Interactive Troubleshooter

Published: Oct 30, 2026

VIDEO_TDR_FAILURE BSOD When Playing Video: TDR Delay Diagnostic

Time: 15 mins Level: Beginner

By StepDiag Engineering Team

Reviewed by Ibraheem Taofeeq Opeyemi

Comprehensive root-cause diagnostic tree for resolving VIDEO_TDR_FAILURE (0x00000116 / 0x00000117) Stop Code errors during GPU-accelerated video decoding. Resolves kernel-level driver hangs, hardware acceleration conflicts, registry TDR timeout thresholds, and PCI express power state latency.

Full Diagnostic Tree & Step-by-Step Overview

What exact pattern or system state accompanies the VIDEO_TDR_FAILURE (0x116 / 0x117) crash during video playback?

Crash occurs specifically in web browsers (Chrome, Edge, Firefox) or media apps (VLC, Movies & TV) while streaming video.
Crash references specific driver sys files (nvlddmkm.sys, atikmpag.sys, or igdkmd64.sys) immediately after GPU power state changes.
System freezes, stutters, or displays screen artifacts/flicker for 2 seconds before throwing the BSOD.
BSOD occurs during high-resolution (4K/8K HDR) decoding, multi-monitor playback, or custom GPU overclock/undervolt setups.

Which application environment or video decoding pathway triggers the browser or media player TDR freeze?

Chromium-based browsers (Chrome, Edge, Brave) crashing during HTML5 video stream playback (YouTube, Netflix, Twitch).
Desktop media players (VLC, MPC-HC, Windows Media Player) crashing during local hardware-decoded MKV/MP4 playback.
Hardware acceleration crashes occurring exclusively when hardware-accelerated GPU scheduling (HAGS) is enabled in Windows.
Video playback crashes accompanied by web browser ANR (Application Not Responding) and GPU process restarts.

Which specific driver sys file or power state transition is cited in the crash logs right before the VIDEO_TDR_FAILURE stop code?

Crash points to nvlddmkm.sys during NVIDIA driver power state transitions or PCIe ASPM wake events.
Crash points to atikmpag.sys or amdkmdap.sys during AMD Radeon power management or Zero RPM fan toggles.
Crash points to igdkmd64.sys during Intel Iris Xe / Arc integrated graphics power gating or display sleep recovery.
Crash occurs during hybrid graphics (MUX/Optimus) switching between integrated and discrete GPUs during media playback.

What specific visual anomaly or system stutter pattern appears right before the 2-second freeze and TDR crash?

Screen stutters or drops frames severely during smooth scrolling or video playback across multiple monitors.
Display flashes black momentarily, recovers once, and then hard freezes into a VIDEO_TDR_FAILURE BSOD on the second attempt.
Screen flickers with horizontal lines or green/purple macroblock artifacts before locking up.
Audio continues playing normally in the background for 10-15 seconds while the video display remains completely frozen.

What heavy workload, resolution setting, or overclocking condition is active when the TDR failure strikes?

Playback of 4K/8K HDR AV1/HEVC streams with Windows HDR or Auto-HDR enabled.
Active GPU core/memory overclock or aggressive undervolt causing voltage instability during low-frequency video states (P-States).
PCI Express bus overclocking or riser cable signal attenuation operating at PCIe Gen 4/Gen 5 speeds.
Insufficient power supply transient response or 12VHPWR/PCIe power rail droop during sudden GPU load changes.

Fix Chromium Browser Hardware Acceleration & ANGLE Backend Timeout

Solution:

Root Cause: Chromium ANGLE Graphics API & HW Decode Timeout

When streaming high-bitrate HTML5 video (VP9/AV1/HEVC), Chromium delegates decoding execution to the GPU via the ANGLE (Graphics Engine Interface) translation layer. If the GPU pipeline takes longer than 2000 milliseconds (2 seconds) to process a batch of video frames due to driver scheduler latency or corrupt DirectX 11/12 command buffers, the Windows Timeout Detection and Recovery (TDR) subsystem assumes the GPU is unresponsive, resets the driver, and triggers a VIDEO_TDR_FAILURE (0x00000116).

