BOOKS...

I I .I I I .I V

SCIENTIFIC
DEVELOPMENT
CORPORATION
Watertown
Mass.

W hat is a Digital Computer?

How Computers M ake Logical Decisions
How Computers do Arithm etic

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.
2.

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

1
5
5
7
12
15
16
16
21
24
25
27
27
30
29
30
32

BOOK III: HOW COMPUTERS MAKE LOGICAL DECISIONS

Preface

35

1.

35
35
38
39
41
42
43
43
44
44
44
44
47
48
49
50
50
54
57
60
63

2.

3.

4.

BASIC OPERATIONS
The O peration " A N D "
The O peration "O R "
The O peration " N O T "
The O peration "EITH ER BUT N O T B O T H "
RELAYS FOR MORE FLEXIBLE T H IN K IN G
The Relay A N D c irc u it
T he Relay OR c irc u it
The Relay N O T c irc u it
The Relay EITHER BUT N O T BOTH c irc u it
T H IN K IN G A N D D E C IS IO N -M A K IN G
Boolean A lge bra
D ecision-M aking w ith In s u ffic ie n t In fo rm a tio n
S im ulation
Thoughts A b o u t T h in k in g
SOME COMPUTER PROBLEMS
A M in d Reading Program
Q u a n tity Recognition
A Problem Involving Three G irls
The Farmer, the Goose, the Corn and the W o lf
The Television Problem
i

BOOK IV: HOW COMPUTERS DO ARITHMETIC

1.

THE B IN A R Y NUM BER SYSTEM

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

67
67
68
71
73
73
74
74
75
76
77
78
78
78
79
79
80
81
81
82
82
83
84
84
85
86
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
The
The
The

in p u t u n it
processing u n it
storage u n it
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.

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 inputo 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.

I N P U T -O U T P U T
AND
SECONDARY STORAGE

STORAGE
AND
PROCESSING

IN P U T -O U T P U T
AND
SECONDARY STORAGE

OUTPUT

IN P U T
AND
OUTPUT

IN P U T

IBM 7090 D A T A PROCESSING SYSTEM

O U T PU T
SECONDARY
STORAGE

STORAGE
AND
PROCESSING

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

M IN IV A C 601

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 .

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-

1 X /2 X
2 X /3 X
3 X /4 X
4 X /5 X
5 X /6 Y

6 Z /6 +
6 Y /5 H
5 H /4 H
4 H /3 H
3 H /2 H

3-

2 H /1 H
1Y /1G
1G /1 C
1F / 1
2 Y /2 G

4-

2 G /2C
2 F /2
3 Y /3 G
3 G /3C
3F /3

5-

4 Y /4 G
4 G /4C
4 F /4
5 Y /5 G
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).

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 TO 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-

1 H /2 H
2 H /3 H
3 H /4 H
4 H /5 H
5H /6 Y

1.

3-

6 Z /6 +
6 Y /5 X
5 Y /D 1 6
D 1/1G
1 G /1 C

IF /1
D 2 /2 G
2 G /2C
2F /2
D 3/3G

4-

3G /3C
3 F /3 D 4/4G
4 G /4C
4 F /4

5-

D 5/5G
5G /5C
5 F /5

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.

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

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.

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
1S/2S
2 R /D 2
2 S /3 S
3 R /D 3

3 S /4 S
4 R /D 4
4S/5S
5 R /D 5
5 S /6 S

6 R /D 6
6S /D 19
D 18/D 16
D 1 6 /M
D 1 7 /6 X

6 Y /6 +

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
BY THE 2 6

PRINTING
CARD PUNCH

0123456769

ABCDEFGHIJKLMNOPQRSTUVUXYZ

IBM DATA PROCESSING

II

II I

III

II

III

tilpS o d

I I

O'fl ooooooooooooooo| ooooooq| | oooooo|

n111ii11111111111111niii 11111 n i i

I
11
222

000
M
tt
11
2 22 2

3 3 3 3 | jj

33333

33

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

98666

6666

66666S6666666666666666|666666666666

777777

7 77 77

77177717J7J777177777|J117J777|77777

8868886

888888

lllllllllllllllllllllllllllllllllll

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
4243 44454* 47 l SI S H H H R D IR R III1 D H B N 17HU 7t 71 7373 74

