Subject: Re: these RTL or what?
From: "William Donzelli" <wdonzelli at gmail.com>
Date: Fri, 05 Oct 2007 13:55:28 -0400
To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at
classiccmp.org>
Using ultra-fast ECL doesn't make much sense
when you've got nanoseconds
of delay to the backplane, to the next board, and back to the part that
needs the signal.
That depends on how tight everything is. Thermal Conduction Module, anyone?
At 1ns/ft even the TCMs were not tight/dense enough.
The ECL
technology used in the VAX9000 was gate arrays with roughly the
same timing parameters as 100K ECL (0.5 to 1.0 ns propogation delays).
Yes, but I do not think that was the cutting edge anymore. Considering
the 9000 was supposed to be the machine that finally convinces the
mainframe world to accept DEC, it may have been a poor choice. We
probably will never know. 9000 may have been as big of an
embarrassment as the KC10.
Even though the 9000s were bombs, they are one of the few VAX machines
I would chase after.
;) the problem is the 9000 by time it got out the door CMOS system on a chip
was around the corner. With the ability to put transistors literally next to
each other it was easier to achieve overall speed. When you consider that
you could put multiple systems in less space than a 9000 well, quantity has
it's own special quality.
> Responsiveness of a computer system depends on a
lot more than the
> speed of the semiconductors used to build it. Plenty of modern examples
> of how to make fast silicon seem slow are coming out of Redmond I
> notice :-).
;) I was always amazed that a dozen users could be on an 11/34 but one
user could bring a 486 to it's knees.
I am thinking raw horsepower - all the benchmarking stuff. Looking at
the KL10 (or the other DEC ECL machines), it justs seems like they
should have been better number crunchers.
Crunching numbers was a part of the task. The other part was moving
and handeling data in mass storage and memory. Most of the DEC hardware
moved data pretty fast. What was the demise of PDP-10 was simple, megabytes.
A PDP-10 could not address the huge volumes of data in one chunk comming
from the more complex models and programs in use. VAX
offered a 32bit
address, PDP-10 was basicially 18bits with memory extension. If
your
munching a model or database of millions of elements that is as important
as the time to add two word size numbers. It's why PDP-11 was replaced
with VAX and why VAX was replaced with Alpha.
In the end ECL was a way to speed but always at such a high system cost
and complexity it was often behind the curve for integration and delivery.
Usually other technologies were close enough behind and had the higher
density or ease of integrationin to larger systems needed to offset the
per gate speed with sometimes better complex function speeds.
Allison
--
Will