Power Reporting Deviation Crashing My PC?

[tantishin]

Hello All,

Ryzen 3700X
Corsair Vengeance 32 GB RGB
MSI 2060 Super
MSI MPG x570 Gaming Edge Wifi
Corsair H100i

I have had problems with this PC in the past, getting a WHEA_UNDETECABLE_ERROR BSOD. Went to a couple places and they thought it was my CPU. Replaced the CPU, went to boot the PC, and it didn't boot. Took it to Geeksquad and they suggested trying a new PSU with the old CPU, and sure enough, it worked. Been running smoothly for about 1.5 months now, but whenever I am in games (ONLY IN GAMES, HASN'T CRAHSES AT IDLE, yet...) it just crashes. Turns off randomly with the following event logs from windows.

Event log 1

Log Name: System
Source: Microsoft-Windows-Kernel-Power
Date: 3/31/2021 4:45:48 PM
Event ID: 41
Task Category: (63)
Level: Critical
Keywords: (70368744177664),(2)
User: SYSTEM
Computer: DESKTOP-8F4PDNP
Description:
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 Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Microsoft-Windows-Kernel-Power" Guid="{331c3b3a-2005-44c2-ac5e-77220c37d6b4}" />
<EventID>41</EventID>
<Version>8</Version>
<Level>1</Level>
<Task>63</Task>
<Opcode>0</Opcode>
<Keywords>0x8000400000000002</Keywords>
<TimeCreated SystemTime="2021-03-31T20:45:48.7856231Z" />
<EventRecordID>9472</EventRecordID>
<Correlation />
<Execution ProcessID="4" ThreadID="8" />
<Channel>System</Channel>
<Computer>DESKTOP-8F4PDNP</Computer>
<Security UserID="S-1-5-18" />
</System>
<EventData>
<Data Name="BugcheckCode">0</Data>
<Data Name="BugcheckParameter1">0x0</Data>
<Data Name="BugcheckParameter2">0x0</Data>
<Data Name="BugcheckParameter3">0x0</Data>
<Data Name="BugcheckParameter4">0x0</Data>
<Data Name="SleepInProgress">0</Data>
<Data Name="PowerButtonTimestamp">0</Data>
<Data Name="BootAppStatus">0</Data>
<Data Name="Checkpoint">0</Data>
<Data Name="ConnectedStandbyInProgress">false</Data>
<Data Name="SystemSleepTransitionsToOn">39</Data>
<Data Name="CsEntryScenarioInstanceId">0</Data>
<Data Name="BugcheckInfoFromEFI">false</Data>
<Data Name="CheckpointStatus">0</Data>
<Data Name="CsEntryScenarioInstanceIdV2">0</Data>
<Data Name="LongPowerButtonPressDetected">false</Data>
</EventData>
</Event>

Event log 2

Log Name: System
Source: Microsoft-Windows-Kernel-Power
Date: 3/31/2021 11:58:33 PM
Event ID: 41
Task Category: (63)
Level: Critical
Keywords: (70368744177664),(2)
User: SYSTEM
Computer: DESKTOP-8F4PDNP
Description:
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 Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Microsoft-Windows-Kernel-Power" Guid="{331c3b3a-2005-44c2-ac5e-77220c37d6b4}" />
<EventID>41</EventID>
<Version>8</Version>
<Level>1</Level>
<Task>63</Task>
<Opcode>0</Opcode>
<Keywords>0x8000400000000002</Keywords>
<TimeCreated SystemTime="2021-04-01T03:58:33.4722128Z" />
<EventRecordID>9698</EventRecordID>
<Correlation />
<Execution ProcessID="4" ThreadID="8" />
<Channel>System</Channel>
<Computer>DESKTOP-8F4PDNP</Computer>
<Security UserID="S-1-5-18" />
</System>
<EventData>
<Data Name="BugcheckCode">0</Data>
<Data Name="BugcheckParameter1">0x0</Data>
<Data Name="BugcheckParameter2">0x0</Data>
<Data Name="BugcheckParameter3">0x0</Data>
<Data Name="BugcheckParameter4">0x0</Data>
<Data Name="SleepInProgress">0</Data>
<Data Name="PowerButtonTimestamp">0</Data>
<Data Name="BootAppStatus">0</Data>
<Data Name="Checkpoint">0</Data>
<Data Name="ConnectedStandbyInProgress">false</Data>
<Data Name="SystemSleepTransitionsToOn">0</Data>
<Data Name="CsEntryScenarioInstanceId">0</Data>
<Data Name="BugcheckInfoFromEFI">false</Data>
<Data Name="CheckpointStatus">0</Data>
<Data Name="CsEntryScenarioInstanceIdV2">0</Data>
<Data Name="LongPowerButtonPressDetected">false</Data>
</EventData>
</Event>

Went to Tom's Hardware Forums to try to get to the bottom of this, since it appears that no tech store understands the issue. They suggested I download HWInfo and just keep an eye on it. Sure enough, I get into a league game, play it out, and all goes fine. Then I look over and see the PRD minimum is at 78.7%. After reading the thread posted here about PRD, and how manufactures are using it to boost benchmark numbers, I assume what is happening is my motherboard is telling my CPU it can draw more, and more, and more power until the unit just can't hold it anymore. So my old PSU was fine, but when it crashed something about the hardware would tell it to BSOD, and this new CPU doesn't use that same process so instead it just shuts off.
Edit: Just glanced at the number again when loading a reddit page on google chrome, and it briefly dropped to 88% not sure if relevant.

