VIDEO_TDR_FAILURE: Video Driver Crashes (How to Fix)

What Does Error 0x000000116 Mean?

The Timeout Detection and Recovery (TDR) mechanism is built into Windows to prevent a complete system freeze when the graphics processor fails. When a video driver does not respond to operating system requests within a specified time (usually 2 seconds), Windows forcibly stops the driver, clears the video memory, and restarts it. You see the characteristic message in the lower-right corner: "Display driver stopped responding and has recovered".

The error manifests as a brief black screen, interface flickering, game crashes, or File Explorer freezes. In the Event Viewer (eventvwr.msc), the codes 0x00000116 (VIDEO_TDR_FAILURE) or 0x00000117 are logged, along with references to files like nvlddmkm.sys (NVIDIA) or atikmpag.sys (AMD).

Causes

• Driver version conflict. Residual files from a Windows Update installation overlap with components from the vendor's official package.
• Insufficient TDR timeout. Modern games and rendering tasks require the GPU to take longer than 2 seconds to process a frame. The Windows mechanism mistakenly interprets this as a hang.
• Aggressive overclocking or undervolting. Unstable voltage or increased core/memory frequencies cause hardware ECC errors that the driver cannot handle correctly.
• Overlay and background utility conflicts. Discord, MSI Afterburner, RivaTuner, or screen recording programs intercept the video stream, creating race conditions in the driver.
• Insufficient PCIe bus power. The power supply experiences voltage droop during sudden load spikes, causing the graphics card controller to reset the connection.

Solutions

Method 1: Clean Driver Installation Using DDU

Automatic installers often leave behind corrupted registry entries. A full cleanup resolves the issue in most cases.

1. Download the Display Driver Uninstaller (DDU) utility and the latest driver for your card from the NVIDIA, AMD, or Intel website.
2. Disconnect from the internet (to prevent Windows from automatically downloading a driver) and restart your computer.
3. Boot into Safe Mode: Start → Settings → System → Recovery → Advanced startup → Restart now → Troubleshoot → Advanced options → Startup Settings → Restart → press 4.
4. Run DDU, select GPU and your vendor, then click Clean and restart.
5. In normal mode, run the downloaded installer. During the process, choose Custom Installation and check the box Perform a clean installation.

Method 2: Increase TDR Timeout in the Registry

If the driver is stable but heavy in-game scenes trigger a reset, increase the response timeout limit.

1. Press Win + R, type regedit, and confirm running as administrator.
2. Navigate to: HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\GraphicsDrivers
3. Right-click in the empty space on the right pane → New → DWORD (32-bit) Value.
4. Name the value TdrDelay (for 32-bit systems) or TdrDelay (DWORD also works for 64-bit).
5. Double-click the newly created value, select the Decimal base, and enter 8.
6. Restart your computer. Windows will now wait 8 seconds before forcibly resetting the driver.

💡 Tip: If error 0x00000116 is accompanied by a complete hard freeze, do not increase the value above 10. This may indicate a hardware defect or insufficient PSU power.

Method 3: Disable PCI Express Power Saving

Aggressive power schemes put the graphics card slot into a low-power state during idle. A sudden load spike causes a current delivery delay, triggering a reset.

1. Open Control Panel → Power Options (or type powercfg.cpl in the search bar).
2. Next to the active plan, click Change plan settings → Change advanced power settings.
3. Expand the PCI Express branch → Link State Power Management.
4. Set both values (On battery / Plugged in) to Off.
5. In the same branch, find GPU Power Management (if available) and set it to Maximum Performance.
6. Click Apply and close the window.

Prevention

• Use WHQL-certified drivers. Beta versions and Game Ready updates marked "experimental" may contain regressions. Monitor stable releases on support forums.
• Monitor thermal conditions. GPU overheating above 85°C causes throttling and memory instability. Regularly clean radiators of dust and reapply thermal paste if the card is over 3 years old.
• Disable overlays during diagnostics. If the problem disappears without Discord, Xbox Game Bar, or MSI Afterburner running, update the conflicting software or disable hardware acceleration in its settings.
• Check your power supply. For modern mid-range and high-end graphics cards, a power reserve of at least 20% above peak system consumption is recommended.