• On March 29th we will be migrating the Anandtech Forums to a new host. We expect the forums to go down at 13:00 GMT (05:00 PST, 08:00 EST) and anticipate a downtime of roughly five hours. Thank you for bearing with us during this maintenance period.

Question Computer freezes thinking it's my gpu

Neil Tropp

Junior Member
Dec 2, 2022
18
12
36
Hello, my computer freezes while playing games, the screen will freeze up and no inputs work I have to hard restart.I do not get a blue screen and the sound will freeze as well and repeat itself. I thought it was my PSU (rm750) but I installed my old one (cx650) which I know is good and I still have the same problem. I did a memtest on both rams sticks and they both passed. All bios setting are default except I have enabled ReBar. I installed all updates from the motherboard website and latest bios, windows updates and the latest gpu driver. I only have one minidump because I reinstalled windows 10 a few days ago. I have hwmonitor and all temps look good Iam running an aida64 stability test for gpu only for 3.5 hours now and no crashes, max temp on gpu was 81c hotspot. I'm guessing its my GPU but I am not sure.

I have uploaded memory.dmp , minidump, dxdiag, and msinfo to one drive.
https://1drv.ms/u/s!AmE-FdmvC7votlewoX--_C06OK-T?e=dIh7aU

Let me know if you need any more info.

Thanks.

Operating System: Windows 10 Home 64-bit (10.0, Build 19044) (19041.vb_release.191206-1406)
System Manufacturer: Gigabyte Technology Co., Ltd.
System Model: Z590 AORUS ELITE
BIOS: F6a (type: UEFI)
Processor: Intel(R) Core(TM) i5-10600K CPU @ 4.10GHz (12 CPUs), ~4.1GHz
Memory: 16384MB RAM
Available OS Memory: 16252MB RAM
Page File: 9255MB used, 17236MB available
Card name: AMD Radeon RX 6600
PSU: corsair cx650
 

Ajay

Lifer
Jan 8, 2001
12,362
5,895
136
Sorry, but I'm not going to click that link. I often start by looking at the system event just prior to the lockup for any clues. You should also run a RAM test.
1670017845464.png
 

Neil Tropp

Junior Member
Dec 2, 2022
18
12
36
I have run a memtest86 on each ram stick and they both passed 4 passes no errors. Should I run the built in microsoft ram test? I have attached a picture of my event viewer, the crash happened at 9:03 am. I also uploaded a picture of a previous crash in reliability manager that says hardware error.

How would you like me to upload my dump files as the site does not let me upload them.

Thanks.
 

Attachments

  • Like
Reactions: Leeea

Ajay

Lifer
Jan 8, 2001
12,362
5,895
136
So, you'll be wanting to look at critical errors. In the windows reliability monitor, there is a critical hardware error. I'm not used to using this tool - but you'll want to dig into that as it's likely the cause of your lockups.
 
  • Like
Reactions: JayMX and Leeea

Neil Tropp

Junior Member
Dec 2, 2022
18
12
36
I took my gpu out and re-seated it. I downloaded DDU went to safe mode and used DDU to uninstall amd drivers. I reinstalled the recommended drivers this time instead of the latest version. I will report back.

Thanks for your time.
 
  • Like
Reactions: Leeea and Ranulf

Neil Tropp

Junior Member
Dec 2, 2022
18
12
36
So, you'll be wanting to look at critical errors. In the windows reliability monitor, there is a critical hardware error. I'm not used to using this tool - but you'll want to dig into that as it's likely the cause of your lockups.
I did look into it and I found a.tmp folder that was the same time as the hardware error, I will upload the folder contents to one drive link below.
 
Last edited:

Neil Tropp

Junior Member
Dec 2, 2022
18
12
36
Well, I downloaded the dmp file - that link resolves to MS' live.com site. It's a video card/driver error. Just for giggles, you may want to reseat your GPU. Most likely, it's some sort of driver error (could be a conflict, not sure). I would use this tool: https://www.guru3d.com/files-details/display-driver-uninstaller-download.html
and then reinstall your drivers. Best guess I have at the moment.
Ever since I re-seated my gpu and reinstalled the drivers I have had no freezes or crashes, I have been playing games for the past 3 days and no issues. This seems to have fixed it so far but I will post again if I have another crash.

