Is My RX 6800 XT Still Worth Repasting After Running at 110°C for Two Years?
Asked by
MellowPine47
I built my first PC about two years ago and bought a used RX 6800 XT because I was on a budget. In demanding games, its temperature often reaches 105–112°C and sometimes peaks around 114°C. It has only crashed a few times, and it performs normally when temperatures stay lower. I recently bought a newer 9070, but my Wi-Fi stops working whenever I launch a game with it installed. Ethernet isn't practical where I live, so I temporarily switched back to the 6800 XT. Would it be worth opening the old card to clean and repaste it, or could two years of high temperatures have caused permanent damage?
Answers
Before returning the 9070, troubleshoot the Wi-Fi issue separately. Use a clean graphics-driver installation, update the motherboard chipset and Wi-Fi drivers, and check for PCIe or power-management conflicts. A GPU-related Wi-Fi problem isn’t normal, so it may be a driver, slot, or system compatibility issue rather than a reason to give up on the new card.
Reply to this answer
Repasting is worth trying as long as the card isn’t showing artifacts, frequent crashes, or performance problems. Also check that the heatsink is dust-free and that the case has good intake and exhaust fans. The temperatures may not have permanently damaged the card, but repeatedly hitting the thermal limit can reduce boost clocks and cause throttling.
Reply to this answer
That 110–114°C reading is probably the GPU hotspot or junction temperature, not the average core temperature. The 6800 XT is designed to throttle around that junction limit, which is why it may have survived without immediately failing. Since it is a used card, dried or pumped-out thermal paste is a realistic possibility. Cleaning it, improving case airflow, and replacing the thermal paste—and thermal pads if they’re damaged—could bring the temperatures down. If you aren’t comfortable opening it, an electronics repair shop can handle it.
Reply to this answer
There’s no practical way to reverse actual heat damage, but high hotspot readings alone don’t prove that the card is damaged. If it still runs without artifacts or repeated crashes, start with cleaning, airflow improvements, and a repaste. Monitor both the core and hotspot temperatures afterward; a large gap between them can indicate poor contact between the cooler and the GPU die.
Reply to this answer
Add your answer
Answers are reviewed before publication.
Related Discussions