Random Restarts, No BSOD, Kernel Power Event ID 41 Task 63

[SignorMagnifico]

Hello!

Recently I've been experiencing random shutdowns of my entire system exclusively while gaming. I would normally expect a B.S.O.D. from an unstable system, but, unfortunately, no screen appears. The only thing that happens is that the system restarts itself suddenly after two to three hours of gaming.

The strange thing is that the shutdowns only happen during gaming, but they never happen during stress tests or video conversions. It's just when I'm gaming. I was also unable to find any minidump files in the root Windows folder. The only symptom I get is a generic error log in the Event Viewer.

I have already tried some troubleshooting to narrow down the issue.

I verified that I have "Automatically Restart" turned off in System Properties > Startup and Recovery. I have Complete Memory Dump turned on in the same section, and I set the Page File to the level recommended by the Memory Dump setting.

Here's everything I've tried so far to troubleshoot:

1) Stress tested the entire system using OCCT. I ran the test for almost 2 hours, and I checked in periodically to ensure none of my components were overheating. The voltages from my power supply also looked correct on the 12V, 3V, and 5V rails (I saw a max of .03V deviation). I was able to game for about two hours until my system rebooted again. I can let the system sit and run video conversions in VidCoder X for about 10 or 12 hours without a crash.
2) Ran a full Windows Memory Diagnostic. I did the extended test with two passes. No errors.
3) Physically swapped RAM sticks and checked all connections to PSU, GPU, etc. Verified that PSU fan turns on. No dust inside PSU or GPU heat sink.
4) The system uses about 480W during a stress test of both the CPU and GPU. The most I see when gaming is somewhere around 400W.
5) CPU temp never exceeds 60C when gaming, but it maxed out at 75-80C on the stress test. The GPU maxes at about 79 or 80C, but I've had it running at that temperature before without an issue.
6) I did a complete reinstall of my Nvidia graphics drivers. I noticed that after uninstalling them, Windows reinstalled the originals without my consent. I had to disable the automatic driver download, kill all Nvidia processes, uninstall all Nvidia items in Programs, and uninstall the graphics card from Device Manager.
7) I reinstalled my graphics driver, but I left out the 3D Vision and the NVidia HD Audio drivers to see if it would make a difference. The reboot continues.
8) I updated to the latest BIOS and adjusted some settings. I disabled anti-surge support and the VFR for the Intel chip. Neither setting made a difference in the old BIOS or the newer BIOS versions. I disabled Fast Boot in the BIOS, and I also disabled Fast Startup in CP > HW&Sound > Power Options > System Settings.
9) These crashes didn't start until upgrading to Windows 10 from Windows 8.1. I was using Windows 10 for about a month or two before the crashes started. My system used Windows 8.1 for about 8 months without this type of crash happening.
10) I have an overclock on the chip to 4.6GHz, but when my system has crashed due to instability in the past (when testing the overclock), there was always a BSOD with a different error message than 41. Generally, it was a watchdog error or temperature. I'm using a closed water loop with push/pull fans.
11) I made sure that the outlet isn't resetting. I have a backup UPS attached to the computer, and I verified that the battery kicks in when I unplug from the outlet. The computer keeps running if I disconnect electrical power.
12) It's probably not the UPS resetting because it makes a series of beeps when it turns off and on. I only hear the beep from the motherboard speaker when my computer restarts.

If anyone has any suggestions on what else I can do to solve this, or if you know where I can go to get more information, I would really appreciate the help. I'm out of ideas on what else I can try to diagnose the problem because the error is really generic. At this point, I'm unsure whether it's my PSU, my GPU, Windows 10, or drivers. Any advice would be appreciated. Thank you for the help.

System:
i7-4790K OC to 4.6GHz
Asus Maximus VII Formula
Corsair H110 (Push/Pull)
Corsair HX850W PSU
Corsair Dominator 8GB 2133MHz
EVGA GTX 780 SC
Creative SoundBlaster X-Fi Titanium
Crucial M4 SSD 512GB
Seagate 1TB SSHD
Western Digital Caviar Black 1TB HDD
ROG Front Base
Samsung Blu Ray Disc Drive
Acer XB270H GSync Monitor
Windows 10
CyberPower 1500VA (900W) AVR UPS

Crash log:
- System

- Provider

[ Name] Microsoft-Windows-Kernel-Power
[ Guid] {331C3B3A-2005-44C2-AC5E-77220C37D6B4}

EventID 41

Version 3

Level 1

Task 63

Opcode 0

Keywords 0x8000200000000002

- TimeCreated

[ SystemTime] 2015-12-03T12:54:34.356762300Z

EventRecordID 16021

Correlation

- Execution

[ ProcessID] 4
[ ThreadID] 8

Channel System

Computer MegumiReborn

- Security

[ UserID] S-1-5-18


- EventData

BugcheckCode 0
BugcheckParameter1 0x0
BugcheckParameter2 0x0
BugcheckParameter3 0x0
BugcheckParameter4 0x0
SleepInProgress 0
PowerButtonTimestamp 0
BootAppStatus 0

 

[Grimwinder]

The first thing I'd do, if you haven't already, is remove the overclocking and see if the problem persists at stock clocks. That would eliminate that as a suspect. You've got a good motherboard, but it almost sounds like the VRMs might be overheating after long gaming sessions.

