Re: Intel vs. AMD performance

Article View: alt.games.quake2

Article #883

#883

From: Thomas Womack
Date: Thu, 14 May 1998 00:00

14 lines
1342 bytes

In comp.sys.intel Doug Chism <chismdm@earthlink.net> wrote:

: Of course well written code could avoid this by executing
: similar instructions in serial ( i.e. transform all the vertices in the
: scene in 1 loop ), but this often results in cache thrashing since its hard
: to store polygon and vertex data together unless strips and fans are used
: exclusively.

I thought the process was usually to transform all the vertices together and
then draw the polygons - and this sounds as if vector FP ops would be very
useful.

Tom

Technical Details

Message-ID: <6jfnk7$n7m$2@news.ox.ac.uk>

Path: arc1-ca.pugleaf.net!unknown!not-for-mail

References: <6jag25$69v$1@nnrp1.dejanews.com> <355923C5.5B9F66A7@home.com> <6jdqkq$l9s@argentina.earthlink.net>

Show Full Headers

Complete Article Headers

X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 110292,ad77f22a85cd2d5c
X-Google-Attributes: gid110292,public
X-Google-Thread: 113685,ad77f22a85cd2d5c
X-Google-Attributes: gid113685,public
X-Google-Thread: 1045b9,ad77f22a85cd2d5c
X-Google-Attributes: gid1045b9,public
X-Google-Thread: f9d87,ad77f22a85cd2d5c
X-Google-Attributes: gidf9d87,public
From: Thomas Womack <mert0236@sable.ox.ac.uk>
Subject: Re: Intel vs. AMD performance
Date: 1998/05/14
Message-ID: <6jfnk7$n7m$2@news.ox.ac.uk>
X-Deja-AN: 353315761
References: <6jag25$69v$1@nnrp1.dejanews.com> <355923C5.5B9F66A7@home.com> <6jdqkq$l9s@argentina.earthlink.net>
Organization: Oxford University, England
Newsgroups: alt.games.quake2,comp.sys.ibm.pc.hardware.chips,comp.sys.intel,intel.microprocessors.pentium_ii

Reply to Parent Message