Periodic BSOD with random error dumps

For quite some time we've had a Core 2 Duo machine acting up. It gives me BSOD at random. Nothing in particular seems to trigger it. It'll often freeze up on the internet as well from time to time on sites like Hulu. When it does this a loud buzz comes from the speaker and the screen just freezes completely instead of giving me a BSOD. I've run stress tests on the GPU and RAM and all temps are normal on the system. I'm starting to suspect the MOBO itself. These are the system specs.

http://dl.dropbox.com/u/6392841/PCSpecs.pdf

It's running Windows 7 x64 Pro now.

And the minidumps
http://dl.dropbox.com/u/6392841/CrashDumps.zip

For the first year it was fine, running Windows Vista. We did an upgrade to Win7 and that's when the problems started. I've since blanked it and done a fresh install 3 times and the BSOD problems still persist. I'm hoping someone can spot something that I'm missing here.

Any help is appreciated.

 

Update the graphic card driver with the latest one available for your OS (from the graphic card vendor site)

Right-click over a video in Zulu for instance, select "settings" and them un-tick "Enable hardware acceleration".

Hope it helps!

 

A qualified Microsoft Engineer at Microsoft Answers would help.

 

Resolving stop (blue screen) errors in Windows 7

 

Windows BSOD analysis - A thorough usage guide

 

I've tried updating drivers multiple times and just tried your suggestion to disable hardware acceleration. No luck. BSOD still persists. I've also done as much analysis of the dumps as I can, but they are just too random. If it kept pointing to one particular thing it wouldn't be so bad but it points to something different almost every time.

Since then, I have installed Ubuntu on a separate HDD in the same computer. So far it's been running smoothly for the past 4 days with no kernel panics at all. I thought this was a hardware issue, but if Ubuntu continues to run smoothly it may be a conflict in Windows 7. Unfortunately I'm not sure how to resolve it. All drives have been updated periodically since this started happening several months ago and still nothing has changed. The errors are so random I can only assume it's not a particular hardware driver but maybe a chipset driver or something to do with the mobo in general that is causing this.

 

One time I had BSODs because of problems with RAM memory sticks. To test them you probably have to leave the diagnostic test running over night to catch any errors (at least I did), and then you may have to test the ram cards individually to see which ones are at fault if you have multiple cards. But as you said, it's probably not a hardware issue anyway since Ubuntu works.

If everything else fails you may just have a install a fresh copy of Windows over the BSODing one. (I've had to resort to that several times).

 

Is it a laptop or desktop?
If a desktop, have you tested the power supply? Dr. Power II is pretty sweet.
If it is a laptop, has it been overheating?
Graphics can affect the PCI bus bandwidth.

 

I've run Memtest overnight twice now with no errors shown so I don't think RAM is the issue. It's a desktop PC. I went out of my way when I built it to insure more than adequate cooling. The core temp rarely goes above 36C and GPU temps are in the 76C area when running dual monitors. I haven't yet tested the PSU. Does the GPU temp seem abnormally high at idle? Both monitors are identical and running the exact same resolution and refresh rate. I don't recall what it was before setting up dual monitors. Windows was reporting the temperature lower than this via SpeedFan. nvclock in linux is showing 76c.

 

Try to simply remove the graphic card and the memory sticks and place them again. While you do that check for inflated capacitors in the mobo.

 

I believe I found the problem. It was the video card after all. Turns out that the fan was slowly dying. It finally died completely and that's why the temp is hovering at 76c idle all the time now. I think Ubuntu just handles the hardware better than Win7 and isn't causing the GPU to kick up as much and thus crash. I have a new card on the way so hopefully that'll fix it.

 

Good to know

 

I've got the new card in. It's an EVGA GT 440. Installation went fine. Ubuntu picks it up no problem and works great. Temps are now reading around 30c idle. Windows 7 installed it automatically as well from Windows Update. Everything seems smooth with no BSOD so far. Back to Hulu and even the Weather Channel videos and we still get lockups. Now the screen just freezes indefinitely requiring a hard reset of the system. I disabled hardware acceleration in Flash again and I guess we'll see what happens. I might have to disable hardware acceleration in Firefox completely. Our estimating program no longer seems to crash the system at least. I hope I can get the remaining issues sorted out. If it continues like this it'll be very difficult to diagnose because it never fully crashes to the point that I can get any kind of dump from it.

