was "b "

To check an applicant's qualifications, enter his replies to the Personal Qualifications form as
follows:

Push the pushbutton corresponding to the question number for each question which was
answer " a " . If the question is answered " b " do not push the corresponding pushbutton.
A s the answer to each Personal Q ualification question is communicated to the machine
note whether the corresponding light goes on or off. (Note: some of the output lights will be
on when only the job evaluation information has been indicated. Record these lights only if
they do not go off when the applicant's replies are entered.) A fter the replies to all six ques­
tions have been tested, record the total number of lights that went or stayed on while the ap­
plicant's qualifications were tested.

Those persons most highly qualified for the specific job will be those with the greatest
number of positive comparisons with the job requirements as indicated by the number of lights
recorded.)

Mate Selection
The job selection technique has been used on one well-known television program and in some
social science situations to provide a means of comparing the interests and orientations of a
number of couples to determine their compatibilities. The assumption in this case is that com­
patibility will determine m arital happiness.
The job selection program can be used to test the compatibility of a couple by changing the
questionnaires. When used as a mate selection program, the replies of one individual are entered
on the slide switches and the replies of the second individual are checked against the first indi­
vidual's replies using the pushbuttons. The number of lights recorded as on will give an indication
of compatibility. (Total compatibility is indicated by all lights recorded as on.)
Two questionnaires are presented below as examples of the characteristics which are com­
pared in this type of program. The replies to questionnaire A are first compared, then the replies
to questionnaire B. The total compatibility figure is reached by adding the results of A and B.

42
 Q U ESTIO N N A IRE A
check only one
for each question
1. Which is more important to you:
a. A person's level of education a.
b. A person's physical appearance b. ---------
2. Do you feel that you have:
a. A liberal political orientation a.
b. A conservative political orientation b. ---------
3. Would you prefer to have:
a. A large fam ily a.
b. A small fam ily b.
4. Would you prefer to live in:
a. An urban area a.
b. A suburban or rural area b.
5. Do you think a wife should be:
a. Career-oriented a.
b. Family-oriented b.
6. Do you feel that you have:
a. A fundamentalist religious orientation a.
b. A liberal religious orientation b.

Slide switch setting for questionnaire A :

Set the slide switch corresponding to each question

L E F T if the reply is " a "
R IG H T if the reply is "b "
Push a pushbutton for each " a " reply.

Q U ESTIO N N A IRE B
Indicate whether or not you like: YES NO
1. Outdoor activities a. b.
2. Music and art a. b.
3. Sports a. b.
4. Political activity a. b.
5. The theater a. b.
6. Reading a. --------- b. ---------

Slide switch setting for questionnaire B:

Set the slide switch corresponding to each question

L E F T if the reply is "ye s"
R IG H T if the reply is "n o "
Push a pushbutton for each "ye s" reply.

43
Behavioral Simulations
Some research is being done in the use of computers to simulate various behavioral phe­
nomena normally associated with intelligent beings. T h is work raises interesting questions about
the ability of a computer to exhibit certain kinds of intelligent behavior. For the purposes of
this discussion, we will consider only two specific examples of machine simulation of intelli­
gent behavior.
Associative Memory
The psychological process of conditioning can be simulated by M IN IV A C through the pro­
gram which follows. This program enables the computer to associate a stimulus and a response
once the two elements have been presented together.
M IN IV A C is programmed so that if pushbuttons 1, 2 or 3 are pushed separately lights 1,
2 or 3 respectively will come on. However, if two or three pushbuttons are pushed simultane­
ously, the computer will remember them in association. Once any combination of pushbuttons
has been pushed, the computer will turn on all associated lights whenever any one of the associ­
ated pushbuttons is pushed. M IN IV A C will remember the association until instructed to forget it.
This is analogous to the famous experiments carried out with dogs by the psychologist Pav­
lov. Like Pavlov's dog, the computer "le a rn s" to associate stimulus and response. If two stimuli
are presented simultaneously, the computer "learn s" to associate the two responses originally
associated with each of the two stim uli. Thereafter, the computer will execute both responses
whenever either stimulus is presented. Unlike Pavlov's dog, the computer can be instructed to
remember or forget various stimuli at w ill.
The program for an associative memory circuit is:

1A/1K 1Y/2Y 3A/3K 4H/6G

1B / l— 2A/2K 3B/3— 4L/5L
1C/2C 2B/2— 3C/4C 5C/6C
1F/4H 2C/3C 3F/6H 5F/5K
1F/1X 2F/5H 3F/3X 5G/6H
1G/6F 2F/2X 3G/5F 5L/6L
1H/3K 2G/4F 3L/4L 6 C /6 -
1K/2H 2K/3H 4C/5C 6F/6K
1L/2L 2L/3L 4F/4K 6L/6Z
1L/1Y 2Y/3Y 4G/5H 6 Y /6 +

To use the program:

Turn power on. Push pushbutton 1, 2 and 3 individually. Notice that each pushbutton turns
on the single light associated with it. Now push two pushbuttons simultaneously— for exam ­
ple, pushbuttons 2 and 3. From now on, pushing either of the associated pushbuttons will
turn on both of the associated lights.
Push all three pushbuttons simultaneously. Now pushing any one of the pushbuttons will
turn on all three lights. To instruct the computer to forget the associations, push the "fo r­
get" button— pushbutton 6. The computer is once again ready to "le a rn " associations.

A simulated M aze Solver

The program which follows permits the computer to "le a rn " which of three possible paths
is the correct path to a goal, and to "rem em ber" the correct path until instructed to "fo rg et" it.
This is analogous to experiments in which an animal learns by trial and error the correct path to
a goal through a maze.
The "m a ze " will consist of a start position and three possible "p a th s." The start position
will be the " 1 " setting on the rotary switch dial. The possible "p ath s" will be:

Path A : from 1 to 3 on the rotary switch dial

Path B: from 1 to 5 on the rotary switch dial
Path C : from 1 to 7 on the rotary switch dial

44
The computer will be allowed to "exp lore" the possible paths. Then one path will be indicated as
the correct path. The computer will find this correct path and will then always take that path
until it is instructed to forget it.

The program for a simulated maze solver is:

1A/D8 3E/4E 4H/D7 6H/5Z

1B/1 + 3E/6N 4L/D5 6Y/D1
2E/4Z 3F/4F 4Y/5G 6Z/D19
2 F/2 + 3F/3 + 5B /5+ D2/D4
2G/6L 3H/D3 5E/6E D4/D6
2H/2— 3J/5A 5F/6F D6/D8
2K/4Y 3K/D2 5H/5— D16/M—
2L/2— 3L/3— 5Y/5— D17/M +
3A/4J 4A/4N 6 F/6 + D18/M—
3B/3 + 4 B /4 + 6G/D19 6E/free*

* This programming wire may be connected to D3, D5 or D7 to indicate the goal.

