116 BSOD, even after replacing graphics card and reinstalling drivers

[subzerogts]

Here's my original thread at MS forums:

Lots of random BSODs, trying to figure out what is going bad - Microsoft Community

There are links to dump files and bluescreenview screenshots there and a description/history of the problem.

Turns out my old GTX 680 did have some issues which were made clear when it crashed during benchmark on my second system (which has no issues otherwise). So I bought a new reference GTX 970 and it ran great on the second system. Now after coming back to the first system, it's still crashing!

Here's the info from the utility in the sticky in this forum:

https://onedrive.live.com/redir?resid=1B39E0EE98653070!269&authkey=!AIs8xDken4FCp4Y&ithint=file,zip

EDIT: Attached file to this post as well.

Anything in July or after is with the new GTX 970.

I'm beginning to think it either really is a PSU issue (as per original thread, I tested the PSU with a digital voltmeter and it was fine, but I did not test it under load), some freak motherboard issue (but it didn't crash once while using onboard intel graphics), or driver issue (looks like maybe directx?).

I'm waiting to upgrade to Windows 10 to see if the DX12 will end the crashes. My friend had BSOD issues on his new build and upgrading to Win 10 insider preview fixed it. I'm reluctant to upgrade to the insider preview, but if it keeps crashing to the point of being unusable, I might try it, barring better advice obtained here.

Thanks for any help you can provide, this has driven me crazy and I've already spent $350 trying to fix it!

The problem suddenly appeared in May 2015 out of nowhere (system was built/installed in summer of 2011). It is not overclocked anymore, and to be doubly sure, I reset BIOS to defaults today and it still crashed (actually, it just plain shut off!).

System:

Maximus IV Gene-Z Micro-ATX
Core i5 2500k
16gb ram
gtx 970
wd 1tb with 128gb sandisk ssd for intel bios caching (64gb set for caching)
Windows 7 x64
corsair ax750 psu from summer 2011

EDIT: Note, the GTX 970 has not crashed while gaming once. These were all from Windows desktop/Google chrome/VLC/whatever

 

[subzerogts]

Oh and I ran SFC /scannow and stuff. When I tried to do it specifically on some of those directx files (mms), it said it couldn't do it. Even in safe mode.

System is stable in Safe Mode too.

I've also tried various Nvidia drivers, using driver sweeper/uninstaller (DDU one) each time. Including before installing the latest drivers when I put in the new video card.

 

[Boozad]

Can you upload the files directly to the site please. http://www.sevenforums.com/tutorials/9733-screenshots-files-upload-post-seven-forums.html

 

[subzerogts]

Ok, I've attached it to this post.

 

[subzerogts]

Hi, can someone take a look at the logs and tell me what they think?

 

[Boozad]

I'll get round to this tonight, bear with me please I apologise for the delay.

 

[subzerogts]

No problem, thanks for taking the time!

 

[Boozad]

I've just looked at the logs and there are a couple of important files missing that we use to get the full picture. Could you run the tool again please and upload the resulting .zip file. Don't worry, sometimes this can take two or three times to capture everything.

 

[subzerogts]

Okay, how's this

 

[Boozad]

Still some files missing, give it another try please mate.

 

[subzerogts]

Ok, third try

 

[Boozad]

That's got everything. I'll start taking a look now, bear with me and I'll get back to you.

 

[Boozad]

Right, every single dump is a 0x116 VIDEO_TDR_ERROR (Timeout Detection Recovery). These point to a problem with the GPU whether it's driver related, power to the card, overheating or the PCI slot among other things. I'm presuming your temperatures are fine and that you've tried clean installs of different drivers? Given that this has happened on two cards now I'd be leaning towards the PCI slot or PSU. Do you have a spare PCI slot you can try the GPU in?

 

[subzerogts]

Okay, wow. My computer just flat out died. It powered off like it has the last time I mentioned, but when I press the power button, the power supply just makes a clicking sound.

Here's a video:

https://drive.google.com/file/d/0B-XBmvmgdkfVVjZvbHNiZ2lwQzA/view?usp=sharing

So... maybe it was a PSU issue after all?

I'm thinking I should buy a PSU first and then if the system starts up, great. If not, I continue by buying another motherboard/CPU?

Btw, yes, I tried clean installs and temperatures are fine. I even switched to a blower style reference card second time around which stays cool and quiet. Unfortunately it only has one PCIE x16 slot I think. I'm going to double check.

 

[subzerogts]

Okay there's a second PCIE slot, but there's no room since I have a Sound Blaster Z in there and the system won't even turn on now. So first order of business is PSU I assume?

Before I do anything else I'm going to test that PSU on the voltmeter again.

 

[subzerogts]

Oh and the motherboard (and attached USB accessories) lights up when the PSU switch is on, the little start button is on and everything. It's just that the clicking starts and the system won't boot.

 

[subzerogts]

Okay, what the hell. I took out the PSU to test it, and it seemed to be okay. Voltages were normal. So I plugged it back into the motherboard and now the system starts up! I'm at a loss here...

I'm going to leave it on a while and see what happens... Could it have been a short in the case or something? Because the motherboard tray is now outside of the case and all the case fans are unplugged (only CPU fan is plugged in). The case is very old and dusty (even though I use an air duster, there's still a layer of dust everywhere).

I don't see why a possible short in the case would cause a 116 TDR error over and over though?

 

[subzerogts]

I am at wit's end here. I put the system into a new case, a Corsair Carbide 200R. Very roomy. Made sure motherboard was nice and secure. I turn it on and several hours later, it crashes with another 116 error after almost one week without any problems with the motherboard outside of the case.

I have a week long trip which makes this frustrating to deal with. When I come back I'll try unplugging the front panel USB headers and see if that resolves it. If not, I'll try unplugging the case fans. If not, I'll try running it with the case laying flat on its side. Last scenario, I'll use the second PCIE slot, although that means I'll need a new motherboard/CPU anyway in order to fit my sound card which I need.

Any other ideas?

 

[subzerogts]

I was told by the tech at the store that the front panel USB 3.0 headers can sometimes cause problems with the motherboard. Has that been true in anyone else's experience too? It's strange that it would keep causing a 116 BSOD.

Or that this would happen out of nowhere. I mean at this point it's looking like the motherboard's having issues with the PCIE slot on its own, perhaps due to age (2011). It's under warranty but I don't have the time/patience to follow through on that.