Path: utzoo!attcan!uunet!lll-winken!lll-tis!helios.ee.lbl.gov!pasteur!ames!nrl-cmf!cmcl2!rutgers!bellcore!tness7!tness1!nuchat!sugar!karl
From: karl@sugar.uu.net (Karl Lehenbauer)
Newsgroups: comp.arch
Subject: Re: Self-modifying code
Message-ID: <2379@sugar.uu.net>
Date: 1 Aug 88 03:20:39 GMT
References: <5254@june.cs.washington.edu> <76700032@p.cs.uiuc.edu> <1988Jul28.170834.6949@utzoo.uucp>
Organization: Sugar Land Unix - Houston, TX
Lines: 20

In article <1988Jul28.170834.6949@utzoo.uucp>, henry@utzoo.uucp (Henry Spencer)
writes some more flamelets, and...
> ...You get a much more useful system if you break down and
> admit that you're building a multiprocessor machine, and make all the
> CPUs general-purpose.

I don't know at what point this becomes weirding out.  Isn't it OK to have a
dedicated processor on the multiport serial I/O boards?  Isn't it OK to have
microprocessors in the disk controllers?  ...in a graphics terminal?  So why
not on a graphics board?  Why can't we put a processor anywhere that's useful
and have general multiprocessing as well?  Why should my graphics CPU, with
dedicated memory and, eventually, parallel processing, hardware transforms
and such have to be complicated by the need to run user programs?  Your 
statement could be read to criticize every engineer designing a part of a 
system who used a uP for something for not implementing with random logic 
instead and adding another general purpose processor.  That's absurd, right?  
OK then, so why is the display among all things sacrosanct from the use of 
custom or dedicated hardware?
-- 
-- 
-- Karl Lehenbauer, karl@sugar.uu.net aka uunet!sugar!karl, +1 713 274 5184