Diagnostic Verification:

Open chrome://gpu or edge://gpu in your browser.

Search for Log Messages and look for GpuProcessHostUIShim: GPU process crash count or GL_OUT_OF_MEMORY / D3D11 device lost errors.

Open Windows Event Viewer (eventvwr.msc), navigate to Windows Logs > System, and check for Event ID 4101: *"Display driver nvlddmkm stopped responding and has successfully recovered."*

Step-by-Step Fix:

1Reconfigure Browser ANGLE Graphics Backend:

Open your browser and navigate to chrome://flags/#use-angle (or edge://flags/#use-angle).

Change the Choose ANGLE graphics backend setting from *Default* to *OpenGL* or *D3D9* to bypass corrupt DirectX 11 hardware rendering pipelines.

Click Relaunch.

2Disable Hardware-Accelerated Video Decode via Browser Flags:

Navigate to chrome://flags/#disable-accelerated-video-decode.

Toggle the setting to Disabled (which turns off GPU video decode while retaining GPU UI rasterization).

Relaunch the browser.

3Reset DirectX Shader Cache:

Press Win + R, type cleanmgr, and hit Enter.

Select your system drive (C:), check the box for DirectX Shader Cache, and click OK to purge corrupted pre-compiled video shaders.

4Clear Browser Hardware State Cache:

Close all browser instances.

Press Win + R, input %localappdata%\Google\Chrome\User Data\ShaderCache, and delete all subfolders (GPUCache, GrShaderCache).

Prevention & Long-Term Monitoring:

Maintain updated browser releases and monitor GPU process memory utilization via the built-in Browser Task Manager (Shift + Esc). Avoid running third-party video overlay hooks (such as Discord overlay or RivaTuner Statistics Server) concurrently with browser video streams.

Fix Local Media Player Hardware Codec Decoding Crash (VLC/MPC-HC)

Solution:

Root Cause: Direct3D11 / DXVA2 Hardware Video Decoder Kernel Lockup

Local media players leverage Direct3D11 Video Acceleration (DXVA2) or NVDEC/AMF hardware decoders to pipeline H.264, H.265 (HEVC), and AV1 decoding directly through dedicated ASIC blocks on the GPU. When playing media files with non-standard pixel formats (e.g., 10-bit YUV 4:2:2 or variable frame rate streams), decoder thread deadlocks can occur inside kernel-mode video drivers, exceeding the Windows TDR execution threshold and causing atikmpag.sys or nvlddmkm.sys blue screens.

Diagnostic Verification:

Reproduce the crash in VLC or MPC-HC while running official GPU monitoring utilities or GPU-Z to observe GPU Video Engine Load.

Review the Windows Reliability Monitor by typing perfmon /rel in the Run dialog. Check for technical details under Windows - Hardware Error listing LiveKernelEvent Code 141 or 116.

Step-by-Step Fix:

1Modify VLC Video Decoder Preferences:

Open VLC, click Tools > Preferences (Ctrl + P), and select the Input / Codecs tab.

Locate Hardware-accelerated decoding and change it from *Automatic* to *DirectX Video Acceleration (DXVA) 2.0* or *Disable*.

Click Save and restart VLC.

2Reconfigure MPC-HC / MPC-BE LAV Video Decoder Settings:

Open MPC, navigate to View > Options > Internal Filters, and click LAV Video Decoder.

Change Hardware Decoder to use from *D3D11* or *NVIDIA CUVID* to *DXVA2 (native)* or *Software (Plain C)*.

Uncheck 10-bit color depth options under hardware acceleration fallback settings.

3Update or Reinstall System Media Codec Packages:

Uninstall outdated third-party codec packs (e.g., legacy K-Lite Codec Pack variants) via Settings > Apps > Installed Apps.

Install the modern K-Lite Codec Pack (Standard/Basic) with default LAV Filter bindings to restore stable DirectShow filter pipelines.

Prevention & Long-Term Monitoring:

Use native media renderers like MPC Video Renderer or MadVR configured with software decoding fallbacks for high-bitrate media formats. Regularly clean up registry codec entries using tools like Codec Guide Tweak Tool.

