Computer loses power during gaming

B

BauerUK

Geek Trainee
Note: I wasn't sure how to classify this problem, since I suspect it could either be overheating, power or maybe another problem. I trust that the post will be moved to the correct forum once more is known.

About 1 year ago, I bought a new computer (specification below) with the primary intention of playing games, coding, watching HD movies, and it worked splendidly for that time.

About 2-3 weeks ago (maybe a little more) whenever I play games for around 30-40 minutes my computer will completely lose power. No error message, no blue-screen of death (even though that option is turned on) -- nothing. Just shuts out completely.

I suspected that maybe it was down to the game I was playing, but that doesn't seem to be true as the same thing happens on all games I've tried playing.

My next guess was maybe a faulty power supply. I don't have a spare and didn't want to purchase a new one just to test the theory, so I'm not sure about that. However, I did lower the graphics options to medium/low for one of the games ('Left 4 Dead') and I'm able to play that just fine now (I'm not sure that is the cause of the problem though, since--just a month or so, ago--I was able to play these games, including Crysis, Far Cry 2, GTA IV, Left 4 Dead at maximum settings without a problem).

After installing my motherboards (Asus) 'Probe II' software, I noticed that when I'm playing a game, the Vcore, +3.3, +5.0 and +12.0 settings fluctuate and my CPU temperature begins to rise from about 48C (which it is at right now, when I'm just idling/chatting/using the web) to around 57-60C (at which point I closed the game to avoid the same problem).

Computer Specification:
  • Antec EarthWatts 430W PSU
  • 2x Western Digital WD5000AAKS 500GB SATA II 7200RPM HDD
  • Intel Core 2 Quad Q6600 G0 CPU
  • Asus P5N-E SLI 650i Socket 775 PCI-E ATX Motherboard
  • 3x Kingston 1GB DDR2 800MHz/PC2-6400 Memory Non-ecc CL5 Unbuffered 1.8V RAM
  • ASUS NVIDIA GeForce 8800GTX GPU
  • 2x 22" LCD Monitors
  • Coolermaster Cosmos RC-1000
No extra cooling devices, other than the CPU fan and chasis fans that came with the case.

I'm torn between whether this is overheating or something wrong with the power supply unit, which in turn is causing overheating (is that even possible?). I'd be happy with having to pay for a new power supply unit, or cooling system (or even both, I guess). I'd just rather avoid the cost if possible.

Any help would be great. Apologies if I've left out any information, I'll be happy to provide any further information about the problem if something is needed.
 
Deffinately sounds like the PSU, i would guess that processor alone requires at least 90-100 watts for operation, perhaps core temperature being raised quite high for more demanding applications could be down to the heatsink not getting enough power to cool the processor sufficently. For the amount of hardware installed in your current setup i wouldn't use anything less than 550w.
 
Vcore, +3.3, +5.0 and +12.0 rails will fluctuate under load, that's nothing to be concerned about (well, within reason, some PSUs handle it better than others). Your CPU temps look fine as well, although the idle is a little high. 60C is nothing to worry about, these chips have a high tjmax.

Honestly, it could be a handful of problems, so you'll just have to start troubleshooting. I've known people in that identical situation and it was caused by faulty RAM. Burn a disk with memtest86+ and test each stick. Check your GPU temps and make sure it's not overheating. Try taking the sides off your case and moving it to an area with more ventilation to rule out overheating.

Start with one component and work your way down the list. PSU is a possibility, but 430W should be adequate for that system.
 
Firstly, thanks to both for your responses.

HardwareAffair said:
Vcore, +3.3, +5.0 and +12.0 rails will fluctuate under load, that's nothing to be concerned about (well, within reason, some PSUs handle it better than others). Your CPU temps look fine as well, although the idle is a little high. 60C is nothing to worry about, these chips have a high tjmax.

Honestly, it could be a handful of problems, so you'll just have to start troubleshooting. I've known people in that identical situation and it was caused by faulty RAM. Burn a disk with memtest86+ and test each stick. Check your GPU temps and make sure it's not overheating. Try taking the sides off your case and moving it to an area with more ventilation to rule out overheating.

Start with one component and work your way down the list. PSU is a possibility, but 430W should be adequate for that system.
Click to expand...

I took the side-panels off my case, and plugged in a standard fan and pointed it at the motherboard. This dropped the CPU and MB temperatures down significantly, and I was able to play a game at full spec's for a while without either temp's going over 40C.

I'll keep trying this for a while, and if I can't replicate the problem -- I'll have to assume it's over-heating which is causing this. (Which component is overheating? I don't know. I'm going to say CPU since that seems to be the one that gets highest before the problem arises.)

If that is the case, where do I go from there? Is there a recommended cooling method for such a system? Am I going to have to make significant modifications to my case, maybe? While I have significant experience with general hardware replacements and upgrades, cooling and case-modification is completely new to me.

Thanks again, guys.
 
First of all, make sure that all cables are tied together with tie-wraps to increase the airflow. Secondly, consider buying a better CPU cooler and some 120mm case fans (IF these can be put into your case).
I used to have the same CPU as you got, and I can tell you that it's really hard to cool it :)
Might even suggest a tower-heatsink to cool it, like the ones from Coolermaster.
 
I've been reading more and more into that Coolermaster case and it looks like general cooling is a very common problem. Most people toss another 120mm fan in that optional hard drive space, and then mod the case to allow for another one or two 120mm intake fans. Checkout some videos on youtube to get some ideas. Remember, you just need to remove dead spots and have a little air running across all major components. You don't need to turn your system into a tornado. In my Lian-Li case, after the overclock, my northbridge and ram heatsinks could literally burn your finger if you touched them longer then a couple seconds. I just strapped the lowest speed, quietest 120mm fan I could buy on them, and now they're barely warm. I don't know the specifics of your Cosmos case, but you might be able to hookup a fan near the bottom of the motherboard, blowing upwards to the top exhaust fans.
 
HardwareAffair said:
I've been reading more and more into that Coolermaster case and it looks like general cooling is a very common problem. Most people toss another 120mm fan in that optional hard drive space, and then mod the case to allow for another one or two 120mm intake fans. Checkout some videos on youtube to get some ideas. Remember, you just need to remove dead spots and have a little air running across all major components. You don't need to turn your system into a tornado. In my Lian-Li case, after the overclock, my northbridge and ram heatsinks could literally burn your finger if you touched them longer then a couple seconds. I just strapped the lowest speed, quietest 120mm fan I could buy on them, and now they're barely warm. I don't know the specifics of your Cosmos case, but you might be able to hookup a fan near the bottom of the motherboard, blowing upwards to the top exhaust fans.
Click to expand...
And to get the best airflow, remember that the cold air needs to get sucked in at the lower front and the hot air gets blown out of the higher rear of the case :)
 
You must log in or register to reply here.
Back
Top