f



Madison offically announced

Reported performance is 1322 SPECint_base2k/ 2119 SPECfp_base2k and
for the 1.5 GHz, 6.0 MB L3 version. BTW, TDP max power is 107 W.

--
Paul W. DeMone     The 801 experiment SPARCed an ARMs race of EPIC
Kanata, Ontario    proportions to put more PRECISION and POWER into
pdemone@igs.net    architectures with MIPSed results but ALPHA's well
                   that ends well.




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----
0
Paul
6/30/2003 4:50:33 PM
comp.arch 7611 articles. 0 followers. carchreader (32) is leader. Post Follow

8 Replies
883 Views

Similar Articles

[PageSpeed] 20

In article <3F006A59.EDAE5B8D@igs.net>,
Paul DeMone <pdemone@igs.net> writes:
|> Reported performance is 1322 SPECint_base2k/ 2119 SPECfp_base2k and
|> for the 1.5 GHz, 6.0 MB L3 version. BTW, TDP max power is 107 W.

We won't know until we see which compilers and systems those are
for, but that looks rather better than clock rate for the former
and about clock rate for the latter.  I don't know the tests well
enough to know if the former was expected.

Definitely an impressive juggernaught, if single processor speed
is your main criterion.

You can probably heat your cocoa on it, too :-)


Regards,
Nick Maclaren.
0
nmm1
6/30/2003 5:03:25 PM
In article <bdpqgt$p3n$1@pegasus.csx.cam.ac.uk>, Nick Maclaren <nmm1
@cus.cam.ac.uk> says...
> 
> In article <3F006A59.EDAE5B8D@igs.net>,
> Paul DeMone <pdemone@igs.net> writes:
> |> Reported performance is 1322 SPECint_base2k/ 2119 SPECfp_base2k and
> |> for the 1.5 GHz, 6.0 MB L3 version. BTW, TDP max power is 107 W.
> 
> We won't know until we see which compilers and systems those are
> for, but that looks rather better than clock rate for the former
> and about clock rate for the latter.  I don't know the tests well
> enough to know if the former was expected.

Intel's press release:

  http://www.intel.com/pressroom/archive/releases/20030630comp.htm

answers some of this.  In footnote (5), it says:

  www.spec.org: Itanium� 2 processor result of 1322 on SPECint_base2000*
  measured on HP Server rx2600 using Itanium� 2 processor 6M at 1.5GHz,
  HP-UX operating system and submitted to SPEC. SPECint* is a trademark
  of SPEC*.

  www.spec.org: Itanium� 2 processor result of 2119 on SPECfp_base2000*
  measured on HP Server rx2600 using Itanium� 2 processor 6M at 1.5GHz,
  RedHat Linux AS2.1 operating system and submitted to SPEC. SPECfp* is
  a trademark of SPEC*. 

The complete results will appear on the SPEC web site after the usual 
SPEC review period.

 -- Jim
    HP PA-RISC/Itanium Processor Architect
0
Jim
6/30/2003 7:23:24 PM
In article <MPG.196a2e0ac76fe45598968e@usenet01.boi.hp.com>,
Jim Hull  <jim.hull@hp.com> wrote:
>In article <bdpqgt$p3n$1@pegasus.csx.cam.ac.uk>, Nick Maclaren <nmm1
>@cus.cam.ac.uk> says...
>> In article <3F006A59.EDAE5B8D@igs.net>,
>> Paul DeMone <pdemone@igs.net> writes:
>> |> Reported performance is 1322 SPECint_base2k/ 2119 SPECfp_base2k and
>> |> for the 1.5 GHz, 6.0 MB L3 version. BTW, TDP max power is 107 W.
>> 
>> We won't know until we see which compilers and systems those are
>> for, but that looks rather better than clock rate for the former
>> and about clock rate for the latter.  I don't know the tests well
>> enough to know if the former was expected.
>
>Intel's press release:
>
>  http://www.intel.com/pressroom/archive/releases/20030630comp.htm
>
>answers some of this.  In footnote (5), it says:
>
>  www.spec.org: Itanium� 2 processor result of 1322 on SPECint_base2000*
>  measured on HP Server rx2600 using Itanium� 2 processor 6M at 1.5GHz,
>  HP-UX operating system and submitted to SPEC. SPECint* is a trademark
>  of SPEC*.
>
>  www.spec.org: Itanium� 2 processor result of 2119 on SPECfp_base2000*
>  measured on HP Server rx2600 using Itanium� 2 processor 6M at 1.5GHz,
>  RedHat Linux AS2.1 operating system and submitted to SPEC. SPECfp* is
>  a trademark of SPEC*. 
>
>The complete results will appear on the SPEC web site after the usual 
>SPEC review period.

Interesting.  That is very similar to the slightly bizarre way that
the results were announced for the McKinley.  While this could be
"for compatibility", there could also be deeper reasons.  I won't
bother to ask you, because your response will be "No comment".  If,
at any stage, you can comment, please do so!


Regards,
Nick Maclaren.
0
nmm1
6/30/2003 7:29:26 PM
Paul DeMone (pdemone@igs.net) wrote:
: Reported performance is 1322 SPECint_base2k/ 2119 SPECfp_base2k and
: for the 1.5 GHz, 6.0 MB L3 version. BTW, TDP max power is 107 W.