11

6666666
66666666
77777777
777777777
8888888 311 8888888881
9 9 9 9 9 9 9 9 9 91
>I I 1 I I I I I IR

UMiiauunanaa

PRINTED
- BY THE 26
PRINTING
CARD PUNCH

aaaaaaaaanaaMaaaaaanaaMaanaaaTinniin

222222222222| 222222222222||22222222

aanaaana'aaaaaaa

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)
(B)

1010
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 | ^ + + + + + + +

j_ _ _ I

J ______

I
I
j r | - | | 1 i i
1
0 0 0|0 0 o'o 0 0 0 1 0| 0 0 DOOOOOOO

g g f e j+ f l

I I I
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
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 43l4 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 [ 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
| l 1 1 1 i'll
I
I
I 1 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 1 I
I
I
I
I
1 .
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
1
,
1
I
I
| 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
I
1
I
1
I

5 515 5 515 5 5,5

J | 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
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
T
I
I
I
I
I
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
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 |||
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

66686666

'

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
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------------------------------s 4 9 4 | 7 I I p * II 1 2 1131)4 15 |1 l i t 11

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.

0%
H^5IScLQ-Q.Q-ujt/>LLJ</H^ Jnmi
Mi l l
i i
I
X-

()-

CHECK*

8*

4 *

2*

A B C D E F G H I JK LM N O PQ R S TU V W X YZO

EI_-

2 3456789

/& i

TTTTII M T
I I I

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.

I" State

"0" State

IBM PUNCHED CARD

.......

/L
0

MAGNETIC CORE

RELAY OR SWITCH

TUBE OR TRANSISTOR

ELECTRICAL PULSES

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
Drum

Reading from a
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
blank
1
2
3
4
5
6
7
8
9
0
+0

-0

BCD
110
000
000
000
000
000
000
000
001
001
000
011
101

Code
000
001
010
011
100
101
110
111
000
001
000
010
010

Char.
A
B
C
D
E
F
G
H
1
J
K
L
M

BCD
010
010
010
010
010
010
010
011
011
100
100
100
100

Code
001
010
011
100
101
110
111
000
001
001
010
011
100

Char.
N
O
P
Q
R
S
T
U
V
w
X
Y
z

BCD
100
100
100
101
101
110
110
110
110
110
110
111
111

Code
101
110
111
000
001
010
011
100
101
110
111
000
001

Char.
+

/
=
-

)
$
*
(

BCD
010
100
110
001
001
011
011
101
101
111
111

Code
000
000
001
011
100
011
100
011
100
011
100

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

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

lllllllll

1 2 3 4567390

= $.-< *> +

MINIVAC 601

II

I 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

________________ IM8081

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

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 382 Paper Tape Reader

IBM 729 IV Magnetic Tape Unit

IBM 7500 Card Reader

IBM 1403 Prihter

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
STORAGE U N IT

DRUM
STORAGE U N IT

5.

DISC
STORAGE U N IT

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.
2.
3.
4.

32

Set a scale to read zero

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

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.
2.
3.
4.

Set an am m eter to zero

Send 2 amps o f cu rren t through the am m eter
Send 3 more amps o f c u rrent through the am m eter
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 onepound 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.
B.
C.

If I go outdoors
and
If it is raining
then
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.

B.

C.

If it is tru e th a t
I go outdoors
AND
If it is tru e th a t
it is raining
TH EN
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

PUSHBUTTON 2

LIG H T 3

Up or O ff

Statem ent A
is false

Statem ent B
is false

Statem ent C
is false

Down or On

Statem ent A
is true

Statem ent B
is true

Statem ent C
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
true) respectively, the conditions expressing the
above can be expressed as:

d e fin itio n s o f 1 and 0

as true and false (not
relationships lin k in g the three statem ents

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.
B.
C.
A.
B.
C.
A.
B.
C.
A.
B.
C.

If 1 go out
AND
If it is raining
1w ill get wet.
If 1 do not go out
AND
If it is raining
THEN
1w ill not get wet

(1 X 1 = 1)

(O X 1 = 0)

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

( 0 X 0 = 0)

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.
B.
C.

If I stand in the rain

OR
If I stand in the shower
TH EN
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.
B.
C.
D.

