Geforce GTX 760 Running 102 Degrees Celsius and Crashing

grizzlymatt

Distinguished
Dec 19, 2011
27
0
18,530
I've been having this problem since the computer came straight out of the box, I tried raising fan speeds to 100% as well as playing with the case open. When I sit in the lobby of one of my favorite games, the gpu temp steadily raises to 90+ and eventually it hits 102 and crashes, I took a screenshot when it was at 97 and shortly after it hit 101 and wasn't crashing so I panic closed the game and it dropped down to normal levels within seconds. I've updated drivers, flashed my bios, tried re plugging my psu cords without really knowing what I was doing, I even tried posting in a BSOD forums since thats the error I got every time my computer crashed.

Is it possible that the video card being improperly seated could cause the overheating? It happens in all games I have installed after a while and in the windows assessment when it touches graphics. I haven't tried taking the video card out because I'm scared of messing something up, I know it's silly but still. I'm on the verge of paying someone 50 bucks to re plug and re seat all of my internal hardware properly.

I'm extremely lost on what I should do here, been at my wits end for 9 days now working on it for several hours a day trying to figure it out. I know Tom's hardware has some extremely knowledgeable people and that's why I'm shooting in the dark here, sorry to bother you but thank you if you can help even a tiny bit.

Computer Specs:
intel i5 quad core 3.4ghz cpu
nvidia gtx 760 gpu
kingston 8gbs ram 2 sticks
gigabyte lg1150 GA-H87M-D3H mobo
corsair 600w PSU

Gpu screenshot after a short time in the lobby(under a minute):
f4m.png


BSOD post on sevenforums:
http://
 

Blecter

Honorable
Jan 21, 2014
59
0
10,660
Have you tried applying new thermal paste? Occasionally some cards will come from the factory with badly applied thermal paste, or it'll be super dried out and be losing it's effectiveness. It's simply to do; shut down the system, let it cool for a couple minutes, unplug the power cables and remove the card from the motherboard. Unscrew the screws that hold the heatsink on (there are 4 on my EVGA ACX cards) and any power plugs. Clean the GPU itself, and the heatsink (I used break cleaner and it worked great - put some on a q-tip and clean the old paste on), then spread a thin layer of new paste on the GPU. Reassemble in the opposite manner.

I dropped 10C when I did it on my first 760. Granted my temps were not as high as yours, but it's a good step to start checking for faults. If that doesn't drop temps then call up your manufacturer. Who makes your card?

Another question, does your case have good airflow? If not then you need to install fans to move air from the front to the back. I dropped a buddy's GPU temps 20C simply by installing 2 120mm fans in the front of his case (he had 1 120m on his blocked up all in one CPU liquid cooling radiator and a 200mm on the top - not enough for even decent airflow).
 

grizzlymatt

Distinguished
Dec 19, 2011
27
0
18,530


I've got 4 fans or so inside the machine and it has lots of holes for airflow, there was no change when I opened the case compared to when I kept it closed. Where would I get thermal paste, best buy? That sounds simple enough and if its a possible solution I'd be more than willing to give it a try, would any thermal paste do?
 
If this is a new machine (you say it was happening 'straight out of the box') you should begin a warranty claim.
Do NOT attempt to remove or disassemble anything because you will void the warranty.
The problem is either the card heatsink is not mounted correctly or it has been dislodged in transit causing it to make poor or no contact with the graphic card main chip or the fans are not actually turning but at 100% you'll KNOW if they're turning, trust me. ;)
 

grizzlymatt

Distinguished
Dec 19, 2011
27
0
18,530


The system was all new parts put together by the company I ordered them from for a short fee. I guess I should start that warranty claim then ;/, just as a last check, there's nothing else anyone can think of? A fault in the card seems to be the problem? Going to talk to both Directron and possibly Gigabyte depending on how receptive Directron is. Apparently the RMA place for Giga is in CA but the Directron warehouse is all the way in Texas. I can still RMA a product if purchased through a 3rd party if I can find the S/N right? God I'm so stupid asking these questions. Sorry.

EDIT: The fans get SUPER loud when I turn them to 100% so I think that's working, since they make next to zero noise on idle.

EDIT2: I guess I've been asking a lot of stupid questions, a bad fan or improper cooling wouldn't make it hit 102C in 30 seconds would it ;/?
 

grizzlymatt

Distinguished
Dec 19, 2011
27
0
18,530


Thanks everybody who replied. I'm pursuing an RMA right away. Have a good day and hopefully if anybody else experiences a similar problem this'll pop up on google for them.