GPU Overheating

Mister Walrus

Junior Member
Jun 16, 2005
23
0
0
*This is a repost, as it is now in the correct forum.

So recently while playing the fabulous Guild Wars, I have started crashing often, from 5 minutes to 60 minutes in the game. I suspect that something is heating up, as I have tried running the game on either of my two hard drives and they both have the same problem (and they are radically different - one is a WDRaptor 74GB and another is a Samsung 40GB). The computer essentially freezes in place, user input is no longer acceptable, and physical reboot must be performed. So I thought something must be overheating, possibly?

I've put some data of temperatures as recorded by SpeedFan. The decline in temperatures in Temp2 starts IMMEDIATELY with the closing of the Guild Wars program, they have to be correlated. Conversely, while playing low-maintenance games like Diablo II or running any other standard application, there is no significant rise in temperature nor is there any ever such crash.

Data:
Playing GW (before crash) -
Temp1 - 82 F / 28 C
Temp2 - 106-110 F/ 41-43 C
Temp3 - 109 F / 43 C
After quitting GW and running menial tasks for 2 mins -
Temp1 - 82 F / 28 C
Temp2 - 92-94 F / 33-34 C
Temp3 - 109 F / 43 C

My stats are as follows:
CPU: 2.4 GHz P4 w/Hyperthreading
GPU: ATI Radeon 9800 Pro
RAM: 1022 MB DDR Kingston or Corsair, forget which.
Sound: Audigy 2 ZS, 24-Bit.

So first of all, I have no idea where the Temp2 indicator is located. Secondly, if it is a GPU problem, is there a cheap (~$25), efficient remedy for the solution?

Please let me know!
Thanks!
 

Steaksauce

Senior member
Feb 15, 2005
255
0
0
Why don't you use Everest so that you can pin point to where the heat problem is or check your BIOS.

It looks to be your CPU though, but 43 is not hot at all.

Have you updated the GPU driver to the latest?
 

Mister Walrus

Junior Member
Jun 16, 2005
23
0
0
I downloaded Everest - what an awesome program! It appears the Temp2 is actually the CPU, as you said. My GPU driver is indeed the latest.

Any suggestions as to why this might be occurring?

I've had GW for a year now and the problem only started up a week ago or so.
 

Steaksauce

Senior member
Feb 15, 2005
255
0
0
What Power supply are you using? Do you run "updated" virusscans and antispyware often and defrag the hard drive? Have you tried to run memtest86 to detec RAM errors? Let's start with the basics and report back.
 

letdown427

Golden Member
Jan 3, 2006
1,594
1
0
Those CPU temps definately shouldn't be anything to worry about. In fact, unless it's a really faulty sensor, they're something to be proud of! lol

What's temp 3? As that doesn't appear to change, it's possible the sensor is fubar'd. It could be the NB chipset temp, and if that's too high, that'll cause instability? What mobo is it you're using? Maybe worth checking the NB heatsink, see if it's getting toasty hot when you play GW.
 

Mister Walrus

Junior Member
Jun 16, 2005
23
0
0
The Motherboard is: ASUS P4S800D
NB: SiS 655FX
SB: SiS 964

I will check where Temp3 is (currently it says "Aux" on EVEREST).
 

pelikan

Diamond Member
Dec 28, 2002
3,118
0
76
Sounds a video card issue. Is the fan still working on it? Does it feel super hot on the back of the card while running a game (on the opposite side of the heatsink)?

If the temp seems ok then I would try underclocking the card to see if that solves the problem. It also may be a good idea to use Driver Cleaner to completely uninstall the drivers and then reinstall them and see if that helps.
 

Mister Walrus

Junior Member
Jun 16, 2005
23
0
0
I reproduced the problem today, except this time in a room about 7 degrees F cooler. It took about 1.5 hrs of running the game for the computer to freeze in the same manner.

The video card's fan is running. Immediately after shutdown I touched the back of the video card - it was hot to the touch, couldn't hold my hand there for more than a few seconds (but it cooled down shortly after that). The North Bridge was warm-hot-ish but certainly not as hot as the video card. Any suggestions in light of this? Thanks.
 

zagood

Diamond Member
Mar 28, 2005
4,102
0
71
Check to see if your card supports temp monitoring:
http://forum.techpowerup.com/showthread.php?t=92

If it does, use ATiTool or RivaTuner in the background to chart temps.

Crack open your case and check around for dust buildup on the 9800 heatsink/fan. Blow out with compressed air. Hold the fan steady so it doesn't spin while you're cleaning it.

Possible it's a power supply issue, even though the TB should be able to handle your system with no problems. Chart the voltages with speedfan, or better yet, a multimeter.

Defrag your hard drive.

EDIT: cooler for under $25
http://www.directron.com/vf700alcu.html


-z
 

Mister Walrus

Junior Member
Jun 16, 2005
23
0
0
I don't think that my card supports temp, as neither RivaTuner or AtiTool read temperatures. I defragged my hard drive yesterday to no avail. However, when using AtiTool, it said Core: 378, when I raised the core using the tool slowly, at around 395 the computer experienced the same crashing problem. Any thoughts? Just buy the cooler?