Digital PDFs

DIGITAL-8-25-F

December 1965

70 pages

Original

2.7MB

Document:	digital-8-25-f-d
Order Number:	DIGITAL-8-25-F
Revision:
Pages:	70
Original Filename:	https://svn.so-much-stuff.com/svn/trunk/pdp8/src/dec/digital-8-25-f/digital-8-25-f-d.pdf

OCR Text

8-5-5

DIGITAL EQUIPMENT CORPORATION. MAYNARD, MASSACHUSETTS

PDP-8 FLOATING-POINT SYSTEM
PROGRAMMING MANUAL

September 1965

DIGITAL EQUIPMENT CORPORATION

•

MAYNARD, MASSACHUSETTS

Reprinted January 1967
Reprinted June 1967
Reprinted November 1967

PREFACE

The PDP-8 comes to the user complete with an extensive selection of system programs
and routines making the full data processing capability of the new computer immediately
available to each user, eliminating many commonly experienced initial programming
delays.
The programs described in these abstracts come from two sources, past programming
effort on the PDP-5 computer, and present and continuing programming effort on the
PDP-8. Thus the PDP-8 programming system takes advantage of the many man-years
of program development and field testing by PDP-5 users.
Although in many cases PDP-8 programs originated as PDP-5 programs, all utility and
functional program documentation is issued in a new, recursive format introduced with
the PDP-8.
Programs written by users of either the PDP-5 or the PDP-8 and submitted to the users'
library (DECUS - Digital Equipment Corporation Users' Society) are immediately available to PDP-8 users.
Consequently, users of either computer can take immediate advantage of the continuing
program developments for the other.

iii

CONTENTS
Chapter

Page
INTRODUCTION.......................................................

1-1

FLOATING-POINT REPRESENTATION ................................... .

2-1

Arithmetic ..................................................... " . .

2-4

Basic Floating-Point Commands .......................................

2-4

Interpreter .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2-4

Floating-Point Instructions ...........................................

2-4

FLOATING-POINT INPUT/OUTPUT ..•...................................

3-1

Floating-Point Input ................................................

3-1

Flags .......•.•.•..••...................................... " . .

3-1

Rubout .............•......................•••••...••.•.•.•..•.

3-2

Floating-Point Output. .. .•• .. .•.••••••.••••••... .•. .•... .•. ..•. .• ...

3-3

Entry. • . . . . . • • . • . . . • . . • . • • • • . . . . . . . . . • • • . . . . . . • . . . . . . . . . . . . . . . .

3-3

Program Example .•.......•.......•..•.•.•.••........•..........

3-3

The Basic Package ....•.•..•.......•......•......•......•....•......

3-5

Subroutines •......•.........•.........•...•...•.......•........

3-5

Description of Basic Function.. . . . . . •. .• ... . ... .••... .. . ... •.. .•. .. .. .

3-6

Addition .•..••••...•..••.•....••••......•.•.•......•..........

3-6

Subtraction ..•...••••..•.•.•••••...••.•.•.....••......•........

3-6

Multiplication ...••.••...•.•..•..•.•..........................•

3-6

Division

..........•...•........•..............................

3-6

Square ...............•..............•...•..........•..•.......

3-7

Square Root .. . . . . • . . . • • . . . . . . . . . • . • . . . . . . . . . . • . . . . . . . • . . . . . . . . .

3-7

Error Flag

.....••.•.••.......•.......•.........................

3-7

Summary of Basic Package ......•...............•....................

3-8

Entry Points ......•.•.•.•..............•..•................. " . .

3-8

Flags .....................•.••.........•...........•..•.......

3-8

Commands .........•...........................................

3-8

Storage ......•........•...........•.....•.....................

3-9

Fixed to Floating/Floating to Fixed Point ..............................

3-9

Extended Floating-Point Package .....................................

3-10

Sine .......•....•..•... ...••....... ..•...... .... ..............

3-11

CONTENTS (continued)
Chapter
3 (cont)

Page

........................................................

3-11

Arc Tangent ...................................................

3-11

Logarithm .....................................................

3-12

Floati ng Exponent ..............................................

3-13

..............................................

3-14

Floating Point Package Versions ......................................

3-16

Cosine

Output Controller

PROGRAM LISTINGS ....................•...........•............•.....

4-1

Floating Point Arithmetic Interpreter ....•..•.••..••......•.•...•...••.

4-1

Floating Point I/O Routines... ............. ..........................

4-14

Double Precision Decimal-Binary •.•••••••••.•••••••...•••••.••••.••.•

4-15

Input and Conversion.......... ..................................

4-15

F looting Output "E" Format......................................

4-17

Floating Point Input ......•.••..•...•.....•........•..••..•...•..

4-19

Floating Point Package.. . . .. . . . . . . . ... ........ . .... . .. .. .. . ... .... ..

4-23

Floating Point Exponential.... .••.. .•..•.••...••..•..••••••••••••

4-23

Negation Subroutine ............................................

4-24

Floating Point Arc Tangent.................................... .••

4-25

Floating Logarithm.... ... . ......................................

4-26

Floating Point Sine .............................................

4-27

Floating Point Cosine ...........................................

4-28

Floating Output Program...... ..... ................ ..................

4-31

CHAPTER 1
INTRODUCTION

The PDP-8 Floating-Point System enables the programmer to concentrate on the logic of his computation
rather than on decimal points. The system maintains a constant number of significant digits throughout
the computation, thereby enhancing the accuracy of the result.
Floating-point notation is particularly useful for computations involving numerous multiplications and
divisions where magnitudes are I ikely to vary widely and where only crude predictions can be made as to
the amount of variation involved. The main advantage of the PDP-8 Floating-Point System is derived
from the ability to store very large or very small numbers by storing only the significant digits together
with the exponent for that number. In an integTal or fractional fixed-point mach ine, the programmer
must either include all of the O's between the significant digits and the decimal point (and in so doing
loose considerable accuracy), or account for the point in each step of his programming.
The PDP-8 Floating-Point System is constructed as a self-contained package which includes its own input,
arithmetic, and output routines.

It allows the programmer to use floating-point arithmetic without having

to construct his own arithmetic subroutines. For example, the polynomial:
Y = AX 3 + ax2 + CX + D
may be programmed as follows (the variables X and Y, and the parameters A,
registers of the same name):

JMS
FGET
FMPY
FADD
FMPY
FADD
FMPY
FADD
FPUT
FEXT

I 7
A
X

a
X
C
X
D
Y

/ENTER FLOATING PACKAGE
/LOAD PSEUDO AC
/MULTIPLY
/ADD
/MULTIPLY
/ADD
/MULTIPLY
/ADD
/STORE
/EXIT FLOATING PACKAGE

1.-1

a, C, and D are stored in

CHAPTER 2
FLOATING-POINT REPRESENTATION

A floating-point number consists of a mantissa and an exponent. For example, the decimal number 12 may
be represented as:
1200.0'10- 2
120.00'10- 1
12.000'100
1.2000.10 1
.12000'10 2
etc.
where the exponents are the numbers - 2, -1, 0, etc.
Since the PDP-8 is a binary machine, floating-point numbers are stored as floating-point binary internally.
For example, the binary number 110 (6 decimal) may be represented as:

11000.0.2- 2

(24·1/4 = 6)

1100.00.2- 1

(12·1/2 = 6)

110.000.20

(6·1 = 6)

11.0000.2 1

(3.2=6)

1.10000.l

(3/2·4 = 6)

.110000.2 3

(3/4'8 = 6)

etc.
Notice that the binary exponent is always a signed integer. The PDP-8 Floating-Point System uses the
convention:
1/2 ~

I MANTISSA I < 1.

When this is true, the word is said to be normalized.
The value of the number is then:
MANTlSSA·2EXPONENT
where the MANTISSA is a signed quantity. The result of this convention is that more significant bits are
retained. For example, the number. 10 (decimal) is equal to:
2-1

10.00 011 001

100

I 110 011 001 100 I 110 011 001 100 110 ..••

in binary.
When this word is stored in two 12-bit words (24 bits), the leading O's are nonsignificant and only 20 bits
of significance are maintained.
2- 3 ., 0.11 001

If the number is rewritten as:

100 110

I 011 001 100 110 1011 001 100 110 ....

and the mantissa stored in two 12-bit words, 23 bits of significance are maintained.
The exponent is stored in a third word making a total of three words for storage.
sign of exponent

binary exponent 2's complement
signed quantity

sign of mantissa

-H

high order mantissa word

low order mantissa word

The number .1 (decimal) would be stored as:
III

III

101 1011001100110 1011001100110

or written in octal:
7775 3146 3146
The address of this number would be considered 2146 if it were stored as follows:
2146/7775
2147/3146
2150/3146
The number -.1 would be:
7775
4631

(double precbion 2 's complement of mantissa)

4632
Since the mantissa is greater (in magnitude) than 1/2 and less than 1, its binary point is considered to be
between bit 0 and bit 1 of the high order mantissa word.

2-2

Further examples:
12 (decimal)
1100. = (24) (0.11000000000000000000000)
would be stored as (written in octal):
0004
3000
0000
Octal

Decimal

Floating Binary

-1

0001

6000

0000

-10

-8

0004

6000

0000

2000

0000

0002

3000

0000

0004

3000

0000

.25

7777

2000

0000

Convert the decimal number 1.6 to PDP-8 floating-binary form for storage.
An easy method to convert a decimal fraction to an octal fraction is illustrated below:

.6
8
4):8

1.610 = 1 .46314631 8

x 8
6):'4

= 1.100110011001100110011001
= (21) (0.11001100110011001100110)

3D"

= 0001

x 8
1}""7;

3146

8
4~

3146

6).4
8

3;-:2
8

1)":'6

2-3

ARITHMETIC
Since floating-point numbers are stored in a three-register format, the floating-point system uses a "psuedo"
floating accumulator (FAC) which consists of three registers in the floating-point package: 44, 45, and 46.
Register 44 contains the exponent; 45 and 46 contain the high and low order parts of the mantissa, respectively.

BASIC FLOATING-POINT COMMANDS
The basic floating-point commands include the following:
load floating accumulator
store floating accumulator
add to floating accumulator
subtract from floating accumulator
multiply by floating accumulator
divide into floating accumulator
normalize floating accumulator
All arithmetic operations are called through an interpreter. The command codes have a format that is
almost identical to the format of the PDP-S memory reference instruction, namely:

where the op code is from 000 to 111 or Os - 7 S .

INTERPRETER
The interpreter contains, at all times, the address of the memory location containing the next pseudo instruction to be executed. This is initially stored when the program enters the interpreter using a JMS
instruction. When the interpreter encounters an instruction with an op code of 0 and with bits S-11 of
the pseudo instruction equal to 0, it exits to the next memory location.

FLOATI NG-POI NT I NSTRUCTI ONS
The floating-point instructions are:
Op Code

Mnemonic

Effect

FADD

Floating Addition
Add the contents of the effective address
to the floating accumulator.

2-4

FSUB

Floating Subtract
Subtract the contents of the effect address
from the floating accumu lator.

FMPY

Floating Multiply
Multiply the floating accumulator by the
contents of the effective address.

FDIV

Floating Divide
Divide the floating accumulator by the
contents of the effecitve address.

FGET

Floating Get
Load the floating accumulator with the
contents of the effective address.

FPUT

Floating Put
Store the contents of the floating accumulator
at the locations specified by the effective
address. The contents of the floating accumu Iator are unchanged.

FNOR

Floating Normalize
Normalize the contents of the floating
accumulator.

is decoded as follow~:

Bits 8-11 = 0000

Floating Exit
Return control to following instruction

= 0001

Floating Square
Square the contents of the floating accumulator

= 0010

Floating Square Root
Take the root of the absolute value of the
floating accumu lator

=0011-1111

Expandable commands

This may be summarized:
FADD

Yi 1000i C(F AC) + C(Y) ~ C(FAC)

FSUB

Yi 2000i C(FAC) - C(Y) ~ C(FAC)

FMPY

Yi 3000i C(FAC) x C(Y) -? C(FAC)

FDIV

Yi 4000i C(F AC) -=: C(Y) -? C(FAC)

FGET

Yi 5000i C(Y) -? C(FAC)

2-5

Result is
C(Y)
unchanged

FPUT

Y; 6000; C(FAC) ~ C(Y)

FNOR

; 7000; C(FAC) normalized

f,EXT

; 0000; exit from interpreter to instruction following this command

~ C(FAC)

SQUARE; 0001; C(FAC)2 ~ C(FAC)
SQROOT; 0002; C(FAC) 1/2 ~

C(FAC)

The assembler recognizes all of these mnemonics except for SQUARE and SQROOT which may be defined
as:
S QUARE=OOOl
SQROOT=O002
The floating-point interpreter is entered with an effective JMS 5600.
For example:
SQROOT=OO02
*7
5600
*200
JMS 17
FGET A
SQROOT
FPUT I B
FEXT
HLT
0003
A,
2000
0000
300
B,

200/
201/
202/
203/
204/
205/
206/
207/
210/
211/

4407
5206
0002
6611
0000
7402
0003
2000
0000
0300

$
When this program is started at 0200, it will halt at location 205. The state of the machine will be:

44/
45/
46/

0002
2000
0000

floating accumulator contains
2.0, i.e., (4.0)1/2

206/
207/
210/

0003
2000
0000

300/
301/
302/

0002
2000
0000

answer stored here

2-6

CHAPTER 3
FLOATING-POINT INPUT/OUTPUT

FLOATING-POINT INPUT
The basic floating-point package contains an input routine to read characters from the 33 ASR keyboard.
Numbers are read in in decimal and converted to the internal floating-point binary format.

