GPU load fluctuates

[sgt_bombulous]

Hello,

I've noticed that the GPU load is fluctuating between 75 & 95%, rapidly, in sort of a jagged sawtooth shape. I recently have have had a mishap with overclocking my PCI frequency (not positive it if took or not, since it never made it to boot with those settings), but I'm just curious if anyone else has this or not. Is it possible my graphics card or motherboard has something wrong with it? I do remember it hovering closer to 99% before, but I think I've seen it do this as well.

Thanks!

[PantherX]

will it actually depends on the WU that you have received, if i am not mistake, core FahCore_14 will behave like this. I too have encountered WUs where the GPU usage fluctuates between 65% - 99%.

Regarding your overclocking attempt, if i am not mistaken, if your PC won't boot then the motherboard will automatically revert to "safe settings" (it happened to me a couple of times when i was overclocking my CPU).

[sgt_bombulous]

will it actually depends on the WU that you have received, if i am not mistake, core FahCore_14 will behave like this. I too have encountered WUs where the GPU usage fluctuates between 65% - 99%.

Regarding your overclocking attempt, if i am not mistaken, if your PC won't boot then the motherboard will automatically revert to "safe settings" (it happened to me a couple of times when i was overclocking my CPU).

Actually it was a bit more hair-raising than that... After I received the "overclocking failed" message I had to reset the board by removing the CMOS battery because I couldn't get into the bios at all. I think I'll be a little more careful next time

[PantherX]

Well from what i have read on the net, by removing CMOS battery, all BIOS settings are cleared so everything is back to factory setting. I believe that you are in the clear and if you really want to be sure, you can download Furmark (http://www.geeks3d.com/20100202/furmark ... available/) and run it so see if your GPU can take the load or not. If you noticed and artifacts or if the system hangs and crashes then you might have a problem. Don't worry if your GPU reaches high temp as Furmark is designed for this. On my system 9600 GT 1 GB 100% @ 81C Furmark but while using F@H it would be at 61C. If you don't want to do this (you can damage your hardware if used incorrectly!!!) you can check your F@H log and if you find "NANs detceted on GPU" or "Unstable_Machine" or other errors then your GPU is facing some problems

[sgt_bombulous]

Well from what i have read on the net, by removing CMOS battery, all BIOS settings are cleared so everything is back to factory setting. I believe that you are in the clear and if you really want to be sure, you can download Furmark (http://www.geeks3d.com/20100202/furmark ... available/) and run it so see if your GPU can take the load or not. If you noticed and artifacts or if the system hangs and crashes then you might have a problem. Don't worry if your GPU reaches high temp as Furmark is designed for this. On my system 9600 GT 1 GB 100% @ 81C Furmark but while using F@H it would be at 61C. If you don't want to do this (you can damage your hardware if used incorrectly!!!) you can check your F@H log and if you find "NANs detceted on GPU" or "Unstable_Machine" or other errors then your GPU is facing some problems

FAHMON will display an active log right? If so I'll check that out and see if anything looks fishy. I use a GTX260 BTW, seems to hover around 71 C with this silly fluctuating, but when it runs hard at over 95% it stays around 79C. Any software that gives me an opportunity to damage my hardware is something I should stay away from with the way my luck has been lately. Thank god it's an eVGA with a lifetime warranty... Thanks for your help!

[PantherX]

No problem, wish you happy folding!

[sgt_bombulous]

No problem, wish you happy folding!

You were 100% right! I'm running a core 11 project now and it's pegged at 99%. I guess loading that wacky vore 14 one right after my OCing mishap was just Murphy's law!