GPU temps are suddenly trash

[Cupcakez]

I had this RTX 2070 for almost a year and until now I haven't had any kind of trouble with it.

 

Today I had to take my pc to work and I took an uber both ways. On my way home the driver smashed the break pedal a bit too hard and my pc fell from the backseat of the car inbetween the backseat and frontseat.

The pc itself is still running, no visible damage or anything, but my gpu temps got worse all of a sudden.

 

During my light gaming sessions my gpu used to sit at a cosy 62 deg celsius with the hotspot at 75-85 deg without the fans having to ramp up.

Now I had to manually set the fan speed to 100% and the gpu temps are 66deg with hotspot at 100-105 deg with the same workload.

 

I checked the fans, they still look and run properly, no weird noise, no nothing. 

I read somewhere that old dried up thermal paste might crack and get disloged, causing temperature problems.

Is that true? Is there another obvious problem I don't see? Should I void my warranty and replace the thermal paste myself?

 

 

 

[Zando_]

2 minutes ago, Cupcakez said:

Should I void my warranty and replace the thermal paste myself?

Do a quick Google on the laws in your country, in most places the OEM cannot actually enforce the "warranty void if removed" sticker thing without running afoul of the law. 

[RAS_3885]

What do you mean by hotspot at 100-105 degress?

 

What are you using to read temps? If something like Afterburner or HWInfo65 is reading a GPU temp of 66 while gaming then you have no issues.

[Cupcakez]

4 minutes ago, RAS_3885 said:

What do you mean by hotspot at 100-105 degress?

Dunno exactelly what that sensor measures. I found it in GPU-Z. (Idle temps)

 

Quote

What are you using to read temps? If something like Afterburner or HWInfo65 is reading a GPU temp of 66 while gaming then you have no issues.

Still doesn't change the fact that I have to ramp up the fans to 100% to keep up when before they could keep up just fine at 60% or so

[RAS_3885]

2 minutes ago, Cupcakez said:

Dunno exactelly what that sensor measures. I found it in GPU-Z. (Idle temps)

 

Interesting. Looks like that's taking the hottest of all the temp sensors on the die, whereas GPU Temp is more of an average. Didn't know GPU-Z had that. I also don't know if that measurement is anything to be worried about. Do you have any other software you can verify temps with?

 

It's POSSIBLE the bump causes the GPU cooler to shift a little, or loosen a bit, but that seems highly unlikely to me. Thermal paste also is good for many years before needing to be replaced, so it should not be dried up after just a year.

[8-Bit Ninja]

I doubt its thermal paste. If this "hotspot" reading is correct and you have an increase of 20+ degrees then its possible a thermal pad has been knocked. But unless you're willing to take the card apart and possibly void the warranty then there isn't much you can do apart from attempting to RMA 

[Kilrah]

1 hour ago, Cupcakez said:

Should I void my warranty and replace the thermal paste myself?

Well your warranty wouldn't cover damage from the PC being knocked anyway, so...