It pops up for me from time to time too (1 time in 100), but I haven't played with the voltages (I prefer not to take risks). I just cranked up the fan speed so that the GPU doesn't lower it when the temperature reaches 70 °C. The top RTX's are walking wavetables, they get very hot and to keep them from damaging themselves they undervoltage.
Currently on my RTX2080 the temperatures don't exceed 69°C and the voltage and power consumption doesn't drop, so I don't have any artefacts etc. and most importantly, FPS don't drop by GPU undervolting.
The default GPU cooling configuration is generally too weak, unlike new CPUs , GPU cores are more delicate and should be cooled at all costs, unless you want to have artifacts and other problems after the warranty expires because the section/memory will start to break down due to damage from too high temperatures especially as you block the graphics undervolting. Just because the GPU core doesn't exceed 71°C doesn't mean the memory doesn't have much higher temperatures 🙂
GPU manufacturers have a sick rule of thumb :
First undervolting the voltage and power consumption, then the clocks, and if that doesn't help then only the GPU overclocks the fan speed 🙂 What they don't do to make a card fit for scrap after warranty.
Since the Radeon HD 3 series died as a result of overheating, I have used a manual cooling chart on every GPU since. Since I have been using this, I have not had any problems with the GPU.
On my RTX2080 Super the only problem is the noisy fans, but as far as graphics go I have no problems.
EA AC causes a ton of bugs because its whitelist is mega limited and until EA's studio expands it, problems like this will occur.
My Fan config in screen below:
GL