To use the program:

Turn power on. The computer will go directly to the start position. Push pushbutton 6. The
computer is now "exploring" the possible paths. Release pushbutton 6 and indicate a "co r­
rect" path by connecting the wire from 6E to D3, D5 or D7. Push pushbutton 6 again. The
computer selects the "C orrect" path and stops.
Return the computer to its start position by pushing pushbutton 5. If you now push push­
button 6, the computer will go directly to the "co rrect" path without exploring possible
paths.
To set the computer to the "exploring" stage again, remove the connection from 6E to the
path you selected and push the "forg et" button— pushbutton 4.

5. COMPUTER APPLICATIONS IN SCIENCE FICTION FILMS

Science fiction fans seem always to be entranced by equipment which flashes lights and
makes peculiar sounds. To satisfy these fans, the film industry has designed a great variety of
"com puters" for leading roles in science fiction film s. Computers are well-adapted for such roles
since it is possible to produce flashing lights and peculiar noises while operating a real computer.
However, the latest computers are made up of solid-state elements which operate silently;
and output is more likely to be a reel of magnetic tape than a series of blinking lights, blips on
a monitor screen and wierd choking and whirring noises.
Lest your friends— or you yourself— be disappointed at the relative quiet and tranquility
of the preceding programs, the following two programs are presented. Neither program per­
forms a useful function, but both provide the elements of science fiction machines.

The Flashing Lights Circuit

This program provides a maximum number of flashing lights. W ith this circuit programmed
on M IN IV A C , turn the power on. M IN IV A C will perform as a blinking light machine.

The program for flashing lights is:

1A/D1 4A/D4 D1/D12 D8/D14

1B / l— 4B/4— D2/D11 D10/D0
2A/D2 5A/D5 D3/D10 D11/D15
2B/2— 5B/5— D4/D9 D16/M+
3A/D3 6A/D6 D5/D8 D17/M +
3 B /3 - 6 B /6 - D6/D13 D18/M —

45
The "Super" Circuit
This program is presented for those who are not satisfied with the quiet operation of the
previous program. Like the program above, this one does absolutely nothing constructive, a l­
though it appears to be diligently operating to solve the problems of the world. Program this cir­
cuit on M IN IV A C and enjoy the ultimate in science fiction machines.

The program for the "Sup er" C ircuit is:

1A / IE 2H/4N 4E/4J 6H/D16

IB/1 — 2K/D4 4H/5K 6J/6K
1C/2C 2L/3N 5B/5— 6J/D17
1E/2G 3A/3E 5C/5com 6L/M—
1F/1H 3B/3— 5F/5— 6N/D18
1F/2F 3C/4C 5G/6E D0/D2
1G/1 + 3E/4G 5H/4— D2/D6
1J/2H 3F/3H 5 L /5 + D3/D15
1K/D1 3F/4F 5com/Dl 3 D3/D5
1L/3K 3G/3 + 6A/DO D5/D9
1N/D8 3H/4L 6B/6— D6/D7
2A/D14 3J/4H 6C/6G D7/D11
2B /2- 3L/6com 6E/6com D9/D10
2C/3C 4A/4E 6F/6— D11/D12
2E/5A 4B/4— 6G/6N D14/D15
2E/2J 4C/3— 6H /6+

APPENDIX

PROGRAMMING LANGUAGES

In the discussions of programming in this series we have examined instructions which are
communicated to a computer as part of its program. It has been emphasized that a computer
must be given complete instructions for each step which it is to perform in solving a problem.
In Book II, three types of programs commonly used by electronic data processing machines
were discussed: W ired programs, Coded programs, and Stored programs.
W e have not yet commented on the various coding systems developed to permit instructions
to be communicated to a computer in other than the l's and 0's of a binary code. It is important
to remember that once communicated to the computer, all instructions will eventually be reduced
to 1's and 0's which will be interpreted by the computer as on-off switches. The computer cannot
directly "understand" the meaning of the word "a d d ." However, it can interpret the letters
"A D D " as 3 binary coded decimal or "B C D " characters according to the BCD code presented in
Book II. A computer programmed to understand BCD can translate the three letters of the word
"A D D " into a combination of 0's and 1's.

Letters BCD Equivalents

ADD 0 1 0 0 0 1 0 1 0 10 00 10 100

Once the word "A D D " has been converted into binary information using the BCD code, the
BCD equivalent of the word "A D D " may be stored in a storage register of the computer. An in­
terpretive program may then translate the BCD into a binary operating code which the com­
puter interprets as an instruction directing it to utilize certain pre-programmed circuits in a speci­
fied way.
There are as many different interpretive routines as there are different computers. Most
large computers have interpretive programs written for them so that they are able to receive in­
structions in the form of letters such as the "ad d " combination above and translate these letters
into binary bits using BCD. Through the interpretive program, the computer can then translate the
coded binary bits into a single instruction code which the computer interprets as an instruction

46
directing it to use certain logical or arithm etic circuits which have been pre-programmed into the
computer. Once the interpretive program has reduced the letters into a simple instruction code,
the large computer is operating with a basic input signal such that a 0 in the instruction code is
comparable to a pushbutton in the up position on M IN IV A C and a 1 in the instruction code is
comparable to a pushbutton in the down position on M IN IV A C .
The interpretive program converts information communicated to it in letters which the hu­
man operator can understand into simple 0 and 1 signals which the machine can interpret as
basic instructions. These basic instructions determine which pre-programmed circuits will be used,
to effect the action which the human operator desires.
Some examples from the basic language routine used in conjunction with the IBM 7090
computer will be used to demonstrate the functions of an interpretive program. The combina­
tions of letters below are a portion of a program prepared for communication to the 7090 using
the alphabetically coded language which is translated by the interpretive program prepared for
operation on that computer.

CAL A
SUB B
STO C
A PTH
B PTW
C PZE

A "SC A T" CODED 7090 PROGRAM

The letters on each line of the program are punched on a single input card. A s the cards
are read into the computer, each of the letters is translated into a BCD representation which is
stored in the computer. A decoding program then interprets the BCD code and establishes the
meaning for each of the instructions explained below.
CAL A This instruction is interpreted by the computer as: define one register in storage as
register " A " . Clear the processing register (set all of the bits of the processing register to zero)
and add the contents of storage register " A " to the processing register. If we consider the push­
buttons of M IN IV A C to be register " A " , this is equivalent to telling the 7090 to select a circuit
comparable to one which will permit the contents of the pushbuttons to be entered in the proc­
essing relays.
SUB B This instruction is interpreted by the computer as: define a single register of storage
as register " B " . Then subtract the contents of storage register " B " from the contents of the proc­
essing register. If we consider the pushbuttons of M IN IV A C to be storage register " B " , this in­
struction is equivalent to telling the large computer to perform that operation wherein the contents
of the pushbuttons are subtracted from the contents previously stored in the processing relays.
STO C This instruction is interpreted by the 7090 as: define a single storage register as reg­
ister " C " . Then transfer the contents of the processing register to storage register " C " . If the
binary output lights of M IN IV A C are defined as storage register " C " , this is equivalent to di­
recting M IN IV A C to display the contents of the processing relays on the binary output lights.
The last three instructions in the program indicated above are used to direct the 7090 to
store specific numerical values in the registers which it has defined as A , B, and C, respectively.

