[SOLVED] Event 41 kernel power while gaming, by now built two separate PCs with identical faults

Aug 17, 2019
21
0
10
Hi all,

My PC has been shutting off and restarting when gaming since I first built it a couple of months back. There are no error messages, screen freezes, visual glitches or anything. Just a simple restart as if I pressed the off and on button.

Event log: The previous system shutdown at ... on ‎... was unexpected. The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

Event 41 kernel power. BugcheckCode 0


I recently noticed that in the event viewer, a couple of seconds before the time it says it crashed there always seems to be an Event 34, BTHUSB:

The local adapter does not support an important Low Energy controller state to support peripheral mode. The minimum required supported state mask is 0x2491f7fffff, got 0x1fffffff. Low Energy peripheral role functionality will not be available.”

Although, I’m not sure if this is related to the cause or just part of the PC is shutting down.

This only occurs while gaming and only with certain games. Certain maps/areas/saves are particularly prone. What mostly has happened is that I can play a game for a while with a PC crash maybe 2-3 times on a weekend. But then get to a particular area/save/fight and crash within a minute. Making and further progress impossible. I can sometimes reload an earlier save and avoid that area and be fine again for a while. Warhammer Vermintide 2 has been the most reliable trigger. I just cannot complete the first missions. There is one where it always happens around the same spot.
It seems to occur with more visually taxing games. However, I can turn the setting all the way down and it will still happen. No issues with any other use besides gaming, although it has happened while merely changing the resolution of one game in the main menu.

Temperatures seem fine to me. These were logged ~10 secs before a recent crash.
System: 41C (105 degF)
PCH: 49C (120 degF)
MOS: 46C (114 degF)
CPU: 63C (145 degF)
GPU: 67C (152degF)

I have spent a lot of time troubleshooting and weeks on waiting for returns but have failed to figure it out.
---------------------------------------
Specifications:
CPU: AMD Ryzen 5 3600 3.6 GHz 6-Core Processor
Cooler: BE QUIET Dark Rock Pro 4 135 mm Silent Wings Fan CPU Cooler
Video Card: RTX 2070 SUPER VENTUS OC
Power supply: Corsair CXM 750 Semi-Modular 80 PLUS Bronze Power Supply Unit, 750 W
Motherboard: MSI B450 GAMING PLUS Motherboard
Memory: Crucial Ballistix Sport LT 16 GB (2 x 8 GB) DDR4-3200
Hard drive 1: Crucial MX500 500GB SSD
Hard drive 2: Seagate ST2000DM006 Barracuda 2000 GB Hard Drive
Windows 10, Education
Nothing 2nd hand.
Desktop is plugged directly into the wall socket.
---------------------------------------
What I've tried so far and hasn't helped:

Reinstalled Windows on the SSD.
Used DDU to uninstall and reinstall my drivers.
Updated BIOS (misinterpreted the versions and did the original flashback with an outdated one).
Checked Chipset drivers are up to date.
Checked SSD drivers are up to date.
Checked Realtek audio drivers are up to date.

Memtest86 to test RAM. Left it overnight. No errors.
Used one RAM stick at a time.
Used prime95 to test my CPU. No issues. CPU temps will go up to around 70°C when under full load.
Ran 3D mark demo. No issues.
Reset Radeon settings to default.
Turned off fast-boot.
Set Power plan to high.
Set DRAM to 2800HHz (instead of auto).
Had a big fan blowing into the case to cool things down.
Hooked up the video card with two separate power cables.
No overclocking. I recently tried to overclock a bit to see if that could help. Nope. Reset to defaults after.
2nd parallel Windows install on the Seagate. Installed a game on the Seagate as well. Still crashes. Did not remove SSD for this so maybe not the perfect test.

