I'd just like to point out some issues here and correct some misinformation that has been spreading in this thread.
If you guys actually bothered to read the original thread over at XS, which can be located here, you would have noticed a couple of things.
4.2GHz on air wasn't even close to stable. It was a suicide run. Getting a CPU-z validation is not hard. I was also able to do a suicide run of 4.2GHz on the stock cooler of my i7 920. The Ph2 result was was done on very good air.
The 4.7GHz result was not on air. It was achieved on dry ice.
There actually is a legitimate disadvantage to a 3-core CPU. Many applications are programmed to only use 2^n cores, and in the case of 3-cores, they will only use 2 of the cores.
These points aside, he has achieved a very respectable overclock.
Yes, but 1 is still a power of 2. 2^0 = 1.