Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!bloom-beacon!mit-eddie!killer!tness7!tness1!sugar!peter
From: peter@sugar.uu.net (Peter da Silva)
Newsgroups: comp.sys.amiga
Subject: Re: Amiga Dealers/Commodore
Message-ID: <2384@sugar.uu.net>
Date: 2 Aug 88 11:20:14 GMT
References: <4322@cbmvax.UUCP> <1209@flatline.UUCP> <4353@cbmvax.UUCP> <4372@cbmvax.UUCP>
Distribution: comp
Organization: Sugar Land Unix - Houston, TX
Lines: 25

In article <4372@cbmvax.UUCP>, marc@cbmvax.UUCP (Marc Rifkin CATS) writes:
> Additionally, and most important, [the Amiga's] architecture is based on
> such operation (COprocessors, separate busses, DMA).  Without these
> features, multitasking on the Amiga would be slow and impractical.  

Without these features, moving windows around would be quite a bit slower...
however multitasking would be quite reasonable and practical. For any CPU
with more than 64K of address space (and a good many with less) multitasking
is such an obvious win that I'm still flabbergasted that anyone is satisfied
with less. I've written a basic timesliced multitasker (no preemption, no
realtime) in Forth that runs under CP/M on an 8080. About 50 lines of code
total. I've used a Forth development system that supports multiple users
with less than 64K, on a computer with a 4-bit ALU. I've run UNIX on a stock
IBM-PC/XT, and it was quite a bit faster than MS-DOS.

In fact you can buy a computer with a very similar operating system to the
Amiga's with no coprocessors... the Sinclair QL.

I suspect that you're looking at Multifinder and OS/2 and saying "there but
for the grace of Agnes go I". It's not the lack of CPU power that cripples
these systems, it's software... the need to retain compatibility with an
obsolete and inefficient software architecture.
-- 
		Peter da Silva  `-_-'  peter@sugar.uu.net
		 Have you hugged  U  your wolf today?