SFC /SCANNOW , no issues
WhoCrashed, does not return any reports.
Replaced my power supply. Same model. Still crashed.
Swapped graphics cards from AMD to NVIDIA. Still crashed.
Replaced motherboard with MSI B450M PRO-VDH MAX, still crashed. (did not reinstall windows at the same time though, went back to the more spacious GAMING PLUS motherboard when installing new CPU).
Replaced CPU. Same model. Still crashed.

Literally, the only components I haven’t swapped now are the SSD, peripherals (mouse/keyboard/screen), and case. I feel by now I’ve built two separate pc with the identical faults…

Initially, the power supply seems the most likely. Returning that and getting a replacement didn’t fix it though. Having tried what, I feel is everything else, I’m back to thinking it’s the culprit. But that would mean I got two identically faulty ones in a row or more likely some compatibility issue? However, the crashes happen even if I set the graphics to low, which makes me think it’s not a power draw issue on the PSU.

Any help you can provide would be very very much appreciated.
 
Solution
Do not know about a double RTM - seller sets the rules there....

As for selecting a new power supply, Do not skip on either wattage or quality.

Tally up the required wattages for all build components. If a wattage range is provided use the high end value. For a GPU use the manufacturer's recommended PSU wattage.

After totaling up all wattages then add 25% more for a final total wattage.

There are online calculators to help with such calculations. Run two or three to get a consensus wattage value.

Also:

https://www.tomshardware.com/reviews/best-psus,4229.html
Aug 17, 2019
21
0
10
When I run FurMark and prime95 simultaneously my pc crashes immediately. However not when run individually.
Does that indicate its still the PSU despite it being swapped already?
 
Aug 17, 2019
21
0
10
Hi all,
My PC has been shutting off and restarting when gaming since I first built it a couple of months back. There are no error messages, screen freezes, visual glitches or anything. Just a simple restart as if I pressed the off and on button.

Event log: The previous system shutdown at ... on ‎... was unexpected. The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
Event 41 kernel power. BugcheckCode 0

I recently noticed that in the event viewer, a couple of seconds before the time it says it crashed there always seems to be an Event 34, BTHUSB:
The local adapter does not support an important Low Energy controller state to support peripheral mode. The minimum required supported state mask is 0x2491f7fffff, got 0x1fffffff. Low Energy peripheral role functionality will not be available.”
Although, I’m not sure if this is related to the cause or just part of the PC is shutting down.

This only occurs while gaming and only with certain games. Certain maps/areas/saves are particularly prone. What mostly has happened is that I can play a game for a while with a PC crash maybe 2-3 times on a weekend. But then get to a particular area/save/fight and crash within a minute. Making and further progress impossible. I can sometimes reload an earlier save and avoid that area and be fine again for a while. Warhammer Vermintide 2 has been the most reliable trigger. I just cannot complete the first missions. There is one where it always happens around the same spot.
It seems to occur with more visually taxing games. However, I can turn the setting all the way down and it will still happen. No issues with any other use besides gaming, although it has happened while merely changing the resolution of one game in the main menu.

Temperatures seem fine to me. These were logged ~10 secs before a recent crash while gaming.
System: 41C (105 degF)
PCH: 49C (120 degF)
MOS: 46C (114 degF)
CPU: 63C (145 degF)
GPU: 67C (152degF)

I can crash it on command:
Used prime95 to test my CPU. No issues. CPU temps will go up to around 70°C when under full load.
Ran FurMark for 30 mins. GPU will go to ~80°C but not crash.
When I run prime95 while running FurMark it will crash less than 5 seconds later.

This is a HWiNFO log every 250ms as I start prime95 and FurMark.
http://s000.tinyupload.com/?file_id=07365758737484040249

I have spent a lot of time troubleshooting and weeks on waiting for returns but have failed to figure it out.

Initially, the power supply seems the most likely. Returning that and getting a replacement didn’t fix it though. Having tried what, I feel is everything else, I’m back to thinking it’s the culprit. But that would mean I got two identically faulty ones in a row or more likely some compatibility issue?

