Code 43 - GPU Stopped Windows

Description

Windows stopped the GPU in Device Manager. The driver reported an unrecoverable error to the kernel. The most common GPU error after driver issues.

Common Causes

• Damaged or incompatible GPU driver causing Windows to stop the graphics card with code 43.
• Insufficient GPU power supply — the GPU needs more watts than the PSU can stably deliver.
• GPU overheating triggering emergency shutdown and Windows blocking.
• Incompatible driver version — a too-new or too-old version causes conflicts.

Recommended Solutions

• Solution: Go to Device Manager, right-click the GPU, and select 'Update driver' — try automatic search.
• Solution: Uninstall the driver using DDU in Safe Mode and install a clean version from the manufacturer's website.
• Solution: Check GPU temperatures in HWMonitor — if they exceed 90 °C, clean the GPU cooling.
• Solution: Try an older stable version of the GPU driver (e.g., Studio Driver for NVIDIA).
• Solution: Check that the GPU is firmly seated in the PCIe slot, and try a different PCIe x16 slot on the motherboard.
• Solution: If the problem persists, it may be a GPU hardware defect — file a warranty claim.

Diagnostic Commands

Understanding Severity: GPU Rendering Failures

Graphics processor (GPU) faults directly impact display output rendering, system graphics acceleration, and compute APIs like DirectX. When a GPU encounters a calculation exception, memory mismatch, or thermal limit, the display driver stops responding to kernel requests. To prevent a full system freeze, Windows activates Timeout Detection and Recovery (TDR), restarting the display driver and closing active graphics applications. If the GPU fails to recover, a fatal kernel crash (BSOD) occurs. Ensuring your GPU operates stably is crucial for gaming, system navigation, and hardware accelerated workloads.

Safety & Prevention Guidelines

Before removing or installing a graphics card in the PCIe slot, ensure the computer is powered down and the PSU switch is set to the OFF position. Disconnect all PCIe external power cables. Touch the metal case chassis to discharge static electricity before touching the card. When installing, verify the GPU is seated completely into the motherboard's primary PCIe x16 slot and the locking tab has clicked in. Always use independent power cables from the PSU for high-draw GPUs rather than pigtail splitters to avoid connector melt.

Windows Version & Compatibility Notes

Modern OS architectures like Windows 10 and 11 heavily rely on graphics driver models (WDDM 3.0+). Advanced rendering features like Hardware Accelerated GPU Scheduling (HAGS), Auto HDR, and DirectX 12 Ultimate require strict coordination between Windows updates and the graphics card driver, making legacy hardware prone to rendering timeouts.

Diagnostic Tools & Log Analysis

For GPU diagnostics, use GPU-Z to check core features, PCIe lane speeds, and thermal sensors. Use MSI Afterburner to monitor real-time clocks, voltages, and memory usage. Run FurMark or Superposition to stress test GPU stability under heavy rendering loads. Use WhoCrashed to read minidump files after GPU BSODs.

When to Seek Professional Hardware Help

If graphical artifacts, black screens, or TDR driver timeouts continue to occur at stock clock speeds after a clean driver reinstall with DDU and on a quality PSU, the graphics card might suffer from VRAM chip degradation or physical silicon damage. Contact an authorized repair center for diagnostic testing.

Frequently Asked Questions

Detailed Troubleshooting Guide Available

We have written a comprehensive, step-by-step diagnostic guide covering these types of issues in depth.

Need more help?

If these steps didn't resolve your issue, try searching our database for related symptoms or hardware components.