Replies to post #40972 on Advanced Micro Devices Inc (AMD)

[bobs10]

Nit picking. I know very well the difference between IPC and ghz. Faster was used in the sense that k7/k8's run programs faster than INTC's best. But then you knew that didn't you?

Really, to me it doesn't make much difference whether INTC's chips suck because the have a lousy design or because they don't use SOI(though I suspect some of both). If it's true that the Prescotts suck because they have a bazillion stages and a gazillion transistors then why doesn't INTC just drop the hot potato and go with the Dotham you seem so happy with. Could it be that maybe that chip has problems of its own that only start to show up when the ghz is increased?

INTC's Prescotts have a stench worse than a 2 holer in the Sierras at the end of summer and the Noconas look to be able to apply for EPA superfund status.

[Jules2]

SOI has nothing to do here, and neither does any of the other red herrings invented to throw more mud at Intel for what is really just a single - gigantic - mistake in planning.

I would have sworn Mr. Barret stated otherwise in his memo to *ntels employee's.
As McGee would say,"must be shom mishtake!

Regards

Jules

[fastpathguru]

The myth is that SOI is the reason, but this is absurd.

What is absurd is making a definitive statement like that when you have no idea.

To know that SOI has no benefit, you'd need to compare two architecturaly-identical chips, alike in all respects save SOI.

Hammer runs at faster clockspeeds than Athlon, a simpler design but 130nm nevertheless. That's enough proof for me that SOI may provide some non-negligible benefit.

Who are you to say that Opteron would be just as fast without SOI?

Do you know for a fact that Prescott or Dothan wouldn't improve with SOI?

Or do you just take Intel's word for it?

fpg