Post by Brian W Spoor
I've just posted a copy of the ICT 1500 Programmers Reference Manual on
my website, a rebadged RCA 301 I believe (or was it the 501? - before my
time I will admit).
See:-
http://www.icl1900.co.uk/techpub/1500prm.pdf
Not strictly ICL1900, but part of the history of the company.
The ICT 1500 was indeed a rebadged RCA 301 (not 501). The same
machine was also rebadged by De La Rue Bull as their Gamma 30,
(French computer company - later absorbed into/as Honeywell-Bull)

The ICT variant of the machine read the more usual encoding of
80-column Hollerith cards, where the rows were called
10,11,0,1,2,3,4,5,6,7,8,9 - this in deference to British Sterling
currency where the upper two rows allowed for encoding of
pennies (12 pennies to a shilling) in a single column - that only
worked on the ICT tab equipment - not on the 1500.

The French Bull machine had a modification where the card-reader
circuitry was tweaked to correctly read their version of encoding
on an 80-column punched card. I can't recall the exact details of
how the French version differed, but I think they moved the top two
rows to the bottom of the card so that the rows represented
0,1,2,3,4,5,6,7,8,9,X,Y or something like that. (X, Y and 0 being the

more usual designators for the overpunch rows when encoding
Alpha strings)

The manual has a nice diagrammatic layout and explanation
of the operator's "console" - a wonderfull array of blinkenlights
some of which were also keys that could be pressed to enter values
into registers etc. A pictures of same can be seen in the second
image on the page at

http://home.znet.com/couperus/ICT1500.htm

This manual curiously enough gives not only the single-character
op-codes as used in actual machine code, but also their
mnemonic equivalents as used in the FAS language. (FAS =
Fifteenhundred ASsembler) but no mention is made of
the FAS language itself. This was a very simple assembler
that did little more thatn translate mnemonic codes to machine
codes and allowed the programmer to address memory via
named variables instead of hardware addresses. The concept
of macros or micro-substitution was unknown in that environment
and only became commonplace in the next generation 1900
machine with the PLAN assembler. The ability to pull in
run-time routines from a tape library and link them dynamically
via a loader also eventualy became available - until then
one took a source deck of the required routine and merely
added it to ones main program to be assembled in all as one
big program.

Jitze