Fix Display Driver Stopped Responding in Windows 11

•
Ankit Kumar
Windows 11DisplayDriversTroubleshootingFix

The Problem

The message "Display driver stopped responding and has recovered" appears after the screen flashes black for a second or two. That is Windows' Timeout Detection and Recovery (TDR) doing its job: the graphics driver failed to respond within roughly two seconds, so Windows reset it instead of freezing the whole PC.

The recovery is a symptom, not the fault. What actually happened is one of: a driver bug or a corrupted driver install, the GPU running outside stable clocks or voltages, the card overheating or starved of power, or a faulty PCIe connection.

Screen goes black and never comes back? That is a different failure — see fix a black screen on boot. Getting a blue screen instead of a recovery message? See fix VIDEO_TDR_FAILURE.

Symptoms

  • Screen goes black for one to three seconds, then returns with a notification balloon.
  • Games minimise, crash to desktop, or report a "device removed" / DXGI_ERROR_DEVICE_REMOVED error.
  • Event Viewer logs Display source events with "Display driver nvlddmkm / amdkmdap stopped responding and has successfully recovered".
  • Happens more often under load, or after the PC has been running for a while.

Step 1: Clean Install the Graphics Driver

A driver installed on top of a broken one keeps the broken parts. Removing it completely is the single most effective fix here.

  1. Download the current driver for your exact GPU from NVIDIA, AMD, or Intel first, so you have it offline.
  2. Follow the full removal procedure in clean install GPU drivers with DDU — boot into Safe Mode, run Display Driver Uninstaller, reboot, then install the driver you downloaded.
  3. Choose the clean installation option in the vendor installer and skip the extra bundled software.
  4. If the problem began right after a driver update, install the previous version instead of the newest one. Newer is not always more stable for older cards.

Step 2: Rule Out Overheating

  1. Install HWiNFO or the vendor tool (NVIDIA App, AMD Adrenalin) and watch GPU Temperature and GPU Hot Spot during a game.
  2. Sustained core temperatures above 83 °C, or hot spot deltas above about 20 °C, mean cooling or paste.
  3. Blow dust out of the card's fans and the case filters with the PC off: clean your laptop and devices safely.
  4. Check the case has an intake path — a GPU starved of air throttles then stalls. Related: fix loud fan noise and fix CPU overheating and thermal throttling.

Step 3: Remove Any GPU Overclock — Including the Factory One

  1. Close MSI Afterburner, EVGA Precision, or the vendor tuning panel, and reset every profile to default.
  2. Make sure the tool is not reapplying a profile at startup: Task Manager → Startup apps, disable it while testing.
  3. On NVIDIA cards, try a small negative offset to the core clock (for example -100 MHz) or a curve capped slightly lower — many cards ship boosting past what they are stable at.
  4. On AMD cards, in Adrenalin → Performance → Tuning, disable any automatic overclock and set the power limit back to zero.
  5. If the card is stable when clocked slightly lower, the silicon is boosting too aggressively — that is a tuning fix, not a Windows fault.

Step 4: Check Power Delivery to the Card

  1. Shut down, unplug the power, and hold the power button for five seconds.
  2. Confirm each PCIe power connector on the card is fully seated and clicked in.
  3. Use separate cables from the PSU where the card needs two connectors — daisy-chaining one cable with two ends overloads a single rail on higher-draw cards.
  4. On 12VHPWR / 12V-2x6 connectors, push until it clicks and confirm no gap is visible.
  5. A PSU that is old or borderline for the card causes transient drops that show up exactly as TDR resets.

Step 5: Reseat the Card and Test Another Slot

  1. With the PC unplugged, unscrew the bracket, release the PCIe latch, and remove the card.
  2. Clean the gold contacts gently with isopropyl alcohol on a lint-free cloth, and let it dry.
  3. Reseat firmly until the latch clicks. If the board has a second x16 slot, try it.
  4. Remove any PCIe riser cable from the equation while testing.

Step 6: Turn Off Hardware Acceleration in Browsers and Chat Apps

If the resets happen on the desktop rather than in games, an accelerated app is usually involved.

  1. Chrome / Edge: Settings → System → turn off Use graphics acceleration when available, then restart the browser.
  2. Discord: Settings → Advanced → turn off Hardware Acceleration.
  3. Microsoft Teams / Office: check the equivalent hardware acceleration setting in each app's options.
  4. Test for a day. If the resets stop, re-enable them one at a time to find the offender.
  5. Related: fix Chrome crashing or not responding.

Step 7: Check Windows Power and Graphics Settings

  1. Settings → System → Power & battery → Power mode → Best performance. On a laptop, set this for Plugged in at minimum.
  2. In the NVIDIA Control Panel → Manage 3D settings → set Power management mode to Prefer maximum performance for the affected game.
  3. Settings → System → Display → Graphics → Change default graphics settings, and try toggling Hardware-accelerated GPU scheduling off — some driver versions are less stable with it on. Background: hardware-accelerated GPU scheduling.
  4. Related: optimize Windows 11 for gaming.

Step 8: Repair Windows and Confirm It Is Not the Panel or Cable

  1. In Terminal (Admin): 
    sfc /scannow

    DISM /Online /Cleanup-Image /RestoreHealth
  2. Reboot and re-run sfc /scannow. Details: how to run SFC and DISM.
  3. Swap the display cable and try a different port (DisplayPort instead of HDMI, or the reverse). A marginal cable at high refresh rates causes blackouts that look identical.
  4. Related display faults: fix screen flickering and flashing and fix a second monitor not detected.

What Not to Do

  • Don't edit the TdrDelay registry value to extend the timeout. It hides short stalls and turns them into full freezes; it fixes nothing.
  • Don't install the driver over the top of the broken one and call it clean — use DDU.
  • Don't assume the newest driver is best. For older cards the previous branch is frequently more stable.

Still Not Working?

If a clean driver, stock clocks, good temperatures, and solid power connections do not stop it, test the card in another PC — or test another card in yours. If the fault follows the card, it is failing. If it stays with the PC, suspect the PSU or the motherboard slot.

Related: fix low FPS and stuttering in games and fix NVIDIA Control Panel missing.

Related guides