Xref: utzoo comp.graphics:16917 comp.benchmarks:487
Path: utzoo!mnetor!tmsoft!torsqnt!news-server.csri.toronto.edu!rpi!usc!cs.utexas.edu!hellgate.utah.edu!cs.utah.edu!thomson
From: thomson@cs.utah.edu (Rich Thomson)
Newsgroups: comp.graphics,comp.benchmarks
Subject: Truth or Dare?
Summary: 1 M polys/sec on VGX real or theoretical?
Message-ID: <1991Mar27.174104.26867@hellgate.utah.edu>
Date: 28 Mar 91 00:41:04 GMT
Followup-To: comp.benchmarks
Distribution: comp
Organization: Computer Science Department, University of Utah, SLC, UT
Lines: 30

Several times on comp.graphics the SGI folks have offered up the
famous "1 million polygons/second" quotation from marketing literature
when referring to peak performance on the VGX series of machines.

The polygons in question are 50 pixel triangle strips that are flat
shaded.  Also, they may have rendered them without the Z buffer turned
on (the spec sheet is away from me at the moment).

What I would like to know is: does anybody have a program, no matter
how contorted, that actually achieves this number?  As far as I can
tell the number is derived from theoretical peak performance of the
VGX pipeline, or some portion of it.

This highlights a sore point in graphics benchmarking in general,
which is getting anybody to agree on the same thing to be measured.
Hopefully the PLB and GPC efforts of NCGA will alleviate this problem,
or at least translate it into ``who has the most optimized GPC
program''.

When I asked the local SGI rep for more information about these
numbers, he sent me a data sheet that described the particular
polygons in question (revealing that they were as described above),
but no working code.

So: is this a "theoretical" number, or is it attainable by a program,
no matter how contorted?

						-- Rich
Rich Thomson	thomson@cs.utah.edu  {bellcore,hplabs,uunet}!utah-cs!thomson
    ``Read my MIPs -- no new VAXes!!''  --George Bush after sniffing freon