: --
: Paul W. DeMone     The 801 experiment SPARCed an ARMs race of EPIC
: Kanata, Ontario    proportions to put more PRECISION and POWER into
: pdemone@igs.net    architectures with MIPSed results but ALPHA's well
:                    that ends well.


	     CINT2000     
  Company    Base Peak  System

  HP         1322  --   HP Server rx2600 (1.5GHz Itanium2 6M, HP-UX)
  Intel      1164 1200  Intel D875PBZ motherboard (3.0 GHz Pentium 4)
  Intel      1099 1107  Intel D850EMVR motherboard (3.06 GHz Pentium 4)
  AMD        1095 1170  A4800
  Dell       1089 1138  Precision WorkStation 650 (3.06 GHz Xeon)
  Dell       1085 1130  Precision WorkStation 350 (3.06 GHz P4)
  IBM        1077 1113  IBM eServer pSeries 690 Turbo (1700 MHz, 1 CPU)
  AMD        1044 1080  ASUS A7N8X Motherboard rev. 2.0, AMD AthlonXP 3200+
  Sun        1024 1066  Sun Fire V65x (3.06 GHz Xeon)

	     CFP2000      
  Company    Base Peak  System

  HP         2119  --   HP Server rx2600 (1.5GHz Itanium2 6M, RedHat AS2.1)
  IBM        1598 1699  IBM eServer pSeries 690 Turbo (1700MHz Power4+)
  IBM        1432 1520  IBM eServer pSeries 670 (1500MHz Power4+)
  HP         1431 1431  hp server rx5670 (1GHz Itanium2)
  IBM        1221 1295  IBM eServer pSeries 650 Model 6M2 (1450MHz Power4+)
  Intel      1213 1229  Intel D875PBZ motherboard (3.0 GHz Pentium 4)
  IBM        1202 1266  IBM eServer pSeries 690 Turbo (1.3GHz Power4)
  HP         1151 1151  hp server rx2600 (900MHz Itanium2)
  HP         1124 1482  AlphaServer GS1280 7/1150
  AMD        1122 1219  A4800

--
 Hai  Vo-Ba
 not speaking for HP

0
hai
6/30/2003 8:23:11 PM
Jim Hull wrote:

>   www.spec.org: Itanium� 2 processor result of 1322 on SPECint_base2000*
>   measured on HP Server rx2600 using Itanium� 2 processor 6M at 1.5GHz,
>   HP-UX operating system and submitted to SPEC. SPECint* is a trademark
>   of SPEC*.
> 
>   www.spec.org: Itanium� 2 processor result of 2119 on SPECfp_base2000*
>   measured on HP Server rx2600 using Itanium� 2 processor 6M at 1.5GHz,
>   RedHat Linux AS2.1 operating system and submitted to SPEC.
     ^^^^^^^^^^^^^^^^^^ fascinating - why not HP-UX (as above) ?

[ It can't be "the great Fortran 90 compiler that comes bundled with
   Linux", because it isn't ready yet :-) ]

-- 
Toon Moene - mailto:toon@moene.indiv.nluug.nl - phoneto: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
GNU Fortran 95: http://gcc-g95.sourceforge.net/ (under construction)

0
Toon
6/30/2003 8:37:32 PM
On Mon, 30 Jun 2003 12:50:33 -0400, Paul DeMone <pdemone@igs.net>
wrote:

>Reported performance is 1322 SPECint_base2k/ 2119 SPECfp_base2k and
>for the 1.5 GHz, 6.0 MB L3 version. BTW, TDP max power is 107 W.

Holding my breath to find out how much SPEC voodoo went into this
impressive result.

RM

0
Robert
7/1/2003 3:26:36 AM
Paul DeMone <pdemone@igs.net> wrote in message news:<3F006A59.EDAE5B8D@igs.net>...
> Reported performance is 1322 SPECint_base2k/ 2119 SPECfp_base2k and
> for the 1.5 GHz, 6.0 MB L3 version. BTW, TDP max power is 107 W.
> 
  But is there any changes in microarchitecture
in comparison w/1 Ghz Itanium 2 excluding
increase of cache size ? Or, for example, some changes in delays
(in the numbers of cycles) ?

Mikhail Kuzminsky
Zelinsky Institute of Organic Chemistry
Moscow
0
kus
7/4/2003 5:35:04 PM
In article <b61cf532.0307040935.5ec88233@posting.google.com>,
Mikhail Kuzminsky <kus@free.net> wrote:
>Paul DeMone <pdemone@igs.net> wrote in message news:<3F006A59.EDAE5B8D@igs.net>...
>> Reported performance is 1322 SPECint_base2k/ 2119 SPECfp_base2k and
>> for the 1.5 GHz, 6.0 MB L3 version. BTW, TDP max power is 107 W.
>> 
>  But is there any changes in microarchitecture
>in comparison w/1 Ghz Itanium 2 excluding
>increase of cache size ? Or, for example, some changes in delays
>(in the numbers of cycles) ?

Almost certainly, but quite likely nothing more than that.  On
the other hand, there has been a considerable speed up, and it is
possible that many of the bottleneck areas have been redesigned.
I just don't know.

When the original Madison project was cancelled, something had to
be hastily cobbled up to replace it.  And that was a speeded up
and much expanded McKinley.  That can be seen by the fact that
there are NO specification changes in the programmers' manuals.

However, that doesn't prove anything about the aspects which you
are asking about.


Regards,
Nick Maclaren.
0
nmm1
7/4/2003 7:20:37 PM
Reply: