GTX 1070 getting hotter and hotter

Status
Not open for further replies.

emailadress2018

Prominent
Jul 7, 2018
27
1
545
Can it be that the slow degrading of the circuits (chip, vrms etc) make a gpu run increasingly hotter with time?
I bought my gtx 1070 (KFA2) mid 2017. The first few months the card never even reached 60° Celsius under load in the summer time (I mostly play Battlefield Games: 3,4 und 1). I know that since I changed my amd fx8370e to i7 4790K (running at 5.0ghz@1.275v on a ROG Maximus VII) the cpu is no longer bottlenecking the gpu(especially in bf1) but still, now in the almost-winter temperatures are 65+ when ambient is 13° when I dont turn on the heater. When my wife sometimes does turn on the heater, and the ambient goes to 24° then the Gpu gets to 70° in Bf1 .Even if its not too much for the card, I would like to know if this increase in temperature difference (compared to ambient) compared to the first months after I bought it is normal.
I even replaced the thermal paste in the gpu with Arctic Mx4,but it didn't help at all.The heatsink and blackplate DO get hot and the fan that I have in my case's side panel does spit out warm air to my feet, so it means that the heat IS being transferred from the chip. I was thinking about using Liquid metal on the gpu as i did with my delided i7 (under the IHS and between IHS and cooler) but I dont know if I shouldrisk damagind the gpu heatssink since its only copper, not nickel.
For comparison, the gtx1050ti I just bought for my wife a few weeks ago is behaving exactly as my 1070 when it was new a year ago- doesnt even reach 60° ..and my wife only has 2 case fans, I have 7 , of which one is 2cm away from the gpu

Don't bother to say that the OCed i7 is making air hot in my case, because the cpu has a gigantic Scythe Mugen cooler that barely feels warm to touch when I stress test cpu for 10 minutes (msi afterburner confirms: 29° idle, max 50° full load), and there are 3 powerful fans with no filter that push and pull air all in the proximity of the cpu cooler and vrms.
I made no other to pc since last year, except adding another 8gb ram for a total of 24gb with 4 sticks , because I wanted to use all sticks in dual channel, but this cant affect gpu.
 
They are normal temperatures.
GPus, Specially 1000 Series, like 1070,1080,etc runs up to 83 degrees Effectively. When they reach more than 83 they Downclock themselves to avoid any possible damage on it. Even throught they support up to 95 Degrees.
If you want them to keep them as cool as possible my recomendations are:
-Maintain your PC clean and free from dust.
-Replace thermal paste in case is needed.
-Try to keep other parts that are close to the GPU with the coolest temps possible.

Still... There is nothing to worry here for now...
 
Another key point is you are now utilizing more of the card with an upgraded CPU, it should be receiving, processing and pushing more data requiring more voltage and an increase in temps. It's doing exactly what it should be.
 
One night, about a month after building my last monster rig, i forgot to put msi afterburner on "auto" and left Farcry 4 running overnight. My card ran at +90c for like 9 hours in a row....after that my card was always like 10 degrees hotter. So i think there is some wear and tear after a while. My situation was just a sped up example lol.
 


this is exactly what i suspect and what I am afraid of.. BUT as this video shows: https://www.youtube.com/watch?v=44JqNJq-PC0&t=289s (at 4:53) this shouldn't be happening.
Maybe the problem with our Cards is not that the CHIPS themselves are getting worse, or even that the fans are failing (mine still spin perfectly fine and yours didnt even ramp up because Afterburner was off) but maybe the VRMs\Power delivery are degrading or something similar, because we got lesser quality units.
We really need the opinion of an EXPERT on this topic (at least GamerNexus experience level)
 
Status
Not open for further replies.