Display drivers/dwm crashing and freezing screens...

Question

Hiya, first off, sorry if this is long and fluffy. I'm not good at being concise.

I've recently (Jan 2021) updated my platform from an lga1155 to a AM4 system, and I'm running into display driver/dwm crashes which freeze what's on the screen but keep the PC running.
I can hear videos/games/chat continue, and I can manipulate them. I just can't see anything.
To gain full control back I have to turn it off with the PSU switch as just smothering the power button causes it to crash at the windows loading icon, the little pale blue one, but the system continues to boot to desktop behind a frozen screen.

This is what event viewer looks like during or around the time of a crash:

The info 1001s read:
Fault bucket 2112677666287471416, type 4
Event Name: APPCRASH
Response: Not available
Cab Id: 0

Problem signature:
P1: dwm.exe
P2: 10.0.19041.746
P3: 6be51595
P4: KERNELBASE.dll
P5: 10.0.19041.662
P6: ec58f015
P7: e0464645
P8: 000000000010bd5c
P9:
P10:

Attached files:
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER91CB.tmp.dmp
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER922A.tmp.WERInternalMetadata.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER923B.tmp.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER9245.tmp.csv
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER9255.tmp.txt

These files may be available here:
\?\C:\ProgramData\Microsoft\Windows\WER\ReportArchive\AppCrash_dwm.exe_eef8fd3d8dc253702159e243a3766f344950851c_c2f10bab_13270018-379d-4e05-8c6a-c9392e6710a3

Analysis symbol:
Rechecking for solution: 0
Report Id: 0761daf8-d534-4aee-a972-7fa6a261b507
Report Status: 268435456
Hashed bucket: 472699661c87084acd51bd24dff74b38
Cab Guid: 0

The Error 1000s read:
Faulting application name: dwm.exe, version: 10.0.19041.746, time stamp: 0x6be51595
Faulting module name: KERNELBASE.dll, version: 10.0.19041.662, time stamp: 0xec58f015
Exception code: 0xe0464645
Fault offset: 0x000000000010bd5c
Faulting process ID: 0x2a70
Faulting application start time: 0x01d6fd981280edc8
Faulting application path: C:\WINDOWS\system32\dwm.exe
Faulting module path: C:\WINDOWS\System32\KERNELBASE.dll
Report ID: 0761daf8-d534-4aee-a972-7fa6a261b507
Faulting package full name:
Faulting package-relative application ID:

The Warning 0s read:
The Desktop Window Manager process has exited. (Process exit code: 0xe0464645, Restart count: 6, Primary display device ID: NVIDIA GeForce GTX 1080)

and the Info 9027s read:
The Desktop Window Manager has registered the session port.

I cannot figure this out. I've rolled back drivers, uninstalled all I can and reinstalled them. Tried 3 different (current, last and a few versions back) Nvidia drivers using DDU each time.
I've run sfc/scannow and a few other commands to wipe and repair Windows update and check that all the Windows files are fine, they all report that everything is fine, or that they corrected something and when run a second or third time come up clean.
I've booted to safe mode a few times, not spent much longer than a few minutes there so I can't really tell if it happens in that environment.
I've rebuilt my W10 install USB stick with an updated version from the Windows support page and run Windows start up repair.
I've tested the RAM, using windows memory tester and also using 1 stick, 2 sticks in each socket
I've reseated the CPU, and also replaced the CPU with a new one
I've reseated the GPU a few times as it's a recently new build and I had to go back to my old one for a few days during a return window(old pc runs fine, no issues)
I've reseated all cables at both ends
I've remounted the motherboard to make sure nothing was touching underneath
Basically I've taken it all apart twice and put it back together.
I've use the DP to HDMI cables between 2 machines no issues
All hard drives have been used across both machines, no issues

I've faffed about uninstalling a much fluff drivers/software as I can. So many RGB things these days
I check temperatures and make sure they don't get too high. A bit much actually, I hate how spiky the ryzen chips are.
I've checked power draw for cpu and gpu, they seem alright.
I've mined some crypto fine, bit warm but fine.*
I can play games, watch videos, browse the web no issues*

*after a random(?) time it'll crash no matter what I'm doing.

I want to point a finger at the 20H2 update but I've no grounds to base that on.
I found a folder, C:\ProgramData\Microsoft\Windows\WER\ReportArchive, that has loads of AppCrash_dwm.exe folders, I assume logs? I can't view them as I don't have nor can I gain permission. Odd that.

I reinstalled windows the day after I finished putting this machine together, I think to fix this same issue, but obviously it didn't work and like I said, I put together a newer W10 install usb stick. It's the last thing I want to do but I'm willing to reinstall again. It's just a bit of a let down doing all that for the same issue to raise it's head again.

More than happy to supply driver numbers or any info needed. Thank you for taking your time to read... this mess :)

Answer 1

Having the same issue, Different hardware, In addition to all the stuff the OP posted i too have had this issue. To bring light on what the OP see's i will read it better for him.

1. What originally happens is windows has a LIVE kernel event crash.
2. Then the desktop windows Manager logs a service restart in the event logs as "informational". Most people look only for errors but this one is always missed since it's tagged as info. What it means is the service died and has restarted it's self (Normally a very bad thing)
3. The the very next Second you display driver crashes since it's dependent on the DWM (point number 2)
4. Your game Crashes since it sees the "Display device has been removed" since the driver was forcibly Stopped.

I only have the issue when playing games. Some days I can go 4hrs no issues others it takes 5 min and can happen and on logout and back in I can then play for another 3-4 hrs. I have replaced power supply and replaced memory GSkill CL14 4000 downclocked to 3600 with matching FCLK/UCLK/MCLK, No XMP all manual clocking. Updated to latest BIOS for Gigabyte.

Also tweaked using the TDRDELAY, besides disabling the option no change. There are alot alot alot of people with the same issues as we have. It's normally found on anyone who runs on a display at high refresh rate 144 and up. I am on a Samsung G9 5120 x 1440 240Hz

AMD Ryzen 9 3900x, 32 GB GSkill, AMD RX 6800 XT, Gigabyte X570 MB

**The Guess is from all the different forums is windows has some issues with high refresh rates and the amount of data over Display port 1.4. 2x I had display corruption having to DDU my driver when using Display power savings on. Had to turn it off to prevent driver issues.

Answer 2

I have same issue with Windows 11 and GeForce 1050TI and it seems there is no cure

Answer 3

I'm having the same issue, started a week ago(17/12/2021). Black screen only when opening games. So far, i tried everthing you did too.

Build:
Ryzen 5 1400,
Gigabyte AB350m - Gaming 3,
8GB RAM Ballistix 2400Mhz,
1050TI GALAX.
PSU EVGA 500W

Answer 4

Hey all, I have the same issue happening to my computer. It just started doing it last night and now I am unable to play games. I can surf the web, stream movies and music with no issues but as soon as I run a game, my display crashes but I can still hear audio.

Windows 11 Home, Ryzen 7 3700x, 16GB RAM, GeForce RTX 2060. Two monitors, 32" 120hz.

Answer 5

These last days this problem has been happening to me many times