Any help you can provide would be very very much appreciated.
---------------------------------------
Specifications:
CPU: AMD Ryzen 5 3600 3.6 GHz 6-Core Processor
Cooler: BE QUIET Dark Rock Pro 4 135 mm Silent Wings Fan CPU Cooler
Video Card: RTX 2070 SUPER VENTUS OC
Power supply: Corsair CXM 750 Semi-Modular 80 PLUS Bronze Power Supply Unit, 750 W
Motherboard: MSI B450 GAMING PLUS Motherboard
Memory: Crucial Ballistix Sport LT 16 GB (2 x 8 GB) DDR4-3200
Hard drive 1: Crucial MX500 500GB SSD
Hard drive 2: Seagate ST2000DM006 Barracuda 2000 GB Hard Drive
Windows 10, Education
Nothing 2nd hand.
PSU is plugged directly into the wall socket.
---------------------------------------

What I've tried so far and hasn't helped:

Reinstalled Windows on the SSD.

Used DDU to uninstall and reinstall my drivers.
Updated BIOS (misinterpreted the versions and did the original flashback with an outdated one).
Checked Chipset drivers are up to date.
Checked SSD drivers are up to date.
Checked Realtek audio drivers are up to date.

Memtest86 to test RAM. Left it overnight. No errors.
Used one RAM stick at a time.

Ran 3D mark demo. No issues.

Reset Radeon settings to default.

Turned off fast-boot.
Set Power plan to high.

Set DRAM to 2800HHz (instead of auto).

Had a big fan blowing into the case to cool things down.

Hooked up the video card with two separate power cables.

No overclocking. I recently tried to overclock a bit to see if that could help. Nope. Reset to defaults after.

2nd parallel Windows install on the Seagate. Installed a game on the Seagate as well. Still crashes. Did not remove SSD for this so maybe not the perfect test.

SFC /SCANNOW , no issues
WhoCrashed, does not return any reports.

Ran everything outside the case on a box. Still crashed.

Replaced my power supply. Same model. Still crashed.
Swapped graphics cards from AMD to NVIDIA. Still crashed.
Replaced motherboard with MSI B450M PRO-VDH MAX, still crashed. (did not reinstall windows at the same time though, went back to the more spacious GAMING PLUS motherboard when installing new CPU).
Replaced CPU. Same model. Still crashed.
 

Ralston18

Titan
Moderator
Bluetooth ?

I googled:

"The local adapter does not support an important Low Energy controller state to support peripheral mode. The minimum required supported state mask is 0x2491f7fffff, got 0x1fffffff. Low Energy peripheral role functionality will not be available.”

Is that a consistent error?

Do you see any similar errors via Reliability History? Or other errors that correspond with the crashes?

Not really sure about some of the links I found but I did scan through a few.

For example:

https://forum-en.msi.com/index.php?topic=332482.0

https://www.tenforums.com/general-support/137074-warning-event-id-25-bthusb.html

However, it all may be moot if you are not using any Bluetooth devices...

And Windows 10, Education: Not sure about its ability to support heavy gaming. Source?
 
Aug 17, 2019
21
0
10
Bluetooth ?

I googled:

"The local adapter does not support an important Low Energy controller state to support peripheral mode. The minimum required supported state mask is 0x2491f7fffff, got 0x1fffffff. Low Energy peripheral role functionality will not be available.”

Is that a consistent error?

Do you see any similar errors via Reliability History? Or other errors that correspond with the crashes?

Not really sure about some of the links I found but I did scan through a few.

For example:

https://forum-en.msi.com/index.php?topic=332482.0

https://www.tenforums.com/general-support/137074-warning-event-id-25-bthusb.html

However, it all may be moot if you are not using any Bluetooth devices...

And Windows 10, Education: Not sure about its ability to support heavy gaming. Source?

I use a Bluetooth USB adapter. Removed it and still get the crashes. I think it did stop the 'local adapter' warning from being listed this time. Removing my network card didn't help either.

