NTOSKRNL.EXE error when starting 32bit games.

[onetoomanypcs]

As the title says I've been getting this error everytime I try to start a game. This started after installing Windows 8.1 update and Batman Arkham Origins (Steam).

112113-11406-01.dmp 2013-11-21 12:12:30 PM 0x00000124 00000000`00000000 ffffe000`024958f8 00000000`00000000 00000000`00000000 ntoskrnl.exe ntoskrnl.exe+5b27b9 x64 ntoskrnl.exe+5b27b9 C:\Windows\Minidump\112113-11406-01.dmp 8 15 9600 262,144 2013-11-21 12:12:37 PM

The machine automatically reboots as if the reset button has been pressed. No BSOD or anything displayed but a log is dumped. I've tested my RAM, and changed out the PSU and the error continues, I've also reinstalled Windows 8.1 fresh and clean, reverted back to display drivers that worked with the games previously with no luck. I've removed my CPU, RAM and GPU overclocks with the same end result. As an after thought I disabled Crossfire and the 32bit exes worked, so I pulled the 7950 out of my wifes computer and rotated the ones in my system out one by one and tried with hers (to see if one of the cards was faulty) and it still resets. Everything was working fine on the 18th, I'd installed the Windows 8.1 update, played a few games, on the 19th I played Battlefield 4 for a bit, then on the 20th I installed Batman Arkham Origins and when I went to play it the system restarted. I figured it was with the game at first, until I tried another 32bit game (Neverwinter Online) and another and another, currently all I can play is Battlefield 4 64bit and Crysis 64bit without issue.

So, that is about it. I've never been stumped by anything like this before. Almost all sites I am reading say that it is PSU related, but that is ruled out. So, if any of the gurus here have any ideas, let me know.

System:
AMD FX-8350
Gigabyte 990FXA-UD7
Dual Gigabyte 7950s
16GB Corsair Vengeance 1600Mhz
Corsair TX850 PSU
Windows 8.1
Catalyst 13.11 Beta9.2 (Reverted back to 13.9 and issue still continued though it worked with these drivers previously)

All system drivers etc are up to date, I've also ran a bunch of different malware scanners and antiviruses, but all come up clean.

Help... please.

 

[The_Prophecy]

Wow... I both love it, and hate it people are very thorough with their troubleshooting. It's nice since you don't have to ask a bunch of basic questions, but the answer may be very difficult to pinpoint.

The only thing that sticks out at me at this point is the fact that you haven't mentioned thermal readings for the system components. Run HWMonitor to keep an eye on the temps and see how high they go both at idle, and under heavy load. Post back once you have them.

 

[onetoomanypcs]

Wow... I both love it, and hate it people are very thorough with their troubleshooting. It's nice since you don't have to ask a bunch of basic questions, but the answer may be very difficult to pinpoint.

The only thing that sticks out at me at this point is the fact that you haven't mentioned thermal readings for the system components. Run HWMonitor to keep an eye on the temps and see how high they go both at idle, and under heavy load. Post back once you have them.

Something I forgot to post, but had already checked as well. Overclocked the GPUs ran around 72C after a good long time under full load (Crysis 3, Battlefield 3 and 4) CPU ran up to 74C overclocked to 5.0Ghz and 62C stock, keep in mind I also have a +10C offset because the chip seems to report idle at 7-10C, which is pretty impossible. Just for grins I checked just now, CPU was pretty idle (was watching a movie) and at 31C the GPU at 45C but when I attempted firing up Neverwinter; reset the PC immediately. Idle temps are around 39C GPU1, 37C GPU2 and 25-27C for the CPU. Running a Corsair H100i for the CPU.

I do realise that the temps may seem rather high, but I've checked with both AMD and Gigabyte and they say that those are well within their thresholds, though neither gave me a max temp, so 6 months ago I forced my 8350 to safety shutdown, which it did at 96C (didn't even start throttling until 90C) So I figure as long as she is running 20C cooler than the safety she is in decent temps.

I also forgot to mention in the OP that I am running Eyefinity. But, even when I disable it the issue still occurs. As if that wasn't enough, my Metro Apps aren't working now either, though they are not forcing a restart they dump me back to desktop.