Replies to post #13863 on Intel Corp (INTC)

[chipguy]

Intel has a core in Dothan that has a lot more potential to be a great server chip, better than Itanium (see SpecInt)

Sigh. Haven't we beaten around this particular bush already?

1) Too bad PC myopic people think that SPECint accurately
reflects the commercial or technical workloads that most
people buy large scale servers to run. If that was the case
Alpha would have been much more successful than SPARC
instead of the other way around.

2) Who says that on equal process footing that P-M will be
better on SPECint than IPF? At 1.5 GHz the 130 nm Banias
gets 995 SPECint_base2k compared to 1408 for the 130
nm Madison 6M. IPF's advantage in SPECint IPC should
increase in 90 nm with Montecito vs Dothan as well as the
IPF chip likely having better frequency scaling at 90 nm. All
that despite IPF chips not being optimized for SPECint like
x86 desktop/mobile chips.

Summary:
- any architectural based SPECint gap favors IPF going
foward
- SPECint doesn't even matter for current IPF target markets.
If and when it does (Tukwila derivatives for desktop?) one
could expect suitably optimized implementations.


provided that Intel makes some adjustments to it to target the server market.

Extend P-M to server class features and you seriously
close the gap in cost and power vs IPF that makes the
Dothan so intriguing as a fast cool little chip.

[Tenchu]

Joe, Intel has a core in Dothan that has a lot more potential to be a great server chip, better than Itanium

What isn't really fair about that comparison is that the McKinley/Madison core isn't the best that Itanium could be. In fact, that core is rather basic, despite it being a "2nd-generation design."

I think there's still a lot of room for improvement in Itanium, enough to justify continued R&D in a proprietary architecture.

Tenchu