The windows education is just because I can get a free licence from my uni. From a quick google, it seems to me people have used it for gaming with no problem.
 
Last edited:
Aug 17, 2019
21
0
10
Yes. People do seem to be able to use Education for gaming but my general sense is that there is likely some limit to it....

Will defer to the PC Gaming community for comments on the overall suitability of Windows 10, Education for gaming.

"sfc /scannow" noted:

DISM?

Reference:

https://www.windowscentral.com/how-use-dism-command-line-utility-repair-windows-10-image

Thanks for your replies.
My access to the Windows education licence has lapsed (seems really stupid to me that this is a thing).
So for the next clean install, I can't use it anyway. I'll see if a different version helps.

I just ran DISM Scanhealth. It returned " The component store is repairable"
I guess that means it found an error.

Just ran RestoreHealth. "The restore operation completed successfully "
I can do another stability test in a little bit. Currently in the middle of creating a windows 10 media on a different USB than the last times (suggested on Reddit).

Or was this just for the diagnosis and is there a better fix than the Restore Health command?

Thanks,
 
Aug 17, 2019
21
0
10
Test to see if DISM etc. fixed anything. Hopefully so.

If not, then proceed with a clean install.

Do you think I should install Windows on the SSD again? Or in case that is the issue install on the HDD?
I guess I have to bite the bullet and do that. Remove the SSD completely and make sure that is not the issue. :(
 
Aug 17, 2019
21
0
10
Clean install.

SSD is 500 GB if I have followed correctly. Plenty of room and I would use the SSD for the boot drive.

https://forums.tomshardware.com/faq/how-to-do-a-clean-installation-of-windows-10.3170366/

Follow the link's instructions (including read the instructions first).

Plan it out, get everything ready, take your time, keep notes of what you do and what happens.

Still crashes. Will do the clean install. But not much hope. Nothing has really changed.
Thank you for your help so far though.
 
Aug 17, 2019
21
0
10
Clean install.

SSD is 500 GB if I have followed correctly. Plenty of room and I would use the SSD for the boot drive.

https://forums.tomshardware.com/faq/how-to-do-a-clean-installation-of-windows-10.3170366/

Follow the link's instructions (including read the instructions first).

Plan it out, get everything ready, take your time, keep notes of what you do and what happens.

Just reinstalled windows on the SSD. Only 3 things I did was: Downloaded the GPU driver, prime95, and FurMark.
Crashed just as fast. :(
Will try an install on the hdd. Will order a new PSU after that. Guess lightning strikes twice and I got two identically broken PSUs... Or more likely its a compatibility issue. Arrrrrrrrrrrrrrrrrrrrrrgh
 
Aug 17, 2019
21
0
10
Just reinstalled windows on the SSD. Only 3 things I did was: Downloaded the GPU driver, prime95, and FurMark.
Crashed just as fast. :(
Will try an install on the hdd. Will order a new PSU after that. Guess lightning strikes twice and I got two identically broken PSUs... Or more likely its a compatibility issue. Arrrrrrrrrrrrrrrrrrrrrrgh

Crash on a clean install on the hdd too. I give up...
 

Ralston18

Titan
Moderator
If you skip installing Prime 95 and FurMark will the system boot and stay stable?

Take a very close look at your build. Look for signs of a short. Ensure that all standoffs are in place, that no wires are being pinched or bare metal showing/touching where it should not be.

Check around the I/O back panel (shield) and all other ports.

Test and try all components outside of the case.
 
Aug 17, 2019
21
0
10
If you skip installing Prime 95 and FurMark will the system boot and stay stable?

Take a very close look at your build. Look for signs of a short. Ensure that all standoffs are in place, that no wires are being pinched or bare metal showing/touching where it should not be.

Check around the I/O back panel (shield) and all other ports.

Test and try all components outside of the case.

