Question Very rare lockups with unknown cause

Fastfishy2

Honorable
Apr 20, 2020
128
7
10,585
I don't have much in the way of log files for this so I have to just describe it as best I can.

B550-A Strix (BIOS at the time was ASUS 2806 / AGESA 1207; has since been updated to Asus 3002)
RTX 4070, latest drivers
R7 5800X3D
32gb Ripjaws 3200mhz CL16 (16x2gb)
Boot drive: Samsung 860 EVO 500gb SATA
Cooler: Noctua NH-D15S Chromax w/ AS Ceramique-2 paste
PSU: Corsair RM750X 2021

CPU, GPU, and VRAM temps never even get close to concerning. RAM is fully inserted and is running at its rated speed due to me telling it to in BIOS. RAM Slots are populated as per motherboard manual too. GPU drivers are up to date as of when this happened.

The lockup itself:
Primary monitor display basically freezes to what it was last showing. I can bring up task manager onto my second monitor but total usages are pegged at a fixed number (in this case 69% for CPU and maybe 32% for RAM) and all the application tabs underneath have no Icons and show 0% under every category. In this case, I was running 3Dmark Port Royal stress test and it was at loop 18/20 when the primary monitor basically froze, although the PC otherwise seemed to be running normally. I had HWINFO64 running in the background.
I was able to shut down the PC using the start menu on my other monitor, so I didn't have to hard reset, and while I was clicking into it the Windows search bar flashed up a few times on the primary monitor, indicating it was still getting some sort of display updates. Only thing abnormal I noticed during shutdown was something called "DXGI-somethingsomething" in the list of programs waiting to be shut down. On reboot, I re-ran the same stress test without issue for all 20 loops, although 3Dmark neglected to flash up my result screen afterward (something it does sometimes, this doesn't worry me too much), this time without HWINFO running.

LOGS:
No WHEA errors have been logged and the only logs in Event Viewer around that time are 3 'error' logs, all under DistributedCOM with eventID 10010. All 3 mention a server response issue with DCOM and 2 of the 3 specifically mention Microsoft processes: MicrosoftOfficeHub and Microsoft.Windows.Search. I can reply with more details on them if this is needed.

I also checked my 3Dmark logs and although it has a log file for the relevant stress test that froze, I can't read it.

History
I haven't worried about this lockup because the last time it was happening was a couple months ago when I was playing The Last of Us Part I, and a couple of times it did this when I closed down the game. It never happened since, so I put it down to just being the game, and moved on. I'm a bit confused why it suddenly popped up again. I've run every manner of game, benchmark, and general PC use case in the meantime and it's never happened until today.

What I've done since then:
Updated motherboard BIOS
Updated AMD chipset drivers
Restarted PC
Ran 3Dmark Speed Way stress test w/HWINFO64 in the background (passed)
Ran Cinebench R23 w/ HWINFO in background (normal behaviour)



I really don't know if this is enough to go on but it's confusing the heck out of me. If anyone knows what it could be, I'm all ears.
 
Last edited:
Updated motherboard BIOS
Did you clear the CMOS for your motherboard after you'd verified that the BIOS was successfully updated to the latest?

RTX 4070, latest drivers
What GPuU were you working with prior to the RTX4070?

As for your GPU drivers, did you use DDU to remove all GPU drivers, then manually reinstalled with the latest drivers sourced off of Nvidia's support site?
 
Updated motherboard BIOS
Did you clear the CMOS for your motherboard after you'd verified that the BIOS was successfully updated to the latest?

RTX 4070, latest drivers
What GPuU were you working with prior to the RTX4070?

As for your GPU drivers, did you use DDU to remove all GPU drivers, then manually reinstalled with the latest drivers sourced off of Nvidia's support site?
I have not cleared the CMOS yet, and these crashes happened prior to the BIOS update. Do I need to clear my CMOS after a BIOS update? I can do that no problem.
I was using an RTX 3060ti and wiped the drivers with DDU when I put the 4070 in, then got my new drivers via geforce experience

EDIT: What I've done so far:
- Switched out boot drive Sata cable for another
- DDU'd my GPU drivers and reinstalled via Geforce Experience custom install
- Found out I'd manually set my ram to 3200mhz instead of using DOCP like I should have and letting that do the OC; it's on DOCP now.
- Reset CMOS via jumper pins and reapplied DOCP + Fan curves

So far the computer seems to be dandy so I'll just have to run some stress tests to check.
 
Last edited: