Random BSOD while gaming or idle

[xTko]

Usually it happens when I am playing a game (BF3, CS:GO, GW2) but sometimes it happens while idling or while I am browsing the internet. I have tried to resolve the issue myself but without success so here is everything I have done recently to fix the issue.

Reformat did not work.
Default BIOS settings did not work (no overclocking anyways)
Tweaked BIOS ram settings to XMP which made the issue happen less frequent. (went about a two weeks with no restarts)
I know that voltage readings from BIOS aren't very reliable but the 3.3v is running around 2.8-2.9v for the past couple of months. I tested the PSU with a multimeter and everything was OK.
I switched out my GTX570 for a spare 9800GT+ and did not get a restart but with how random they are I can't blame the video card definitively.

I am at a loss for ideas. Any help would be appreciated. If you need anymore info let me know.

I thought the issue had been resolved previously in:
http://www.sevenforums.com/crashes-...ile-playing-any-game-no-bsod.html#post1674860

Sorry for the repost but I was unable to get help in the previous topic.

Thanks in advance.

Regards,

xTko

 

[carwiz]

The last couple of dumps indicate a timeout after an attempt to reset the video dirver. Are you sure you have the correct drivers for your OS and card?

Loaded symbol image file: nvlddmkm.sys
    Image path: \SystemRoot\system32\DRIVERS\nvlddmkm.sys
    Image name: nvlddmkm.sys
    Timestamp:        [B]Tue Oct 02 13:21:13 2012[/B] (506B3099)

 

[xTko]

I was using 306.97 GeForce Drivers for my GeForce 570GTX. nVidia says that this is the driver for my GPU.

Updating to 310.90 tonight.

 

[xTko]

The issue still persists after the driver update. Any other suggestions?

 

[carwiz]

Was it a Bug Check Code 116 again? (TDR Timeout)

If so, try disabling the on-board HD3000 in the BIOS for the time being. I know you're supposed to be able to run both the HD3000 and a video card but some have seen problems running both. Also, if you are OCing the video card or the processor, set them to their default parameters.

If you still see the 116 Bug Check, provide a new dump.

 

[xTko]

I have disable the on-board HD3000. I set the BIOS to default recently so there should be no overclocking.

I will see if I can get it to crash again. If I see the 116 Bug I will post a new dump as you requested.

Thanks for the help.

 

[xTko]

Well...

Well that did not take long. Still receiving the Bug Check 116.

Here are the new dump files:

 

[carwiz]

Loaded symbol image file:[B] nvlddmkm.sys
[/B]    Image path: \SystemRoot\system32\DRIVERS\nvlddmkm.sys
    Image name: nvlddmkm.sys
    Timestamp:        Tue [B]Oct 02 13:21:13 2012[/B] (506B3099

Driver Name: nvd3dumx.dll,nvwgf2umx.dll,nvwgf2umx.dll,nvd3dum,nvwgf2um,nvwgf2um
Driver File Version: 9.18.0013.1090 (English)
     Driver Version: 9.18.13.1090
        DDI Version: 11
       Driver Model: WDDM 1.1
  Driver Attributes: Final Retail
   Driver Date/Size: [B]12/29/2012[/B] 02:34:47, 18054312 bytes

Still the same TDR error too.

There's a discrepancy between the version dates on the drivers. Not totally out of the realm of reality but still suspect. Also, the video card is not responding to a reset. There's a couple of things to try.

Shutdown the system completely (power down) and re-seat the video card. While you're in there, check for a build up of dust and clean it. Check all the intake vents too and clean.

It could be that the card is over heating and shutting down. It may be a good idea to run a temperature monitor for a while and log the temps until a shutdown.

Constant TDRs could be a sign of a dieing video card too but try the above for now--They're cheap.

 

[xTko]

What is the best way to remove all the graphics drivers and ensure only the newest drivers are reinstalled?

I just completed a full system clean less than a month ago, but I will re-seat the card and ensure that it is still clean.

I will keep a temp monitor up and see if that is an issue.

 

[xTko]

Update:

It does not seem to be an issue of overheating. At idle the temp is sitting at about 40. I loaded up a game and the temp went to about 50 and the PC crashes within 5-10 min before I can push the card.

 

[carwiz]

Is that CPU temps or video card temps?

Except for a Nvidia daemon (possibly some type of auto update) it appears the drivers are the 310.90 release.
I'm drawing a blank except for Spybot.

While looking at your system in detail, I did see that you are running Spybot with the Tea Timer turned on. That's a no-no for Windows 7. It could have prevented critical updates to the Registry. It will most likely slow down most modern AV monitors too. You also have SD helper turned on. This is not necessary and is probably incompatible with your ESET AV. You can use Spybot to keep your Hosts file updated but I advise against running the resident portions.

Start the Spybot program then navigate to Tools > Resident. Un-check both boxes, exit then reboot.

Run SFC /scannow from an elevated command prompt. Report back on console errors.

 

[xTko]

That is GPU temps. CPU temps are at ~20-27 at idle. Again I can't check my temps under load because it always crashes.

I uninstalled SpyBot and removed any traces I could find of it.

Ran SFC /scannow: No integrity violations found.

Since SpyBot could have blocked some of my registry updates should I reinstall certain drivers or reformat?

 

[carwiz]

I use Spybot to keep my Hosts file updated but that's all. Updates usually come out on Tuesday or Wednesday so I run it once a week just to get updates then "Immunize". That's what updates the Hosts file. It's called Passive Protection". It stops a lot of junk from even being read.

The scan will attempt to repair Windows standard registry entries but I don't think it knows about any type of third party driver parameters other than their registration. One way to check would be to open the graphics properties and make sure all the parameters are set to match your display. If any Reg entries are missing, it should create them if not a default option.

The next item to check is memory. Fire up Memtest86+ and let it run over night or until you start getting a lot of errors if you're watching it. You should run it for at least 6 or 7 passes and it could take up to an hour per pass.

Forgot to mention: The temps look good.

 

[xTko]

That's good information to know about SpyBot, thanks.

I will run a memory test tonight and see what happens.

On a side note. Since your initial guess was my video card I decided to replace my GTX570 with my 9800GTX+ from my old build. The issue seems to have stopped but I cannot say for certain without running it for a longer period of time. While using my GTX570 I would get a restart within 1-10 min of starting a game and sometimes randomly without even loading a game, just running videos in flash. While I am using the 9800GTX+ I have yet to receive a random restart. I will use this card for the next couple of days to see if I get any errors. If I do not receive any errors would you agree that the video card is most likely the issue?

 

[carwiz]

Hardware can certainly be an issue. The TDR error is caused by the driver not responding back to Windows after it issued a reset to the video card. If the card doesn't respond, the driver lags telling the OS it's complete so Windows issues the TDR (timeout). The first choice it to make sure the drivers are compatible and up to date since this is usually the cause. Since yours are, we start looking at hardware issues like dirty slots and/or flaky memory. These are cheap tests before condemning a piece a hardware though. Having a spare card sure makes it easier.

Keep us up to date on the outcome.

 

[xTko]

8 Passes 0 Errors on the Memtest.

No restarts last night with the 9800GTX+, so we will see how tonight goes.

 

[carwiz]

Great so far.

 

[xTko]

Almost 2 full weeks without a restart while running my 9800GTX+ but there is a new symptom.

When I start my PC I am receiving an anti-surge warning from my BIOS every time I boot my computer.

Could it be my PSU is the issue and not the video card? My GTX570 does require two 6-PIN connectors while my 9800GTX+ only requires one.