Input format

is floating decimal. The number 726.7 may be read in in any of the following forms:
726.7
. 7267E3
. 7267Et<>3
+7267E-1
etc.
Input is terminated when a character is typed that is not a part of this format. The conversion of "12.0."
would be terminated on the second "." and the binary number:
000000000100

(octal 0004)

011000000000

(octal 3000)

000000000000

(octal 0000)

would be in the floating-point accumulator upon completion of the conversion.

There are several flags associated with the input routine that are useful to the programmer.
0056

This register is a switch that has the following meaning:
If C(56) = 0, do not type a line-feed after a carriage-i"eturn was read.
If C(56) = 7777, type a line-feed when a carriage-return is inputted.
This switch is initially set to 7777.

0057

This register contains the character that terminated the input conversion.

0060

This register contains 0000 if no input conversion was made; i.e., a
space or other terminator was initially typed.

3-1

The input routine is entered with on effective JMS 7400.

It returns control to the instruction following

the colling JMS upon receipt of a terminator. The floating accumulator contains the input number in
normalized floating-binary.

dicates whether or not there was a vol id input.
For example;
*5
7400
*7
5600
*200
JMSI5
JMSI7
FPUT A
FEXT
JMSI5
JMSI7
FPUT B
FEXT
HLT

/INPUT ROUTINE
/CALL FLOATING POI NT

A, 0

o
o
$
When this program is started at 0200 and the following is typed;
X2.0Y
The program will holt at location 0210, and A and B will contain

A, 0

o
o

and C(57) = 0331, the second TERMI NA TOR

0002
2000
0000

The first input was considered a 0 because a terminator "X" was inputted.

3-2

This program could be written to ignore the non-numeric information as follows:

7400
*7
5600

*200
/CALL INPUT ROUTINE
JMSI5
/ ANY VAll D INPUT?
TAD 60
SNA CLA
JMP .-3
/NO - IGNORE
JMSI7
/YES
/STORE IT
FPUT A
FEXT
/GET NEXT
JMSI5
TAD 60
SNA CLA
/VAlID?
JMP .-3
/NO - IGNORE
JMSI7
/YES
FPUT B
/STORE IT
FEXT
HLT
/HALT

o
o

$
Register 57 may be used for integrating control characters into the input.

Rubout
There is one special input character, the rubout. If it is struck before an input delimiter, the input routine is restarted and previous numbers deleted.
For example, assume the input routine has been entered.
+-

::l

276 ~1T
c:::

The input routine would exit with 1 (decimal) in the floating accumulator or:

44/

0001
2000
0000

3-3

FLOATING-POINT OUTPUT
The basic floating-point package contains an output routine. Upon entry the contents of the floating accumulator are converted to floating-point decimal and typed out on the ASR 33 in the following format:
±O. XXXXXXXE±XX
For example, if the floating accumulator contained:

44/

0002
2000
0000

and the output routine were entered,
t{).2000000Et{)1
would be typed.

The output routine is entered by an effective JMS 7200. After outputting the contents of the floating accumulator, it returns control to the instruction following the calling JMS instruction. The contents of the
floating accumulator are lost. The floating-output routine has a flag or program switch in location 0055
on page O.
If C(0055) does not equal 0, type a carriage-return/line-feed following each output. This location initially contains 7777.

Program Example
/SOLVE THE QUADRATIC EQUATION

/ AX 2+ BX+C=O FOR VALUES
/OF A, B, C.

TYPE OUT ROOTS

SQROOT=O002
S QUARE=OOOl
*5

7400

/POI NTER TO INPUT ROUTI NE
/POI NTER TO OUTPUT ROUTI NE
/POI NTER TO INTERPRETER

7200
5600
*200
KCC
TLS
BEGIN,

JMSI5
JMSI7

/INPUT A
/ENTER INTERPRETER

3-4

TWO,

MIN1,

TEMP,

FMPY TWO
FPUT A
FEXT
JMSI5
JMSI7
FPUT B
FEXT
JMSI5
JMSI7
Ftv\PY TWO
FMPY A
FPUT TEMP
FGET B
SQUARE
FSUB TEMP
SQROOT
FPUT TEMP
FGET B
FMPY MINl
FPUT B
FSUB TEMP
FDIV A
FEXT
JMSI6
JMSI7
FGET B
FADD TEMP
FDIV A
FEXT
JMSI6
JMP BEGIN
0002
2000
0000
0001
6000
0000
0
0
0
0

/2.0
/STORE 2.0·A
/INPUT B
/ENTER INTERPRETER
/STORE B
/INPUT C
/ENTER INTERPRETER
/MULTIPLY BY 2.
/MULTIPLY BY 2A
/4AC
/B SQUARED
/B SQUARED -4AC
/TAKE SQUARE ROOT
/STORE IT
/-l·B=-B
/STORE
/-B-SQUARE ROOT
/ANSWER IN FAC
/OUTPUT IT
/ENTER INTERPRETER
/-B
/-B + SQUARE ROOT
/ANSWER IN FAC
/OUTPUT IT
/CONTINUE
/CONSTANT 2.0

/CONSTANT -1.0

0
0
B,

0
0
C,

0
0
0

$
Both the input-conversion routine and the output-conversion routine use a typeout subroutine that contain~
the Instructions:

3-5

TSF
JMP .-1
TLS
CLA
This means that the teleprinter flag must be set when these routines are entered or it must be in the process
of typing. A TLS instruction can be put into the initial ization section of a program.

THE BASIC PACKAGE

Subroutines
When the floating-point interpreter encounters a 0 op code, it further decodes bits 8-11. If these bits
equal 0, the interpreter exits. If the bih are nonzero, they are used in a table look-up to specify the
address of a subroutine. The called subroutine may use the floating-point interpreter, the input, or the
output, but it may not u~e bits 8-11 to call another subroutine. The look-up table is on page 16 of the
interpreter listing. For example, if the user wished to define a routine that negated the contents of the
floating accumulator, the following ~teps would be taken:
On page 6 of the interpreter listing, there is a routine to negate the floating accumulator. It is a subroutine and has, as entry point, the address 6000. If the negate subroutine were to be called by 0010 in
the interpreter mode, the word 6000 wou Id be placed in address 6554 of the call ing table.

Input and out-

put could be called in the interpreter mode if 7400 were placed in 6555 (0011) and 7200 were placed in
6556 (0012).
NEGATE = 0010
INPUT = 0011
OUTPUT = 0012
SQUARE = 0001

*7
5600
*200
KCC

TLS
JMSI7
INPUT
SQUARE
NEGATE
OUTPUT
FEXT
HLT

/ENTER INTERPRETER
/CALL INPUT ROUTINE
/SQUARE IT
/CALL OUTPUT ROUT! NE
/CALL OUTPUT ROUTI NE
/EXIT
/HALT

3-6

DESCRIPTION OF BASIC FUNCTION

Addition
Floating-point addition is carried out by first aligning the binary points of the two numbers. This is accomplished by scaling the smaller number to the right. Then the mantissas are both scaled right once 50
that overflow wi II not occur into the sign bit. A 2 1s complement addition of the mantissas is then made.
The result is normalized and control returns to the interpreter. This may be represented as:

Can operands be aligned?

Yes

Align Exponents
Add Mantissas
Normalize
Return to Interpreter

No
Put larger number in FAC

<l1li<l1li1-------....

Subtraction
Floating-point subtraction is accomplished by negating the operand, and then calling .the addition subprogram.
Negate Operand
Call Addition

Multiplication
Floating-point multiplication is accomplished by adding the exponents together and then performing a
double-precision multiplication. The result h normalized and control returns to the interpreter.
Add Exponents
Multiply Mantissas carrying result to 35 bits
Normalize
Return to Interpreter

Division
Floating-point division is accomplished by subtracting the exponent of the divisor from the exponent of
the dividend. The mantissa is divided and the result is normalized. Control returns to the interpreter.

3-7

Is divisor = O?

•

Yes

Subtract Exponents
Divide Mantissas
Normalize
Return to interpreter •

+ 1 to error fl ag
Set resrlt to highest possible

The ~quare routine calls the mu Itipl ication routine internally.

Square Root
The square root is calculated using Newton's method in which an initial approximation is made and then
each succeeding approximation is calculated. The routine exits when two successive approximations are
equal to within the least significant bit of the mantissas.
Form first approximation, Xl' of .,t'N';"
then:

X.+1 = 1/2 (X. + -X )
I

I .
I

IXi - Xi +1 I < least significant bit?

~-~o !
Yes

Return to interpreter

Error Flag
Division by 0 causes C(61) to be incremented by 1. The quotient is set to the highest positive number.
Attempting to extract the square of a negative number causes C(61) to be incremented by 1. The root of
the absolute value is taken.
Thecontentsof61 are set to 0 at the beginning of each divide and square root operation.

3-8

SUMMARY OF BASIC PACKAGE

Entry Points
5600

Arithmetic Interpreter

7400

Floating Input

7200

Floating Output

NOTE: Both the input and the output routines require that register 7 contains
the interpreter entry point: 5600.

If 10, type carriage-return/line-feed after output.

If 10, follow each input carriage-return by a line-feed.

Contains the input terminating character.

Equals 0 if no val id input.

Is nonzero if (a) divide by 0 or (b) square root of a negative number.

Commands
FADD

1000

Floating Add

FSUB

2000

Floating Subtract

FMPY

3000

Floating Multiply

FDIV

4000

Floating Divide

FGET

5000

Floating Get

FPUT

6000

Floating Put

FNOR

7000

Floating Normalize

FEXT

0000

Floating Exit

SQUARE

0001

Square

SQROOT

0002

Square Root

0003
Expandable
0017

3-9

Storage
The floating accumulator is in registers:
44 - Exponent
45 - High Order Mantissa
46 - Low Order Mantissa
The basic floating-point package (Digital 8-5A-S) uses memory locations:
7; 40-61; 5600-7577

FIXED TO FLOATING/FLOATING TO FIXED
Since the floating-point package stores numbers in the following format:

Signed Exponent

"
,J

2 1s Complement Signed Mantisa

llmplied Binary Point

and ~ince the exponent indicates where the real exponent is, this information may be used to either convert a floating-point number to fixed point or to convert a fixed-point number to floating point.
For example, assume that there is an integer in the accumulator that is less in magnitude than 2047. To
float this number, the following sequence of deps may be employed:

DCA 45
DCA 46
TAD C13
DCA 44
JMSI7
FNOR
FEXT

/PUT INTO HIGH-ORDER MANTISSA
/PUT 0 INTO LOW-ORDER MANTISSA
/11 (10) INTO
/EXPONENT
/CALL INTERPRETER
/NORMALIZE
/LEAVE INTERPRETER

C13,

0013

/11 (DECIMAL)

At point C, we have set the binary point of the integer to the right end of the high order mantissa word,
or eleven (decimal) locations to the right of the implicit binary point.
To float this number, the floating accumulator is scaled right until the exponent contains the location of
the desired binary point. To fix the floating accumulator as an ll-bit signed integer, the following sequence of coding may be employed:

3-10

DONE,

CLA
TAD 44
SZA SMA
JMP .+3
CLA
JMP DONE+l
TAD M13
SNA
JMP DONE
SMA
JMP ERROR
DCA 44
CLL
TAD 45
SPA
CML
RAR
DCA 45
ISZ 44
JMP 00
TAD 45

M13,

-13

GO,

IFETCH EXPONENT
liS THE NUMBER <1?
INO:
IYES: FIX IT TO 0

INO: SET BINARY POINT AT
III (10) PLACES TO RIGHT OF CURRENT POINT
liT IS ALREADY THERE: ALL DONE
ITEST TO SEE IF IT IS TOO LARGE
IYES: NUMBER >2**1)
INO: SET SCALE COUNT
10 TO C(L}
IFETCH MANTISSA
lis IT <O?
IYES: PUT A I IN LEFT BIT
ISCALE RIGHT
IRESTORE IT
ITEST IF SHIFTED ENOUGH
INO: CONTI NUE
I ANSWER IN C(AC)

1-11 (DECIMAL)

This may be coded as a subroutine.

EXTENDED FLOATING-POINT PACKAGE
The extended floating-point package contains the following additional commands:
0003

Floating Sine
Take the sine of the angle in the floating accumulator (considered to be in
radians) and leave the result in the floating accumulator.

0004

Floating Cosine
Take the cosine of the angle in the floating accumulator (considered to be
in radians) and leave the result in the floating accumulator.

0005

Floating Arc Tangent
Take the arc tangent of the number in the floating accumulator and leave
the resu It (in radians) in the floating accumu lator.

