Sudden Shutdown : My half-informed diagnosis for your edification/amusement

[docbadwrench]

Tom's Hardware is my EMT. I never visit except when something bad has happened. I hope that's okay. Life has taken me away from the ready-knowledge of system builders and troubleshooters.

The Problem

My computer has taken to shutting down during games, especially when things get intense. My system is connected to a heavy PSU that has been happily absorbing my minor Seattle-area brownouts. Click…click. But with this new error, a loud, persistent beep irritatingly coughs its throat. I have to shut down the PSU. Turn it back on. Then boot the computer.

For the past week, I've been google-flailing at the symptoms, wondering if it was my PSU or my video card. Then I realized: There are way smarter people out there who have the benefit of not being me. Hi People! Here are things!

Mobo: Gigabyte GA-MA770T-UD3P (AMD Phenom II X3/X4)
CPU: Athlon II X4 630 @ 2.8 Mhz
RAM: 4 GB (2GBx2) DDR3 (1333Mhz)
Video Card: NVIDIA GeForce GTX 465 (1GB)
PSU: OCZ600MXSP (ModXStream Pro)

Here are some more salient things I've observed. I'll just put them here and you can quietly judge me. If I've neglected a fact, it's because I dismissed it as not important enough (face it, you have other things to read).

Who Doesn't Love Lists?

1. There have been zero cases of a BSOD accompanying a sudden shutdown.
2. I can't seem to track down the appropriate event code number, but there are no driver failures.
3. It has been somewhat hot lately (late summer in Seattle), but not nearly as hot as it was a few months ago and there was only one case of that happening, which I chalked up to heat.
4. My video card (specs below) was purchased and installed (new) not more than four months ago.
5. My PC is a homebuilt system which I've been updating and maintaining for… god, I don't know anymore. A long time. The mobo was spanking new and the shiniest thing out there.
6. My PSU has aged between 1.5 and 2 years.
7. The shrill PSU beep I mentioned at the start. It tells me that some damned thing just stopped drawing power after sucking a whole lot of it.

Possibly, Probably PSU?

These facts, together, indicate to me that this must be a PSU issue. I've excluded a video card error because I'd be waist-deep in BSOD's and there would be tons of driver failures. I thought it couldn't be a heat issue because, as I noted, it's been hotter. I have fans that run just fine, too, though poor ventilation in my area isn't doing me any favors.

My PSU is an OCZ. Durable stuff, by my understanding. However, I'm playing games that tax my system, and have done the occasional video editing. I'm wondering if the thing has just hit an early end of life. It's a 600W. However, when I plug in my stats into two watt-load calculators, I get the following recommended wattage:

Extreme Outervision: 459 watt
NewEgg: 531 watt

That's quite a discrepancy, and all the more confusing since I was able to account for more 'stuff' in the EO calculator. Could I just have a bad PSU? Are the calcs more than detritus?

Feel free to grade me. I'm much more interested in taking the right action than being right. I don't have wads of cash to throw at the problem, so I must purchase with precision.

Many thanks, in advance, for your help.

 

[docbadwrench]

Windows is clean of gunk. I now have a liquid cooling apparatus (Corsair H80i). I have confirmed the problem.

My CPU was running a little hot, but not enough to explain the stupid amount of heat I was reading. Once I installed the H80i stuff, I had a PC that was running very smooth and very cool, but this was with my old Nvidia 9600 GT card. When I swapped it with my - uh, new - videocard, it was like dropping a flaming brick into my PC.

Every damned readout was too high. After being on about a minute, mind you, I couldn't even remove the damned thing. I had to wait for it to cool down. So, I have learned that my purchase from five months ago was a straight up lemon.

At the very least, I'm happy to know that I shouldn't need to worry about CPU heat for a while. It's been hovering about 30 deg C and barely bumps upward even when under a pretty big load.

Thanks to all of you for your help in bouncing ideas around. It surely helped me uncover the facts that filled in the picture.

 

[scoobydenon]

Windows is clean of gunk. I now have a liquid cooling apparatus (Corsair H80i). I have confirmed the problem.

My CPU was running a little hot, but not enough to explain the stupid amount of heat I was reading. Once I installed the H80i stuff, I had a PC that was running very smooth and very cool, but this was with my old Nvidia 9600 GT card. When I swapped it with my - uh, new - videocard, it was like dropping a flaming brick into my PC.

Every damned readout was too high. After being on about a minute, mind you, I couldn't even remove the damned thing. I had to wait for it to cool down. So, I have learned that my purchase from five months ago was a straight up lemon.

At the very least, I'm happy to know that I shouldn't need to worry about CPU heat for a while. It's been hovering about 30 deg C and barely bumps upward even when under a pretty big load.

Thanks to all of you for your help in bouncing ideas around. It surely helped me uncover the facts that filled in the picture.

Great to hear you have found a solution. Now remember to keep it clean. Run those tools weekly, enjoy.