If I may revive this thread....
I'm hitting a wall overclocking my 4200+ X2 around the 2400 mhz range and I'm hoping some of you who successfully oc'd into the 2.5 - 2.6 range will divulge your secrets (system specs are linked in my sig: My Builds / Main Workstation). For CPU cooling, I have an XP-90 with a Panaflo medium speed fan. Heat doesn't appear to be an issue, as my max load temps are <50 degrees C. I've tried memory timings of 2-3-2-5 at 2T (initially) and 3-3-3-8 at 2T.
Going through the steps in
Zebo's overclocking guide, I am stuck on step 2: setting the multiplier to maximum (11 for the 4200 X2): upping CPU volts to the 1.45 range (adding +0.05 volts in BIOS; Magic Health reports 1.45 V but CPU-Z shows a little higher), upping the DIMM voltage to 2.8, dropping memory speed to 100 mhz, and reducing the LDT Multiplier (aka HT Frequency in my BIOS) to 3x. With those settings, I can't get an FSB higher than 220 x 11 = 2420 mhz. I tried dropping the multiplier down to 10x, but the best I could do without upping the volts is 240 x 10 = 2400 mhz.
I tried loosening the memory timings to 3-3-3-8 and boosting the core voltage to +0.10 in BIOS (1.5 - 1.52 volts), but that only gets me to 250 x 10 = 2500 mhz, and Prime95 won't run for more than a few minutes before one core craps out on me (I've since returned to the 240 x 10 @ 1.45 volts, which is proving to be the most stable). I tried dropping the multiplier to 9x and raising the FSB further, but no matter what combination of FSB and multiplier I use I cannot consistently boot into Windows--let alone run two instances of Prime95--above 2500 mhz (230 x 11, 250 x 10, 280 x 9, adding up to +0.125 volts or core voltage reported in the 1.55V range, which is as high as I'm willing to go).
What has me stumped is that other folks are reporting success in the 2500 - 2600 mhz range while holding core voltages in the 1.45 range. Any ideas what I'm doing wrong? Thanks!