Fix nvlddmkm.sys Errors on Windows (NVIDIA)

If you’re seeing nvlddmkm.sys errors on Windows, the fix usually involves updating or cleanly reinstalling your NVIDIA graphics driver, but the underlying cause can also be overheating, unstable overclocks, or a failing GPU. This article walks through the most effective solutions step by step, from driver reinstallation to hardware checks.

What Is nvlddmkm.sys and Why Does It Crash?

nvlddmkm.sys is the NVIDIA Windows Kernel Mode Driver, version 1.0. It’s the core driver component that lets Windows communicate with your NVIDIA graphics card for rendering, video playback, and compute tasks. When it fails, you typically see a Blue Screen of Death (BSOD) with a stop code like VIDEO_TDR_FAILURE (0x116), VIDEO_SCHEDULER_INTERNAL_ERROR (0x119), or a message saying “nvlddmkm.sys has stopped responding and has successfully recovered.”

The error means the GPU didn’t respond to the Windows Display Driver Model (WDDM) within a set timeout (usually 2 seconds). The system then attempts to reset the driver. If the reset succeeds, you get a notification; if it fails, the system crashes.

Common Causes of nvlddmkm.sys Errors

Corrupted or outdated NVIDIA drivers – incomplete updates, leftover files from old drivers, or a bad install.
Overheating GPU – dust-clogged fans, dried thermal paste, or poor case airflow.
Unstable overclocks – either from factory settings or manual overclocking tools like MSI Afterburner.
Power supply issues – insufficient wattage or a failing PSU can cause voltage drops under load.
Faulty GPU hardware – memory or VRAM defects, often worsened by heat.
Windows updates or system file corruption – a damaged OS can interfere with driver loading.

Step-by-Step Fixes for nvlddmkm.sys Errors

Start with the simplest solution and work down. Do not skip steps unless you have a clear reason.

1. Clean Reinstall of the NVIDIA Driver

Download the latest driver for your GPU model from NVIDIA’s official website. Use the “Advanced Driver Search” if you have an older card.
Run DDU (Display Driver Uninstaller) in Safe Mode. This removes all traces of previous drivers.
After DDU finishes, reboot normally and install the downloaded driver. Choose “Custom” installation and check “Perform a clean installation.”
Restart your PC and test with a stress tool like FurMark or a demanding game.

If the error persists, try an older, stable driver version. Sometimes the latest driver has bugs; NVIDIA often releases hotfixes or Game Ready drivers that address specific issues.

2. Check GPU Temperatures and Cooling

Download a monitoring tool like HWiNFO64 or MSI Afterburner. Run a stress test and note the maximum GPU temperature. For most modern NVIDIA cards, anything above 90°C (194°F) under load is concerning. If temperatures are high:

Clean the GPU fans and heatsink with compressed air.
Ensure case fans are working and airflow isn’t blocked.
If the card is old, consider replacing thermal paste (this requires disassembly and voids warranty on some cards).

3. Disable Overclocking

If you’ve overclocked your GPU (via Afterburner, EVGA Precision, or BIOS), revert to stock clocks. Also, check if your GPU has a factory overclock; try underclocking slightly with a tool like MSI Afterburner (e.g., -50 MHz on core and memory) to see if stability improves.

4. Update Windows and Check System Files

Go to Settings > Update & Security > Windows Update and install all pending updates.
Open Command Prompt as administrator and run sfc /scannow to repair system files.
Then run DISM /Online /Cleanup-Image /RestoreHealth to fix Windows image issues.

5. Adjust Power Management Settings

In the NVIDIA Control Panel, go to Manage 3D Settings > Power Management Mode and set it to “Prefer Maximum Performance.” Also, in Windows Power Options, set the plan to High Performance. This prevents the GPU from downclocking aggressively, which can cause TDR errors.

6. Test with a Different GPU or Power Supply

If none of the above works, the issue may be hardware. Try your GPU in another PC, or swap in a known-good GPU. Also test with a different PSU if possible. A failing PSU can cause intermittent crashes under load.

7. Check for Overheating of the Power Delivery (VRM)

Some GPUs have weak VRM cooling. If you have a card with a blower-style cooler, it’s more prone to VRM overheating. Use HWiNFO to monitor VRM temperatures if available. If they exceed 100°C, improve case airflow or consider an aftermarket cooler (if applicable).

What to Do If the Error Occurs Only in Specific Games or Apps

If nvlddmkm.sys crashes happen only in one game, the issue might be the game’s graphics settings or a conflict with overlays (Discord, GeForce Experience). Try:

Disable in-game overlays.
Lower graphics settings, especially anti-aliasing and shadow quality.
Verify game file integrity (Steam/Epic).
Check if the game has a known issue with NVIDIA drivers – search for the game name + “nvlddmkm” to see if others have found a specific driver version that works.

Advanced Troubleshooting: Event Viewer and Memory Dumps

To pinpoint the exact cause, check the Windows Event Viewer. Press Win + X and select Event Viewer. Go to Windows Logs > System and look for errors with source “Display” or “nvlddmkm” around the crash time. The error details may indicate a specific hardware failure (e.g., “GPU at PCI bus 1, device 0, function 0”).

Also, analyze the memory dump file from the BSOD. Use a tool like BlueScreenView to read the dump. It will show the bug check code and which driver was involved. If the dump points to nvlddmkm.sys, it’s almost certainly a GPU/driver issue; if it points to another driver, the problem might be elsewhere (e.g., a faulty RAM stick causing data corruption).

Frequently Asked Questions

What does “nvlddmkm.sys has stopped responding and has successfully recovered” mean?

This is a TDR (Timeout Detection and Recovery) event. Windows detected that the GPU didn’t respond within the timeout period, so it reset the driver. It’s a warning sign that the GPU is unstable, but not necessarily fatal. If it happens repeatedly, follow the fixes above.

Can a faulty PSU cause nvlddmkm.sys errors?

Yes. If the PSU cannot deliver stable voltage under load, the GPU may crash. Symptoms often appear during gaming or benchmarking. Test with a known-good PSU if possible.

Is nvlddmkm.sys a virus?

No. It’s a legitimate NVIDIA driver file located in C:\Windows\System32\drivers. However, malware can sometimes disguise itself with similar names. If you’re concerned, run a full antivirus scan and verify the file’s digital signature (right-click > Properties > Digital Signatures).

How do I update nvlddmkm.sys manually?

You don’t update the file directly. You update the NVIDIA driver package, which includes nvlddmkm.sys. Use GeForce Experience or the NVIDIA website to download the latest driver.

Will resetting Windows fix the error?

It can, if the cause is corrupted system files or driver conflicts. But if the issue is hardware (overheating, failing GPU), a reset won’t help. Try all other fixes first before resorting to a full reset.

Should I disable the TDR timeout to fix the error?

Some guides suggest increasing the TDR delay via registry (TdrDelay). This can mask the symptom by giving the GPU more time to respond, but it doesn’t fix the underlying issue. Only use this as a temporary workaround while troubleshooting.

When to Replace the GPU

If you’ve tried all software fixes and the card still crashes under load, the GPU may be dying. Signs of hardware failure include:

Artifacts (visual glitches) on screen.
Crashes even at stock clocks and low temperatures.
Errors in Event Viewer pointing to hardware memory (e.g., “nvlddmkm” with a memory address).

Before replacing, test the GPU in another PC with a fresh Windows installation. If it crashes there too, it’s almost certainly faulty. If the card is under warranty, contact the manufacturer for a replacement.

General steps for current mainstream versions; menus can differ by device and version. See the disclaimer or report an error.

More in this bed

The whole Tech How-To bed →