mjt@super.ORG (Michael J. Tighe) (04/08/89)
Does anyone have any information on the CDC 8600? I saw a picture of it recently. About all I know is it was built around 1973, was an N-gon type shape (about the same size as a Cray-3, but with more sides). And the modules were quite large (approximately 3" x 8" x 5"). I also heard it didn't work. That's why it was never put into production. Any info on architecture, instruction set, chip type, clock times or memory would be nice. -- ------------- Michael Tighe internet: mjt@super.org uunet: ...!uunet!super!mjt
jps@wucs1.wustl.edu (James Sterbenz) (04/11/89)
In article <7948@super.ORG> mjt@super.UUCP (Michael J. Tighe) writes: >Does anyone have any information on the CDC 8600? I saw a picture of >it recently. About all I know is it was built around 1973, was an >N-gon type shape (about the same size as a Cray-3, but with more sides). >And the modules were quite large (approximately 3" x 8" x 5"). I thought that the 8600 was the name for Cray's 7600 successor machine, while he was still at CDC. This was in compitition with the STAR-100, which Norris decided was the way CDC wanted to go. Cray took the 8600 (with CDC's blessing and support) to become the Cray-1. This would be about the right time, but was there any working hardware before Cray broke off? Or is it possible that CDC considered calling the STAR the 8600 (but doesn't sound like the right foot-print)? -- James Sterbenz Computer and Communications Research Center Washington University in St. Louis 314-726-4203 INTERNET: jps@wucs1.wustl.edu UUCP: wucs1!jps@uunet.uu.net
seanf@sco.COM (Sean Fagan) (04/11/89)
In article <7948@super.ORG> mjt@super.UUCP (Michael J. Tighe) writes: >Does anyone have any information on the CDC 8600? I saw a picture of >it recently. About all I know is it was built around 1973, was an >N-gon type shape (about the same size as a Cray-3, but with more sides). >And the modules were quite large (approximately 3" x 8" x 5"). Dredging through my memory, I come across a few third+-hand stories about Seymour (Cray, that is), CDC, and the future of computing. The stories I heard had Seymour Cray working on a successor the the 7600, which was a successor to the 6600, which was to be called the 8600. Supposedly, this marvelous (being designed by Seymour, could it be anything else?) device would have a neat new idea, called "vectors," but would, otherwise, resemble the 7600 (except for being 64-bits instead of 60). At the same time, CDC had a project called the "Star," which also had vectors, but of a different sort. The Star's vectors were memory-to-memory, and it had virtual memory (something that Seymour either doesn't believe in or doesn't understand 8-)). After both groups spent gobs of money, Seymour was told that the Star would get their full attention now, not the 8600. Seymour then left, worked as a consultant for IBM, and founded Cray Research, Inc. The rest is history, of course... >I also heard it didn't work. That's why it was never put into >production. Any info on architecture, instruction set, chip type, >clock times or memory would be nice. "Chip"? Surely you jest... The Star, however, never worked properly. Evenutally, a successor called the 205 was built, and that worked properly (it was fast, but still slower than the Cray-1). Eventually, another company split off from CDC, called ETA Systems, and built a *very* fast 205 clone, called the ETA-10. I think the machine you saw a picture of was probably a Star, not an 8600... -- Sean Eric Fagan | "An acid is like a woman: a good one will eat seanf@sco.UUCP | through your pants." -- Mel Gibson, Saturday Night Live (408) 458-1422 | Any opinions expressed are my own, not my employers'.
