[SOLVED] Win 10 BSOD when gaming

Dec 11, 2020
14
0
10
0
Hi.
In the last month or so I've started getting BSOD "critical process died" errors, but ONLY whilst gaming.
In fact the message does change from crash to crash, but that seems the most common one.

Things that I've tried:
memtest which showed no errors,
reinstalling the video drivers
updated all of my other drivers using Drive Booster 8
Perfomed a windows 10 reset.

None of these seem to have helped.
On Cyberpunk 2077 (running at 1920x1080) I can't even get more than 5 minutes inbefore it happens.

PC Specs:

CPU: AMD Ryzen 7 2700X Eight Core/Sixteen Thread 4.35GHz Processor (Tray) YD270XBGM88AF
MOBO: Asus ROG Strix X470-F Gaming AMD X470 (Socket AM4) DDR4 ATX Motherboard
VIDEO: Gigabyte GeForce RTX 2070 SUPER Gaming OC 8192MB GDDR6 PCI-Express Graphics Card
RAM: Team Group Vulcan T-Force 16GB (2x8GB) DDR4 PC4-19200C14 2400MHz Dual Channel Kit - Grey (TLGD416G24
SOUND: Asus Strix SOAR 7.1 PCI-E Sound Card (90YB00J0-M1UA00)
DRIVES: Samsung 970 EVO Plus Polaris 250GB M.2 2280 PCI-e 3.0 x4 NVMe Solid State Drive
Seagate 1TB 2.5" BarraCuda Solid State Drive - System Stock
Seagate 4TB BarraCuda 5400RPM 256MB Cache Internal Hard Drive (ST4000DM004)
PSU: Kolink KL-850M 850W 80 Plus Bronze Modular Power Supply

Desperate for any suggestions at this point.
Thanks
D.
 

Colif

Win 10 Master
Moderator
updated all of my other drivers using Drive Booster 8
wish i had got to you before you did that, DB can make it worse.


Can you follow option one on the following link - here - and then do this step below: Small memory dumps - Have Windows Create a Small Memory Dump (Minidump) on BSOD - that creates a file in c windows/minidump after the next BSOD

  1. copy that file to documents
  2. upload the copy from documents to a file sharing web site,
  3. and share the link in your thread so we can help fix the problem
critical process died isn't a normal BSOD, its actually one that only happens if special windows files crash. there are a number of them so its not easy to finger exact cause.

right click start button
choose powershell (admin)
type SFC /scannow and press enter
once its completed, copy/paste this command into same window:

Repair-WindowsImage -Online -RestoreHealth
and press enter


SFC fixes system files, second command cleans image files, re run SFC if it failed to fix all files and restart PC

may not help but checks if windows is okay.
 
Dec 11, 2020
14
0
10
0
Hi. It doesn't seem to be creating a minidump.
I've got:
Default operating System: Windows 10
Time to display list of operating systems: 30 seconds
Write an event to system log checked
Automatically restart checked (unchecked also doesn't help)
Small memory dump (256 KB) on the drop down menu
Small dump dir = $SystemRoot%\minidump (which does exist at C:\Windows\minidump)
Overwrite any existing file checked

All other boxes are unchecked.
 
Dec 11, 2020
14
0
10
0
They've mostly been the same ("critical process died" ) but I have had an "Unexpected Store Exception" once or twice as well.
The machine has also just reset with a BSOD a couple of times.

Still no luck getting a minidump though
 

Colif

Win 10 Master
Moderator
DRIVES: Samsung 970 EVO Plus Polaris 250GB M.2 2280 PCI-e 3.0 x4 NVMe Solid State Drive
try running Samsung Magician on your nvme and run the tests it includes.

one of the files that is included as a critical file that causes that error is to do with storage and booting.

"Unexpected Store Exception" is a error in virtual memory. Virtual memory is essentially either ram or page file. Page file is probably on C

you have checked ram so lets check storage.

If its the nvme I can see why it doesn't create dumps. as it can't record them if the device it records them onto is the problem.

I doubt its the hdd but you can run this on them - https://www.seagate.com/au/en/support/downloads/seatools/seatools-win-master/
 
Reactions: Drexel_D
Dec 11, 2020
14
0
10
0
OK, the NVME did need a firmware update, but otherwise it was all good. The other drives seem fine too.
It's still BSODing though and not producing a minidump.
 

