lol i think we should have a burn in comp forum wide :P see who can get the best speed:voltage ratio.
Those are just rough estimates based on the potential this has demonstrated thusfar...core1 is rather weak however and core0 shows potential of hitting 2.6ish with no vcore change
Burning in is not neccesary and there's nothing wrong with overclocking right away...generally what you might notice is you might initially reach a limit but then try again a week later or so and find that you can squeeze a little bit more out of it
If you're talking about burning in, in almost all situations burning in at stock speeds is useless and you can base this on your stepping...either your core has been binned and stressed at the specified speeds already OR your core got really lax binning in which case you probably have a lot of headroom with stock vcore to bump up the speeds as I demonstrated above
So yeah, IMO you should determine how good your stepping is...if you find it is a highly acclaimed stepping you should bump up clock frequencies as high as stock vcore will take them and THEN start burning in...burning-in is often widely confused with allowing thermal compound to settle, in which case after a week or so it should drop your temperatures a few degrees and possibly allow a higher clock
So there are too many variables to determine if stressing a core increases its threshold...I would personally say yes, but at the same time overclocking out of the box isn't going to have any effect on decreasing that threshold
And FYI the results you see above are from overclocking a core out of the box...I really pretty knew what to expect though based on the stepping though