PC turns off after a few minutes of launching a game, whats the problem?

[kidgarrett]

I've had my PC from MESH, running VISTA for 2 years now and everything works fine under normal load / tasks such as office, outlook, internet browsing etc. However, when I'm playing games - any games at that, within a few minutes the PC will just turn off - completely dead. Powers back on fine but there are no errors reported in the Windows Event Log. This has only started happening recently, approx. last 3 months.

The only changes I have made to the PC in this time are Windows Updates and Nvida Graphics card driver updates. I have rolled back the graphics card drivers to see if this was the problem but the fault still happened so I don’t think it is a driver issue. I have since gone back to the latest drivers.

I can only assume that when playing games, the CPU is working harder so it is a power or heat problem? Does any one else have the same issue - and how did they fix it?

Any help or advice would be really helpful and going mad now! :-(
Mark

[Splash]

Firstly... cross posting isn't necessary.

Secondly - chances are it's the temps, though I guess it could also be a power issue. Post your full spec and temps at idle and load of your CPU and GPU.

[WhiteGiant]

Yep, temps or PSU.

Give the inside of your case a good clean.

[peterb]

I woul sy overheating somewhere - the problem is determinng where!

Operate the computer with teh case off and check that both the CPU coller fan and PSU fan are working correctly - and also see if the problem still happens with teh case side off. If it doesn't - then it is a heat problem.
Do you have a ny temperature monitoring software installed/ If so that will give you a clue if the CPU tempo is going high.

If neither of those, then I would suspect teh PSU, but the only way to check that really is by substitution. What make ids the one you have and how old is it?

(Thread in Hardware has been closed)

[kidgarrett]

Thanks very much to everyone who has replied - great knowing people are here to help. Sorry about the cross posting - kinda new to forums!

Have taken the hoover to the PC this morning! - inside and out! so agree this must help!

Have been running this morning without the cover off - only turned off once in the 3 hours as opposed to every 20 minutes so sound advice on the heat being an issue.

I managed to install some software called "RivaTuner" to monitor the GPU temp and fans - the idle temp is now at 71 for the GPU and with the lid on it went up to 80 when playing a game, - with the lid off it peaked at about 74.

The current power supply was shiped with the PC, purchased from MESH, and is now 2 years and 2 months old - it is: OCZ 1000PXS. I only have one graphics card, a NVIDA 8800GTX.

IS there any software to monitor the general PC temp / fans / CPU that anyone would recommend?

So to go forward I guess I need some software for CPU temp monitoring - but find it stange that no probs for nearly 2 years and now it's every day! - could this be simply cause of a build up of dust around the chips / PSU that I can't get into to clean? or could it simply mean that the PSU under load is going faulty / brushes wearing?

Again, I really appreciate everyones time so thanks in advance.

Mark

[Splash]

If the fans on the CPU and GPU heatsinks are spinning without any nasty noises then it's likely just a buildup of dust, which a liberal application of compressed air will help with. I personally use Everest to monitor temps and voltages - you can get a free 30 day trial (I think!) from www.lavalys.com

[Blitzen]

I've had my PC from MESH

Mistake number one

I agree with the above, this is almost certainly a temperature issue.
After 2 years, even if the house is immaculate, the dust build up would be enormous. Also, you may have to place a few fans in there and fiddle around with the exhaust so get rid of as much heat as possible.

Then, the GPU and CPU fan/Heatsink need a good 'blow out'.

BE CAREFUL using a vaccum cleaner to do this though as moving the dust in this way, charges the dust particles and you may blow something inside the PC.

[piggin23]

I would get one of these
http://cgi.ebay.co.uk/COMPRESSED-AIR...1%7C240%3A1318
Give all the fans, psu, cpu & gpu heat sink a good blow out.

[GoNz0]

yeh get compressed air, those 8800gtx's are terrible for dust getting stuck inside the heatsink assembly

[kidgarrett]

Great advice guys - thanks for your time helping with this. Will get myself some air in a can and give the PC a good blast.

Cheers. Mark

[TickledPink]

Two things that don't appear to have been mentioned. First, on most modern motherboards you can view the system's temperature in the BIOS. This won't show your peak temperatures under load since you have to wait for the machine to switch itself off and thus cool down slightly before you can switch on and get to the BIOS. However, if when you get to the BIOS any of the temperatures seem suspiciously high (if the cooling is really bad then even under minimal load you could see the CPU temp especially continuing to rise) then you have a fair idea of where the problem is.

In my experience, this kind of issue usually (read 90%+ here) stems from poor CPU cooling. I've yet to come across a graphics card causing this kind of issue. If it was the graphics card overheating then I'd probably expect to see some kind of artifacting on the screen and maybe a system lockup before a complete shutdown - unless it's an integrated GPU. Shutdowns are initiated by the motherboard to protect itself and the CPU from damage.

Secondly, something else you can try is applying some thermal paste between the CPU and the heatsink. Sometimes it seems that simple wear and tear on a board or the heatsink retainer brackets can cause them to flex ever so slightly and lose contact. By applying the thermal grease almost as a mortar you're improving the connection between the two components. I've had this issue a couple of times where it's been fixed simply by adding the paste.

And if it isn't that, it's either the power supply or the motherboard. Power supplies tend to be the first things to go on a PC (hard drives are a very close second) and it could be a sign that it's on its way out.

[jordan427]

This is very likely to be heat issue, give it a good clear out and provide good ventilation

[smithy1158]

I would buy a set of cheap 120mm fans and cable tie them wherever you can facing your GPU.

Just a cheap cooling solution that usually works great.

[kidgarrett]

Thanks for your update Tickledpink. I did go and have a look in the BIOS under "Hardware Monitoring" and soon after a power off (even with the lid / side panal off) the CUP temp was 83. After sitting in the BIOS for prob 5 mins this went down to 74. Does 83 seemly wildly too high?

Like you mention, it would have been higher at the time of the crash and with the time I spent swearing and restarting the machine / entering the BIOS would have allowed it to cool a little I guess.

So my actions are to use the air can I have ordered from ebay (thanks to Piggin23) but you think it may be worth swapping out the PSU anyway?

Really amazed at the amount of helpful responses so many thanks to everyone on the page.

Mark

[Blitzen]

but you think it may be worth swapping out the PSU anyway?

Seeing the ones placed into MESH systems then yes. But only for the safety of hter components and piece of mind.
I doubt it will effect your temps much though.

Is the machine out of warranty? If so, replace the CPU HS/Fan.

Bios update possibly would help also. <--long shot but worth a try

[shadowmaster]

Seeing the ones placed into MESH systems then yes. But only for the safety of hter components and piece of mind.

He has an OCZ 1000PXS 1000watt psu (manufactured by FSP), so unless its causing the problem which I doubt, then there is no need to change it.