From owner-freebsd-hackers Sat Apr 19 01:17:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA01895 for hackers-outgoing; Sat, 19 Apr 1997 01:17:13 -0700 (PDT) Received: from rf900.physics.usyd.edu.au (rf900.physics.usyd.edu.au [129.78.129.109]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA01867 for ; Sat, 19 Apr 1997 01:16:44 -0700 (PDT) Received: (from dawes@localhost) by rf900.physics.usyd.edu.au (8.8.5/8.8.2) id SAA10612; Sat, 19 Apr 1997 18:16:15 +1000 (EST) Message-ID: <19970419181615.45445@rf900.physics.usyd.edu.au> Date: Sat, 19 Apr 1997 18:16:15 +1000 From: David Dawes To: hackers@freebsd.org Subject: Re: Video Card Q References: <199704182313.QAA27968@crab.whistle.com> <199704182338.QAA01538@rah.star-gate.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69 In-Reply-To: <199704182338.QAA01538@rah.star-gate.com>; from Amancio Hasty on Fri, Apr 18, 1997 at 04:38:40PM -0700 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Fri, Apr 18, 1997 at 04:38:40PM -0700, Amancio Hasty wrote: >Well, I still swear by my old S3 968 with 4MB and by the way the card >is faster than the Matrox Millenium when is operated at depths of 24 or 16 >which is what are normally use over here. > >Not sure however I think that XFree86 does not fully support most of >the accelerated features on the Matrox Millenium. That is true for the 3.2 release, but not for the 3.2A beta. The Millennium is the fastest card supported by 3.2A by a large margin. Just digging a few numbers out from www.goof.com's xbench archive shows that a Millennium at 32bpp (packed 24bpp is buggy and/or interacts poorly with some clients) comes in at between 366k-444k xstones on a Pentium 133, while a #9 Motion 771 (S3 968) at 8bpp comes in at 273k xstones on a Pentium 100. Unfortunately they don't have results for the two cards running with the same speed CPU (or better still, on exactly the same machine), but it does indicate that the Millennium at 32bpp beats the 968 at 8bpp in the overall xstone result. For reference the 8bpp Millennium results shown there for a Pentium 133 are between 733k and 940k xstones. Note, I just took these results from the "3.2 Betas" summary page at www.goof.com. Although they seem give a reasonable general indication of the relative performance that is consistent with my experience with both cards, they are only "benchmarks" and should be treated as such. David