0006

Floating Exponent
Raise e to the C(FAC) power. Leave the result in the floating accumulator .

.0007

Floating Logarithm
Take the natural logarithm of the absolute value of the number in the floating
accumu lator.
3-11

Sine
The sine routine uses the following identities:
SIN(-X)=-SIN(X)
SI N(X)=51 N(2 IT N+F)=51 N(F)
where O<F<2lTand N is an integer:
51 N(lT -F)=-51 N(F)
Using these identities, F is reduced to the range -IT/2<F<1V'2.
If Y = 2F/lT so that -1 <F<l , then:
51 N(X)=(C 1F+Cl 3+C5 F5 +Cl7+C9F9)
where:
C 1 = 1.57079631847
C3 = -0.64596371106
C5 = +0.07968967928
C7 = -0.00467376557
C9 = +0.00015148419
The sine subroutine is valid over the range:
-2lT.2047<X<2lT.2047

Cosine
The cosine routine uses the identity:
C05(X)=51 N(lT /2-X)

I t then uses the 5 I NE routi ne .

Arc Tangent
The arc tangent routine uses the following identities:
ARCTAN (-X)=-ARCT AN(X)
IF X>, 1; then ARCTAN(X)=lT/2-ARCTAN(l/X)
FOR 0<X<1,

3-12

where:

Aa = +0.6402481953
A1 = +0.4229908144
A2 = +0.0264694361
BO = +0.6402487022
B1 = +0.6363779373
B2 = +0. 1108328778

The result is in the range:
-1T /2<ARCTAN (X)< 1T/2

-oo<X<oo
Logarithm
The logarithm routine uses the identities:
For X<l;
LOG(X)=LOG(2 N . F) (l<F<2)
=NLOG(2)+LOG(F)

8
LOG(l+Y)=

L
C=l

where:
A1 = +0.9999964239
A2 = -0.4998741238
A3 = +0.3317990258
A4 = +0.2407338084
A5 = +0.1676540711
A6 = -0.0953293897
3-13

~ = +0.0360884937

Aa = -0.0064553442
for:
0<X<1, the identity:
LOG(X)=-LOG(l/X) is used.

Floating Exponent
The exponent routine uses the identity for X)(),

where N is an integer and <F<l

l =1 + _ _2_F_ __
C_
A-F+BF2 _ _
D+F

where:
A = +9.95459578
B = +0.03465735903
C =+617.97226053
D =+87.417497202
Log 2e=1.4426950409
If X<O

x
1
e =--x
e
Example, input A, B, and output:
Y = LOG(COS(A/B)+I'A'=B)
SQROOT=O002
COS=0004
LOG=0006

/DEFINITIONS TO ASSEMBLER

*5
7400
7200
5600
/USES EXTENDED INTERPRETER
3-14

*200
KCC

TLS
BEGI N,

A,
B,
TEMP,

JMS I 5
JMSI7
FPUT A
FEXT
JMSI5
JMSI7
FPUT B
FMPY A
SQROOT
FPUT TEMP
FGET A
FDIV B
COS
FADD TEMP
LOG
FEXT
JMSI6
JMP BEGIN

/INPUT A
/ENTER INTERPRETER
/STORE A
/EXIT
/INPUT B
/ENTER INTERPRETER
/STORE B
/A'B
/EXTRACT ROOT
/STORE IT
/LOAD FAC WITH A
/DIVIDE BY B
/TAKE COSINE
/ADD
/TAKE LOG.
/OUTPUT ANSWER

a
a
a
a
a
a
a
a
a

Output Controller
There is an additional routine available that formats floating-point output. It is available with both the
basic floating-point package and the extended package.
The controller routine requires two parameters for output formatting. It is called by an effective JMS7200.
At this point:
C(62) = total number of digits to be outputted. If C(62)=O, output in E format.
C(AC) = number of digits to the right of the decimal point. If it equals 0, do
not type a ".".
If the number in the floating accumulator is larger than the field width allows, "XIS" will be typed. The
sign is typed and leading OIS are suppressed.
Example: If the contents of the flodting accumulator are 678.234 (decimal).

3-15

C(62)

C(AC)

Output

3
4
4
5
6
3
2
0

0
0
1
2
2
2
0
1

+ 678
+ 678
+ 678.2
+ 678.23
+ 678.23
+ 678.
+XX
+ 0.678234E+OJ

FLOATING POINT PACKAGE VERSIONS
Four versions of the floating-point package are available:
Digital 8-5A-S
This is the basic floating-point package. It consists of the input/output package and the basic arithmetic
instructions. Its core limits are:
7; 40-61; 5600-7577

Digital 8-58-S
This is the bClsic floating-point package with the output controller. Its limits are:
7; 40-62; 5400-7577
The output controller does not type a carriage-return/line-feed after output and, hence, the programmer
must provide his own routine to do so. Because of the way in which the typeout routine in the floatingpoint packages are constructed, the user must construct similar typeout routines in order to avoid timing
errors in the teleprinter. In other words, the floating-point package uses a routine simi lar to:
TSF

JMP .-1
TLS
to do its typing. The user must use a similar routine, i.e., one that waits for the teleprinter flag to be
set before executing the TLS instruction.

Digital 8-5C-S
This is the basic floating-point package with the extended functions. In addition, two interpretive commands are provided for input and output. The additional interpret ive commands are:

3-16

SINE
COSINE
ARCTANGENT
EXPONENTIAL
LOG
INPUT
OUTPUT

3
4
5
6
7
13
14

Its core limits are:

7; 40-61; 4757-7577
Using this package, the routine on page 3-14 and 3-15 for calculating:
Y=LOG(COS(A/B)+

IA-B) may be rewritten as:

SQROOT=2
FCOS=4
FLOG=6
INPUT=13
OUTPUT=14
*7
5600
*200
KCC

TLS
BEGI N,

A,
B,
TEMP,

JMSI7
/CALL INTERPRETER
INPUT
/READ A
FPUT A
/STORE IT
INPUT
/READ B
FPUT B
FMPYA
/A·B
SQROOT
FPUT TEMP /(A· B)** .5
FGET A
FDIV B
FCOS
FADD TEMP
FLOG
OUTPUT
FEXT
JMP BEGIN

o
o
o
o
o
o
o
o
o

$
3-17

~igital 8-50-5

This is the basic package, the output controller, and the extended functions. Its core limits are:
7; 40-62; 4557-7577

3-18

CHAPTER 4
PROGRAM

0040
0041
0042
0043
0044
0045
0046
0047
0050
0051
0052
0053
0054

0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000
0000

0061

0000

5600
5601
5602
5603
5604
5605
5606
5607
5610
5611
5612
5613
5614
5615
5616
5617
5620
5621
5622
5623
5624
5625
5626
5627
5630
5631
5632

0000
7300
3043
3047
1600
3253
1253
0263
7650
5214
1261
0200
3256
1262
0253
1256
3256
1264
0253
7650
5227
1656
3256
2200
1656
3040
1256

LISTINGS

/FlOATING POINT ARITHMETIC INTERPRETER
*40
EX1,
0
/OPERAND STORAGE
AC1H,
0
AC1 l,
0
OVER 1,
0
EXP,
0
/F.A.
HaRD,
0
lORD,
0
OVER2,
0
EXP1,
0
QUal,
0
FPAC1,
0
0
0
*61
FLAG,
/ARITHMETIC ERROR FLAG
0
*5600
FPNT,

lOOP01,

0
ClA Cll
DCA OVER1
DCA OVER2
TAD I FPNT
DCA JUMP
TAD JUMP
AND PAGENO
SNA ClA
JMP .+3
TAD MASK5
AND FPNT
DCA ADDR
TAD MASK7
AND JUMP
TAD ADDR
DCA ADDR
TAD INDRCT
AND JUMP
SNA ClA
JMP lOOP01
TAD I ADDR
DCA ADDR
ISZ FPNT
TAD I ADDR
DAC EX1
TAD ADDR

/GET NEXT INSTRUCTION

/GET PAGE BIT
/PAGE ZERO?
/YES
/NO
/C(FPNT)O-4 CONTAINS PAGE BITS
/GET 7 BIT ADDRESS

/INDIRECT BIT=1?

/NO-GO ON
/YES DEFER

/EXPONENT

4-1

5633
5634
5635
5636
5637
5640
5641
5642
5643
5644
5645
5646
5647
5650
5651
5652
5653
5654
5655
5656
5657
5660
5661
5662
5663
5664
5665
5666
5667
5670
5671
5672
5673
5674
5675
5676
5677
5700
5701
5702
5703
5704
5705
5706
5707
5710
5711
5712
5713
5714
5715
5716
5717
5720

3257
2257
1657
3041
2257
1657
3042
1253
7106
7006
0260
1265
3254
1654
3254
5654
0000
0000
0000
0000
0000
0017
7600
0177
0200
0400
5666
5742
5716
5737
5761
6305
5676
5705
5773
1040
3044
1041
3045
1042
3046
5201
1044
3656
2256
1045
3656
2256
1046
3656
5201
4771
5201
4772

JUMP,
JUMP2,
G02,
ADDR,
SAVE,
MASK3,
MASK5,
MASK7,
PAGENO,
INDRCT,
TABLE,

FlGT,

FlPT,

FlAD,

DCA SAVE
ISZ SAVE
TAD 'SAVE
DCA AC1H
ISZ SAVE
TAD I SAVE
DAC AC1l
TAD JUMP
Cll RTl
RTl
AND MASK3
TAD TABLE
DCA JUMP2
TAD I JUMP2
DCA JUMP2
JMP I JUMP2
0
0
0
0
0
0017
7600
0177
0200
0400
.+1
EXIT
FlAD
FlSU
FlMY
FlDV
FlGT
FlPT
NORF
TAD EXl
DCA EXP
TAD AC1H
DCA HORD
TAD ACll
DCA lORD
JMP FPNT+1
TAD EXP
DCA I ADDR
ISZ ADDR
TAD HORD
DCA I ADDR
ISZ ADDR
TAD lORD
DCA I ADDR
JMP FPNT+1
JMS I AlGN
JMP FPNT+1
JMS I UNORM

/HIGH ORDER MANTISSA

flOW ORDER MANTISSA

/GET BITS 0-2, IE OPCODE
/lOOKUP IN TABLE

/GO THERE

/TABlE USED IN INTERPRETING
/BITS 0-2 OF PSEUDO
/INSTRUCTION
/IF OPCODE=O, GO TO EXIT
/AND INTERPRET BITS 8-11

/FGET=5

/FPUT=6

/FlAD=l - FIRST ALIGN EXPONENTS
/RETURN IF NO ALIGNMENT IS POSSIBLE
/lARGER OF THE TWO IS IN F.A.
4-2

5721
5722
5723
5724
5725
5726
5727
5730
5731
5732
5733
5734
5735
5736

7100
1043
1047
3047
7004
1042
1046
3046
7004
1041
1045
3045
4770
5201

5737
5740
5741
5742
5743
5744
5745
5746
5747
5750
5751
5752
5753
5754
5755
5756
5757
5760
5761
5762
5763
5764
5765
5766
5767
5770
5771
5772
5773
5774

4741
5316
6400
1253
0260
7450
5600
1360
3254
1654
3254
1200
3255
4654
1255
3200
5201
6544
7201
1040
1044
3044
4767
5201
6221
6600
6020
6564
4770
5201

6000
6001
6002
6003
6004
6005
6006
6007

0000
7300
1047
7041
3047
1046
7040
7430

Cll
TAD OVER1
TAD OVER2
DCA OVER2
RAl
TAD AC1l
TAD lORD
DCA lORD
RAl
TAD AC1H
TAD HORD
DCA HORD
JMS I NORM
JMP FPNT+1
FlSU,
OPMINS,
EXIT,

ACON6,
FlMY,

MUlT,
NORM,
AlGN,
UNORM,
NORF,
*6000
ACMINS,

/TRIPlE PRECISION ADDITION
/SINCE BITS ARE SHIFTED
/RIGHT

/NORMALIZE THE RESULT

JMS I OPMINS
JMP FlAD
MINUS2
TAD JUMP
AND MASK3
SNA
JMP I FPNT
TAD ACON6
DCA JUMP2
TAD I JUMP2
DCA JUMP2
TAD FPNT
DCA G02
JMS I JUMP2
TAD G02
DCA FPNT
JMP FPNT+1
TABlE6-1
ClAIAC
TAD EX1
TAD EXP
DCA EXP
JMS I MUlT
JMP FPNT+1
DMUlT
DNORM
ALIGN
DUNORM
JMS I NORM
JMP FPNT+1

/FSUB=2 - NEGATE THE OPERAND
/ADD

0
Cll ClA
TAD OVER2
CMA lAC
DCA OVER2
TAD LORD
CMA
SZL

/ROUTINE TO PERFORM

/OPCODE=O
/ARE BITS 8-11=0?
/YES=FEXT
/lOOKUP ON TABLE

/CAll AS A SUBROUTINE
/RESTORE F.P. POINTER
/GET NEXT PSEUDO INSTRUCTION
/CAlLiNG CAN BE TO A DEPTH ONE
/FMPY=3
/ADD EXPONENTS TOGETHER
/MUlTIPlY

