Date: Fri, 31 Aug 2001 05:10:43 -0400 From: Allen Landsidel <all@biosys.net> To: freebsd-stable@FreeBSD.ORG Subject: Re: FreeBSD and Athlon Processors Message-ID: <5.1.0.14.0.20010831044409.00c26c38@rfnj.org> In-Reply-To: <200108310822.BAA03593@mina.soco.agilent.com> References: <Your message of "Fri, 31 Aug 2001 03:12:50 EDT." <5.1.0.14.0.20010831024941.00c3fe18@rfnj.org>
next in thread | previous in thread | raw e-mail | index | archive | help
At 01:22 8/31/2001 -0700, Darryl Okahata wrote: > Do a google groups search on "via chipset ide corruption". You >might be surprised at what you find. Just because you don't have >problems, doesn't mean that the problem doesn't exist. Ok I'll give you that much. But the flip side of that coin is that just because some people are having problems doesn't mean that it's defective hardware, or even if it is hardware, it doesn't by definition mean it's the chipset. I remember when the Athlons first came out, and before I ordered my MSI, I did a usenet search on "athlon problems" or something to that extent. I picked the board that seemed to only appear in response to peoples problems, with the owner saying "well I have an MSI-6167 and I'm not having any of those problems." Soon after the athlons came out people were complaining all over the place about stability issues, hangs, crashes, and all kinds of problems. As it turns out, the great majority of these problems were -other- cheap hardware used in building the machines and other forms of user error such as : 1. Overclocking without adequate knowledge of how to do so effectively (I won't say overclocking in general because a great many people, including myself, overclock without any problems.) 2. An in adequate power supply. AMD is very specific about how much power the Athlon needs to operate, and so is nVidia about their AGP graphics cards. Ignore these -requirements- at your peril. 3. Incompatible or defective memory. The original athlons required unbuffered PC100 memory. People using PC66 memory or buffered dimms had problems, and they deserved them. On rare occasions there are incompatibilities, and on even rarer ones they cannot be fixed by a new driver/firmware. One such case I have encountered, between the KT133/686 and the Aureal Vortex2 sound cards. This is a known problem, and will probably never be fixed, seeing as how Aureal is no more. Since I wasn't going to be seeing any more great chips out of them, or even any more driver updates, I felt no great loss in replacing my Aureal based card with another brand. >[ After scanning some of the hits, it seems that a new BIOS might fix > the problem. ] That's good to know. I have probably never noticed any of these problems because I only have one Athlon in the house with an IDE drive, and using the HPT370 I would assume it's not subject to the same problems as the non-raid integrated controller; there have been no problems with it, in any event. > Heh. I'm writing this on an Asus A7M266 (I didn't want to wait for >the nForce). Well I'm writing this on an Abit KT7, and it will be sent through my MSI-6167 smtp server when I click the button. ;) I did not want to wait for nForce either, but I'm not in a big hurry to get more memory bandwidth.. I'm more awaiting more SMP options. The Tyan boards may have a good reputation for being stable and reliable, but their current board is a reference design, and I'm sure other manufacturers will offer higher performance boards, hopefully with less integrated crap. If I were to suggest a chipset manufacturer to avoid, it would be the ALi chipsets. The "ALiMAGiCK 1" (they insist you write it that way) has been plagued with problems, as was their K6-2 chipset (ALADDiN). OTOH, my VIA based motherboards (KT133 and MVP3) have all been trouble free. > Darryl Okahata > darrylo@soco.agilent.com Just seeing your email address made me think of something funny but totally unrelated. At my office, the idiots who set up the internal network downstairs used IP addresses that were used as examples in some TCP/IP book or another, inside the 130.30/16 netblock, owned by aligent. Do they change it when I bring this to their attention? Hell no. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5.1.0.14.0.20010831044409.00c26c38>