rpeglar@como.unix.eta.com (Rob Peglar) (04/11/89)
In article <791@wucs1.wustl.edu> jps@wucs1.UUCP (James Sterbenz) writes: >In article <7948@super.ORG> mjt@super.UUCP (Michael J. Tighe) writes: >>Does anyone have any information on the CDC 8600? I saw a picture of >>it recently. About all I know is it was built around 1973, was an >>N-gon type shape (about the same size as a Cray-3, but with more sides). >>And the modules were quite large (approximately 3" x 8" x 5"). > >I thought that the 8600 was the name for Cray's 7600 successor machine, >while he was still at CDC. This was in compitition with the STAR-100, >which Norris decided was the way CDC wanted to go. Cray took the 8600 >(with CDC's blessing and support) to become the Cray-1. This would >be about the right time, but was there any working hardware before >Cray broke off? The 8600 was indeed the intended successor to the 7000 series machines. Cray did indeed take the ideas and designs of the 8600 with him, to eventually become the Cray-1. If you are in doubt, take a look at the 7000 series instruction set and the Cray-1 instruction set. To say that "Norris decided..." is a bit overstated. There were three distinct camps within CDC in those days; the 7000 series people, with Cray, Davis, etc.; the STAR-100 people, led by Lincoln and Thornton; and the 180 series people, some in Minnesota, some in Canada. The feud between the 7000 and the 180 people was intense, and eventually led to CRI and what we now know as the CDC Cyber 800 series, both in 170/800 and 180/800/900 models. The STAR people were (kind of) background participants in those days, trying to sell their ideas on vector computation to anyone who would listen. As CRI formed in 1972, CDC chose the STAR path almost by default for vector machines. As you know, the STAR-100 series begat the Cyber 203 begat the 205 begat the ETA-10, although the ETA-10 is a far cry from the 205. In relative terms, the STAR-100 and 205 are closer cousins than the 205 and ETA-10, architecturally speaking. > >Or is it possible that CDC considered calling the STAR the 8600 >(but doesn't sound like the right foot-print)? > Nope. Rob Rob Peglar internet: rpeglar@woods.unix.eta.com ETA Systems, Inc. ETC10C uucp: {amdahl,rutgers}!bungia!eta!rpeglar 1450 Energy Park Drive fax: (612) 642-3448 St. Paul, MN 55108 voice: (612) 642-8386
rvk@twitch.UUCP ( Bob Kline) (04/11/89)
I seem to remember that the 8600 was supposed to be a cluster of four 7600's sharing a memory. At the time (about 1973), Cray was said to already be working on a 9600. Shortly after this he left CDC. I never heard that any 8600's were shipped; rumor had it that the plans for the 9600 went with Cray. maybe someone else can take this up to a 10600....
mjt@super.ORG (Michael J. Tighe) (04/19/89)
Thanks to all of those that sent me mail about the CDC 8600. I
received plenty of responses, and was eventually pointed in the
direction of someone who actually worked on the machine with Seymour
Cray. So to sum up, here is what I have:
The CDC 8600 was a follow on to the 7600, but not like the 7600 was a
follow on to the 6600. The 8600 was different in several ways. It had
4 CPU's running at 8 nanoseconds with 250 Kwords of memory (20 bit
address space). Two types of memory were designed. Core memory with
about 20 nanoseconds access time. Semiconductor memory with about 22
nanoseconds access time.
The modules had discrete transistors, 18 boards/module, 4 layer
boards. 13 modules/CPU. Approximate speed was 2.5 x 7600 speed/CPU.
Total speed was about 10 x 7600 speed.
It also had a 64 bit word, not 60, which was another departure from
the 6600/7600. It used ones complement arithmetic with 2 sign bits, 14
bit exponents and a 48 bit coefficient.
There were no vector registers. It had 16 general purpose registers.
Operation speed: (in clock periods)
Boolean 2
shift 3
long add 3
multiply 8
floating add 8
read 15
write 1
branch in stack 7
fall through 3
out of stack 15
The 8600 project was terminated due to financial problems in CDC.
Also, the Cray-1 was not an 8600. It did evolve from the 8600 though.
It also had ideas taken from the Star 100, such as vector registers.
(this was stated by Seymour at Supercomputing 88)
--
-------------
Michael Tighe
internet: mjt@super.org
uunet: ...!uunet!super!mjtrchrd@well.UUCP (Richard Friedman) (04/23/89)
In article <791@wucs1.wustl.edu> jps@wucs1.UUCP (James Sterbenz) writes: >In article <7948@super.ORG> mjt@super.UUCP (Michael J. Tighe) writes: >>Does anyone have any information on the CDC 8600? I saw a picture of >>it recently. About all I know is it was built around 1973 > >I thought that the 8600 was the name for Cray's 7600 successor machine, >while he was still at CDC. This was in compitition with the STAR-100, >which Norris decided was the way CDC wanted to go. Cray took the 8600 >(with CDC's blessing and support) to become the Cray-1. This would >be about the right time, but was there any working hardware before >Cray broke off? > >James Sterbenz Computer and Communications Research Center I actually saw a mock up of the 8600 in CDC's Chippewa Falls lab in '72 (or was it '74?). Then Cray was at CDC Chippewa lab and the 8600 was indeed to be the follow on of the 7600, with vector instructions (somewhere in my archives is a pre-release of the 8600 instruction set) The lab engineer who showed it to me said that CDC was going to have to hire a lot of very short engineers with very long narrow arms to service it. It was about waist high and looked alot like a miniature Cray-1. When I saw my first Cray-1 4 years later I was surprised that it was just like the 8600, but it had grown up. Those were heady days! Gawd we're getting old. (rip cdc/eta). -- ...Richard Friedman [rchrd] Berkeley, CA. rchrd@well.uucp -or- {ucbvax,lll-lcc,pacbell,hplabs}!well!rchrd