Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!husc6!think!ames!pioneer!eugene
From: eugene@pioneer.UUCP
Newsgroups: comp.arch
Subject: Re: chewing up mips with graphics
Message-ID: <1803@ames.UUCP>
Date: Wed, 17-Jun-87 18:19:32 EDT
Article-I.D.: ames.1803
Posted: Wed Jun 17 18:19:32 1987
Date-Received: Fri, 19-Jun-87 05:32:21 EDT
References: <8270@amdahl.amdahl.com> <359@rocky2.UUCP> <2120@dg_rtp.UUCP> <1658@umn-cs.UUCP>
Sender: usenet@ames.UUCP
Reply-To: eugene@pioneer.UUCP (Eugene Miya N.)
Organization: NASA Ames Research Center, Moffett Field, Calif.
Lines: 37

In article <1658@umn-cs.UUCP> herndon@umn-cs.UUCP (Robert Herndon) writes:
>
>  Cray has been selling multiprocessor systems for a few years now.
>Even the machines Seymour Cray designed while at CDC were multiprocessors
>(central processor and lots of little peripheral processors) and have
>come in dual central processor models for some time.

Evans and Sutherland has also been selling high performance multiprocessor
graphics systems for years as well.  In some ways they outperform Crays
in floating point multiplication (send inquires to esvax).

>  Many applications would benefit significantly from parallelism.
>I'm informed that ray-tracing graphics is trivially parallelizable

Trivial eh?  This is not a pursuit game.  Trivial ray examples
parallelize trivially.  You only need look at the world around you.

>to many many machines.  Many types of simulation are also parallelizable
>(weather prediction, thermal simulations, flow simulation, ...).
>Most physical simulations can be mapped onto a planar 3D mesh for
>significant speed gain.

While you use the word "most" understand, many 3-D simulations and
analyses use higher-order dimensions to hold other useful data: alpha
channel for transparency in graphics systems, homogeneous coordinate
systems, audio channel sync stuff, and so forth.  At LANL and LLNL there
are 7-D problems which is limited to the FORTRAN.  Don't restrict
your view to 3-D only because you can conceptualize in 3-D.

From the Rock of Ages Home for Retired Hackers:

--eugene miya
  NASA Ames Research Center
  eugene@ames-aurora.ARPA
  "You trust the `reply' command with all those different mailers out there?"
  "Send mail, avoid follow-ups.  If enough, I'll summarize."
  {hplabs,hao,ihnp4,decwrl,allegra,tektronix,menlo70}!ames!aurora!eugene