Fix Hardware-Accelerated GPU Scheduling (HAGS) TDR Interruption

Solution:

Root Cause: HAGS Memory Management & Context Switching Collision

Hardware-Accelerated GPU Scheduling (HAGS) offloads memory management tasks from the CPU kernel scheduler directly to a dedicated RISC core on the GPU. However, when switching between low-power video playback pipelines (p-states P8/P5) and high-performance 3D workloads, HAGS can fail to handle preemptive GPU context switches smoothly. This race condition causes the GPU command queue to stall, triggering an immediate Windows TDR kernel crash.

Diagnostic Verification:

Open PowerShell as Administrator and execute:

Get-ItemProperty -Path 'HKLM:\SYSTEM\CurrentControlSet\Control\GraphicsDrivers' -Name 'HwSchMode'

A return value of 2 confirms HAGS is actively enabled.

Step-by-Step Fix:

1Disable HAGS via Windows Settings:

Press Win + I to open Settings, then navigate to System > Display > Graphics.

Click Change default graphics settings.

Toggle Hardware-accelerated GPU scheduling to Off.

Restart the computer.

2Override HAGS Settings in the Registry (Alternative Method):

Press Win + R, type regedit, and press Enter.

Navigate to HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\GraphicsDrivers.

Locate HwSchMode, double-click it, and change its Value Data from 2 to 1 (Disabled).

Click OK and reboot your machine.

3Disable Variable Refresh Rate (G-Sync / FreeSync) for Video Applications:

Open NVIDIA Control Panel (or AMD Radeon Software).

Navigate to Display > Set up G-SYNC.

Select Enable for full screen mode only instead of *Enable for windowed and full screen mode* to prevent display frequency switching during video playback.

Prevention & Long-Term Monitoring:

Keep Windows 11 feature updates current, as Microsoft continuously refines WDDM (Windows Display Driver Model) scheduling logic. Disable G-Sync/FreeSync overlays on desktop applications to prevent refresh rate desynchronization.

Fix Chromium GPU Process Memory Leak & ANR Pipeline Timeout

Solution:

Root Cause: VRAM Fragmentation and Chromium Shared Memory Deadlock

Extended video streaming sessions across multiple open tabs can cause severe VRAM fragmentation within the browser's GPU rendering process. When the allocated shared video memory space hits physical or virtual limits, the Chromium GPU process fails to submit frame draw calls within the time window required by the WDDM driver interface, prompting a system-wide driver reset via VIDEO_TDR_FAILURE.

Diagnostic Verification:

Open Windows Task Manager (Ctrl + Shift + Esc), click the Performance tab, and select GPU.

Observe Dedicated GPU Memory and Shared GPU Memory usage while playing video in multiple tabs to check if VRAM usage approaches 100% capacity.

Step-by-Step Fix:

1Enable Memory Saver Mode in Browsers:

Open Chrome/Edge Settings and navigate to Performance.

Toggle Memory Saver to On to immediately discard memory for inactive video tabs.

2Configure Chromium Command-Line GPU Execution Flags:

Right-click your browser desktop shortcut and select Properties.

In the Target field, append the following parameter to the end of the executable path (after a space):

--disable-gpu-memory-buffer-video-frames

Click Apply and OK.

3Terminate and Restart the GPU Process On-the-Fly:

If video stuttering begins, press Shift + Esc inside Chrome/Edge to open the internal Task Manager.

Locate GPU Process, select it, and click End process. The browser will safely recreate the GPU context without crashing Windows.

Prevention & Long-Term Monitoring:

Limit excessive active video streams in background tabs. Use extension tools like *The Great Suspender (Community Edition)* to automatically unload background media contexts.

Fix NVIDIA PCIe ASPM Link State Power Management TDR

Solution:

Root Cause: PCIe Active State Power Management (ASPM) Latency

Active State Power Management (ASPM) and Link State Power Management allow PCIe graphics cards to step down to lower power states (L1/L1.substates) when idle or during low-load tasks like video playback. When media decoding demands a sudden burst of compute power, waking the GPU across the PCIe bus can introduce latency that exceeds the 2-second TDR window, triggering nvlddmkm.sys crashes.

Diagnostic Verification:

Open Device Manager, expand System devices, and check properties of PCIe root ports for power management flags.

