From owner-freebsd-hardware Sun Sep 13 07:09:13 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA05606 for freebsd-hardware-outgoing; Sun, 13 Sep 1998 07:09:13 -0700 (PDT) (envelope-from owner-freebsd-hardware@FreeBSD.ORG) Received: from geo.geocast.net (geo.geocast.net [128.177.240.120]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA05595 for ; Sun, 13 Sep 1998 07:09:08 -0700 (PDT) (envelope-from castor@geocast.net) Received: from localhost (castor@localhost) by geo.geocast.net (8.8.8/8.8.8) with ESMTP id HAA20795; Sun, 13 Sep 1998 07:08:42 -0700 (PDT) Date: Sun, 13 Sep 1998 07:08:39 -0700 (PDT) From: Castor Fu To: Duncan Barclay cc: spork , freebsd-hardware@FreeBSD.ORG Subject: Re: "Cacheable memory"?? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 12 Sep 1998, Duncan Barclay wrote: > > The Ali chipsets, however, look much better. In general all of the buffers etc. > on the ALi 5 are twice as big as the MVP3, this may result in better system > performance. The ALi chipset also has part of the RAM needed for the cache on > chip (the tag RAM). This will help cost/stability, but not necessarily speed. > > Given a choice I would buy a Ali 5 chipset (Aladdin also re-spun the chipset to > help a boot problem with FreeBSD, that's what I call support). Has anyone done any actual performance measurements with these two chipsets? I was rather leary of the Ali5 because ALI is an acer subsidiary and Acer's motherboard arm, AOpen, uses the MVP-3 in their super 7 board. I figure if ALI couldn't get a design win with their own companies, there's something you don't want to find out. I also had a friend with a cheap ALI board which had various nasty problems which only cropped up as one started adding cards. -castor To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message