Archived from groups: alt.comp.periphs.mainboard.asus (
More info?)
Thanks for the info will post my results when i am done but not till
next weeked
Paul wrote:
> In article <10evhoeklinkm07@corp.supernews.com>, "D.M."
> <"fartman"@(Nospam).charter.net> wrote:
>
> > Anyone know any good settings for overclocking a p4 3.2 i have
> > watercooling and ocz 4400 ram at 550 mhz thanks
>
> Look up your processor here:
>
> cpudatabase.com
>
> Pick some entries that have water cooling or are
> representative of your hardware.
>
> Say the 3.2GHz will do 4.0Ghz. That is a 25% overclock.
> Raise FSB from 200 to 250. Leave the RAM setting in the
> BIOS at 400, and the RAM will actually be running at
> 400*250/200 = DDR500. You'll probably end up leaving
> the RAM at 3-4-4-8, as I doubt it can be tightened up
> to CAS2.5 . It might not even overclock that far, as
> there are no guarantees. Your ram will need 2.6 to 2.8V
> applied to it - do a Google search to see if anyone has
> a good starting value for you to try (or consult the
> datasheet from the manufacturer).
>
> While you can punish your processor by raising Vcore,
> check the datasheet for the processor to see what the
> maximum voltage is. Don't go all the way to the max
> value, staying maybe 0.1V on the low side of max.
>
http://developer.intel.com is a good place to start.
>
> If you have a P4P800 board, they don't like high 1:1
> overclocking, and you will see video artifacts on the
> screen under the conditions mentioned above. For such
> a board, you need to run 5:4, in which case your BIOS
> RAM setting will be DDR333 (actual DDR320) and the
> RAM runs at 320*250/200 = DDR400. You might be able to
> tighten CAS to 2.5 if you end up running the RAM that
> slow.
>
> You will need many experiments to find stable and
> aggressive settings. Each time, the BIOS will complain
> about being corrupted, when it is not. Simply use the
> clear CMOS procedure if this happens. Redo all your
> settings etc. (I got sick of that after doing only
> one experiment - hope you are more persistent
🙂
> Do not boot into Windows, until you can pass memtest86
> on a floppy diskette, for a pass or two. No sense corrupting
> a Windows install, unless you've set up a virgin
> disk for just this purpose. I'd keep only one
> disk and a CDROM connected during the experiments, just
> to prevent any accidents. The worst that could happen to
> the disk, is you'll need to reformat it.
>
> Paul
Facebook
Twitter
Instagram