Re: building a PDP-8 computer in your kitchen

From: "Charles N. Choukalos" <chuckc@btv.ibm.com>
> All I can say is WOW...  That's some serious persiverance and patience
to
> do that kinda a work.

As I said, the guy whose project this was was a nutter.

> I think I would scream, decide I wasn't getting
> paid for this horse droppings and run back to the FPGA a happy man.

It was a labour of love. Which is to say, that after the nth batch with
0% yield we both said F*&^% this for a joke <g>
We did it to see if we could, basically. A hobby.

> I checked with some of the old time IBM process
> guys... And they were like "How did you think we made chips?"  and the
> famous line... "Chuck, you know that we used looms to make memory back
in
> the 60's don't you?"

And the UYK-7 computer, still to be found in many combat ships, still
uses
Core memory. It's very resistant to EMP, doesn't require a refresh (so
if
power is lost, you still have the memory as it was at the time), and
degrades gracefully - shoot a bullet through it, and only the bits
actually
damaged by the projectile are lost.

Of course, this means a CPU with all the power of a 286 weighs about a
tonne...
you'd be surprised at how primitive some of the "advanced" military
electronics is.
That's cause they keep it in service for decades.

> Thanks again Alan for explaining... I was foolishly envisioning
something
> far more simple...

Simple it ain't. But it *is* doable.

Guess I'm showing my age. My father was involved with computing wayyy
back in the early days, and my Uncle before that (as in University of
Manchester, Bletchley Park and all that). I can still remember when I
got
my first FORTRAN subroutine to run, at Harwell Atomic Research
establishment
in the mid 60s. I was 8 at the time. IIRC the system was a DEUCE, or was
it
a PEGASUS? I used its big brother ARGUS at the AIOTT in the 80s, when it
was being retired.