Seeking help with PC Build that keeps resetting

To Whom may want to be Concerned with this nightmare

Recently i built a new Desktop pc

Buil info - https://pcpartpicker.com/list/VP88vW

the computer kept resetting when trying to install windows from a usb ( İ also tried other OS’s it still resets while installing the os) after some tinkering for reasons i still have not been able to figure out i managed to install windows, however the system resets itself with a Kernel-Power error 41(63)
details are below

Log Name: System
Source: Microsoft-Windows-Kernel-Power
Date: 03/11/2020 16:15:26
Event ID: 41
Task Category: (63)
Level: Critical
Keywords: (70368744177664),(2)
User: SYSTEM
Computer: DESKTOP-IRD1M09
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
Version>8
Level>1
Task>63
Opcode>0
Keywords>0x8000400000000002
TimeCreated SystemTime=“2020-11-03T13:15:26.4079847Z” />
EventRecordID>17358
Correlation />
Execution ProcessID=“4” ThreadID=“8” />
Channel>System

Computer>DESKTOP-IRD1M09
Security UserID=“S-1-5-18” />
/System>
EventData>
Data Name=“BugcheckCode”>0
Data Name=“BugcheckParameter1”>0x0
Data Name=“BugcheckParameter2”>0x0
Data Name=“BugcheckParameter3”>0x0
Data Name=“BugcheckParameter4”>0x0
Data Name=“SleepInProgress”>0
Data Name=“PowerButtonTimestamp”>132488828963579454
Data Name=“BootAppStatus”>0
Data Name=“Checkpoint”>0
Data Name=“ConnectedStandbyInProgress”>false
Data Name=“SystemSleepTransitionsToOn”>0
Data Name=“CsEntryScenarioInstanceId”>0
Data Name=“BugcheckInfoFromEFI”>false
Data Name=“CheckpointStatus”>0
Data Name=“CsEntryScenarioInstanceIdV2”>0
Data Name=“LongPowerButtonPressDetected”>false
/EventData>
/Event>

and sometimes when im playing a game instead of a full reset the game crashes triggering an event called Display 4101(Display driver amdkmdag stopped responding and has successfully recovered.) the details are as follows

Log Name: System
Source: Display
Date: 02/11/2020 22:35:09
Event ID: 4101
Task Category: None
Level: Warning
Keywords: Classic
User: N/A
Computer: DESKTOP-IRD1M09
Description:
Display driver amdkmdag stopped responding and has successfully recovered.
Event Xml:
Event xmlns=“http://schemas.microsoft.com/win/2004/08/events/event”>
System>
Provider Name=“Display” />
EventID Qualifiers=“0”>4101
Version>0
Level>3
Task>0
Opcode>0
Keywords>0x80000000000000
TimeCreated SystemTime=“2020-11-02T19:35:09.3368528Z” />
EventRecordID>16967
Correlation />
Execution ProcessID=“0” ThreadID=“0” />
Channel>System
Computer>DESKTOP-IRD1M09
Security />
/System>
EventData>
Data>amdkmdag
Data>
/Data>
/EventData>
/Event>

İ have tried various methods mentioned in other posts on this site and many others to no avail the only thing that worked for a bit was changing the PSU to an 350w non modular old PSU with it i could not connect the gpu but the system would no longer reset when playing a game or without the GPU installed however when i plugged the GPU to my 750w psu while the system was running of the 350w one it would still reset when playing a game or give the aforementioned display error i have changed the PSU under warranty with a new one but that did not solve the problem.

One other thing is i cannot see the post screen when the system is booting or the bios screen when connected to the gpu output and the internal output does not display anything the only way i can go into the bios is by unplugging the GPU and using the İGPU to get into the bios but as i said i can only get into the bios with this method as if i log into windows afterwards the system resets itself.

İ would greatly appreciate any and all input as a this stage a have no more ideas about what to do.

