Solved BOSD about week after new video card

[Metzler]

The backstory is my old video card a Geforce GT 220 had been over-heating and causing crashes. I replaced it early in the month (I think on the 10th,) with a Gefore 660 ti. It ran fine for a week, but then I started getting the BSOD.

I read that if the video card was bad, the computer wouldn’t crash to a BSOD, rather it would be a black or pixilated screen (as happened with my earlier video card.) However, it does seem to crash more often if I am in a game (but not always.) I am hoping it is just a driver issue. I ran memory test, a hard drive integrity test, windows integrity test, an antivirus test, and so far everything has come back clean. Windows and the video card drivers are updated to most recent. I really have no idea what to do next, so if anyone can help I would be very grateful. Thank you!!!

 

[Northernsoul55]

Hi,

Your latest crash is related to your Nividia Drivers and DirectX issues:

*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************
Use !analyze -v to get detailed debugging information.
BugCheck 3B, {c0000005, fffff880042cff06, fffff8800915c0e0, 0}
Probably caused by : dxgkrnl.sys ( dxgkrnl!DXGDEVICE::Lock+1e6 )
Followup: MachineOwner
---------
3: kd> !analyze -v
*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************
SYSTEM_SERVICE_EXCEPTION (3b)
An exception happened while executing a system service routine.

Relates to: DirectX Graphics Kernel

:ar: http://windows.microsoft.com/en-gb/windows7/diagnosing-basic-problems-with-directx

Also your Graphics driver is causing issues:

fffff880`0915b848 fffff880`049344beUnable to load image 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
nvlddmkm+0x834be
nvlddmkm nvlddmkm.sys Sat Feb 09 23:13:08 2013

Try updating to this version:

:ar: http://www.geforce.com/drivers/results/57096

or auto detect latest:

:ar: http://www.geforce.com/drivers

Stop messages 0x00000023 and 0x00000024 Descriptive text: FAT_FILE_SYSTEM or NTFS_FILE_SYSTEM
Usual cause: The hard disk is fragmented or there is excessive file I/O. This Stop error may also be cause by drive-mirroring or some antivirus software.

Please uninstall Secunia from your system as you have Microsoft Security Essentials installed which is sufficient.

You say you have ran disk check / windows integrity check, please run again after the above has been carried out:

Run SFC to check the integrity of all Windows 7 system files:
Click on the Start > Type in search cmd
Right click cmd at top and Run as Administrator
Type SFC /scannow at the prompt (note the space between SFC and /scannow)

:ar: http://www.sevenforums.com/tutorials/1538-sfc-scannow-command-system-file-checker.html

Also please Run chkdsk /f/r in an elevated command prompt.

Click on the Start > Type in search cmd
Right click cmd at top and Run as Administrator
At the command prompt please type: chkdsk /f/r
(Note: there is a space between chkdsk and /f/r)

:ar: http://www.sevenforums.com/tutorials/433-disk-check.html

Also it wont harm to Defragment your hard drive.

:ar: http://www.sevenforums.com/tutorials/11733-disk-defragmenter-open-use.html

Please post back any results / observations

cheers
Dave

 

[Metzler]

Thanks! So I:

1. verified direct X (no problems detected)

2. I already have most recent video card driver (314.07)

3. Uninstalled Secunia (goodnight sweet prince! haha)

4. Ran windows 7 integrity (clean)

5. Ran chkdsk, results are:
[FONT=&quot]
CHKDSK is verifying files (stage 1 of 5)...[/FONT]
[FONT=&quot] 563712 file records processed. File verification completed.[/FONT]
[FONT=&quot] 1920 large file records processed. 0 bad file records processed. 0 EA records processed. 93 reparse records processed. CHKDSK is verifying indexes (stage 2 of 5)...[/FONT]
[FONT=&quot] 690306 index entries processed. Index verification completed.[/FONT]
[FONT=&quot] 0 unindexed files scanned. 0 unindexed files recovered. CHKDSK is verifying security descriptors (stage 3 of 5)...[/FONT]
[FONT=&quot] 563712 file SDs/SIDs processed. Cleaning up 72 unused index entries from index $SII of file 0x9.[/FONT]
[FONT=&quot]Cleaning up 72 unused index entries from index $SDH of file 0x9.[/FONT]
[FONT=&quot]Cleaning up 72 unused security descriptors.[/FONT]
[FONT=&quot]Security descriptor verification completed.[/FONT]
[FONT=&quot] 63298 data files processed. CHKDSK is verifying Usn Journal...[/FONT]
[FONT=&quot] 36468160 USN bytes processed. Usn Journal verification completed.[/FONT]
[FONT=&quot]CHKDSK is verifying file data (stage 4 of 5)...[/FONT]
[FONT=&quot] 563696 files processed. File data verification completed.[/FONT]
[FONT=&quot]CHKDSK is verifying free space (stage 5 of 5)...[/FONT]
[FONT=&quot] 25429531 free clusters processed. Free space verification is complete.[/FONT]
[FONT=&quot]Windows has checked the file system and found no problems.[/FONT]

Is the “[FONT=&quot]Cleaning up 72 unused index entries from index $SII of file 0x9”. anything to worry about? I suspect no, but wanted to check. [/FONT]

[FONT=&quot]6. Disks are only 2%, and 6% fragmented (C: and system reserve respectively,) but I will defrag tonight. [/FONT]

[FONT=&quot]Fingers crossed, and I will keep you posted! [/FONT]

 

[Northernsoul55]

Thanks for the feedback.

