CPU Burn in?

[Peckers]

I've been thinking about this and im wondering if anyone knows how this really works

I think it's strange how a CPU hence transistors could work better after a burn in. Is it just that the thermal compund fill the cavities better when working at higher temps and therefore giving better temps. Or is the increased proc performance only a "sideeffect" of other components burning in, mobo, mem etc ?

[CCW]

burning in is to test for stability

[Peckers]

CCW: yes that too, but was I am seeking is an explanation to why a processor can work in higher speeds after leting it burn in for a day or two with high voltage. There shouldn't be any "running-in" needed like with a car for example, since there is no mechanical wear with transistors(as far as I know).

Anyone who can give som clarity to this? It would be nice to know, so that you can optomize this procedure(know what compoments are vital to stress).

[MrLavender]

Here is a somewhat technical (and unproven AFAIK) theory. Sounds plausible though.

[Peckers]

Thanks for the link MrLavender, just what i was looking for

[Major]

Good reading.

[CCW]

sorry