My ASUS RTX 4080 Super reaches a 105–109°C hotspot almost immediately after launching a game, which puts it right around the throttling range. I've repasted it and also tried a phase-change material, but the temperatures didn't improve. At idle, the GPU core is around 44°C and the hotspot is about 54°C. The original thermal pads are still installed, and the card is out of warranty. I'm not sure whether this is a mounting problem, incorrect pad thickness, or an airflow issue. What should I check before giving up on the card?
3 Answers
The thermal pads could be the problem even if they’re original. A graphics card can use several different pad thicknesses, and a pad that’s slightly too thick can prevent proper die contact. Before replacing everything, check the cooler contact pattern and verify the correct thickness for each location. Thermal putty can be an alternative if the exact pad sizes are difficult to source.
A hotspot that’s 30°C or more above the core usually points to poor contact between the GPU die and the cooler, especially when the jump happens almost instantly after a game starts. Check the imprint left in the paste or phase-change material. If coverage is uneven or missing along one edge, the cold plate may not be sitting flat. Thermal pads that are too thick can also push the cooler away from the die.
Try running the card briefly with the case side panel removed and watch both the core and hotspot temperatures. If the hotspot still shoots to 105–109°C, airflow probably isn’t the main issue and the cooler mounting or pad pressure is more likely. Also check that the fans ramp up normally and that the card isn’t overclocked.
I switched to PTM7950 to test it, but the original thermal pads are still in place. The idle core temperature is about 44°C, with the hotspot around 54°C.