So my question is; is this a CPU or Motherboard problem, and what is my fix? Is this even indicative of anything needing to be fixed? I never have overclocked, and don't even touch my bios because I am terrified I will screw something up. I think the only reason you can't boil water on my CPU is because my cooler has done a pretty kick ass job of keeping temps extremely low so it doesn't heat up too much, but just draws a shit ton of power instead.

Thank you in advance!

Just got a crash, file is linked below with what HWinfo got during the crash. The first ten lines are the moments leading up to the crash and everything below that is from when I was running my PC fine.

Google Drive: Sign-in

Access Google Drive with a Google account (for personal use) or Google Workspace account (for business use).

drive.google.com

 

[Zach]

PRD "issue" can't do any damage to the CPU, no matter how a board treats it. Its very difficult to crash it either. What you describe as "my motherboard is telling my CPU it can draw more, and more, and more power until the unit just can't hold it anymore" can't happen. No matter how much a board is lying to a CPU, the CPU itself has internal sensors and metrics that prevents damage or crash to occur. It goes as far as it can. Yes it may exceed the stock limits of power but still everything is under control and out of harm's way.

Also, PRD has value only under 100% CPU load. Check it when(during) running Cinebench R20.
When the system is idle, or surfing the net, watching videos, or even gaming you can't evaluate PRD. It's useless. Only under 100% CPU load matters. On all other situations PRD could be anything from 0% to 600% and doesn't mean anything.

Your crashing problem is elsewhere. Could be hardware related or software(most probably).
You should update to the latest version the:
1. Board BIOS (non beta)
2. Chipset drivers (from AMD and not from MSI)
3. GPU drivers
4. Windows10 v20H2

What you describe with CPU temp and power draw is typical Ryzen behavior and as intended by AMD engineers. The lower the temp the higher the clock and voltage will go until certain limits reached without compromising longevity or stability. Your CPU has a stock power draw limit of 88W (PPT), a current limit of 90A (EDC) and a thermal throttle current limit of 60A (TDC).
What are you seeing on HWiNFO sensors only mode from these sensors under various and on 100% load

 

[tantishin]

PRD "issue" can't do any damage to the CPU, no matter how a board treats it. Its very difficult to crash it either. What you describe as "my motherboard is telling my CPU it can draw more, and more, and more power until the unit just can't hold it anymore" can't happen. No matter how much a board is lying to a CPU, the CPU itself has internal sensors and metrics that prevents damage or crash to occur. It goes as far as it can. Yes it may exceed the stock limits of power but still everything is under control and out of harm's way.

Also, PRD has value only under 100% CPU load. Check it when(during) running Cinebench R20.
When the system is idle, or surfing the net, watching videos, or even gaming you can't evaluate PRD. It's useless. Only under 100% CPU load matters. On all other situations PRD could be anything from 0% to 600% and doesn't mean anything.

Your crashing problem is elsewhere. Could be hardware related or software(most probably).
You should update to the latest version the:
1. Board BIOS (non beta)
2. Chipset drivers (from AMD and not from MSI)
3. GPU drivers
4. Windows10 v20H2

What you describe with CPU temp and power draw is typical Ryzen behavior and as intended by AMD engineers. The lower the temp the higher the clock and voltage will go until certain limits reached without compromising longevity or stability. Your CPU has a stock power draw limit of 88W (PPT), a current limit of 90A (EDC) and a thermal throttle current limit of 60A (TDC).
What are you seeing on HWiNFO sensors only mode from these sensors under various and on 100% load

Thank you! I figured out the problem, and well I'll look stupid. My power strip got wedged in between my desk and wall in just the right way that if my feet hit the back of my desk while I move in the slightest, it turns off.

 

[Zach]

Well it can happen to any one, don’t feel like that. Just be glad that you found it.

 

[Dalai]

Although the problem is solved, JFTR: The MPG X570 Gaming Edge Wifi is known to have extremely hot VRMs as can be seen in this Hardware Unboxed video. Though you only have a 65 W CPU, I recommend to check the airflow in your case, especially over the VRMs and VRM heatsinks. Since you use an AIO liquid cooler, there's no airflow by default in that regard compared to an air cooler. Too hot VRMs could have been (one of) the culprit(s).

Regards
Dalai

 

[Zach]

In reality the 3000series 65W TDP CPUs are 88W PPT. But still compared to the 3900X of 105W TDP and 142W PPT can make a big difference in VRM temps.
PPT is the max power draw and TDP is just for the minimum cooler heat capacity for the Ryzens under specific conditions.

I agree that especially when an AIO water cooler is used, the airflow around the CPU socket should be amplified somehow. And even more when VRMs are a little weak.