Re: Future development of PEX/PHIGS

Article View: comp.windows.x.pex

Article #53

Back to comp.windows.x.pex Previous Next

#53

From: Stuart Levy
Date: Wed, 26 Aug 1992 01:07

11 lines
1297 bytes

In the debate over efficient transport of graphics data, I'm
still puzzled by the fact that GL & OpenGL both use a procedure call
for each low-level data item.  On the current MIPS processors,
does a procedure call + jump through shared-library table + return sequence
plus something like store-pointer-into-magic-location really come out
faster than, say, copying three or four words of memory plus 1/N
procedure calls for an object with N vertices?  Or is this a design that
will win on future systems, but doesn't especially on current ones?

    Stuart Levy, Geometry Center, University of Minnesota
    slevy@geom.umn.edu

Technical Details

Message-ID: <1992Aug26.010706.2647@news2.cis.umn.edu>

Path:

arc1-ca.pugleaf.net!sparky!uunet!charon.amdahl.com!pacbell.com!mips!news.cs.indiana.edu!umn.edu!noneuclid.geom.umn.edu!slevy

References: <or5vc3k@fido.asd.sgi.com> <1992Aug24.190727.19006@shograf.com> <p1f880s@fido.asd.sgi.com>

Show Full Headers

Complete Article Headers

Newsgroups: comp.windows.x.pex,comp.graphics,comp.sys.sgi
Path: arc1-ca.pugleaf.net!sparky!uunet!charon.amdahl.com!pacbell.com!mips!news.cs.indiana.edu!umn.edu!noneuclid.geom.umn.edu!slevy
From: slevy@noneuclid.geom.umn.edu (Stuart Levy)
Subject: Re: Future development of PEX/PHIGS
Message-ID: <1992Aug26.010706.2647@news2.cis.umn.edu>
Keywords: PEX, OpenGL
Sender: news@news2.cis.umn.edu (Usenet News Administration)
Nntp-Posting-Host: noneuclid.geom.umn.edu
Organization: Geometry Center, University of Minnesota
References: <or5vc3k@fido.asd.sgi.com> <1992Aug24.190727.19006@shograf.com> <p1f880s@fido.asd.sgi.com>
Date: Wed, 26 Aug 1992 01:07:06 GMT
Lines: 11

Reply to Parent Message