Question random crashes

Oct 1, 2022
19
1
15
i've been experiencing random crashes since i built my PC 2 years ago. i got it working after a year with very unusual crashes (3-4 times a year) with the dev version of windows 11. happened with windows 10 too. now they started again, every day, usually when playing but also when doing non-heavy tasks like web browsing. It seems something related to drivers because I tried replacing every component, one at a time and keeps doing it

dumps with verifier enabled
https://1drv.ms/u/s!AqnmCxeffsQMg31q2FDx1EkrrVrv?e=ctUOsS

system info
https://1drv.ms/u/s!AqnmCxeffsQMg37OlGxIdQXJCRZT?e=hmivwc
 

Colif

Win 11 Master
Moderator
Still on dev version? I see you aren't.
Conversion of dumps

report - Click run as fiddle to see report

File: 100122-4875-01.dmp (Oct 2 2022 - 04:34:10)
BugCheck: [SYSTEM_SERVICE_EXCEPTION (3B)]
Probably caused by: memory_corruption (Process: svchost.exe)
Uptime: 0 Day(s), 0 Hour(s), 00 Min(s), and 23 Sec(s)

File: 100122-11078-01.dmp (Oct 2 2022 - 05:00:35)
BugCheck: [DRIVER_VERIFIER_DETECTED_VIOLATION (C4)]
*** WARNING: Unable to verify timestamp for AmdTools64.sys
Probably caused by: AmdTools64.sys (Process: amdvbflash.exe)
Uptime: 0 Day(s), 0 Hour(s), 10 Min(s), and 45 Sec(s)

File: 100122-10281-01.dmp (Oct 2 2022 - 04:48:55)
BugCheck: [DRIVER_VERIFIER_DETECTED_VIOLATION (C4)]
*** WARNING: Unable to verify timestamp for AmdTools64.sys
Probably caused by: AmdTools64.sys (Process: amdvbflash.exe)
Uptime: 0 Day(s), 0 Hour(s), 00 Min(s), and 23 Sec(s)

DV blames AmdTools64.sys
May 27 2020AmdTools64.sysAMD Special Tools driver
its possibly part of the GPU drivers, its used to flash the GPU BIOS

Try running DDU in safe mode, remove all AMD drivers, log back into normal and reinstall drivers
top crash also relates to GPU drivers.
 

Colif

Win 11 Master
Moderator
its possible its not part of the gpu Drivers

did you look in settings/apps and see if its listed there? might be able to just uninstall it.

Download https://learn.microsoft.com/en-us/sysinternals/downloads/autoruns and run it as admin
it shows every item that auto loads with windows.
GO to drivers tab
find AMDTools64.sys and either untick it which will stop it loading with windows or right click it and uninstall it from inside program.
 

Colif

Win 11 Master
Moderator
The date on top of the dumps is when they happened, they all seem to be today.
Conversion of dumps

report - Click run as fiddle to see report


File: 100122-5187-01.dmp (Oct 2 2022 - 11:33:39)
BugCheck: [WHEA_UNCORRECTABLE_ERROR (124)]
Probably caused by: AuthenticAMD (Process: System)
Uptime: 0 Day(s), 1 Hour(s), 05 Min(s), and 23 Sec(s)

File: 100122-5187-02.dmp (Oct 2 2022 - 12:22:10)
BugCheck: [WHEA_UNCORRECTABLE_ERROR (124)]
*** WARNING: Unable to verify timestamp for amdkmdag.sys
Probably caused by: AuthenticAMD (Process: FIFA23.exe)
Uptime: 0 Day(s), 0 Hour(s), 23 Min(s), and 02 Sec(s)

i could have done without WHEA errors
at least its blaming the AMD GPU drivers. they normally not so revealing. the number of itmes I see a driver or anything in a WHEA error is less than 1% so I will take that.

WHEA - Windows Hardware Error Architecture
Error called by CPU but not necessarily caused by it.
Can be any hardware, can sometimes be hardware drivers.
I normally say more about what it could be but yours pointed at GPU drivers.
Can be caused by heat so how hot is it inside PC?


I can't tell you if its just the driver or the card itself.
So if you run ddu multiple times and it keeps happening, where are you getting drivers from?
IF AMD, instead of getting them from there again, run windows update after exiting safe mode and see if the older drivers that Microsoft have from AMD work any better

Do you have another PC you can try GPU in and see if it BSOD in it as well?
 