Colif

Win 10 Master
Moderator
does it crash in safe mode?

is it only cyberpunk? i hear its buggy but not sure about causing any bsod. likely a good stress test of any PC

can you run this, it creates a zip file, can you upload it to a file sharing website and show link. It might not help since PC isn't creating any dumps so it might not show me anything = https://www.sysnative.com/forums/pages/bsodcollectionapp/

I have to go soon as its getting late.
 
Dec 11, 2020
14
0
10
0
No worries, I'll run the exe and send you a link. Whenever you free to have to look. Thanks!

It first started when I got Star Wars Squadrons installed.
It also occurs with Witcher 3, but only since that first time. Back in May I was able to play Witcher 3 for hours at a time without problems.

I'll need to check about safe mode.
 

Colif

Win 10 Master
Moderator
WUDFHost - WIndows User mode Driver Framework Host Process is constantly crashing. its used mainly for USB devices
The Windows Driver Foundation (WDF) contains a framework for the creation of User Mode Drivers. The User Mode Driver Framework (UMDF) is designed to support protocol device classes such as cameras and portable music players. It integrates the installation and management of these devices with standard operating system facilities, such as I/O and Plug and Play and power management.
https://www.oreilly.com/library/view/windows-7-device/9780321670540/ch04.html

You have a headset showing as "device cannot start" that could be the cause of the WUDFHost errors.

this might not have anything to do with BSOD, its difficult to tell without a dump file.
 
Reactions: Drexel_D
Dec 11, 2020
14
0
10
0
I disconnected the headphones but unfortunately it still BSODed after 10 minutes.
This time I got an Unexpected Store Exception.
I turned off the autorestart so I could see the exception.
It says that it is collecting data but the counter always just stays at 0% (even after 30 minutes).

I guess I could install windows on my SSD and disable the nvme to see if that's the issue.
 

Colif

Win 10 Master
Moderator
Its hard to know what cause is when PC won't tell us.

that error can be hardware or drivers, but as its not creating a dump.... can only guess. not many software reasons for lack of a dump. the only software one is the page file and thats on nvme. if the page file is too small it won't make one.

install is a test I guess.
 

Colif

Win 10 Master
Moderator
have you ever changed your page file size or location?

do you have any 3rd party disk cleanup utilities like ccleaner?

can you take a photo of the next BSOD message? upload to an image sharing website and show here.

is that an event 161 from volmgr?

Volmgr.. guesses initials - Volume Manager. it looks after the disks is a basic way to say it... its taken me a while to find a basic description. so its having a non specific problem with the location of page file.
 
Dec 11, 2020
14
0
10
0
I have changed the page size once since this started in the hope it would make a difference. I set it to 24GB for initial size and 64GB for the maximum (1.5x and 4x mem), but all that happened was the machine just rebooting instead of BSODing, so I changed it back to system managed. See here:
https://drive.google.com/file/d/1HVP5EC6RPP4fitWdJwk4ApGYapok2i7Z/view?usp=sharing

Screen shot of the BSOD is here:
https://drive.google.com/file/d/1gOWHp5dmEvlV0egjy0EEc7sqLxPL_QQW/view?usp=sharing

Just for info, these are my current minidump settings:
https://drive.google.com/file/d/16OJHWdHJWYTyJBAifNmC7lfe1B73Aqhg/view?usp=sharing

I don't have ccleaner or any other 3rd party disk cleaners other than Norton 360.
I've disabled the part that deletes temporary windows files thinking that it might be removing the minidump, but I'm still not seeing one.

The event is a 161 from volmgr.
\Device\HarddiskVolume4
See here:
https://drive.google.com/file/d/1GsCTl8JDg1hikS-L0P7AZ4dzK1AemQWZ/view?usp=sharing
 

Colif

Win 10 Master
Moderator
grumbles at critical error not showing a file name.

system managed is fine. Nothing wrong with dump settings.

try installing on ssd and see if its same result I guess

unless @johnbl has any better ideas and doesn't mind me asking him.
 
Dec 11, 2020
14
0
10
0
OK, might have to leave it a few days as I use it for work as well (which I can do for the whole day without issue).
Thanks for everything.
 
OK, might have to leave it a few days as I use it for work as well (which I can do for the whole day without issue).
Thanks for everything.
you might want to google "how to force a windows memory dump via keyboard"
and make the registry settings and use your keyboard to force a memory dump.

