home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!olivea!sgigate!odin!sgi!rhyolite!vjs
- From: vjs@rhyolite.wpd.sgi.com (Vernon Schryver)
- Newsgroups: comp.arch
- Subject: Re: Computer architects forced to change careers. NOT!
- Message-ID: <pnjn93s@rhyolite.wpd.sgi.com>
- Date: 11 Sep 92 17:28:49 GMT
- References: <1992Sep9.061933.28304@news.eng.convex.com> <1992Sep10.194733.8001@boole.uucp>
- Organization: Silicon Graphics, Inc. Mountain View, CA
- Lines: 36
-
- In article <1992Sep10.194733.8001@boole.uucp>, John@boole.uucp (John Ahlstrom) writes:
- > ...
- > >Take a look at the history of the Intel 432. Was there ever a computer
- > >architecture that was a worse disaster?
- >
- > Was there any "new technology" required "to be invented on
- > the spot by the" logic designers, circuit designers, process
- > engineers (the analog to the construction engineers)?
- >
- > I think the problem was what was seen as the requirements
- > for the machine. Particularly very small-grained security
- > and the religious refusal to use faster mechanisms (e.g. jsr
- > rather than protected call) even when they existed. Other
- > things like bit addressability were so new technology (cheap
- > ram) would NOT have to be invented on the site by construction
- > engineers.*
-
-
- The talks at SOSP and the INTEL Traveling Road Show were exciting.
- Then I got manuals, and found that of the many different, redundant,
- and contradictory operating system mechanisms that I, at least, knew
- about back then, the 432 seemed to implement each and every one.
-
- I've forgotten the details, but I remember running around, waiving the
- manuals and challenging people to think of a queueing, scheduling,
- messaging, or exclusion mechanism that was not all but completely
- implemented in the silicon. And this was years after the Needham-Lauer
- dichotomy was popular.
-
- Whatever else it's faults or virtues, the 432 demonstrated that the
- kitchen sink design school doesn't work, that anybody can add features
- to anything, but that the real work and value is in leaving the
- right things out.
-
-
- Vernon Schryver, vjs@sgi.com
-