psu or gpu?

[ganon11000]

When I'm gaming I get a Kernel Power Loss (the power cuts) and it's fine afterwards but it's very annoying.
Am I overloading the PSU?

Specs:
Seasonic s12II 430W
AMD Sempron 145 overclocked and unlocked to AMD Athlon II X2 4500e 3.5GHz
ATI Radeon HD 3870
ASRock N68-VS3 FX
4GB DDR3 RAM 1333MHz
1TB Seagate HDD
1x DVD Burner
1X 120MM case fan
1X PCI fan controller and a 80MM fan
Netgear WIFI USB Stick
Apple Pro Keyboard
Logitech USB mouse

Thanks


Swapped PSU to my 750W PC Power and Cooling Silencer and all is good!

[AETAaAS]

Really doesn't look like it. From what I have seen, your proc is a 45W and the graphics are another 75W or so. Even if you overclock them both, as long as you don't have insane case mods or cooling, you are happily under your 430W supply.

It is interesting that you are finding it cuts with gaming. Maybe run CoreTemp and something like MSI Afterburner to keep an eye on your temps? Give your case a good clean out (especially the heatsinks) to get the air moving and make sure there isn't anything contacting which may be causing shorts.

[Tom94]

This type of problem can be caused by many things. It could be hardware related or software

I notice that your processor is overclocked. How long have you had it overclocked for? Overclocked components can cause this problem aswell.

Have you got any BSOD codes?

[ganon11000]

it's a new build so it's not dirt, don't know about shorts?

It's happened twice, once on DiRT Showdown and once on Age Of Empires 3.

tempretures are great with speed fan, I'll download core temp to test
I get 60~70Oc with FurMark

[ganon11000]

No BSOD Just a pure power cut!

It's passed 2 Hours of prime 95, 1 hour of LinX and 7+ Hours of endless mincoin mining

[Tom94]

Is your graphics card overclocked? Also have you checked the temperatures of it aswell?

Could also be conflicts with drivers. Has this just started happening?

[ganon11000]

Is your graphics card overclocked? Also have you checked the temperatures of it aswell?

Could also be conflicts with drivers. Has this just started happening?

temps 60~70oC FurMark

Its happened 2 times in the 2~4 weeks of having this GPU

If it's drivers surely it would BSOD not fully cut? (it's like the plug has been pulled out!)

And the GPU is all stock

[Tom94]

temps 60~70oC FurMark

Its happened 2 times in the 2~4 weeks of having this GPU

If it's drivers surely it would BSOD not fully cut? (it's like the plug has been pulled out!)

And the GPU is all stock

Sorry i should have asked. When you say the power cuts, does it return to the desktop or does it just restart?

[ganon11000]

Sorry i should have asked. When you say the power cuts, does it return to the desktop or does it just restart?

It goes out like a light. No reboots etc... It does however start when pressing the power button like nothing happened?

[Tom94]

I highly doubt it is a software problem then. As you said, there would be a bsod but the fact it just shuts off is strange.

It could be that your psu is overheating or is failing as i doubt you will be using more then 430 watts

Is there another psu you can use to test your system or another machine you can test your current psu with?

[mikeo01]

I've had this issue before but there can be many causes.

Note your power supply seems to have more than 1 rail, make sure you are not overloading a rail's maximum amps somewhere.

Check your northbridge temperatures as this is directly related to hardware, a northbridge fault/overheat can cause random issues.

May be worth checking your RAM out also.


Other than that if there hasn't been any other time when it shuts down randomly then it sounds like it may be something to do with the GPU, whether this be drivers or an overload on PSU rail somewhere.

Double check your physical PCI-E connectors are not causing the issue, for instance use the provided molex to 6-pin to see if a simply faulty cable is causing power loss to the card

EDIT: Also your power supply is bronze rated, so I wouldn't let the maximum power go above 344w as to not overload it, but you don't seem to be near that HD 3870 is quite a hungry card it draws quite some amps

[ganon11000]

I've had this issue before but there can be many causes.

Note your power supply seems to have more than 1 rail, make sure you are not overloading a rail's maximum amps somewhere.

Check your northbridge temperatures as this is directly related to hardware, a northbridge fault/overheat can cause random issues.

May be worth checking your RAM out also.


Other than that if there hasn't been any other time when it shuts down randomly then it sounds like it may be something to do with the GPU, whether this be drivers or an overload on PSU rail somewhere.

Double check your physical PCI-E connectors are not causing the issue, for instance use the provided molex to 6-pin to see if a simply faulty cable is causing power loss to the card

EDIT: Also your power supply is bronze rated, so I wouldn't let the maximum power go above 344w as to not overload it, but you don't seem to be near that HD 3870 is quite a hungry card it draws quite some amps

I was thinking that but I couldn't find what rail the PCI-E connectors are on? and if a molex to PCI-E connector is needed?

BTW all temps are as desired, no over heats

[Pob255]

does sound like psu off hand, that psu should be able to cope, probably lost power over time.

try running a gpu stress test at the same time as running a cpu stress test, eg prime95 & furmark
And keep an eye on cpu and gpu temperatures as well as voltages

[mikeo01]

You might be able to tell from the wire colouring to identify what rails they are on sometimes they are labelled.

I assume the PCI-Es will be split on both rails, but can be still worth checking out with a molex-6-pin adapter. The PCI-E pins may not be making a good connection.

However you did say this is a new build, so it is very unlikely the PSU is failing to provide on one of the rails.

Instant shutdown when running games does sound like an overload, but you could try completely removing the graphics drivers, taking out the GPU then reinstalling it with previous drivers just to be sure

[Tom94]

Are you sure no components are causing power shorts?

[Lee`]

I'm having the same problem with a new graphics card. Using a molex to PCIe for the second PCIe needed for the card, using two seperate lines for the molex, with nothing else on them, and the power is cutting when I start games up, presumably when the GPU is drawing more power. I checked my event viewer and it showed;

Critical 26/05/2013 09:05:06 Kernel-Power 41 (63)

Might be worth checking if event viewer has anything to say about your black outs.