/TRIPlE PRECISION NEGATION
/OF FLOATING AC

4-3

6010
6011
6012
6013
6014
6015
6016
6017
6020
6021
6022
6023
6024
6025
6026
6027
6030
6031
6032
6033
6034
6035
6036
6037
6040
6041
6042
6043
6044
6045
6046
6047
6050
6051
6052
6053
6054
6055
6056
6057
6060
6061
6062
6063
6064
6065
6066
6067
6070
6071
6072
6073
6074
6075

7101
3046
1045
7040
7430
7101
3045
5600
0000
1045
7640
5227
1046
7650
5337
1041
7640
5235
1042
7650
5620
1040
7041
1044
7450
5314
7500
7041
3342
1342
1343
7700
5256
4316
7430
1350
1347
5325
4316
7420
1350
1347
3344
1342
7041
1744
3744
2344
1344
3345
2345
1345
3346
2346

ALIGN,

CLL lAC
DCA LORD
TAD HORD
CMA
SZL
CLL lAC
DCA HORD
JMP I ACMINS
0
TAD HORD
SZA CLA
JMP .+4
TAD LORD
SNA CLA
JMP NOHERE
TAD AC1H
SZA CLA
JMP .+4
TAD AC1L
SNA CLA
JMP I ALIGN
TAD EX1
CMA lAC
TAD EXP
SNA
JMP DONE
SMA
CMA lAC
DCA AMOUNT
TAD AMOUNT
TAD TEST2
SMA CLA
JMP .+6
JMS OUTGO
SZL
TAD TAG2
TAD TAG1
JMP NOGO
JMS OUTGO
SNL
TAD TAG2
TAD TAG1
DCA TEST3
TAD AMOUNT
CMA lAC
TAD I TEST3
DCA I TEST3
ISZ TEST3
TAD TEST3
DCA TEST4
ISZ TEST4
TAD TEST4
DCA TEST5
ISZ TEST5
4-4

ISUBROUTINE TO ALIGN
IBINARV POINTS FOR
ADD-SUBTRACT

liS MANTISSA ZERO?
/yES

liS OPERAND ZERO?

IBOTH ARE ZERO-EXIT

I ARE EXPONENTS EQUAL?
IVES

ICAN EXPONENTS BE ALIGNED?
IVES

INO

ISET UP ADDRESSES

6076
6077
6100
6101
6102
6103
6104
6105
6106
6107
6110
6111
6112
6113
6114
6115
6116
6117
6120
6121
6122
6123
6124
6125
6126
6127
6130
6131
6132
6133
6134
6135
6136
6137
6140
6141
6142
6143
6144
6145
6146
6147
6150
6151
6152
6153
6154
6155
6156
6157
6160
6161
6162

7100
1744
7510
7020
7010
3744
1745
7010
3745
1746
7010
3746
2342
5276
2220
5620
0000
1040
7041
1044
7004

SHIFT,

DONE,
OUTGO,

7200
5716
3344
1744
3044
2344
1744
3045
2344
1744
3046
5620
1040
3044
5314
0000
0030
0000
0000
0000
0044
7774
5601
1362
3044
1362
3045
7040
3046
2061
5751
3777

NOGO,

NOHERE,

AMOUNT,
TEST2,
TEST3,
TEST4,
TEST5,
TAG1,
TAG2,
RETN2,
ERROR1,

GOOF,

CLL
TAD I TEST3
SPA
CML
RAR
DCA I TEST3
TAD I TEST4
RAR
DCA I TEST4
TAD I TEST5
RAR
DCA I TEST5
ISZ AMOUNT
JMP SHIFT
ISZ ALIGN
JMP I ALIGN
0
TAD EXl
CMA lAC
TAD EXP
RAL
CLA
JMP I OUTGO
DCA TEST3
TAD I TEST3
DCA EXP
ISZ TEST3
TAD I TEST3
DCA HORD
ISZ TEST3
TAD I TEST3
DCA LORD
JMP I ALIGN
TAD EXl
DCA EXP
JMP DONE
0
0030
0
0
0
EXP
EX1-EXP
FPNT+l
TAD GOOF
DCA EXP
TAD GOOF
DCA HORD
CMA
DCA LORD
ISZ FLAG
JMP I RETN2
3777

/THIS ROUTINE DOES
/THE ACUTAL SHIFTIN G

/DETERMINE WHICH TO SHIFT

/CAN'T BE ALIGNED
/LARGEST GOES INTO FAC

/MANTISSA=O

/DIVISION BY ZERO
/SET TO LARGEST + VALUE

/SET FLAG

4-5

6163
6164
6165
6166
6167
6170
6171
6172

0000
4407
6052
3052
0000
5763
0000
5771

6200
6201
6202
6203
6204
6205
6206
6207
6210
6211
6212
6213
6214
6215
6216
6217
6220
6221
6222
6223
6224
6225
6226
6227
6230
6231
6232
6233
6234
6235
6236
6237
6240
6241
6242
6243
6244
6245
6246
6247
6250
6251
6252
6253

0000
7300
1045
7510
7120
7010
3045
1046
7010
3046
1047
7010
3047
7100
2044
7000
5600
0000
7300
1363
3347
4336
1042
3756
1046
4755
7200
1757
3047
1045
3756
1042
4755
1047
3047
7004
1757
3367
7004
3370
1041
3756
1046
4755

SQUARE,

EXIT6,
*6200
DIV1,

DMUlT,

0
JMS 10007
FPUT FPACl
FMPY FPACl
FEXT
JMP I SQUARE
0
JMP I EXIT6

/DUMMY SUBROUTINE

0
ClA Cll
TAD HaRD
SPA
Cll CMl
RAR
DCA HaRD
TAD lORD
RAR
DCA lORD
TAD OVER2
RAR
DCA OVER2
Cll
ISZ EXP
Nap
JMP I DIV]
0
ClA Cll
TAD SMAClA
DCA SNSWIT
JMS SIGN
TAD ACl l
DCA I MP2PT
TAD lORD
JMS I MP4PT
ClA
TAD I MP5PT
DCA OVER2
TAD HaRD
DCA I MP2PT
TADACll
JMS I MP4PT
TAD OVER2
DCA OVER2
RAl
TAD I MP5PT
DCA D
RAl
DCA KEEP
TAD AC1H
DCA I MP2PT
TAD lORD
JMS I MP4PT

/SHIFT FAC RIGHT

/DOUBlE PRECISION MULTIPLY
/SAVE PRODUCT TRIPLE PRECISION
/CAllS A SINGLE PRECISION
/MUlTIPl Y 3 TIMES

4-6

6254
6255
6256
6257
6260
6261
6262
6263
6264
6265
6266
6267
6270
6271
6272
6273
6274
6275
6276
6277
6300
6301
6302
6303
6304
6305
6306
6307
6310
6311
6312
6313
6314
6315
6316
6317
6320
6321
6322
6323
6324
6325
6326
6327
6330
6331
6332
6333
6334
6335
6336
6337
6340

1047
3047
7004
1757
1367
3367
7004
1370
3370
1045
3756
1041
4755
1367
3046
7004
1757
1370
3045
4760
3047
2365
5621
4773
5621
1041
7640
5313
1042
7650
5774
1040
7041
1044
7001
3044
1362
3347
4336
4761
1757
1041
7630
7001
3042
3041
2365
4773
5735
5720
0000
1366
3365

FLDV,

SIGN,

TAD OVER2
DCA OVER2
RAL
TAD I MP5PT
TAD D
DCA D
RAL
TAD KEEP
DCA KEEP
TAD HORD
DCA I MP2PT
TAD AC1H
JMS I MP4PT
TAD D
DCA LORD
RAL
TAD I MP5PT
TAD KEEP
DCA HORD
JMS I NORMF
DCA OVER2
ISZ SGN
JMP I DMULT
JMS I MINS
JMP I DMULT
TAD AC1H
SZA CLA
JMP .+4
TAD ACl L
SNA CLA
JMP I ERROR
TAD EXl
CMA lAC
TAD EXP
lAC
DCA EXP
TAD SPACLA
DCA SNSWIT
JMS SIGN
JMS I DIVIDE
TAD I MP5PT
TAD AC1H
SZL CLA
lAC
DCA ACl L
DCA AC1H
ISZ SGN
JMS I MINS
JMP I .+1
FLAD+2
0
TAD REST
DCA SGN

/DIVISION BY ZERO

/SUBTRACT EXPONENTS

/SET UP SIGNS
/DIVIDE
/ROUND OFF IN 23RD BIT

/TEST SIGN
/NEGATE
~DD ROUNDING
/TEST SIGN OF RESUL T
/SET UP BY MULTI PL Y AN D
/DIVIDE

4-7

6341
6342
6343
6344
6345
6346
6347
6350
6351
6352
6353
6354
6355
6356
6357
6360
6361
6362
6363
6364
6365
6366
6367
6370
6371
6372
6373
6374

1045
7700
5346
4773
2365
1041
7700
5736
4771
2365
7000
5736
6437
6471
6465
6600
6472
7710
7700
5601
0000
7776
0000
0000
6400
6420
6000
6152

6400
6401
6402
6403
6404
6405
6406
6407
6410
6411
6412
6413
6414
6415
6416
6417
6420
6421
6422
6423
6424
6425
6426
6427

0000
7300
1043
7041
3043
1042
7040
7430
7101
3042
1041
7040
7430
7101
3041
5600
0000
7300
1041
7510
7120
7010
3041
1042

SNSWIT,

MP4PT,
MP2PT,
MP5PT,
NORMF,
DIVIDE,
SPAClA,
SMAClA,
RETURN,
SGN,
REST,

0,
KEEP,
MINS2,
RAR2,
MINS,
ERROR,
*6400
MINUS2,

DIV2,

TAD HORD
SMA ClA
JMP .+3
JMS I MINS
ISZ SGN
TAD AC1H
SMA ClA
JMP I SIGN
JMS I MINS2
ISZ SGN
NOP
JMP I SIGN
MP4
MP2
MP5
DNORM
DUBDIV
SPA ClA
SMA ClA
FPNT+l
0
7776
0
0
MINUS2
DIV2
ACMINS
ERRORl

lOR SPA ClA

I-AC1H,ACll
I ACl H, ACl L/2

0
ClA Cll
TAD OVERl
CMA lAC
DCA OVERl
TAD ACll
CMA
SZl
Cll lAC
DCA ACll
TAD AC1H
CMA
SZl
Cll lAC
DCA AC1H
JMP I MINUS2
0
ClA Cll
TAD AC1H
SPA
Cll CMl
RAR
DCA AC1H
TAD ACl l

INEGATE OPERAND
ITRIPlE PRECISION

ISH 1FT OPERAND RIGHT
ITRIPlE PRECISION

4-8

6430
6431
6432
6433
6434
6435
6436
6437
6440
6441
6442
6443
6444
6445
6446
6447
6450
6451
6452
6453
6454
6455
6456
6457
6460
6461
6462
6463
6464
6465
6466
6467
6470
6471
6472
6473
6474
6475
6476
6477
6500
6501
6502
6503
6504
6505
6506
6507
6510
6511
6512
6513
6514

7010
3042
1043
7010
3043
7100
5620
0000
3266
3265
1270
3267
7100
1266
7010
3266
1265
7420
5255
7100
1271
7010
3265
2267
5245
1266
7010
7100
5637
0000
0000
0000
7764
0000
0000
7300
3051
1344
3267
5306
1046
7004
3046
1045
7004
3045
1042
1046
3271
7004
1045
1041
7420

MP4,

MP5,
MP1,
MP3,
THIR,
MP2,
DUBDIV,

DV3,

DVX,

RAR
DCA AC1 l
TAD OVER1
RAR
DCA OVER1
Cll
JMP I DIV2
0
DCA MP1
DCA MP5
TAD THIR
DCA MP3
Cll
TAD MP1
RAR
DCA MP1
TAD MP5
SNl
JMP .+3
Cll
TAD MP2
RAR
DCA MP5
ISZ MP3
JMP MP4+6
TAD MP1
RAR
Cll
JMP I MP4
0
0
0
7764

/SINGlE PRECISION MULTIPLY
/12 BITS BY 12 BITS

0
ClA Cll
DCA QUOl
TAD MIF
DCA MP3
JMP DVX
TAD lORD
RAl
DCA lORD
TAD HaRD
RAl
DCA HORD
TAD AC1l
TAD lORD
DCA MP2
RAl
TAD HaRD
TAD AC1H
SNl

/DOUBlE PRECISION DIVIDE

4-9

6515
6516
6517
6520
6521
6522
6523
6524
6525
6526
6527
6530
6531
6532
6533
6534
6535
6536
6537
6540
6541
6542
6543
6544
6545
6546
6547
6550
6551
6552
6553
6554
6555
6556
6557
6560
6561
6562
6563

5321
3045
1271
3046
7200
1051
7004
3051
1047
7004
3047
2267
5300
1051
3046
1045
7106
3265
1047
3045
3047
1265
5672
7751
6163
6656
6171
6171
6171
6171
6171
6171
6171
6171
6171
6171
6171
6171
6171

