"Burning In" Myth or Truth

[Same]

I've heard people say that, if you run you max your overclock, run it for afew days on that and "burn it in," you can then OC higher. I have never experienced this and it sounds like abunch of crap. But what do you guys think?

 

[Ben333]

Well supposedly it helped you gain stability but many people doubt it works. Kinda like how they said grounding your heat sink to reduce EMI would give better overclocks.

 

[juane414]

I used to think it was crap, but I'm starting to believe it. When I first got my Opteron 1212 it walled at 300x10. I could accomplish this OC with stock volts (1.35) but I couldn't even get to 305x10 with 1.55 volts. It seemed that I would never get past 3Ghz. After running the computer at 3Ghz for a few months, including stress testing and folding, I decided to give it another shot last week. To my amazement, Vista booted at 310, 315, and 320 HTT requiring 1.475v at 320x10. It wasn't stable, but it posted and booted. I'm still not convinced that Vista hasn't contributed to this, because when I first installed Vista I couldn't get my system stable at stock. I'm not sure if Vista has a hardware "learning curve" or if I was successful at "burning in" my system. I'm certainly going to give burning in a chance, and I think I'll run my system at 305 or 310 for a while to see what happens.

 

[Ben333]

Thinking back, my Opteron 175 had a similar issue. At 1.5V it would only do 2.6 GHz and after a couple months of folding 24/7 it went with EASE to 2.8 GHz.

 

[apintojr]

well usually when i knew CPU i always stress test for a couple hours to burn it in before i atempt to OC

 

[Same]

Hmm interesting. Its just strange that that would happen I can't understand why. You use a cpu for a month or two and then it can go higher.

 

[Diddyu]

I usually get a few degrees less after applying paste, OC'ing, and running coredamage for 24 hours, but supposedly it continues setting in for a long time. I think it has something to do with the paste expanding or flowing into all the microscopic grooves, and increasing the surface area contacted.

 

[Same]

I usually get a few degrees less after applying paste, OC'ing, and running coredamage for 24 hours, but supposedly it continues setting in for a long time. I think it has something to do with the paste expanding or flowing into all the microscopic grooves, and increasing the surface area contacted.

Yeah, but unless tempature was holding you back initially it shouldn't have an effect on an FSB wall.

 

[Diddyu]

What I meant is that the next time I started the computer after running it hot for a day and started it after letting it cool for a night, my temps were about 5-6c less than the day previous, and the ambient was near the same (0.2 higher I think).

 

[Sarsbaby]

Interesting, I would think this to be the opposite. Anyone with actual proof?

 

[Diddyu]

Interesting, I would think this to be the opposite. Anyone with actual proof?

About the temps? I was planning on putting a bigger fan on my heatsink over the weekend, Ill record the temps before and after when I do that since I need to remove the HS to put the fan on.

 

[apintojr]

about temps its true, AS5 takes several amount of hours to settle so the temps do get lower with time. i beleive thats on the AS5 site.

edit: this is off the AS5 site "During the CPU's initial use, the compound thins out to enhance the filling of the microscopic valleys and ensure the best physical contact between the heatsink and the CPU core. Then the compound thickens slightly over the next 50 to 200 hours of use to its final consistency designed for long-term stability."

 

[Sarsbaby]

Sorry, I meant about the burn-in process and it getting better speeds after months of use. I know the TIM takes time to settle in and do it's job. I shoulda specified, sorry....

 

[AlabamaCajun]

I would attribute the stability to the BIOS adjusting itself after several reboot cycles. In most cases I've seen systems degrade as caps and other parts age noise builds up. It may be possible that a little improvement in the VRegs occurs after several hours of running where they improve in performance but that's theory. As for the CPU getting better doubtful at best scary at worst.

 

[Albuquerque]

Overclocking isn't specific to your CPU; you also must consider the motherboard and all it's little analog chemical components (read: capacitors) that can drastically affect overclocking potential.