If I go outside
AND
If it is raining
OR
If I stand under the shower
TH EN
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
IX /2 Y
2 Z /3 A
3 B /3

1Z

1+

1Y

2Z
1X

2Y

3B

3-

2X

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.
B.
C.
D.
E.

If it is raining
AND
I am outside
AND
If I am not under cover
OR
If I am in the shower
TH EN
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.
B.
C.

If I tu rn on the hot w ater

OR
If I tu rn on the cold water
TH EN
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"

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 =
ing through switch B).

1 means "sw itch B is CLOSED" (th a t is, current is flo w

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

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:

0
1
0
1

0
0
1
1

0
0
0
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

0
0
1
1

0
1
0
1

0
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:

1)

46

AXA = 0

2)

A + A = 1

M athem atical laws:

the com m utative law:
1)
2)

A + B= B+ A
A X B = BX A

the associative law

1)
2)

A + (B + C) = (A + B) + C
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)
2)

if the com puter has s u ffic ie n t capacity fo r a ll conditions of the problem.

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
1 am not under cover

1am wet

1am inside
OR
1am under cover

1 am not wet

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.
B.
C.
D.

I can stay inside

I can go outdoors
I can wear a raincoat
I can carry an um brella

W e m ust now fin d the results o f each possible solution:

Possible Solutions
A.
B.

Stay inside
Go outdoors

C.
D.

W ea r a raincoat
C arry an um brella

Results
I w ill not get wet
I w ill not get wet if it is not raining or if I am un
der cover.
I w ill not get wet
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
A.
B.
C.
D.

Stay inside.
Go outdoors.
W ea r a raincoat.
C arry an um brella.

Results
I w ill not get wet.
I may not get wet.
I w ill not get wet.
I w ill not get wet.

Other Factors
have an appointm ent I m ust keep,
cannot a ffo rd a new suit,
do not have a raincoat,
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.
B.

50

Is the num ber greater th an 3?

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.)

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 =
A =
B=
B=
C=
C=

the num ber is greater than 3

the num ber is not greater than 3
the rem ainder is greater than 1
the rem ainder is not greater than 1
the num ber is odd
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 BXC~= 6
A X B X C = 5
A X B X 'C 'r r 4

A X B X C = 3
A X B X C = 2
A X B X C = 1
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
A X J3X C T= 2
AXBXC =1
T x^
o

a x

becomes
becomes
becomes
becomes

A X CX B= 3
A X ~C X J3 = 2
AX C XB = 1
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
B
C

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

(slide switch 5): W hen the num ber is divided by 4, is the rem ainder greater than 1?
(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
4 R /4 A
4 B /4 5 C /5 A
5 B /5 6 C /6 A
6 B /6 6 X /6 V
6 V /5 V
6 U /6 C

5 U /5 C
5 F /6 F
6 F /6 M /D 1 8
D 1 8 /4 V
4 U /6 S
4 W /5 S
5 R /6 H
5 T /6 L
6 R /5 H

6 T /5 L
M + /6 Y
6 X /D 1 7
D 1 6 /D 1 9
D 0 /6 N
D 1 /6 K
D 2 /6 J
D 3 /6 G
D 4 /5 N
D 5 /5 J

D6/5K
D7/5G

To use the program :

1.
2.

A sk a frie n d to th in k o f any num ber between 0 and 7.

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
L ig h t
L ig h t
L ig h t
L ig h t
L ig h t

1 ON
2 ON
3 ON
4 ON
5 ON
6 ON

indicates th a t 1 switch is ON
indicates th a t 2 switches are ON
indicates th a t 3 switches are ON
indicates th a t 4 switches are ON
indicates th a t 5 switches are 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
B
C
D
E

=
=
=
=
=

the
the
the
the
the

condition
condition
condition
condition
condition

th a t
th a t
th a t
th a t
th a t

slide
slide
slide
slide
slide

switch
switch
switch
switch
switch

2
3
4
5
6

is ON
is ON
is ON
is ON
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
A X B ^ X C " X D X F = light
A X fX C ^ X D X E _ = light
A X F X X D X F = light
A X B X C X D X E : light

1 ON
1 ON
1 ON
1 ON
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

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
2 B /2
3 B /3
4 B /4
5 B /5
6 B /6
2 F /3 F
3 F /4 F
4 F /5 F
5 F /6 F

