The key (IMHO) is the same as always - the multiplier + the stepping of the CPU.
This doesen't seem to have changed a bit since the days of the Celeron 300A - locked multiplier, but stuff darn near any of them in a Abit BX6 motherboard (with the stock heatsink/fan), adjust the FSB from 66 to 100, and
whosh - instant 50% overclock!
Some could get a bit more, but it took a
LOT more effort and equipment.
The AMD chips let you unlock the multiplier, so a "faster" newer (higher inital speed) chip can be effectivly dropped back to a "slower" multiplier, but with a better (later) yeld. Even so, they only OC a little better than their older cousins.
So just what am I trying to say?
That those who try to overclock higher multiplier (locked or unlocked) CPU's and are expecting to achieve better results than someone with a lower multiplier CPU (of the same manufactered date) are bound to be disapointed.
IT"S THE SAME CHIP!
But you just paid more for a chip that Intel or AMD has ALREADY "OVERCLOCKED".
In fact, when we hit a batch of chips like the old 300A's or the new 1.8A-C1's its more like "re-clocking" or "right-clocking" than "over-clocking".
No need to apply more voltage.
No need to buy screaming heatsink/fans, much less chilled water blocks.
Just drop the chip in a good motherboard, adjust the BIOS settings and enjoy the ride.
If there were 1.6A-C1's I'm pretty sure they would OC even better than the 1.8A's.
It sure can be fun trying to squeeze the very last MHz from whatever you are running, but don't expect the same kind of percentage increase form a new 3GHz ($700) chip as the guy with the 1.8A-C1 will get for (maybe) $150 when they hit the mainstream.
Of course, getting a $500 PC just to test the chip like I did
is a little nuts!
I just had to see if those 1.8A's were able to do it.
They do.
Just my two cents.