GPU (I believe) is overheating

Mblack04

Honorable
Feb 27, 2013
16
0
10,510
Recently my computer started shutting down while playing games or recording gameplay. Specifically it shut off three times while playing World of Warcraft on the highest settings and also shut off while recording League of Legends gameplay on the highest settings. I am almost positive it is the GPU, I've been monitoring temps via CPUID Hardware Monitor and noticed my GPU spikes to over 100c, while my processor will sometimes reach 90 but usually stays around 80 and everything else is low.

I've tried cleaning out every bit of dust from my computer case as well as my heatsinks, I've re-applied the thermal compound to my GPU, I've tried putting a fan next to my computer case, but the temperatures still reach over 100c at times while gaming. It hasn't overheated in a while, but I haven't really given it the opportunity as once I see the temps start to reach the high 90s I usually stop playing (this only takes somewhere between 30 minutes to an hour depending on the game)

I really don't know what else to do so any suggestions would be great, my GPU is only about a year old and here are my PC specs:

Mobo: MSI 760GM-E51
GPU: ATI Radeon Sapphire HD 6850
CPU: AMD 8-Core FX-8120
RAM: 2x4GB (8GB RAM)
^^ All of these components installed Jan. 2012

PSU: XPOWER780 600W(780W Peak) ATX12V v2.3 - Installed Sept. 2011
HDD: Western Digital 120GB 7200RPM SATA 3Gbps 8MB Cache 3.5-inch - This is pretty old, probably installed early '09

I have a 120mm fan on the rear of my case and a large fan on the side, not sure exactly how big probably around 200mm (not very knowledgeable on fans)

With my setup I should easily be able to run WoW on ultra settings without overheating and I've done it many times before in 25 man raids that are more graphically intense, but I was overheating in starting zones on a new character.
 
not specifically gpu related... most likely summer related and temps in general.

not sure why, but it sounds like your system isn't venting air at all... you shouldn't be seeing 80-90 in your cpu... which is why your system is shutting down... the mb is protecting your computer.

nor should you be seeing 100 on your gpu.

that speaks to me either you're living in a sauna or your pc case has the worst airflow ever.

did you move your case into a wooden desk or closet recently? are your fans actually working? All fans run on the 12v rail on your psu... if that rail fails you can lose your system fans fast (as well as your disk drives)
 


its not just his gpu fan, his whole system is running VERY hot. i mean 90C for a cpu will trigger the automatic safeties and shut down even an Intel... and it will kill an AMD cpu. Hell that amd cpu shouldn't be over 60C, but its running up to 90?!!

god... you're gonna burn something out. i can't believe your fans are working at all anywhere in that system.
 

Mblack04

Honorable
Feb 27, 2013
16
0
10,510
Thanks for the responses, I'll try to answer all the questions.

Trust me it is not summer related, I live in the awful state of Minnesota and it is 30 degrees out right now, it is around 68 in my house at the moment I think. Also, my computer has been in the same spot since I moved into my house almost a year ago.

All of my fans are running, according to my AMD Vision Engine Control Center, the GPU fan is usually around 40% speed, but goes to 100% during high performance times.

I'm uploading some pictures of my case to see if that will help shed some light.
 

Mblack04

Honorable
Feb 27, 2013
16
0
10,510


^^ This. Your fan setup is good and even with really hot seasonal weather your PC has enough cooling.
 

Mblack04

Honorable
Feb 27, 2013
16
0
10,510
Not really sure what else, I know my mobo has two fried RAM slots but that is the way it came and I've never had this problem before (I probably should have requested a new one but I only use two ram sticks so I didn't bother) and since my PSU isn't even two years old I would really hope it isn't already going bad.
 
mmmm...

gotta say i'm a bit stumped... i suppose the gpu/cpu fans could be dying... it happens from time to time... all it takes is a little dust in the bearings...

if the system is off, and you flick the fan blades, does it have a nice snappy "spring back" in the opposite direction after spinning freely and silently in the direction you flicked it?

if there is no "back-spring" at the end of the spin, the fan is on its last legs... if the fan doesn't spin freely and quietly it's probably on its last legs as well... lets start with that.
 

vipervoid1

Distinguished
Feb 10, 2009
323
0
18,860
What is your environment temperature ?
Here in Asia, is 32 C
Which is normal for us , And my setup without air conditioner, my Asus EAH6850V2 normal operate at 45-52C
And 70-80 C for full loading ~
I think the GPU temperature is quite high for 6850 ~
 

Mblack04

Honorable
Feb 27, 2013
16
0
10,510
I've tested the fan blades and they do come back after spinning, I wasn't aware that was a method of testing but I happened to notice it. I'm watching the fans spin and they both seem to be normal, I've wiped all dust off of them and dusted out both heatsinks. My only thought is a bad PSU, I've had it happen before so maybe I just have bad luck with them. I'm not sure how to interpret the voltages on the HW Monitor logs so if someone wants to check that out for me and let me know whats up there that'd be great, I posted it a few posts above.

Edit: It is 0c outside right now, lol, I live in Minnesota in the US.
Also yes the temps were bad before I re-applied the thermal paste, this has been going on for a couple days and I just re-applied the paste today.
 


since apparently temps don't change with a fan stuck in the side of the case, with the case door off, i'd say intake/exhaust in the case isn't the issue.

i'd clean your cpu/heatsink and gpu/heatsink, reapply the paste, and reseat the sink.

unless your temp monitors are way off your system is running way too hot for an AMD... the GPU is hot... but not "break the gpu" hot, just crash the system hot... the cpu however is hot enough to break it.

Are you using the stock cpu cooler? that's what it looks like in the pics... cause the stock amd cpu cooler is not good enough for an FX 8 core.
 


That rules out the chance of any user error then. You said that this has been going on for a couple of days? I assume your temps have been much lower before then?
 

Mblack04

Honorable
Feb 27, 2013
16
0
10,510


I'm going to head up to my brother's tomorrow and try his PSU and see what happens. There's not really much else I can think of to try at this point.
 

spat55

Distinguished


If it isn't that then it must be the motherboard, but I think it is the PSU as it isn't the most wonderful thing in the world, and MSI isn't bad.

 

Mblack04

Honorable
Feb 27, 2013
16
0
10,510


I've re-applied the paste on my GPU already, but I can do the CPU as well, and yep it is the stock heatsink, I could look into upgrading that if nothing else is working. My CPU doesn't seem to be getting as hot as I thought it was though, I just played Devil May Cry for about 10 minutes and watched my GPU go up to 92c and my CPU stayed right around 60c and maxed out at 65.




To be honest I never really checked my temps before I started having the issue, but I've never had any real problems before.
 
I checked your voltages and they are within range of others. The only common factor that comes to mind that would cause these issues is either motherboard issues or poor ventilation.

Try putting a floor fan up next to the exposed case and see if that has an effect. It could just be a faulty motherboard reporting false temps.

*Edit
MSI isn't the best mobo brand, I have to rma allot more from that brand than say Asrock.