 

Does he same problem occurs with Ubuntu?

And with Google Chrome?

Try now to update the graphic card driver with the latest one available for your OS, from the graphic card vendor site, and check if you have the latest version of Flash Player x64 installed (maybe others are experiencing the same problem with the x64 version of Flash Player?) Also install the drivers for your mobo, from the vendor site. If you don't know what's your mobo, Speccy might be usefull - hxxp://www.softpedia.com/get/PORTABLE-SOFTWARE/System/System-Info/Speccy-Portable.shtml

 

At this point all driver have been brought up to date. I built the computer so I have all the info I need on it. Ubuntu still runs fine. Windows 7 just BSOD again.

This is the error message:

 

If you google for "ntkrnlmp.exe" you'll find many people complaining precisely about the same problem you're having now...

 

Yeah I've googled every error message I've gotten and there are always a ton of people having the same issue. Never any definitive fixes though. The problem is that seldom is the error the same. I guess I'm basically at the point that I either have to start swapping out part after part to find the issue or just scrap the core of the system and start with a new build, which sucks since the PC is only a couple of years old. What really annoys me is Gigabytes motherboard warranty. It should still be under warranty since their warranty is for 3 years. But it isn't. That three years is from date of manufacture, not from the date I bought it. So I actually had less than two years since the board was evidently older inventory when it was bought.

I just looked as the Asus warranty. It's the same way. So I guess warranties of motherboards don't mean much.

 

The strange thing is that if it is a hardware problem, you should be experiencing problems with Ubuntu either


Is a long shot, but have you tried to scan your HD for errors?
Try to run SFC /SCANNOW Command (as Administrator) and also check your HD for damaged sectors, MHDD is a excellent tool for that.


(download and burn Hirens (hxxp://www.hirensbootcd.org/download/). Boot from the CD, select Hard Disk Tools > MHDD > choose the number that correspond to the HD, press F2 > F4 > F4 to run a full scan.

If your HD is not displayed in MHDD, you might have to go to the BIOS and change from ACHI to compatible.

In Hirens you can also find tools to repair bad sectors in your HD, if found, like HDAT2 and Vivard)

 

Actually yes I have run a scan for bad sectors with MHDD and with the manufacturer's tools (western digital). No errors reported. I agree that it shouldn't be hardware if Ubuntu has been fine. I ran the memory dumps through WhoCrashes and it reports that all errors seem to be driver errors and not hardware problem. But those drivers are always undetermined.

I just went through and made sure all available Gigabyte drivers were installed as the Gigabyte version instead of the most recent, just in case Windows Update might have pulled drivers that I missed. I also installed the latest BIOS, which is a beta, but quite a bit newer than the previous one. I also did a clean install of the NVIDIA drivers. I guess we'll wait and see. If it does BSOD again I hope it keeps pointing to the same error at least. It'll be easier to figure out if I'm not getting different errors each time.

 

The crash dumps say its your video drivers. Update them. Uninstall the old ones then update it

 

I've done this multiple times. Uninstall, reboot, run Driver Sweeper to completely remove them, reboot, cancel Windows Update trying to auto install them, install them directly from NVIDIA. Still no good. However, that was with the old card. Now that the new one is in I've done the same thing and I'll have to wait and see if the problem persists. Thanks for the input though.

Also, is it odd that a reboot is not required after installing the package directly from NVIDIA? Every PC I've ever installed any video driver on has needed a reboot afterwards. Every time I install these though the changes are immediate and no reboot is prompted for. I give it a few minutes and reboot anyway, but it's still odd.

 

Another BSOD over the weekend. Same error message at least. Just not sure what to do next. Maybe I'll check EVGA's website and see what driver they have up for the video card.

 

What driver version were you using?

 

Version 285.62

There are a couple of Beta drivers I can try. I haven't had any BSOD in the past day so I'll wait until it happens again before trying the Beta.

 

Try the newest drivers from the nvidia site or guru3d.com. If your older drivers do it again. Clean install and install the latest ones