home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.arch
- Path: sparky!uunet!pipex!warwick!pavo.csi.cam.ac.uk!cam-cl!rf
- From: rf@cl.cam.ac.uk (Robin Fairbairns)
- Subject: Re: Computer architects forced to change careers. NOT!
- Message-ID: <1992Sep11.133532.19502@infodev.cam.ac.uk>
- Sender: news@infodev.cam.ac.uk (USENET news)
- Nntp-Posting-Host: ely.cl.cam.ac.uk
- Reply-To: rf@cl.cam.ac.uk (Robin Fairbairns)
- Organization: U of Cambridge, England
- References: <1992Sep9.061933.28304@news.eng.convex.com> <1992Sep9.213319.7894@fasttech.uucp>
- Date: Fri, 11 Sep 1992 13:35:32 GMT
- Lines: 40
-
-
- In article <1992Sep9.213319.7894@fasttech.uucp>, zeke@fasttech.uucp
- (Bohdan Tashchuk) writes (quotes omitted):
- |> Take a look at the history of the Intel 432. Was there ever a computer
- |> architecture that was a worse disaster?
-
- I was a manager when the 432 was announced. With commendable
- dedication to principle I delegated attending a seminar Intel had
- invited us to. The woman I sent had (like me) come out of CS
- research. When she came back, she remarked that the whole set up
- was so exciting that she would have jumped at a job if they had been
- offering them.
-
- The 432 attacked problems that I had been involved with in
- architecture research in the 70s. While it obviously failed, it
- seemed like a signpost to me - silicon was going to become more and
- more complex, and be able to solve more and more extreme problems. Of
- course, what has actually happened is that the complexity has gone on
- increasing, but (following the earthquake of the introduction of RISC
- in most of the industry) has been pretty largely devoted simply to
- increasing throughput, ignoring so many of the CS-y ideas that the 432
- addressed.
-
- |> And yet, those most directly responsible were (at least initially)
- allowed an
- |> almost completely free hand to architect what eventually became the
- Intel 960.
- |> [...]
- |>
- |> The concepts behind RISC have helped bring some sanity back into the field.
-
- Hmmm. Depends what you mean by sanity. I still fancy the idea of
- firewalls existing in the hardware I use. Don't tell me it's not
- possible (chip design is so damned clever nowadays). I do understand
- that it's not the way of (today's) world, and I appreciate the quality
- of the thinking behind many of today's designs. But, as a programmer
- (at heart ;-), I still appreciate other things than pure speed...
- --
- Robin (dump Brian Kay) Fairbairns rf@cl.cam.ac.uk
- U of Cambridge Computer Lab, Pembroke St, Cambridge CB2 3QG, UK
-