Consistent LiveKernelEvent 141 errors causing game crashes

Question

Hello,

I have been experiencing frequent LiveKernelEvent 141 errors sometimes accompanied by 117, and occasionally 1a8 and 1b8. My system is only about a year old, with the exception of a used 2080ti that I believe is causing the problems. I have tried seemingly everything I can find here in the forums about similar issues, but it hasn't worked for me and I would appreciate help in troubleshooting the problem.

System specs:

Win 11 Pro

AMD Ryzen 5 9600X

B850M Aorus Elite Wifi6e Ice

2x16gb Corsair Vengeance

2080ti FE

Corsair RM750x PSU

Ultrawide 1440p 144hz monitor

I have done many gpu driver reinstalls with DDU, including rolling back drivers or updating to latest. Tried both HDMI and DP cables. Gpu is powered by two separate PCIe cables. Uninstalled MSI Afterburner. Disabled MPO (I think). Manually set BIOS PCI speed to Gen 5, and then Gen 3 (2080 supports gen 3). Updated gpu bios with the nvflash program. Gpu voltage never went below 11.991v according to hwinfo even during a crash. Performed SFC and DISM commands, as well as windows memory check.

What is interesting is that this problem only appears at low to moderate load, and never during any benchmarks/stress tests/error checks. It is most reproducible by playing modded Minecraft, which crashes with exit code -1073740791. I thought it may be due to heat since the card is pretty old and the hotspot averages 15C over the gpu temp, but again, high stress games or programs don't cause the crash. When it crashes, the screen goes black (occasionally red), and my second monitor connected to integrated graphics freezes. After about 30 seconds, the screens flash and come back on, and the game will have crashed.

At one point I tested swapping in an old Radeon 6870 I had laying around, and after finding the right drivers, it seemed to perform fine without any crashes (albeit extremely slow and choppy). It seems like the problem has to be with the 2080 or the nvidia drivers.

I can provide log files if requested, I don't really know how to read them to find out what is the underlying problem. I am hoping it is not a gpu hardware failure and is unfixable as I don't have the budget for a new card right now.

Any help is appreciated!

Answer 1

The EXPO ram profile is currently enabled, but it was not always. It did not seem to make a difference to disable EXPO and I tested the ram with Windows Memory Diagnostic which did not report any issues.

I did try a mild undervolt with MSI Afterburner to reduce temps, only down to about 0.925v. It did run a bit cooler but did not increase stability. I tried uninstalling Afterburner entirely in case some other overclock setting that I didn't know about was interfering but that also didn't make a difference.

Answer 2

Everything appears OK here. Are you doing any sort of overclocking or XMP?

Answer 3

Thank you for your response!

Here is a screenshot of the system summary:

Answer 4

Hi,

It does point to a hardware defect, since the other GPU has worked and 117 error code directly indicates the GPU. Let us explore what we can do.

Please supply the following Information.

Right-click the Start button and Choose the Run option

Type in msinfo32

Press enter.

Take a screenshot of the System Summary window and post it in the next reply.

I look forward to your response and am eager to collaborate until we find a solution.

Answer 5

LiveKernelEvent 141 and 117 map to VIDEO_TDR (Timeout Detection and Recovery) issues, where the GPU or its driver stops responding in time. 1A8 and 1B8 are also GPU / display pipeline related live dumps. Given that:

• The issue only occurs with the RTX 2080 Ti (not with the Radeon 6870), and
• It happens under light/moderate, not heavy, load, this points to a stability problem with the 2080 Ti or its driver stack rather than Windows core components.

Use the following focused steps to isolate and mitigate:

1. Validate GPU driver and graphics stack

• Since DDU and multiple driver versions were already tried, focus on the guidance for VIDEO_TDR failures:
  • If a driver is identified in the bug check message or dump, disable it or check with the manufacturer for an updated version.
  • Ensure all graphics-related software (DirectX/OpenGL) and the affected games (e.g., Minecraft, mods, Java runtime) are fully patched.
• Confirm that the 2080 Ti is fully compatible with the current Windows 11 build and motherboard BIOS. If the card is older, ensure the motherboard BIOS and chipset drivers are up to date so PCIe power management and link training are stable.

2. Test in Safe Mode / minimal driver environment

• Use Safe Mode to help isolate third‑party software:
  • Go to Settings → Update and Security → Recovery → Advanced startup → Restart now.
  • Choose Troubleshoot → Advanced Options → Startup Settings → Restart.
  • Select option 4, 5, or 6 for Safe Mode.
• In Safe Mode, run a light 3D load (if possible) or at least leave the system idle with the 2080 Ti installed. If LiveKernelEvent 141/117 does not occur there, a third‑party overlay, capture tool, or tuning utility in normal mode is likely contributing.

3. Check for hardware stability and overheating

• VIDEO_TDR issues can be triggered by marginal hardware even if stress tests pass. From the TDR guidance:
  • Run the hardware diagnostics supplied by the system or GPU manufacturer.
  • Run Windows Memory Diagnostics (Control Panel search → “Diagnose your computer’s memory problems”) and review results in Event Viewer under System → MemoryDiagnostics-Results.
• Given the used 2080 Ti and the fact that a different GPU works, consider:
  • Inspecting and, if needed, replacing GPU thermal paste and cleaning the cooler (as in the LiveKernelEvent 1b8 case where dried thermal paste and overheating caused watchdog timeouts).
  • Monitoring GPU hotspot and VRAM temperatures specifically during the low/moderate-load scenarios that trigger crashes.

4. Power and board seating checks

• Even though PSU voltage looks stable in HWiNFO, perform physical checks:
  • Reseat the 2080 Ti in the PCIe slot.
  • Reseat or swap PCIe power cables (still using two separate cables as already done).
  • Power off, unplug, and discharge static electricity by holding the power button for ~30 seconds before reseating components (this helped resolve LiveKernelEvent 1b8 in a similar case).

5. Analyze dumps for TDR specifics

• For VIDEO_TDR-related codes (0x116, 0x117, 0x141, 0x1B8, 0x1A8):
  • Use the bug check references to interpret dumps and confirm the responsible module (typically the display driver, e.g., nvlddmkm.sys).
  • If the same GPU driver module is consistently implicated, treat the card or its driver as the root cause.

6. Decide on next steps if issues persist

• If:
  • Another GPU works fine in the same system,
  • Multiple clean driver versions behave the same,
  • Hardware diagnostics and thermal remediation do not change behavior, then the 2080 Ti is likely failing intermittently at the hardware level (often more visible at certain power states or clocks used under light/moderate load).
• In that case, options are:
  • Underclock/undervolt the 2080 Ti via vendor tools to reduce stress and TDR frequency.
  • If still unstable, plan for repair or replacement when feasible, as Windows and driver-level mitigations will be limited.

---

References:

• Bug Check 0x116: VIDEO_TDR_FAILURE
• Bug Check 0x117: VIDEO_TDR_TIMEOUT_DETECTED
• Bug Check 0x141: VIDEO_ENGINE_TIMEOUT_DETECTED
• Bug Check 0x1A8: VIDEO_DXGKRNL_BLACK_SCREEN_LIVEDUMP
• Troubleshooting Windows unexpected restarts and stop code errors
• Bug checks (stop code errors)
• Recurring "LiveKernelEvent 1b8" hardware error (Watchdog Timeout) - Microsoft Q&A
• My Windows 11 PC Hard Freezes for about 30 seconds to a minute when I play certain games. - Microsoft Q&A
• Kernel_Security_Error BSOD. Driver Verifier causing bootloop. - Microsoft Q&A
• BSODs caused by ntkrnlmp.exe - Microsoft Q&A