Frequency is different for starters

Article: Sizing up the Super Heavyweights
By: Paul DeMone (pdemone.delete@this.igs.net), October 27, 2004 5:05 pm
Room: Moderated Discussions
IntelUser2000 (Intel_user2000@yahoo.ca) on 10/27/04 wrote:
---------------------------
[..]
>Do I see an error you made Paul? Why does your older article on future chips(Link: http://www.realworldtech.com/page.cfm?ArticleID=RWT042704221446)
>> state that Montecito gets
>SPECint_base_rate2k: 62
>SPECfp_base_rate2k: 87
>
>with this quote: "The effects of improvements in compiler technology or platform
>support (chipset, FSB speed, memory etc.) over the next 12 to 18 months were not included in the performance estimates."
>
>While your new article says(Link:http://www.realworldtech.com/page.cfm?ArticleID=RWT100404214638):
>SPECint_rate_base2k, 4P: 96
>SPECfp_rate_base2k, 4P: 120
>
>Do you mean that now your estimation of spec-rate performance is off since you
>estimated back then SPECint_base_rate2k: 62,
>
>and now you say
>
>SPECint_rate_base2k, 4P: 96
>
>So which one is right?

Right? These are estimates based on the best available
information not the voice of god talking to me. Even
Intel won't have definitive numbers until near release. :-/

The earlier estimate is based on a 2.5 GHz clock rate
running SPEC while the newer one is based on a 2.2 GHz
clock rate running SPEC. Also the new estimate takes
into account the much coarser scale of multithreading
in Montecito than was assumed in the earlier article.
This greatly limits its potential to improve throughput
of SPECrate type workloads.

> Oh yeah and another question: I heard Maximum power
> dissipation of Madison 6M was 130W, and its TDP: 107W,

Wrong. The defined maximum power for the IPF family is
130 W. Maximum peak power for I2/1.5/6M is around 120 W.

> and I also heard that Montecito will have 100W power
> dissipation rather than 130W of the previous generation
> Itaniums so does it mean that Montecito will have 100W
> max power dissipation or 100W TDP?

With Foxton technology there is little difference in practice.
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
New Silicon Insider ArticleDavid Kanter2004/10/11 04:43 AM
  Good read. Thanks again Paul. (NT)Michael Menietti2004/10/11 05:36 AM
  New Silicon Insider ArticleSingh, S.R.2004/10/11 05:44 AM
    New Silicon Insider ArticlePaul DeMone2004/10/11 05:55 AM
  New Silicon Insider Articletecate2004/10/11 08:52 AM
  core size: x86 vs IPFanonymous2004/10/11 10:43 AM
    Good Question. (NT)Chuck2004/10/12 07:22 AM
    core size: x86 vs IPFMichael S2004/10/12 09:53 AM
    K8 core size 52mm^2IlleglWpns2004/10/12 11:15 AM
    core size: x86 vs IPFPaul DeMone2004/10/12 12:10 PM
  New Silicon Insider ArticleJS2004/10/13 01:42 AM
  New Silicon Insider Articlemas2004/10/16 08:07 AM
  New Silicon Insider ArticleAnonymous42004/10/16 09:39 PM
    New Silicon Insider ArticlePaul DeMone2004/10/17 05:58 AM
  New Silicon Insider ArticleArun Ramakrishnan2004/10/22 07:48 AM
    New Silicon Insider ArticlePaul DeMone2004/10/22 09:38 AM
      New Silicon Insider ArticleDavid Kanter2004/10/22 10:04 PM
        New Silicon Insider ArticlePaul DeMone2004/10/23 05:02 AM
        New Silicon Insider ArticleArun Ramakrishnan2004/10/23 12:47 PM
      New Silicon Insider ArticleMichael S2004/10/23 10:38 AM
        New Silicon Insider ArticlePaul DeMone2004/10/23 12:10 PM
          New Silicon Insider ArticleMichael S2004/10/23 12:40 PM
  Do I see a flaw here?IntelUser20002004/10/27 04:43 PM
    Frequency is different for startersPaul DeMone2004/10/27 05:05 PM
      Frequency is different for startersIntelUser2004/10/27 05:28 PM
        Frequency is different for startersPaul DeMone2004/10/27 06:49 PM
          Frequency is different for startersIntelUser20002004/10/27 07:45 PM
  New Silicon Insider ArticleIntelUser20002004/11/01 09:13 AM
    New Silicon Insider Articleanonymous2004/11/02 08:46 AM
      New Silicon Insider ArticleChuck2004/11/02 09:33 AM
        New Silicon Insider ArticleSingh, S.R.2004/11/02 11:21 AM
          New Silicon Insider ArticleJosé Javier Zarate2004/11/02 02:14 PM
            New Silicon Insider ArticleSingh, S.R.2004/11/02 06:47 PM
        New Silicon Insider Articleanonymous2004/11/03 02:04 PM
          New Silicon Insider Articleanonymous2004/11/03 02:07 PM
  New Silicon Insider ArticleIntelUser20002004/11/04 10:30 AM
    New Silicon Insider ArticleIntelUser20002004/11/04 08:38 PM
      New Silicon Insider ArticleDavid Kanter2004/11/05 07:44 AM
        New Silicon Insider ArticleIntelUser20002004/11/08 07:23 PM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell purple?