PTH Indicates that binary 3 is to be stored in the register defined as storage register " A " .
PTW indicates that a binary 2 is to be stored in the register defined as storage register " B " .
PZE d irects the 7090 to make the initial contents of storage register C equivalent to a positive
0.
M any instructions in addition to those indicated in the example above may be used to com­
municate programming ideas to the 7090. In each case the instruction is first converted to the
BCD code so that it may be stored in binary form in the computer. The BCD code is then trans­

47
lated by the interpretive routine into a simple binary instruction which can be performed by the
computer.
The binary instruction signals the large computer just as pushing a pushbutton on M IN I-
V A C signals it to select certain pre-programmed circuitry and perform the operations which the
program is prepared to execute.
Coding languages may be used to communicate logical as well as arithm etic instructions to a
computer. The first sequence of instructions below is interpreted by the interpretive program of
the 7090— and a series of binary instructions are generated. These instructions cause the 7090
to perform an "A N D " operation with the contents of storage register " C " set equal to the con­
tents of storage register " B " AND the contents of storage register " A " . The second program causes
the computer to store the contents of storage register " B " combined according to the rules of
the "O R " function with the contents of storage register " A " in storage register " C " .

Program 1: Program 2:
CAL A CAL A
ANA B ORA B
STO C STO C

TW O SCAT-CODED LOGICAL PROGRAMS

The interpretive programs discussed so fa r permit instructions to be given to the computer

expressed as a combination of letters, rather than as binary words. More complex interpretive
programs have been developed to permit the computer to perform an additional step in trans­
lating information communicated to it by the human programmer. These programs translate
information into binary control words which the computer can interpret as control instructions
directing it to select specific pre-programmed operating circuits. An example of one such "com ­
piler program" is the "F o rtra n " System prepared for use with the IBM 7090. The statement be­
low is a Fortran coded representation of the relationships expressed in the S C A T coded 7090
program discussed at the beginning of this section.

C = A —B

The "Fo rtra n " statement which is punched on a single input card is a mathematical state­
ment of the relationship which the computer is to represent by the steps in its program. The For­
tran compiler first reads the alphabetic and special characters punched on the card and inter­
prets them using BCD into a series of 0's and l's which can be stored in the registers of the
computer. The compiler program then interprets each of the special characters (in the example
above the " —" sign and the " = " sign) as directions to establish a series of instructions equiva­
lent to the S C A T coded 7090 program instructions.
Once a series of instructions equivalent to the S C A T coded instructions are established by
the compiler, an interpretive routine directly equivalent to that used to interpret the S C A T coded
program is employed to establish the equivalent instructions which are then executed by the
computer.
The examples above are simple ones. The capabilities of the various programming languages
developed for use with large scale machines are quite comprehensive. Although these languages
solve some problems by making it unnecessary to communicate to the machine directly in binary
language which the machine can understand, they create others by making it necessary to com­
ply with a number of conventions in a specific manner. These conventions are necessary in order
to insure that the interpretive routine will interpert the letters given to it on the punch cards in
exactly the manner in which the programmer wishes to have them interpreted.
The important point to remember in considering the variety of machine coding systems
which exist is that they are not read or executed by the computer in the form in which they are
communicated to it. Compiling and interpretive routines are used to translate the information
presented to the computer into the basic 0 and 1 codes which the computer is able to use as a basis
for its operation.

48
 The example below provides a summary of the basic steps performed by a compiler and
interpretive routine in translating a single card containing a Fortran statement into instructions
which the 7090 is able to perform.

The FO RTRA N Expression

C = (A + B) * D + F /(E - A)

is interpreted by the compiler program as equivalent to the following series of SC A T instructions.

C LA E C LA A
FSB A FAD B
STO Temp XCA
C LA F FM P D
FDP Temp FAD Temp
STO Temp STO C

Once this series has been generated, each instruction is assigned to a storage register in
which the binary code representing that instruction will be stored. The binary code is not the BCD
coding of the letters of the instruction but a code representing the BCD instruction. For example,
C L A E which is BCD coded by the machine as 010 01 1 100 011 011 001 010 101 is coded by the
interpretive routine as 000 000 111— 101 000 000— 000 000 001 where storage register 001 has
been assigned to Variable " E " and the instruction " C L A E " has been assigned storage register
111. The code 101 000 000 is the operation code derived from " C L A " which indicates to the ma­
chine "select the pre-programmed circuitry designed to place the contents of the indicated data
register in the processing register (Accum ulator)."

The Machine language coding of the entire SC A T program appears below:

Instruction Location Operation Code Data Location

000 000 111 101 000 000 000 000 001

000 001 000 011 000 010 000 000 010
000 001 001 110 000 001 000 010 100
000 001 010 101 000 000 000 000 011
000 001 011 010 100 001 000 010 100
000 001 100 110 000 000 000 010 100
000 001 101 101 000 000 000 000 010
000 001 110 011 000 000 000 000 100
000 001 111 001 011 001 000 000 000
000 010 000 010 110 000 000 000 101
000 010 001 011 000 000 000 010 100
000 010 010 100 000 001 000 000 110

This machine language representation of the program is the only form of the program which
the computer can actually execute. Until this form is derived the computer— in running the com­
piler and interpretive programs used to reduce the FO RTRA N statement given above to the M a­
chine language— is simply translating BCD data and coding this data in accordance with a series
of procedures which it has been previously programmed to follow. Once it has produced the ma­
chine language program above it can be given this new program and directed to execute the in­
dicated instructions.

49
 BOOK VI

MIN IVAC Games

PREFACE

This book contains examples of several types of games which can be played using M IN I-
V A C 601. Once you have become acquainted with these games you will discover that many d if­
ferent games can be developed using the gaming capabilities of the M IN IV A C . Realizing that
hundreds of games have been designed using the 52 cards of a deck of playing cards, you can
imagine the number of games which can be developed using the hundreds of contact points on
the M IN IV A C 601. Each of the games appearing in this book demonstrates a principle of com­
puter gaming which may be expanded to produce additional games.
Most of the games with and against M IN IV A C depend either on the players' skill (as in the
"reaction time tester" or "philosophic tug of w a r") or M IN IV A C 's " s k ill" (as in the "M atch
Gam e" or "T ic-Tac-To e"). Some games depend on chance and make use of M IN IV A C as a source
of random numbers (as in the "Fortune T e lle r" program).

