Path: utzoo!attcan!uunet!portal!cup.portal.com!bcase
From: bcase@cup.portal.com (Brian bcase Case)
Newsgroups: comp.arch
Subject: Re: A simple question on RISC
Message-ID: <10937@cup.portal.com>
Date: 5 Nov 88 20:57:57 GMT
References: <6544@xanth.cs.odu.edu> <75577@sun.uucp> <10802@cup.portal.com>
Organization: The Portal System (TM)
Lines: 41

In article <10802@cup.portal.com> bcase@cup.portal.com (bcase) writes:
>>Notice also that Seymour's first machine came out in '76 (I believe).
>
>Wrong. UNIVAC, the all of the CDC machines until he left to found Cray
>Research. The CDC 6600 is often listed as the first supercomputer. 

Yes, yes, yes.  Ok, please, please, don't send me any more mail about
the CDC6600.  Things that need to be said:  I just assume that when 
people talk about Cray's machines they are talking about Cray Inc.'s
machines.  Thankfully, there are those out there who remember (better
than I) that he did the CDC6600.

The 6600 is somewhat close to being a RISC machine.  It is hardwired,
has three-address operations (is that right?), some registers, sorta
fixed-length instructions, etc.  But I find it interesting that subsequent
architectures, even Seymour's, diverged from this simplicity.  This
tells me that the architects of those days didn't "get the point."  The
fact that nearly all new architectures are RISCs or RISCy these days, tells
me that the formulation known as RISC has enabled architects to "get the
point."  RISC is a logical, reasoned approach that didn't exist in 1966,
even though someone might have had a simple machine.  This makes me tend
to *dis*believe that those early simple machines were RISCs:  they were
experiments that were not deemed successful enough to change the way
things were done.  RISC was an experiment too, but someone could demostrate
why it was good, and the architects saw that it was good, and it was good.
Par-ump-a-dum-dum.  The mouse and lamb kept time, par-ump-a-dum-dum....

>>The 801 research began in 1975, though it was perhaps not as focused
>>as it later became.  I'd say this is a pretty close call to who was first.
>
>6600 was a 1960's machine. So if it's RISC, it won by a decade.
                              ^^
Right.

>Seymour designs simple things which go fast. He
>assumes that compiler writers can cope. This is, perhaps, not the best
>approach...but his track record is pretty good (CDC owned the high
>performance computer market during his long reign). The Multiflow and
>Cydra machines are, arguably, RISC's .

Well put, I think.