On Mon, 6 Aug 2001, Gunther Schadow wrote:
Re-implementations of PDPs, VAXen or other computers are one thing,
also new IO modules for old computers are another
interesting use are
conceivable.
That's right. For each part, there will come a time when there are no
more original, functioning instances. Our ability to continue the
tradition will then lie in our ability to put something else in their
place.
Right, Jeff.
And this was a conversation I meant to finish with Tony Duell. I made
a remark about the speed I was seeing in a particular simulator of old
iron as run on a 233MHz Pentium-1 PC. Tony remarked that he didn't have
a machine of the host class (the Pentium-1 PC), so the simulator was
therefore not available to him, but that it didn't matter much to him
because he prefers the *real* original iron anyway.
And that's fine... I too, would love to own a Cyber 172, but even if I
ever find one (doubtfull), I wouldn't be able to afford to turn it on
even one time. So I wouldn't be able to enjoy it as a *computer*, but
rather as an *object* (d'art or whatever).
I suppose for someone who didn't have a history with a particular
machine, that is, an old beast like a CDC 6600 or similar, that being
able to recreate the *software* environment by having a simulator just
doesn't mean much. But for mean, it means everything. The operating
system and the language translators and utilities were my points of
familiarity as a programmer; again, had I been an engineer on one of
these systems, having something I could bootstrap might be less important.
Now, the FPGA route is even better than a simulator, of course, because
it's faster. Also, getting real old peripherals to work with a re-
implemented CPU is also promising- in many cases, the FPGA board could
be set in open space in old iron and the system be made to appear
operational even if not!
Oops, time for more coffee...
-dq