• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

7900XTX Hotspot Temperature Issue tested (der8auer)

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
I just want to comment, cutting the cooler open doesn't give any indication as to whether there is insufficient fluid in the vapor chamber. The amount used is so small and since it's in a vacuum most of it would evaporate as soon as air enters the chamber. I don't know about the lack of sintering on most of the posts and the presence over the GPU, maybe the idea is to collect liquid from the whole chamber to then bring it back across to the GPU in that location. Or is the sintering only necessary because the depth of the chamber is larger there. These are things I have no clue about.

AMD has "struggled" with the hotspot since at least Vega, and a lot of that comes down to temperature reporting. At least 10 series from nVidia simply did not report a temperature to the user that was so "in the thick of it" so to speak, and only gave a more generalized reading. What I'm trying to say is GPU to Hotspot is not an apples to apples comparison on an AMD card, my understanding is that the GPU is more of a package temp while HotSpot is a die temp. I think people equate this with the core to core delta seen on a lot of 14nm Intel CPUs, and immediately think the cooler is not even. Granted a GPU is a huge bare die and a proper mount can be elusive and absolutely effective in lowering temperatures. I think the real issue is that PR over promised (again) and at the last minute the engineering team were expected to get more out of the cards than the coolers were designed for. The cards are obviously throttling. And of course a better cooler will bring down your delta from hotspot to die, and mainly will hopefully keep you from throttling.

My question which I've wasted a lot of words getting to, is it a design flaw in the cooler or is it an engineering error in not matching the cooler to the application due to pressure from bad marketing at the last minute?

The orientation thing can happen with heatpipes as well as vapor chambers, did they not put the thing in a case to test it horizontally? If so what a huge oversite!
 
Back