Random BSODS, Particularly when playing Crysis Warhead.

[kjclark]

I just recently built my system and have been experiencing several annoying BSODs, particularly when playing Crysis Warhead. Sometimes I can play the game for several hours without problems, and sometimes running the games for 5 minutes results in a SYSTEM_SERVICE_EXCEPTION error. I also experienced one BSOD from my desktop when changing my background.

Here are my system specs:
Asus P6X58D-E motherboard
G.Skill 6GB DDR3 1600Mhz RAM
2 x EVGA GTX 460s via SLI
Intel i7-950 Processor
XFX 750W Black Edition PSU
SAMSUNG Spinpoint f3 1TB HDD

 

[TorrentG]

Hey there.

Visit the manufacturer of your modem to download and install the latest driver to update this old one:

ALSysIO64 ALSysIO64.sys Sun Apr 19 15:41:04 2009

----------

Update the Marvell Yukon network driver by installing from my attachment.

yk62x64 yk62x64.sys Wed May 20 04:14:06 2009

---------

Install from the following link to update you Realtek HD Audio driver:

RTKVHD64 RTKVHD64.sys Fri Jan 29 01:48:35 2010

RtkHDAudio

-----------

That should fix it up. If you have further crashes, you can post the new crash dumps.

Enjoy.

 

[kjclark]

thanks for the reply. I'm not sure how to update that Marvell Yukon Driver. The file you sent me does not contain any executable.\
EDIT: derp, it has a readme. sorry.

 

[kjclark]

just had another blue screen (System_service_exception) while playing Bad Company 2.

 

[TorrentG]

Update the NEC USB 3.0 driver from motherboard manufacturer's website:

nusb3hub nusb3hub.sys Thu Jan 21 22:22:18 2010

If crashes persist, test the memory overnight with Memtest86+ as it is likely faulty. Replace RAM if any errors show.

 

[kjclark]

Ok, so I just had this Bsod. I have tried use Memtest86 several times but the system always freezes around 7% and the clock stops. Any idea why this is? Should i try to remove sticks one by one or something? I'm not sure about that method because Bsods occur pretty randomly.

 

[TorrentG]

This latest crash is with the video card driver and the driver itself would simply never cause this.

nvlddmkm nvlddmkm.sys Fri Jul 09 17:15:58 2010

Use !analyze -v to get detailed debugging information.

BugCheck 1000007E, {ffffffffc0000046, fffff800036aa034, fffff8800645a3b0, fffff8800645a450}

Unable to load image \SystemRoot\system32\DRIVERS\nvlddmkm.sys, Win32 error 0n2
*** WARNING: Unable to verify timestamp for nvlddmkm.sys
*** ERROR: Module load completed but symbols could not be loaded for nvlddmkm.sys
Probably caused by : nvlddmkm.sys ( nvlddmkm+24732 )

Followup: MachineOwner
---------

Make sure optimized defaults are set in the motherboard bios but other than that, this confirms that there really is a RAM problem.

Not sure why it's freezing when running Memtest but that also shows there is a hardware problem and not with Windows and/or its installed drivers.

I don't want to make you waste money, but I highly suggest replacing RAM if optimized defaults continue to give problems. Of course, that decision is yours to make though.

 

[kjclark]

ok, I realize that my RAM is probably the problem, but I just ran Memtest86 (v2.11 instead of 4.1) for 8 hours and it returned no errors. Just set my bios settings to XMP so that my RAM will run at 1600MHz (it was previously running at 1033). If I blue screen again I will probably have to replace my memory, but I probably wont be able to afford it.

 

[kjclark]

Perhaps I should have phrased this differently. If I ran Memtest86+ for 8 hours and encountered no errors, is it still likely that the problem was with the RAM?

 

[TorrentG]

Not necessarily...many times, modules need to be tested one at a time with others removed, to show errors.

If you have any new crash dumps, you're also welcome to post them.

I would use Driver Sweeper in safe mode to clear all NVIDIA video files and settings. Reboot, install the latest driver from NVIDIA's site again, and don't change any CCC settings at all.

 

[kjclark]

Ok, I have experienced two more BSODs since I last posted. I will now reinstall the nVidia drivers and test my RAM sticks separately. The dump files are attached.

 

[TorrentG]

Both of the latest crash dumps involve video; one directly with the video driver and another 0x3b stop which involves video data and the kernel code.

I noticed you're running SLI....do you have only one bridge attached instead of two? If 2 try running with only 1 connected. Otherwise, perhaps the clean install of the video driver will help. Your psu seems to be sufficient.

You'd also do well to update this NEC USB 3.0 driver from the motherboard website:

nusb3hub nusb3hub.sys Thu Jan 21 22:22:18 2010

 

[kjclark]

I'm not sure what you mean by one/two bridges. The cards are attached by a kevlar bridge that came with the motherboard, but thats it.

 

[TorrentG]

Do you happen to have any SLI bridges that came with one of the cards. Very doubtful using that instead will make any difference but I guess worth a try.

Also make sure heat isn't any type of issue.

Or take one card out and run like that for purely temporary testing purposes to see if bsods occur.

Also continue testing RAM sticks separately if that's what you're doing already.

 

[kjclark]

Just got another one, this time Memory Management error. I haven't tried taking one card out yet, but this seems to confirm that it is a memory problem.

 

[kjclark]

After testing each memory stick individually, I have received no errors on any of them. However, when running applications with only 1 stick in, BSODS do not occur (as of yet) I wonder whether a memory slot on the motherboard might be the problem

 

[clifford_cooley]

I'm wondering if there might be a problem with the memory being paired together causing the problem.

Do you have an option in the BIOS to run the memory in single or dual channel when both modules are installed? If so you might consider running the modules in a single channel mode for a little while. Who knows, that could change how the memory reacts in the system.