Why OS X Sucks

Discussion in 'NZ Computing' started by Lawrence D¹Oliveiro, Aug 1, 2005.

  1. So, did anybody else see this Anandtech review
    <http://www.anandtech.com/mac/showdoc.aspx?i=2436>? While the reviewer
    was impressed with Apple's PowerPC G5 hardware, the OS X software came
    out rather less well.

    The problem is that its multithreading performance is appallingly bad.
    And the blame seems to lie squarely with the Mach microkernel. The odd
    thing is, this problem has been known for years, right since the
    beginning of OS X <http://developer.apple.com/technotes/tn/tn2028.html>,
    yet Apple hasn't fixed it. Either because they won't, or perhaps because
    they can't.
     
    Lawrence D¹Oliveiro, Aug 1, 2005
    #1
    1. Advertising

  2. Lawrence D¹Oliveiro

    Mercury Guest

    old news. I.E. yes.

    IMHO "easily" fixed too.

    "Lawrence D¹Oliveiro" <_zealand> wrote in message
    news:...
    > So, did anybody else see this Anandtech review
    > <http://www.anandtech.com/mac/showdoc.aspx?i=2436>? While the reviewer
    > was impressed with Apple's PowerPC G5 hardware, the OS X software came
    > out rather less well.
    >
    > The problem is that its multithreading performance is appallingly bad.
    > And the blame seems to lie squarely with the Mach microkernel. The odd
    > thing is, this problem has been known for years, right since the
    > beginning of OS X <http://developer.apple.com/technotes/tn/tn2028.html>,
    > yet Apple hasn't fixed it. Either because they won't, or perhaps because
    > they can't.
     
    Mercury, Aug 2, 2005
    #2
    1. Advertising

  3. In article <>,
    Lawrence D¹Oliveiro <_zealand> wrote:

    > The problem is that its multithreading performance is appallingly bad.
    > And the blame seems to lie squarely with the Mach microkernel. The odd
    > thing is, this problem has been known for years, right since the
    > beginning of OS X <http://developer.apple.com/technotes/tn/tn2028.html>,
    > yet Apple hasn't fixed it. Either because they won't, or perhaps because
    > they can't.


    In what circumstances would the average Mac user note this appalling
    deficiency?

    --

    Chas Right
     
    Charlie Right, Aug 2, 2005
    #3
  4. Lawrence D¹Oliveiro

    ofn01 Guest

    Charlie Right wrote:
    > In article <>,
    > Lawrence D¹Oliveiro <_zealand> wrote:
    >
    >
    >>The problem is that its multithreading performance is appallingly bad.
    >>And the blame seems to lie squarely with the Mach microkernel. The odd
    >>thing is, this problem has been known for years, right since the
    >>beginning of OS X <http://developer.apple.com/technotes/tn/tn2028.html>,
    >>yet Apple hasn't fixed it. Either because they won't, or perhaps because
    >>they can't.

    >
    >
    > In what circumstances would the average Mac user note this appalling
    > deficiency?
    >


    While loading up fireworks and at the same time taking the shrinkwrap
    off a new dreamweaver box.
     
    ofn01, Aug 2, 2005
    #4
  5. Lawrence D¹Oliveiro

    AD. Guest

    On Tue, 02 Aug 2005 19:41:16 +1200, Charlie Right wrote:

    > In article <>,
    > Lawrence D¹Oliveiro <_zealand> wrote:
    >
    >> The problem is that its multithreading performance is appallingly bad.
    >> And the blame seems to lie squarely with the Mach microkernel. The odd
    >> thing is, this problem has been known for years, right since the
    >> beginning of OS X <http://developer.apple.com/technotes/tn/tn2028.html>,
    >> yet Apple hasn't fixed it. Either because they won't, or perhaps because
    >> they can't.

    >
    > In what circumstances would the average Mac user note this appalling
    > deficiency?


