To Burn In Or Not To Burn In Das Cpu

[bugmage]

Burning in sonds like a stupid idea. What it accomplish? So what if you can get to a higher cpu clock, what are the costs. Woulden the added heat simply blow out more componets, possibly short some safeguards allowing for higher clocks?.

Ive read a lot of articals on people who swear by it, and Ive seen cpu performance charts by age, and none of the CPUs get faster with age, things burn out and they always slow down.

Can someone help me out here, on a physical lvl, how would this help the cpu? :confused:

[|MaguS|]

Burn In is good because it can detect if your CPU has in flaws in it, It will also help to determine its limits in temp and possible OCing.

Burn In and Burn Out are not related, Burn Out is when the CPU dies due to over voltage or too much heat.

For more info...
http://en.wikipedia.org/wiki/Burn_in

[bugmage]

What about those people who say they can hit higher speeds after burning in a cpu

[|MaguS|]

You can sometimes, Burn In sometimes gives the CPU (or RAM) more overclock flexiblity.

[bugmage]

yes thats what they say but how is that physically possible, componets loose conductivity with heat, the only thing i can think of is that some parts might weld better together but thats a half ass explanation, Ive read that theres no proof either way, so lets speculate on it

[Roadhog]

its like a new car, you have to break everything in for it to work right.

[bugmage]

that doesnt make sense in a car you want the fitting to melt together for a better fit on the pistons or w/e (computer nerd not car mechanic) there are no moving parts on your cpu, so heat expands the internal "wires" in the cpu, so maybe those expanding force ends together giving them better contact thus high clock speeds idk,

comments anyone?

[-=DVS=-]

I know for a fact some thermal grease need time to settle down for max effectiveness so thats might be the reason ,as for burn in for computer hardware i think thats myth.

[bugmage]

nice icon ;)

yeah that sounds like a myth, I use the artick silver 5 and with a litte time it does get better because it takes a while for the different sized particals to fill in the cpu cracks, however i dont think you burn in a cpu for that reason, better to just run it and let fan vibrations do it for U

[zachary]

I'm pretty sure burn-in is real. An example: When I got my ram, my system wouldn't boot at 2-2-2-5 timings, no video whatsoever, but a few months down the road, it did boot at those settings, but wasn't all that stable. I've heard of burn-in existing with pretty much everything where a clock is involved. I can't prove anything, but I bet it has something to do with the effect of voltage over time on transistors.

[CaptNKILL]

I deffinitely believe in burning in CPUs...

When I got this 3000+ Venice, I couldnt pass prime95 for a minute at 2.2Ghz. A month later, I could run it for hours at 2.5Ghz. If that isnt burning in, I dont know what is... my temps have stayed about the same, so I doubt its anything cooling related (ie, the thermal pad settling).

Heh, I better go do something constructive before I decide I want to try to overclock more... every time I read threads like this, its hard to resist cranking it up another 100Mhz ;)

[j0j081]

You might have to get some expert advice on this one. I've heard the rumors to but never really understood it.

[Absolution]

heh, well, when i first started OC'ing my friends 4200 + OCZ VX 4000, nothing would run at the rated speeds (cept the CPU)., i tried cranking up the FSB to 250 (not right away), of course not stable. i then ran memtest for about a day on everything stock, then prime 95 for about 8 hours. started ocing again, got the memory to its rated 2-2-2-5 timings (finally, couldnt before even on ddr400 speeds) and upped the fsb to 250 and its been stable ever since =\ (slight CPU oc too at 2.25ghz)

[Nutty]

so maybe those expanding force ends together giving them better contact thus high clock speeds idk,

comments anyone?

Utter bollox.. Theres no such thing as getting better performance for a given clock speed by burning in a cpu.

[bugmage]

Utter bollox.. Theres no such thing as getting better performance for a given clock speed by burning in a cpu.


Ok so if you can reach a higher overclock after a burn in like the other guys posted wouldent you be getting better performance?

I understand what you saying that the cpu will be slower that the burn in clock speed, like it should, but what is adding the extra stability? :eek:

[-=DVS=-]

Thing is there is no scientific proof , for all we know that guy changed something in the system/hardware/software/bios ect that contributed to higher OC and just doesn't remember :p or paid attention.

Now if someone made experiment with dedicated test system , new hardware max stable OC , leave PC untuched for a week 24/7 and try OC again , that would be somewhat more reliable info ;)