Would a 2.4C OC'd to 3Ghz be faster than a stock 3Ghz?

[ArcticFox]

I was thinking, if you took a P4 2.4c and pushed it's FSB to 940-something to overclock it to 3Ghz, would it be faster than if you were to buy a 3Ghz chip with an 800Mhz FSB? You'd think the faster FSB would make it quicker.

[ooops]

Yes it would be faster because you would be increasing the front side bus.

Internal clock of the 2.4 =12 ( 12 @ 200mhz (quad pumped 800mhz).

But the 2.4 @ 3ghz would be 12 @ 250 mhz

where the stock 3ghz is 15 @ 200mhz.

[ArcticFox]

That's what I thought, but some hacker in Canada said that the chip isn't actually running any faster, just the communications link. I called bunk on that, but just wanted to make sure.

[emcron]

I dont know how much of a difference in performance it would make but I'd imagine that it would be a bit faster. But the stock 3Ghz would probably be more stable.

[Bullion]

dude your arguing semantics, the system is faster, so yes the comms link runs faster, the actual chip isnt as the clock frequency is derived from the fsb prior to any calculations, so a chip running at 3g on a 250 fsb is still running at 3g, but the memory interface and bus comms are faster.

eg
fsb----->multiplier------>chip frequency/operation.

changing the multiplier and fsb concurrently doesnt change the final operation of the chip, it does affect the system though

[ArcticFox]

Which is exactly what I told "Pimp Con Seannery" up north.

I am running my P4C 2.4 at 3GHz with the stock retailer cooler. The additional 50MHz front-side bus rate dramatically improves bandwidth. Read my findings at ~2.9GHz here at TechIMO....

http://www.techimo.com/articles/index.pl?photo=140

Actually, this same test system will push well beyond 3.4GHz with better cooling and relaxed memory timings. I just can't deal with the fan noise in my current environment.

[HeadBand]

i dont think it would accually be any faster because the multiplier is different and the faster fsb is making up for that making it the same speed

but you never know...someone should do some benches on this

Unless you find MIPS numbers informative, benchmarking the core is rather useless. The overall performance curve will be faster for the overclocked system, especially for memory intensive operations.

The actual core operational frequency is the same for a P4C 2.4 overclocked to 3GHz as it is for a P4C 3.0 chip. There is no difference in internal frequencies except for a slight variation imposed by the motherboard's PLL IC generator.

Assuming the bandwidth difference was not present, there would be no performance difference in the two systems. However, the bandwidth difference is unavoidable, thus the overclocked P4C 2.4 will win in every benchmark.

Robert Richmond