Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!decwrl!sgi!shinobu!odin!bennett
From: bennett@sgi.com (Jim Bennett)
Newsgroups: comp.sys.sgi
Subject: Re: old vs. new style drawing on the Personal IRIS
Message-ID: <1991Apr23.223748.28054@odin.corp.sgi.com>
Date: 23 Apr 91 22:37:48 GMT
References: <9104222144.AA26082@chem.chem.ucsd.edu> <1991Apr23.174445.21278@odin.corp.sgi.com>
Sender: news@odin.corp.sgi.com (Net News)
Organization: Silicon Graphics, Inc., Mountain View, CA
Lines: 30

In article <1991Apr23.174445.21278@odin.corp.sgi.com> kurt@cashew.asd.sgi.com (Kurt Akeley) writes:
>In article <9104222144.AA26082@chem.chem.ucsd.edu>, sdempsey@UCSD.EDU (Steve Dempsey) writes:
>
>|> Will the performance penalty for using the "high performance" vertex drawing
>|> functions on the Personal IRIS go away with IRIX 4.0, or is this embedded
>|> in the graphics hardware of the PI?
>|> ---------------------------------------------------------------------------
>
>I know of no plans to retune the Personal Iris family for the 4.0 release.
>Such tuning would require microcode changes, so the penalty is related to
>the graphics hardware, if not "embedded" in it.
>
>Perhaps someone from the Entry Systems Division will comment?
>
>-- Kurt

You're right, I should have commented earlier, but you seemed to be
handling this OK.  Anyway, I have run many benchmarks on the PI, and
the difference between v3f and move-draw lines is not significant.

That is, the performance varies more from one run to the next of the
SAME benchmark, than it does from a v3f to a move-draw benchmark.
So I have always recommended the v calls as the way to go, on all of
our machines.

If anyone has a benchmark that shows a significant difference between
move-draw and v3f, I would like to take a look at it.  You can mail
it to me at my address below.

Jim Bennett				(bennett@sgi.com)