Ps: if i should move the post to somewhere else feel free to point it out this seemed like an appropriate place for it.

have you tried installing with a differet USB, maybe some data keeps getting corrupted

as far as bios goes, try changing “display first” to PEG or something similar

Sound like you narrowed it down to the GPU being part of the issue. Could be a power delivery or the GPU itself.

  • Try switching out the power cables, might be a bad power cable going to GPU
  • Try a different PCI slot on the motherboard incase of bad connection/bad slot
  • A little bit of a long shot, your motherboard had a new bios release in October, update that and try above again
  • Instead of running a game, you can load and run a benchmark - furmark and 3d mark will stress cpu and gpu. That may help consistantly finding the crash point.
  • Try running same cables/slot with another GPU (if you have an older GPU availible) See if something crashes same time with a different card.
  • If none of above work, see if you can RMA the GPU. That might be the issue.

Also guessing you are fully updated (Windows, drivers, game files), those could also be reasons for instability. Also guessing you are running stock speed on cpu and gpu (no over clocks). If these don’t apply, go to stock speed, update everything, reboot a few times and try again from there.

İ tried with a few different usb drives that did not change anything,

And as for changing the bios settings to post from the dGPU i will give it a try when i get back home.

edit: I checked the bios settings the IGPU was set to auto and the dGPU was already set as first device to post from, i changed the IGPU to enabled leaving the dGPU as the post device but it still would not post from it until the win_logon screen however if i plug the cable to the motherboard display output i can now get into the bios without unplugging the dGPU

For now i have been just unplugging the gpu from the psu which then lets me use the İGPU to boot into bios and for reasons i have not been able to figure out the computer does not reset while in the bios but if i boot into the os with the gpu unplugged it resets with a kernal-power error in a minute or two concistently.

Beyond that my main isue is the system resetting itself when playing a game with the gpu plugged in or giving the display error as mentioned above but if i dont try playing a game and just use the computer for rowsing the web, listening to musc etc. it does not reset and works normally if that could be said in this situation.

thanks for the fast replies

  1. İ tried the different cables that came with the psu along with different ports on the psu itself, no difference

  2. İ will try a different slot to see if that changes anything

Edit:Tried a different slot certain games still crash but the error message changed to a different one saying the application has been blocked from accessing the graphics hardware

Edit3: The pc still resets when playing some games but it is quite random one game does not cause resets for certain or crashes sometimes the same game can crash a few times than cause a reset or vice-versa

  1. İ updated the bios to the latest version already

4.İ will try running benchmarks to see if it crashes under load

Edit 2: I. ran furmark but it didnt crash or reset just in case i keot it running for longer than it takes most of the games i tried to crash but it ended up running smoothly for a while, and since i wasnt sure what to look for i closed it afterwards.

5.Unfortunatly i dont have another gpu so i can’t try this one

İ kept checking for windows updates until there was nothing left same fro the drivers, reinstalled games a few times as well.

Also everything is running with stock speeds and bios settings i have not changed anything yet.

Did you use the latest windows media creator to make the install media ? Seems there was a bug here recently.

I did get a fresh download at the time when i installed it, i have been trying to figure this out for quite some time and reinstalled windows a few times (getting a new installer every time just in case there were updates) but my last clean install was a while back so i dont know if i got the bugged version or an even earlier one.

Are you using anything to monitor temps ?

https://www.hwinfo.com/

tested memory

update bios/ default bios to stock settings.

It seems to point to gpu drivers. Maybe DDU is in order. Clean driver install / not allowing windows ten to update graphic drivers.

https://community.amd.com/thread/252725

İ was using aida64 to monitor temps, games crash early enough that it doesn’t cause the temps to rise noticeably and when i was running furmark the temps were stable around mid 30’s for cpu and around 70’s for gpu and as i said on a previous post Furmark did not crash or cause a reset

Bios is already updated as ı said on previous posts with everything running on stock settings.

