Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!ames!hc!beta!cmcl2!rna!rocky2!cowburn
From: cowburn@rocky2.UUCP
Newsgroups: comp.arch
Subject: Re: chewing up mips with graphics
Message-ID: <359@rocky2.UUCP>
Date: Wed, 10-Jun-87 08:53:39 EDT
Article-I.D.: rocky2.359
Posted: Wed Jun 10 08:53:39 1987
Date-Received: Sat, 13-Jun-87 08:20:09 EDT
References: <8270@amdahl.amdahl.com>
Organization: Rockefeller Univ.,N.Y.C 10021
Lines: 23
Summary: graphics displays need more power, and ...

In article <8270@amdahl.amdahl.com>, chuck@amdahl.amdahl.com (Charles Simmons) writes:
> In my recent posting, I forgot to mention that I was exposed to video
> tapes of computer generated animation by Eugene Miya at NASA.  Eugene
> also mentions that I didn't nearly cover the possible applications
> for high-powered computers running graphics programs.  Some of the
> other applications he mentions are weather forecasting, medical imaging,
> CAD/CAM, computer enhancement of satellite photos, and I think there
> were one or two more that I'm still forgetting.
> 

perhaps another point to consider is that the need for graphics in the
applications mentioned is still just the first part of the
interpretative effort which has yet to be reduced satisfactorily to
machine activity.  Oil companies like to look at the maps before oil
wells are dug, the day when the Cray spits out "Drill Here" ( and is
accepted) is distant; the radiologist still likes to hold the film to
the lighbox, and no doubt we would all be unhappy with some SUN in the
operating room directing the surgeon where to make the first incision
for the appendectomy.  Yet these tasks are clearly reducible to machine
activity by appropriate scene analysis and "AI"-style incorporation of
preexisting expertise.  There is, of course, some severe lag in
incorporating even existing algorithms into satisfactory code for
supercomputers.