Computer Restarts While Playing Games

[xI Insidious Ix]

My computer spontaneously restarts while playing almost any game I own (Starcraft II, SimCity, Diablo III, and a few others) and comes up with a Kernel-Power 41 error in the event viewer that looks like this:

- System

- Provider

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

EventID 41

Version 2

Level 1

Task 63

Opcode 0

Keywords 0x8000000000000002

- TimeCreated

[ SystemTime] 2013-03-07T02:38:25.083211600Z

EventRecordID 5419

Correlation

- Execution

[ ProcessID] 4
[ ThreadID] 8

Channel System

Computer Workstation

- Security

[ UserID] S-1-5-18


- EventData

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

The computer is about two years old and it started doing this maybe 6 months ago with increasing frequency to the point where its pointless to even start playing a game because it restarts so quickly. My PC's specs are:

Intel Core i7 2600k @ 3.4GHz Processor
Gigabyte GA-Z68 Motherboard
EVGA GTX560 Ti Graphics Card
Corsair Dominator 16GB (4x4) RAM
Seagate 1TB HDD
Crucial 128GB SSD
Corsair 750HX PSU

I've run prime95 for about 30 minutes with no issues. The CPU runs at about 35C idle and up to 80C under 100% load. The GPU runs at about 70-80C under load (read while playing games of SCII). I've run windows memory test without errors, as well as uninstalling and reinstalling windows both with and without using the SSD. I've disassembled and reassembled the entire thing making sure everything was squeaky clean. I've installed the newest beta drivers for the graphics card to no avail. I've turned off the restart on system power off function and it still restarts without a BSOD.

I'm guessing the issue is with the PSU which is why its listed under the PSU subcategory, but I was just looking for another opinion or more tests to do considering I've been out of the computer repair industry for around 7 years and don't really know these types of things for sure.

[EDIT]: I would also like to add that I'm realizing that for some reason this never happens with torture testing applications. I just ran around 30 minutes of FurMark getting only 10fps and GPU temp around 85 solid for that half hour with absolutely no problem. Then I turn around and boot up SimCity and it crashes in under 2 minutes.

[EDIT2]: I just checked the power supply for the correct voltages and the 12V rail is exactly 12.00V and the same is true for the 5V rail (its supplying 5.00V). Does this mean all is good with the power supply? Or could there be other PSU issues that only happen when a game is running?

 

[XYMan]

Defected PSU unit.

 

[simon12]

Try running Prime 95 or IBT & Furmark at the same time.

 

[darth pravus]

Do as simon says 😉

If the combined load of power on the PSU causes you to lockup or crash then it is certainly the PSU.

 

[apache_lives]

ram can pass memtest but fail stability tests

lower the speeds, up the voltage and manually set the timings see if it helps

also force your processor (no AUTO settings) to the correct settings

check your HSF is installed and seated correctly also

 

[fauzan_22]

You can check the psu's power cord as the same thing happened to me once and changing the power cord fixed the problem.

 

[xI Insidious Ix]

Ok thanks all I will try these things and report back.

@apache_lives - The CPU is manually set to 3.4GHz right now (its normally 3.7 w/TurboBoost).

Also I had a dual monitor set up (an old Hyundai that was going bad) and removing the other monitor seems to help the problem a bit. Now the games take 30-40 minutes to crash instead of the usual 2-15. Does this point to the video card being the problem?

 

[Flakkattack]

I would also say that your psu is failing. If you are using two connections for your displays through your gpu, it could either be the psu not pushing enough power to the card. You could try to hook one monitor through mobo, and one through gpu. If it goes back to the way it was (crashing 2-15 mins) then I can almost guarantee that its the psu going out.

 

[xI Insidious Ix]

So I just ran Furmark and Prime95 for 3.5 hours and it ran flawlessly. It is very frustrating to not be able to replicate the problem, especially since the games im playing don't stress the system nearly as much as those two programs. Will try the monitor switch and report back again.

 

[Flakkattack]

I think that OCCT has a PSU stress test on it. You could also try that and see what luck you get as well, but it sounds to me like that would be the problem. Either that or you have some sort of manufacturer defect in your GPU. If you figure out that its not your psu, go to the next, test and resolve if the problem is found. If you bought the gpu within a certain time, I am pretty sure that they will let you RMA it.

 

[xI Insidious Ix]

Thanks thats exactly what I need. Downloading it right now, still have yet to test the monitor switching, so I'll be back with both of those results.

 

[Flakkattack]

Awesome! Good luck Insidious!

