PC Graphics

[Jose_Zé]

Hi everyone!

I bought a card with a 6750 XT GPU from the power color brand model RED DEVIL because it comes with a really generous heat dissipation system.
My intention was to have a device capable of keeping the chip cool and increasing longevity.
The hot spot remained at approximately 90º C maximum and everything was fine.

It happened that from 2023 onwards, in a strange way, the chip on my board and that of dozens of other users began to reach hot spot temperatures above 100º C
And now it reaches 105º C with peaks between 110~112º C.

This coincided with some new Adrenalin driver updates.

I changed thermal paste and thermal pads, all of good quality and correct measurements.
No results.

What is happening with AMD???

Someone?

[Earnhardt]

ASUS STRIX RX6750XT here, Benching Time spy @1440p latest Drivers

[johnnyenglish]

The RoG STRIX is a bad example, way overbuilt, they just dont overheat unless you clock it like a maniac.

My old RoG STRIX RX480 could go 1500 OC'ed and the RX6800 wont even touch 70C

[elstaci]

First your GPU card is engineered to run up to a Maximum Operating Temperature of 110c. So any temperatures at 110c or below is within the safe temperature range. But once it reaches and surpasses 110c it will automatically start to throttle or slow down to cool down the GPU card.

One way to eliminate the AMD driver as the culprit for your GPU card starting to overheat is by removing the current AMD Driver with DDU and installing a previous AMD driver and see if the issue continues or not. If it doesn't I suggest you open a AMD SUPPORT TICKET and let them know from here: https://www.amd.com/en/forms/contact-us/support.html

Here is AMD PREVIOUS DRIVERS from AMD download page: https://www.amd.com/en/support/previous-drivers/graphics/amd-radeon-rx-6000-series/amd-radeon-rx-670...

NOTE: A GPU's temperature changes can be due to many factors such as PC internal Air Flow, Hardware Fan issues, PSU issues, Hot Ambient room temperatures, etc.

[Fimbulvarg]

Hey there,

I have similar hotspot temperatures on my AsRock RX 6750 XT Challenger Pro OC while playing Witcher 3 (only modded slightly, I swear). Bought the card in May last year, so it's probalby no ageing problem, but I only recently upgraded to a QHD (2560x1440) monitor, that's when the temps went up...

My hotspot goes up to about 102...105°C (highest I've seen was 109°C but only once), while the average GPU temp is reported as 68...72°C. That means, I have a temp difference of ~30°C between the edges and the center of the GPU chip. These values are reproducible with FurMark (v 2.60) on 1440p with Anti-Aliasing set to x8, though my reseller couldn't or wouldn't when I tried to RMA the card for the issue.

Are these temps still in the "safe to operate" zone? Are there specs available anywhere? I've read on many other forums, that when repasting your GPU you'd aim for 15...20°C between hotspot and the rest, and that anything above 95°C will damage the card in the long run (as in "it won't live much longer than the warranty time").

The sensor readings imply that 110°C is the maximum operating temp, but you don't push the hardware to nigh 100% thermal capacity if it's supposed to work for more than a year or two...

Thanks for reading this, even though I'm late to the party 🙂