Computer shutting off during certain games ..

Yes.. Another problem with my computer. This time, my computer shuts off randomly, as if someone is just ripping out the power cable, during games like GTA:IV, Assassin's Creed and Dead Space. It happens in GTA:IV when I try and max out the settings (ignoring the warning in red), and then perform a benchmark. As soon as the game is about to load .. BOOM .. It shuts off. If I lessen the graphics (when I tested, I turned everything down to the lowest setting except my resolution), it runs fine and performs just fine, but as soon as I turn everything back up or even a little bit less than what my "max memory of my graphics card allows", it will just shut off completely again. Assassin's Creed does it randomly, Shadowrun does it randomly and so does Dead Space. Crysis did it as well when I went into nVidia's settings and turned on "Maximum Power Performance" under Power Managerment (or something like that) for specific program settings. Right when I double-clicked on the game, my computer shut off. I turned it back on, loaded up Windows, and changed my nVidia setting back to default (adaptive power management) and then the game worked fine again.

This ONLY happens when playing those specific games. If I never start up a game and just decide to browse the Internet and listen to music using iTunes, or even watch a Blu-Ray, nothing goes wrong - all is good. I've already went to my system's properties and unchecked 'Automatically Restart' when an error occurs. I haven't received any error messages when Windows boots back up and if I check out Event Viewer, it just states the event as "Event 41, Kernel-Power". My GPU fan idles at about 42 celsius, and averages at about 55 to 75 when playing a game. The highest I've ever seen it hit is about 79 celsius, and that's when I decided to play Crysis maxed out. I've recently downloaded SpeedFan, in order for me to check my CPU temps., and it states my core temp is at 54 celsius, even though all I'm doing is typing this post and .. that's it. My CPU usage right now, while the temp. is 54 celsius, is averaging between 1.7% and 7.6%.

I'm guessing this is either a GPU problem, CPU problem or a power supply problem. The power supply is brand new, I just bought it about 2 months ago. It's an OCZ GameXStream 850w power supply. The CPU is using the stock cooler and has never been overclocked and was also purchased brand new. The GPU, as some of you may know, has been factory-overclocked - though, my temps. seem to be doing alright in that area. Memtest86+ will not run for me anymore, as it just shuts itself off as well, about 10 seconds after the test starts. So I used the next .. best thing: Windows Memory Diagnostics Tool. I ran the test about 2 or 3 times on Extended and Default (Cache?), and Windows didn't prompt me about any errors being found. I randomly checked how the tests (on the RAM) were going, and everytime I checked, it said no errors have been found. With that information, I've eliminated RAM out of the equation.

What's the next step? What could be causing this?

Find out what your CPU temps are reaching from http://www.overclockersclub.com/reviews/phenom2_955/ :

The Phenom II 955 is manufactured using a 45nm process, has 758 million transistors and has a maximum TDP of 125 watts, using a voltage of 0.875 - 1.5v with a maximum operating temperature of 68C degrees.

Stock coolers are CRAP

Find out what your CPU temps are reaching from http://www.overclockersclub.com/reviews/phenom2_955/ :

The Phenom II 955 is manufactured using a 45nm process, has 758 million transistors and has a maximum TDP of 125 watts, using a voltage of 0.875 - 1.5v with a maximum operating temperature of 68C degrees.

Stock coolers are CRAP

Snotweasel530

amen to the above.. ah freaking men..

I don't understand the purpose of this link. I appreciate the reply, but what exactly am I supposed to be looking for in this link? Do you think the CPU could be the cause of my problem?

Alright, so SpeedFan WAS (and still is) reading my CPU temps correctly. I entered my BIOS and it said that my CPU was running at 51 Celsius, just like SpeedFan reported. When I decided to play Prototype, I watched the temps. by alt+tabbing every now and then out of the game, and the temp. were averaging about 67 degrees. Could this be hot enough to just shut off my PC completely?

Alright, so SpeedFan WAS (and still is) reading my CPU temps correctly. I entered my BIOS and it said that my CPU was running at 51 Celsius, just like SpeedFan reported. When I decided to play Prototype, I watched the temps. by alt+tabbing every now and then out of the game, and the temp. were averaging about 67 degrees. Could this be hot enough to just shut off my PC completely?

ATLReppa770

Instead of continuously bringing your cpu to the threashold, effectively shortening the life span of it, why don't you just get a cooler for it and find out.

[QUOTE="ATLReppa770"]

Alright, so SpeedFan WAS (and still is) reading my CPU temps correctly. I entered my BIOS and it said that my CPU was running at 51 Celsius, just like SpeedFan reported. When I decided to play Prototype, I watched the temps. by alt+tabbing every now and then out of the game, and the temp. were averaging about 67 degrees. Could this be hot enough to just shut off my PC completely?

Snotweasel530

Instead of continuously bringing your cpu to the threashold, effectively shortening the life span of it, why don't you just get a cooler for it and find out.