This will confirm that they system is correctly setup for a memory dump.
ie you have the correct settings, have enough free space to save the dump file and you are looking for the file in the correct space.

Also, you should scan your system for *.dmp files. sometimes these will get migrated to windows error reporting and end up in a different directory.

there are many reasons that you will not get a saved memory dump even if you have the correct settings:
for example: your drive where the memory dump should be saved stops responding. this can happen with certain error, then windows resets the drive but
the drive can not reconnect for some reason. some time later (could be hours later if you have a lot of ram) you get a bugcheck.

for this you would want to put your device on a different sata port (if it is a sata drive) and you want to make sure the bios thinks the device can be removed. (even if it should never be removed) IE hot swap enable the port. This way if the port is reset then the bios will allow it to reconnect and you don't get the bugcheck later)

other examples, are when you plug into sata ports that have special purpose for your motherboard (most often the first or last port)

other issues can be the drivers not matching the actual hardware on your machine.

Unexpected Store Exception : this one is also hard to figure out without looking at the actual memory dumps. It can be bad drivers that write over other drivers memory locations. it can be poorly written drivers that talk to a service where the service just always assumes that the driver is never moved in memory. The windows memory manager is free to move the driver out of memory to virtual memory, when it gets move back in to RAM it may be in a different spot. Windows now compresses the memory so you can have drivers that have services that write over uncompressed data on top of compressed data and when windows memory manager attempts to uncompress it again it gets Unexpected Store Exception . Basically, any thing that the data goes thru can cause the corruption.
(cpu can corrupt the data, the memory timings can be bad and corrupt the data as it is being moved to storage, the storage controller can corrupt the data, the firmware of the SSD can corrupt the data. Other drivers with bugs next to the data can corrupt it in the PAGEFILE.sys or while it is in RAM.

looking at bunch of the memory dumps you can sometimes figure out the likely cause of the corruption. for example
if a memory address should be in kernel memory space but it looks like a user mode address then you look for a service/ bad programming or maybe malware. if the address looks like a small number then you look for a programming mistake or another driver that has over written another drivers data (kind of common)

not getting a actual dump file makes it hard to figure out. be sure to search for *.dmp files on your machine. if you have windows error reporting on it will detect problems and might do a automated live debug of your machine. by looking at these you can at least have an idea of what microsoft automated debugging machines think is wrong with your system.

I found one on a machine I set up a month ago, and It was debugging a intel wireless driver. I ended up downloading a fix from intel for it.
the one I looked at before that was microsoft debugging a USB controller. windows detected that the controller stopped functioning, and needed to have a bios update. The only symptom I was given by the owner of the machine was his machine would reboot in the middle of the day. (windows live debugger could not fix the problem until the bios was updated, so it rebooted the machine and the machine would work again until it hit the problem again) when i looked at it, it turned out to be a actual bug in a USB chip version that was on his motherboard and he had to only run the driver from the motherboard vendor. Newer generic drivers would not work. he could only use the old driver or just reboot his machine more often.
 
Last edited:
Reactions: Drexel_D and Colif
Dec 11, 2020
14
0
10
0
Just to confirm, I CAN manually trigger a dump and produce a minidump file, it's just when the machine BSODs in the wold so to speak.
The boot drive is an NVME drive, so I'm not sure that I can change the port.

I'll be trying a reinstall onto my SATA in a couple of days. I'll let you know how things go.
 
Just to confirm, I CAN manually trigger a dump and produce a minidump file, it's just when the machine BSODs in the wold so to speak.
The boot drive is an NVME drive, so I'm not sure that I can change the port.

I'll be trying a reinstall onto my SATA in a couple of days. I'll let you know how things go.
you might be able to add the second drive, change the memory dump options to dump the memory dump on the second drive rather than on the nvme drive
this way if the interface to the nvme drive breaks the bugcheck dump will still be stored.
 
Reactions: Drexel_D

Colif

Win 10 Master
Moderator
Makes me wonder where the blame lies. If it can't create a dump on a different drive, the problem occurs before it reaches the drive. I don't know where that somewhere is.
nvme might not be at fault since same thing happens on ssd - i might be victim blaming.
memtest shows no problems with ram
CPU & MB are involved.

guess we see what it does on ssd install.
 

ASK THE COMMUNITY

TRENDING THREADS