6564
6565
6566
6567
6570
6571
6572

0000
4220
4772
2040
7000
5764
6200

6600
6601
6602
6603
6604
6605

0000
7300
3255
3254
1045
7510

DV2,

MIF,
TABLE6,

DUNORM,

RAR1,
*6600
DNORM,

JMP DV2-1
DCA HORD
TAD MP2
DCA LORD
CLA
TAD QUOL
RAL
DCA QUOL
TAD OVER2
RAL
DCA OVER2
ISZ MP3
JMP DV3
TAD QUOL
DCA LORD
TAD HORD
CLL RTL
DCA MP5
TAD OVER2
DCA HORD
DCA OVER2
TAD MP5
JMP I DUBDIV
7751
SQUARE
SQROOT
EXIT6
EXIT6
EXIT6
EXIT6
EXIT6
EXIT6
EXIT6
EXIT6
EXIT6
EXIT6
EXIT6
EXIT6
EXIT6

/TABLE FOR INTERPRETATION
/OF BITS 8-11
/CONTAINS ABSOLUTE ADDRESSES
/OF PROGRAMS CALLED AS
/SUBROUTINES
/EXIT6=A DUMMY OR NOP

0
JMS DIV2
JMS I RAR1
ISZ EX1
NOP
JMP I DUNORM
DIV1

/SHIFT OPERAND RIGHT

/SUBROUTINE TO NORMALIZE
/FLOATING ACCUMULATOR

0
CLA CLL
DCA AMTl
DCA SIGN1
TAD HORD
SPA

/15 MANTISSA NEGATIVE?
4-10

6606
6607
6610
6611
6612
6613
6614
6615
6616
6617
6620
6621
6622
6623
6624
6625
6626
6627
6630
6631
6632
6633
6634
6635
6636
6637
6640
6641
6642
6643
6644
6645
6646
6647
6650
6651
6652
6653
6654
6655
6656
6657
6660
6661
6662
6663
6664
6665
6666
6667
6670
6671
6672
6673

2254
7640
5217
1046
7640
5217
1047
7650
5251
1254
7640
4653
1045
7104
7710
5241
1047
7104
3047
1046
7004
3046
1045
7004
3045
2255
5222
1255
7041
1044
3044
1254
7640
4653
5600
3044
5600
6000
0000
0000
0000
3362
4407
6052
0000
1045
7710
5351
1044
7100
7510
7020
7010
3356

G06,

lOP,

EXITl ,

EXIT2,
NEG,
SIGN1,
AMTl,
SQROOT,

ISZ SIGNl
SZA ClA
JMP G06
TAD lORD
SZA ClA
JMP G06
TAD OVER2
SNA ClA
JMP EXIT2
TAD SIGNl
SZA ClA
JMS I NEG
TAD HORD
RAl Cll
SPA ClA
JMP EXITl
TAD OVER2
Cll RAl
DCA OVER2
TAD lORD
RAl
DCA lORD
TAD HORD
RAl
DCA HORD
ISZ AMTl
JMP lOP
TAD AMTl
CMA lAC
TAD EXP
DCA EXP
TAD SIGNl
SZA ClA
JMS I NEG
JMP I DNORM
DCA EXP
JMP I DNORM
ACMINS
0
0
0
DCA FlAGl
JMS 1007
FPUT FPACl
FEXT
TAD HORD
SPA ClA
JMP SQENDl
TAD EXP
Cll
SPA
CMl
RAR
DCA ITERl

/YES
/IS MANTISSA=O?
/NO

/YES
/NO
/NEGATlVE?
/YES
/Wlll SHIFT BE TOO FAR?

/YES

/NO SHIFT MANTISSA lEFT

/COUNT NO. OF TIMES SH IFTED
/CORRECT EXPONENT

/NEGATlVE?
/YES
/SET TO ZERO

/FlOATING SQUARE ROOT
/TAKES ROOT OF ABSOLUTE
/VAlUE
/NEWTON'S METHOD IS USED

/NUMBER IS NEGATIVE

/MAKE FIRST APPROXIMATION
4-11

6674
6675
6676
6677
6700
6701
6702
6703
6704
6705
6706
6707
6710
6711
6712
6713
6714
6715
6716
6717
6720
6721
6722
6723
6724
6725
6726
6727
6730
6731
6732
6733
6734
6735
6736
6737
6740
6741
6742
6743
6744
6745
6746
6747
6750
6751
6752
6753
6754
6755
6756
6757
6760
6761
6762

7430
2356
7000
1361
3357
3360
1053
7640
5310
1054
7650
5354
4407
5052
4356
1356
0000
7240
1044
3044
1044
7041
1356
7640
5345
1045
7041
1357
7640
5345
1046
7041
1360
7500
7041
7001
7710
5345
1362
3061
5656
4407
6356
0000
5310
4653
2362
5260
3044
5656
0000
0000
0000
3015
0000

CLCU,

ROOTGO,

SQEND1,

SQEND,
ITER1,

SQCON1,
FLAG1,

SZL
ISZ ITER1
NOP
TAD SQCON1
DCA ITER1+1
DCA ITER1+2
TAD FPAC1+1
SZA CLA
JMP CLCU
TAD FPAC1+2
SNA CLA
JMPSQEND
JMS 10007
FGET FPAC1
FDIV ITER1
FADD ITER1
FEXT
CLA CMA
TAD EXP
DCA EXP
TAD EXP
CMA lAC
TAD ITER1
SZA CLA
JMP ROOTGO
TAD HORD
CMA lAC
TAD ITER1+1
SZA CLA
JMP ROOTGO
TAD LORD
CMA lAC
TAD ITER1+2
SMA
CMA lAC
lAC
SPA CLA
JMP ROOTGO
TAD FLAG1
DCA FLAG
JMP I SQROOT
JMS 10007
FPUT ITER1
FEXT
JMP CLCU
JMS I NEG
ISZ FLAG1
JMP SQROOT+2
DCA EXP
JMP I SQROOT
0
0
0
3015
0
4-12

INUMBER=O

I ARE EXPONENTS EQUAL?
INO
IARE HIGH-ORDER MANTISSAS EQUAL?

INO
IDO LOW-ORDER MANTISSAS AGREE?
/WITHIN ONE BIT?

INO

INEGATE FAC

ACMINS
ACON6
AC1H
AClL
ADDR
ALGN
ALIGN
AMOUNT
AMT1
CLCU
D
DIVIDE
DIV1
DIV2
DMULT
DNORM
DONE
DUBDIV
DUNORM
DVX
DV2
DV3
ERROR
ERROR1
EXIT
EXITl
EXIT2
EXIT6
EXP
EXP1
EX1
FLAD
FLAG
FLAG1
FLDV
FLGT
FLMY

6000
5760
0041
0042
5656
5771
6020
6142
6655
6710
6367
6361
6200
6420
6221
6600
6114
6472
6J64
6506
6522
6500
6374
6152
5742
6641
6651
6171
0044
0050
0040
5716
0061
6762
6305
5676
5761

FLPT
FLSU
FPAC1
FPNT
GOOF
G02
G06
HORD
INDRCT
ITER1
JUMP
JUMP2
KEEP
LOOP01
LOP
LORD
MASK3
MASK5
MASK7
MIF
MINS
MINS2
MINUS2
MP1
MP2
MP2PT
MP3
MP4
MP4PT
MP5
MP5PT
MULT
NEG
NOGO
NOHERE
NORF
NORM

4-13

5705
5737
0052
5600
6162
5655
6617
0045
5664
6756
5653
5654
6370
5627
6622
0046
5660
5661
5662
6544
6373
6371
6400
6466
6471
6356
6467
6437
6355
6465
6357
5767
6653
6125
6137
5773
5770

NORMF
OPMINS
OUTGO
OVER1
OVER2
PAGENO
QUOL
RAR1
RAR2
REST
RETN2
RETURN
ROOTGO
SAVE
SGN
SHIFT
SIGN
SIGN1
SMACLA
SNSWIT
SPACLA
SQCON1
SQEND
SQEND1
SQROOT
SQUARE
TABLE
TABLE6
TAG1
TAG2
TEST2
TEST3
TEST4
TEST5
THIR
UNORM

6360
5741
6116
0043
0047
5663
0051
6572
6372
6366
6151
6364
6745
5657
6365
6076
6336
6654
6363
6347
6362
6761
6754
6751
6656
6163
5665
6545
6147
6150
6143
6144
6145
6146
6470
5772

IFLOATING POINT 1/0 ROUTINES
IREQUIRES FLOATING POINT INTERPRETER
IENTRY AT 0007
0001

5600

*7
FPNT,

5600

0044
0045
0046

0000
0000
00((J0

*44
EXPONT,
HORDER,
LORDER,

0
0
0

0052
0053
0054
0055
0056
0051
0060

((J((J((J0
((J0((J0
000((J
7111
1117
0000
0000

16161
16170
6111
6172
6173
6174
6115
6176
6111

0000
1056
7550
5767
1377
4776
5767
7344
0212

*52
FPAC1,
SWI Tl ,
SWIT2,
CHAR,
DSWIT,
*6767
PRCHAR,

OPUT,
LFED,

0
0
0
7771
7717
0
0

IIF : ((J, NO CH-LF AFTER OUTPUT
IIF = 0, NO LF AFTER CR IN INPUT
ICONTAINS LAST CHARACTER READ
1= 0 IF NO CONVERSION TOOK PLACE

0
TAD SWIT2
SNA CLA
Jr1P 1 PRCHAR
TAD LFED
JMS I OPUT
JMP I PRCHAR
OUT
0212

4-14

IDOUBLE PRECISION DECIMAL-BINARY
IINPUT AND CONVERSION
*1000
11,lrlJ0
1001
1002
1003
1004
701d5
10"'6
116"'1
1"'10
7011
1012
7"13
7"'14
1015
1016
7" 17
7"'210

00"0
12i10
3045
3046
3265
3266
4342
1336

DECONV,

145~

5217
1335
1440
5220
7240
3265
4342
721d0
h)57
1337
75"'''
5600
1340
1510
560fd
3263
11lJ45
1lJ341

1021
11rl22
7023
1024
7"25
7"'26
7027
71cl3"
7031
71lJJ2
7"'33
7k.lJ4
7kJ35
7"'36
71637
71cJ4.:J
1041
7"'42
7043
7044
7045
7046
7"'47
7050
7051
7052
71lJ53
7054
1055
7056
7057
106"

5217
2060
2266
4241
5217
0"'160
1046
3261
1"45
3262
3264
4267
4267
4303
4267
1263
3261
3262
4303
1264
5641

7061
1062
1063
712164
1065
71666
1061

001210
0000
0000
01210121
kl001c!)
0000
0000

DECON,

744;0

MULTI0,

LTEMP,
HTEMP,
DIGIT,
REMAIN,
SIGN,
DNUMBR,
MULT2,

0
CL.A
DCA HORDER
DCA LORDER
DCA SIGN
DCA DNUMBR
JMS INPUT
TAD PLUS
SNA
JMP DECON
TAD MINUS
SZA
JMP .+4
CLA CMA
DCA SIGN
JMS INPUT
CLA
TAD CHAR
TAD MIN9
SMA
JMP I DECONV
TAD PLUS12
SPA
JI'lP I DECONV
DCA DIGIT
TAD HaRDER
AND MASK
SZA
JMP DECON
ISZ DSWIT
ISZ DNUMBR
Jl1S MUL Tl L1
JMP DECON
0
TAD LaRDER
DCA LTEMP
TAD HaRDER
DCA HTEMP
DCA REMAIN
JMS NULT2
JMS MULT2
JMS DUBLAD
JMS MULT2
TAD DIGIT
DCA LTEMP
DCA HTEMP
JMS DUBLAD
TAD REMAIN
JMP I MULTlI2I

IINITIALIZE MANTISSA

ITEST FOR SIGN

IIF-, SET SWITCH
lIS IT A DIGIT

INa
INO
IYES
IOVERFLOW1
IYES-IGNORE
IINDEX NUMBER OF DIGITS
ICONTINUE
IROUTINE TO MULTIPLY
IDOUBLE PRECISION wORD
IBY TEN (DECIMAL)
IREMAIN=REMAINDER
ICALL SUBROUTINE TO
IMULTIPLY BY TWO
ICALL DOUBLE ADD

IADD LAST DIGIT RECEIVED

IEXIT WITH REMAINDER
lIN AC

0
0
121
0
0
0

IDOUBLE PRECISION WORD
/STORAGE FOR DIGIT
1=0 IF PLUS: =7777 IF MINUS

I=NUMBER OF DIGITS
IMULTIPLY LaRDER, HaRDER BY 2

4-15

7070
7Ii') 71
7072
7073
7074
7075
7076
7100
71f211
7102
7103
7104
7105
7106
7107
7110
7111
7112
7113
7114
7115
7116
7117
7120
7121
7122
7123
7124
7125
7126
7127
7130
7131
7132
7133
7134

7300
1046
7004
3046
1045
7004
3045
1264
7004
3264
5667
liHHI10
7300
1046
1261
3046
7004
1045
1262
3045
7004
1264
3264
5703
0000
7300
2265
5720
1046
7041
3046
1045
7040
7430
7001
3045
5720

