GPU artifacting describes unexpected visual corruption, such as colored blocks, flashing shapes, broken textures, or stray lines. These symptoms can involve the graphics card, but they do not by themselves prove it is failing. A game bug, unstable settings, a driver problem, a display cable, or the monitor can produce similar results.
Record what the visual problem looks like
Note whether the corruption appears in one game, several applications, the desktop, or before Windows starts. Record the resolution, refresh rate, driver version, and any recent settings changes. A photograph or short recording can help support staff understand an intermittent symptom.
Try a screenshot as well. If the problem appears in a captured image viewed on another working display, that is a useful clue that it occurred before the final display output. If it is missing from the screenshot, the monitor or signal path remains a possibility. Neither result is a complete hardware diagnosis.
Start with reversible checks
- Check the application. If only one game is affected, test another game and review the affected title's patches, known issues, and file-verification option.
- Return custom tuning to a stable baseline. Undo GPU overclocks or undervolts first. If other custom CPU or memory settings are involved, follow the motherboard instructions for restoring the relevant defaults.
- Check the driver. Use the GPU or PC manufacturer's appropriate driver. If the symptom started immediately after an update, check its release notes and the supported rollback or reinstall procedure.
- Check the signal path. With the relevant equipment powered off, reseat the display cable. Test a known-working compatible cable or another display/input when available.
ASUS's abnormal-display troubleshooting guide covers drivers, unstable overclocking, temperature, software, and power as separate possibilities. Change one variable at a time so that a successful result tells you something useful.
Check cooling, power, and physical connections
Temperature and airflow
Observe the temperatures and fan behavior reported by appropriate monitoring software during the workload that causes trouble. Compare them with guidance for the exact graphics card. There is no single temperature number that diagnoses every model, and a normal reported core temperature does not certify every component on the card.
Check for obstructed ventilation and excessive dust without dismantling the cooler as the first step. Follow the card and case makers' cleaning guidance. Our guide to case fan layout explains why airflow paths matter more than a universal fan count.
Power and seating
Shut down and disconnect mains power before inspecting internal connections. Check that the card and its required power plugs are fully seated, using the GPU and PSU instructions. Use cables approved for the exact power supply; modular PSU cables are not interchangeable merely because a connector fits. A symptom under load is a reason to investigate power delivery, not proof that a higher-wattage PSU will fix it.
Custom settings
A configuration that appeared stable in one game can fail in another. Restore the affected settings and repeat the same workload before assuming permanent damage. Do not apply a universal voltage recommendation or increase power limits to conceal the symptom.
When hardware failure becomes more plausible
A repeatable fault that follows the card across known-working, compatible configurations is stronger evidence than a single glitch. Corruption before the operating system loads also changes which explanations are plausible, but the display and cable still need to be considered.
If problems persist at supported settings after basic checks, document the results and contact the manufacturer or a qualified repair service. Do not keep running demanding stress tests on hardware showing signs of physical damage, an unusual smell, or repeated shutdowns.
Frequently asked questions
Can background applications cause artifacts?
Software conflicts, overlays, and application rendering bugs can cause display problems. Close optional overlays and compare the result. Their presence does not establish that the GPU is defective.
Should I replace thermal paste immediately?
No. Check settings, drivers, cables, and airflow first. Cooler disassembly introduces additional work and requires model-specific instructions; it is not a universal cure for visual corruption.
Does artifacting qualify for a free replacement?
Warranty coverage depends on the product, purchase, region, and diagnosis. Keep your evidence and ask the manufacturer or seller about service. An expired warranty does not automatically mean replacement is the only repair option.
