[SOLVED] Rtx 2070 super high temps but strange

Jan 6, 2022
4
1
15
Hi everyone. Couldn't find any answer to my question, so decided to ask here.
So I have gigabyte rtx2070 super gaming oc for 2 years.
First I made pretty stupid decision - I put it into the mini-itx case with a very bad airflow and complete lack of space. I was getting temps around 78C under load (I did my best with airflow in that case)
So I decided that was ok, cause I wasn't playing a lot under full load and stopped monitoring temps at all.
But probably 5-6 months ago I heard my fans revving under load periodically. I checked the temps and they were 85C under load. And the case was extremely hot under load, and exhaust air was pretty hot either.
Today I finally got a new case, I choose Lian Li lancool 2 mesh for better airflow.
I set all up, even added additional two 120mm fans under the gpu. I thought that will resolve the issue.
But my temps still goes up to 80-81C under load. And hotspot goes to 105C.
But strange thing - the case is absolutely cool, all exhaust air is very cool either.
I checked with MSI afterburner, when I just load a game (witcher 3 for example) - temps run from 32C idle to 78C in 3-4 secs, just after clock boost. And when I shut it down - It gets back to 38C in 3-4 secs.
Can anyone explain me, are these temperature changes truthful enough? I thought probably that I need to change thermal paste, but I don't want to disassamble my gpu without sure, that the problem is in paste, cause it's pretty risky for me.
Can there be problems with temps sensor or anything else? Can i check it somehow?
Thanks in advance.
P.s. no overclock, never changed firmware. adaptive power settings in control panel.
 
Solution
the almost instant ramping up & down of temps is normal.
but reaching >70°C is not normal in a well ventilated case with a good GPU fan curve.

it's very possible that running over higher temperatures for extended amount of time will "wear out" the thermal paste leaving it dry and hard and not as effective.
but i wouldn't recommend disassembling and reapplying if the card is still under warranty, usually 3 years.

if you really don't mind just searching for a new card if it malfunctions though, go ahead and do it.
i would first just setup a more aggressive fan curve for both the GPU and those fans pumping cool air directly into it.
if that doesn't offer the temperatures you'd like to see then maybe go through it.

but you really aren't...
the almost instant ramping up & down of temps is normal.
but reaching >70°C is not normal in a well ventilated case with a good GPU fan curve.

it's very possible that running over higher temperatures for extended amount of time will "wear out" the thermal paste leaving it dry and hard and not as effective.
but i wouldn't recommend disassembling and reapplying if the card is still under warranty, usually 3 years.

if you really don't mind just searching for a new card if it malfunctions though, go ahead and do it.
i would first just setup a more aggressive fan curve for both the GPU and those fans pumping cool air directly into it.
if that doesn't offer the temperatures you'd like to see then maybe go through it.

but you really aren't reaching anything that's physically dangerous to the card so if you can stand some louder fans and some higher ambient temps around the system, just leave it as is.
 
  • Like
Reactions: TetraSlider
Solution
Jan 6, 2022
4
1
15
the almost instant ramping up & down of temps is normal.
but reaching >70°C is not normal in a well ventilated case with a good GPU fan curve.

it's very possible that running over higher temperatures for extended amount of time will "wear out" the thermal paste leaving it dry and hard and not as effective.
but i wouldn't recommend disassembling and reapplying if the card is still under warranty, usually 3 years.

if you really don't mind just searching for a new card if it malfunctions though, go ahead and do it.
i would first just setup a more aggressive fan curve for both the GPU and those fans pumping cool air directly into it.
if that doesn't offer the temperatures you'd like to see then maybe go through it.

but you really aren't reaching anything that's physically dangerous to the card so if you can stand some louder fans and some higher ambient temps around the system, just leave it as is.
Thanks for your answer!
New card is a big problem these days, especially in my country. Even this 2070 super now cost 3-4x more, than I bought it 2 years ago lol.
And ther is no malfunctions.
I set up pretty agressive curve in MSI Afterburner already. In terms of the case - I did my best, I still had to install my AIO rad in front for intake (cause I have 280mm rad and it doesnt fit at top exhaust in this case unfortunately), but it feels that it all works great still, I feel big improvement with new case (very cool side panel and exhaust air in this case in comparison with previous, but somehow gpu temps are still almost same)
And 2 years ago max gpu temps in that mini'itx case with awful airflow were lower than now.
So I can feel that there is some problem with gpu, and I would like to fix it somehow..
 
Jan 6, 2022
4
1
15
i don't mean if it is malfunctioning now.

but disassembling and reapplying thermal paste will void the manufacturer's warranty.

so if it does happen to malfunction before that warranty period is over you will not be able to RMA for replacement or repair.
Ah sorry, got you wrong, I'm not a native speaker.
I think I still will go for paste change. The warranty is almost out so it should be fine I guess.
And I don't see any other logic reasons for temps that high in such a good case like that.
I just hope that thermal pads will be fine after disassembley, cause I really don't want to mess with them.
 
Jan 6, 2022
4
1
15
the almost instant ramping up & down of temps is normal.
but reaching >70°C is not normal in a well ventilated case with a good GPU fan curve.

it's very possible that running over higher temperatures for extended amount of time will "wear out" the thermal paste leaving it dry and hard and not as effective.
but i wouldn't recommend disassembling and reapplying if the card is still under warranty, usually 3 years.

if you really don't mind just searching for a new card if it malfunctions though, go ahead and do it.
i would first just setup a more aggressive fan curve for both the GPU and those fans pumping cool air directly into it.
if that doesn't offer the temperatures you'd like to see then maybe go through it.

but you really aren't reaching anything that's physically dangerous to the card so if you can stand some louder fans and some higher ambient temps around the system, just leave it as is.
Just if you're interested. I've changed my thermal paste today (I used Kryonaut). The old one was dry AF.
My temperatures in load dropped from 81C to 54C
Hot spot dropped from 105 to 67.
I'm happy with that now, thanks for confirming my thoughts about thermal paste!
 
  • Like
Reactions: JohnBonhamsGhost