Solved BSOD Apparantly random

[essenbe]

You're welcome, as always Arc.

 

[Layback Bear]

Just some things I noticed in Speccy. They may or may not be causing a problem. I'm not sure how any of them would cause a video card problem.

1. Bitdefender and Windows Defender are both running.

2. You are also using AVG Tune up

3. You are using Team Viewer at this time.

3. You have a Asus P9X79-WS Quad channel with 3 of the 8 ram slots used. It should be 2, 4, 6, or 8 slots used.

4. Your ssd in hook to Sata II at 3.0 gb. instead of Sata III at 6.0 gb.

With all that information we still need to know what power supply you are using.
You state that it is a custom build so if you could put your power supply in your system specs.

You need at a minimum a 550W quality power supply with two 6 pin connectors for the video card.

 

[Owen27]

Icit2lol, when I clicked the Yellow button in your recommended program I get this:

I assume disabling it would be a good idea?

I think this is what you wanted to see:

Layback Bear, I've got a 750w power supply from Cooler Master. Also, the RAM setup isn't causing the issue, I only took one of my RAM sticks out to see if it would stop causing this issue, so it was happening before hand.

Thanks for the help guys.

 

[essenbe]

have you had any more BSODs? If you have upload the same file again. Did you reinstall the driver like the tutorial said and run Furmark again? Set the custom fan profile?

 

[Owen27]

I installed the driver as recommended in the tutorial. As soon as I went to run Furmark, Blue screen.

Rebooted and ran it again. Got to the 13 Minute mark and it was at 89 Degrees. Went to take a picture and Blue Screened again.

Dump files attached.

 

[essenbe]

Same thing again Video TDR

BugCheck 116, {fffffa8013fc4010, fffff8800ff7851c, ffffffffc000009a, 4}
*** 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+91951c )
Followup: MachineOwner
 
nvlddmkm
start             end                 module name
fffff880`0f65f000 fffff880`10333000   nvlddmkm T (no symbols)           
    Loaded symbol image file: nvlddmkm.sys
    Image path: \SystemRoot\system32\DRIVERS\nvlddmkm.sys
    Image name: nvlddmkm.sys
    Timestamp:        Sat Sep 13 14:59:50 2014 (5414A236)
    CheckSum:         00C8E3A6
    ImageSize:        00CD4000
    Translations:     0000.04b0 0000.04e4 0409.04b0 0409.04e4

I'm not really sure reinstalling the drivers would help or not. If you do, try a different driver and use the same procedure. http://www.sevenforums.com/tutorials/327414-nvidia-drivers-avoid-problems.html

Also, you may check and see if you have the latest DirectX How to install the latest version of DirectX

Also please go to device manager and expand graphics adapters and post a screenshot of the wholw device manager. This will tell you how to post a screenshot.
http://www.sevenforums.com/tutorials/9733-screenshots-files-upload-post-seven-forums.html

 

[Owen27]

I assumed you meant expand Display Adapters in the Device Manager.

I tried using Arc's idea again, installing the older driver. While this does seem to make the crashes less frequent, it still happened.

Already have the latest DX.

 

[essenbe]

Unless you are overheating, the problem is your Card, Motherboard or PSU. You may try putting it in a different slot on your motherboard, just to see.

 

[Arc]

I installed the driver as recommended in the tutorial. As soon as I went to run Furmark, Blue screen.

Rebooted and ran it again. Got to the 13 Minute mark and it was at 89 Degrees. Went to take a picture and Blue Screened again.

Dump files attached.

Though it is not an even regularity, but it is there. BSOD during Furmark. That is the point.

Your GPU is failing to handle a stress. Sometimes it cannot, sometimes it can, but not for a long time, and sometimes it manages the stress for a fairly good time. Overall, it has started failing.

Still, a good firs step will be to try the GPU in a different PCIe slot, as essenbe suggested. You have alternative slots ....

View attachment 337564​

If the issue continues even in the other slot, you will have to apply a different card, at least on a test basis.

 

[Owen27]

I hoovered out the PC. (Paying particular attention to in and around the graphics card)

I also pushed it in, to ensure it was fully connected. No crashes since then.

It's still early days and it could very well happen again. At the very least it seems to of improved the stability. Or it could be a coincidence.

 

[essenbe]

Let us know how it does.

 

[Owen27]

I can confirm that after a week there has been no more crashes. Never underestimate the power of a good cleaning, I guess.

Thanks for all the help everyone, very much appreciated.

 

[essenbe]

You're welcome, and I'm glad you have it running well. Good job.

 

[Arc]

Great news!

to the team! essenbe, Owen27, Layback Bear and ICit2lol!

 

[ICIT2LOL]

Sorry really late reply Owen have been off doing stuff (tax man etc ) Glad it has got fixed and the readout for the PSU was spot on no worries there.

As for the actual cleaning a lot of people don't like vacuuming but it is no different in my mind than blowing air across the board or whatever it still causes friction either way and I made up a small bore tube fitting for my el cheapo vacuum cleaner to reach down into the very narrow spots in and around the different components.

It is very easy to make up all one needs is a length of clear poly tube say about 1.5 metres long and 3/4 inch diameter or smaller and some duct tape to make a seal to shove in the end of the vacuum main tubing then it is so easy to clean the machine. If I was at home I could send a pic to show you but may do if you like by message later on.