Analysis of minidump needed?
Analysis of minidump needed?
Hi,
So I helped my mate build a gaming PC and he said everything was ok until he started play Battlefield, he would be playing then the 2 screens would go black and the fans go full speed but he could still hear and talk to people on Discord.
Then last night it did the black screen and the fans go full speed again but this time when he was on Youtube, and again the audio continued ok
I looked at the PC today and updated the BIOS, updated chipset and other drivers and thought everything was ok, I had it playing youtue for about 45 mins no problem then plugged the mouse back in a USB port at the back of PC and black screen and the fans go full speed.
He has tried different nvidia drivers and used DDU to uninstall, I was tempted to go back to 566.36 but haven't yet
I checked with Whocrashed and it said hardware or potentially driver, so a bit vague really.....
😉
Spec
9800X3D
32GB RAM 6000MHz
5070Ti
Minidumps are here, shared from my OneDrive -
Minidump
Thanks
System specifications should be provided alongside the troubleshooting thread. Include details such as CPU, CPU cooler, motherboard, RAM, SSD/HDD, GPU, PSU, chassis, OS, monitor. Mention the age of the power supply and the BIOS version for the motherboard at this time.
I didn't set up Windows myself; I'll ask him about the installer source.
I obtained the installer from someone else. My friend installed Windows, but he only recalls using a USB drive without knowing its source.
From the dumps, the identical issues point to the WiFi7 adapter or its driver. The system struggles to switch from low power to high power (idle to running) within an acceptable time. The relevant IRP is visible, showing a failing driver (qcwlan64.sys)...
Great!!
Thanks for the clear and straightforward guide.
Yes, it's an onboard wifi adapter.
After the last BSOD, I've already updated the chipset, wifi and a few other drivers, so I just double-checked and everything is up to date.
I'll return the tower and see if he can fix it again.
😊
Thanks a lot
Can you help again please, my computer was working fine until the other day and then it started crashing again, 3 times in quick succession
Here are the new minidump files -
Minidump 2
Also can you recommend a minidump analysis guide so I can do this myself, I remember following one a few years ago but it wasn't very clear
Thanks
I don't believe there is a dump reading guide, not one that would be of much use anyway. Not only do you need to understand how to use WinDbg but, much more importantly, you need a really good grasp of Windows internals in order to understand what the dump is telling you. In addition there is a great deal of experience behind dumps analysis, I started reading IBM mainframe dumps almost 40 years ago for example. It's not something you can learn quickly TBH.
In all those recent dumps the problem is with the graphics card or driver. However, in the list of unloaded drivers, these are drivers that ended and unloaded immediately prior to the bugcheck, we see vgk.sys in every dump. From one of your earlier dumps we can see that vgk.sys is relatively recent...
Rich (BB code):
3: kd> lmvmvgk
Browse full module list
start end module name
fffff805`48b20000 fffff805`4a14b000 vgk (deferred)
Image path: vgk.sys
Image name: vgk.sys
Browse all global symbols functions data Symbol Reload
Timestamp: Fri Jul 11 00:07:48 2025
(68702BA4)
CheckSum: 0165430E
ImageSize: 0162B000
Translations: 0000.04b0 0000.04e4 0409.04b0 0409.04e4
Information from resource tables:
The vgk.sys driver is a component of the Riot Vanguard anti-cheat tool, it's used in several Riot games. Sadly, this driver is known to cause BSODs, all the anti-cheat tools do on some systems at one time or another. I note also that the Nvidia graphics driver nvlddmkm.sys was updated recently...
Rich (BB code):
1: kd> lmvmnvlddmkm
Browse full module list
start end module name
fffff803`91ca0000 fffff803`98806000 nvlddmkm T (no symbols)
Loaded symbol image file: nvlddmkm.sys
Image path: nvlddmkm.sys
Image name: nvlddmkm.sys
Browse all global symbols functions data Symbol Reload
Timestamp: Fri Aug 8 01:05:19 2025
(6895231F)
CheckSum: 0695CF57
ImageSize: 06B66000
Translations: 0000.04b0 0000.04e4 0409.04b0 0409.04e4
Information from resource tables:
It's not impossible that this latest graphics driver conflicts in some way with vgk.sys.
In order to see whether or not vgk.sys is the root cause you need to reboot, to ensure that vgk.sys is not in memory, and the use the PC as extensively as you can WITHOUT starting any Riot games at all (so that vgk.sys is never loaded). If it will BSOD without vgk.sys running then we need to look more closely, but if it doesn't BSOD then vgk.sys is a major suspect.
If it does appear to be vgk.sys then use
DDU
to uninstall the latest graphics driver and install whatever version you were using prior to that. If vgk.sys was stable with that earlier graphics driver then you may be better off sticking with that.
most recent memory dump was taken. Older versions were not reviewed.
the main issue is three distinct copies of AMDRyzenMasterDriver.sys are present (three separate dates). you must uninstall all three and install only the latest version.
remember to repeat this for your razer drivers as well.
AMDRyzenMasterDriver.sys Mon Sep 2 23:16:28 2024
AMDRyzenMasterDriver_fffff80592e40000 AMDRyzenMasterDriver.sys Thu Jul 11 22:53:41 2024
AMDRyzenMasterDriver_fffff80592e80000 AMDRyzenMasterDriver.sys Wed Oct 30 01:52:15 2024
download microsoft autoruns64.exe to remove the drivers. (removing these extra copies should be the first attempt)
check your video card and verify it is correctly seated in the PCI/e socket.
unexpected GPU removal? dxgkrnl!DpiFdoHandleSurpriseRemoval
(possibly just windows attempting to resolve a failing GPU)
razer drivers:
RzDev_0067.sys Mon Mar 22 18:25:06 2021
RzDev_00cc.sys Thu Jul 25 03:09:18 2024
RzDev_02a1.sys Wed Aug 3 03:26:14 2022
I removed the additional razer and ryzen master drivers as requested.
I believe this is the right approach since I stood up and knocked over the table where the tower was, and it performed normally—black screen and GPU fans running at full speed.
I swapped out the two thumbscrews that secured the GPU with regular screws, tested by hitting the table repeatedly; after about six or seven hits, it went black and the fans sped up.
However, no minidump files were generated.
I then tried hitting the table again more than twenty times, but this time it didn’t crash.
This motherboard has a new lock for the GPU, and it seems to be securely fixed.
Since hitting the table didn’t produce a crash, it’s likely the crashes are caused by something else.
If it fails again, I’ll reinstall Windows and the latest drivers, then run a stress test to rule out other software issues.