I'm new to the forum and I'm hoping that someone can help ease the sense of dread I have about my computer.
Recently (about 3-6 weeks ago) my computer started restarting sometimes during games and later other applications, even minor ones. At first I thought I was hitting the power cord with my foot, but after moving all the cords, it was still occurring. I have the option to restart for errors turned off. This is something else. No blue screens involved.
After a lot of reading, I've become fearful that the problem are the cpu temperatures.
MY SYSTEM (self assembled with help of a friend in April of 2009):
core i7 940 (not overclocked) @ 2.93 GHz - stock cooling without the optional fan installed.
Nvidia 285 GTX
Creative SB Xfi
Coolermaster Cosmos case
Thermaltake 850w Toughpower PS
I have 4 fans (sort of): The cpu/heatsink fan, another fan on the back of the case perpendicular to the cpu fan, a fan on the front of the case near the bottom, and a fan on the top of the case over the cpu.
I keep the side off my computer because it appears to keep it cooler than with it on. I have blown compressed air around the cpu and other places and it appears to be relatively free of dust aside from a light scattering.
The thermaltake is a wire monster, but none of them are anywhere near the cpu as the case is pretty gigantic.
At idle (0-3 percent CPU) my core temperatures are about 68 Celcius.
I am afraid to run any tests as the cores reach about 85-95 Celcius at even 4-10 percent loads.
Interestingly, the GPU idles at about 47 Celcius and never goes near the CPU temperature when gaming.
I understand that the thermal paste can be on wrong (I am using the stock cooling for the cpu), I know that airflow is important, but this system has run largely flawlessly for the most of a year. I don't have temperature data before the last couple days unfortunately. I seem to recall a few restarts here and there maybe as far back as 2 months ago, but VERY rarely if ever. This is something that has really started to go wrong after nearly a year in the last 3-6 weeks (at least noticeably).
No changes have been made in the case since purchase/assembly with the exception of the new power supply. I previously ordered a 650 which couldn't hope to run the 285 graphics card. This replacement happened in April of 2009.
Could anyone help me understand my options and the reasonable chances of possible causes? I'm really stressed about this because I'm out of work and the last thing I need is for this thing to go belly up as my entire world is on this computer.
I'd be almost embarrassingly appreciative of assistance. If you require any additional info, please be kind and tell me how to obtain it. I'm a professional software guy, but woefully amateurish about hardware.
Well since your cpu is cleary running way to hot and overheating there is only really one thing that could be the problem.
Your cooler is not functioning properly. It is not seated properly anymore. Remove the heastink and clean the old paste of the cpu and the base of the cooler. Apply some new paste and make reinstall the cooler.
The cause is most likely one of the push pins on the Intel cooler is no longer secure. And the cpu and cooler no longer have good contact which is why the temps are high.
Another unlikely cause in this case is your cpu is massively overvolted.
The paste does go bad and degrade overtime. But not in this short amount of time.
The HSF will not become unseated overtime by itself. But if the original install was not in there perfectly overtime it may give way. Especially if you move the system around or when you installed the psu. All it takes is for one of those pins to not be secure.
In the mean time when the machine is powered down check and see if all the push pins are locked in and not moving around. You may be able to tell just by that if they are in there properly. What sucks about these coolers is the actual best way to check that they seated properly is to look at the back of the motherboard. Which requires removing the motherboard from the case.