From owner-freebsd-chat Fri Jul 9 6:20:50 1999 Delivered-To: freebsd-chat@freebsd.org Received: from gate2.consol.de (gate2.consol.de [194.221.87.12]) by hub.freebsd.org (Postfix) with ESMTP id 5CBF514C87 for ; Fri, 9 Jul 1999 06:20:34 -0700 (PDT) (envelope-from Michael.Elbel@consol.de) X-Envelope-Sender-Is: Michael.Elbel@consol.de (at relayer gate2.consol.de) Received: from msgsrv.bb.consol.de (root@msgsrv.bb.consol.de [10.250.0.100]) by gate2.consol.de (8.9.1a/8.9.1) with ESMTP id PAA19097; Fri, 9 Jul 1999 15:20:22 +0200 (CEST) Received: from fourier.int.consol.de (fourier.int.consol.de [10.0.1.17]) by msgsrv.bb.consol.de (8.8.8/8.8.8) with ESMTP id PAA08093; Fri, 9 Jul 1999 15:19:42 +0200 Received: (from me@localhost) by fourier.int.consol.de (8.9.3/8.8.7) id PAA14454; Fri, 9 Jul 1999 15:19:41 +0200 (CEST) (envelope-from me) Date: Fri, 9 Jul 1999 15:19:41 +0200 From: Michael Elbel To: brett@lariat.org Cc: chat@freebsd.org Subject: Re: IA64 Message-ID: <19990709151940.A13563@consol.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.5i Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In lists.freebsd.chat you write: >At 05:52 PM 7/8/99 +0300, Nadav Eiron wrote: >Funny how Intel rewrites history to avoid antitrust problems. ;-) Fact is, >they wanted the i860 to be used on graphics coprocessor boards but NOT >as the system's main CPU. I may still have the handouts which said so. >At the presentation, I asked an Intel rep whether the i860 could be used >as a general purpose CPU for a workstation; he replied that Intel WOULD >NOT SELL the CPU to a company that wanted to use it for that purpose. Well, all I can say is that in a former life at a former company we were indeed building workstations with i860 processors and only those. Intel *themselves* supplied a SYSVR4 port to the i860 that we used as the base of our own version. Deficiencies in the processor design aside (context switches e.g. were horrendously expensive due to the awkward way you had to save the processor state), they ok for their time, at least on floating point. Porting a X server to the thing was also not much fun what with the difficulties we had mapping a frame buffer uncached. I don't know if they actually produced a somewhat bug free processor stepping before discontinuing them. I remember that we were pretty proud of outperforming almost everything else on mandelbrot calculations back in 1992 or so. This was before the HP snake and Alpha processors came along, of course. I believe that Gary Jennejohn still uses one of the beasts as an X terminal. So, yes, there may have been a Intel rep who claimed that the things should not be used as general purpose CPU. Fact is also that Intel indeed was selling them as such, even doing a UNIX port themselves. The left not knowing what the right is doing? Ask three people at a large company and get 5 different answers? Surely doesn't happen only at Intel. Michael -- \|/ -O- Michael Elbel, ConSol* GmbH, - me@consol.de - 089 / 45841-256 /|\ Fermentation fault (coors dumped) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message