If you have the latest Nvidia drivers installed can you roll back the driver to the previous one,
as your system will crash again with the latest driver installed?

:ar: http://www.vistax64.com/tutorials/165382-device-manager-roll-back-driver.html

If you are unable to roll back or rolled back driver causes issues, please install this driver:

Download and install
:ar: Driver Fusion.
Reboot the computer in
:ar: Advanced Boot Options, safe mode. Search Driver Fusion in your start menu,
and remove all components of your nVidia display driver.
If it says the the free version of Driver Fusion cannot remove all the elements, stop there and follow
:ar:Drivers - Clean Left over Files after Uninstalling
Boot normally now. Go to:
:ar:Drivers - Download NVIDIA Drivers, Drivers > Beta and Legacy; search there with your cards particulars for:
:ar:GeForce 306.23 Driver, dated 13.9.2012 and install it.

Cheers

Dave

 

[Metzler]

I haven't had any more problems so I'm going to leave it be for now. But if it comes back I will follow your instructions above. Thanks again!

 

[Northernsoul55]

No problem, glad to help,
Monitor the situation for a week and post back any feedback.

Cheers

Dave

 

[Metzler]

Well unfortunately the problem is back. Starting last night, it crashes 2-3 minutes into a game (both in Starcraft 2 and DOTA 2.) The strange thing is I used the computer heavily on Sunday with no problems, including both those games. I can't think of anything that changed on the computer from Sunday to Monday. I will roll back the drivers and try 306.23 as you recommend. Updated logs are attached.

 

[Northernsoul55]

Hi,

Unfortunately your new dump files appear corrupt and are unusable, but I suspect same issue as the same Nvidia
drivers are installed that were initially causing issues.

Please post back once you have rolled back the driver as suggested and see if your crashes have stopped.
If the issue remains the same then use the stable 306.23 driver.

cheers

Dave

 

[Metzler]

So before I rolled back the drivers I had another BSOD, and this one gave some useable info, it was dxgmms1, which a quick Google search looks like it is direct X related. (minidump attached.)

I tried rolling back drivers, but computer crashed, so I just took them all out and installed 306.23 using driver sweeper. Jumped in a game for like 20 minutes and it looked fine, but I'm sure its too soon to tell for sure yet. Will keep you posted!

 

[Metzler]

Darn, had another crash with the 306.23 driver. minidump file attached, looks to be similar error.

 

[Northernsoul55]

Well it does not appear the Nividia Driver is a fault, the error is pointing toward DirectX again:

Stop message 0x00000050 Descriptive text: PAGE_FAULT_IN_NONPAGED_AREA
Usual cause: Requested data is not found in memory. This might be caused by a printer driver or remote access driver.

*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************
Use !analyze -v to get detailed debugging information.
BugCheck 50, {fffff8a0131e72e0, 0, fffff88004223d78, 2}

Could not read faulting driver name
Probably caused by : dxgmms1.sys ( dxgmms1!VIDMM_GLOBAL::IsPagingOperationPending+4c )
Followup: MachineOwner
---------
0: kd> !analyze -v
*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************
PAGE_FAULT_IN_NONPAGED_AREA (50)
Invalid system memory was referenced. This cannot be protected by try-except,it must be protected by a Probe. Typically the address is just plain bad or it is pointing at freed memory.

DirectX Graphics MMS

Attempt to update DirectX if it will allow you to do so, either by Windows updates or from the following link:

:ar: http://www.microsoft.com/en-gb/download/details.aspx?id=17431

If that fails try the Windows hotfix found here:

:ar: http://support.microsoft.com/kb/974431

There might be a possibility of bad memory.
Take memtest. Run for 8 passes and test each stick in a know good slot for an additional 6 passes.

:ar: http://www.sevenforums.com/tutorials/105647-ram-test-memtest86.html

Please post back any progress / observations.

Cheers

Dave

 

[Metzler]

The direct X link above just looks to be a power point and audio presentation. I checked online myself though, and don't see a update or repair module for it. The Windows hotfix also said it is not applicable when i tried to install it, (I don't know if I'm already up to date.) I left the Memtest on overnight and had no errors after 11 passes.

