X-pop3-spooler: POP3MAIL 2.1.0 b 4 980420 -bs-
X-Authentication-Warning: basis.Desk.nl: listserv set sender to owner-beastium-list AT basis DOT desk DOT nl using -f
Message-ID: <19981208212652.J8178@cerebro.laendle>
Date: Tue, 8 Dec 1998 21:26:52 +0100
From: Marc Lehmann <pcg AT goof DOT com>
To: beastium-list AT desk DOT nl
Subject: Re: 3dnow in pgcc
Mail-Followup-To: beastium-list AT desk DOT nl
References: <Pine DOT GSO DOT 4 DOT 05 DOT 9812072058470 DOT 5488-100000 AT stekt3> <Pine DOT GSO DOT 3 DOT 96 DOT 981208181014 DOT 11281B-100000 AT gauge4>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
In-Reply-To: <Pine.GSO.3.96.981208181014.11281B-100000@gauge4>; from Jacek Pliszka on Tue, Dec 08, 1998 at 06:11:52PM +0100
X-Operating-System: Linux version 2.1.131 (root AT cerebro) (gcc version egcs-2.92.25 19981130 (gcc2 ss-980609 experimental)) 
Sender: owner-beastium-list AT basis DOT Desk DOT nl
Status: RO
Content-Length: 965
Lines: 20

On Tue, Dec 08, 1998 at 06:11:52PM +0100, Jacek Pliszka wrote:
> > theoretically generate 3dnow opcodes and take advantage of it.
> > I don't know how easy that would to do, and if it would actually
> > be useful optimization.
> 
> Well, if it speeds up floating (or even better - complex) point matrices
> multiplying it would be very, very usefull. That's what I am doing most of the
> time.

Just that it doesn't speed up this in almost all cases, except in situations
where you don't care for precision (3d engines and the like)

      -----==-                                              |
      ----==-- _                                            |
      ---==---(_)__  __ ____  __       Marc Lehmann       +--
      --==---/ / _ \/ // /\ \/ /       pcg AT goof DOT com       |e|
      -=====/_/_//_/\_,_/ /_/\_\                          --+
    The choice of a GNU generation                        |
                                                          |