Diagnosing a motherboard problem

[oron]

A while ago I had an old Asus GeForce2 card in my newly built sytem (athlon64 + msi k8n neo). The card made the system run VERY unstable and just before I pulled the card to get rid of it, I heard a loud "POP" come from near my machine. I smelled and visually checked for any physical damage, but didn't find anything wrong. I pulled the card and replaced it with a new GeForce6800. The machine ran great... for a while.

It slowly started behaving strangely with random read/write errors (corrupt files) and occasional BSOD's.

I replaced the PSU (now an Antec 450). The problems persisted, and slowly got worse.

Finaly my machine would barely even run windows, so i downloaded memtest86 and ran a test. Both my chips (tested individually) reported 1000's of errors. An identical chip from a friend reported 400 errors, and a higher quality Corsair XMS chip reported 0 errors. I sent the RAM off to be replaced and am currently running a fresh install of windows on that XMS chip.

So far so good.

However, I have received 2 BSOD's just recently with the error "PFN_LIST" I did check on this error and it reports either faulty drivers (all drivers are updated) or bad RAM as the source of the problem.

I re-ran memtest and it still reported 0 errors.

I'm deathly afraid that my mobo is going to kill my friend's RAM chip. Is this possible?
Could it be the onboard memory controller of the CPU?
What else could cause this?

Thanks!

 

[jammydodger]

increase the RAM voltage to 2.7, see if that helps. What are your CPU temps? What are your voltage rails like? Use motherboard monitor to find out (do a search on google to get it).

 

[jammydodger]

increase the RAM voltage to 2.7, see if that helps. What are your CPU temps? What are your voltage rails like? Use motherboard monitor to find out (do a search on google to get it).

 

[snotling]

have you checked for blown capacitors?
this kind of "slow deterioration" is often related to bad capacitors.

BIOSTAR M7NCG Athlon XP 2400+
ATI RADEON 9600SE
Adaptec 29160 - Ultra160 SCSI
2 x Quantum Atlas 10K3 18 WLS
WDC WD1200JB-00CRA1
Toshiba 8x4x12x DVD±RW
AOpen Micro-ATX Desktop 200W

 

[oron]

MBM reports:
current - low - high
Case: 39c - 39c - 41c
CPU: 38c - 34c - 39c
Core 0: 2.63v - 2.55v - 2.76v
3.3: 3.25v - 3.25v - 3.26v
5: 4.97v - 4.95v - 4.97v
12: 11.8v - 11.8v - 11.86v
cpu fan: 3183rpm - 3183rpm - 3214rpm

This is under the load of high resolution raytracing (Flamingo raytracer with Rhino 3D). 100% cpu utilization.

 

[oron]

I haven't really been able to see anything. I checked the capacitors and they look fine. What should I be looking for?

 

[snotling]

do they look as fine as those?

<A HREF="http://www.auroracomputer.ca/bad_capacitors.htm" target="_new"> full article </A>
<A HREF="http://www.auroracomputer.ca/images/bad_cap.jpg" target="_new"> Blown capacitor picture </A>

BIOSTAR M7NCG Athlon XP 2400+
ATI RADEON 9600SE
Adaptec 29160 - Ultra160 SCSI
2 x Quantum Atlas 10K3 18 WLS
WDC WD1200JB-00CRA1
Toshiba 8x4x12x DVD±RW
AOpen Micro-ATX Desktop 200W

 

[oron]

Definitely no cracks or leaks. The "bulge" looks pretty subtle in that picture - I'll have to take a closer look at the board.

Thanks

 

[oron]

I haven't physically checked the board yet, but in using the machine for the last 3 hours here's the mbm specs for the RAM:

Core 0 voltage:
current - low - high
2.63v - 2.49v - 2.82v

2.82?! and a low of 2.49? Doesn't that range seem, well, out of range?

Do you think the videocard explosion caused the mobo to not regulate ram voltage very well?

Thanks

 

[oron]

What is more dangerous for RAM - too much or too little voltage? What's an acceptable range?

 

[RichPLS]

To much voltage is far more dangerous to the hardware, to less voltage creates errors. Best to only use as much voltage as it takes for stabile operations.

<font color=red><pre>_____________________________________
And the sign says "You got to have a membership card to get inside" Huh
So I got me a pen and paper And I made up my own little sign</pre><p></font color=red>

 

[jammydodger]

Your RAM is suposed to use 2.6v but urs is fluctuating like crazy.

 

[oron]

I decided to not take a chance and am going to just have the mobo replaced. Sucks, though, to be without a machine for a week (or more).

I pulled the motherboard out and carefully inspected it but couldn't find any physical damage. Doesn't mean there isn't any, though, I suppose...

Thanks for all your help, guys.

I'll tell the MSI people about the voltage fluctuation.