Hard limit for 800 FSB CPU?

[Vatt]

Could it be that increasing VCore will not help anymore once you get beyond a certain point?

If have a P4 3.0(800) that will do 3900 Mhz with 1,725 volt Prime stable, but fails Prime at 3915 Mhz (FSB +1). Even increasing the vcore all the way up to 1,90 doesn't help either.
Memory is not an issue as I have already tried the 3:2 divider and doesn't help. Heat is no problem because I use a Prometeia.
3DMark can be run upto around 4.0 Ghz, but not much more.
Is this your experience also, or should I look for some problem with my Abit IC7-G ? I have read about the 275 'brick wall' but I'm not even close to that.

Ofcourse 3.9 is not bad and I can live with it ;) , but maybe I'm missing something here.....

[Major]

At some point Vatt increasing the Vcore will no longer yeild any increase in OC. I'm at 3705 with 1.7 vcore with my current cooling and gain nothing by increasing vcore.

Yea 3.9 is nothing to complain about IMO !!! hehe

[Tom Holck]

quite stable

[Tom Holck]

The IC7 G can go far higher than 283

[OPPAINTER]

Tom,

Nice Clock :)

What sort of cooling you have there??

OPP

[Amoeba Assassin]

p4c800 pwns all though. im at 3.6 on air with my 2.6c:)

[Tom Holck]

Originally posted by OPPAINTER
Tom,

Nice Clock :)

What sort of cooling you have there??

OPP


My "Old" Prometeia on R404

can go below -53 degree at 1.81V load.

[Liquid3D]

Originally posted by Major Slaughter
At some point Vatt increasing the Vcore will no longer yeild any increase in OC. I'm at 3705 with 1.7 vcore with my current cooling and gain nothing by increasing vcore. Yea 3.9 is nothing to complain about IMO !!! hehe I have to agree with the Major. Especially after writing that (poorly executed) article last week about wafer fab.

If it's the case that a manufacturer aims for a specific, consistent yield throughout the chips design. There will be a certain performance ceiling dependent upon plant, wafer quality, photolithographic process etc. I beleive once the voltage is maxed out, there is still room for increase, the obvious method being thermallly related. If the temperature can be lowered, it's almost certain a few Mhz can be squeezed otu without rasing Vcore. I imagine a processor running a 3.9GHz @ 1.7V and 33C, will run slower then the same processor at -33C.

One question, what is the S-code, Plant info?
And Tom I have the same question for you? I'd really appreciate seeing the results as my 2.4C is sitting in the FedeX wharehouse just minutes away! BTW Nice overclock!!!