 

[Metzler]

Another thought I had, this video card comes with ASUS "gpu tweet" which is overclocking software. I figure overclocking has to be bad in a situation like this. I havn't messed with those setting yet, but i'm not sure if it came factory overclocked. It has a GPU Boost clock, Min GPU VOltage, Memory Clock, and Power target %. I figure I would turn the GPU Boost clock and Memory Clock all the way down right? Note that they have an "offset" number. So for GPU boost, if I turn it all the way down it lists an offset of -221. If I have zero offset, the boost is about halfway on the screen. So I'm not sure if I would actually be underclocking if I turned it all the way down.

video card is GTX 660 TI


EDIT:

Another thought. About 3 months ago, I bought 16 gigs RAM, and swapped them out. They didn't work because the motherboard could only take 8 max. So I returned them and put my 4 gigs back in. Is it possible this could have changed the motherboard ram settings, something I would have to set back for the 4 gigs? Thanks!

 

[Northernsoul55]

Hi,

Please download and run Speccy to monitor your temps and post back a screenshot.
:ar: http://www.piriform.com/speccy
Carry out a Video Card - Stress Test with Furmark:
:ar: http://www.sevenforums.com/tutorials/100356-video-card-stress-test-furmark.html

Also please download and run Prime95 to further test hardware:

:ar: http://www.sevenforums.com/tutorials/100352-hardware-stress-test-prime95.html

Post back any results

cheers

Dave

 

[Metzler]

Hey Northern, btw thanks so much for your continued assistance with this.

So I already have CPUID HW Monitor, which looks to be similar to Speccy, so I skipped that step, but let me know if you want me to go back.


I ran furmark for 10 minutes with no problems. It stabilized pretty quickly at 75 Celsius, and I had no “artifacts” ("artifacing?" Lol )

However, when I run Prime 95, my computer freezes within the first minute. By freeze I mean that my computer monitor stays the same, but my mouse locks, and Ctrl-alt-delete does nothing (had to manually restart.) I'm not sure what this means.


For Furmark I ran burn in, xtreme burn-in, and post FX
For Prime 95 I ran Small FFTs, and 4 torture test threads.

 

[Metzler]

New BSOD and new error, this time ntkrnlmp.exe

 

[Metzler]

Another BSOD, this one is memory_corruption. Does that mean RAM? Maybe I shoudl rerun the memtest?

 

[Northernsoul55]

Hi,

From your latest crash dumps:

*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************
Use !analyze -v to get detailed debugging information.
BugCheck 50, {fffff8a002f3a134, 0, fffff8000356330e, 2}

Could not read faulting driver name
Probably caused by : ntkrnlmp.exe ( nt!CmpKcbCacheLookup+1de )
*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************
Use !analyze -v to get detailed debugging information.
BugCheck A, {51, 2, 1, fffff800032b7821}
Probably caused by : memory_corruption
Followup: memory_corruption
---------
2: kd> !analyze -v
*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************
IRQL_NOT_LESS_OR_EQUAL (a)
An attempt was made to access a pageable (or completely invalid) address at an interrupt request level (IRQL) that is too high. This is usually caused by drivers using improper addresses.

The analysis was somewhat inconclusive for your latest crashes.

Run Driver Verifier to scan for any corrupted drivers which may be causing problems, this program works by running various stress tests on drivers, in order to produce a BSOD which will locate the driver; run for least 24 hours:

:ar: http://www.sevenforums.com/tutorials/101379-driver-verifier-enable-disable.html

   Tip

Do this after you have carried out my other suggestions below as the test may cause lagging/slow computer

Please ensure you Reset GPU Tweak to run at factory default settings.
Any overclocking will cause the system to become less stable.

:ar: http://www.geforce.co.uk/hardware/desktop-gpus/geforce-gtx-660ti/specifications

Uninstall SuperAntispyware from your system.

Please update your BIOS as it is outdated:

[BIOS Information (Type 0) - Length 24 - Handle 0000h]
Vendor American Megatrends Inc.
BIOS Version 0503
BIOS Starting Address Segment f000
BIOS Release Date 01/05/2010
BIOS ROM Size 100000

A newer stable version can be found here:

:ar: ASUSTeK Computer Inc. -Support- Drivers and Download M4A78LT-M LE

   Warning

Please follow the manufacturer's instructions carefully when updating the BIOS, as incorrectly updated, may render your computer unusable!!

What is your PSU Power rating?

Look forward to your reply

Cheers
Dave

 

[Metzler]

Thanks, Power Supply is 500 W. The unit ID # is rs-500-pca-d3. Yeah I was wondering about that. I can see the voltage numbers of the MOBO and GPU on CPUID Hardware monitor, but I don't know how to read the data to see if I'm near the limit.
eXtreme Power Plus 500W - Cooler Master - Leading Provider of Computer Case, Cooler, Power Supply, Laptop Cooling, Gaming Peripherals

Uninstalled AdAware and updated BIOS. I will check GPU overclock and run driver verifier now and report back.

 

[Northernsoul55]

Hi,

From your GPU specs it seems your 500W PSU just meets above the minimum:
View attachment 261837

Cheers

Dave