From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.1.3 (2006-06-01) on yquem.inria.fr X-Spam-Level: X-Spam-Status: No, score=0.0 required=5.0 tests=AWL autolearn=disabled version=3.1.3 X-Original-To: caml-list@yquem.inria.fr Delivered-To: caml-list@yquem.inria.fr Received: from concorde.inria.fr (concorde.inria.fr [192.93.2.39]) by yquem.inria.fr (Postfix) with ESMTP id C50FBBC69 for ; Thu, 5 Jul 2007 09:45:20 +0200 (CEST) Received: from mail15.bluewin.ch (mail15.bluewin.ch [195.186.18.63]) by concorde.inria.fr (8.13.6/8.13.6) with ESMTP id l657jKps031452 for ; Thu, 5 Jul 2007 09:45:20 +0200 Received: from [10.0.1.2] (85.2.25.144) by mail15.bluewin.ch (Bluewin 7.3.121) id 467A58C40040B7A4 for caml-list@yquem.inria.fr; Thu, 5 Jul 2007 07:45:20 +0000 Mime-Version: 1.0 (Apple Message framework v752.2) In-Reply-To: <200707050405.12031.jon@ffconsultancy.com> References: <002501c7be76$d0348630$ac01a8c0@johnyaya> <200707042222.13783.jon@ffconsultancy.com> <3E92B237-EEF9-48BF-A0B9-C3793C1B3D9D@epfl.ch> <200707050405.12031.jon@ffconsultancy.com> Content-Type: text/plain; charset=ISO-8859-1; delsp=yes; format=flowed Message-Id: <1F16F766-1C00-4C69-9154-9A2A1528FB5C@epfl.ch> Content-Transfer-Encoding: quoted-printable From: =?ISO-8859-1?Q?Daniel_B=FCnzli?= Subject: Re: [Caml-list] ANN: Chess III Arena 0.5 Date: Thu, 5 Jul 2007 09:45:49 +0200 To: caml-list@yquem.inria.fr X-Mailer: Apple Mail (2.752.2) X-Miltered: at concorde with ID 468CA190.000 by Joe's j-chkmail (http://j-chkmail . ensmp . fr)! X-Spam: no; 0.00; bunzli:01 buenzli:01 sig:01 node:01 caml-list:01 epfl:02 ensuring:02 opengl:02 opengl:02 objects:02 objects:02 optimization:03 optimization:03 dependent:04 daniel:04 Le 5 juil. 07 =E0 05:05, Jon Harrop a =E9crit : > There's no way this optimization should slow the program down =20 > unless its > software rendered, [...] > Yes but the performance difference between texture objects and =20 > display lists > should be tiny compared to the difference with immediate mode, =20 > which is what > we're comparing here. No, it can result in a slow down even if it is harwdare rendered, the =20= effect of your "optimization" is implementation dependent. The =20 problem is that glTexImage2d cannot be compiled efficently in a =20 display list (see the bottom of this page [1] for why this is the =20 case) and it may be that in the implementation of OpenGL he uses the =20 cpu to gpu texture transfer is done each time the display list is =20 called. In that case if your bottleneck is in the cpu to gpu transfer =20= of textures, the performance advantage of retained vs immediate is =20 lost. The only platform independent way of ensuring that cpu to gpu =20 texture transfers are not performed each time the display list is =20 called is to use texture objects. Daniel [1] http://www.bluevoid.com/opengl/sig00/advanced00/notes/node60.html