Oct 1, 2022
19
1
15
Can be caused by heat so how hot is it inside PC?
no, I checked the temperature. CPU no more than 50 celsius and GPU no more than 70

So if you run ddu multiple times and it keeps happening, where are you getting drivers from?
from and. i tried with the recommended (WHQL) and the latest. also with window ones. now I'm trying with the pro version

Do you have another PC you can try GPU in and see if it BSOD in it as well?
i don't. but I assume it's driver related. it's the same kind of issue that 1 year before. that time, I tried the GPU in another PC and didn't have any issue
 

Colif

Win 11 Master
Moderator
Apart from AMD drivers, you only have 3 others
Mar 11 2020amdgpio2.sysAMD GPIO Controller Driver from Advanced Micro Devices http://support.amd.com/
Jun 25 2021AMDRyzenMasterDriver.sysAMD Ryzen Master driver
Jul 19 2021RTKVHD64.sysRealtek Audio System driver https://www.realtek.com/en/
Aug 17 2021amdxe.sysAMD Link Xinput Emulation driver
Oct 28 2021AtihdWT6.sysAMD High Definition Audio Function driver http://support.amd.com/
Nov 03 2021amdsafd.sysAMD Streaming Audio driver
Dec 10 2021amdfendr.sysAMD Crash Defender Service driver
Dec 10 2021amdfendrmgr.sysAMD Crash Defender Manager driver
Jan 14 2022amdgpio3.sysAMD GPIO Controller Driver from Advanced Micro Devices http://support.amd.com/
Apr 27 2022CtiIo64.sysCreative Audio Driver
Apr 28 2022amdkmdag.sysAMD Graphics driver
Jun 01 2022AMDPCIDev.sysAdvanced Micro Devices PCI Device driver
Jul 21 2022rt68cx21x64.sys
Aug 19 2022GUBootStartup.sysGlary Utilities Startup Manager (Glarysoft Ltd.)
Oct 01 2022eaanticheat.sys
can update chipset drivers = https://www.amd.com/en/support/chipsets/amd-socket-am4/b450
doubt either realtek drivers be cause
Doubt its Easy Anti Cheat to blame.

It seems something related to drivers because I tried replacing every component, one at a time and keeps doing it
3 different OS with errors. Have they all been with same GPU?

I don't think this description is right
Apr 27 2022CtiIo64.sysCreative Audio Driver
I have been trying to work out what it really is though. Its even on my pc. And I don't have any creative sound devices... (getting de ja vu, feel I have looked at this before)
 
Oct 1, 2022
19
1
15
can update chipset drivers = https://www.amd.com/en/support/chipsets/amd-socket-am4/b450
I did. tried with the latest version and also, the previous one from the gigabyte pate

3 different OS with errors. Have they all been with same GPU?
same GPU but also tried with a Nvidia 2070 with a clean Windows install

idk. doesn't make any sense. i replaced every component 1 by 1 a year ago. same error
all the components work just fine in other pcs. I kept getting the same kind of crash
tried with windows 7, 10 and also ubuntu. same
touched every bios option, CPU, GPU, ram. undervolt, underclock. same
but for some reason worked in w11 dev for a year, with the original components. just 3 crashes (same kind) in a year
installed FIFA 23 and started again (I've been playing the 22 without any problem)
installed 22h2. same
i said drivers because of the windows dev working. but I'm not sure
I'm very close to throw this PC to the trash
 
Oct 1, 2022
19
1
15
same motherboard each time?
no. had an msi 550 that i don't remember the model but changed it for the gigabyte and still same error

Ram - 16gb
2 hyperx 8gb 3600 (running without xmp at 2400)

PSU?
cooler master 750 bronze, brand new. had a power cooler 1000
 

Colif

Win 11 Master
Moderator
replaced CPU?
Is there anything you haven't replaced?

