Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!rutgers!cbmvax!jesup
From: jesup@cbmvax.UUCP (Randell Jesup)
Newsgroups: comp.arch
Subject: Re: Oh, not again! Re: DARPA supercomputer
Message-ID: <6657@cbmvax.UUCP>
Date: 21 Apr 89 02:06:28 GMT
References: <9951@polyslo.CalPoly.EDU> <6504@cbmvax.UUCP> <16715@cup.portal.com> <149@bashful.ACA.MCC.COM> <3210@eos.UUCP>
Reply-To: jesup@cbmvax.UUCP (Randell Jesup)
Distribution: usa
Organization: Commodore Technology, West Chester, PA
Lines: 20

In article <3210@eos.UUCP> eugene@eos.UUCP (Eugene Miya) writes:
>In article <149@bashful.ACA.MCC.COM> rsb@bashful.ACA.MCC.COM (Richard S. Brice) writes:
>>"Advanced Military Computing", a darpa director named Raymond Colladay is
>>quoted as claiming that DARPA plans to have a billion operation per second
>>supercomputer packaged in 10 cubic inches by next year.  I don't have
>>access to "Advanced Military Computing" so I can't say whether a
>>longer article appeared there.
>
>Ray Colladay is the acting DARPA head who used to sign my travel at NASA HQ.
>Are you sure it's not a cube "10 inches on a side?"  That is completely
>foreseeable.  SO WHAT?

	This matches what I remember from my GE R&D days (rpm-40), when the
government put out an RFP for 1000 cubic inch (10x10x10), 1 gigaflop
machine.  You can do it (theoretically and using peak flops) right now.
Practicly and sustained flops are another issue, as well as inter-processor
communication.

-- 
Randell Jesup, Commodore Engineering {uunet|rutgers|allegra}!cbmvax!jesup