CPU burn in, myth or reality?

[Budwise]

yup, from my experience, total myth. Scientifically and otherwise...

[benefit14snake]

and of course all CPUs are not same!
For my personal data gains for AMDs are 30-50 MHz OK this is nothing.....
The most important is Vcore decrease for the same MHz.
Anyway if someone follow my guides then he/she must read carefully the instructions and of course he/she should visit OcTech forum for latest updates etc....
Moreover I stack Gorki's thread see it here

MrSeanKon is my nick name!

Good luck!

Can you give any reason as to why this would help lower voltage? I want some scientific reasonings, not the "it just does." crap. If you can provide me with a scientific reasoning, I can find some one to confirm this reasoning. Only then will I be convinced, although when I do get this opty 165, I will undoubtedly be conducting my own tests, and will post short after my results. If this in fact DOES work, then I will be super stoked and will bow down to the almighty Mr Sean Kon ... Hey, I really hope it works actualy, but untill I have tested it my self, and gotten scientific reasoning to back up my results, I will still have many doubts about it.

The main reason that I believe it is a myth is, the person I see that supports this thereoy most actively, Mr Sean Kon, has never ONCE (from all that I have read), given any reasoning behind this proccess, other then "Follow my instructions carefully and it will work; Not all CPU's will have the same results; and CPU's are like a spring, you have to stretch them."

[MrSeanKon]

and test for reliability

e.g. Toast, BurnMax my CPU Stretcher are not stability programs like Prime95 or OCCT etc....

most components that will fail due to being loaded will fail in the first 24hrs......

CPU Stretcher pushes your CPU but it has the proper routines inside and the CPU temperature will not arise like other programs --> try FPU long test of S&M for example...
My idea was as more as possible stress keeping as lower as possible temps.

benefit14snake:
You should realise that I don't press someone to use this program.
First of all I warn the users before use this program so I am not responsible for any damages.
My MSI is dead now before burn-in my CPU just passed SuperPi 8M at 2650 MHz (Vcore=1.663) and for 32M needed 1.716 Volts (+50mVolts)...
After this process became stable at 2680 MHz (1.663 Volts) or I could run it at 2650 MHz with Vcore < 1.663 and this was the difference...
But I would to play more and more and CPU became unstable anyway it is not necessary to repeat the progress.
Of course I respect your ideas.

[benefit14snake]

e.g. Toast, BurnMax my CPU Stretcher are not stability programs like Prime95 or OCCT etc....

CPU Stretcher pushes your CPU but it has the proper routines inside and the CPU temperature will not arise like other programs --> try FPU long test of S&M for example...
My idea was as more as possible stress keeping as lower as possible temps.

benefit14snake:
You should realise that I don't press someone to use this program.
First of all I warn the users before use this program so I am not responsible for any damages.
My MSI is dead now before burn-in my CPU just passed SuperPi 8M at 2650 MHz (Vcore=1.663) and for 32M needed 1.716 Volts (+50mVolts)...
After this process became stable at 2680 MHz (1.663 Volts) or I could run it at 2650 MHz with Vcore < 1.663 and this was the difference...
But I would to play more and more and CPU became unstable anyway it is not necessary to repeat the progress.
Of course I respect your ideas.

So wait, is your MSI board broken, or your 3200+ broken? Or is it both...? Is there a possibility that the burn in process could have been one of the factors that led to the failure of your proccessor? (although that doesnt seem likely......) unless your temps were horrible. Also, how many proccessors (amd more specificaly) have you tested this on? Im not talking about you hearing on a forum some one else had success doing it, im talking the real-deal here.. good ol real life experience. Also, what made you think to try to burn in a CPU in the first place?

I respect your opinions as well, but I want to know for sure if there is any truth to this whole CPU burn in thing. I don't want the little noobs running around burning out their CPU's. I don't know if you've noticed, but us enthusiest don't have a good rep among the non overclockers, because of the common fear "My CPU will MELT if I overclock at ALL!!!!!!"