6 F /6 6 + /6 S
6 S /4 V
4 V /3 V
3 V /2 V
2 U /2 C
2 C /3 C
3 U /4 C
4 C /5 C
4 U /6 C

5 A /2 G
4 A /2 K
3 A /2 N
2 A /2 T
1 A /3 J
6 A /3 N
6 T /5 V
6R /5S
5 U /5 T
5 U /4 S

5 R /6 H
5 W /6 L
4 T /6 K
4R /6 J
6 N /5 L
6 J /5 H
6 G /4 H
6 K /4 L
5 J /4 K
5 N /3 L

5 K /4 N
5 G /4J
4 N /3 H
4 J /2 L
4 G /2 H
4 K /2 S
3 J /3 K
3 G /2 T
2 J /2 K
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 65F 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 65F. 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 65F and would a u to m a tica lly
tu rn up the furnace u n til the tem perature reached 68F.
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 =
s_=
s =
w =

C am ille could not

Sara should
Sara should not
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.
B.
C.

The goose and the w o lf cannot be le ft alone w ith o u t the farm er.
The goose and the corn cannot be le ft alone w ith o u t the farm er.
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~
G
G
W
W
C
C
E

represent the absence o f the farm er

represent the presence o f the goose
represent the absence o f the goose
represent the presence o f the w olf
represent the absence o f the w olf
represent the presence o f the corn
represent the absence o f the corn
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

Goose

W o lf

Corn

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

1A /1 J
1B / l
1C /1G
1E/1 U
1F /2F
1 H /2 H
1 K /2 K
1K /1 +
1 L /2 L
1 N /3 K

1 U /2 V
1 V /1 X
1 X /6 X
1 Y /1 +
2 A /2 J
2 B /2 2 E /2 U
2 C /2 G
2 F /3 F
2 H /3 H

2 N /3 N
2 W /3 V
3 A /3 J
3 B /3 3C /3 G
3E /3U
3 F /4 F
3 H /4 H
3 K /4 K
3 L /5 A

3 N /4 N
3 W /4 V
4 A /4 J
4 B /4 4 C /4 G
4 E /4 U
4 F /4 4 H /4 +
4 L /6 A

5 B /5 6 B /6
6 Y /6

To use the program :

1.

2.
3.
4.
5.

62

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.
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)
Choose another object and place it in th e boat by pushing the appropriate slide switch
to the LEFT.
M ove the boat across the river by pushing pushbutton 6. If lights 5 and 6 rem ain OFF,
no error has been made.
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)

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.
2.
3.
4.
5.

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

Either Edward or Francis, or both, are w atching television.
Either C lara or David, but not both, is w atching television.
Edward and David are e ither both w atching or both not w atching television.
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:
A nd:

B = the condition th a t Barry is N O T w atching television, etc.

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 ---------------------

64

2.

E+ F= T

3.

CXD + CXD = T

4.

EXD + EXD = T

5.

F + ( BXE) = T

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
Slide
Slide
Slide
Slide

switch
switch
switch
switch
switch

1 represent Barry
2 represent C lara
3 represent David
4 represent Edward
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
1S/2S
1 T /2 R
2 R /2 V
2 U /3 W

2 W /3 U
3 R /4 R
3 S /3 V
3 T /4 T
4 A /4 S

4 B /4 4 U /5 V
4 U /1 V
4 V /4 +
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
0
1
2
3
4
5
6
7
8
9

B inary
0
1
10
11
100
101
110
111
1000
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
5 H /6 H
5 G /5 F

6G /6 F
5 C /6 C
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
6 X /5 L

5K/6F
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
5 H /5 F
5 F/6F
5 C /6C
6 C /6
5 J /6H
6G /5E

6J/6E
6E /6A
6 B /6
5 H /5 A
5B /5
6 + /6 Y
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.
2.
3.

T u rn power ON. N otice th a t both relays 5 and 6 are OFF.