Review Event Viewer for kernel power warnings regarding PCIe link state transitions during video start events.

Step-by-Step Fix:

1Adjust Windows Power Plan Link State Power Management:

Open Control Panel > Power Options > Edit Plan Settings > Change advanced power settings.

Expand PCI Express > Link State Power Management and set both *Battery* and *Plugged in* to Off.

2Modify NVIDIA Control Panel Power Management Mode:

Open NVIDIA Control Panel > Manage 3D Settings > Global Settings.

Locate Power management mode and change it from *Normal* or *Optimal power* to Prefer maximum performance.

3Update Motherboard Chipset Drivers:

Download and install the latest Intel or AMD chipset drivers to ensure proper ACPI and PCIe power delivery management tables.

Prevention & Long-Term Monitoring:

Avoid aggressive third-party power saving utilities that force deep PCIe link states without proper BIOS ACPI support.

Fix AMD Zero RPM Fan & Radeon PowerTune TDR

Solution:

Root Cause: AMD Zero RPM Thermal Hysteresis & PowerTune Throttling

AMD Radeon graphics drivers utilize a Zero RPM fan feature that keeps cooling fans stationary until GPU junction temperatures cross 50°C–60°C. During video playback, GPU core temperatures hover right around this threshold, causing fans to constantly spin up and spin down. This thermal hysteresis combined with aggressive PowerTune clock adjustments can stall the driver command processor, resulting in atikmpag.sys TDR faults.

Diagnostic Verification:

Open AMD Software: Adrenalin Edition, navigate to Performance > Tuning, and monitor GPU temperature and fan speed fluctuations during video playback.

Step-by-Step Fix:

1Disable Zero RPM in AMD Software:

Open AMD Software: Adrenalin Edition > Performance > Tuning.

Toggle off Zero RPM to keep cooling fans spinning continuously at a low baseline speed (e.g., 25-30%).

2Adjust Minimum GPU Frequency State:

In Radeon Tuning, set a minimum GPU frequency state or disable advanced power-saving states for media workloads.

3Perform Clean Driver Installation with DDU:

Boot into Safe Mode, run Display Driver Uninstaller (DDU) for AMD, and reinstall the latest stable Adrenalin WHQL driver package.

Prevention & Long-Term Monitoring:

Maintain stable case temperatures and clean dust filters to prevent thermal oscillation around fan activation thresholds.

Fix Intel Integrated Graphics Power Gating & C-State TDR

Solution:

Root Cause: Intel Graphics Power Gating & Display C-State Latency

Intel integrated graphics (Iris Xe / UHD Graphics) heavily rely on aggressive power gating and Display C-States (DC states) to conserve battery life and reduce desktop power draw. When video decoding requests hardware context changes, waking the display engine from deep C-states can cause synchronization timeouts in igdkmd64.sys.

Diagnostic Verification:

Check Device Manager under Display adapters to confirm Intel graphics driver version and status.

Monitor system energy reports using powercfg -energy via an administrative command prompt.

Step-by-Step Fix:

1Update Intel Graphics Driver:

Download and install the latest official Intel Graphics Driver via Intel Driver & Support Assistant (IDSA).

2Disable Panel Self-Refresh (PSR) in Intel Graphics Command Center:

Open Intel Graphics Command Center > System > Power.

Toggle off Panel Self-Refresh and Display Power Savings.

3Adjust Windows Processor Power Management:

Open Power Options advanced settings and set Minimum processor state to 5% or 10% to prevent core parking lockups during media playback initiation.

Prevention & Long-Term Monitoring:

Keep Intel graphics drivers synchronized with certified OEM releases for laptop platforms.

Fix Hybrid Graphics (Optimus / MUX) Frame Buffer Handshake TDR

Solution:

Root Cause: Hybrid Graphics Muxless Frame Buffer Routing Deadlock

On laptops with hybrid graphics (e.g., Intel/AMD integrated GPU paired with an NVIDIA discrete GPU), video playback hardware acceleration is frequently routed through the integrated GPU frame buffer while rendering display outputs via the discrete GPU. If this muxless handoff fails or deadlocks during dynamic context switches, a TDR crash occurs.

