Coma

Maybe other loads DON'T get it to run at 100%. Even if the FPS is low, it doesn't mean it's running at full force to get those few frames. Bad programming sometimes also means unused resources.

And for the record, it was not excessively hot or anything.

__________________
When asking for help: state the goal, not the step.

Clockadile Dundee

I am in agreement with you guys. I guess I am just wanting something a little more technical to read, with regards to this.

I think the fact that heat was not a problem, is even more interesting; what do you think the chain of events was here?

Perhaps the fine distinction, however, is that I define the word "unstable", to include internal events of electromigration, which are totally undetectable, and result in no perceivable effects, but cause accelerated degradation.
If your device's processor is degrading faster than what the internal manufacturer docs say it should, something is up right (not that we would ever have access to that info)? Could something not be improved, on the user end, to extend the life of these devices?
For example, my GTX 295 never goes over 33c, won't it last longer, just because of that, probably? (Totally an excuse to just mention this fact to someone on here, for the first time, since I overclocked the GTX 295, mobo is rma right now, but I want to say 743 / 1600 / 1100.


I will stand by my statement, that it is not folding, per se. Lets just grant you for a moment that nothing else but folding loads a given gpu to 100%. If some application did exist, which loaded your gpu to 100%, for the same amount of time, anything with the same demands on the gpu, how will it not have the same effect? I would agree with the statement : "To degrade your gpu/cpu as fast as possible, run it at 100%, all the time, too hot, too much volts, and an unstable OC that still functions mostly, so it can actually rack up those hours.

I am just having fun here, we are basically all in agreement, as far as I am concerned.

__________________

Clockadile Dundee

Some cards are just unstable at stock settings, brand new in box.

By unstable, I am not necessarily referring to a crash...are not elevated, but undetectable levels of electromigration instability? Yes, because the electrons are behaving erroneously, causing long-term gradual degradation...that is unstable. (right?)

What are all these gpu stability testing programs? I am not very educated about OC'ing gpus.

__________________

Benladesh

I've been folding on my card for 4-5 months 24/7 and I haven't had any problems as of now. You might have gotten a bad card. I also have a friend where all his cards fry when he OCs and folds. We think it's the mobo being faulty... he's rma'd 2-3 cards now...

Clockadile Dundee

Right, it is hard to prove a negative, in this case, that it was NOT something other than folding.

Unless something is abnormal, which it often is, the normal process of degradation should take many years, to have a noticeable effect.

__________________

corky dorkelson

My 9800GTX+ slowly became unstable and I had to back off of the clocks to keep it stable, but that is about all the problems I have had. I am sure gaming would have done the same thing to the card, although it may have taken a while longer.

I am firm in my belief that any damage that occurs from folding would have also happened from gaming. It may happen a bit sooner from folding(if folding 24/7), but that just means that if there IS a problem, it will be exposed sooner and (hopefully) within the warranty period. I still say f@h is a great stress-test.

__________________

markt

I have not had even one card or cpu break during the entire time I've been folding. I used to overclock my 8800gs's and gtx+'s but now everything is stock speeds so I don't worry about eue's and such. my 9600 gt's are actually underclocked.

__________________
Worldwide f@h Rank; 55 of 1,303,174
WU 60575 (certificate)

T D

My HD4870 died during folding but would have died anyway since VRM temps were up. RMA ftw.

__________________
Last.fm

SniperXX

From my testing folding seems a sure fire way to test overclocks. I've had some be fine with other apps but the load FAH puts on the card makes it unstable.\\

As for me, no problems on any of my gpus. I do run the fans @ 100% to keep the chips as cool as possible and I monitor my temps.

My 8800GTX has folded for almost a year straight with zero problems. The shaders are even bumped up 200MHz. My 8800GT cars are fine @ stock speeds due to the stock heatsink sucking with overclocks. Heck I even fold on my 285s.

Those with problems, was your fan speed at 100%?

__________________
Want to get ~25,000PPD folding on your i7 running a Virtual Machine?

EVGA 680i FOR SALE

i7 920 @ 4.2Ghz
HT off for now, still testing.

HEATWARE

O o
/¯/___________________________ _\
| BBBBBAHHHHHHHHHHHHHHHHH
\_\¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯ ¯/

prosser13

You're completely correct. It is not folding in particular - however, folding will cause it, whether the overclock is stable or not. Simply overclocking will cause electromigration to occur faster (doubling the clock speed doubles the TDP, while doubling the voltage quadruples the TDP), but it will still occur over time

__________________