Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site utastro.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!gamma!epsilon!zeta!sabre!bellcore!decvax!linus!philabs!cmcl2!seismo!ut-sally!utastro!nather
From: nather@utastro.UUCP (Ed Nather)
Newsgroups: net.arch
Subject: Re: RISC/CISC - IBM mainframes & ancient history
Message-ID: <243@utastro.UUCP>
Date: Sat, 15-Jun-85 16:22:30 EDT
Article-I.D.: utastro.243
Posted: Sat Jun 15 16:22:30 1985
Date-Received: Tue, 18-Jun-85 05:35:17 EDT
References: <1452@ecsvax.UUCP> <1062@peora.UUCP>
Organization: U. Texas, Astronomy, Austin, TX
Lines: 33

> A reasonable approach to producing CISC machines, then, would be to have the
> compiler writers design the instruction set; and to have the instruction set
> be changeable for different languages.  

The Packard-Bell PB-440 computer (what, never heard of it? well ...) was a
Complex Instruction Set Computer with instructions realized in microcode
stored in a writable control store.  The instruction set was designed by
the compiler writing team (all 3 of us) and had a different instruction set
for different jobs (e.g. the Fortran compiler used a different instruction set
from the instructions generated for run-time use).

> This could be extended by having the
> instruction set be further changeable by the individual user, 

We did an instruction set that emulated a popular IBM computer of that era,
and was clocked to run faster than the original.

> or automatically
> by a program that analyzes instruction usage and moves frequently-used
> instructions into microcode (vertical migration).

Neat idea.  We didn't think of that one.

> (These are not new ideas, of course ...
> -- 
> Full-Name:  J. Eric Roskos

The prototype PB-440 first ran in 1962.
-- 
Ed Nather
Astronony Dept, U of Texas @ Austin
{allegra,ihnp4}!{noao,ut-sally}!utastro!nather
nather%utastro.UTEXAS@ut-sally.ARPA