Fix 'DXGI_ERROR_DEVICE_HUNG' in Games: 3 Causes

Cause #1: Corrupt or Conflicting GPU Drivers (Most Common)

Nine times out of ten, DXGI_ERROR_DEVICE_HUNG is a driver issue. Not just any driver issue — it's almost always leftover driver files from a previous install conflicting with the current one. Windows Update also likes to sneak in its own driver version, which can stomp all over your Nvidia or AMD driver.

Fix: Clean Driver Removal Using DDU

Don't bother with the normal "uninstall" in Add/Remove Programs. It leaves registry keys and driver store files behind. Use Display Driver Uninstaller (DDU) in Safe Mode. Here's how:

1. Download the latest DDU from Guru3D.
2. Boot into Safe Mode (hold Shift while clicking Restart, then Troubleshoot > Advanced Options > Startup Settings > Restart > press 4).
3. Run DDU, select your GPU vendor (Nvidia or AMD), and choose Clean and Restart.
4. Once back in normal Windows, download the latest driver for your card — but don't install the GeForce Experience or Adrenalin bloatware. Use the "Driver Only" option if available.

This kills the error for about 70% of people. If it still happens, move to cause #2.

Cause #2: Unstable GPU Overclock (Factory or Manual)

Here's the one people miss: even a factory overclock from MSI, ASUS, or EVGA can be unstable in certain games. The card might pass FurMark but choke on a DirectX 12 title because of how voltage drops under sustained load.

Fix: Underclock or Undervolt Your GPU

You don't need to go full stock. Just dial it back a bit:

1. Download MSI Afterburner (works on any brand).
2. Lower the Core Clock by -50 MHz and the Memory Clock by -100 MHz.
3. Apply and test the game. If the error stops, your card can't hold its stock boost. Leave the underclock in place or RMA the card.
4. For AMD cards, use the Adrenalin software to set a slight undervolt (say 1100 mV instead of 1150 mV) — this reduces heat and voltage ripple.

If you're running a manual overclock, that's almost certainly the culprit. Back it off by 25 MHz increments until stable. Don't skip this step just because your OC seemed fine in other games.

Cause #3: GPU Overheating or Thermal Throttling

Modern GPUs will hit TDR (Timeout, Detection, and Recovery) before they reach thermal shutdown. The driver literally says "I can't finish this frame because I'm too hot, so I'm resetting." That throws DXGI_ERROR_DEVICE_HUNG.

Fix: Check Temps and Improve Cooling

1. While gaming, run HWMonitor or GPU-Z in the background. Watch the GPU Hot Spot temperature — it often hits 105°C on cards that otherwise show 75°C on the main sensor.
2. If your GPU core is above 85°C or hot spot above 100°C, you've found the problem.
3. Clean the dust from your GPU heatsink. Use compressed air, hold the fans with a toothpick so they don't overspin, and blow from the back of the card outward.
4. Re-paste the GPU die if the card is over 2 years old. It's not hard — watch a teardown video for your specific model. Use Thermal Grizzly Kryonaut or Arctic MX-6.
5. Improve case airflow: front intake fans should push fresh air toward the GPU's intake fans. If your case has a solid glass front panel, remove it or run with the side panel off for testing.

Quick-Reference Summary Table

That's it. Three causes, three fixes. Start with the DDU cleanout — it works for most people. If it doesn't, you're looking at hardware instability or heat. And for the love of all that's holy, don't reinstall Windows unless you've exhausted these steps.