Question Been trying to fix for 7 days! Persistent different DPC Watchdog Crashes RTX 4090/MSI 321URX OLED

Moonhawkmanton

Reputable
Jun 15, 2020
2
0
4,510
I recently got the new MSI OLED Monitor. Plugged in and ran Cyberpunk 2077 ran it in background while watching a video on setting up HDR, started using the Monitor settings on the actual monitor and got a blackscreen freeze and audio freeze crash. When it rebooted, it had to go into recovery. It repaired itself enough to start up but I noticed that it thought Cyberpunk, Icloud, and a few other programs were no longer on the c: drive. Since then I have been getting persistent DPC Watchdog crashes during gameplay only, internet and video playback is fine. Please help I am getting OCD'd out right now and have lost about a week of free time trying to correct this.

I have restore windows and updated fully from 10 to 11

System Configuration

  • Operating System: Windows 10 (first) now 11
  • GPU: Zotac RTX 4090
  • CPU: Intel i9-9900K
  • PSU: Corsair RM1000x
  • Cooling: Air cooling with case fans and GPU/CPU fans
  • Monitor: Two monitors tested, including an AOC U34G2G4R3 and a new MSI 321URX QD-OLED

Issues Experienced

  • System Crashes: Occurred during gameplay across multiple games, including Cyberpunk 2077, Green Hell, and Hogwarts Legacy.
  • Inconsistency: Crashes often happen before the 1-hour mark but not always, making it difficult to pinpoint the exact cause.

Steps Taken

1. GPU Testing

  • FurMark Stress Test:
    • Ran FurMark (GL) and FurMark Knot (GL) for over an hour with no issues. GPU was stable, with temperatures peaking at 73.8°C.
    • GPU memory clock consistently at 2625 MHz during gameplay, and VRAM usage peaked at 7753 MB.

2. CPU & PSU Testing

  • OCCT Power Supply Test:
    • Ran the test for 1 hour with CPU max MHz at 4551.9 and 100% utilization. No errors detected.

3. Memory Testing

  • MemTest86:
    • Ran MemTest86 with no errors detected.

4. Driver Reinstallation

  • NVIDIA GPU Drivers:
    • Used DDU (Display Driver Uninstaller) to uninstall GPU drivers and performed a clean installation of the latest NVIDIA drivers.
  • Windows Event Viewer Logs:
    • Analyzed Event Viewer logs, with critical errors such as Event ID 41 (Kernel-Power), Event ID 1001 (BugCheck), and others pointing to potential hardware or driver issues.

5. Crash Analysis

  • BlueScreenView & WhoCrashed:
    • Analyzed crash dumps, with results pointing to ntoskrnl.exe, which typically indicates underlying driver or hardware issues.
    • BugCheck code 0x00000133 identified as DPC_WATCHDOG_VIOLATION, often related to driver or hardware issues.

6. General Troubleshooting

  • HWiNFO Monitoring:
    • Monitored GPU, CPU, and other system metrics during gameplay and stress tests, looking for anomalies leading up to crashes.
  • Tested on Different Monitors:
    • Tested with both the new HDR monitor and the older AOC monitor, with crashes occurring on both.
  • Checked BIOS & Firmware:
    • Verified that BIOS and firmware were up to date.
  • Checked Windows Updates:
    • Ensured that Windows 10 was fully updated.
 
Last edited:

rustigsmed

Distinguished
i would try a different cable. then i would try different ports on the gpu and then monitor. are you using those monitors in a dual setup? maybe just try one at a time til you get sorted. also that new monitor had two firmware updates since june with a heap of fixes have you applied them? if the psu is from the original 9900k build it may be getting a bit old now - possibly on the way out - i'd try and test out another one after you exhaust your easier options.