1. M IN IVAC AS AN OPPONENT

The Secret Code

The object of the game is to find the code; that is, the correct order in which the push­
buttons must be pushed to make the "Code Solved" light come on. Using the slide switches, you
can generate 63 different codes; by simple re-programming you can generate 720 different
codes.
Basic to this game as well as to the two games which follow, is a sequence recognition circuit.
Once this basic circuit is wired, different games can be created using the pushbuttons, game
m atrix, rotary switch and slide switches. For the secret code, the slide switches and pushbuttons
are used.
The program for the basic sequence recognition circuit is:
1A/1G 2F/3K 3com/4L 5F/6K 6F/6com
IB/1— 2H/3G 4C/5C 5H/6G 6H /6+
1C/2C 2N/3N 4F/4G 5N/6N
1F/1G 2com/3L 4F/5K 5com/6L
1F/2K 3C/4C 4H/5G 6A/6J
1H/2G 3F/3G 4N/5N 6 B /6 -
lcom/2L 3F/4K 4com/5L 6 C /6 -
2C/3C 3H/4G 5C/6C 6E/6N
2F/2G 3N/4N 5F/5G 6F/6G

Code, add the following connections:

1R/6V 2V/3R 4T/4U 6R/6W
1R/1W 2Y/3Y 4V/5R 6T/6U
1S/1X 3R/3W 4 Y /5 Y 6 Y /6 +
1T/1U 3S/3X 5R/5W 2T/1com*
1T/2S 3T/3U 5S/5X 6W/2com‘
1V/2X 3T/4S 5T/5U 4T/3com*
1Y/2Y 3V/4X 5T/6S 2W/4com*
2R/2W 3Y/4Y 5V/6X 6T/5com*
2T/2U 4R/4W 5Y/6Y 4W/6com'1

51
* These connections may be interchanged for new codes. Be sure though, that there are connec­
tions to each of the six common terminals.

To use the Secret Code program:

W ire the basic circuit onto the M IN IV A C ; then add the connections directly above. Pro­
gram a code by setting the slide switches in various positions. For example, with the above pro­
gram and all six slide switches to the right, the code is solved by pushing the pushbuttons in
sequence from 6 through 1. When the code has been correctly solved, light 1 comes on indicat­
ing "code Solved." If light 6 comes on, an error has been made and M IN IV A C will autom atically
re-set for the next attempt.
Once you have set the code by moving the slide switches, you can challenge your friends
to "b reak the code." They can then try to find the correct sequence of pushing the six pushbut­
tons to break the code. W hen your code is broken, you can change it by moving the slide switches.
If you want to create more new codes, you can interchange the starred connections to the
common terminals. For example, by changing:
2 T /1 com to 2T/2com
and
6W /2com to 6 W / 1com
you will be able to generate 63 different codes.

The Combination Lock

Using the basic sequence recognition circuit of the Secret Code, in combination with the
rotary switch dial, you can design an electronic combination lock which works just like those
used on vaults, padlocks, etc.
Light 1 comes on to indicate that the code has been solved; light 6 comes on when an
error is made, and the machine then autom atically re-sets for the next attempt.

First, program the basic sequence recognition circuit:

IA/1G 2H/3G 4F/4G 5N/6N
IB/1 — 2N/3N 4F/5K 5com/6L
1C/2C 2com/3L 4H/5G 5A/6J
1F/1G 3C/4C 4N/5N 6B/6—
1F/2K 3F/3G 4com/5L 6C/6—
1H/2G 3F/4K 5C/6C 6E/6N
lcom/2L 3H/4G 5F/5G 6F/6G
2C/3C 3N/4N 5F/6K 6F/6com
2F/2G 3com/4L 5H/6G 6H /6+
2F/3K 4C/5C

(This is the same basic circuit as was used in the previous game)
Now add the connection D16/M-j-

Now, design a combination and program it as follows:

Combination Connections
left to 13 D13/6com
right to 2 D2/5com
left to 10 D 1 l/4co m
D10/3com
right to 6 D6/2com
left to 9 D 9 /lco m
D 8/6E*

* Th is last connection is optional; it will cause the machine to indicate an error if 8 is passed at
any point.

52
 When you are designing a combination, you must not include any connection which will
touch off a connection further along. Also, you must make connections to each of the six common
terminals in order.
Once you have designed a combination and programmed it onto M IN IV A C , you can chal­
lenge your friends to attempt to solve the combination. Light 1 will come on when the combina­
tion is successfully solved; light 6 will come on to indicate an error, and the player must start
over again.
You may wish to add dummy connections to the rotary switch dial so that the pertinent con­
nections will not be obvious.

The Electronic Maze

Again using the basic sequence recognition circuit, this time in combination with the game
m atrix, we can design another game. The object of this game is to find the correct path through
the game m atrix. Light 1 indicates success; light 6 indicates error and re-set.
First, program the basic sequence recognition circuit:
1A/1G 2H/3G 4F/4G 5N/6N
1B/1 — 2N/3N 4F/5K 5com/6L
1C/2C 2com/3L 4H/5G 6A/6J
1F/1G 3C/4C 4N/5N 6 B /6 —
1F/2K 3F/3G 4com/5L 6C /6—
1H/2G 3F/4K 5C/6C 6E/6N
lcom/2L 3H/4G 5F/5G 6F/6G
2C/3C 3N/4N 5F/6K 6F/6com
2F/2G 3com/4L 5H/6G 6H /6+
2F/3K 4C/5C
(This is the same basic circuit as was used in the two previous games.)
Now add a programming wire at M-f-, leaving one end free.
This is the "p lay w ire ."
Design a path through the game m atrix. For example: from 8 to 1 to 2 to 9 to 6 to 5

This is programmed as follows:

from 8 M 8t/6com
to 1 M 1t/5com
to 2 M 2t/4com
to 9 M 9t/3com
to 6 M 6t/2com
to 5 M 5 t/ 1com

53
 The player now takes the free end of the "p la y w ire" which is connected to M-f- and touches
it to the top terminals of the game m atrix squares. (The wire need not be plugged into the ter­
m inal; touching the terminal will make the contact.) You may tell the player his S T A R T and
G O AL points; if you wish to make the game more d ifficu lt, however, you need not give him this
information.
Since connections are made only to those squares which are a part of the path, you may
wish to add dummy connections to those not involved in the actual maze. For example, in the
above program, you could add:

M 3t/6 R
M 4t/6S
M 7 t/6 T

This will not affect the maze, but conceals the fact that squares 3, 4 and 7 are not part of
the path.
If you wish to program a path which has less than six steps, connect this path to the lower
common terminals and m anually push the remaining relays to the left before beginning play. For
example, consider the following path:
from 8 to 9 to 4:

M ake the connections:

from 8 M 8t/3com
to 9 M 9t/2com
to 4 M 4 t/lco m

Before beginning the play, m anually push relays 4, 5 and 6 to the left. Whenever an error is
made, relays 4, 5 and 6 must be pushed to the left before play is resumed.

The Match Game

In the original game two players alternate in taking matches from a pile. A t each move a
player can take either one or two matches. The object is to take the last match. This program per­
mits the M IN IV A C to play against a human opponent. The number of matches is represented by
the rotary switch dial.
The game is started with the rotary switch dial at 15. The player takes the first move and
indicates his move (1 or 2 matches taken) by pushing the appropriate pushbutton. M IN IV A C 601
will always win.

54
The program for the M atch Game is:

3A/3E 4 Y /5 Y 6Y/&+
3B/3— 5C/6C D1/D4
3E/3G 5E/D13 D2/D5
3F/3— 5F/5X D3/D6
3G/D0 5F/5G D4/D7
3H/3Z 5H/6H D5/D8
3Y/4Y 5K/6K D6/D9
4C/5C 5L/6L D7/D10
4E/D14 5Y/6Y D8/D11
4F/4X 6 C /6 - D9/D12
4F/4G 6E/D15 D10/D13
4H/4L 6F/6X D11/D14
4H/5H 6F/6G D12/D15
4K/5K 6H /6+ D16/M +
4L/5L 6K/D18 D17/M—

Pushbutton 3 is the re-set button.

Pushbutton 4 is used for "player removes 1 m atch."
Pushbutton 5 is used for "player removes 2 m atches."
Pushbutton 6 is used for "m achine moves."

Light 3 is the "w in " indicator. The last player to move before light 3 comes on wins the game.

To play the Match Game:

Push the re-set button (pushbutton 3). The rotary switch dial will turn to 15, indicating
that there are 15 matches in the pile.
Player makes the first move and indicates:
removing 1 match by pushing pushbutton 4
or
removing 2 matches by pushing pushbutton 5
A fte r the player indicates his move, the rotary switch dial will indicate the number of
matches remaining.
M IN IV A C moves next: player must push pushbutton 6; the rotary switch will turn to the
number of matches remaining after the machine's move.
Play continues as above, with the player and M IN IV A C removing matches in turn.
Light 3 will come on after the last move has been made. The last player to move before light
3 comes on wins the game.
To start another game, push the re-set button (pushbutton 3).

Tic-Tae-Toe
M IN IV A C can be programmed to play tic-tac-toe against a human opponent. A warning
though: with this program, M IN IV A C can not lose. The human opponent may tie the game, but
he can never win. Th is is because of the decision rules which are the basis of the program.
The M IN IV A C is so programmed that the machine will move 5 squares to the right of its
own last move if and only if the human opponent has blocked that last move by moving 4 squares
to the right of the machine's last move. If the human player did not move 4 squares to the right
of the machine's last move, M IN IV A C will move into that square and indicate a win. If the hu­
man player consistently follows the "move 4 to the right" rule, every game will end in a tie.
Th is program requires that M IN IV A C make the first move; the machine's first move will
always be to the center of the game m atrix. A program which would allow the human opponent
to move first would require more storage and processing capacity than is available on M IN IV A C
601. Such a program would, of course, be much more complex than the program which permits
the machine to move first.

55
The program forTic-Tac-Toe is:

1A/1N 2L/D18 4F/5F 5L/M4t

1B /l — 2N/D19 4G/D5 5G/6N
1C/2C 2X/3X 4H/M1 b 5Y/M10
1F/M11 2 Y /3 Y 4H/M5t 6+ /6Y
1G/2A 3C/4C 4J/4K 6— /6C
1H/5X 3F/4F 4K/D6 6G/D1
1J/2F 3G/D7 4L/M2b 6H /M lt
1L/1Z 3G/4N 4L/M6t 6H/M5b
1N/3F 3H/M3b 4G/5N 6J/6K
1X/2H 3H/M7t 4Y/5Y 6K/D2
1X/2X 3J/3K 5C/6C 6L/M2t
1Y/2Y 3K/D8 5F/6F 6L/M6b
1Y/2G 3L/M4b 5G/D3 6X/M10
2B/2— 3L/M8t 5H/M7b 6com/D9
2C/3C 3N/6G 5H/M3t M9b/Ml 1
2E/D16 3X/4X 5J/5K M—/D17
2F/2H 3 Y/4Y 5K/D4 Ml 0/free’
2G/2K 4C/5C 5L/M8b Ml 1/free*

* this is the machine's "p la y " wire

** this is the player's "p la y " wire
To play:
(Note: it will probably be most convenient to keep track of the game on a piece of paper.)
Turn power ON.
Place the free end of the machine's "p la y " wire in 6 com.
Push pushbutton 6 and hold pushbutton 6 down while you push pushbutton 1. The pointer
knob will turn to indicate the machine's first move.
Release both pushbuttons 1 and 6.
A ll succeeding moves are made as follows:
Place the free end of the machine's "p la y " wire in the bottom contact of the m atrix ter­
minal corresponding to the machine's last move (as indicated by the pointer knob).
Player now selects his move and indicates this move by placing the free end of the player's
"p la y " wire in the top contact of the m atrix terminal corresponding to his move.
Push pushbutton 6 and hold pushbutton 6 down while you push the pushbutton correspond­
ing to the move (pushbutton 2 the second time, pushbutton 3 the third time, etc.)
The play continues as above until either light 1 comeson— indicating that M IN IV A C wins;
or light 2 comes on— indicating a tie.
To begin another game, be sure that the player's "p la y " wire and the machine's "p la y "
wire are both free.

2. M INIVAC AS A REFEREE

The Philosophie Tug of War

This game is played with two persons; M IN IV A C acts as the game board and referee. The
object of the game is to make the rotary switch dial stop at the W IN position for the player. If the
dial stops at 9, player 1 wins; if the dial stops at 7, player 2 wins.
Player 1 uses pushbutton 1; player 2 uses pushbutton 6. The rotary switch knob turns:
counterclockwise (towards 9) if pushbuttons 1 and 6 are in the same position (both up or
both down.)
clockwise (towards 7) if pushbuttons 1 and 6 are in different positions.
In other words, player 1 tries to match his pushbutton position to that of player 2's. A t the
same time, player 2 tries to keep his pushbutton position opposite to that of player 1's.

56
 The program for the Philosophic Tug of W a r is:

IX /6 X 4Z/D9 6G/6N 6N/D18

1Y/1 + 6E/6Y 6H /6+ D7/D8
1Z/6Z 6 F/6 - 6J/6K D8/D9
4Y/D18 6F/6L 6J/D17 D16/D19

To play:

Player 1 uses pushbutton 1

Player 2 uses pushbutton 6

To start:

Push pushbutton 4; this is the re-set button. Release pushbutton 4 as the rotary switch knob
starts to turn. A s soon as the knob starts to turn, each player tries to control its direction
using his pushbutton. Player 1 will win if the knob reaches 9; player 2 will win if the knob
reaches 7.