7135
7136
7137
7140
7141

7776
7525
7506
0012
7600

7142
7143
7144
7145
7146
7147
7150
7151
7152
7153
7154
7155
7156
7157
7160

0000
7200
6031
5344
6036
3057
1057
4766
1057
7450
5343
1370
7450
5767
1371

1~11

DUBLAD,

MSIGN,

CLA CLL
TAD LORDER
RAL
DCA LORDER
TAD HORDER
RAL
DCA HORDER
TAD REMAIN
RAL
DCA REMAIN
JMP I MULT2
0
CLA CLL
TAD LORDER
TAD LTEMP
DCA LORDER
RAL
TAD HORDER
TAD HTEMP
DCA HORDER
RAL
TAD REMAIN
DCA REMAIN
JMP I DUBLAD
0
CLA CLL
ISZ SIGN
JMP I MSIGN
TAD LORDER
CMA lAC
DCA LORDER
TAD HORDER
CMA
SZL
lAC
DCA HORDER
JMP I MSIGN

IDOUBLE PRECISION ADDITION

IROUTINE TO FOR[1
12'S COMPLEMENT IF
IMINUS
ISIGN=0000: EXIT

253-255
MINUS,
ITEST FOR SIGN
PLUS,
-253
-272
ITEST FOR DIGIT
MIN9,
272-260
PLUSI2,
760iO
MASK,
ITEST FOR OV~RFLOill
IINPUT A CHARACTER, IF CR, TEST
IINPUT SWITCH TO SEE IF LF SHOULD
IBE TYPED. IF RUBOUT, RESTART INPUT
13
INPUT,
IINPUT A CHARACTER
CLA
KSF
JMP .-1
KRB
DCA CHAR
TAD CHAR
JMS I OUTPUT
TAD CHAR
SNA
JMP INPUT+1
IIGNORE BLANKS
TAD MRBOUT
SNA
JMP I RESTRT
IRUBOUT-RESTART INPUT
TAD MINCR
4-16

7161
7162
7163
7164

76510

7165
7166
7167
717ft)
7171

6767
7344
7401
7401
10162

SNA CLA
JMS 1 PRINT
TAD CHAR
JMP 1 INPUT

4765
1057
5742
PRINT,
OUTPUT,
RESTRT,
MRBOUT,
MINCR,

ICR - SEE IF TO BE FOLLOWED
IBY l.F
IEXIT ROUTINE

PRCHAR
OUT
FLINTP+1
-0377
377-215

IFLOATING OUTPUT "E" FORMAT
TSF
IUSES:
JMP .-1
I
I
TLS

720k:>
7201
72(.12
7203
7204
72~5

7206
7207
72110
7211
7212
7213
7214
7215
7216
7217
722iil
7221
7222
7223
7224
7225
7226
7227
7230
7231
7232
7233
7234
7235
7236
7237
724(.1
7241
7242
"243
7244
7245
7246
7247
72510
7251

o"tHO

*72100

FLOUTP,

JMS FOUTCN
ICONVERT MANTISSA AND OUTPUT
TAD BEXP
DCA EXPONT
TAD CHE
JMS OUT
JMS 1 FEXPPT
ICONVERT EXPONENT AND OUTPUT
TAD SWITl
IPRINT CR-LF?
1~55
SNA CLA
7650
JMP 1 FLOUTP
56i,"!)
INO-EXIT
TAD CARRTN IYES
1341
JMS OUT
4344
TAD LNFEED
1342
JMS OUT
4344
JMP 1 FLOUTP
IEXIT
5600
ITHIS WHOLE SUBROUTINE MAY BE ALTERED TO BUFFER
ITHE OUTPUT DIGITS: CHANGE JMS OUTDG TO DCA 1 110, ETC.
IclltllOk:)
FOUTCN,
'"
CLA CLL
73I.!J1cl
INUMBER>07?
TAD HORDER
1045
SPA CLA
7710
I NO SET LINK
CLA CML
7220
TAD SPLUS
IYES
1327
7430
SZL
TAD SMINUS INO
1330
JMS OUT
4344
10UTPUT "0"
JMS OUTDG
4352
TAD PERIOD
1331
4344
JMS OUT
10UTPUT • "
73100
CLA CLL
TAD HORDER
11045
771010
SMA CLA
5242
JMP FG01
CMA
710410
INUMBER IS NEGATIVE
3733
DCA 1 SNPT
INEGATE
JMS 1 MSNPT
4732
CLA CMA
ISUBTRACT 1 FROM BINARY EXPONENT
72410 FG01,
TAD EXPONT
11044
ICOMPENSATE AT FG04
DCA EXPONT
3"'44
3324
DCA BEXP
IINITIALIZE DECIMAl. EXPONENT
TAD EXPONT
lIS -4<EXPONENT<-1
11044 FG02,
75010
SMA
JMP FG03
ITOO LARGE: MUl.TIPLY BY 1/1V
5263
TAD FOUR
1326

4217
1324
3044
1343
4344
4737

4-17

1252 1100
1253 5270
1254 4401
1255 3740
1256 0000
7257 7240
7260 1324
7261 3324
7262 5246
1263 4407
7264 3372
7265 0000
7266 2324
7267 5246
7270 3734
7271 4736
7272 4135
7273 7410
7274 4357
7275 2044
7276 5274
1277 7450
7300 5311
7301 4352
7302 1325
7303 3044
7304 4735
7305 4352
7306 2044
7307 5304
7310 5617

FG03,

FG04,

SMA Cl.A
JMP FG04
JMS I FPNT
FMPY I TENPT
FEXT
CLA CMA
TAD BEXP
DCA SEXP
JMP FG02
JMS I FPNT
FMPY C.10
FEXT
ISZ BEXP
JMP FG02
DCA I OPT
JMS I M2PT
JMS I MI0PT

ITOO SMALl.-TIMES TEN
ITEN

lONE TENTH
IMULTIPLY BY TWO
IIE.SHIFT LEFT
IMULTIPLY BY TEN

SKP

FG05A,

FG06,
FG06A,

7311
7312
7313
7314
7315
7316
7317
7320
7321
7322
7323

7240
1324
3324
1045
7640
5322
1046
7650
3324
7240
5302

FG07,

1324
1325
1326
7327
1330
7331
7332
7333
7334
7335
7336
7337
7340
7341
7342
7343

0000
1772
0004
0253
0002
0256
7120
7065
7063
7041
7067
7522
7504
0215
0212
0305

BEXP,
MINUS7,
FOUR,
SPLUS,
SMINU~,

PERIOiJ,
MSNPT,
SNPT,
OPT,
MI0PT,
M2PT,
FEXPPT,
TENPT,
CARRTN,
LNFEED,
CKE,

JMS DIVTWO
ISZ EXPONT
JMP FG05A

ICOMPENSATE FOR
IBINARY EXPONENT

SNA

115 FIRST DIGIT A ZERO
IYES, IGNORE
IMULTIPl.ICATIONS YIEl.D
IDECIMAL DIGITS AS HIGH
IORDER REMAINDERS
lIE •• 672XI0:6+.72 •• ETC

JMP FG07
JMS OUTDG
TAD MINUS7
DCA EXPONT
JMS 1 MI0PT
JMS OUTDG
ISZ EXPONT
JMP FG06A
JMP I FOUTCN

17 DIGITS OUTPUT??

INO: CONTINUE
IYES:EXIT

CLA CMA
TAD BEXP
DCA BEXP
TAD HORDER
SZA Cl.A
JMP .+4
TAD LORDER
SNA CLA
DCA BEXP
CLA CMA
JMP FG06+1

IIGNORE FIRST DIGIT
ISUBTRACT I FROM
IDECIMAL EXPONENT
115 MANTISSA ZERO?

IYES: EXP:0

ICONTAINS DECIMAl. EXPONENT
INUMBER OF DIGTS OUTPUT

"1172

QJ004
G53
255-253
256
MSIGN
SIGN
DIGIT
MULTl0
MULT2
FEXC
TEN
0215
0212
305

IPOINTERS

4-18

7344
7345
7346
7347
735121
7351

121121121121
612141
5345
612146
721210
5744

OUT,

121
TSF
JMP .-1
TLS
CLA
JMP lOUT

10UTPUT ONE ASCII CHARACTER

7352
7353
7354
7355

12101210
1356
4344
5752

OUTDG,

121
TAD C260
JMS OUT
JMP I OUTDG

10UTPUT ONE DIGIT

7356

0260

C260,

0260

7557
7360
7361
7362
7363
7364
7365
7366
7367
737121
7371
7572
7573
7374

121000
711121
3344
1045
712110
3045
1046
7"'1121
312146
1344
5757
7775
3146
3147

DIVTWO,

III

7400
74131
7402
7403
741214
7405
741216
741217

00"'121
7240
3313
306121
4716
7200
1057
1312
7640
5220
1313
765121
5222
3721
3313
5717

74H}

7411
7412
7413
7414
7415
7416
7417
7420
7421
7422
7423
7424
7425
7426
1427
743121
7431
7432
7433
7434
7435

1313
7650
1721
712141
3314
472121
1311
312144
441217
700121
612152
9099

1051
1319

IDIVIDE BY TWO IE.
IROTATE RIGHT
ITEMPORARY STORAGE

CLL RAR
DCA OUT
TAD HORDER
RAR
DCA HORDER
TAD LORDER
RAR
DCA LORDER
TAD OUT
JMP I DIVT\tIO
C.10,
7775
3146
3147
IFLOATING POINT INPuT
*7413121
FLINTP,
CLA CMA
DCA PRSW
DCA DS\tIIT
JMS I DPCVPT
CLA
TAD CHAR
TAD PER
SZA CLA
JMP FIGOI
TAD PRSW
SNA CLA
JMP FIG02
DCA I DPN
DCA PRSW
JMP I DPCSPT

ICONSTANT .10 USED IN
IFlOATING OUTPuT-PROVIDES

FIG01,
FIG02,

FIG03,

IINITIALIZE ·PERIOD SWITCri"
17777

= NO PERIOD

IPERIOD FOUND
ISECOND PERIOD
IYES,TERMINATE
INO - SET NUMBER OF DIGITS TO 0
ISET PERIOD SWITCH TO 121
ICONVERT REST OF STRING

TAD PRSW
SNA CLA
TAD I DPN
CMA lAC
DCA SEXP
JMS I MSGNPT
TAD C27
DCA EXPONT
JMS I FPNT
FNOR
FPUT FPACI
FEXT
TAD CHAR
TAD MINUSE

IPERIOD READ lN PREVIOUSLY?
IYES:-NUMBER OF DIGITS IN SEXP
INO

ITEST SIGN
INORMALIZE F.P. NUMBER
ISAVE NUMBER

4-19

7436
7437
744121
7441
7442
7443
7444
7445
7446
7447
7450
7451

7640
5252
4716
4720
If1:145
7510
7leW 1
1640
5277
1046
1314
3314

SZA CLA
I"E" READ IN?
JMP ENDFl
INO
JMS 1 DPCVPT
IYES - CONVi:RT DECIMAL EXPONENT
JMS 1 MSGNPT
ITEST SIGN
TAD HORDER
IEXPONENT TOO LARGE??
SPA
lAC
SZA CLA
J:1P EXCESS
IYES
TAD LORDER
INO:DECIMAL POINT IS
TAD SEXP
IC(SEXP)PLACES TO RIGHT
DCA SEXP
IOF LAST DIGIT
lEND OF FLOATING POINT INPUT
ICOMPENSATE FOR DECIMAL EXPONENTS

7452
745.3
7454
7455
7456
7457
7460
7461
7462
7463
7464
7465
7466
7467

44i17
5 ic:l 52
0000
1314
7450
56kH1
7700

ENDF'I,

7470
7471
7472
7473
7474
7475
7476
7471
751d1d
7501
75.)2
75"'3
75;04
75"'5
75106
75i07
7510
7511
7512
7513
7514
7515
7516
7517
7520
7521
7522
7523

527~

4407
3707
"1Il~0

2314
5255
560kl
4407
3304

F'I G04,

001/10

7240
1314
.3.314
5255
1.315
3044
1315
3045
561/)101
101O~4

EXCESS,

TiN,

2410id
101.11010

7372
747.3
iQld27
7522
kHHH'

fdfd"''''
3777

PC.lt1,
MINUSE,
C27,
PiR,
PRSW,
SEXP,
C3717,

Ji'lS I FPNT
F'GET FPACI
FEXT
TAD SEXP
SNA
JMP I FLINTP
SMA CLA
JMP FlG04
JMS 1 FPNT
FMPY I PC. lI/J
FEXT
ISZ SEXP
JMP ENDF'I+3
JMP 1 FLINTP

