You are not getting the STOP 0x116 error, but the steps in STOP 0x116: VIDEO_TDR_ERROR troubleshooting are very helpful for determining display card issues and may tell us if there is an incompatibility between the card and motherboard or a problem with the card itself (which is unlikely since it was working on another board unless it was damaged in some way during transfer [ESD]).
Not much luck in there so far though I'll keep going through the checklist. Furmark ran its Burn-In test for an hour straight with no issues.
The latest batch of crashes:
Attachment 192771
BlueScreenView is also mentioning something called ntoskrnl.exe. I have no idea what that is though.
Both recent crashes\bsods points to gpu.
First read carefully this tutorial then run NVIDIA/AMD Video Card - Test with OCCT
Occt automatically create a folder with screenshots\results in your documents folder.
Let the test\stress finish,zip that folder and upload it here.
During the gpu test\stress have always in mind the temperature,if the temperature of your GPU exceeds 80C, stop the test.
OCCT isn't giving me any output files after testing (unless it's hiding them somewhere bizarre), and the monitoring function doesn't seem to work at all, even after installing Fanspeed. I assume I'm just doing something wrong but couldn't finding anything I missed in the tutorial.
I let the test run for an hour anyway and it reported no errors, but I can't find any output files to upload.
Last edited by EffinCraig; 12 Jan 2012 at 21:34.
I prefer HWiNFO64 Download over fanspeed for getting hardware sensor information. You can start the logging in sensor only mode to generate a comma separated value (CSV) file that you can upload on here.
Run the log in HWINFO while you run OCCT, and also keep HWINFO logging enabled and go through the rest of the steps outlined in that thread for STOP 0x116 I linked in my previous post (the ones you have not already done, that is) and post back the results. You want to go through the first three posts in that link and follow each step closely until you have completed all steps that you have not already done to this point. I am with usasma in that I do not approve of registry cleaners, so steer clear of that step.
Good luck! :)
Attachment 192922
HWINFO output during the OCCT test attached.
I'll continue with that checklist as well.
Temperatures and Voltages look clean. Proceed to the next step in the STOP 0x116: VIDEO_TDR_ERROR troubleshooting link. :)
I've finished most of the checklist now. I haven't tried reinstalling Direct-X just yet, and I don't have a multimeter to measure voltages with (though I can go out and get one if necessary). Ran sfc.exe /scannow, which returned normal results. Reset my page file according to the steps in the link. Crash keeps happening like clockwork.
I haven't tried streaming video from a different web browser (just in case that makes a difference) or by playing a DVD; I'll see if those also cause the crash, just out of curiosity if nothing else.
I had HWINFO recording during the last crash, the log is attached. I was running Dead Island while streaming youtube videos to reproduce the error.
Attachment 193042
EDIT: Tried using a different browser for video playback. Same old thing, got the crash.
Last edited by EffinCraig; 14 Jan 2012 at 09:37.
It definitely appears to be display card related for the crashes, and it may very well be hardware related. When the crash occurred, the GPU sensors gave very strange readings. The fan's RPMs dropped to 0, yet the GPU sent that the fan was running at 100%. The memory clock changed from 1000.0 MHz to 3196.9 MHz, the GPU clock changed from 775 MHz to 13.7 MHz, and the GPU utilization dropped from 100% to 0%. Temperatures and Voltages look good.
It could be a faulty power supply (though as I said, Voltages look good, so I don't think this is the case), a bad motherboard slot, or the graphics card itself was damaged during transfer. The only way to really test these is to find another graphics card, another power supply, or use a multimeter on the power supply to rule them out.
I've opened a ticket with AMD tech support, I'm hoping I'll get an RMA out of this (because it sucks to go buy another video card just for testing purposes). I'd try my old EVGA 8800GT to test the slot, but it is sadly dead. If no RMA is forthcoming I'll get a new card to test (return policy permitting of course).
I'm going to move the card to a different PCI-e slot and see if that changes anything.
Quote