Originally Posted by xxbassplayerxx
History doesn't point to these 100MHz increments doing much at all for overclocking. A new stepping, however, would most likely prove to be quite an upgrade.
When I moved from my Core 2 Duo E8400 C0 to my Core 2 Duo E8600 E0, it definitely seemed like the stepping made a bigger impact rather than the default clock speed. Granted, the multipliers were locked, so the higher default multiplier aided me in getting higher since the RAM became a limit, but since that's much less so with these chips, I'd imagine you're right.
Originally Posted by Don Karnage
Ivy Bridge 1.4B Transistors > Sandy Bridge 1.16B Transistors
Ivy Bridge 63X Multiplier > Sandy Bridge's 57X Multiplier
Ivy bridge's DDR3 2800 Memory Support > Sandy Bridges DDR3 2133 Memory Support
This is a lot more then a die shrink
The extra transistors are likely due largely to the IGP and/or cache, no?
The multiplier is an artificial limit, no?
RAM support is merely an official sanction and little more, so far as I know (I could be wrong).
3D transistors are a sizable deal, yes, but we saw big changes with the 45nm Wolfdale too with it's then new high K metal gate transistor implementation. Whether this new 3D transistor translates to another Conroe to Wolfdale type of increase remains to be seen, however.
Bclk tinkering is a part of the LGA 2011 platform chips. It has nothing to do with Ivy Bridge. Sandy Bridge-E will have it, and Ivy Bridge on LGA 1155 will not.