ECS K7S5A with Data corruption/memtest errors? READ THIS!

[TheBaumSquad]

Hey guys. If you are having these problems while running at 133/133, especially on higher speed chips (1.2 - 1.4 TBirds) you REALLY need to read this thread! An extrememly experienced engineer has found a problem and a hardware fix for these problems. He has written a very well rounded FAQ for these problems, and it is something all K7S5A owners should read. Of course, you should run Memtest to check out your own system, then read this thread to get an idea of what your problems are from. This looks to be a HUGE problem in the board design. Anyway, read it Here

FAQ 1.0 on K7S5A (M830) Data Corruption - Memtest86, 1400, CONFIGMG, 133/133 Problem

-Paul Baumgaertner

 

[Buz2b]

For a "View from the other side" of this debate, I offer the following:

I respect the work that went into that testing and do not doubt that the individual is both knowledgable and sincere in his work. However, I do have some issues with his "testing".

First, the entire foundation of his research is based on errors that showed in the Memtest86 memory test. The problem is that many other reliable test platforms could not duplicate these findings. In addition many many individuals that have shown these apparent "problems" in Memtest have perfectly solid, excellent performing boards. While he is showing "Data Corruption", there has been no such substantiated reports of any corruption of data, with the exception of the Win2K floppy bug that was later solved with a BIOS update. There are other reasons that I will not go into right now. I am sorry but I just do not buy into it. I also see many who are preaching the gospel according to his testing as the "holy grail" to solving problems you might have with the K7S5A. It is simply not that at all IMHO. It is indeed a very time consuming report that he has posted. It is also a valuable research tool for some. I just have a problem with what appears to be "chasing Ghosts". I would encourage you to read this "test and FAQ" material very carefully and pay particular attention to exactly who has the problem and where it shows up and on what testing platform it shows up on. There are just too many holes in this work. A whole lotta something about nothing! Enjoy the read though.

For those who will undoubtedly say that I have "always been a supporter" of this board and that as such my opinion might be questionable, I say only that I stand on the voice of reason. I refuse to get caught up in a "frenzy" every time someone announces that they have all the answers for this board or for any other product for that matter. I've always been that way and I do not apologize for it.

 

[TheBaumSquad]

LOL! We replied to eachothers threads. Heh heh, now that's good discussion! Heh heh, you rock Buz! Okay, I'll read what you said now..

Okay, I already pretty much responded to this in the final official thread. Heh heh, probably better there anyway.

-Paul

 

[emjem]

K7S5A love is blind.

MrAthlon has taken "opinion" out of the lengthy list of problems with this mobo. It doesn't matter if there is proven data corruption or not, he has fixed a number of k7s5a mobos by correcting what he found to be wrong with the basic design of the board.

Anyone having any problem with this board should immediately go and read his FAQ. That will save you time, and money.

Anyone comtemplating buying this mobo -- don't unless you feel lucky.

I HAD one of them for the better part of 2 weeks before sending it back to where it belongs. Since this is my hobby I didn't mind reading posts for several hours and trying this and trying that. But no way was I going to keep a mobo that was basically unstable and quirky.

I now have a Shuttle AK31a. I installed it in the very same case with the very same components that the k7s5a mobo didn't like, including a TB 1400, Enlight 300 watt power supply and a $9 Volcano2 hs/fan.

For just $20 more, the ak31a booted the first time, OC'd the first time (with shipped bios) and has not crashed once in 3 days. So much for the BS about needing more power and/or better cooling to run a fast Athlon based system.

 

[TheBaumSquad]

Emjem, glad to hear that the AK is working out well for ya!

I am taking my chance on another K7S5A. It's a great board, when it works. So I'm RMAing mine, and crossing my fingers for the next one (at least now I have a good method of testing with memtest as well as attemtping the installs that constantly corrupted for me in the past)

Anyway, I guess I'll see what happens, but many people are running fine with their setups, so I'll take my chances on getting a good board, and if all else fails, I'll get some good new info for what's going on. Heh heh.

Good luck to all!

-Paul

 

[bevo]

<< Anyone comtemplating buying this mobo -- don't unless you feel lucky. >>

This is Hogwash! Luck has very little to do with it. I've now configures 10 k7s5a systems. All run like gazelles...