I've gotten a pc from a good friend of mine. He used it for a year or so. I've used it for about two years. I have had no issues with it. The past few months, though, it started to shut down in the middle of some intensive games. I figured it was overheating. It steadily got worse. It's now so bad that I can't effectively play games and am forced to finally deal with it.
My pc is a custom build with an Intel Core 2 Quad PC @ 2.40 GHZ
4 GB RAM
NVIDIA GeForce GTX 260
Windows 7 Ultimate
So, I did get the latest Speedfan and noticed that my CPU was running around 62C just idling! OK. I went ahead and cleaned out all the dust from the case and fans. I replaced the CPU fan with a brand new Alpine 11 Pro Rev.2 (Arctic Cooling). I bought new thermal paste (Dynex). Cleaned off the old and put in the new. When running, it was a lot cooler. It was idling around 50C. Not great, but better!
At this point, the GPU was listed as 50 or 51C, which was the highest thing. The main CPU heat would dip to upper 40's and turn green, but the GPU would not drop below flame icon. I went ahead and decided to play a game and take a screenshot of the heat at that point. So....in the middle of the game, I tab out and see this:
So......I shut 'er down and pop out the video card. I try to open it up, but this video card isn't easy to get into . I'm able to open it up (mostly), clean out the dust, and replace the thermal paste. So, it all seems the best that I can do, but when I boot it up, I still see this when it's just idling.
What can I do? I know as soon as I start playing a game it's going to freak out again. Even at idle it's pretty hot. What can I do? Maybe I should even ask.....what can I do THAT IS COST EFFECTIVE?
It's not that bad, those 'older' cards often idled around thse temperatures, try running a game with GPU-Z running, it's a free app and will continue to log temperatures if it's left running in the taskbar.
The CPU idle is a little high, though, check those **** pushpins have locked fully into place it's a known issue with this design.
Well, I would be more than happy to hear that the older cards run hotter and it should be fine....except I'm still shutting down mid game. It doesn't even take that long.
As for the push pins....yes, they are a pain. In fact, I broke them when I was trying to clean it and had to buy a whole new fan for the unit. The Arctic Cooler has these odd little push in things that were a bit of a pain, but not as much as the regular pins. I did get them in after some colorful words and a few sore finger tips.
Ah, but which is causing the problem?
Grab a download of Prime95 or Intelburntest and run it with Speedfan open as well, see how hot the CPU gets.
If that's OK get a download of Furmark and run that, if GPU-Z is installed it'll automatically show you te GPU temperature.
I installed Prime95, but I have to admit, I have no idea what it's doing. The help function doesn't work and the wiki page is useless. Any idea what I am suppose to do with it? I know since I started running it, my temps have gone higher.
IMO your GPU temps are fine. GPU's can get hot and generally do. Your problem lies in your CPU. If I'm looking at your picture when you were running P95 correctly, I'm not familiar with speedfan, your CPU temps are WAY too high. Plus I can see BF3 in your background, which is a CPU intensive game.
If you had seen random video glitches and artifacts, which you haven't mentioned so I'm assuming you haven't seen any, that would hint to GPU issues.
I would suggest check airflow to your CPU and check your CPU fan for proper speeds. Also make sure you have enough but not too much thermal paste and your CPU heatsink is sitting properly on your CPU.
Have to agree there, the card seems to be running fine but those CPU and system temperatures are well above normal.
With the system off, remove the side panel and check the rear fan is clear, dust can clog the blades and vents and it's often difficult to see if the vents themselves are clear.
Boot up and check the fans are turning, it's not unknown for monitoring software to misread so a physical check is warranted.
If all else is fine, I suggest you remove and reinstall the CPU cooler, I've had this with a 775 build myself, I was convinced the pins were fully seated but they were not and the system shut down less than a minuet after booting up.
BTW, make sure all the old thermal paste is removed, even small traces can prevent good contact between the CPU and cooler. Computer shops sell cleaning fluids or you can use pure alcohol-no, not Vodka .