Thank you for your time and help, I appreciate it.
 

aigomorla

Cases and Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
20,026
2,409
126
One of his event errors had a Kernal-power.
It could of been this videocard seatted like you stated, but it could also point to a ton of other things.
I would keep checking logs and see if you have a repeat in the Kernal-Power error, as it could be an underlying issue of something else, and just resitting the GPU might of been a temporary fix.
 
  • Like
Reactions: Leeea

Neil Tropp

Junior Member
Dec 2, 2022
18
12
36
One of his event errors had a Kernal-power.
It could of been this videocard seatted like you stated, but it could also point to a ton of other things.
I would keep checking logs and see if you have a repeat in the Kernal-Power error, as it could be an underlying issue of something else, and just resitting the GPU might of been a temporary fix.
How do I monitor logs? and which logs?

Thanks.
 

aigomorla

Cases and Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
20,026
2,409
126
How do I monitor logs? and which logs?

Thanks.
Its the event viewer.png you uploaded in the second post.
You see under critical it says kernal-power (41)
This means hardware fault, or bad hardware.

It could mean as i said your GPU Was improperly seated like Ajax stated, or it could mean bad memory / pci-e slot / ram / cpu / bad overclock.

If you see that error again, it could require more diagnostics, if you never see that error again, then it was most likely the GPU being improperly seated.
 

Neil Tropp

Junior Member
Dec 2, 2022
18
12
36
Its the event viewer.png you uploaded in the second post.
You see under critical it says kernal-power (41)
This means hardware fault, or bad hardware.

It could mean as i said your GPU Was improperly seated like Ajax stated, or it could mean bad memory / pci-e slot / ram / cpu / bad overclock.

If you see that error again, it could require more diagnostics, if you never see that error again, then it was most likely the GPU being improperly seated.
Okay, I just checked and there has been no more kernal power errors or any other critical errors since that last one on December 2nd. I will keep an eye on it though.

Thanks.
 

Tech Junky

Platinum Member
Jan 27, 2022
2,349
804
96
One of his event errors had a Kernal-power.

You see under critical it says kernal-power (41)
That's just the hard power cycle. It's a good time stamp event though to pin down the actual event prior to that though.

A while back I was having spontaneous reboots on a laptop and pinned it down to a driver issue after a couple of months. The "kernel-power" error is a bit misleading when trying to pint things down. I went as far as swapping the power brick but, eventually just booted into Linux for a couple of days and the issue went away. Thus it's a WINDOWS issue not a HW issue. Since then I block windows updates through DNS completely and use a different program instead to check for updates. When I do let WU through it tends to install outdated drivers and mess things up.
 
  • Like
Reactions: JayMX

Ajay

Lifer
Jan 8, 2001
12,362
5,895
136
Is there a particular game that is causing the crash?
The last crash actually was initiated by the Steam web client making a directX call.

You can try this generalized procedure to see if Windows itself is needs fixing.
 

Neil Tropp

Junior Member
Dec 2, 2022
18
12
36
Is there a particular game that is causing the crash?
The last crash actually was initiated by the Steam web client making a directX call.

You can try this generalized procedure to see if Windows itself is needs fixing.
Yes it does only happen during one game that is on steam (lost ark) but this is the game I play the most so I didn't know if it was just a coincidence or part of the problem.

I installed my old gpu (gtx 780) and I called the place where I bought my gpu and it is still under their warranty and they will test it for free should I have them test it?
 

Neil Tropp

Junior Member
Dec 2, 2022
18
12
36
So far I have no crashes with my old gpu (gtx 780) and my games dont hitch or stutter like they used to on my rx6600. I will keep you updated.
 