Need to figure out where to start

Have you run Prime95 on CPU?
Prime 95 Bootable - https://www.infopackets.com/news/10113/how-fix-bootable-prime95-stress-test-hardware
Prime 95 Instructions - https://appuals.com/how-to-run-a-cpu-stress-test-using-prime95/

Try running memtest86 on each of your ram sticks, one stick at a time, up to 4 passes. Only error count you want is 0, any higher could be cause of the BSOD. Remove/replace ram sticks with errors. Memtest is created as a bootable USB so that you don’t need windows to run it
 
Oct 1, 2022
19
1
15
replaced CPU?
yep, for a ryzen 5 1600

Is there anything you haven't replaced?
i don't think so. 2 different SSD, rams, disconnected all the cables. nothing
also tried running all outside of the case, over a box, just to make sure there wasn't anything touching and causing a short

Have you run Prime95 on CPU?
i didn't. I'll try
btw I left heaven benchmark running last night for 7 hours. the PC is still running it

Try running memtest86 on each of your ram sticks, one stick at a time, up to 4 passes. Only error count you want is 0, any higher could be cause of the BSOD. Remove/replace ram sticks with errors. Memtest is created as a bootable USB so that you don’t need windows to run it
I pass the windows one without errors. that counts?
 

Colif

Win 11 Master
Moderator
memtest runs more tests than the windows does. It can test ram windows locks up due to some areas being protected while its running

Run both overnight, as both Prime & Memtest take a long time. Not on same nights.

I have asked if others have any ideas.
 
Oct 1, 2022
19
1
15
i've been able to play fifa 23 for over 2 hours without any issue. could not pass the 30 minutes before
I have to go out so I'll let the memtest running with the 2 rams. then I'll try one by one and the prime tonight
 

Colif

Win 11 Master
Moderator
The problem with testing 2 sticks is if you do have an error, you can't tell which stick was to blame. So it could be a waste of time. True, if they don't have errors it could show they fine... but i don't know if I would bother.
 

Colif

Win 11 Master
Moderator
Conversion of dumps

report - Click run as fiddle to see report


File: 100222-5109-01.dmp (Oct 3 2022 - 06:13:12)
BugCheck: [WHEA_UNCORRECTABLE_ERROR (124)]
Probably caused by: AuthenticAMD (Process: System)
Uptime: 0 Day(s), 0 Hour(s), 21 Min(s), and 42 Sec(s)

so that one was caused by a device on PCI Express not working... my guess would be GPU given whats happened before
PRIMARY_PROBLEM_CLASS: 0x124_AuthenticAMD_PCIEXPRESS

does it crash in safe mode?
run prime yet?

Summary:
CPU Had 1600, has 2600
MB: Had MSI 550, has GB Aorus B450 Elite
Ram: had 2 Hyperx 3600, has Kingston KHX3600C17D4/8GX
GPU: Has RX 5500 XT, had 2070
Storage: 250gb sandisk, I assume you tried another drive
PSU: has Cooler master 750 bronze, had Power cooler 1000?
Cooling? what fans have you got?
What case?

did I miss anything?
 
Oct 1, 2022
19
1
15
does it crash in safe mode?
not sure, I'll check

run prime yet?
yep, for a few hours. no errors nor crashes

Summary:
CPU Had 1600, has 2600
MB: Had MSI 550, has GB Aorus B450 Elite
Ram: had 2 Hyperx 3600, has Kingston KHX3600C17D4/8GX
GPU: Has RX 5500 XT, had 2070
all right

Storage: 250gb sandisk, I assume you tried another drive
yep

PSU: has Cooler master 750 bronze, had Power cooler 1000
yep

Cooling? what fans have you got?
had 3 120mm fans at the front, 1 at the back and this one on the cpu
https://www.idcooling.com/Product/detail/id/47/name/SE-903
stock cooling on the gpu

What case?
have the MAG VAMPIRIC 010

did I miss anything?
i dont't think so. i keep thinking why works fine on the dev version. whats the difference? different TDR config or way to handle driver errors?
 

Colif

Win 11 Master
Moderator
i never used Dev version, I ran insiders for a while just before they released 11. I don't know what the differences are as Dev more close to truly new and untested compared to the beta channel. Could be features they haven't quite finished playing with in it. Could be it handles errors differently but you should still get them, only difference I see would be in how they reported.

So we basically have to test everything but it be nice if it stopped blaming GPU. I don't expect you have the 2070 still?
Shame neither CPU has an internal GPU... but they becoming rare now.

You don't have any dongles that you always use? although its unlikely, all hardware in PC is linked via same motherboard chips, so its possible a USB device could disrupt the GPU.

tried running with bare minimum parts?