    Running a server. The issues are less apparent for workstation type loads.
    Or at least that's the case if it's the article I think it is ;)

    --
    Cheers
    Anton
     
    AD., Aug 2, 2005
    #5
  6. In article <>,
    Charlie Right <> wrote:

    >In article <>,
    > Lawrence D¹Oliveiro <_zealand> wrote:
    >
    >> The problem is that its multithreading performance is appallingly bad.
    >> And the blame seems to lie squarely with the Mach microkernel. The odd
    >> thing is, this problem has been known for years, right since the
    >> beginning of OS X <http://developer.apple.com/technotes/tn/tn2028.html>,
    >> yet Apple hasn't fixed it. Either because they won't, or perhaps because
    >> they can't.

    >
    >In what circumstances would the average Mac user note this appalling
    >deficiency?


    How about the one where they run it on an upcoming multicore Intel
    processor, and discover that the performance improvement is precisely
    zero?
     
    Lawrence D¹Oliveiro, Aug 3, 2005
    #6
  7. In article <>,
    "AD." <> wrote:

    >On Tue, 02 Aug 2005 19:41:16 +1200, Charlie Right wrote:
    >
    >> In article <>,
    >> Lawrence D¹Oliveiro <_zealand> wrote:
    >>
    >>> The problem is that its multithreading performance is appallingly bad.
    >>> And the blame seems to lie squarely with the Mach microkernel. The odd
    >>> thing is, this problem has been known for years, right since the
    >>> beginning of OS X <http://developer.apple.com/technotes/tn/tn2028.html>,
    >>> yet Apple hasn't fixed it. Either because they won't, or perhaps because
    >>> they can't.

    >>
    >> In what circumstances would the average Mac user note this appalling
    >> deficiency?

    >
    >Running a server. The issues are less apparent for workstation type loads.


    I expect this is likely to change in the near future. With the
    increasing emphasis on multicore processors in lieu of further clock
    speed increases, multithreading performance is going to have to become
    important on the desktop too.
     
    Lawrence D¹Oliveiro, Aug 4, 2005
    #7
  8. Lawrence D¹Oliveiro

    AD. Guest

    On Fri, 05 Aug 2005 09:47:54 +1200, Lawrence D¹Oliveiro wrote:

    >>Running a server. The issues are less apparent for workstation type
    >>loads.

    >
    > I expect this is likely to change in the near future. With the increasing
    > emphasis on multicore processors in lieu of further clock speed increases,
    > multithreading performance is going to have to become important on the
    > desktop too.


    You're right, but I reckon a little hasty. I think it will take a bit
    longer for multithreading to sink in as a 'mainstream' performance
    concern. Users are still going to keep getting faster chips and probably
    won't notice for a while that they are missing out.

    Maybe Apple should've gone with BeOS rather than Next after all? ;)

    BeOS was pretty handy with multithreaded stuff.

    --
    Cheers
    Anton
     
    AD., Aug 4, 2005
    #8
  9. In article <1123192988.3a5c5960d2f1ff0e54042a8974760884@teranews>,
    "AD." <> wrote:

    >On Fri, 05 Aug 2005 09:47:54 +1200, Lawrence D¹Oliveiro wrote:
    >
    >>>Running a server. The issues are less apparent for workstation type
    >>>loads.

    >>
    >> I expect this is likely to change in the near future. With the increasing
    >> emphasis on multicore processors in lieu of further clock speed increases,
    >> multithreading performance is going to have to become important on the
    >> desktop too.

    >
    >You're right, but I reckon a little hasty. I think it will take a bit
    >longer for multithreading to sink in as a 'mainstream' performance
    >concern.


    "Mainstream" is not "leading-edge". Is 64-bit hardware "mainstream"? Not
    in the closed-source world, because the software is taking so long to
    come out. But in the open-source world it's already commonplace, with
    software to match.

    Same thing with multicore--it's just another packaging of
    multiprocessor, which Linux has been doing for years.
     
    Lawrence D¹Oliveiro, Aug 5, 2005
    #9
  10. