The game constantly crashes the whole PC

[AkiA]

The game constantly crashes the whole PC

I'm starting to get frustrated that I will never be able complete this game a second time. Just today I tried to play one arena game and it crashed the whole computer after starting the second battle (well I died in the first 15 seconds of the first one). Sometimes I manage to play 20-30 minutes, but then I get too scared that it will crash and exit. Many times it crashes within five minutes of starting the game. I don't think it's heating issue, since it sometimes occurs so fast after starting the game and I have no problem with other games (Don't remember when some other game has crashed the whole computer, well only Witcher 1 ;/)

My pc is
MSI P45 NEO3-FR
Intel Q6600
Club 3D HD 4850
6GB 800MHZ DDR2
WD Caviar 640GB
64-bit Windows 7

Stock clocks, but coolers changed for cpu and gpu.

 

[AkiA]

*bump*

 

[gnoupi]

If it's related to heating, most likely it will trigger the whole computer to turn off in emergency (well, at least it does for my laptop, I don't remember for desktops). TW2 is very demanding, and it tends to heat the card quite fast.

Just to be sure, you can try to use GPU-Z.

Click the "continue refreshing in background" and "log to file" checkboxes, and let it run while you play. That way, if it crashes again, you will be able to check the file, to be sure about the temperature at this moment.

 

[AkiA]

Okay, finally got a round to try it once more. My PC had been stable for the week and on the whole time. Tried two Arena-battles and it crashed after I tried to buy something from the merchant. Here's the GPU-Z log, just before I fired up the game and the last line of the log.

Date , GPU Core Clock [MHz] , GPU Memory Clock [MHz] , GPU Temperature [°C] , Fan Speed (%) [%] , Fan Speed (RPM) [RPM] , GPU Load [%] , GPU Temp.(DISPIO) [°C] , GPU Temp.(MEMIO) [°C] , GPU Temp.(SHADERCORE) [°C] , Memory Usage (Dedicated) [MB] , Memory Usage (Dynamic) [MB] , VDDC [V] ,

2011-10-06 23:01:25 , 625.0 , 990.6 , 59.0 , 0 , - , 1 , 58.5 , 66.5 , 56.0 , 112 , 46 , 1.123 ,

2011-10-06 23:07:49 , 625.0 , 990.6 , 76.0 , 16 , - , 99 , 76.5 , 85.5 , 74.0 , 494 , 79 , 1.123 ,

Maybe confusing to read, but the highest temperature value I see is 85,5°C and that shouldn't be too high.

 

[GuyNwah]

I'm confused by the fan speed on the second line: is that 16%? If that measurement is true, it suggests you still have some of the fan profile trouble that the 4850 was known for when it came out. It should be higher than that when the GPU is hot. And there's still the possibility that the reading you see was not the reading at the point of failure. The GPU heats up and cools down quickly, and unless you see it right at the point of failure, you'll get a reading that is lower than truth.

Anyway, a way to assure yourself that it isn't GPU overheating is to download and run Furmark. This will give you an overlay with the GPU temperature on a stripchart while it is running. You want to see the GPU temperature come up to a safe maximum (95C or less) and stay there.