The game continues with player 1 trying to match his pushbutton position to that of player
2, while player 2 is trying to keep his pushbutton position different from that of player 1.
The game ends when the pointer knob stops— either at 9 (player 1 wins) or at 7 (player 2
wins).
To start another game, push the re-set button (pushbutton 4).

The Mind Reading Trick

The "M ind Reading Program" of Book III is repeated here with a variation you may wish to
try. (See Book III for the development of the program.)

The program for the Mind Reading T ric k is:

5V/4S 6V/5V 4W/5S D0/6N

4R/4A 6U/6C 5R/6H D1/6K
4B/4— 5U/5C 5T/6L D2/6J
5C/5A 5F/6F 6R/5H D3/6G
5 B /5 - 6F/6 - 6T/5L D4/5N
6A/6C M—/D18 M + /6 Y D5/5J
6B/6— D18/4V 6X/D17 D6/5K
6X/6V 4U/6S D16/D19 D7/5G

To use the program:

A sk a friend to think of a number between 0 and 7. A sk your friend to answer "ye s" or
"n o " to the following questions about the number:

A. Is the number greater than 3?

B. When the number is divided by 4, is the remainder greater than 1?
C. Is the number odd?

Indicate the answers to the questions as follows:

If the answer to question A is Y ES, move slide switch 4 to the LE FT .
If the answer to question A is NO, move slide switch 4 to the R IG H T.
If the answer to question B is Y E S , move slide switch 5 to the LE FT .
If the answer to question B is NO, move slide switch 5 to the R IG H T.
If the answer to question C is Y ES, move slide switch 6 to the LE FT .
If the answer to question C is NO, move slide switch 6 to the R IG H T.

A fte r the replies to the questions have been indicated, push pushbutton 6. The pointer knob
of the rotary switch will turn to the number your friend had in mind.

57
A variation:
The same program can be used substituting names for numbers as follows:

0= Karen
1= David
2 = Catherine
3= Robert
4= Peggy
5= Andy
6 = Dorothy
7= Hartley

A sk a friend to mentally select one of the above names, and then ask him to answer the following
questions:

A. Does the name end in " Y " ?

B. Does the name have more than 5 letters?
C. Is the name a boy's name?

Indicate the replies to the questions exactly as before. Push pushbutton 6. The pointer knob will
turn to the number corresponding to the name which your friend selected.

The Fortune Teller

W ith this program, M IN IV A C will answer questions— after “ thinking" about them. The an­
swer to the question will appear on the lights as follows:
light 1— Absolutely No
light 2— Doubtful
light 3— Perhaps
light 4— Definitely Yes
Lights 5 and 6 are M IN IV A C 's "contem plation" lights.

The program for the Fortune Teller is:

1A/3G 3A/3J 5C/6C

IB/1 — 3B /3 - 5E/6K
1C/3E 3C/4C 5F/6N
1E/2E 3F/5G 5F/5G
2A/3K 3H/4K 5H/5 +
2B/2— 3L/4N 5K/6F
2C/4E 3N/4A 5N/6E
2E/3C 4B/4— 6A/6E
2F/2X 4C/5C 6B/6—
2G/6L 4F/6G 6C/6—
2H/2L 4L/4X 6F/6G
2K/5L 4Y/4 + 6 H /6 +
2L/2Y 5A/5E
2 Y /2 + 5B /5 -

To use the Fortune T eller program:

A sk M IN IV A C a yes-or-no question.
Push the "contem plate" button (pushbutton 2). Hold the "contem plate" button down while
M IN IV A C considers the question. When you feel that M IN IV A C has had sufficient time to con­
sider the question, release the "contem plate" button and push the "answ er" button (pushbutton
4).
M IN IV A C 's answer will appear in the lights as indicated above. (You may wish to label the
lights by placing a piece of paper with the appropriate indication above the lights).

58
The Random Number Generator
The program for the Fortune Teller can also be used for any game which requires generating
a random number from 1 through 4. Using the preceding program as a random number generator
is just the same as throwing a four-sided die, or spinning a dial with four possible stops.
To generate a random number:
Program M IN IV A C as for the Fortune T e lle r program. Push pushbutton 2 to allow M IN I­
V A C to generate a random number. Release pushbutton 2 and push pushbutton 4 to allow M IN I­
V A C to display the number it has generated. (Light 1 indicates " 1 " , light 2 indicates " 2 " , light
3 indicates " 3 " and light 4 indicates " 4 " .)

Scissors, Paper or Stone

In this modern version of a very old game, M IN IV A C acts as referee and indicates "w in " or
" tie " . Two people play the game, and each player indicates his play (scissors, paper or stone) by
pushing the appropriate pushbutton.

Player 1 Player 2

Scissors pushbutton 1 pushbutton 4

Paper pushbutton 2 pushbutton 5
Stone pushbutton 3 pushbutton 6

The rules of the game are:

Scissors cut paper (scissors will win over paper)

Paper covers stone (paper will win over stone)
Stone breaks scissors (stone will win over scissors)

A fte r both players have indicated their moves, M IN IV A C will indicate that:

Player 1 wins: light 1 comeson

or
Player 2 wins: light 6 comes on
or
T ie : lights 3 and 4 come on

The program for Scissors, Paper or Stone is:

1A/1G 2C/4X 3F/4F 4G/6G
1B / 1— 2F/3F 3G/5G 4Y/5Y
1C/2C 2G/4G 3H/6H 5C/6C
1F/2F 2H/5H 3K/5K 5F/6F
1G/3G 2X/3L 3L/5H 5Y/6Y
1H/4H 2 Y/3Y 3X/5L 6A/6G
1H/5L 3A/5K 3 Y/4Y 6B/6—
1K/3K 3A/4A 4B/4— 6C/6X
1L/1X 3B/4B 4C/5X 6F/6—
1L/3H 3C/AC 4F/5F 6 Y /6 +
1Y/2Y

To Play:

Each player decides upon a move (scissors, paper or stone) and indicates it by pushing his
appropriate pushbutton. Player 1 should move ju st before player 2 moves since the relays will move
after player 2 indicates his move. M IN IV A C w ill indicate a win or a tie on the lights.
You may find it convenient to fix a tall piece of cardboard between pushbuttons 3 and 4 to
keep players from observing their opponent's move in advance.

59
Reaction Time Tester
This game uses M IN IV A C to judge which of two players has the faster reaction time. When
lights 3 and 4 come on, each player tries to push his pushbutton down first. M IN IV A C will indi­
cate a winner, and will also indicate a foul— that is, when a player moves before the signal lights
(3 and 4) come on.
Player 1 uses pushbutton 1
Player 2 uses pushbutton 6
Pushbutton 3 is the re-set button
Light 1 indicates that player 1 wins.
Light 2 indicates that player 1 has fouled.
Light 5 indicates that player 2 has fouled.
Light 6 indicates that player 2 wins.
Lights 3 and 4 give the signal " S T A R T " .
The program for the Reaction Tim e Tester is:
1A/3F 2B/2— 3G/5G 5 B /5 -
I B/) — 2E/3E 3H/4H 5E/6G
1E/2E 2F/2G 3J/6Y 5F/5—
1E/1—- 2F/5A 3K/4K 5L/6X
1F/2A 2G/5N 3L/4L 6B/6—
1F/1G 2H/3H 3 Y /3 + 6E/6G
1G/5J 2K/3K 3Z/6H 6E/D2
1H/2H 2L/3L 4B/4— 6 F/6-
IH /3Z 3A/6K 4F/6A 6 L /6 +
1K/2K 3A/4A 4F/4G 6L/D17
1L/2L 3B/4B 4G/5K D16/D17
1X/5H 3E/4E 4K/D19 D18/M—
1Y/4J 3F/3G 4L/4—
To play:
Each player should be ready at his pushbutton.
Push the re-set button (pushbutton 3).
A s soon as lights 3 and 4 come on, each player tries to be the first to push his pushbutton.
M IN IV A C will indicate the winner on the lights.
If either player pushes his pushbutton before lights 3 and 4 come on, M IN IV A C will indi­
cate who has fouled and the game must be begun again.
To start another game push the re-set button and proceed as above.

60
1

J
M IN IV A C 601 A N D THE M IN IV A C M A N U A L

AKE PRO D UCTS O F :

C O R P O R A T IO N

372 Main Street, W atertown, Massachusetts

Minivac 601
in *

maintenance
manual
SC IEN TIFIC DEVELOPMENT CORP.
3 7 2 M a in S t r e e t , W a t e r t o w n , M a s s a c h u s e t t s
 M INIVAC 601 has been carefully designed to give consis­
tent and reliable operation. Like any large-scale computer,
M INIVAC requires routine maintenance to keep it in top operat­
ing condition. This routine maintenance is easy to perform and
takes little time. Conscientious maintenance w ill insure that
M INIVAC is always ready for operation.

ROUTINE MAINTENANCE

Cleaning: Keep the computer console clean by carefully wiping

it with a damp cloth.

Storage: When the computer is not in use, store the power

cord in the compartment under the power panel. Access to the
storage compartment is through the three-inch hole in the bot­
tom of the computer.
Program w ires should be kept in the transparent bag sup­
plied. Keep the wires free from kinks and knots. When the
computer is put aw ay, it should be kept in its cardboard ship­
ping box.

Relay Contacts: Clean the relay contacts by drawing a piece

of clean, dry paper between the closed contacts. Do not use a
cleaning solution. To clean the normally open contacts, gently
push the plastic crossbar to the left to close, the contacts. Then
draw a piece of clean, dry paper between them. The relay con­
tacts w ill require cleaning once or twice a year.

Oiling: The rotary switch shaft and the motor shaft w ill require
oiling once or twice a year. Remove the bottom panel from the
computer by unscrewing the screws from the rubber feet.- Place
one or two drops of a high-grade, lightweight machine oil (not
a heavy household oil) on the motor bearings at each end of the
motor shaft. Replace the bottom panel.
Remove the pointer knob of the rotary switch and put two or
three drops of oil on the bearing. Work the oil into the bearing
by turning the shaft back and forth a fe w times.
 Replace the pointer knob and realign it by connecting a light
through one of the switch contacts on the rotary switch dial.
Adjust the position of the knob on the shaft so the light comes
on when the knob is pointing at the connected switch position.

Adjustment and Repair

In addition to routine maintenance, M INIVAC may occasion­

ally require minor adjustment and/or repair. A screwdriver is
normally the only tool required for repair. Under some circum­
stances, however, a soldering iron or a drill may be necessary.
All M INIVAC parts are conservatively designed to give last­
ing service. The lights are operated at less than their rated
voltage to extend their operating life. The transformer and the
rectifier are capable of accepting repeated overloads.
The power supply is rated at 1.5 amps continuous with a
nominal 12 volts output. All switches are rated at 3 amps; relay
contacts are rated in excess of 3 amps. To provide low contact
resistance, the rotary switch uses silver-plated circuits. The rotary
switch contact fingers are made of berryllium copper; the drive
shaft is made of stainless steel running in an oil-impregnated
bronze bearing. The drive motor is operated at less than one-
third rated voltage.
Should you find that your M INIVAC is not operating prop­
erly, check first to see that routine maintenance has been carried
out. The following sections describe trouble-shooting and repair
techniques for the various parts of the M INIVAC.
Power Supply:
If the power cord is plugged into a live 110-volt AC outlet
and the power light does not come on when the power
switch is turned on, there may be a fault in the power sup­
ply. Proceed with the following steps until the fault is
found, then follow the repair instructions:

1. Check to see that a short-circuit is not connected across

the power supply terminals (+ and —): remove all pro-

2
 gram w ire connections to the power supply terminals.

2. Push the circuit breaker button to re-set the circuit breaker

if it has "blow n."
3. Check to see that power is actually being supplied to
the + and — terminals by connecting an output light to
the terminals + and — with programming wires. The
unloaded power supply outlet voltage should be ap­
proximately 13 volts.

4. Check the 110 volt outlet with a lamp or other appliance

to see that power is being supplied to the outlet.

5. Unplug the M INIVAC power cord. Remove the bottom

panel from the computer by unscrewing the rubber
feet. Push the power panel out of the computer console
through the holes in the dividing section.

6. Check all power connections to be sure they are secure.

The power supply circuit diagram is:

Power S u p p ly Circuit Diagram

7. With the power cord unplugged, connect a binary output
light between the transformer center top (yellow wire)
and the center terminal on the rectifier. Program wires
with alligator clips attached are convenient for this pur­
pose.
Turn the power switch on and plug in the power cord.
Do NOT touch any power supply components while
power is connected. If the binary output light comes on,
the trouble is in the circuit breaker, the power light or the
power switch. If the binary output light does not come
on, the trouble is in the transformer, the rectifier or the
power switch.
8. Check the power light by connecting a binary output
light to the power light terminals. If the binary output
light comes on when power is on, the power light is
burned out. Remove the burned-out power light and
replace it with a new light assembly, connecting the leads
to the power supply terminals.
9. Check the circuit breaker:
Connect a binary output light between the + power
supply terminal and the circuit breaker terminal which
leads to the transformer. If the output light does not
come on when power is on, the trouble is the transformer
or the rectifier. If the output light comes on, proceed
with the next step. Connect a binary output light be­
tween the + power supply terminal and the circuit break­
er terminal which leads to the power light. If the output
light comes on when power is on, the trouble is not in
the circuit breaker. If the output light does not come
on, even when the circuit breaker is pushed, the circuit
breaker is defective.
To replace the circuit breaker: remove the leads from
the terminals; bend up the tabs which hold the circuit
breaker in place and remove the defective unit. Insert
a new circuit breaker and replace the leads.
10. Check the power switch as follows:
Using program wires with alligator clips, connect the DC
output contacts (located beneath the power panel) to an
output light. Turn power on; if the output light does
not come on, the power switch is defective.
To replace the power switch, remove the leads from the
terminals and drill out the rivets which hold the switch.
Insert a new power switch, fastening it with machine
screws and nuts. Replace the leads.
Check the AC input contacts with an ordinary household
lamp. Connect the lamp to the black (primary) trans­
former leads. It w ill be necessary to pierce the insulation
of one of the leads. After the A C input contacts have
been checked, the insulation must be repaired.
Turn the power switch on and plug the power cord in.
Do not touch any of the power supply components while
the power is connected. If the lamp comes on, the AC
imput contacts are in correct working order.