Push and release pushbutton 6. N otice th a t both relays 5 and 6 are now ON.
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
5 L /6 L

6 K /4 A
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 5

Light 4

Ligh

FLIP-FLOP

carry

FLIP-FLOP

carry

FLIP- F .OP

B to A

C to B

pushbutton

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).
Pushbutton
carry

Times
Pushed

0
1

UP
DOWN

NO
YES

1
1

UP
DOW N

1
1

NO
NO

UP
DOWN

NO
YES

1
1

NO
YES

1
1

UP
DOWN

1
1

NO
NO

0
0

NO
NO

UP
DOW N

1
1

NO
NO

1
1

NO
YES

1
1

UP
DOWN

1
1

NO
NO

1
1

NO
NO

UP
DOWN

1
1

NO
YES

1
1

NO
YES

1
1

UP
DOW N

NO

NO

UP

0 or 8

Carry
C to B

Flip-Flop
A

Carry
B to A

Flip-Flop
B

0
0

NO
NO

0
0

NO
NO

0
0

NO
NO

0
1

0
0

NO
NO

0
1

Flip-Flop
C

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
IB /1
1C /2C
1 E/2 G
1F/2F
1F/1H
1G /1 +
1J/2H
1J/4K
2 A /4 E
2B /2

2 C /3 C
2E/2J
2 F /4 A
3A /6E
3B /3
3 C /4 C
3E /4G
3 F /3 H
3 F /4F
3G /3 +
3 J /4 H

3 J /6K
3L/4 L
3N /4 H
4 B /4
4 C /5 C
4 E /4 J
4F /5 A
5 B /5
5 C /6 C
5E /6G
5F/6 F

5F /5 H
5 G /5 +
5 J /6H
5 J /6 X
5 L /6 L
5 N /6 H
6 A /6 F
6 B /6
6 C /6
6 E /6 J
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
3 N /4 N
6 R /6 K
4 K /6 U

5 N /6 N
5 K /6 T
6S/3J
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.
d.

M ove slide switch 6 to the RIGHT.

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

+_]_

+0

-M

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

00

74

Sum

0
1

Carry Out

1
0

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:
6 R /6 Z
6 T /6 X
6 T /6 V
6 Y /6 +

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

The three fu ll adders which

perim ent 5. However, the c irc u it
lays. The high order stage o f the
T his allows us to display a 4 -b it

3L /4 X
4 A /4 V
4 B /4
4E /5F
4F/5S
4 G /5 U
4 H /4 N
4 K /5 Z
4 L /5 A
4 N /5 X
4 Y /5 Y
5 B /5
5C /6K
5F/6E
5G /6S
5 H /5 +

5K /6 W
5 L /6 A
5 N /6 U
5 R /6 T
5T /6 R
5 T /5 V
5 Y /6 V
6B /6
6E /6
6 G /6R
6 H /6 X
6J/6K
6 L /6 Z
6 N /6 T
6 V /6 Y
6 Y /6 +

m ake up th is c irc u it are very much like the single adder o f ex

has been m odified s lig h tly so th a t three adders use only five re
adder is wired so th a t the carry from it is displayed by lig h t 3.
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
-4

1101
-0 1 0 0

decim al

bin ary

1101
1011

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_
1

1 0

1
0

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
3 L /4 X
3 H /4 Z
4 N /5 X

4 K /5 Z
6 H /6 X
6 L /6 Z

Now A D D these connections:

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

4 K /5 X
6 C /6 F
6 H /6 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
1X 0=0

0 X 1 = 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
1B / 1
1C /3U
1 F /2 F
1G /2 W
1 H /2 A
1J/2U
1K /1T
1L/2R
1 N /1 R
1 N /2 T
1U / M 1 0
1V /1 T
2 B /2
2 C /3G
2F/3 F
2G /4S

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

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

5L /6 A
5 N /6 U
5R /6 T
5 T /6 R
5 T /5 V
5 Y /6 V
6 B /6 6 C /M 1 0
6E /6
6G /6R
6 H /6 X
6J /6K
6 L /6 Z
6 N /6 T
6 V /6 Y
6 Y /6 +

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
1B / l
1R /3K
1S/2G
1 T /3 J
1 U /2 N
1W /2K
1 X /3 H
1 Y /2H
1 Y /2 Y
1 Z/3 L
2 A /4 L
2 B /2
2C /3G

2F/3F
2 L /2 +
2R /3 T
2S/4F
2T /3 R
2 T /2 V
2 U /4 G
2 X /4 K
2 Y /6 S
2 Z /4 N
3A /5 L
3 B /3
3 C /4G
3F /4E

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

4 H /4 N
4S /4 +
5 A /5 R
5 B /5 5 C /6 K
5F/6E
5 H /5 +
5 S /5 +
6 A /6 R
6B /6
6E /6
6 F /M 1 0
6 J /6K
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
01 0

answer
101

11011 H- 101 =
27

101 plus a rem ainder o f 10.

