[Help] PC restarting for no apparent reason.

[Cortexiphan]

Greetings.

I've been having a random reboot issue with my machine. There's no warning whatsoever and no BSOD, the PC simply shuts down and turns itself on again after 3-5 seconds.

My first thought was temperature, because the lockpins in my CPU fan are broken (it's properly inserted in the motherboard sockets and working, it's just not locked) but it has happened at 34º CPU temp on idle as well as at  60º while playing, so I'm ruling it out as the cause for the resets. There's no overheating detected by BIOS or AsusProbe and the fan's RPM seem to be fine.

The odd thing is, it usually resets when I pause the games to access menus or loading screens, meaning whenever it stops processing graphics. A practical example would be Skyrim, I can roam the open world for hours without a single reset but if I head to town and start opening doors or pressing quicksave too often it appears I have 2/3 chance of rebooting my system. Other games, like Max Payne 3 for instance, I finished in a single setting of well over 6 hours playing without a single reboot. I've already re-installed my graphic drivers to no avail.

I'm thinking it's a power issue. I've been told my PSU is rather crappy (Hantol ATX v2.3 800W) however my graphic card isn't known for being too power demanding. I have another PSU backed up but it's even worse I believe (Eurotech 580W).

My question is, looking at my specs, can I at least switch the PSU's so I can run a couple of tests and see if the reboots still happen? Could a 800W PSU simply not be enough for my gaming hence the reboots?

My voltages are as follow:

+3.3V: 3.34V
+5V: 5.02V
+12V: 12.15V

Specs:

Windows 7 64bits Enterprise SP1
Intel Core i5 650 @ 3.20Ghz
Asus Motherboard
8.00Gb RAM
AMD Ati Radeon HD 5700
Samsung HD103SI ATA Device 1TB

Windows Event Log doesn't warn me of anything irregular happening other than the Kernel-Power failure when the reboots happen.

Anyone have some insight on what could be causing this? I'm not very tech savvy but I will replace whatever needed if just pointed in the right direction. I'll be happy to provide screenshots or any other information required.

Thanks in advance!

[DaveLembke]

I'd run MEMTEST86 on this system just to make sure that its not a GPF that is causing the reboots. More on GPF's here: http://en.wikipedia.org/wiki/General_protection_fault

GPF could be caused my the RAM with an address that is flawed. Memtest86 will detect any flawed addresses in memory and if found it may be as easy as identifying a single troublesome RAM stick or replacing RAM stick.

Check all capacitors on the motherboard for any that are leaking as well around the CPU as for I have seen capacitors cause too much ripple when electrolyte is depleted and cause for system reboots. Any capacitors with swollen tops or leaking electrolyte would be a sign of this although in very rare occasions electrolyte can dry up and not show physical damage to the can body of the cap.

Do you get a BSOD flash right before the reboot or does it always go to black screen and reboot.

If its a black screen reboot without a quick BSOD before reboot it could be the power supply, however your voltages stated look healthy, and as long as the voltages are maintained and not fluctuating the power supply should be healthy from info you shared.

[Cortexiphan]

I'd run MEMTEST86 on this system just to make sure that its not a GPF that is causing the reboots. More on GPF's here: http://en.wikipedia.org/wiki/General_protection_fault

GPF could be caused my the RAM with an address that is flawed. Memtest86 will detect any flawed addresses in memory and if found it may be as easy as identifying a single troublesome RAM stick or replacing RAM stick.

Check all capacitors on the motherboard for any that are leaking as well around the CPU as for I have seen capacitors cause too much ripple when electrolyte is depleted and cause for system reboots. Any capacitors with swollen tops or leaking electrolyte would be a sign of this although in very rare occasions electrolyte can dry up and not show physical damage to the can body of the cap.

Do you get a BSOD flash right before the reboot or does it always go to black screen and reboot.

If its a black screen reboot without a quick BSOD before reboot it could be the power supply, however your voltages stated look healthy, and as long as the voltages are maintained and not fluctuating the power supply should be healthy from info you shared.

Thank you for the swift reply.

I've just ran Memtest86 and it passed all tests without any errors found. I also ran a burn in test on the CPU, GPU and Memory test without any errors or resets as well, while monitoring the voltages that didn't suffer any significant change.

There is no BSOD, it just plain goes black screen and reboots after 3-5 seconds. I've swapped the components to a different cabinet I had with less static a while ago but the problem remains.

Do you reckon a 580W PSU will do the trick? At least to try to reproduce the error and see if it's what's at fault.

[patio]

I think the CPU holdowns are still the culprit...

[Cortexiphan]

I think the CPU holdowns are still the culprit...

Could you elaborate please?

I'm seeing it rise to 65º when playing which I'm guessing it's hot. There's no dust in the cabinet and the thermal paste is not dry but yes, the cooler isn't strapped on as it should, it's simply inserted in the sockets, which I'm guessing could cause some warm up but not these random reboots when the CPU's temp isn't even above 40º.

Either ways, I'm getting it replaced, I'm looking at these Artic ones and they seem way better than the stock intel I have.

[patio]

Just a quick question...where are your temp readings coming from ? ?
If the Fan/Heatsink isn't properly seated it may affect how they are being reported...just sayin.

[Cortexiphan]

Just a quick question...where are your temp readings coming from ? ?
If the Fan/Heatsink isn't properly seated it may affect how they are being reported...just sayin.

The BIOS and AsusProbe, they're reporting very similar values.

[Geek-9pm]

Some information. A motherboard may have one or more probes built-in. But the CPU itself has a temperature probe independent of the heat sink and motherboard.

All major processor manufacturers have implemented a "DTS" (Digital Thermal Sensor) in their products. The DTS provides more accurate and higher resolution temperature readings than conventional onboard thermal sensors.

Find a utility that reads directly from the CPU.
Key phrase: CPU core temperature
Also:
Tom;s Hardware - CORE temp.

[Cortexiphan]

Some information. A motherboard may have one or more probes built-in. But the CPU itself has a temperature probe independent of the heat sink and motherboard.Find a utility that reads directly from the CPU.
Key phrase: CPU core temperature
Also:
Tom;s Hardware - CORE temp.

Yeah well funnily enough the CORE temps showing are even lower than BIOS and AsusProbe. Everyone's focusing on the *censored* temperature being the cause when I've had resets right after turning on the computer and simply browsing in Windows. I've ruled it out, period.

What I'd love for someone to answer my ORIGINAL question, which was would a 580W PSU be replacement enough for my 800W at least for testing purposes, that's what I'd like to know really.