Ajay

Lifer
Jan 8, 2001
12,362
5,895
136
Here is what happened (in part) from the dmp file:

DPC_WATCHDOG_VIOLATION (133)
The DPC watchdog detected a prolonged run time at an IRQL of DISPATCH_LEVEL
or above.
Arguments:
Arg1: 0000000000000001, The system cumulatively spent an extended period of time at
DISPATCH_LEVEL or above.
Arg2: 0000000000001e00, The watchdog period (in ticks).
Arg3: fffff807724fb320, cast to nt!DPC_WATCHDOG_GLOBAL_TRIAGE_BLOCK, which contains
additional information regarding the cumulative timeout
Arg4: 0000000000000000
Looking this stuff up, and a few other bits from the debugger, this is typically a problem with the driver. Weirdly, the problem seems to have been caused by a Steam web client call to directX - not the game (???). I couldn't find anything about this type of error occuring due to a hardware problem, but who knows. Sorry for your problems. I'm no expert when it comes to analyzing dump files and I don't know how to solve this.

Given your success with swapping back to the old GFX card, I'd suggest going with an nVidia card for your system, if you can. Seems like you're not having a very good experience with AMD on this particular system.
 
  • Like
Reactions: Neil Tropp

Neil Tropp

Junior Member
Dec 2, 2022
18
12
36
Here is what happened (in part) from the dmp file:



Looking this stuff up, and a few other bits from the debugger, this is typically a problem with the driver. Weirdly, the problem seems to have been caused by a Steam web client call to directX - not the game (???). I couldn't find anything about this type of error occuring due to a hardware problem, but who knows. Sorry for your problems. I'm no expert when it comes to analyzing dump files and I don't know how to solve this.

Given your success with swapping back to the old GFX card, I'd suggest going with an nVidia card for your system, if you can. Seems like you're not having a very good experience with AMD on this particular system.
I dropped my rx6600 off at the store they sad it would take up to 5 business days to test and get back to me so latest friday dec 16. I will report back with what they find.

A little more info:

After I re-seated my gpu last week ago I used DDU in safe mode to uninstall the gpu drivers and I reinstalled the recommended drivers as well as the AMD software: Adrenalin. All was great and no issues for 5 days (I would get a freeze everyday before this) until I tried to launch another game Forza Horizon 5, where it gave me the error (see uploaded forza driver error) and then I tried Battlefield 5 and it gave me the error (see battlefield 5 driver error pic), lost ark gave me no driver errors. What was weird was that in the adrenalin software it said my drivers where up to date and above the version that forza and battlefield where detecting.

So I removed the AMD software and the drivers and I let windows install the recommended driver for the gpu and forza and battlefield 5 both worked with no driver errors so I didn't install the drivers from the AMD website. Then later that day I had the same crash where it would freeze while playing Lost Ark and I had to hard reset. So I removed the old drivers and installed just the recommend drivers from the AMD website no AMD software and still had freezes.

The only time I went more then one day with no problems was with the weird driver install that worked for lost ark but the other games I tried said my drivers were out of date. How does this even happen? I'm so confused lol.

I really appreciate your time thank you for all your help so far.
 

Attachments

Last edited:

alexruiz

Platinum Member
Sep 21, 2001
2,832
554
126
I am late to the party.
Radeon drivers are extremely sensitive to system RAM stability.
If you have XMP enabled, disable it, reinstall adrenalin 22.11.1 WHQL from AMD's website, run the games again.
If RAM is the issue, running it stock will fix it.
 
  • Like
Reactions: VirtualLarry

Neil Tropp

Junior Member
Dec 2, 2022
18
12
36
I am late to the party.
Radeon drivers are extremely sensitive to system RAM stability.
If you have XMP enabled, disable it, reinstall adrenalin 22.11.1 WHQL from AMD's website, run the games again.
If RAM is the issue, running it stock will fix it.
I have tried it stock (default) still had issues and then I down clocked it to 2600mhz the max for my cpu is 2666mhz. I even put the voltage up to 1.35 v, the max for the Ram is 1.4v.
I have run windows memory diagnostics and memtest86 both passed no errors. It is default right now and has been for a while still have the freezes.

I will have to wait for the testing from the store I bought it at they have my gpu right now. should be done latest friday dec 16.

When I bought the parts for my system I was sure to look up the compatibility of the ram and motherboard maybe I made a mistake, did I do this right?
This is the module P/N: ax4u413338g19j-dt60
list of compatible ram here:

It is on that list but maybe I am missing something?
 
Last edited:

ASK THE COMMUNITY