Three years ago and while I was playing Stalker: Shadow of Chernobyl, suddenly my nVidia 8800 GTS 640 MB got fried. It was summer and hot actually, and the card used to get very hot +80 C. After one month I purchased a brand new ATI Radeon 4890 1 GB and installed it on the same system. It's been three years and I am still afraid to play Stalker fearing it will damage my newer card.
Can a particular PC game really be the reason for a damaged card?
ASUS A8N32-SLI DELUXE mainboard with nVidia nForce4 chipset
2 GB DDR-1 400 MHz RAM
AMD ATHLON X2 4800+ @ 2.4 GHz
ATI Radeon 4890 1 GB
Thermaltake 650 watts PSU
Not really, if the game is too demanding the card will get hotter, but before frying the pc should restart, freeze or turn off, it's a measure to save the card, BUT if you insist on playing the game when the pc restarts it will end up as a nice rock.
This happened to me when I played WoW and Sunwell patch went live, my card for no reason started to heat and my computer froze or restarted, the solution was taking off the lateral panel and adding an extra fan until I changed my graphic card
Heat or static can permanently damage the chips on a video card or in a computer (i.e.,CPU). Performing tasks that require a high amount of computations like playing games or render video causes the GPU and/or CPU to work harder, generating more heat, so on a hot day with not enough ventilation, or a failed fan, you could lose a component. But it might have occurred with any game you played. Inspect your case, make sure there is proper ventilation, all fans are clean, remove any layers of dust, monitor the CPU and GPU temperatures while playing, open the case and add a fan blowing directly on the motherboard...
It is possible. For example - Battlefield 3 is a very GPU intensive (as well as CPU depending in situation) and will make your GPU draw more power. You can google it if you want to know more about it. Basically if your PSU is not up to the task and your GPU is trying to draw more power than your PSU can provide - that can cause problems (however your PSU seems to be enough).
However in your case it could have been multiple factors. That temp you mentioned 80+ is quite hot for a GPU and overtime it can cause it to degrade to the point of no return.
Another thing too - as your PSU heats up, it will become less efficient at delivering power.
I highly doubt you would have a problem with your "new" card playing stalker. What you should do is try playing it - and after about 1/2 hour to an hour check your GPU's temperature, or even watch the temperature by alt tabbing in the middle of the game.
Do you have good cooling in your computer? Whats your idle GPU temperature? Whats the load GPU temperature?
@ Dangi, actually my video card's RAMDAC got damaged during gameplay, but back then the PC did not do anything as a protective measure. My PC does shut down when CPU temps are so high, but as for my old GPU, I don't think there was any kind of such measure.
@ EnderWiggin: the case is clean and I recently cleaned the fans. Additionally, I removed the case's side cover.
@ Chainzsaw: I managed to take note of some of the temps and fan speeds while idle and while playing as shown in the graph below. The room temperature is 30.5 C, and the time of test is 2 pm.
The numbers for the under load test are for Mass Effect, all settings are maxed at 1280,800 pixels.
I actually always override AA, AF, and VS using AMD Catalyst. The settings for the test are:
Anti aliasing filter: Edge-detect, samples: 12X
Anisotropic filtering: 8X
Vertical sync: Always on
Adaptive AA: off
Catalyst version: 2012.0405.2154.37503
As you can see, the automatic setting lead to temps that are nearly 80 C. I think of turning on manual fan control every time I want to play a graphics intensive game and set the fan speed to 42 percent (around 3500 RPMs). Will this speed damage the fan? What do you think of my card's temps? are they really high?
80C is a bit on the high side. However setting your fan speed manually to 42% shouldn't hurt it. And 72C at 42% is actually pretty good. IMO if your GPU is running at 72C under full loud, that temperature should be pretty safe for long periods. Just try to keep the temps within 5C of 70C.
The only tradeoff between fan speed and cooling is loudness. GPU fans can get quite loud when cranked up.