Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 alpha 4/15/85; site mordred.purdue.UUCP
Path: utzoo!watmath!clyde!cbosgd!ihnp4!inuxc!pur-ee!pucc-j!purdue!cak
From: cak@purdue.UUCP (Christopher A. Kent)
Newsgroups: net.arch,net.micro.pc
Subject: Re: "Origins of an Architecture: 801/RISC"
Message-ID: <490@mordred.purdue.UUCP>
Date: Sat, 15-Feb-86 19:05:07 EST
Article-I.D.: mordred.490
Posted: Sat Feb 15 19:05:07 1986
Date-Received: Mon, 17-Feb-86 04:58:26 EST
References: <1194@ecsvax.UUCP>
Distribution: net
Organization: Department of Computer Science, Purdue University
Lines: 18
Xref: watmath net.arch:2529 net.micro.pc:7002
Summary: Seymour Cray was there first.

In article <1194@ecsvax.UUCP>, hes@ecsvax.UUCP (Henry Schaffer) writes:
> The original idea [of RISC] is credited to Dr. John Cocke, IBM Fellow in
> 1975. 
> 
> However this may be, is the RISC architecture an example of IBM
> leading the way in an architectural innovation?

I dunno. I think an argument can be made that Seymour Cray was
designing RISC machines more than 10 years earlier. Look at the CDC
6000 series, started in 1960, first delivered in OCtober 1964 (a 6600).
It has a small number of simple instructions, efficiently implemented, and
simple addressing modes (the processor is a three address machine, but
the addressing modes correspond to register, direct, and
register-index). Those seem to be the salient features of the RISC 
architecture idea.

It's just my opinion,
chris