You might want to clean the interior of your computer, reapply the thermal compound and add some vents into your case. Phenom II shouldn't be that hot even with stock cooler.

You might want to clean the interior of your computer, reapply the thermal compound and add some vents into your case. Phenom II shouldn't be that hot even with stock cooler.

jernas

Could anyone just tell me if there is even a remotely chance that this could be a problem caused by my graphics card?ATLReppa770

Yes. There is a remote chance that this problem could be caused by your graphics card. pebkac is more likely however.

http://products.amd.com/en-us/DesktopCPUDetail.aspx?id=682. I've seen alot of people having this problem. The misconception is between the 2 latest revisions of the 955. You need to find out if your running a revision 2 [rb-c2] or a revision 3[rb-c3] the revision 2 had a higher temp threshold then the revision 3. Also in the meantime you might just wanna reinstall your drivers from scratch. uninstall from the control panel then reboot in safe mode and use driver cleaner from guru3d.com. Reboot. reinstall and see if the problem persists. Honestly your at redline on that processor you should not be getting that high. I also think you might not have seated the heatsink correctly.

http://products.amd.com/en-us/DesktopCPUDetail.aspx?id=682. I've seen alot of people having this problem. The misconception is between the 2 latest revisions of the 955. You need to find out if your running a revision 2 [rb-c2] or a revision 3[rb-c3] the revision 2 had a higher temp threshold then the revision 3. Also in the meantime you might just wanna reinstall your drivers from scratch. uninstall from the control panel then reboot in safe mode and use driver cleaner from guru3d.com. Reboot. reinstall and see if the problem persists. Honestly your at redline on that processor you should not be getting that high. I also think you might not have seated the heatsink correctly.

skinntech

To be honest man i can tell you right now that your proc is running way too high and its protocol on the motherboard to just cut power to save the cpu. Thats what I think it is. I dont know for certain so i suggested all that little stuff to do first just as a process of elimination. if you do all the little things and its still goin black then youll know for certain its the cpu overheating and shutting down

To be honest man i can tell you right now that your proc is running way too high and its protocol on the motherboard to just cut power to save the cpu. Thats what I think it is. I dont know for certain so i suggested all that little stuff to do first just as a process of elimination. if you do all the little things and its still goin black then youll know for certain its the cpu overheating and shutting down

skinntech

Alright, so I've just purchased the Hyper 212+ CPU cooler, received an approval for my GTX 275 and now, in 2 weeks, we'll (I'll) see if the problem has gone away for good.

Update:

Received my RMA'ed 275 back from eVGA two days ago. After I installed it, I formatted my main drive (Win 7), and then completely re-installed Windows 7 on it. So now, I have a fresh install of Windows with the latest chipset and graphics drivers, a new GTX 275 (card still had the original plastic over the stickers, so I'm assuming it's new) and a cooler that I need to install. But I went ahead and installed Crysis, SF IV, Bioshock 2 and some other games, and well .. my computer has shut off on me 3 times. Twice during Crysis (as soon as I double-clicked DX 9 mode) and once (5 minutes ago) on Bioshock 2. When Bioshock 2 did it, I didn't see my computer shut off, it seemed more of a restart vs. the Crysis-type shut down, where it seems someone is ripping out the power cable. I don't know what to do at this point. I don't think it's the CPU, but temps should be around 61 Celsius. Is my power supply supplying enough voltage or whatever my graphics card needs? I don't know how to read the stats on a power supply to make sure it supports enough for my card. Could it be my motherboard? What!?

NEEEDS MORE POWA

NEEEDS MORE POWA

NailedGR

[QUOTE="NailedGR"]

NEEEDS MORE POWA

ATLReppa770

It's probably the motherboard.

[QUOTE="ATLReppa770"][QUOTE="NailedGR"]

NEEEDS MORE POWA

NailedGR

It's probably the motherboard.

Are all the power cords plugged into everything? I know some psu/motherboards have their power connectors split up. Also maybe you aren't getting all the power you need to each device.

Also maybe your motherboard or psu are broken. Do you have any other place to test them out? Like another computer you can swap components with?

The only shut-off problems that I've ever had was with the power supply, and considering you have an OCZ just further solidifies that possibility. Do you have any other power supplies on hand that you could use?

The only shut-off problems that I've ever had was with the power supply, and considering you have an OCZ just further solidifies that possibility. Do you have any other power supplies on hand that you could use?

ravenguard90

[QUOTE="ravenguard90"]

The only shut-off problems that I've ever had was with the power supply, and considering you have an OCZ just further solidifies that possibility. Do you have any other power supplies on hand that you could use?

ATLReppa770

Put in your old crappy psu and see what happens, then post results.

[QUOTE="ATLReppa770"][QUOTE="ravenguard90"]

The only shut-off problems that I've ever had was with the power supply, and considering you have an OCZ just further solidifies that possibility. Do you have any other power supplies on hand that you could use?

NailedGR

Put in your old crappy psu and see what happens, then post results.

