Fix 0xC0262101: Graphics can't lock memory on Windows 10/11

Quick answer for pros

Run DDU in Safe Mode to fully remove the GPU driver, then reinstall the latest driver from NVIDIA or AMD. That fixes 80% of these cases. If it doesn't, check for a memory leak in a game or app using GPU-Z.

What the hell is this error?

Error 0xC0262101 (ERROR_GRAPHICS_CANT_LOCK_MEMORY) happens when DirectX or Vulkan tries to lock video memory for rendering, and the GPU driver can't pin that memory down. The system returns a "timeout, detection, and recovery" (TDR) or a crash. I've seen this on everything from a GTX 1060 to an RTX 4090, across Windows 10 22H2 and Windows 11 23H2.

The culprit here is almost always a corrupted or mismatched driver. Second most common: a memory leak in the game or app that exhausts VRAM. Rarely, it's a hardware defect — bad VRAM chips or an overheating card that's causing memory controller errors.

Fix it step by step

1. Run Display Driver Uninstaller (DDU) in Safe Mode. Boot into Safe Mode with networking. Download DDU from Guru3D. Run it, select "Clean and restart" for your GPU vendor (NVIDIA or AMD). This nukes every trace of the driver — registry keys, leftover DLLs, the works. Don't skip this. Normal uninstallers leave junk.
2. Install the latest driver fresh. After reboot, grab the newest Game Ready or Pro driver from NVIDIA or the Adrenalin driver from AMD. Do a clean install (check the box if NVIDIA, or use Factory Reset if AMD). Don't use Windows Update drivers — they're often stale.
3. Test with a different workload. Run a known stable game or a benchmark like Unigine Heaven. If the error only appears in one game, the issue is that app. Check for patches or updates. For example, I've seen this happen in Starfield before a patch fixed it.
4. If it persists, check VRAM usage. Open GPU-Z and watch the Memory Used graph. If it's hitting 95-100% during the error, you've got a leak. Close background apps (Chrome tabs with video, Discord hardware acceleration). Lower texture quality in the game.

Alternative fixes when the main one fails

• Disable hardware-accelerated GPU scheduling. Go to Settings > System > Display > Graphics > Default graphics settings. Turn off Hardware-accelerated GPU scheduling. Reboot. I've seen this cause memory lock issues on some AMD and Intel GPUs.
• Increase TDR timeout. This is a registry tweak. Open Regedit, go to HKLM\SYSTEM\CurrentControlSet\Control\GraphicsDrivers. Create a DWORD called TdrDelay, set it to 10 (decimal). Reboot. This gives the driver more time to lock memory before Windows kills it. Rarely helps, but when it does, it's a lifesaver for older cards.
• Underclock VRAM slightly. Use MSI Afterburner. Drop the memory clock by 100-200 MHz. Test. If the error stops, your VRAM might be borderline unstable from the factory or overheating. This is a band-aid, but it works for some RTX 3080s with Hynix memory.

Prevention tips

Don't let your GPU run over 85°C for extended periods. Heat kills memory stability. Keep drivers updated, but don't install every beta release — stick with Game Ready or WHQL drivers. Monitor VRAM usage with MSI Afterburner OSD; if a game eats 90%+, drop textures a notch. Also, avoid overclocking VRAM unless you've stress-tested it with OCCT's VRAM test for an hour.

If you've done all this and still get 0xC0262101, you're likely looking at bad hardware. Run MemtestG80 or OCCT's VRAM test in loop mode. Failing that, RMA the card. I've had to do that twice in 14 years — both were faulty VRAM chips that no driver ever fixed.