The Core i7 has the advantage that the memory controller is part
of the processor chip. That reduces the latency or delay, for
info to get from memory, into the processor. That is the main
benefit of Core i7. Basically, Core i7 is doing the same thing
that AMD has been doing for a while - direct memory interface.
*******
With regard to your question about the older Extreme...
http://en.wikipedia.org/wiki/Core_2
"This means that the only major difference between the regular Core 2 Duo
and Core 2 Extreme, is the clock rate and unlocked multiplier....
The unlocked upward multiplier is of use to enthusiasts as it allows the
user to set the clock rate higher than shipping frequency without modifying
the FSB..."
So the Extreme is mainly intended for rich guys, who like to overclock
without having to work very hard. You basically dial up the multiplier
until it crashes.
With regular processors, the multiplier is limited (my processor only
goes to x13 for example). If I want to overclock, I have to increase the
FSB frequency. Some chipsets can support FSB settings all the way to
FSB2000, before there is trouble. So if you purchase a nominal FSB1333
processor, the "FSB headroom" makes it possible to do a 50% overclock,
if the processor will allow it.
The owner of an Extreme, on the other hand, doesn't have to torture
the chipset on the motherboard. By adjusting the multiplier freely,
they can increase the core frequency inside the processor, without
affecting other subsystems. For the privilege of doing that, Intel
wants more money.