There have been cases of real overclocks being sold as "faster" chips in the past (in the PC market)...all those companies are no longer in business. Fraud is what that is called.
g-news is correct. used to be vendors would order a huge number of chips, then take the time to see how high they would clock and still be stable. they could clock quite a few well above the rated Mhz. The chip manfs. stamped that out quickly though, as they were selling these chips as a legit speed when it was set to something well above where they were rated. would likely make the chip unstable.
If you overclock a chip beyond it's rated specs and try to repackage it as a faster chip you are breaking the law. The chip manufacturer puts the speed rating on a chip to protect itself.
If Apple was to overclock a chip beyond it's stamped rating it would have to disclose it. If they didn't they would be liable for any failures because they put a UL rating on their product and by going outside the "rated" limits of the components used they would have violated the law.
That's why you used to see 3rd party PC manufacturers selling "overclocked systems" but they had to state that they were overclocked and either had to warranty all parts themselves or sell it as is. A vendor will not cover warranty items for parts that have been misused.
As for the rest of the article, the author is a complete moron. He says
[quote]
A new mobo controller chip isn't something that you conjure up in two weeks, and the fact that they actually designed an entirely new piece of silicon to support this kludge means that Apple has been making its engineering plans based on the assumption of another Moto speed stall.
<hr></blockquote>
Doesn't he realize this is the custom chipset used in the Xserve, Apple made this public knowledge and it's on their freakin' website. He makes statements that the L3 cache doesn't help alleviate bus traffic, not a bright boy.
Let's just put it this way....... He seems's to know something about computers, but he's either flat out lying or making it up as he goes.
(PS to anyone: the P4 does NOT support L3 cache, is that a true statement? I think so, appreciate a response)
G-News, I know the Xeon is different than the P4, but that also covers the big price difference. Does anyone know if the P4 supports level 3 cache?
I'm not concerned over a head to head (G4 vs P4) comparison (enough have been done). I'm only looking to debunk this really crappy review that started this thread. To say the L3 cache doesn't relieve the bus from traffic enough to make a big differenece is a foolish statement. (By the way, I'm intoxicated)
IIRC processors are rated for a particular speed and recommended use (at a given temperature) Thusly, a bunch of procs rated to run at a nominal temp found in the average home/office setting could come of the same run as another group rated to run much slower but at a much higher average temp (perhaps found in an industrial setting.)
The issue came up a few months ago when one of the upgrade manufacturers was using slightly clocked PPC's. They were using industrial ratings and the slight clocking they recievedd was acceptable practice because they were running at much cooler temperatures in the desktop setting.
It's possible that a huge cooling improvement could guarantee a lower nominal temperature whereby the CPU could run faster WITHOUT exceeding it's rated specification. Not likely what's going on, but possible.
Comments
G_News
companies don't do that kind of thing.
If Apple was to overclock a chip beyond it's stamped rating it would have to disclose it. If they didn't they would be liable for any failures because they put a UL rating on their product and by going outside the "rated" limits of the components used they would have violated the law.
That's why you used to see 3rd party PC manufacturers selling "overclocked systems" but they had to state that they were overclocked and either had to warranty all parts themselves or sell it as is. A vendor will not cover warranty items for parts that have been misused.
As for the rest of the article, the author is a complete moron. He says
[quote]
A new mobo controller chip isn't something that you conjure up in two weeks, and the fact that they actually designed an entirely new piece of silicon to support this kludge means that Apple has been making its engineering plans based on the assumption of another Moto speed stall.
<hr></blockquote>
Doesn't he realize this is the custom chipset used in the Xserve, Apple made this public knowledge and it's on their freakin' website. He makes statements that the L3 cache doesn't help alleviate bus traffic, not a bright boy.
Let's just put it this way....... He seems's to know something about computers, but he's either flat out lying or making it up as he goes.
(PS to anyone: the P4 does NOT support L3 cache, is that a true statement? I think so, appreciate a response)
Caches aren't an x86 strength anyway, neither are registers.
G-News
I'm not concerned over a head to head (G4 vs P4) comparison (enough have been done). I'm only looking to debunk this really crappy review that started this thread. To say the L3 cache doesn't relieve the bus from traffic enough to make a big differenece is a foolish statement. (By the way, I'm intoxicated)
The issue came up a few months ago when one of the upgrade manufacturers was using slightly clocked PPC's. They were using industrial ratings and the slight clocking they recievedd was acceptable practice because they were running at much cooler temperatures in the desktop setting.
It's possible that a huge cooling improvement could guarantee a lower nominal temperature whereby the CPU could run faster WITHOUT exceeding it's rated specification. Not likely what's going on, but possible.