or
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
1 F /2 F
1G /3com
1 H /1 L
1 K /6 K
1 L /2 L
2 E /D 9
2 F /3 F
2 G /5 G
2 H /2 L
2 H /3 H
2 K /3 c o m
3 A /3 c o m

3 B /3
3 E /D 7
3 F /4 F
3 G /6 E
3 H /3 L
3 K /4 K
3 L /4 L
4 A /4 c o m
4 B /4
4 E /D 6
4 F /5 F
4 G /5 co m

4 H /4 L
4 H /5 H
4 K /4 c o m
5 A /5 c o m
5 B /5
5 E /D 5
5 F /6 F
5 G /6 G
5 H /5 L
5 K /4 c o m
5 L /6 L
6 A /6 c o m

6 B /6
6 E /D 3
6 F /6
6G /6co m
6 H /6 L
6 H /6 +
6 K /5 c o m
D 1/ 6com
D 2 /5 c o m
D 4 /4 c o m
D 8 /3 c o m
D 1 6 /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
1 F /2 F
1 G /6 H
1 H /5 U
1 J /5 H
1 K /6 L
1 L /5 W
1 N /5 L
2 C /5 R
2 F /3 F
2 G /3 H
2 H /4 U
2 J /4 H

2 K /3 L
2 L /4 W
2 N /4 L
3 C /6 R
3 C /4 C
3 F /4 F
3 G /D 1 5
3 J /D 1 4
3 K /D 1 1
3 N /D 1 0
3 U /4 V
3 V /3
3 W /5 V

4 C /5 C
4 F /5 F
4 G /D 1 3
4 J /D 1 2
4 K /D 9
4 N /D 8
4 S /4 +
4 S /5 S
5 C /6 C
5 F /6 F
5 G /D 3
5 J /D 2

5 K /D 1
5 N /D 0
5 S /6 S
6 F /6
6 G /D 7
6 J /D 6
6 K /D 5
6 N /D 4
6 X /D 1 7
6 Y /6 +
D 1 6 /D 1 9
D 1 8 /M

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
1B / l
1C /2C
1E/2N
1 F /1 G
1F/2K
1 G /1X
1 H /2 H
1L/D1
1 Y /2 Y
2 A /2 X
2 B /2
2C /3C
2 E /3 N
2 F/2G
2F /3 K
2 G /2 X
2 H /3 H
2 L /D 2
2 Y /3 Y

86

3 A /3 X
3B /3
3 C /4 C
3 E /4N
3F/3G
3 F /4 K
3 G /3 X
3 H /4 H
3L /D 3
3 Y /4 Y
4 A /4 X
4 B /4
4C /5C
4E /5N
4F/4G
4 F /5 K
4 G /4 X
4 H /5 H
4L /D 4
4 Y /5 Y

5 A /5 X
5 B /5
5 C /6C
5 E /6 N
5F/5G
5F/6K
5 G /5 X
5 H /6 H
5 L /D 5
5 Y /6 Y
6 A /6 X
6 B /6
6 C /6
6E /D 7
6F/6G
6 G /6 X
6 H /6 +
6L /D 6
6Y /D 1 6
D 1 6 /M +

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
1 F /2 F
1 G /3H
1H /6G
1 J/2H
1K /3L
1 L/6J
1N /2L
2 C /4 K
2C /3C
2 F/3 F
2 G /D 5
2 J /D 1
2 K /D 4
2 N /D O
3 F/4 F
3 G /D 7

3J/D 3
3 K /D 6
3 N /D 2
4 A /5 R
4 A /4 K
4B /4
4 C /5 V
4 F/5 F
4G /5J
4 H /5 A
4J/5G
4L/4J
5 B /5 5 C /6R
5F/6F
5 H /5 +
5 S /5 W

5 U /5 Z
5 W /5 X
5 Y /6 V
6 A /6 C
6B /6
6C /6 V
6 F /6
6 H /5
6S /6 W
6 U /6 Z
6 W /6 X
6 Y /6 +
D 16/D 19
D 1 7 /M +
D 1 8 /M

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