Path: utzoo!attcan!uunet!wuarchive!cs.utexas.edu!tut.cis.ohio-state.edu!ucbvax!pasteur!cory.Berkeley.EDU!navas
From: navas@cory.Berkeley.EDU (David C. Navas)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: (B) 68040 vs. gfx coprocessor <pipe(lined) dreams>
Message-ID: <21663@pasteur.Berkeley.EDU>
Date: 1 Feb 90 22:25:55 GMT
References: <633@xdos.UUCP> <3046@pur-phy> <16389@boulder.Colorado.EDU> <677@h.cs.wvu.wvnet.edu> <641@xdos.UUCP>
Sender: news@pasteur.Berkeley.EDU
Reply-To: navas@cory.Berkeley.EDU.UUCP (David C. Navas)
Organization: University of California, Berkeley
Lines: 20

In article <641@xdos.UUCP> doug@xdos.UUCP (Doug Merritt) writes:
>In article <677@h.cs.wvu.wvnet.edu> dbl@h.cs.wvu.wvnet.edu (David Lawson) writes:
>As I was just discussing with David Sinz in another article, there's a question
>as to how many memory cycles are available beyond what the 040 uses. If there

It says in the Byte article that the 68040 was designed to stay *off* the bus
better than its predecessors.  Therefore, it would seem, that if it makes sense
to have co-processors with the other members of the 680x0 then it makes even
more sense with the 68040.

I think the real issue is the cost/benefit ratio.  IE: after considering the
price of such a co-processing system, does it deliver [at least linearly with
cost] an increase in throughput?

Otherwise we're back up to out-of-reach workstation-type pricing.
>Doug Merritt		{pyramid,apple}!xdos!doug
>Member, Crusaders for a Better Tomorrow		Professional Wildeyed Visionary

David Navas
navas@cory.berkeley.edu