İ tried different drivers for the gpu but i cant recall if i used DDU while installing the new drivers since its been a while so İ’ll give that a try when i have time along with MemTest86

Okay, thread says last post was 4 hrs. ago. I read through most of the thread but not every single post so ignore my suggestion if you’ve already tried this. Just for the heckuvit try reseating your CPU. Make absolutely sure there is no contamination in the socket and carefully inspect the pins. It may not even be your GPU but they do talk to each other and I once had a similar problem on a build that remedied this problem. Even though I found no contamination the reseating seemed to do the trick. Also keep all the hardware as minimized as possible so that you can narrow it down to precisely where you are having the problem and what triggers this kernel error. Not all GUID errors are generated by a graphics card. Just a suggestion. I hope this helps.

Certainly not a bad idea…and go over all the connections and make sure everything is seated correctly.

Thanks again for all the suggestions from everyone.

The problem was present the first time i put the system together as well so i tried running it with just the CPU and a single stick of ram and even tried running the ram at different slots and tried it with all 4 sticks separately just to make sure, but it was still resetting while trying to install the OS.

And İ have tried cleaning and reseating the CPU (İ used Isopropyl alcohol to clean the CPU pads and made sure there was no thermal paste or anything else stuck to the CPU, but i don’t think i checked the pins on the MOBO so i’ll give that a try when i get back.

Just got back a while ago

I ran MemTest86 it passed all tests with no errors, Also checked the MOBO CPU Socket i could not see anything abnormal with the pins or any contamination and just for the sake of it unplugged and plugged back all the PSU cables will check to see if the problem persists and let you know.

Edit: The problem still persists the games still crash with the Display error saying “Application has been blocked from accessing Graphics hardware.”

I just rand DDU and did a clean install of the GPU drivers as well but the problem still persists the games are still crashing but at some point the error generated changed back to “Display driver amdkmdag stopped responding and has successfully recovered.”

Out of ideas other than maybe looking all issues listed in windows event viewer . Look at all running tasks/applications for something dodgy / or causing issues. It is possible you have something not playing nice with other things. Device manager should tell you if a driver somewhere is having an issue and not working. I would make sure everything on your system /drivers / software / everything is fully up to date. A dodgy monitor cable or monitor not running in spec would also be something to check.

Device manager does not give any errors

But while i was messing around with the amd drivers to see if changing the settings would do anything i found something called amd link which would let me monitor the gpu from my phone and since i had no idea what else i could do, decided to take a look at it and it had a list of games which ı thought would give me performance information about the games like the adrenaline app on the pc,
instead it turned out to be a streaming app for phones and launched the game i clicked on so i decided to wait and see when the game would crash. But to my surprise as long as the amd link was open on my phone the game would not crash, However when i turned of the amd link app on my phone the game crashed after a few seconds like before, so i have no idea whats going on but at least now i am pretty sure the problem is caused by the amd drivers. though i have no idea how to fix it as i already ran ddu and did a clean install of the latest drivers.

1 Like

That is interesting… Another clean install and not using the link would be first guess. I would google around about AMD Link for other perspectives about this issue. Rolling back the previous known working drivers release is standard policy with such cases. In this case the latest released drivers may not help you.

https://community.amd.com/thread/246401 but this is old. beginning of the year.

Have look to see if the antivirus is flagging the app and shutting it down.

A little research about the drivers will be helpful to you.

Thanks for all the ideas
ill be going through the event viewer in detail to see if there is anything i missed around the time the errors were generated.

For now i did a clean install of the previous drivers but that didn’t seem to change much i still get the “Display driver amdkmdag stopped responding and has successfully recovered.” error at this point i’m starting to think something else might be causing the GPU or the drivers to fail, just to see if the resets were gone i removed the GPU after cleaning up the drivers but the machine still resets if i boot into windows without the GPU attached which is the most confusing part.

If the GPU or the drivers were causing the resets it should work fine without them being present, though the two issues may be completely unrelated as well i guess(the drivers causing the game crashes and something else within the system causing the resets)

also new drivers dropped.

I am not familiar with intels igpu but I make it is shutoff when using the amd gpu. I am completely out of ideas .

Problem Solved i changed the psu to a different model and so far the resets are gone including the one that was caused by the GPU being absent, and the crashes stopped as far as

1 Like