News Focus
News Focus
icon url

wbmw

04/29/03 1:44 AM

#3344 RE: Dan3 #3336

Dan, Re: Given that Pentium III-M on .13, with half as much L2 and no SSE2, was about the same size, it's reasonable to assume that they've used some (or much) of their 90nm process technology to produce the chips.

So what's going to happen with Dothan, the Banias successor with 2MB cache. Will Intel have to use 65nm transistors to keep the die size small, or will it blow up to twice the size. Option 3 is that Dothan uses 90nm transistors, while Banias has a very efficient layout at 130nm. I'll admit that it's too soon to tell.
icon url

fyodor

04/29/03 5:40 AM

#3355 RE: Dan3 #3336

Dan3: Given that Pentium III-M on .13, with half as much L2 and no SSE2, was about the same size, it's reasonable to assume that they've used some (or much) of their 90nm process technology to produce the chips.

I don't buy this "they're burning their candle from both ends" argument (I think Paul E. put it that way when the argument was applied to AMD). The basic assumption in that line of reasoning is that progress stops, or at least slows dramatically. In fact, the progress in process technology is quite smooth and a company should use the smallest possible gate sizes they can at any given point. If you *can* use 90nm structures, why on Earth shouldn't you? You will gain useful experience and when the "official" 90nm process comes along, it will be that much more mature.

-fyo