[prev in list] [next in list] [prev in thread] [next in thread] 

List:       classiccmp
Subject:    Re: VAX9000 (was: RE: kda50 manual and sdi cables & questions)
From:       Paul Koning <pkoning () equallogic ! com>
Date:       2005-04-30 21:39:17
Message-ID: 17011.64261.549933.424951 () gargle ! gargle ! HOWL
[Download RAW message or body]

>>>>> "Allison" == Allison  <ajp166@bellatlantic.net> writes:

 Allison> One of the interesting aspects is the collision of
 Allison> technology at that point in time.  People that thought they
 Allison> needed a bigger meaner VAX got it with the 9000.  However in
 Allison> the same timeframe there was increased demand for smaller
 Allison> ligher VAXen and use more of them.  What was happening in
 Allison> the background was the speed/power product of CMOS was
 Allison> getting better and ECL (ECL100k) was was poor in the
 Allison> density/power product. 

True.  Although interestingly the reason that is true is because the
ECL chip builders never built anything other than SSI parts with ECL.

DEC Research in Palo Alto did a neat project to do VLSI ECL, which no
one had ever done.  It turns out that ECL is very efficient and
compact for many logic functions if you use the right circuits.  The
WRL team designed its own design tools because nothing existed to do
the job.  They actually did build a MIPS processor that way that ran
at a half GHz or so (back in 1992) and were going to do a GHz Alpha.
They almost did.  What killed the project is that all the ECL
foundries kept closing down.  The design tools actually allowed
switching to another fab with different design rules, but at some
point they ran out of fabs.

They also did some very slick work to handle 100 watt chips, something
that Intel later made routine (some years after DEC was ridiculed for
building 100 watt Alphas).

 Allison> So while the 9000 was the fastest
 Allison> vax.. for a while the later CMOS versions was very close and
 Allison> a smaller plus cheaper was it.  Add to that that a
 Allison> Multiprocessor VAX using chip technology was not only
 Allison> practical and well supported it offered a better bang for
 Allison> the buck. If that weren't enough the same ugly dragon that
 Allison> helped kill Jupiter (PDP-10) was back to haunt VAX.  At some
 Allison> point in the near future it was clear that 32bits no matter
 Allison> how fast was going to be inefficient and 64bits was a very
 Allison> viable answer, that begat Alpha.  Who'd have though in 1987
 Allison> that a gigabyte was not enough and terabytes were only a few
 Allison> years away.

Too bad that DEC canceled two perfectly good RISC architectures before
finally deciding to ship RISC in the guise of Alpha (and, for a while,
MIPS because they had canceled their two earlier internal efforts that
were better).

     paul

[prev in list] [next in list] [prev in thread] [next in thread]