Diagnostic Verification:

Check laptop manufacturer control software (e.g., ASUS Armoury Crate, Lenovo Vantage) to see if GPU working mode is set to Hybrid, Optimus, or dGPU-only.

Step-by-Step Fix:

1Force Discrete GPU Mode in BIOS or Control Software:

Reboot into UEFI BIOS or open your laptop management app, and switch the display mode from Hybrid/Optimus to Discrete GPU Only (or dGPU mode).

2Clean Reinstall Both Integrated and Discrete Drivers:

Use DDU in Safe Mode to remove both integrated (Intel/AMD) and discrete (NVIDIA/AMD) drivers, then reinstall vendor-certified packages.

3Configure Global Preferred Graphics Processor:

Open NVIDIA Control Panel > Manage 3D Settings, and set Preferred graphics processor to the high-performance GPU for all media apps.

Prevention & Long-Term Monitoring:

Keep vendor-provided system management software updated to ensure proper ACPI switching method execution.

Fix Multi-Monitor Mixed Refresh Rate WDDM Composition Stutter

Solution:

Root Cause: WDDM Multi-Monitor Refresh Rate Desynchronization

When running a multi-monitor setup with mixed refresh rates (e.g., a 144Hz primary gaming monitor paired with a 60Hz secondary display playing a YouTube video), the Windows Desktop Window Manager (DWM) struggles to synchronize composition vsync pulses. This compositor bottleneck causes micro-stuttering and command buffer delays that trigger TDR recovery crashes.

Diagnostic Verification:

Disconnect the secondary monitor and test video playback on the primary display alone to confirm if multi-monitor composition desync is the root cause.

Step-by-Step Fix:

1Match Display Refresh Rates:

Open Display Settings > Advanced display and set all active monitors to integer-divisible refresh rates (e.g., 60Hz, 120Hz, 240Hz).

2Disable Hardware-Accelerated Video on Secondary Displays:

Move video windows strictly to the primary display or disable hardware acceleration in media apps running on secondary screens.

3Update Graphics Drivers to Latest WDDM Build:

Install current GPU drivers featuring Microsoft's multi-monitor independent refresh rate scheduling fixes.

Prevention & Long-Term Monitoring:

Keep Windows display drivers updated and avoid mixing extreme refresh rate disparities across displays without independent VSync enabled.

Fix Cascading TDR Failure & Driver Recovery Hang (LiveKernelEvent 141)

Solution:

Root Cause: Cascading WDDM Driver Recovery Hang

When a minor GPU timeout occurs, Windows attempts to recover the display driver via TDR. However, if the underlying driver state is corrupted or third-party monitoring hooks intercept the recovery routine, the recovery attempt itself hangs. This triggers a cascading failure that escalates from LiveKernelEvent Code 141 into a full VIDEO_TDR_FAILURE (0x116) Blue Screen of Death.

Diagnostic Verification:

Check Reliability Monitor (perfmon /rel) for repeated LiveKernelEvent 141 warnings occurring right before system crashes.

Step-by-Step Fix:

1Perform Clean Driver Installation via DDU:

Boot into Safe Mode, run Display Driver Uninstaller (DDU), and perform a complete clean removal of all GPU display drivers.

Reinstall the latest stable WHQL driver package in Normal Mode.

2Remove Overlay and Hook Software:

Completely uninstall or disable third-party overlay utilities like RivaTuner Statistics Server (RTSS), MSI Afterburner hardware monitors, Overwolf, and Discord overlay.

3Increase TdrDelay Registry Timeout Threshold:

Open regedit, navigate to HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\GraphicsDrivers.

Create a new DWORD (32-bit) Value named TdrDelay and set its Value data to 10 (decimal), giving the GPU 10 seconds to recover instead of 2.

Prevention & Long-Term Monitoring:

Avoid running outdated hardware monitoring software that hooks directly into low-level WDDM rendering APIs.

Fix VRAM Overclock & Memory Bus Instability Artifacting TDR

Solution:

Root Cause: Unstable VRAM Overclock or GDDR Error Correction (ECC) Saturation

An overly aggressive VRAM memory overclock or degraded thermal pad contact on GDDR6/GDDR6X memory modules causes internal error correction codes (ECC) to saturate during high-throughput video decoding and texture caching. When memory read/write errors exceed fault tolerance limits, the GPU driver crashes and invokes a TDR stop code.