 

[SignorMagnifico]

The first thing I'd do, if you haven't already, is remove the overclocking and see if the problem persists at stock clocks. That would eliminate that as a suspect. You've got a good motherboard, but it almost sounds like the VRMs might be overheating after long gaming sessions.

Thanks for the advice. I could try without the overclock, but I haven't yet done so because I've been using this same overclock setting for close to a year. Only the CPU is over clocked. The GPU is running at stock. I was able to game for hours without this kind of restart until about a month ago, including after I upgraded to Windows 10.

 

[Grimwinder]

Another thing you can try is to take off the case side and blow a floor fan directly onto the motherboard. If this improves the problem, then you know you probably have a heat issue with the VRMs (since you've tested the CPU/GPU temps). You can then look at aftermarket heatsinks for them specifically, or add more cooling to your case. Outside of that, my other suspect would be the PSU getting hot after long gaming use. Those would be my two best guesses based on your other troubleshooting.

 

[SignorMagnifico]

Good suggestions. My case actually has two 240mm blowing onto the motherboard from the side. I'm curious as to whether my PSU could be wearing out, but I don't want to put money down on a new one unless it's an absolute must.

 

[Grimwinder]

Unfortunately, the only way to know would be to swap it out or have it tested. You might try getting some monitoring software and see if you can see anything weird happening in your voltages, that would reinforce the thought that it's either the PSU or VRMs going bad/overheating.

 

[SignorMagnifico]

I could try some long term monitoring or something. When I ran a stress test, the voltages all looked fine. They were within .03v of the target throughout the test with max cpu and gpu usage. Is there a way to find out if I have too much current on one rail? Then again, my system would not have been stable for 8 months from the last major hardware change I made. I still have a feeling it's the NVidia drivers, but I have no idea how to prove/fix it.

 

[Grimwinder]

Well, you could do a uninstall of the Nvidia drivers with a complete sweep, then install an older driver version, and even try one for a different Windows version, just run the install as Admin in Compatibility mode. At worst you'd just have to uninstall and reinstall again.

 

[SignorMagnifico]

Well, you could do a uninstall of the Nvidia drivers with a complete sweep, then install an older driver version, and even try one for a different Windows version, just run the install as Admin in Compatibility mode. At worst you'd just have to uninstall and reinstall again.

That's a good suggestion. I did try doing a clean install of the drivers themselves, but I still have the problem. I also tried using an older driver version. I've been reading reports of other people having random restarts in Windows 10, so it could be an issue with the OS. I now know for a fact that my CPU clock is stable because I ran a stress test for two hours from OCCT, and I also left it encoding video overnight in Vid Coder X (CPU usage was 100% for about 10 hours). That leaves the GPU, the PSU, or the OS. After finals next week, I may try doing a complete clean reinstall of Windows 10 from scratch to see if it solves the problem.

 

[SignorMagnifico]

The first thing I'd do, if you haven't already, is remove the overclocking and see if the problem persists at stock clocks. That would eliminate that as a suspect. You've got a good motherboard, but it almost sounds like the VRMs might be overheating after long gaming sessions.

I tried disabling my OC completely in the BIOS by setting the core ratio to auto and voltage to auto. The BIOS set the voltage at 1.07V, but AI Suite showed 1.15V after opening Windows. I launched one of my games, and the system restarted without warning in less than a minute after getting into the game. I reloaded my stable OC of 4.6GHz and bumped up the VCore from 1.29V to 1.325V, and I was able to play for about 4 hours without a crash happening. Normally, the system would have restarted within about 1 to 2 hours. I'm guessing this is just due to wear on the CPU after a year? Thanks for your advice.

 

[Talavar]

Had this same problem. A stink bug got into my computer and crawled inside my graphics card.. apparently he was causing fluctuating shorts. I cracked it open, found that little ah heck and removed him. Problem solved. If all else fails, pop it open and take a gander. make sure you have some thermal paste available though, as you should replace it after doing this. Just putting this up for reference as a possible solution for others.. (blockages BAD!)