From: pontius@btv.MBI.com.invalid (Dale Pontius) Newsgroups: alt.folklore.computers,comp.arch Subject: iAPX-432 (was: 36 to 32 bit transition Date: 31 Aug 2000 13:03:47 GMT Organization: IBM Global Services North -- Burlington, Vermont, USA Lines: 24 Message-ID: <8oll3j$ote$2@news.btv.ibm.com> References: <8o7ju4$jvf$1@murrow.corp.sgi.com> <8o8idl$b6r$1@lisa.gemair.com> <39A875A3.B548A192@ibm.net> <8ocfdo$ij6$1@xuxa.iecc.com> <8ocvhc$24c$1@murrow.corp.sgi.com> NNTP-Posting-Host: kimon.btv.ibm.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: news.btv.ibm.com 967727027 25518 9.66.91.77 (31 Aug 2000 13:03:47 GMT) X-Complaints-To: news@btv.ibm.com NNTP-Posting-Date: 31 Aug 2000 13:03:47 GMT X-Newsreader: knews 1.0b.0 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!fu-berlin.de!news.nikoma.de!tiscalinetde!bignews.mediaways.net!abq.news.ans.net!news.chips.ibm.com!newsfeed.btv.ibm.com!news.btv.ibm.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:62765 In article <8ocvhc$24c$1@murrow.corp.sgi.com>, mash@mash.engr.sgi.com (John R. Mashey) writes: ... > And of course, there have been large numbers of all-out machines > that either never worked, or worked but were hard to maintain, > or worked, but got caught too quickly by rising lower-cost systems. > Most interesting. When you say "all-out", you were obviously referring to performance. But it also makes me think of the old iAPX-432 that Intel tried to do so long ago, and the term 'performance' was never associated with that CPU. You draw an interesting line of heritage, and that basic topic comes up here from time to time. Again that brings me back to the 432, because as far as I can tell, it came from nowhere, and went right back there. It appears to have been an isolated event in computing, quite a rare thing. Did the 432 really have predecessors and successors? Dale Pontius NOT speaking for IBM ###### From: "Larry Elmore" Newsgroups: alt.folklore.computers,comp.arch References: <8ocvhc$24c$1@murrow.corp.sgi.com> <8oll3j$ote$2@news.btv.ibm.com> <3911C34B5AA2DE31.E1C9F3532D62BBF4.D6ED9205B201817A@lp.airnews.net> <8omg7e$1u2l$1@nntp1.ba.best.com> Subject: Re: iAPX-432 (was: 36 to 32 bit transition Lines: 43 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Message-ID: X-Trace: +r4CDioPsFvPSyWBph6WAnMeOm0gR12DdmbJUgSxzx50XhcuGCN00EGIozo5Ch+b9sAU+GXN5CNu!QY+zz6k72q21kaUqkfZtp8K4Nk0wyR8K9bh5YQ98hOebqbw5hWhGt0EFKupJKOJg3hwnEdwSmU9x!bXaOAZg= X-Complaints-To: abuse@gte.net X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly NNTP-Posting-Date: Thu, 31 Aug 2000 22:19:51 GMT Distribution: world Date: Thu, 31 Aug 2000 22:19:51 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!sunqbc.risq.qc.ca!newsfeed.direct.ca!look.ca!paloalto-snh1.gtei.net!news.gtei.net!paloalto-snr1.gtei.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:62797 "Peter van der Linden" wrote in message news:8omg7e$1u2l$1@nntp1.ba.best.com... > > Did the 432 really have predecessors and successors? > > Well, it was a revolution in both software and hardware, and the > effects caused Intel to say "Never again attempt a revolution in both > hardware and software simultaneously." > > One of the problems was that the 432 project took place just as the > fuse was lit on the 8088/8086 (i.e. at the same time as the astonishing > take-up of the IBM PC. At that time, Intel only had so many resources, > and they were keen to divert the major funding to the most successful > projects. The 432 couldn't point to a huge revenue stream any time in > the near future, whereas the x86 was going up like a rocket. No brainer. > > >said, it seemed to disappear as quickly as it came. Has anyone ever > >actually seen working 432 hardware? > > I have. It worked fine. It was slower than anyone thought (you had > to pay for putting all that OOP support into the hardware), not > faster. But it did work fine. It was 20 years ahead of its time. > I don't know what successors it had, but people are trying to solve > the same problems today, with projects like Java chips. I wish I could remember where I read this, but I was browsing through the Montana State's library a couple of years ago and came across one of those cheaply printed collections of articles from some conference or another. There was an analysis of the design and implementation of the 432 and whether or not the goals were achievable or even desirable. Essentially they concluded that the goals might have been achievable then and certainly could be today, but that bad decisions in how the 432 was actually implemented effectively destroyed any chance it might have had. It wasn't just a couple of bad decisions, either, but a whole flock of them that slowed things down to a crawl. Of course, everything's much clearer in hindsight than it is when you're actually doing it. Perhaps someone has read this article and remembers where it might be found? (I've searched for it again and haven't been able to find it (I suspect it's no longer there), and the library catalog's search capabilities are rather scattershot and poor and turns up nothing) Larry ###### From: John Ahlstrom Newsgroups: alt.folklore.computers,comp.arch Subject: Re: iAPX-432 (was: 36 to 32 bit transition Date: Thu, 31 Aug 2000 15:26:41 -0700 Organization: Ye 'Ol Disorganized NNTPCache groupie Lines: 32 Message-ID: <39AEDBA1.A1A0FB62@cisco.com> References: <8ocvhc$24c$1@murrow.corp.sgi.com> <8oll3j$ote$2@news.btv.ibm.com> <3911C34B5AA2DE31.E1C9F3532D62BBF4.D6ED9205B201817A@lp.airnews.net> <8omg7e$1u2l$1@nntp1.ba.best.com> X-Complaints-To: newsabuse@supernews.com X-Mailer: Mozilla 4.72 [en] (WinNT; I) X-Accept-Language: en MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cache-Post-Path: sj-nntpcache-3!unknown@dhcp-171-68-135-37.cisco.com X-Cache: nntpcache 2.4.0b2 (see http://www.nntpcache.org/) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!sunqbc.risq.qc.ca!cyclone.bc.net!newsfeed.stanford.edu!sn-xit-01!supernews.com!sn-inject-01!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:62795 Peter van der Linden wrote: > One of the problems was that the 432 project took place just as the > fuse was lit on the 8088/8086 (i.e. at the same time as the astonishing > take-up of the IBM PC. At that time, Intel only had so many resources, > and they were keen to divert the major funding to the most successful > projects. The 432 couldn't point to a huge revenue stream any time in > the near future, whereas the x86 was going up like a rocket. No brainer. > I may be reading more into what he said but the 432 project (under some different names like 8800, maybe some others) was started immediately after the 8080. It was supposed to be the successor to the 8080. But, trying to do two revolutions at the same time, it was taking much to long - like Stretch??? - and the 8086 was started as a stop-gap. Steve Morse where are you now that we need you? I believe that the 8086 ISA was defined by 2 or 3 guys in a couple of weeks. Many of its characteristics were for backwards compatibility, many more because the really high-power, long-life successor was to be the 432. One of the ISA designers was Bruce Ravenal who, I believe, did have prior B5500 experience. Then the PC happened and the 432 had very disappointing performance and TERRIBLE cost-performance. No brainer. JKA -- Any sufficiently well-rigged demo is indistinguishable from advanced technology.