Hello everybody,
For some time now I have noticed a slight increase of temperature on my RTX2070 XC ultra (buy early 2019, always in guarantee).
 
I'm approaching a pretty hot period in my country and it will be complicated to play if the trend continues (my game room will go from 24 to 30 °C)

• Before (2019-2020) I had : arround 38°C IDLE, arround 68°C - 75°C in game.
• Actually (since three mounth and before the "hot" periode) : arround 42° IDLE, arround 70°C - 80° C in game.

I noticed this on this actual following games (which have evolved and become a little more "greedy"):

• No man's sky (last Updated)
• Crusader kings 3 (probably not yet perfectly optimized)
• Elite dangerous Odyssey (really not yet optimized !)

the annoying point of reaching these "80°C" is the ventilation is activated fully (100%) and it is extremely noisy for a short period (one to two minutes). So I wonder where this increase in temperature can come from ? I realized all the usual solutions:

•  Cleaning of dust in the machine
• Case setup ventilation control (nothing to report, a fan was changed 6 months ago, RPM of all are normal and voltage too)
• Updated driver / software (+ Windows etc ...)

After that I made a lot of observation and tests, problem seems to be related to the solicitation of the card itself. For example I pushed the ventilation of the box to 100%, result I obtained a cold configuration (but very noisy of course) in game the motherboard/CPU/SSD lost arround 5 or 10°C but not for Graphics cards, still ends going to 80°C (??) but after a more longer game time.
 
When I stop the game, the graphics card temperature come down quickly (a few minutes to find an IDLE temperature).
 
So... What conclusion to draw from this? Does the graphics card seem to dissipate its temperature badly (Heatsink or thermal paste problem?) On the ventilation side, they seem to be running very well. Or is there a bad setting somewhere? To be honnest I have always enormous difficulties with the X1 precision settings, some would have precision X1 profiles for me to test for these games ?
 
Thank you very much for your advice...

Replace the thermal paste. It is normal for the thermal paste to pump-out and create voids after enough thermal cycles, and require being replaced.

You won't void your warranty by reapplying the thermal paste as long as you don't physically damage the card.

Alphacentor

bdary
You won't void your warranty by reapplying the thermal paste as long as you don't physically damage the card.

 
I believe this is true in the USA but not in France 

Ok.  My advice would be to contact EVGA directly in your area to verify that.  Redoing the thermal paste could solve your temp issue.
 
https://www.evga.com/about/contactus/
 

This is the first time that I have received such a quick and clear response from customer service...
 

evgaThe temps seem to be fine on the GPU. However, you can of course try and change the thermal pads.
This will not void your warranty !
As a side note, there are a few things that can void our warranties, few examples are, removed serial or manufacturers stickers, liquid damage or physical damage.
The sticker on one of the screws in the back is not a warranty sticker so its safe to remove to apply the thermal paste.

 
if my temperatures are good despite the increase then I come back to my astonishment that at 80°C the graphics card goes to 100% ventilation (with a sound similar to an airplane turbine). I will have to do some tests with horrible X1 precision curves
 

thanks again for your reply :)! For curves... Yes but I must not be good at this point, I have already tried to "play" with these curves but I never really know what to think of the results... The goal is obviously to have a good temperature for a noise "reduced" I supposed...
 
These curves have helped me in the case of "old" games which run at a very low level and on the temperature at which the fans started up (passive / active / passive... -1 ° C, + 1 ° C etc etc, Sound bzzz bzzz bzzz)
 
This time it is a "high" temperature, what do you recommend as a strategy? Push 100% usage FANs from 80° to 85 ° first? And put a level of how much at 80 °? 80% fans for example?
 
Or ... "Attack" harder lower, eg increase ventilation by 20 or 30% at lower temperatures (50=>70°) ?

It will take some trial & error to get it just how you want it.  Noise versus cooling to your liking.

Sounds good...  You can set PX1 to load at Windows startup.  If I recall correctly there is a box to tick for that.  I use MSI AB so it's been a long time since I looked at PX1.  I don't remember exactly the steps to get to that setting.
 
Another good way to control the GPU temp is by limiting the frames.  Check post #3 out here ->  https://forums.evga.com/SuggestionsAdvice-m3311760.aspx#
 

You're welcome and good to hear you got things running better/cooler now...