I. IS TO THE RIGHT:
IMULTIPLY BY 10

ICONTAINS DECIMAL EXPONENT

I. lI/J

3777

DPCVPT,
DECONV
DPCSPT,
DECON
MSGNPT,
MSIGN
DNUMBR
DPN,
IOUTPUT THE EXPONENT

01/)00

FEXC,

7300

I. IS TO THE LEFT:
ITIMES .1000

JMS I FPNT
FMPY TEN
FEXT
CLA CMA
TAD SEXP
DCA SEXP
Jl~P ENDF'1+3
TAD C.3 777
DCA EXPONT
TAD C.3 777
DCA HORDER
Jl'lP 1 FLINTP
010 I/J 4
24i1iO
00130
C.liQ
-305
01/J27
-256
0

71/J17
712O
7066

7~fd0

IRESTORE MANTISSA

CLA CLL
4-20

1524
7525
1526
1521
1530
1531
1532
1533
1534
1535
1536
1531
1540
1541
1542
1543
1544
1545
1546
1541
155i:)
1551
1552
1553
1554
1555
1556
1551
1560
1561
1562
1563
1564
1565

1044
1510
1061
3044
1366
1430
1361
4114
3045
1044
2045
1310

1566
1561
1510
1511
1512
1513
1514

1113
0002
1634
0144
1166
13012
1352

TAD EXPONT
SPA
CMA lAC CML
DCA EXPONT
TAD C253
SZL
TAD C255
Jt1S I DGPT
DCA HORDER
TAD EXPONT
ISZ HORDER
TAD 1'1144
SMA
Jt1P .-3
TAD C144
DCA EXPOtn
CMA
TAD HORDER
SZA
JMS I DGPT
DCA HORDEH
TAD EXPONT
ISZ HORDER
TAD M12
SMA
JMP .-3
TAD C12
DCA LORDER
CLA CMA
TAD HORDER
JI1S I DGPT
TAD LORDER
JMS I DGPT
Jl1P I FEXC

15~0

5336
1311
31044
1040
1045
1440
4114
3 III 45
1044
2045
1312
150~

5352
1313
3046
12413
li045

4114
1046
4114
5122
C253,
C255,
1'1144,
C144,
1'112,
C12,
DGPT,

0253-2611l
255-253
1634
0144
1166
0012
OUTJG

4-21

SEXP
CARRTN
CrlAH
CHE
C. 1t6
C12
C144
~53
~55
~60
~1

C3111
DEGON

DECONV
DGPT
DIG1 T
Dl VTWO
DNUMSH
DPCSPT
DPCV?T
DPN
OPT
DSWI T
J.)UoLAD
ENDFI
EXCt:S~

EXPONT
Ft:XC
FEXP?T
FllOl
FG02
FG03
FG04
Fl105A
FG06
FG06A
FG07
FlGOI
FlG02
FlG03
FlG04
FLINTP
FLOUTP
FOUR
FOUTCN
FPACI
FPNT

7.324
7341
10"'57
7.343
1312
1513
7571
7566
7567
7356
1511
7515
7"'11
7kH:l12J
7574
7063
7357
7066
7517
7516
7521
7334
0060
71ft)3
7452
7471
0044
7522
7337
7242
7246
1263
7210
7274
13161
731c:l4
7311
7420
7422
7426

HORDER
HTEMP
INPUT
LFED
LNFEED
LOfWEt<
LTEMP
~SK

MINCR
MINUS
(fll NUSE
M1 NUS1
£111 N9
('lRBOUT
NSllNPT
MSI GN
lYiSNPT
MULTl ~
MULT2
MildPT
£1112
£11144
M2PT
OPUT
OUT
OUTDG
OUTPUT
PC. 1'"
PiR
PERIO!)
PLUS
PLUS12
PRCHAR
PRINT
PRSW
REMAIN
RESTRT
SEXP
SIGN
SfHNUS
SNPT
SPLUS
SW1 Tl
SW1T2
TEN
TENPT

741~

7400
720'"
7326
7217
0052
0007

4-22

rd045
7062
1142
6771
1342
0046
1061
1141
1111
1135
151rd
7325
7137
7170
7520
71210
7332
7041
7067
7335
7572
7570
7336
6776
7344
7352
7166
7507
7512
7331
1136
7140
6767
7165
7513
7064
7167
7514
7065
7330
7333
7327
rd055
0056
7504
7340

EXTENDED FUNCTIONS
IFLOATING POINT PACKAGE
GETSGN:TAD 45
CLFPNT:JMS 1 7
IFLOATING POINT EXPONENTIAL

51.122
5023
5"'24
5025
5026
5027
503(21
5031
5032
5033
5034
5035
5036
5037
511)40
5041
5042
51643
5044
5045
5046
5041
5050
51.151
5052
5053
511154
5055
51656

0(21(21(21
1045
1110
4310
3304
4401
3213
6315
0(210(21
411116
1(2145
3305
4701
4407
6320
5315
2320
6315
3315
6320
1210
6323
5265
4323
2315
1251
6323
5262
3320
1323
6323
5315
4323
3301
1216
001210
1305
1044
3044
2304
5600
4401
6315
5216
4315
00111(21
5600

5051
5060
5061

0004
2312
14(212

5000
5(2101
5002
5003
5004
5005
501116
50(211
5~10

5011
5i(J 12
5(2113
511H4
5015
511116
511117
502111
5~21

*500111
FEXP,

0
GETSGN
SPA CLA
JMS FNEG
DCA SIGN
CLFPNT
FMPY LG2E
FPUT X
FEXT
JMS 1 FIXIT
GETSGN
DCA FLAG2
JMS 1 FL.OAT
CL.FPNT
FPUT XSQR
FGET X
FSUB XSQR
FPUT X
F:1PY X
FPUT XSQR
FADD D
FPUT TEMP
FGET C
FDIV TEMP
FSUB X
FADD A
FPUT TEMP
FGET B
FMPY XSQR
FADD TEMP
FPUT TEMP
FGET X
FDIV TEMP
FMPY TWO
FADD ONE
FEXT
TAD FLAG2
TAD 44
DCA 44
ISZ SIGN
JI~P 1 FEXP
CLFPNT
FPUT X
FGET ONE
FDIV X
FEXT
JMP 1 FEXP
ICONSTANTS FOR FEXP

IC(SIGN):-1 IF X<0

0004
2372
1402
4-23

5062
5063
5064
5065
5066
5067
50713
5071
5072
5ill 73
5[374
5075
5076
51377
5100
5101
5102
51d3
5104

21300
"".,013
'''''102
2000
1010100
0000

7774
2157
5157
0012
5454
0343
0"'1217
2566
5341
0001
2705
2435
0kHH

51~5

00100

5106
5107

4757
5563

5110
5111
5112
5113
5114

kH.l00
4714
7240
5710
61000

5115
5116
5117
5120
5121
5122
5123
5124
5125
5126
5127
5130

flH'00

5131
5132
5133
5134
5135
5136
5137
5140
5141
5142
5143
5144
5145
5146
5147
5150
5151
5152
5153

01000

0000
00~0

01000

O000
0000
0000
012100
00ic10
{,1000
O000
4407
5315
3315
6320
3373
13710

3320
1365
6323
5362
3320
1357
3320
1354
3315
4323
010100
5753
5226

B,
C,
D,
LG2E,
ONE,
TWO,

7774
2157
5157
0012
5454
0343
0007
2566
5341
O001
2705
2435
~H"'H

2000
0000
00132
213100
0000

SIGN,
121
fLAG2,
0
fI Xl T,
fIX
fLOAT,
fLOA
INEGATlON SUBROUTINE
fNEG,
0
JMS I ACMINS
CLA CMA
Jl1P I fN£G
ACMINS,
61000

ICALL SUBROUTINE IN
IINTERPRET£R
IPOINTER TO INTERPRETER

ITEMPORARY STORAGE
X,
0
0
0
XSQR,
0
0

TEMP,

ZER,

0
0
0

IMAIN ALGORITHM FOR ARCTANG£NT
ARCALG,
CLfPNT
fGET X
HIPY X
fPUT XSQR
fMPY BET2
fADD BETl
fMPY XSQR
fADD BETZ
fPUT TEMP
fGET ALF2
fMPY XSQR
fADD ALfl
H1PY XSQR
fADD ALfZ
fMPY X
fDIV TEMP
fEXT
JMP I .+1
ARCRTN
4-24

5154
5155
5156
5157
5160
5161
5162
5163
5164
5165
5166
5167
5170
5111
5112
5113
5174
5175

0000
2437
1643
7771
3304
4434
1713
3306
5454
0000
2437
1646
000(2)
2421
2323
1715
3427
7052

5200
52(2)1
5202
5203
5204
52(2)5
5206
5201
5210
5211
5212
5213
5214
5215
5216
5217
5221
5222
5223
5224
5225
5226
5227
5230
5231
5232
5233
5234
5235
5236
5237
5240
5241

(2)(2)(2)0
1045
7710
4642
3243
4407
6644
5644
2645
0000
1045
7710
5223
44(2)7
5645
4644
6644
0(2)00
7240
3251
5625
5131
2251
5235
441/)7
6644
5246
2644
0000
2243
5600
4642
7200
5600

5242
5243
5244
5245

511(2)
000(2)
5115
5(2)76

522~

ICONSTANTS - FLOATING ARC TANGENT
0000
ALFZ,
2437
1643
1117
ALF1,
3304
4434
1113
ALF2,
3306
5454
0000
BETZ,
2431
1646
000(2)
BETl ,
2421
2323
BET2,
1775
3427
7052
IFLOATING POINT ARC TANGENT
*5200
0
ARTN,
GETSGN
SPA Cl.A
JMS I NEGIT
DCA FLAG1
CLFPNT
FPUT I Xl
FGET I Xl
FSUB I CONI
FEXT
GETSGN
SPA CLA
JMP GO
Cl.FPNT
FGET I CONI
FDIV I Xl
FPUT I XI
FEXT
Cl.A CMA
DCA Fl.OG
GO,
IFl.AG
JMP I .+1
ICAl.l. Al.GORITHM
ARCALG
ARCRTN,
ISZ FLOG
IRETURN HERE
JMP EXIT
Cl.FPNT
FPUT I Xl
FGET PlOT
FSUB I Xl
FEXT
ISZ Fl.AGl
EXIT,
JMP I ARTN
JMS I NEGIT
CLA
JMP I ARTN
ICONSTANTS FOR ARCTANGENT
NEGIT,
FLAG1,
Xl,
CONI,

FNEG
0
X
ONE
4-25

5246
5247
5250

0001
3110
3755

0001
3110
3755
IFLOATING LOGARITHM

5251
5252
5253
5254
5255
5256
5257
5260
5261
5262
5263
5264
5265
5266
5267
52713
5271
5272
5273
5274
5275
5276
5277
53"'1.1
5,)01
5,)k12
53k:i3
5,)04
5,)10 5
5306
5307
531'"
5311
5312
5313
5') 14
5315
5316
5317
5320
5321
5')22
5323
5324
5325
5326
5327
5331.1
5331
5332
5333
5334
5335
5336
5337
5340

0000
1045
7440
5261
4407
4775
0000
5651
77H'J
4642
4407
6774
2645
0000
1045
7450
5350
7710
5341
3243
71340
1774
31.145
4777
4407
,)776
6644

FLOG,

IJk>k'l0

7001
3774
4407
5774
2645
6774
3773
1772
3774
1771
3774
1770
3774
1365
3774
1362
3774
1357
3774
1354
3774
1644
icHH'Ht'l

2243
5651
4642
720~

5651

PlOT,

OK,

START,

0
GETSGN
SZA
JMP OK
CLFPNT
FDIV I ZERO
FEXT
JMP I FLOG
SPA CLA
JMS I NEGIT
CLFPNT
FPUT I TEM
FSUB I CONI
FEXT
GETSGN
SNA
JMP ZERGO
SPA CLA
JMP INVERT
DCA FLAGI
CMA
TAD ITEM
DCA 45
JMS I FLOATP
CLFPNT
FMPY 1 LOG2
FPUT 1 Xl
FEXT
lAC
DCA ITEM
CLFPNT
FGET I TEM
FSUB 1 CONI
FPUT 1 TEM
FMPY 1 L8
FADD I L7
FMPY I TEM
FADD 1 L6
FMPY 1 TEM
FADD I L5
FMPY 1 TEM
FADD L4
FMPY 1 TEM
FADD L3
FMPY 1 TEM
FADD L2
FMPY 1 TEM
FADD Ll
FMPY ITEM
FADD 1 Xl
FEXT
lSZ FLAGI
JMP 1 FLOG
JMS 1 NEGlT
CLA
Jl1P 1 FLOG

IINDICATE ERROR
IDIVIDE BY ZERO

4-26

5.341
5.342
5.34.3
5.344
5345
5.346
5.341
5.350
5.351
5.352
5.35.3
5.354
5.355
5.356
5351
5360
5.361
5362
5363
5.364
5.365
5.366
5.367
5.3 70
5.3 71
5.372
5.3 7.3
5374
5375
5376
5377

4401
5645
4114
6114
000121
1240
5214
4401
5115
001210
5651
01210121
.3111
1142
1111
400121
4100
7777
2517
121.31217
1776
411.3
1211
5547
5552
5555
5560
512.3
5126
5544
556.3

540121
5401
541212
54".3
5404
541215
5406
5407
5410
5411
5412
541.3
5414
5415
5416
5417
542121
5421
5422
542.3
5424
5425
5426
5421
54.30
54.31
5432
54.3.3
5434

01211210
1045
774121
521"
1045
77121121
561210
4741
.3.34.3
4407
4.315
6124
12112113121
4742
4.36.3
441211
672.3
5724
272.3
.3.315
672.3
2.32121
121000
1045
1710

5241
4407
612.3
01211210

INVERT,

CLFPNT
FGET I CONI
FDIV ITEM
FPUT ITEM
FEXT
CLA CMA
JMP START
ZERGO,
CLFPNT
FGET I ZERO
FEXT
JMP I FLOG
121000
Ll,
.3111
1142
1177
L2,
41210121
41121121
L.3 ,
7777
2517
121.307
7776
L4,
411.3
7211
L5,
LOG5
LOG6
L6,
L7,
LOG7
LOG8
L8,
TEM,
TEMP
ZERO,
ZER
LOG2,
LOGE2
FLOATP,
FLOA
IFLOATlNG POINT SINE
*541210
FSIN,
GETSGN
SMA SZA CLA
JMP MOD
GETSGN
SMA CLA
JMP I FSIN
JMS I NEGT
MOD,
DCA PNTR
CLFPNT
FDIV TWOPI
FPUT I XSQ2
FEXT
JMS 1 FlXR
JMS FLOA
CLFPNT
FPUT I X2
FGET I XSQ2
FSUB I X2
FMPY TWOPI
FPUT I X2
FSUB PI
FEXT
GETSGN
SPA CLA
JHP PCHECK
CLFPNT
FPUT I X2
FEXT

IX>"?