Diagnostic Verification:

Open MSI Afterburner or your GPU tuning utility and check if Memory Clock or Core Clock offsets are applied.

Step-by-Step Fix:

1Reset GPU Clocks to Stock Baseline:

Open MSI Afterburner or your vendor tuning software and click the Reset button to restore default GPU core and VRAM clock speeds.

2Improve GPU Thermal Performance:

Increase fan curves to keep GPU VRAM junction temperatures below 85°C.

3Test Stability:

Play high-resolution video streams for 30 minutes to verify that stock clocks eliminate artifacting and crashes.

Prevention & Long-Term Monitoring:

Avoid unstable VRAM memory overclocks; GDDR memory error correction hides minor instability until heavy multi-task workloads trigger driver crashes.

Fix Audio-Video Synchronization Deadlock & WASAPI Audio Driver Conflict

Solution:

Root Cause: WASAPI Audio Engine & Video Decoder Pipeline Desynchronization

During media playback, audio stream timing (managed via WASAPI / Windows Audio Session API) must lock precisely to video frame presentation timestamps. If audio endpoint drivers (such as Realtek Audio, NVIDIA HD Audio, or USB DAC drivers) experience sample rate buffer mismatches, audio-video synchronization deadlocks can starve the GPU rendering queue, resulting in a TDR driver hang.

Diagnostic Verification:

Check if video playback freezes specifically when switching audio output devices (e.g., from speakers to Bluetooth headphones) or changing sample rates in Sound Control Panel.

Step-by-Step Fix:

1Standardize Audio Sample Rate Settings:

Press Win + R, type mmsys.cpl, and press Enter.

Right-click your default playback device, select Properties > Advanced.

Set Default Format to standard 24-bit, 48000 Hz (Studio Quality) or 16-bit, 44100 Hz.

2Reinstall Audio Drivers:

Download and install the latest official audio drivers from your motherboard or sound card manufacturer.

3Disable Audio Enhancements:

Uncheck any spatial sound or audio enhancement packages that hook into the WASAPI rendering pipeline.

Prevention & Long-Term Monitoring:

Avoid mixing conflicting audio driver suites (e.g., Nahimic, Dolby Atmos, Realtek Audio Console) simultaneously without vendor-approved integration packages.

Fix Windows HDR Metadata & Tonemapping Pipeline TDR

Solution:

Root Cause: High Dynamic Range (HDR) Tone Mapping Metadata Timeout

Streaming 4K or 8K HDR video content (AV1/HEVC streams) requires real-time metadata calculation and tonemapping conversion (PQ/HLG to display color spaces). When Windows HDR or Auto-HDR is active alongside aggressive color management profiles, tone mapping shader pipelines can time out during complex scene transitions, triggering a VIDEO_TDR_FAILURE.

Diagnostic Verification:

Toggle Windows HDR off (Win + Alt + B) and test if 4K video playback runs stably without crashing.

Step-by-Step Fix:

1Recalibrate Windows HDR Settings:

Download the official Windows HDR Calibration app from the Microsoft Store and complete peak brightness calibration for your display.

2Update GPU Display Color Settings:

Open NVIDIA Control Panel / AMD Software and ensure Output dynamic range is set to *Full* and color depth matches display capabilities.

3Disable Browser Hardware-Accelerated HDR if Needed:

If crashes persist during browser HDR video playback, navigate to browser flags and disable hardware-accelerated HDR video decoding.

Prevention & Long-Term Monitoring:

Verify that your HDMI/DisplayPort cables support full uncompressed color bandwidth required for HDR 10-bit/12-bit transmission.

Fix GPU Undervolt P-State Voltage Drop & Minimum Core Clock TDR

Solution:

Root Cause: Aggressive Undervolt Voltage Starvation at Low P-States

Applying a custom GPU curve undervolt to reduce power consumption often sets voltage too low at idle and low-power video playback states (P8/P5 P-States). While the GPU remains stable during 100% full-load gaming, the reduced voltage is insufficient to maintain signal integrity during low-clock video decoding, causing a TDR crash.

