I've been getting some random shutdowns while playing hardware intensive games (like crysis) lately. The computer simply shuts down and I have to wait a while before it can be turned on again. I've been monitoring the temperature of my components and here are the results (all at full load):

motherboard (Gigabyte GA-MA780G-UD3H) - between 45 and 50°C
CPU (AMD phenom II 710, overclocked @ 3.2ghz) - between 40 and 45°C
GFX card (geforce gtx 260) - between 65 and 75°C

I did some research on temperatures for these three and apparently they're all ok. So I'm wondering if anyone could tell me what else could be the reason for this kind of PC "behavior"?

Now I've done some mem-testing and apparently there's something wrong with two of my ram slots (i got like 100 errors per second with one or both of these occupied). Does this necceseraly mean the slots are "dead" or could the reason for this be elsewhere?

Sometimes when you overclock the CPU the Front Side Bus get's overclocked also along with RAM. You might have simply damaged the RAM. What brand of RAM is it. Usually there are only certain brands you want to use for overclocking. OCZ, Mushking, etc...
My guess would be faulty RAM before a faulty motherboard. Are both Dimms giving you errors, you could try switching out the stick that is not giving you errors into the other slot to eliminate a problem with the Motherboard.

I agree with josmcc, your cpu may not be able to take that high of an overclocking level, or your gpu may be the problem. Do you get the BSoD? and if so, what is the error message and how long does it take to dump the physical memory?

I never got a BSOD, the computer just shuts down. About the RAM... I'm quite certain it's not damaged, because every stick passed memtest when in the first two slots, but i get thousands of errors in the other two slots. But wouldn't a faulty ram slot cause me other problems aswell (ive used all 4 slots for about 1 week (the motherboard is new) without any weird behavior, then after the first crash i started looking into it)? I'm getting really confused. I also forgot to mention that my computer has an old PSU (550W), could this be the cause of any problems?

I have just recently been dealing with a busted RAM issue, so I can sympathize. In my case, it was a bad stick, but the result was as you described, i.e. random shutdowns, sometimes of the program, sometimes of the whole computer. Also, in my case, while the computer just restarts, the BSOD does flash up briefly, though you'd have to be staring to catch it.

To answer your newest question, the faulty slots will only cause you trouble if you have RAM in them. Otherwise, no current will be pushed through the bad DIMMs, and no errors will occur. This assumes the problem is isolated to those DIMMs.

The power supply you have should be sufficient for the specs you listed (depending on the rest of the system, of course), and while age could be an issue, this is probably not the root cause of your problems.

If your board is new, you may want to send it in for replacement under warranty. Though if you don't plan to use the two bad RAM slots, you could just ignore it, and you'll likely be fine.

Vedikur - Lanslide Gaming PCs:
Gaming Computers Built For LAN Parties

Interesting stuff happening here. Yesterday i took the RAM out and placed it in the "faulty" slots again, ran memtest and got zero errors.

Interesting stuff happening here. Yesterday i took the RAM out and placed it in the "faulty" slots again, ran memtest and got zero errors.

never mad a good connection the last time ,i use a white [lead pencil] eraser to clean ram connections

I got a new "symptom" today twice. While i was playing a game, the PC froze completely. I did a "hard" reset but it didn't want to restart (just a black screen, no beeps or anything). I unpluged the power for a bit and after that i was able to start the PC again. Any ideas?

EDIT: When the PC didn't want to restart i could hear the fans starting to run, then it sounded like they stopped running for a sec and started again.. then stopped etc.

you could have a bad harddrive ,go to hdd the manufacturers web site and get a hdd utility and burn it to disk and boot to it and do the full scan

I'm getting really confused. I also forgot to mention that my computer has an old PSU (550W), could this be the cause of any problems?

Possibly, check the voltages during intensive use. It suprises me alot how people don't twig on to how often the PSU is actually the cause of problems... It's 1 of the 1st things I check now because alot of computers I've found end up with aging or faulty PSU which isn't supplying enough power to either 1 or multiple components, even 1 component not having enough voltage makes the entire system unstable. Sometimes it's just spikes of voltage, like most of the time it's giving it enough voltage & then it spikes a few times & then the computer goes. It's worth just testing another PSU just to see, it's sometimes a hassle, but a hassle that is worth it if the PSU is the cause of the problem. Save you running around looking at the RAM, HDD & all the other components.

You say it's during hardware intensive gaming, this makes sense, when all components need more voltage & thus power to run, really just try monitoring voltage or testing out a new PSU just to see.

Right, so I've tried putting another PSU in (a 650W) and it seems the problems are gone, so I'm marking this as solved. Thanks all for your help.