On Fri, 2026-06-19 at 09:54 -0400, Paul Koning via cctalk wrote:
On Jun 18, 2026, at 11:33 PM, Van Snyder via
cctalk
<cctalk(a)classiccmp.org> wrote:
On Thu, 2026-06-18 at 16:28 -0400, Paul Koning via cctalk wrote:
I just now had a conversation with Google (not Google Gemini, just
the
search engine AI). I mentioned that I had written an IBM 1130
emulator
in Varian V-73 microcode. It said "You must be Paul Koning." Did
you
wrote one too?
That's a new kind of hallucination!
Nope, I never wrote anything for any Varian machine. I did spend
time in university working on a PDP-11 that sat next to a Varian
machine, a 16 bit machine with user programmable microcode. Is that
the V-73? I remember it had a flat console lights panel -- the
switches were membrane switches so the whole front was flat.
V73 fit in a standard 19" rack, about six inches high. Flat panel. Not
membrane switches, IIRC. Made by Varian Data Machines, not DEC. Varian
sold the Data Machines division to Sperry Univac, who produced the V77
but couldn't actually run with it. Then they pounded Data Machines into
the ground.
I had a handbook for that machine but lost it. But
that's as close
as I got. And besides, I never worked with an 1130, though I did do
some elementary stuff on a 1620.
In the late 1960s one of my colleagues needed to process some telemetry
data. The client couldn't afford time on the IBM 7094 so he wrote the
code for the 1620. To do octal arithmetic he changed the tables. He
exploited the CADET — "Can't Add Doesn't Even Try" — nature of the
1620.