So I'm in a bit of a pickle. I determined I can safely overclock my P4 2.4B to 2.8 using an FSB of 156Mhz and a 1:1 divider on the memory (PC3200 DDR RAM @ 312Mhz). But I need about 1.65v to the CPU to keep it stable.Alternatively, I can go with a 5:4 divider on Mobo, jack the RAM up to 370Mhz, but the FSB can only go to 148Mhz (2.66Ghz). The CPU requires much less voltage, but the memory needs a boost to 2.8V.I'm inclined to think the latter is a better alternative, since my mobo does not have a PCI/AGB bus lock. The HD don't care much for high FSB's.Am I likely to see better overall system performance by favoring faster RAM or faster CPU?Cheers,J

Tough call but I guess it

Ideally, if I could get the FSB up to 150 w/ the 5:4 divider, that would bring my CPU to 2.7Ghz and the memory to 375Mhz (187.5 x 2); which to me seems like the best of both worlds. The most stable configuration I have to date, seems to be the 2.8Ghz setup - but having to underclock PC3200 memory to do it is a darn shame. The other downside to the 2.8 CPU overclock, is that I have the old warm boot problem.I suppose the real answer is get a new MB with a PCI/AGP lock. I could easily get this thing up to 3.0Ghz, I think if it weren't for the hard drives getting cranky on me ;-)J

Yes generally overclocking really don

As it turns out, I had some kind of FSB modulation that was enabled in the BIOS. Apparently, MB manufacturers are enabling this in order to comply with strict FCC regulations about RF interference. This apparently was causing all kinds of instability. I disabled it. I can now run a 150 FSB (2.7Ghz) with the 5:4 clock divider (375Mhz DDR) w/ only a 0.5v increase to the CPU. I could get it running at 156 (2.8 Ghz) but w/o the PCI bus lock it's not the most stable option. Not to mention it takes ~1.65v to keep the CPU stable; the added voltage was raising the heat precariously near 60C under load (up about 13C) and my memory had to be taken down from CL2 to CL2.5. That seemed to me like alot of excess baggage for another 100Mhz. What do you think?For giggles, I did get it booting once at 3.0G w/ the 1:1 divider, but the amount of voltage it took to keep the CPU stable was approaching 1.8v and that's just a bit too rich for my blood any way you cut it. I don't know how some people are getting 3.2G+ and up on these chips w/o running them out of spec. Guess I got a bad die lot. At least I can say I did it :^)JPS. It's amazing to me how a 1C degree change in ambient temperature can cause nearly a 5 degree change in CPU temp.

Yup 1,8 on P 4 is kind of high. A piece of cake for Bartons though.You can 100 % safe go up to 1,85 V on thoseI run mine at 2V (not recommended for the normal user or if you aren

Is 55-60C under load still reasonable for a P4 with standard cooling (well, it's a little better than standard: I have a swiftech heatsink w/ Arctic Silver compound applied). 60C has always struck me as a little high for a Pentium. I'm used to running in the high 40's. But maybe I'm just being paranoid ;^)

Well if you are stable it don

My CPU has just lived through its second summer (no air con). CPU temps averaged 55-57, with the occasional push up to 59-60 on really hot days outside. Intel CPUs have overtemp protection anyway (90C I think), whereby they start throttling back if it all gets too hot. Temp schmemp I say!GaryEdit: sp

However they won