Diagnostic Verification:

Monitor GPU core voltage during video playback using MSI Afterburner. If voltage drops below the stability floor established by your undervolt curve, crashes occur.

Step-by-Step Fix:

1Adjust Undervolt Curve for Low P-States:

Open MSI Afterburner voltage/frequency curve editor (Ctrl + F).

Raise the voltage floor at lower frequency points (e.g., ensure voltage does not drop below 875mV to 900mV at baseline video decoding clock speeds).

2Remove Undervolt Temporarily to Test:

Disable custom curves entirely and test video playback stability at stock voltage settings.

3Set Minimum Power State:

In NVIDIA Control Panel / AMD Software, set power management mode to *Normal* rather than aggressive eco profiles.

Prevention & Long-Term Monitoring:

Test undervolt stability across both idle media playback and heavy gaming workloads rather than gaming alone.

Fix PCIe Riser Cable Signal Attenuation & Gen 4/5 Bus Error

Solution:

Root Cause: PCIe Riser Cable Signal Attenuation and CRC Errors

Using a PCIe riser cable in vertical GPU mounts—especially unshielded Gen 3 cables running on Gen 4/Gen 5 motherboards—introduces signal attenuation and electromagnetic interference (EMI). During high-bandwidth data transfers like video stream decoding and frame buffer swapping, PCIe bus CRC error correction rates spike, leading to bus lockups and TDR crashes.

Diagnostic Verification:

Physically seat the GPU directly into the motherboard primary PCIe x16 slot, removing the riser cable, and test video playback stability.

Step-by-Step Fix:

1Force Lower PCIe Generation in BIOS:

Enter UEFI BIOS -> PCIe Subsystem Settings.

Change PCIEX16_1 Link Speed from *Auto* or *Gen 4 / Gen 5* down to Gen 3 to match riser cable signal limits.

2Replace with Certified PCIe Gen 4/5 Riser Cable:

Upgrade to a high-quality, fully shielded premium riser cable (e.g., LINKUP or Lian Li PCIe 4.0 riser).

3Ensure Secure Direct Connection:

If possible, eliminate the riser cable entirely to ensure clean direct slot communication.

Prevention & Long-Term Monitoring:

Always lock PCIe link speeds manually in BIOS when utilizing vertical GPU riser mounts.

Fix PSU Transient Voltage Droop & Power Delivery Rail Saturation

Solution:

Root Cause: Power Supply Unit (PSU) Transient Load Droop on 12V Rails

High-performance GPUs experience rapid current spikes (transient loads) when waking up video decode engines or switching power states. If the power supply unit (PSU) exhibits slow transient response or voltage ripple on the +12V rail, transient voltage drops can cause the GPU power telemetry to trigger internal protection shutdowns or driver TDR resets.

Diagnostic Verification:

Monitor +12V rail voltages via HWInfo64 during video playback start events to check for voltage sags below 11.4V.

Step-by-Step Fix:

1Use Dedicated PCIe Power Cables:

Ensure your GPU is powered by separate individual PCIe / 12VHPWR cables running directly from the PSU rather than daisy-chained "piggyback" splitter cables.

2Upgrade to ATX 3.0 / PCIe 5.0 Compliant PSU:

If using older power supplies, upgrade to an ATX 3.0 certified PSU engineered to handle 200% power excursions without rail voltage sags.

3Check Power Connections:

Verify that all power connectors are firmly clicked into place on both the GPU and PSU ends.

Prevention & Long-Term Monitoring:

Match your power supply wattage rating and transient excursion headroom to high-draw modern GPU architectures.

Related Diagnostic Guides

MEMORY_MANAGEMENT BSOD While Gaming: XMP Profile Diagnostic

Laptop Screen Goes Black But Power Light On: Lid Sensor & eDP Fix

Mouse Lag On 4K Monitor Only: Polling Rate & DPI Diagnostic

External SSD Not Showing Up But Making Sound: Disk Management Tree

PFN_LIST_CORRUPT BSOD After RAM Upgrade: Page Frame Diagnostic

Subscribe to StepDiag

Embed Guide on Your Website