I still can't quite understand how this L12 trick works. I understand the putting the wire in the socket (I had to do this to get my 2100+ "B" to run on my older Iwill XP333-R, else default to 100FSB x 20
). I just don't understand how it "adds headroom". If the chip and/or DDR is at it's limit, how can changing the "default" speed increase the maximum overclock?
For example, my 2100+ "B" is currently running at 11 x 200. (2200MHz). When it was on the Iwill XP333-R, I had weaker DDR, and ran it at 12.5 x 180. While "faster" than 11 x 200, CPU speed-wise, I make up for it with the faster bus speed/bandwidth at 11 x 200.
I dropped the multi down (on the NF7-S...I got the board before the PC3200) on my current combo to 9x, and the system wouldn't budge past 208, even with 2.9v to the RAM.
Ok...I guess the real crux of the question I have is...changing the "default" FSB of the CPU to 166 (333)...how the heck does this effect how much further the DDR will go?? It should still max out at 208ish, which is over the changed speed of 166/333.
It's a bugger to get my waterblock off in the main rig, so it's a bit more involved for me than those running air. If I
knew...I was "guarranteed" a higher overall FSB after dropping a hunk of wire in the CPU socket, I'd go thru the trouble..... I just can't understand the logic of how it works.....
There's another thread in this topic on the subject, and I asked the same questions there... Some report improvement, some report instability..... That's why I'm hesitant to do this (plus the involved process of pulling the waterblock).
B.