0XC0262003

0xC0262003: Graphics Adapter Reset – Quick Fixes That Work

Windows Errors Intermediate 👁 1 views 📅 May 28, 2026

Your display adapter crashed and reset. Usually a driver glitch or power issue. Start with the 30-second fix—it works most of the time.

The 30-Second Fix: Kill the Hung Driver

This error means Windows saw your display driver hang and force-reset it. Usually it's a one-time glitch. Press Win + Ctrl + Shift + B to restart your graphics driver. You'll see a black flash and hear a beep. That's it. Try reproducing the error now—if it's gone, you're done. I've seen this fix work for a client who was getting it every time they opened a 4K video editor. One keyboard shortcut and no more crashes.

If the error pops right back, or you see it every time you start a game or app, move to the next fix.

The 5-Minute Fix: Disable GPU Overclocking or Power Saving

Most cases of repeated 0xC0262003 come from an unstable GPU—either overclocked too far, or a power-saving profile that throttles it mid-workload. Here's what to check:

  1. Roll back any GPU overclock. Use MSI Afterburner or your card's utility (ASUS GPU Tweak, EVGA Precision) and set core clock and memory clock to defaults. Had a client last month whose card was stable in benchmarks but crashed in specific Unity scenes—defaults fixed it.
  2. Set your power plan to High Performance. Open Control Panel > Power Options, select High Performance. This prevents the OS from cutting power to the PCIe slot.
  3. In Nvidia Control Panel or AMD Adrenalin, set power management to Prefer Maximum Performance. For Nvidia: Manage 3D Settings > Power Management Mode > Prefer Maximum Performance. For AMD: Gaming > Global Graphics > Power Saving Off.

Test again. If the error's gone, you're set. If not, we go deeper.

The 15-Minute Fix: Clean Driver Install and TDR Adjustment

This error is often tied to a TDR timeout (Timeout, Detection, Recovery). Windows detects the driver hasn't responded in ~2 seconds and resets it. Some apps need more time. Here's the fix:

  1. Remove current driver completely. Download Display Driver Uninstaller (DDU). Boot into Safe Mode, run DDU, select Clean and restart. This nukes every trace of the old driver.
  2. Install the latest stable driver. Don't use beta drivers. For Nvidia, use Studio Drivers if you don't game—they're more stable than Game Ready. For AMD, avoid optional preview drivers.
  3. Increase the TDR timeout (if needed). Open Registry Editor (regedit) and go to HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\GraphicsDrivers. If the TdrDelay key doesn't exist, right-click > New > DWORD (32-bit) and name it TdrDelay. Set value to 8 (decimal). That gives the driver 8 seconds before resetting—enough for most heavy renders. Reboot.

If you still hit the error after this, it's likely a hardware issue. Check GPU temps with HWiNFO64—if you're hitting 85°C+ under load, you have thermal throttling. Reseat the GPU, clean the fans, or replace thermal paste. Rarely, the card itself is dying. But in 90% of cases, one of these three steps stops 0xC0262003 cold.

Real story: A small architecture firm called me—their rendering station crashed every 20 minutes with this error. Tried everything. Final fix was a clean DDU removal and setting TdrDelay to 10. The card was fine, just needed more time for complex raytracing jobs.

Was this solution helpful?