EDIT: Also, you said those programs stress your cpu (meaning put them underload) but no extra heat arises? Physicaly IMPOSSIBLE. Now if it puts you CPU under 10%, 20%, hell even 50% load, then sure, no NOTICABLE extra heat would be seen. From what I've heard though, when ever work is done, heat and or sound is usualy released. HAHA but im no where near a physics professor, but merely a high school student who has passed physics class. So if it is possible to stress your CPU and not release extra heat, PLEASE PLEASE PLEEEEEEASE tell me how so I can. (Although I think in a vacuum things are a bit different.)

[MrSeanKon]

I am leaving after some hours for vacation....

1) My MSI is dead --> http://octech.gr/forum/viewtopic.php?t=997 it had problems from beggining.
2) My CPU is OK don't worry.
3) I compare CPU Stretcher's temperatures at full load with other normal programs (video tools, Maths, audio etc) at full load also.
4) My personal databases --> http://octech.gr/forum/viewforum.php?f=24
5) Some stickies --> http://octech.gr/forum/viewforum.php?f=13
6) Also my guides --> http://octech.gr/forum/viewforum.php?f=21

You should realize that I cannot double post everywhere.

Merry Xmas and happy new year.

[alex666]

Just a few minutes ago, I discussed this whole "burning in" topic with my father-in-law, a retired defense communications managing engineer. I'm talking about a guy who earlier in his career managed the development of the communications equipment used in some of the original communications satellites, so he knows his stuff. Anyway, he stated that "burning in", aka "infant mortality", is actually a manufacturing technique used to identify those electronic components that fail to meet spec. Typically parts will be run for, say, 1000 hours, and those that fail never get passed on to customer. In other words, it's a QA process. As for "burning in" a cpu to increase somehow its overclockability, there is no engineering or scientific basis. If a cpu is destined to work only at, say, 2.5 GHz, then "burning it in" will not increase its capability to say 2.6 or higher.

[benefit14snake]

Thanks guys, I'm pretty sure its safe to say that...........CPU BURN-IN IS A MYTH!

[chucko]

i'll probably read this whole thread later, but...

i realize this thread is about "CPU" burn in, but what about the other components? what about the motherboard and ram?

thanks.

[thelmores]

maybe we should send this to "myth busters"!

[madhatter256]

maybe we should send this to "myth busters"!

That wouldn't be a bad idea!!

[alex666]

Great idea. While I totally respect my father-in-law and his engineering knowledge, he ain't God either.

marty

[benefit14snake]

Yea that would be awesome, lets do it hehe

[benefit14snake]

i'll probably read this whole thread later, but...

i realize this thread is about "CPU" burn in, but what about the other components? what about the motherboard and ram?

thanks.

I would imagine, if its a myth with CPU, it would hold true for othercomponents as well. Although... I might try to burn in my new mouse a bit later!!! Make sure to burn in that new monitor too, leaving it on for 6 months straight will result in BETTER performance!!
I left mine on for a year straight, and now the picture is brighter, my dead pixels came back, and the ghosting completly went away!

(hehe not trying to make fun of ya bud, I just think the whole burn in thing is kind of funny.)


Anyways - Im going to go see if I can send to mythbusters. But on a side note, i was thinking of making a overclocking MYTH guide, since theres alot of overclocking guides that tell you HOW to do it, I want to make one that shows how NOT to do it. Obviously I couldnt do it alone, so id need some help - maybe Ill make a short list and let people add from there, and then we can see which ones we want to include on the first pages, which would be like the results page. Some of the things I wanted to do would include like: CPU burn in, running AMD "1:1" (skt 754 and above), memory timmings VS frequency (amd), amd and 1/2 multis and the big one: Overclocking will MELT your proccessor. That is one I personaly hate, when ever I mention overclocking at my school, i get some person (who barley knows how to browse the internet) say "overclocking is stupid, you will burn out your computer!"

[scrandman]

Anyways - Im going to go see if I can send to mythbusters. But on a side note, i was thinking of making a overclocking MYTH guide, since theres alot of overclocking guides that tell you HOW to do it, I want to make one that shows how NOT to do it.

That's a good enough idea, but let's not make it a flaming session, eh?

[Antonik]

You're not really burning in the CPU, but the adhesive, and only if its a paste/liquid, like the AS5 or Ceramique, since the "burn it" allows the adhesive to fill the voids and have a better connection between the heatsink and CPU.