25 Nov
2010
25 Nov
'10
9:58 a.m.
On 11/25/2010 2:20 AM, jim s wrote:
the 800 was deemed not a good design to continue, and
the 1600 was a
redesign by Ken Omohundro. The 800's DTL was a big reason to do the
redesign, as well as other problem with it.
What is the history of your 800?
Jim
Because of the particular microcode, ours was called an 820. There
seemed to be some confusion at the time about that, as ISTR it was
sometimes referred to as an 822. It of course could become an 8xx where
xx is most anything just by plugging in a different microcode board.
Anyway, in the late 60's or early 70's Recognition Equipment, Inc. (REI)
in Dallas/Irving built a number of systems with the model designation
OCR/S-2000. The OCR/S meant Optical Character Reader/Sorter, and the
2000 was the maximum number of documents per minute that could be
processed. The 820 in the OCR/S-2000 was called I think the SSC meaning
SubSystem Controller, and it was interfaced to and controlled various
aspects of the document transport, control panel, etc. This system was
originally provided to the Swedish Post-Giro and was eventually used by
the Danish, Finnish, and Norwegian Post-Giros as well to optically read
and sort their various payment documents. I once knew how many of those
we built in all, but it has been too long now. I'm pretty sure it was
more than 20, which was a quite significant number for the company at
the time. These were not small systems physically! I'm trying to
remember how many sort pockets, and I'm sure at least 12, but I think
actually it was either 18 or 24 max.
After production stopped, some number of the 820's were surplussed and
acquired by one of the Field Service people who gave one to me. One of
the guys gave me several spare boards and other bits for it as well.
The last time I tried to fire one up, which was a lot of years ago, it
did not work and I never got around to finding out why. Now that I'm
semi-retired maybe I'll find the time to look into it again.
Before we did the OCR/S 2000 project some poor engineer did one with an
800. He actually wrote microcode to control some device. I say "poor"
because it was before the PROM microcode board was available, and he had
to solder/unsolder discrete diodes to "load" his code. And I thought
loading code from punched paper tape with an ASR-33 was slow!
The assembler for the 820 was written in Fortran IV, and I did quite a
lot of work on that. I used to think it was really cool to tell someone
I was going to go compile my assembler. :-)
Later,
Charlie C.