IYES
X=9?
IYES SIN(0)=1il
INO:SIN(-X):-SIN(X)
IREDUCE X MODULO 2 PI
INO

IX<PI?

IYES
INa SIN(X-PI)=-SIN(X)
4-27

5435
5436
5431
5440
5441
5442
5443
5444
5445
5446
5441
5450
5451
5452
5453
5454

1343
1653
1040
3343
4401
5123
2114
1045
111B
5255
4401
5320
2123
6123
01111110

5455
5456
5451
5460
5461
5462
5463
5464
5465
5466
5461
541111
5411
5412
5413
5414
5415
5416
54'11
5500
551111
5502

4401
5123
4114
6123
3123
6124
5325
3124
1330
3124
1333
3124
1336
3124
1114
3123
1111110111
2343
5600
4141
120111
5600

PAl..G,

5503
5504
551115
5506
5501
5510
5511
5512
5513

0000
4401
6123
5114
2123
0000
1303
3200
5201

FCOS,

PCHECK,

""""

TAD PNTR
SNA Cl..A
CMA
DCA PNTR
Cl..FPNT
FGET I X2
FSUB I PI2
FEXT
GETSGN
SPA Cl..A
JMP PAl..G
Cl..FPNT
FGET PI
FSUB I X2
FPUT I X2
FEXT

IX<PI/2?

IYES
INO

ISIN(X):SIN(PI-X)

CI..FPNT
FGET I X2
FDIV I PI2
FPUT I X2
FMPY I X2
FPUT I XSQ2
FGET C9
FMPY I XSQ2
FADD C1
FMPY I XSQ2
FADD C5
FMPY I XSQ2
FADD C3
FMPY I XSQ2
FADD I PI2
FMPY I X2
FEXT
ISZ PNTR
JMP I FSIN
JMS 1 NEGT
CI..A
JMP I FSIN
IFl..OATlNG POINT COSINE
0
CI..FPNT
FPUT I X2
FGET I P12
FSUB I X2
FEXT
TAD FCOS
DCA FSIN
JMP FSIN+l

ICOS(X):SlNCPI/2-X)

ICONSTANTS AND POINTERS
5514
5515
5516
5511
552111
5521
5522
5523
5524

5246
0003
3110
3155
0002
311111
3155
5115
5120

PI2,
TWOPI,
PI,
X2,
XSQ2,

PlOT
0003
3110
3155
0002
3110
3155
X
XSQR

IPI/2

4-28

5525
5526
5527
5530
5531
5532
5533
5534
5535
5536
5537
5540
5541
5542
5543

7764
2366
5735
7771
5466
6317
7775
2431
5053
0000
5325
0420
5110
4757
0000

ISINE CONSTANTS
7764
C9,
2366
5735
C7,
7771
5466
6317
C5,
7775
2431
5053
C3,
0000
5325
0420
NEGT,
FNEG
FIXR,
FIX
PNTR,
L:1
ILOGARITHM CONSTANTS

5544
5545
5546
5547
5550
5551
5552
5553
5554
5555
5556
5557
55610
5561
5562

L:1001O
2613
4414
7776
2535
331211
7775
4746

5563
5564
5565
5566
5567
5570
5571
5572
5573
5574

0000
7300
3046
1374
3044
4407
70130
0000
5763
0013

4757
4760
4761
4762
4763
4764
4765
4766
4767
477'.1
4771
4772
4773
4774

010010
1044
7540
5365

10771

7774
2236
4304
7771
4544
1735

72100

5375
1377
3044
1044
7700
5757
4774
5367
621013

LOGE2 ,
LOG5,

0000
2613
4414
7776
2535
33~1

LOG6,

7775
4746
10771
LOG7,
7774
2236
43"'4
LOGS,
7771
4544
1735
I FLOAT C(45)
FLOA,
Id
CLA CLL
DCA 46
TAD C13
DCA 44
CLFPNT
FNOR
FEXT
Jl'1P I FLOA
C13,
0013
*4757
IFIX C(FAC)
13
FIX,
TAD 44
SMA SZA
JMP .+3
CLA
JMP FIXEND
TAD M13
DCA 44
LOOP,
TAD 44
SMA CLA
JMP I FIX
JMS I .+2
JMP LOOP
6200

IDIVl IN INTERPRETER
4-29

4775
4776
4777

3045
5757
7765

FIXEND,
M13,

A
ACI'liNS
ALFZ
ALFl
ALF2
ARCALG
AHCRTN
ARTN

511157
5114
5154
5157
5162
5131
5226
5200

5~62

BETZ
BET!
BET2
C
CLFPNT
CON!
C13
C3

C5
C7
C9
D

EXIT
FCOS
FEXP
FIX
FIXEND
FIXI T
FIXR
FLAGI
FLAG2
FLOA
FLOAT
Fl.OAT?
FLOG
FNiG
FSIN
GET3GN
GO
INVERT
LG2E
LOGE2

DCA 45
JMP I FIX
-13

LOG2
LOG5
LOG6
LOG7
LOGS
LOOP
Ll

12
L3
L4
L5

5165
51713
5173
5065
4407
5245
5574
5536
5533
5530
5525
50713
5235
5503
510100
4757
4775
5106
5542
5243
5105
5563

L7
L8

MOD
M13
NEGIT
NEGT
OK
ONE
PALG
PCriECK
PI
PLOT
PI2
PNTH
SIGN
STAliT
T£M
T£MP
TWO
TWOPI
X
XSQH
XSQ2
Xl

51~7

5,)77
5251
51110
54iOtc1
1fd45
5223
5341
5073
5544

ZEH
ZEHGO
ZERO

4-30

5376
5547
5552
5555
5560
4767
5354
5357
5362
5365
5370
5371
5372
5373
5410
4777
5242
5541
5261
5~76

5455
5441
5520
5246
5514
5543
5l1.14
5274
5374
5123
5HH
5515
5115
5120
5524
5244
5523
5126
5350
5375

IFLOATING OUTPUT PROGRAM
IIF(62):S, THEN OUTPUT IN E FORMAT
IELSE, C(62):NU"8ER OF DIGITS
ICCAC):NUMBER OF PLACES TO RIGHT OF •
IIF C(AC):~, THEN DON'T OUTPUT POINT
ISIGN AND • NOT CQUNTED IN OUTPUT
ICONTENTS OF 15 LOST DURING OPERATION
7201
7202
72143
72S4

4777
4217
1324
4776

7211)7

7200

7221
72314
1231
7232

3415
1000
70140
71it00

7301

3415

7305

3415

7376
7371

5412
5400

54140
5401
5402
5403
5404
5405
5406
5407
5410
5411
5412
5413
5414
5415
5416
5411
5420
5421
5422
5423
5424
5425
5426
5427
5430
5431
5432

0000
3352
1062
7041
3347
1357
3351i
1363
3ftH5
561it0
a0""
31it44
1363
31415
1415
47'1
1062

*7201

*7207

*7227

*7301
*7305
*737'
*5414~

EDIT,

TGO,

76511

5327
2212
2212
1044
7510
530"
1352
7041
10'2

TRYAGN,

JMS 1 7377 IALTERATIONS FLOATING OUTPUT
JMS 7217
TAD 7324
JMS 1 7376
CLA

DCA 1 15
NOP
NOP
NaP
DCA 1 15
DCA 1 15
TGO
EDIT
0
DCA SAC
TAD '2
C"A lAC
DCA COUNT1
TAD "8
DCA COUNT2
TAD SAD1
DCA 15
JMP 1 EDIT
0
DCA 44
TAD SAD1
DCA 15
TAD 1 15
J"S 1 oun
TAD 62
SNA CLA
JMP EFORM
ISZ TGO
ISZ TGO
TAD 44
SPA
JMP "INS
TAD SAC
CMA .1AC
TAD '2
4-31

5433
5434
5435
5436
5431
5440
5441
5442
5443
5444

151/iJ
5266
145111
5245
1041
3351
1353
4321
2351
5241

5445
5446
5441
5450
5451
5452
5453
5454
5455

1044
1041
1450
5256
3351
4331
4321
2351
5252

G01,

TAl) 44
CMA lAC
SNA
JMP G02
DCA CNTR
JMS GET
JMS OUT
ISZ CNTR
JMP .-3

5456
5451
5460
5461
5462
5463
5464
5465
5466
5461
5410
5411
5472
5473

1352
1650
5263
1354
4161
4331
4321
52Q3
1352
1700
5214
1356
4321
5271

G02,

TAD SAC
SNA CLA
JMP .+3
TAD PERIOD
JMS I oun
JMS GET
JMS OUT
JMP .-2
TAD SAC
SMA CLA
JMP ERGO
TAD CHX
JMS OUT
JMP .-2

5414
5475
5476
5477

12451
1352
3352
5225

ERGO,

CLA CMA
TAD SAC
DCA SAC
JMP TRYAGN

5500
5501
5502
5503
5504
55d5
5506
5511)7
5510
5511
5512

12016
1062
7041
1352
745a
5313
3351
1353
4321
2351
5307

MINS,

CLA
TAD 62
CMA lAC
TAD SAC
SNA
JMP G03
DCA CNTR
TAD SPCE
JMS OUT
ISZ CNTR
JMP .-3

5513 1354
5514 4761
5515 4321
5516· 2044
5517 5315
552. 5263
5521 0000
5522 4162

G03,

TAD PERIOD
JMS I OUll
JMS OUT
ISZ 44
JMP .-2
JMp G02+5

SPA
JMP ERR
SNA
JMP GOI
CftA lAC
DCA CNTR
TAD SPCE
JMS OUT
ISZ CNTR
JMP .-3

ERR,

OUT,

JfoJS I OUT2
4-32

552j
5524
5525
5526

2341
5121
1355
5612

5521
55316
5531
5532
5533
5534
5535
5536

4162
1354
4161

5531
55416
5541
5542
554.3
5544
5545
5546
5541
55516
5551
5552
5553
5554
5555
555&
5551
5560
5561
5562
5563

lSZ COUNTl
JMJI lOUT
TAD CriE
JMP 1 TGO
EYORM,

JMS I OUT2
TAD JlERIOD
JMS I OUTl
TAD M7
DCA COUNTl
JMS GET
JMS OUT
JMP .-2

GET,

It)

136~

3347
4337
4321
5334
161616~

235'"
5345
12414
3350
5731
1415
5737
001t)d
00160

Id"'
'''
0000
77&0
"'25&
16305
0050
1770
71"11
7344
7.552
5563

bUI'YEij
Crii
CtiX
CNTH
COUNTl
COUNT2
EDIT
EYORM
EkGO
ERR
GET
G01
G02
G03
MlNS
M7
M8

our
oun

OUT2
PErUOD

;3AC
SADI
~PCE

IdO

T..,YAGN

COUNTl ,
COUNT2,
CNTR,
SAC,
SPCE,
PERIOD,
CHE,
CHX,
M8,
M7,
OUTl ,
OUT2,
SAD1,
iUl'iER,

ISZ COUNT2
JMP .+4
Cl..A CMA
DCA COUNT2
JMP I GET
TAD I 15
JlvtP 1 GET

'00"
0
240-2'0
25&
305
330-260
-10
-7
7344
7352
iUYYER-l

5564
5555
5556
5551
5541
5550
5400
5521
5474
5466
5537
5445
545'
5513
55160
556/d
5551
5521
55"
5562
555'5552
5563
5553
5412
5425
4-33

mamaama

EQUIPMENT
CORPORATION
MAYNA~D.MASSACHUSETTS

PRINTED IN U.S.A.