Date: Mon, 10 May 1999 01:47:48 -0400 (EDT) From: "John T. Farmer" <jfarmer@goldsword.com> To: cjames@opensite.com, freebsd-hardware@FreeBSD.ORG, nicole@nmhtech.com Cc: jfarmer@goldsword.com Subject: RE: Motherboard comparisons Message-ID: <199905100547.BAA07697@rapier.goldsword.com> In-Reply-To: <XFMail.990509181821.nicole@nmhtech.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 09 May 1999 18:18:21 Nicole Harrington said: > On 07-May-99 My Secret Spies Reported That Cathy James wrote: > >>I just ordered 3 motherboards for testing with the AMD k-6III 400. I > >>think they were (PO's at work) TYAN, FIC, ASUS. Since it seems a lot > >>of people are curious, I will post my tests when I get done in abt a week. > > > > Did you ever post this, Nicole, or is it still a work in progress? > > Hmm.. Poke poke... > > Well so far it is sort of work in progress. > > I found that Tyan does not have a 100Mhz buss ATX socket 7 MB, only AT. No > good. Buzz! Not True! Tyan makes the S1590S Trinity 100 AT with the VIA Apollo MVP3 AGPset chipset. This board is solid and _very_ stable. It also supports parity & ECC memory operations at the 100Mhz level (see comment about the ASUS below). <http://www.tyan.com/products/html/s1590s.html>; This is _the_board_ to use (IMHO) for low cost, high performance servers (I've used it in both FreeBSD and Netware5 systems without a probem.) It will also work with either AT or ATX style power supplies (for extra control/monitoring...) The pure ATX version is the S1592S Trinity ATX. Look at: <http://www.tyan.com/products/html/s1592s.html>; I've used the AHA-2940U2W card with the Tyan S1590S in Netware servers were they "just work" (at Mach speed!!). Under FreeBSD, I've settled in the Diamond Fireport series for low cost with super performance. > Asus works great, however with the particulat setup I had I kept getting > stuck at the "waiting 15 secounds for SCSI devices." I had a bunch of LVD > drives hanging off of an adaptec 2940U2W card. I had the same problem with > the FIC board. Thus no tests for either. I have two downchecks on the ASUS P5A (P5A-B for AT format). The first is that it only supports parity & ECC memory operation at 66Mhz. The second is that the very first board I received died during burn-in, delaying the delivery of a system to a client by a couple of weeks. (Picky, I know, but it was the _first_ ASUS board I had do that.) I gave the client a different board & put the replacement P5A-B in my personal WinDoze machine with non-parity SDRAM. So far so good. > Gigabyte MB worked great with a K6/2-400 and the K6/3-400. The K6/3-400 is > worth the price. It is F A S T. I plan on using that with the gigabyte MB for > my servers for a bunch of things. I didn't do any real bench marking as I > didn't have time (and they kicked me out of the lab to put in shelves) but I > found the K6/3-400 visably faster than the K6/2-400. > > I still have a Epox MB to test. Epox has in the past been ok, but at times > unstable. Thus I just decided to go with the Gigabyte Board. It has been rock > solid in all my tests. I haven't tried the Gigabyte or the Epox boards. The farthest I've strayed from ASUS or Tyan is to try in the Tekram and DFI Super Socket7 boards for Win-9x boxes (clients, you know...) > BTW, my application is for the new FreeBSD port of the Inktomi caching > software. Good stuff! I plan on talking about it at the next BAFUG meeting. > (www.bafug.org) > Tell us more!!! John ---------------------------------------------------------------------- John T. Farmer Owner & CTO GoldSword Systems jfarmer@goldsword.com 423-691-6498 Knoxville TN Internet Services & Servers, Network Design, Consulting To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199905100547.BAA07697>