Professional Web Applications Themes

Plea to SPARC laptop designers - Sun Solaris

>>>>> "Chris" == Chris Morgan <cmmihalis.net> writes: Chris> Raymond Toy <toyrtp.ericsson.se> writes: >> >>>>> "Mark" == Mark Gibson <gibsonbluestem.prairienet.org> writes: >> Mark> SPARC architecture is flwawed in many ways, but it does run fast for >> >> Flawed in what ways? Chris> still doesn't do out-of-order execution, still doesn't scale its Chris> clockspeed as well as intel or amd chips, lower bandwidth interface to These seem to be implementation issues, not architecture issues. And the mythical sufficiently smart compiler would make out-of-order execution not an issue. But perhaps I'm confusing architecture with implementation. Ray...

  1. #1

    Default Re: Plea to SPARC laptop designers

    >>>>> "Chris" == Chris Morgan <cmmihalis.net> writes:

    Chris> Raymond Toy <toyrtp.ericsson.se> writes:
    >> >>>>> "Mark" == Mark Gibson <gibsonbluestem.prairienet.org> writes:
    >>
    Mark> SPARC architecture is flwawed in many ways, but it does run fast for
    >>
    >> Flawed in what ways?
    Chris> still doesn't do out-of-order execution, still doesn't scale its
    Chris> clockspeed as well as intel or amd chips, lower bandwidth interface to

    These seem to be implementation issues, not architecture issues. And
    the mythical sufficiently smart compiler would make out-of-order
    execution not an issue.

    But perhaps I'm confusing architecture with implementation.

    Ray
    Raymond Toy Guest

  2. #2

    Default Re: Plea to SPARC laptop designers


    Raymond Toy <toyrtp.ericsson.se> writes:
    > >>>>> "Chris" == Chris Morgan <cmmihalis.net> writes:
    >
    > Chris> Raymond Toy <toyrtp.ericsson.se> writes:
    > >> >>>>> "Mark" == Mark Gibson <gibsonbluestem.prairienet.org> writes:
    > >>
    > Mark> SPARC architecture is flwawed in many ways, but it does run fast for
    > >>
    > >> Flawed in what ways?
    >
    > Chris> still doesn't do out-of-order execution, still doesn't scale its
    > Chris> clockspeed as well as intel or amd chips, lower bandwidth interface to
    >
    > These seem to be implementation issues, not architecture issues. And
    > the mythical sufficiently smart compiler would make out-of-order
    > execution not an issue.
    >
    > But perhaps I'm confusing architecture with implementation.
    Not that I know anything about it, but I compare Sparc to
    all the other high-performance CPUs that TI fabs... oh, yeah...
    TI _doesn't_ fab any other high-performance CPUs, do they?

    Perhaps that's a clue.
    -SEan
    Sean Burke Guest

  3. #3

    Default Re: Plea to SPARC laptop designers

    In article <x7fzlzjik9.fsfbolo.xenadyne.com>, Sean Burke wrote:
    >
    > Raymond Toy <toyrtp.ericsson.se> writes:
    >
    >> >>>>> "Chris" == Chris Morgan <cmmihalis.net> writes:
    >>
    >> Chris> Raymond Toy <toyrtp.ericsson.se> writes:
    >> >> >>>>> "Mark" == Mark Gibson <gibsonbluestem.prairienet.org> writes:
    >> >>
    >> Mark> SPARC architecture is flwawed in many ways, but it does run fast for
    >> >>
    >> >> Flawed in what ways?
    >>
    >> Chris> still doesn't do out-of-order execution, still doesn't scale its
    >> Chris> clockspeed as well as intel or amd chips, lower bandwidth interface to
    >>
    >> These seem to be implementation issues, not architecture issues. And
    >> the mythical sufficiently smart compiler would make out-of-order
    >> execution not an issue.
    >>
    >> But perhaps I'm confusing architecture with implementation.
    >
    > Not that I know anything about it, but I compare Sparc to
    > all the other high-performance CPUs that TI fabs... oh, yeah...
    > TI _doesn't_ fab any other high-performance CPUs, do they?
    No, just the majority of high-performance DSP (Digital Signal
    Processing) chips. Similar speed & process requirements.
    > Perhaps that's a clue.
    Indeed.
    Steve Kappel Guest

  4. #4

    Default Re: Plea to SPARC laptop designers

    Raymond Toy wrote:
    >
    > >>>>> "Chris" == Chris Morgan <cmmihalis.net> writes:
    >
    > Chris> still doesn't do out-of-order execution, still doesn't scale its
    >
    > These seem to be implementation issues, not architecture issues. And
    > the mythical sufficiently smart compiler would make out-of-order
    > execution not an issue.
    Ah, out-of-order execution. I can just see it now - branch and
    then test.

