A question about overclocking (in general)...

[PlunX]

I know this cocky guy that fixes computers that I met at Fry's and started talking to recently.. He's totally against overclocking and is trying to tell me that I shouldn't do it at all. Recently, he told me something that boggles my mind.. He gave me an example that went something like this..

"Let's say you have a Duron 700MHz and a Duron 850MHz. You run them both on the exact same motherboard with the same type RAM, video card, etc. You overclock the Duron 700MHz through the multipler to 850MHz. You leave the Duron 850MHz at it's default speed (850MHz). I can guarantee you that the overclocked processor (Duron 750MHz at 850MHz) will not perform as well as the other processor (Duron 850MHz) running at the same speed but is not overclocked."

Is this true or is this total bullcrap?

 

[Fardringle]

It's bull crap. (I should know, since I'm a bull! ) There may be a SLIGHT difference (as could be expected between ANY two processors running at the same clock speed) and the O/C'd one may have a bit shorter life span, but after all, both processors are operating as Duron 850's...

 

[BA]

It's BS. Just for example, and PII 300 SL2W8 is a 450 part downgraded to 300 because of market demands. I'm pretty sure a 450 part running at 450 is the same speed as a 450. They're all the same chips, they're just graded by how fast they are capable of running(within certain tolerances) and the market.

 

[Coolie]

Yeah its BS. When I was running a celeron 333 @ 500, I was getting the exact same benchmark scores as people with true celeron 500s.

 

[PlunX]

I thought it was bullcrap.. He was trying to say that the chips are made to be ran at one speed and one speed only and the chips themselves can only recieve so many volts or something like that.. Can't really remember exactly what he said.. Just checking, though.

 

[robinhung]

That's bull sh_t, don't trust him. He knows nothing.

 

[GaryTcs]

What his friend is referring to is the fact that the manufacturer tests the chips for speed and gives them a little overhead for safety. The 850 in his example should outperform the overclocked 700, but probably by such a small margin that it is really pathetic to discuss.

 

[Stallion]

I know squat about Durons but from the intel stand point I would say that if you took a 600 and a 750 and you ran the 600 at a FSB of 124 to get 744mhz it would out perform the 750 at 750 as the FSB would be faster.

So from where I stand I would say he's a bonehead too. He was probably just promoted from the dishwasher dept to computers just a few days before, hence why he don't know nothing.

 

[paulip88]

The only way I can see his point of view is if he said that the slower clocked CPU's top speed is not as fast as the faster CPU's top speed (when both are OCed to the max). However, every other sane argument points otherwise unless you cripple the OCed system with considerably slower peripherals.

In fact, often you can get your OCed system faster than an equally clocked system simply by the virtue of the FSB being faster. Of course this varies by the processor you are using...

 

[GeoffS]

Now, if you OCed the 850...