the old "burn-in" question...i know it doesnt help OC, but is there a point to doing it?

[Mav]

the old "burn-in" question...i know it doesnt help OC, but is there a point to doing it?

I know the "burn-in" does not help with OCing (or at least that seems to be the more truthful opinion)...but, is there a point to doing it? Some say that the burn-in is good for testing stability...

If so, how long should I burn-in? And which program is best? Also...when burning in, do you leave the computer running for the entire period (minus troubleshooting time), or just several sessions of "burns"?

And...on the point of stability- If im going to play games, surf the net, listen to mp3's, do school work (word, excel, powerpoint, etc etc)...is an extended burn period of like a week really necessary??

thanks,
-eric

 

[Badger]

'Burn-in' is complete Bull-sh*t, just forget it.

 

[Phil]

Dismissing burn-ins as complete bs is just as bad as the people that swear by it and compare it with running a new car in or breaking in a new pair of shoes (excuse me whilst I just go and laugh at this incredibally scientific theory ahhhhhh hahahaha )
To be honest there isn't proof it does anything and there isn't proof it doesn't, on one hand it's comparable to not walking under a ladded, on on pavement cracks, or it could be looked at in a similar way the the wierd science of socks going missing in the washing machine because of the spacial vortex created by the swirling massof water that sends them to never never land, at the moment people don't beleive that but I'll show em!

 

[dozier768]

"break it in like your gonna drive it" lol burn in is not bullshit and it can help you get higher clock speeds or so the majority claims. and i would tend to agree

 

[ebola]

when i got my celly 566 at first, i had to run it a 1.75v to have stability at 850. now at 850 i can run it at 1.65v with no problems. at 1.85v i'm doing 954. should work at 1.8v in another week or so. soon i'll get 1054. just wait.

burning in is the key to lowering voltage.

 

[$o_dope]

kinda of the same topic, but I was wondering when you get a brand spanking new processor do you have to burn in that baby?

 

[jeff_harrison_344]

Once again, dont have to, but you should do it anyways, I do atleast

 

[Kingslayer]

A lot of people confuse "burn-in" and "stability test". Testing for stablility for some odd reason uses the same process as burning in. Running all different kinds of programs that stress the components.

Now, here's the truth, WITH an explanation. Burning in IS bs. It's a processor not a boat engine. If a burn in was necessary, then there would be a warning from the manufacturer, just like boat engines.

Why do chips work at less voltages as they get older. Look into the effects of the microchannels inside a processor eroding from the electrical current passing through them.

 

[Mav]

so perhaps I should run something like [email protected] Prime95 for a day or two b4 I OC, to make sure all my parts are in working order?

And then...as I OC and such...run those high load progs again to make sure my OC is good?

of-course...monitor temps...wouldnt wanna blow up my thunderchicken

 

[Kingslayer]

SETI@home is alright for a test. But it will only stress certain parts of the CPU. This is what I do to test for stability.

I run Prime95 through 1 complete test.
Run CPU bench, CPU multimedia bench, and memory bench in Sandra for 24 hours.
Run Quake loop for 24 hours.
Run 3DMark 2001 in a few loops.
Then play around with the computer doing your nomal every day things.

This will test your stability in just about every area, and that's what you want to do.

 

[Big Mike]

I'm of the opinion that burn-in for performance is pretty much baloney other than perhaps a bad heatsink compound job where the heating boils it out or what have you and lowers the temps perhaps...However, there is ONE compelling reason to burn in a chip, most ICs have a relatively long life span, however MOST that are going to fail will fail relatively quickly, almost all ICs that dont fail within the first week or so of use will not fail until very near their designed life span ends, blasting a brand new chip hard for a couple days tends to bring out any faults it could have left the factory with and makes it easier to argue a DOA/waranty claim.

 

[Mav]

oh? in that case I'll definitely run some heavy tests...getting replacements from online guys can be difficult...and i want as much on my side as possible

-eric

 

[Badger]

Well my PIII700 overclocks the same today at the same voltage as the day i dropped it in my PC almost a year ago.
Burn in is complete crap, if it worked the manufacturers would do it and sell the chip at a higher speed rating. If it was necassary you would get running in instructions like on a car.
I think some times people change there system settings, hardware or cooling and then start making statements like ....."when i got my celly 566 at first, i had to run it a 1.75v to have stability at 850. now at 850 i can run it at 1.65v with no problems"... and use this as evidence of 'burn-in'.

 

[ebola]

Badger, I guess your name suits your personality. I believe I am a bit more familiar with my machine than you are, so please do not make these assumptions.

 

[Badger]

ebola (Jun 25, 2001 05:47 p.m.):
Badger, I guess your name suits your personality. I believe I am a bit more familiar with my machine than you are, so please do not make these assumptions.

ebola,
I meant no disrespect; can you really say you have made absolutely no system changes since you got your CPU?

 

[Slake]

Badger (Jun 26, 2001 12:55 a.m.):

ebola (Jun 25, 2001 05:47 p.m.):
Badger, I guess your name suits your personality. I believe I am a bit more familiar with my machine than you are, so please do not make these assumptions.

ebola,
I meant no disrespect; can you really say you have made absolutely no system changes since you got your CPU?

CAUTION JOKE AHEAD!
Badger, don't mess with him; He's gouging his OWN eyes out....Just imagine what he'd do to you.. END JOKE

Burn in can't hurt and just may work.

 

[CpuWaterCool.com]

If one does not choose to overclock, burn in is not necessary. There will be no noticable benefit.

If you overclock, however, the more you run the CPU at the edge limit, the higher the edge limit will go over time.