[QUOTE="NailedGR"]

[QUOTE="ATLReppa770"] I just recently upgraded to this power supply, from a 550w non-name brand supply that never gave me this problem. Is there something wrong with the OCZ supply itself, or did I just pick the wrong supply to use? I could always get it power swapped out, according to OCZ.ATLReppa770

Put in your old crappy psu and see what happens, then post results.

You may be learning this now, at this very moment, but you aren't supposed to make things pretty until after you are positive that everything works right.

[QUOTE="ATLReppa770"][QUOTE="NailedGR"]

Put in your old crappy psu and see what happens, then post results.

NailedGR

You may be learning this now, at this very moment, but you aren't supposed to make things pretty until after you are positive that everything works right.

After so many replies already, i'm sure someone mentioned overheating as I think is the main culprit. Sorry, I'm too lazy this afternoon to read through all the messages. :P I've had an older computer do the same when I stuffed a brand new GPU in it (generic desktop) and the poor thing didn't see what's coming... The computer knows to shut itself off when temperatures surpasses it's threshold temp.

Other reasons could be FAULTY PSU, or not enough power.

[QUOTE="NailedGR"]

[QUOTE="ATLReppa770"] I never said it was crappy. Actually, it makes my OCZ supply look crappy, as my old one could do the job. That's a pain in the ... to replace the power supply, since I have all the cables tied up and spent a long time on the cable management. Great .................d-rtyboy

You may be learning this now, at this very moment, but you aren't supposed to make things pretty until after you are positive that everything works right.

[QUOTE="d-rtyboy"][QUOTE="NailedGR"]

You may be learning this now, at this very moment, but you aren't supposed to make things pretty until after you are positive that everything works right.

ATLReppa770

You don't have your old psu sitting around?

you don't even have to put it in your machine, just unplug your new PSU and unplug it from all the other components. Then plug in your old psu (do not install it) and plug in all your components into it and use it to power your rig.

A few of my more problematic set ups, all I had installed into anything was the motherboard on the backplate. Everything else was free from the case. it makes trouble shooting easier that way.

[QUOTE="ATLReppa770"][QUOTE="d-rtyboy"]That's the reason I stopped bothering to make everything neat. The case is closed anyway, and I'm the only one who looks in it. Cable management for anything other than air flow and accessibility is a complete waste of time.NailedGR

You don't have your old psu sitting around?

you don't even have to put it in your machine, just unplug your new PSU and unplug it from all the other components. Then plug in your old psu (do not install it) and plug in all your components into it and use it to power your rig.

A few of my more problematic set ups, all I had installed into anything was the motherboard on the backplate. Everything else was free from the case. it makes trouble shooting easier that way.

[QUOTE="NailedGR"]

[QUOTE="ATLReppa770"] Yeah, but I guess what's really bothering me about this is, I had someone else do the cables for me and it cost about $30 to do that and re-arrange the place of my hard drives. I could have done it myself, but not as nice as them. The place is about 10+ minutes away, and they hold my computer for about 2 days. Now, if I go back, it'll be like I paid them to re-arrange my cables for nothing! So, I think I'm just gonna take out the motherboard, get it RMA'ed, take out the power supply, get that RMA'ed, and install the 212+ CPU cooler. That HAS to fix this problem! If not, my computer will be found scattered around my neighborhood ....... in parts. ATLReppa770

You don't have your old psu sitting around?

you don't even have to put it in your machine, just unplug your new PSU and unplug it from all the other components. Then plug in your old psu (do not install it) and plug in all your components into it and use it to power your rig.

A few of my more problematic set ups, all I had installed into anything was the motherboard on the backplate. Everything else was free from the case. it makes trouble shooting easier that way.

Leave all the cables where they are, just unplug them.

Using your old PSU will be a way to test if your OCZ is dead.

[QUOTE="ATLReppa770"][QUOTE="NailedGR"]

You don't have your old psu sitting around?

you don't even have to put it in your machine, just unplug your new PSU and unplug it from all the other components. Then plug in your old psu (do not install it) and plug in all your components into it and use it to power your rig.

A few of my more problematic set ups, all I had installed into anything was the motherboard on the backplate. Everything else was free from the case. it makes trouble shooting easier that way.

NailedGR

Leave all the cables where they are, just unplug them.

Using your old PSU will be a way to test if your OCZ is dead.

Yessssssssss ! It's definitely the power supply. I'm on the computer with the old (Okia 550w) supply now. I just came back from testing Crysis in DX 9 and DX 10, no issues at all. GTA IV benchmarks run no matter what settings I turn on; though, the results are disappointing, but I think that's due to poor programming. And most of all, I can finally get passed the 2nd level on Force Unleashed 2!

Tomorrow, I'll use the PowerSwap program and exchange this failed supply. You know, it's amazing how this non-existent name brand that I've had for about 5 years outperformed an OCZ branded GameXStream SLi certified supply with more than enough power to work without any issue.

Thanks to all who contributed and helped out.