[anndruu12]

Good evening folks. I have run into a problem with my system and am looking for any help and advice that you guys can give. My system runs fine during normal, everyday use. However, when I start playing a game I usually play, Final Fantasy XIV, it has started shutting the system down after a few minutes of play. It never used to do this, and nothing has been changed to my knowledge. The first time it shut down on me, when I tried to restart the system, it kept shutting itself down before the system even reached the windows screen. I let it sit off for a while, and then it booted fine. Today, when it would shut down on me during the game, it has let me turn the computer back on fine. At first, I was led to believe it may be an overheating issue, but now I am not so sure. In BIOS, the cpu temp seems to be running normal, even after a shutdown. Any ideas or suggestions?

Systems Specs:

-AMD Phenom II X4 955 3.2 GHz 125W Processor

-COOLERMASTER Hyper TX3 92mm CPU Cooler

-Rosewill Green Series 630W PSU

-GIGABYTE GV-R587UD-1GB Radeon HD 5870

-GIGABYTE GA-770TA-UD3 AM3 AMD 770 Mobo

-4GB Corsair memory

 

[Markgg88]

Rosewill PSU's are low quality and possibly the sources of your crashes, I'd replace that with a good quality unit such as Corsair-SeaSonic or XFX.

The GPU maybe overheating, but before we check that, please boot into the bios and check the voltage readings. You'll find them under "system health status" where the CPU temp readings are located.

 

[anndruu12]

I am assuming this is what you need:

VCore: 1.312V
DDR3 1.5V: 1.504V
+3.3V: 3.328V
+12V: 12.112V

 

[Markgg88]

Looking ok, but I think they may not remain so under load.

Let's see the GPU temps. Download HW Monitor.

Take note of the temps while idle. Then play a game for a few minutes, alt-tab out and take a screenshot of HW Monitor. Hopefully you'll get us something before it crashes. HW Monitor will show the voltage readings while the system is under stress as well.

 

[anndruu12]

Alright, first of all a couple notes. Last night I went into BIOS and into PC Health Status and for the option 'CPU Warning Temperature', I changed that from disabled to either 70C/158F or 80C/176F, I can't remember. I also set the SYSTEM FAN1 Fail Warning option from disabled to enabled. Then I knocked my game settings down a bit and everything worked fine. I didn't get booted and was able to play for several hours.

Today, I left the BIOS options as I changed them to above, but turned the settings back up. Again, I was able to play fine. Then I changed the BIOS settings back to disabled, and it crashed on me. Here are a couple HWMonitor logs:

Idle:

In-game:

 

[kalibcrone]

just wanna put this thought out there. the psu is in my opinion the most important base part of a computer. at first i didnt want to spend a lot cuz i thought it didnt really matter, but ive had to deal with the consequences. if you want a great system, you NEED an equally great PSU. kinda sucks but once you get a quality one, you are set to make all the customizations you want to your computer from there without any worries

 

[Markgg88]

That part about the bios is weird, I'm not sure how to explain it.

Now, I'm not sure if the voltage readings are displayed correctly, but if they are then your PSU -12V, +5V and -5V are very unstable. Again I'm not sure if those are correct cause they seem a little extreme, but it wouldn't surprise me with a roeswill PSU.

If you look below the voltage readings, we have the motherboard temp readings for different parts such as the northbridge and southbridge sections. One of them is at 84C which is no good.

Your ATI card is extremely hot at 64C while simply IDLE. That would be a perfect in-game temp but not while idle. It shouldn't be more than 40C.

Now in-game it got to 82C which although high, shouldn't be enough to crash it. So I have a question: Did you continue to play after taking the screenshot to the point where the system crashed?

Also how long did you play before taking the shot?

Your system needs a good cleanup, I suspect there is a lot of dust in there. I'd get a can of compressed air and blow the dust out, specifically the GPU.

And I'd also replace the Rosewill PSU with any of the brands I suggested in my first post.

 

[anndruu12]

I will dust it up and bit and see if the temps drop at all.

I played approx. 10 minutes max before taking those screenshots. Wasn't a long time at all, but it doesn't usually need to run very long to shut itself down.

 

[Markgg88]

While 80C is a little hot, it's not enough to cause a shutdown, so that's why I was asking whether you kept playing after taking the screenshot or did you close the game right afterwards. Although the GPU is likely to freeze if it reaches a critical temp instead of shutting down, so I'm suspecting the PSU.

 

[anndruu12]

I played for a few minutes after the screenshot, not too long. I'll look at upgrading the PSU. Assuming it is the PSU, is this something that could have developed over time, because I didn't used to have the problem.

 

[Markgg88]

Yes. Any PC component will degrade over time. For Rosewill PSU's, you can expect them to fail sooner rather than later, they're not good quality. So even if it's not the source of the issue, I'd replace it regardless. But for now, give the inside of the case a good cleanup and see if it helps the temps.

Also, when PSU's fail, they tend to take other components with them, so keep that in mind.

When and if you decide to replace the PSU, here are a few I would recommend.

Newegg.com - CORSAIR Enthusiast Series TX650 V2 650W ATX12V v2.31/ EPS12V v2.92 80 PLUS BRONZE Certified Active PFC High Performance Power Supply

Newegg.com - XFX Core Edition PRO650W (P1-650S-CAB9) 650W ATX12V v2.2 & ESP12V v2.91 SLI Certified CrossFire Ready 80 PLUS BRONZE Certified Active PFC Power Supply

Newegg.com - SeaSonic X Series X650 Gold ((SS-650KM Active PFC F3)) 650W ATX12V V2.3/EPS 12V V2.91 SLI Ready CrossFire Ready 80 PLUS GOLD Certified Modular Active PFC Power Supply

All three of these are made by SeaSonic, who are the best. You can't go wrong with any of these.

 

[anndruu12]

I greatly appreciate all your help.

 

[gcavan]

Ignore the -5v and -12v readings. ATX power supplies do not have a -5 volt output and I've yet to see a software monitor which will consistently show -12v correctly. HW Monitor, as with all software monitors takes its information from the motherboard sensors not the power supply itself and will often get false readings.

 

[A1tecice]

Yes, i have never heard of your make of psu, and most low quality PSU company's will only test that 650W at idle so there not properly testing how it can preform under load eg, gaming.

I suspect if your not overheating it is your power supply failing to power your rig,I suggest you take Markgg88's advice and get a power supply from a company that he suggested doing this will most likely solve your problems.

 

[Markgg88]

Ignore the -5v and -12v readings. ATX power supplies do not have a -5 volt output and I've yet to see a software monitor which will consistently show -12v correctly. HW Monitor, as with all software monitors takes its information from the motherboard sensors not the power supply itself and will often get false readings.

Thanks for that, I did think they were too extreme to be true, but thought I'd mention them regardless.