I understand that the CPU's nm process determines how effective the frequency becomes. For example a 3.0 Ghz frequency with 32nm process (sandy bridge) would be out done by a 22nm process (Ivy-Bridge or Leaked Haswell). How does it scale, exactly? Is there a direct way to compare a 32nm Sandy Bridge to an Ivy-Bridge / Leaked Haswell's 22nm, some magical formula? Also, why is it that the Haswell is expected / leaked / rumored to use the exact same 22nm process as an Ivy Bridge, what benefits, other than its "3x better" internal GPU, does it have?
While your response does explain the differences between the Ivy Bridge and the Haswell, it doesn't explain HOW it's 10% better. I'd love some more specifics regarding how the nm fits into play and the like.
1) Why does the Haswell have the exact same nm as the Ivy-Bridge, yet will apparently perform 10% better?
2) How does nm effect frequencies at all, how can we determine and compare CPU's with the same frequencies with different nm's?
sterlin, nm has nothing to do with that. Nm is how big the transistors are, and usually, the tinier the transistor, the more there are in a die - the more transistors, the better the performance, which means lower clockspeed - still improved performance, like a 2.8GHz 22nm processor vs a 3.0GHz 32nm processor.