Table of Contents
Advertisement
A 20-bit word can accommodate a range of decimal values from -524,288 to +524,287, sometimes
r e f e r r e d to a s 5-1/2
decimal digits.
Double length word operations permit the efficient pro-
cessing of decimal values between +274,877,906,943 and -274,877,906,944.
The advantages
associated with representing numeric data in the true binary format a r e :
Memory storage efficiency.
An effective increase in the rate with which numeric data can be transferred to and f r o m
magnetic tape.
Increased speed of arithmetic and data handling operations.
Binary and BCD information may be intermixed in memory s o a s to provide the most efficient
mode of representing each field in each application, Subroutines a r e provided to convert numeric
data from BCD to binary, o r vice-versa.
Decimal Arithmetic
An optional Decimal Arithmetic feature of the GE-235 provides a decimal add and subtract
capability without requiring BCD to binary conversion.
These arithmetic operations can be
performed on single decimal words of 3 digits o r on double length decimal words of 6 digits.
Automatic c a r r y i s provided f o r l a r g e r fields.
The examples below illustrate how the decimal
numbers +368 and +I5896 would appear in memory.
Floating Point Arithmetic
The GE-235 Auxiljary ~ r i t i m e t i c Unit can be used to advantage in scientific and engineering
applications where numerous floating point o r double precision arithmetic calculations a r e r e -
quired.
The logic of the AAU performs double precision fixed point and floating point arithmetic
m o r e efficiently than i s possible when using mathematical subroutines. However, subroutines
may be preferable when floating point arithmetic i s done on a limited basis.
Three modes of calculation may be performed by the Auxiliary Arithmetic Unit: unnormalized
floating point, normalized floating point, and fixed point double precision operations. Addition,
subtraction, multiplication, and division may be done under any of the three modes of operation.
All arithmetic i s performed in binary mode.
Advertisement
Table of Contents
