My PC shuts itself off after long periods of gaming.

[The_Absolute]

I played Crysis on all of the "optimal" graphical settings for my Radeon HD 4850 (512 MB) and I got to the point in the game where you enter the frozen alien hive, after killing that North Korean general, and then my PC just shut off by itself. I'm not talking about a crash back to desktop, I'm talking about the power to the PC being turned off for no apparent reason.

This happened to me a few times before, I played STALKER Clear Sky on "maximum" quality settings, and it shut off after about 25 minutes of playing.

I'm running Crysis on 1024x768, 0x AA, Vsync, on all "high" (optimal) graphics settings. Could you please try to explain to me why I am having this problem? Has this ever happened to you?

Every component on my computer is running on it's factory clock speeds. Including my GPU.

 

[mgb_phys]

Check the tempertures and any safety shut off in the bios

 

[story645]

Could you please try to explain to me why I am having this problem? Has this ever happened to you?

Overheating's the standard explanation. Most modern hardware is set to shut down at extreme temperatures as a safety feature.
Take off the side, push everything to max, and see it if still happens. Or just buy a case thermometer and see how high it gets.

 

[vincentm]

your CPU is heating up too much, reseat the cpu with a fresh application of Arctic Thermal five, install Speedfan. Your BIOS should also have cpu temp threshold settings.

 

[dE_logics]

The first step will be to open your PC and clean it.

Next turn it on to see if all the fans are working at reasonable RPM.


If there is no problem...time to change your cabinet to a cooler one.

 

[The_Absolute]

The first step will be to open your PC and clean it.

Next turn it on to see if all the fans are working at reasonable RPM.If there is no problem...time to change your cabinet to a cooler one.

CPUID Hardware Monitor say's my CPU's temperature reaches around 40-45 *C while playing Crysis. It idles at around 32-35 *C. I don't think it's my CPU, it's my GPU that reaches over 100+ *C while playing graphically intensive games. Even with the fan speed running at 100%.

If I were to RMA my Radeon HD 4850 (512MB) I would be out of a graphics card for quite a while. It's probably faulty, and the fact that I bought my computer as a whole, and didn't purchase the GPU individually, I probably can't RMA it unless I sent my whole machine in.

Every part of my computer is using it's stock clock speeds.

 

[Illbilly]

Try resetting the cooler on your GPU.

 

[estro]

As said before me, it is probably a heat issue,

Download and install some kind of stress program (like OCCT - http://www.ocbase.com/perestroika_en/index.php?Download)
Stress GPU and CPU for at least a hour, while following the temperatures.

If CPU overheats:
1. Open the case, clean it.
2. Take off cpu HS clean it and reapply thermal grease.

 

[True Cherokee]

You could also try the program MSI Afterburner. It allows you to manually regulate fan speed and will also show the temperature and performance of your GPU

 

[moogoomonkey]

OK I agree with all of the above. You can also buy auxiliary fans that you can install next to your graphics card in one of the free PCI slots to suck some of the heat out of the case. Also since this is happening sometimes within an hour of booting up (your S.T.A.L.K.E.R example) it could also suggest that there is a problem with your PSU. This happened to me a few years back and I had no idea what was going on then it just died completely. PSU replacements are VERY simple and cost relatively cheap now'a'days however don't just buy the cheapest one you can get your hands on, since many cheapskate manufacturers just pile all the power onto the crappy 3A rails, you need one which has maybe a 80plus certification and you should be good. Also one last note, look inside your case and make sure you buy one with all the correct leads ;)

 

[rhody]

It seems like the OP is no longer with us or banned for s short period, never the less, there is a reliable way to determine of your PC is overheating with software. First, go to BOINC.com and download the user sharing software. The go to Einstein.home and download the work packages to run software to analyze the data. Then, set the config settings in BOINC to use 100% of CPU, and watch as your PC will rapidly heat up, every time. I needed a more robust heat sink and fan, and that solved the problem. This is a great way to prove once and for all if you have a heat problem, it is a slam dunk.

If it weren't for running BOINC and Einstein@home work packages on my two AMD CPU cores for 30 minutes or so, I never would have been able to reliably reproduce the overheating problem. The repeated variability in cpu use and stress is what caused my power supply to shut the PC off. Watch because it jumps repeatedly between 30 - 100%, it is pretty cool.

Rhody...