Yeah, it's fully stable and fine along as I don't stress it. (Although I haven't let it run for a long time since the reinstall).
The only issue is the regular crashes while gaming. And I can replicate those with running FurMark and Prime95 at the same time.

Last night I took it out of the case. It's currently sitting on top of a card box.
Still crashing. I will swap out the oversized CPU cooler, but I can't see anywhere it is making contact with other components.

I've been looking into underclocking it to see if that helps. Or at least that would more clearly point to a power draw issue. Setting the CPU to eco mode didn't help. I may have to underclock the GPU and CPU simultaneously.
 

Ralston18

Titan
Moderator
Fair enough.

I am tempted to move this thread to Overclocking but the thread has gotten rather long and into TLDR territory.

My thought is that once you have finished your ongoing efforts then consider starting a new thread in Overclocking (or elsewhere as applicable).

Basically a fresh start hopefully with things narrowed down somewhat.

Keep track of the clocking and configurations that work/do not work.

Note any error codes, beeps, LED's etc.

And be sure to double check the applicable CPU/GPU User Guides. There may some fine print or other caveats therein. Or links to the manufacturer's website that effectively do the same.
 
Aug 17, 2019
21
0
10
Fair enough.

I am tempted to move this thread to Overclocking but the thread has gotten rather long and into TLDR territory.

My thought is that once you have finished your ongoing efforts then consider starting a new thread in Overclocking (or elsewhere as applicable).

Basically a fresh start hopefully with things narrowed down somewhat.

Keep track of the clocking and configurations that work/do not work.

Note any error codes, beeps, LED's etc.

And be sure to double check the applicable CPU/GPU User Guides. There may some fine print or other caveats therein. Or links to the manufacturer's website that effectively do the same.

Yes, it's gotten quite long. Thank you very much for your help so far.

GPU power limit set to 58%, core clock -23.
CPU in Eco mode (45 watts)

It ran Furmark&prime95 for 1:30 mins instead of 5 seconds before crashing.
I guess that's an improvement...

Anything else's I can do before ordering a new PSU? I've been wrong every single other time. Having gotten two identically faulty PSUs just seems so unlikely. I'm starting to think I accidentally switched it back while installing the new one.
Can I do an RTM twice?
 

Ralston18

Titan
Moderator
Do not know about a double RTM - seller sets the rules there....

As for selecting a new power supply, Do not skip on either wattage or quality.

Tally up the required wattages for all build components. If a wattage range is provided use the high end value. For a GPU use the manufacturer's recommended PSU wattage.

After totaling up all wattages then add 25% more for a final total wattage.

There are online calculators to help with such calculations. Run two or three to get a consensus wattage value.

Also:

https://www.tomshardware.com/reviews/best-psus,4229.html
 
Solution
Aug 17, 2019
21
0
10
Do not know about a double RTM - seller sets the rules there....

As for selecting a new power supply, Do not skip on either wattage or quality.

Tally up the required wattages for all build components. If a wattage range is provided use the high end value. For a GPU use the manufacturer's recommended PSU wattage.

After totaling up all wattages then add 25% more for a final total wattage.

There are online calculators to help with such calculations. Run two or three to get a consensus wattage value.

Also:

https://www.tomshardware.com/reviews/best-psus,4229.html
Thank you!
 
Aug 17, 2019
21
0
10
Do not know about a double RTM - seller sets the rules there....

As for selecting a new power supply, Do not skip on either wattage or quality.

Tally up the required wattages for all build components. If a wattage range is provided use the high end value. For a GPU use the manufacturer's recommended PSU wattage.

After totaling up all wattages then add 25% more for a final total wattage.

There are online calculators to help with such calculations. Run two or three to get a consensus wattage value.

Also:

https://www.tomshardware.com/reviews/best-psus,4229.html

Used a new Corsair HX850 850W. This fixed all my issues. No crashes anymore.
So in conclusion: after retuning my first faulty Corsair CXM 750, the replacement was faulty as well.

Thank you for your help.