    -am © 2003
    Anthony Mandic Guest

  5. #5

    Default Re: Plea to SPARC laptop designers

    Chris Morgan wrote:
    > Raymond Toy <toyrtp.ericsson.se> writes:
    >
    >
    >>>>>>>"Mark" == Mark Gibson <gibsonbluestem.prairienet.org> writes:
    >>
    >> Mark> SPARC architecture is flwawed in many ways, but it does run fast for
    >>
    >>Flawed in what ways?
    >
    >
    > still doesn't do out-of-order execution
    SPARC64 built by Fujitsu is doing out-of-order execution.
    > lower bandwidth interface to cache memory
    SPARC64-V has got 2MB cache on-chip. Cache access is faster than on
    any x86-chip I know of.

    Juergen

    --
    Juergen Gross Telephone: +49 (0) 89 636 47950
    Fujitsu Siemens Computers Telefax: +49 (0) 89 636 40638
    Otto-Hahn-Ring 6 e-mail: [email]juergen.grossfujitsu-siemens.com[/email]
    81739 Muenchen, Germany Internet: [url]http://www.fujitsu-siemens.com[/url]

    Jürgen Groß Guest

  6. #6

    Default Re: Plea to SPARC laptop designers

    Jürgen Groß <juergen.grossfujitsu-siemens.com> writes:
    > SPARC64 built by Fujitsu is doing out-of-order execution.
    Yes, and SPARC64 is outperforming Sun's current SPARC offerings on
    many benchmarks. For example, in the SPECint_rate_base2000 benchmark,
    an 8-processor PRIMEPOWER850 (1.35 GHz) scores 67.3, whereas an
    8-processor Sun Fire V800 (1.05 GHz) scores only 49.1 (though you can
    boost this to 54.5 if you spring for an 8-processor Sun Fire 4800 at
    1.2 GHz). To be fair, the Fujtsu-Siemens box isn't available until
    next month whereas the Sun boxes are available now, but still one has
    to wonder.

    The 8-processor SGI Altix 3000 (1.5 GHz Itanium 2, available now)
    scores 98.3 on the same benchmark, so Sun and Fujitsu are both well
    behind the GNU/Linux competition here.
    Paul Eggert Guest

  7. #7

    Default Re: Plea to SPARC laptop designers

    On Wed, 2 Jul 2003, Paul Eggert wrote:
    > Jürgen Groß <juergen.grossfujitsu-siemens.com> writes:
    >
    > > SPARC64 built by Fujitsu is doing out-of-order execution.
    >
    > Yes, and SPARC64 is outperforming Sun's current SPARC offerings on
    > many benchmarks. For example, in the SPECint_rate_base2000 benchmark,
    > an 8-processor PRIMEPOWER850 (1.35 GHz) scores 67.3, whereas an
    > 8-processor Sun Fire V800 (1.05 GHz) scores only 49.1 (though you can
    > boost this to 54.5 if you spring for an 8-processor Sun Fire 4800 at
    > 1.2 GHz). To be fair, the Fujtsu-Siemens box isn't available until
    > next month whereas the Sun boxes are available now, but still one has
    > to wonder.
    >
    > The 8-processor SGI Altix 3000 (1.5 GHz Itanium 2, available now)
    > scores 98.3 on the same benchmark, so Sun and Fujitsu are both well
    > behind the GNU/Linux competition here.
    >
    Interesting laptops you guys discuss, 8 way SGI's and V880's

    Not that I'm really that into benchmarks, but its fun to see that as
    per usual, all you get in theese discussions is the numbers that work
    for the current speaker, SPECint, what does it really tell?

    Only true benchmark is to try the system with the important app and see
    what outperformes what, and then cut cost into it, not just initial but
    the total cost of ownership into account to get real numbers.

    Nothing special against or for any of the systems here, just think this is
    quite a funny sidethread to laptops...

    /Johan A
    Mr. Johan Andersson Guest

Similar Threads

  1. Yet another plea for feedback
    By moviePig in forum Macromedia Flash
    Replies: 1
    Last Post: May 8th, 01:38 PM
  2. Simple Plea :P
    By pull your pants up in forum Macromedia Flash
    Replies: 1
    Last Post: September 24th, 06:29 AM
  3. Plea for challenge consideration
    By Grant Dixon in forum Adobe Photoshop Elements
    Replies: 0
    Last Post: August 31st, 05:57 PM
  4. Client script opensource plea
    By Larry Brown in forum PHP Development
    Replies: 0
    Last Post: July 14th, 01:42 AM
  5. 64-bit Sparc laptop
    By John P Davis in forum Linux / Unix Administration
    Replies: 9
    Last Post: July 12th, 01:32 PM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139