The VIDEO_TDR_FAILURE (0x116) bugcheck occurs when Windows detects that your NVIDIA GPU stopped responding and failed a driver reset within the timeout threshold. Resolve it by performing a clean driver wipe using Display Driver Uninstaller (DDU) in Safe Mode, inspecting 12V power stability and connector seating, or checking for unstable system RAM profiles.
The Causal Chain: Why BugCheck 0x116 Occurs
Windows relies on a subsystem called Timeout Detection and Recovery (TDR) managed by the DirectX graphics kernel (dxgkrnl.sys). When the GPU hangs during a compute or render operation and does not respond within the allocated operating window (typically two seconds), the OS intervenes to reset the display pipeline and avoid a hard system freeze.
If the display driver (nvlddmkm.sys) fails to recover during this reset sequence, the kernel executes KeBugCheckEx, generating a VIDEO_TDR_FAILURE (0x116) stop code. In less severe cases, this mechanism manifests as a temporary black screen accompanied by Event ID 4101 in the System event log: “Display driver nvlddmkm stopped responding and has successfully recovered.”
While software corruptions frequently cause this timeout, persistent 0x116 errors across clean driver installations point directly to power delivery deficiencies, PCIe bus communication errors, or faulty system memory corrupting driver instructions.
| Diagnostic Parameter | Typical Dump / Log Value | Technical Interpretation |
|---|---|---|
| Bugcheck Code | 0x00000116 |
VIDEO_TDR_FAILURE (display reset failed) |
| Faulting Module | nvlddmkm.sys |
NVIDIA Windows kernel-mode driver |
| Event Viewer Warning | Display Event ID 4101 | GPU driver timed out and attempted self-recovery |
| Event Viewer Error | volmgr Event ID 162 | Dump creation aborted due to hard system reboot |
Step 1: Clean Driver Removal via DDU
Overwritten or misconfigured display drivers frequently leave legacy DLLs, registry keys, and corrupted cache files that fail OS handshakes. Performing a complete removal in an isolated environment eliminates driver corruption as the root cause.
- Download the latest official NVIDIA Game Ready or Studio driver installer, along with the latest release of Display Driver Uninstaller (DDU). Disconnect your network cable or disable Wi-Fi to stop Windows Update from automatically pulling generic drivers.
- Boot Windows into Safe Mode (hold Shift while clicking Restart > Troubleshoot > Advanced options > Startup Settings > press F4).
- Launch DDU, select GPU as the device type and NVIDIA as the device, then click Clean and restart.
- Once rebooted into normal Windows, run the NVIDIA driver package, select Custom (Advanced), check Perform a clean installation, and complete setup.
Step 2: Inspect Power Delivery and 12V Rails
Modern high-draw GPUs—particularly RTX 40-series and high-end 30-series cards—exhibit aggressive transient power spikes. If the power supply cannot sustain dynamic response requirements or if rail voltage drops, the GPU core drops clock states abruptly, missing the TDR recovery window.
Inspect high-draw connectors (such as the 12VHPWR / 16-pin connector or standard 8-pin PCIe cables) for loose seating, backing pins, or thermal scorching. Poor contact resistance causes voltage drops under load that trigger driver dropouts.
Use HWiNFO64 to log your +12V input rail while running a load test. A drop below 11.4V indicates severe power supply sag that starves the VRM and induces driver timeouts.
To validate whether transient power spikes are causing your crash, use an overclocking utility such as MSI Afterburner to set the GPU Power Limit to 80% as a diagnostic baseline. If capping power draw restores stability under heavy workloads, your power supply unit lacks adequate transient handling capacity or the PCIe power cable assembly requires replacement.
Step 3: Test System RAM and Bus Stability
Although nvlddmkm.sys handles graphics processing, the driver actively maps commands through host system RAM. Corrupted memory cells in main system memory—not just GPU VRAM—frequently corrupt driver pointer addresses, causing immediate access violations.
Faulty XMP or AMD EXPO memory overclock profiles are a leading source of system-level TDR errors. Open your motherboard BIOS and revert memory to default JEDEC speeds (such as 4800 MT/s on DDR5 or 2133/2400 MT/s on DDR4) without the profile enabled.
If instability ceases, test individual DIMMs using MemTest86 or the Windows Memory Diagnostic (mdsched.exe). Physically isolate suspect modules by running a single stick in primary memory slots to determine if physical DRAM degradation is triggering the crash.
Step 4: Motherboard Firmware and PCI Express Power Settings
Outdated motherboard BIOS firmware can introduce severe ACPI DPC latency and PCIe bus synchronization faults, particularly when transitioning between low-load idle states and active 3D pipelines.
- Update your motherboard UEFI/BIOS to the latest stable revision using the manufacturer’s flash utility to address known PCIe bifurcation and link-state bugs.
- Update the chipset drivers directly from AMD or Intel to maintain proper PCIe root complex communications.
- In Windows, open Control Panel > Power Options > Change plan settings > Change advanced power settings. Expand PCI Express > Link State Power Management, and set it to Off.
Systems utilizing hybrid graphics configurations (CPUs with integrated graphics paired with a discrete GeForce GPU) occasionally experience driver arbitration conflicts. If crashes occur during simple desktop video playback or browser loading, disable the integrated graphics processor in Device Manager or the BIOS to test for driver conflict.