Continuous Kernel-Power 41 (63). Crash and reboot when gaming.

N17 dizzi · 14 Mar 2016

I have been having continuous problems with my PC crashing, restarting or giving me occasional BSODs. Recently the motherboard bios was stating power-surge warnings when rebooting, so I have since replaced my PSU. The power surge warning has gone, but the 41's and freezing persists.

My system is a Asus X99 Pro, 5960K, GTX Titan X SLI, 16Gb 2,666Mhz Corsair DDR4 RAM, Corsair HX1000i 1KW PSU (Old PSU is a AX860) 512GB NVMe Samsung 950 PRO.

I have quite a lot of experience building PCs, and would say my knowledge is average. This one has stumped me. So far I have:

Reinstalled Windows 10
Change/update/rolled back drivers
Bought a brand new high quality PSU
Used different power sockets
Updated the bios on the motherboard
Returned CPU/GPUs to stock speeds
Ran MEMTEST86 - passed
Ran Intel stress test - passed

Does anyone know what this problem is most likely to be? I am about to try reinstalling Windows 8.1 on a sata SSD and install new power cables through my PC. I have seen many other threads where people guess at it being PSU CPU RAM or motherboard, with no real evidence to suggest which one it may be. The PSU is not the issue.

Many thanks for any help.

Arc · 14 Mar 2016

There is no crash dump in the uploaded zip, but this information is available .....

Event[9302]:
Log Name: System
Source: Microsoft-Windows-WER-SystemErrorReporting
Date: 2016-03-14T09:32:26.376
Event ID: 1001
Task: N/A
Level: Error
Opcode: N/A
Keyword: Classic
User: N/A
User Name: N/A
Computer: X99-5930K
Description:
The computer has rebooted from a bugcheck. The bugcheck was: 0x00000101 (0x0000000000000010, 0x0000000000000000, 0xffffd001888e7180, 0x0000000000000004). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 00000000-0000-0000-0000-000000000000.

Also this one:

13/03/2016 13:50 Windows Error Reporting Fault bucket AV_CODE_AV_NULL_IP_nvlddmkm!CNvLAdapter::processRC, type 0
Event Name: BlueScreen
Response: http://wer.microsoft.com/responses/r...e-07e494b44701
Cab Id: 3ed10ad6-a7ee-46ea-a91e-07e494b44701

Problem signature:
P1: d1
P2: 0
P3: 2
P4: 8
P5: 0
P6: 10_0_10586
P7: 0_0
P8: 256_1
P9:
P10:

Analysis symbol:
Rechecking for solution: 0
Report ID: 33b86752-4f36-419a-a185-bc562bff5a06
Report Status: 0
Hashed bucket:

Stop 0x101 is CLOCK_WATCHDOG_TIMEOUT. You will get a nice explanation of the issue here: STOP 0x101: CLOCK_WATCHDOG_TIMEOUT troubleshtg - Windows 7 Help Forums
It contains the steps you need to follow, too.

I would suggest you ....
► To not overclock. Set all the hardware components like CPU, RAM and Graphics card to default.

► Stress test the Graphics Card using Furmark.

FurMark - GPU Stress Test

Take a screenshot of the furmark window before closing it. Upload the screenshot for us. Also let us know if you have experienced any crash/BSOD and/or artifacts during the test.

► Stress test the CPU.

Prime95 - Stress Test Your CPU

It saves the result as a .txt file in the prime95's folder. Upload the file for us.

Let us know the results.

N17 dizzi · 14 Mar 2016

Hello Arc, thank you for your reply.

I believe that Prime 95 is not a safe stress test for Haswell-E processor family or X99 chipset as it puts too much amperage through the motherboard and indeed chip.

Can I perform a different stress test?

N17 dizzi · 14 Mar 2016

Here is a screen grab of furmark. I had it on for 10 minutes or so. No crashes or artifacts. This screenshot is after I resized furmark to get a screen shot. Temps had stabilized.

Attachment 69440

Arc · 14 Mar 2016

N17 dizzi said:

Hello Arc, thank you for your reply.

I believe that Prime 95 is not a safe stress test for Haswell-E processor family or X99 chipset as it puts too much amperage through the motherboard and indeed chip.

Can I perform a different stress test?

Prime95 is the best test, but Haswell is hot at its own. If you find it risky, avoid it. Opt for IntelBurn.

Furmark looks good. Let us know about your clocks too. If your system is overclocked, you need to revert those back to stock.

N17 dizzi · 14 Mar 2016

Arc said:

Prime95 is the best test, but Haswell is hot at its own. If you find it risky, avoid it. Opt for IntelBurn.

Furmark looks good. Let us know about your clocks too. If your system is overclocked, you need to revert those back to stock.

I reset everything to default. All I have is XMP profile 1 running as default underclocked the memory.

Here is the intel stress test: Attachment 69465

Arc · 15 Mar 2016

OK, it seems good.

Have you got any further BSOD after setting the hardware to default?

N17 dizzi · 15 Mar 2016

Arc said:

OK, it seems good.

Have you got any further BSOD after setting the hardware to default?

Not yet. I need to have a proper gaming session and see what happens.

Does this mean that my system cannot be overclocked at all? That would be rather unlucky. Particularly when you spend £3,000 on a system!

Arc · 15 Mar 2016

Overclocking causes BSODs. Stop 0x101 and Stop 0x124 are almost certain to occur in the overclocked environment. The overclockers are well aware of it, they adjust their OC settings depending on the frequency of those BSODs.

I would not say that you should not go for overclocking at all; but you need to find a OC setting which will minimize the BSODs (see, I mentioned "minimize"; that means all chance of BSODs will be there; but under control).

If you want to discuss about overclocking, get in PC Custom Builds and Overclocking - Windows 10 Forums ..... we have some excellent overclockers like essenbe, dude or paulpicks21.

N17 dizzi · 15 Mar 2016

Hello Arc,

I have had a very long gaming session on a demanding title, Witcher 3. It was rock solid throughout. I can safely say that the overclocking settings were causing the instability in this case.

I will spend some time investigating different or improved settings and voltages when I overclock in future.

Thanks again for your help