New Silicon Insider Article Online

Article: The Battle in 64 bit Land: Merchant Chips on the Rise
By: mas (mas769.delete@this.hotmail.com), May 3, 2004 12:31 pm
Room: Moderated Discussions
Paul DeMone (pdemone@igs.net) on 5/3/04 wrote:
---------------------------
>mas () on 5/3/04 wrote:
>---------------------------
>[..]
>>Quite good in general with good historical analysis. However Sun's future (Serverworks)
>>8/16/32 way Opterons not mentioned and their possible impact on the high-end.
>
>I am sceptical about the time scale of such an impact
>on that end of the market and Sun's remaining influence
>by that time.
>
>>2.5
>>Ghz would be a large (47%) increase for Montecito and
>
>Compared to what? A 1.5 GHz Madison? Or a 1.7+ GHz
>Madison? ;-)

Mind you, you are being pessimistic compared to what HP are now saying. :-)

"Performance: Itanium is more efficient in the utilization of clock cycles. Over the next few years, Itanium will exceed x86 performance by as much as 2X. This will come from two areas: Today, an Itanium 1.5GHz processor has integer performance equivalent to a 3GHz Xeon - approximately twice the performance per clock tick. Design and manufacturing improvements are expected to allow the Itanium clock rate to equal that of Xeon, thus creating a wide gap in performance. In addition, improvements in compilers will further increase the effective performance yield of EPIC."

http://www.hp.com/products1/evolution/alpha_retaintrust/openvms/faqs2.html


>
>> sounds optimistic but it could
>>happen I suppose but is less than the Power increase (32%) and makes me wonder how
>>these figures were derived Also Power 5's performance stats look a bit low compared
>>to the noises IBM have been making.
>
>Keep in mind that a lot of the noise comes from the
>fact that POWER5 allows systems based on 8 MCMs,
>an improvement from the 4 MCM limit of POWER4x.
>
>>Perhaps they may need updating in light of its immediate release ?
>
>I never update predictions in my articles. They stand as
>an indication of my thinking at the time of publishing.
>I also hate revisionist/retroactive prophets so I try to
>avoid being one myself.
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
New Silicon Insider Article OnlineDavid Kanter2004/05/03 12:25 AM
  TDP questionSingh, S.R.2004/05/03 06:18 AM
    TDP questionWouter Tinus2004/05/03 08:11 AM
      TDP questionanonymous2004/05/03 09:12 AM
        TDP questionWouter Tinus2004/05/03 09:51 AM
          TDP questionanonymus2004/05/04 08:26 AM
    TDP questionPaul DeMone2004/05/03 08:16 AM
      TDP questionAlberto2004/05/03 09:16 AM
        TDP questionPaul DeMone2004/05/03 11:11 AM
          TDP questionAlberto2004/05/03 05:25 PM
      TDP questionMaarten2004/05/04 05:22 AM
        TDP questionPaul DeMone2004/05/04 07:54 AM
  New Silicon Insider Article OnlineBob Koz2004/05/03 08:58 AM
    New Silicon Insider Article OnlinePaul DeMone2004/05/03 09:21 AM
      Does your 'evidence' include backstabing corps?Interested in Illinois2004/05/03 10:00 AM
      Does your 'evidence' include backstabing corps? (NT)Interested in Illinois2004/05/03 10:00 AM
        Does your medication cost alot?waitressInGaza2004/05/03 11:16 AM
          I don't need that poison. We have you. (NT)Interested in Illinois2004/05/03 11:21 PM
            I think this can stop now (NT)David Kanter2004/05/04 01:57 AM
    New Silicon Insider Article OnlineDaniel2004/05/03 02:34 PM
  core size -- x86 vs. IPFanonymous2004/05/03 09:02 AM
    core size -- x86 vs. IPFPaul DeMone2004/05/03 09:24 AM
      core size -- x86 vs. IPFanonymus2004/05/04 08:35 AM
  Corrections aboundInterested in Illinois2004/05/03 09:19 AM
    StalkingDean Kent2004/05/03 01:20 PM
      StalkingSalem2004/05/05 07:28 PM
        More irrational people. :-)Dean Kent2004/05/05 07:48 PM
          More irrational people. :-)Interested in Illinois2004/05/05 10:07 PM
            More irrational people. :-)Heath Provost2004/05/05 10:27 PM
            More irrational people. :-)aap2004/05/06 12:27 AM
              More irrational people. :-)Dean Kent2004/05/06 01:45 AM
            More irrational people. :-)Dean Kent2004/05/06 01:38 AM
              More irrational people. :-)Interested in Illinois2004/05/06 06:32 AM
                More irrational people. :-)Beelzebubba92004/05/06 07:49 AM
                http://www.sacklunch.net/Latin/V/venividivici.html (NT)mas2004/05/07 04:56 PM
              More irrational people. :-)Beelzebubba92004/05/06 07:41 AM
            More funBeelzebubba92004/05/06 07:33 AM
              OT: Even More Funhobold2004/05/07 04:56 AM
                OT: Even More FunPaul DeMone2004/05/07 09:10 AM
                  OT: Even More Funhobold2004/05/08 11:59 AM
  New Silicon Insider Article Onlinemas2004/05/03 10:44 AM
    New Silicon Insider Article OnlinePaul DeMone2004/05/03 11:02 AM
      New Silicon Insider Article Onlinemas2004/05/03 11:15 AM
        New Silicon Insider Article OnlinePaul DeMone2004/05/03 11:24 AM
      New Silicon Insider Article Onlinemas2004/05/03 12:31 PM
        New Silicon Insider Article OnlineDavid Kanter2004/05/03 11:16 PM
      New Silicon Insider Article Onlinemas2004/05/03 02:53 PM
        New Silicon Insider Article OnlinePaul DeMone2004/05/03 03:03 PM
          New Silicon Insider Article OnlineMarc M.2004/05/03 10:04 PM
  Opteron CMP performance estimation?Wouter Tinus2004/05/03 12:21 PM
    Opteron CMP performance estimation?Paul DeMone2004/05/03 12:59 PM
      But your CMP scaling is signficantly worseIlleglWpns2004/05/03 02:21 PM
        But your CMP scaling is signficantly worsePaul DeMone2004/05/03 02:56 PM
          Check which clock rate?IlleglWpns2004/05/03 03:41 PM
            Check which clock rate?Paul DeMone2004/05/03 03:49 PM
              You are right, my badIlleglWpns2004/05/03 03:58 PM
  Opterons with more cacheDresdenboy2004/05/04 01:44 AM
    Opterons with more cachePaul DeMone2004/05/04 08:09 AM
      Opterons with more cacheDoug Siebert2004/05/05 04:59 PM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell purple?