[SOLVED] Kernel power 41, event ID 63. Random crashes

[rs123]

Hello, thanks in advance for any advice given here.

My computer keeps crashing with no consistent reasoning due to a Kernel power 41 with event ID 63. The bugcheck code is 0, which I read was a hardware issue. I replaced my PSU but the crashes continued. I've done memtests, I removed all RGB software and updated all drivers, etc... I ran prime95 for an hour last night with no crashes, not sure if I ran it incorrectly or not. I also ran unigine for my gpu twice with no crashes as well. Any help or advice would be greatly appreciated on this. I will post my specs down below.


CPU: AMD Ryzen 5 3600 3.6 GHz 6-Core Processor
Mobo: MSI B550 Tomahawk
GPU: MSI 2060 RTX
RAM: Corsair vengeance 16gb
PSU: Corsair RMx Series (2021), RM850x, 850 Watt, GOLD, Fully Modular

Like I said, thanks in advance for any help. Really appreciate it!

 

[Colif]

41(63) is just the reaction to a restart, its not a cause as such.

I ran prime95 for an hour last night with no crashes, not sure if I ran it incorrectly or not.

one hour isn't long enough for Prime.

Prime 95 how to Guide: http://www.playtool.com/pages/prime95/prime95.html
what other parts in PC as it might be anything causing the crashes.

Do you have to be anything in particular for it to happen?

 

[rs123]

Going to run prime tonight using the instructions from the link.
Other parts are only hard drives, samsung 860 evo ssd and Western digital caviar blue hdd.

 

[rs123]

So I ran prime95 torture test on smallest FFTs and it broke 100 degrees which im guessing is not normal.

 

[Colif]

what cooling do you have? what case do you have?
3600 should max out at 95c, you shouldn't be able to get to 100c

restarts with no BSOD isn't windows, so cause is hardware of some kind.

did prime finish? results can be found in the folder the app is installed in, a text file.

 

[rs123]

I have a cooler master hyper 212. Never had a problem w thermals before, but I didn’t finish prime bc i was concerned w the temp being that high. should i still run it for an extended period of time w a temp that high ?
My case is a phanteks enthoo pro m w 2 front fans and one in the back.

 

[Crosslhs82x2]

Did you disable Avx's ?
If not theres your reason for the extra heat.

 

[rs123]

Avx settings are greyed out when I run torture test. I noticed in task manager my cpu is at 4.2ghz and the base is supposed to be 3.6. Is bios the only way to turn it down? I never overclocked so im not sure why it is that high but im thinking it could be the reason for my crashing.

 

[Crosslhs82x2]

Hi
Setting P95 to not run Avx's should be at the bottom of the of the run dialog box.
You have to click disable Avx2 and then the avx will also be able to be click to disable as well.
Have a look at this

https://imgur.com/a/ilFccZX

View: https://imgur.com/a/ilFccZX

 

[rs123]

Thanks for your help everyone. I do think the problem lies in my cpu being overclocked to 4.2 but I cant seem to get it down to 3.6. the cpu ratio is greyed out in my bios. I tried turning down the voltage a little but it doesnt seem to do anything. Im not sure why the cpu ratio is greyed out, any tips would be greatly appreciated on that.

 

[Crosslhs82x2]

If I remember right on my son's Msi b450 you have to change to like advanced or Expert mode to be able to change any cpu ratio.
So being grayed out tells me it's in Auto.
Check you windows Power Plan.
if you have updated your chipset drivers lately then go to the power plan as the update changed the settings.

In the power plan settings look for
Processor power management
then Minimum Processor Power state and click to expand.
check to see what % it is at.
if left at 99% then your cpu is idling at 4.2ghz
If it is then lower it down to 5% and click apply.
Now it will allow the cores to sleep when not be used.

 

[theboeingman]

I replaced my 12900K with another 12900KS. The problem is gone. No more Event 41 Kernel error and no more stability problems with RAM. The defective CPU was causing memory instability and random black screens and subsequent reboots/bootloops. All this time i thought I had bad ram sticks or faulty motherboards and GPUs. For anyone else with such issues, I recommend you email Intel and demand an RMA replacement. The process takes around 1 month to email back and forth.