someone (someone@somewhere.com) on 11/22/10 wrote:
---------------------------
>Richard Cownie (tich@pobox.com) on 11/22/10 wrote:
>---------------------------
>>Anyhow, they've worked on Itanium for 15+ years and it
>
>Let me correct that for you.
>
>"They" have created just two microarchitectures for IA64
>over the last "15+ years", both designed:
>1) in the 1990s
>2) for 180 nm Al interconnect CMOS
>3) before any IA64 systems were deployed into customer
>sites and actual workload trace data collected
>
>>has never looked good on performance/watt.
>
>Despite issue 3 McKinley did quite well with perf/W vs its
>contemporary conpetitors similarly constrained by 1 and 2.
>
>Poulson will blow past issues 1, 2, and 3 with a vengence.
>It will have much higher performance than Tukwila in the
>same socket power. Will its perf/W be better, comparable,
>or worse than Westmere-EX or Power7+? Only time will tell.

Much higher mtp vs Tuk is a must; it would be beyond ridiculous otherwise. Tell us one thing though: are you saying it will have much higher stp as well? How much higher then?