 

[xI Insidious Ix]

I just ran a quick ~30 minute test of the PSU stress test on OCCT. I will do a much longer one later tonight, but everything looked pretty normal other than this:

I know that those actual numbers are likely very off because I tested the PSU with a multimeter a few days ago and got exactly 12.00V but those fluctuations definitely do not look normal, especially compared to all of the other graphs that are spot on for the entire test at whatever voltage they should be.

Is this normal? I don't know enough about power supplies to know if this type of thing is normal or not.

[EDIT]: Oh and another question: If the computer crashes and restarts, will OCCT still save those log files? I'm thinking about running OCCT for the monitoring and then playing games and then looking at those logs to determine whats wrong.

 

[Flakkattack]

I am pretty sure that it does save the log files. I have never needed to refer back to them but I am pretty sure that if you check OCCT in your program files folder, there should be a subfolder that has all the saved data.
As far as the graph you provided, I will be honest and say that Ive never had to stress test my psu. However, from using OCCT and doing stress testing on my cpu, the graph looks pretty normal to me. Most of the time the fluctuations happen during the testing of times of cooling down and increasing voltage again. If you look at the right it says that its also based on cpu usage, so I am assuming that its testing the voltage along with your cpu. It shows that at max load on your cpu, you are only using about 6.0v. So when I have used OCCT for my cpu testing, it looks roughly the same as far as decreasing/increasing values during the test like that.
On the other hand though, If you notice that you said that you crash at about 12-15 minutes, you are actually losing power right at about every 15 minutes. That could potentially be something to consider.

Do you have a friend or something with a psu that you can borrow to do a little testing? I bet that would be the easiest way to get your answer and either replace yours, or move to the next thing that could possibly be the issue.

 

[urge13]

I was having similar problems. I set my Nvidia control panel 3d settings power management to "prefer maximum performance". I noticed improvement but restarts were still happening while gaming so I dusted out my PC. No restarts while gaming after that for weeks so far.

 

[Majormofo]

I had the exact problem and i flashed the bios to the previous version and it worked. My mobo is brand new gigabyte z170 has version F7 i flashed it with F6. I did the same tests as u did and thru eliminating all other possibilities it had to be the mobo. Remember bios flashing is not for the faint hearted. Best of luck

My computer spontaneously restarts while playing almost any game I own (Starcraft II, SimCity, Diablo III, and a few others) and comes up with a Kernel-Power 41 error in the event viewer that looks like this:

- System

- Provider

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

EventID 41

Version 2

Level 1

Task 63

Opcode 0

Keywords 0x8000000000000002

- TimeCreated

[ SystemTime] 2013-03-07T02:38:25.083211600Z

EventRecordID 5419

Correlation

- Execution

[ ProcessID] 4
[ ThreadID] 8

Channel System

Computer Workstation

- Security

[ UserID] S-1-5-18


- EventData

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

The computer is about two years old and it started doing this maybe 6 months ago with increasing frequency to the point where its pointless to even start playing a game because it restarts so quickly. My PC's specs are:

Intel Core i7 2600k @ 3.4GHz Processor
Gigabyte GA-Z68 Motherboard
EVGA GTX560 Ti Graphics Card
Corsair Dominator 16GB (4x4) RAM
Seagate 1TB HDD
Crucial 128GB SSD
Corsair 750HX PSU

I've run prime95 for about 30 minutes with no issues. The CPU runs at about 35C idle and up to 80C under 100% load. The GPU runs at about 70-80C under load (read while playing games of SCII). I've run windows memory test without errors, as well as uninstalling and reinstalling windows both with and without using the SSD. I've disassembled and reassembled the entire thing making sure everything was squeaky clean. I've installed the newest beta drivers for the graphics card to no avail. I've turned off the restart on system power off function and it still restarts without a BSOD.

I'm guessing the issue is with the PSU which is why its listed under the PSU subcategory, but I was just looking for another opinion or more tests to do considering I've been out of the computer repair industry for around 7 years and don't really know these types of things for sure.

[EDIT]: I would also like to add that I'm realizing that for some reason this never happens with torture testing applications. I just ran around 30 minutes of FurMark getting only 10fps and GPU temp around 85 solid for that half hour with absolutely no problem. Then I turn around and boot up SimCity and it crashes in under 2 minutes.

[EDIT2]: I just checked the power supply for the correct voltages and the 12V rail is exactly 12.00V and the same is true for the 5V rail (its supplying 5.00V). Does this mean all is good with the power supply? Or could there be other PSU issues that only happen when a game is running?