11. Check the transformer with an output light. Disconnect

the two yellow /black w ires from the rectifier.
Connect an output light between the yellow lead and
one yellow /black lead. Turn the power switch on and
plug the power cord in. The light w ill come on if the
transformer is operating correctly. Continue the trans­
former check by connecting an output light between
the yellow lead and the second yellow/black lead.
If the transformer is defective, it must be removed and
replaced with a new unit.

12. If the transformer checks out but there is not proper

DC output between the transformer center top (yellow
lead) and the center terminal on the rectifier, the rec­
tifier is defective and must be replaced.

5
Program Wires:
The program wires are made to last for the life of the com­
puter if they are properly used. When removing program
wires from the computer console, do not pull them out by the
w ire. Grasp the taper pin and remove it from the terminal.
Be sure to keep the wires free from kinks and knots.
If program wires become lost or damaged, new sets may
be purchased from Scientific Development Corporation.
(See spare parts list).
Lights:
Check the binary lights and the relay indicator lights by
connecting them to the 4- and — power supply terminals.
The lights are conservatively operated and should not re­
quire replacement for at least a year. To replace a light,
cut the leads % inch from the terminal and remove the
defective light from the panel. Insert a new light assembly
and solder the new leads in place.
Pushbuttons:
Check a pushbutton by connecting it in series with two
lights as indicated:

Circuit Diagram — Pushbutton Check

6
 Turn power on. Light 1 should come on. Push the push­
button. Light 1 should go off and light 2 should come on.
To replace a defective pushbutton switch:
Pull the red plastic button off the pushbutton; unscrew
the exposed mounting screw. Cut the pushbutton switch
leads (located underneath the console) and remove the
unit. Insert a new pushbutton switch, screwing it into
place. Solder the new leads in place. Replace the red
plastic button on the switch shaft.
Relays;
Check a relay by connecting the coil with power through a
pushbutton as indicated:

Circuit Diagram — Relay Check

Turn power on. Lights 1 and 3 should come on. Push the
pushbutton. Lights 1 and 3 should go off and lights 2 and 4
should come on.
If a relay snaps on and off but does not make proper con­

7
 nection through the relay contacts, the contacts may need
cleaning (see routine maintenance). If the switch arms are
not meeting the contacts with proper pressure, gently bend
the switch arm to increase the contact pressure. The switch
arm should engage the normally open and normally closed
contacts and should push them slightly as they meet.
To replace a defective relay, disconnect the eight connec­
tions to the relay and remove the unit by unscrewing the
relay mounting screws. Insert a new relay and replace the
eight connectors. To assure proper connection, compare
the connections on the new relay with another relay's.
Slide Switches:
Check a slide switch by connecting binary output lights to
the terminals as indicated:

3A 4B 4-
OX 3B 4A,
•+

z
•+

Circuit Diagram — S lid e Switch Check

Set the slide switch RIGHT. Turn power on. Lights 1 and 3
should come on. Move the slide switch to the LEFT. Lights
1 and 3 should go off and lights 2 and 4 should come on.

8
 To replace a defective slide switch:
Cut all six connections to the slide switch. Drill out the
rivets which hold the unit in place.
Install a new switch with machine screws and nuts.
Solder all connections in place, using other slide switch
units as guides.
Rotary Switch:
The circuit diagram for the rotary switch motor drive is:

Circuit Diagram — Rotary Switch M otor Drive

Check the motor drive by connecting power to the RUN

terminals (D17 and D18). Turn power on. The rotary
switch pointer knob should turn. Reverse the power con­
nections: the pointer knob should turn in the opposite di­
rection. Connect a short circuit across the STOP terminals
(D18 and D19): the pointer knob should stop turning.
If the pointer knob does not rotate, connect power momen­
tarily across the STOP terminals: the knob should turn very
rapidly. If the knob does turn, the trouble may be a bad
connection or an open resistor. If the knob does not turn,
the trouble may be a bad connection or a defective motor.
It is also possible that the motor shaft is not properly en-

9
gaging the drive wheel. Check the rotary switch by con­
necting D16 (the switch arm) to + and connecting output
lights through the rotary switch contacts, DO through D15
Check the knob alignment to be sure that the knob has not
turned on the drive shaft. If proper contacts are not being
made, check the assembly under the panel: the rotary
switch contacts may require cleaning.
The rotary switch mechanism is accessible when the bottom
panel of the computed is removed. If the motor does not
turn the rotary switch:

1. Check all electrical connections.

2. Check the drive wheel to be sure that it is free to

rotate.

3. Check each of the two resistors by connecting power

across them. The resistors should heat up in about
20 seconds. If a resistor does not heat up, it is de­
fective. Replace a defective resistor by cutting its
leads and soldering a new unit of the same resistance
in place.

4. Check the motor by unsoldering the motor leads and

connecting them directly to the power terminals. If
the motor does not rotate it is defective and must be
replaced with a new unit.

5. To assure proper friction drive, adjust the motor bracket

and spring the motor shaft against the edge of the
drive wheel.

10
 MINIVAC 601 - SPARE PARTS

The following spare parts are available from Scientific De­

velopment Corporation for repairing M INIVAC 601:

P a rt N o . D e scrip tio n Price

6431 Circuit Breaker T.00

6432 Slide Switch .50

6433 Rectifier 3.00

6 434 Transformer io.oo

6435 Light Assembly 1.00

6436 Pushbutton Switch .75

6437 Relay 6.00

6471 Motor 4.00

6472 Drive W heel Assembly 2.00

6473 Complete Rotary Switch Assembly 20.00

6490 Set of 100 Programming W ires 12.00

6491 Set of 2 test w ires with clips 1.00

€ 1- 3- ^
31 °
6 "V7 S

To avoid C.O.D. charges please send check or money order

with your spare parts order directly to:

Scientific Development Corporation

372 Main Street
Watertown 72, Massachusetts

There is a handling charge of 25 cents for orders less than

$5.00.

11