Path: utzoo!utgpu!water!watmath!clyde!bellcore!tness7!tness1!sugar!karl
From: karl@sugar.UUCP (Karl Lehenbauer)
Newsgroups: comp.arch
Subject: IBM Stretch
Summary: CDC 6600 had 25X better price/performance than the Stretch
Message-ID: <2308@sugar.UUCP>
Date: 14 Jul 88 12:34:20 GMT
References: <28200172@urbsdc> <11458@steinmetz.ge.com> <4480@medusa.cs.purdue.edu>
Organization: Sugar Land UNIX - Houston, TX
Lines: 24

At the Boston Computer Museum, there's a little "shrine" to Seymour Cray,
with a quote attributed to Mr. Watson that goes something like 'how can
this band of upstarts build a computer, with only 35 people, that runs
five times faster than our fastest machine and costs one fifth as much?'
There's a movie about it (at the museum), too.  Brigham Young University
had one and ran it for years and years.

In article <4480@medusa.cs.purdue.edu>, sxr@purdue.UUCP writes:
> There was no 360 model 97.  IBM did introduce a model 90 series of
> very high performance 360 systems, at least in part to help it compete
> with Control Data.  This was the basis of an anti-trust suit by CDC
> against IBM.  IBM built and delivered a few model 91 systems, and at
> least one model 95.  

Yeah, the 360/90 was the machine that IBM was more or less forced to build.
As I understand it, IBM didn't think there was a scientific supercomputer
market and was quite surprised when CDC received a bunch of orders for
the then-new 6600.  Consequently, IBM announced a vaporware product, a high 
speed scientific supercomputer, the 360/90, causing most 6600 orders to be 
cancelled, prompting CDC's antitrust suit against IBM.  IBM settled out of
court with CDC and did finally build the 90, but it was years late and did
not achieve its original stated performance.
-- 
-- karl@sugar.uu.net aka uunet!sugar!karl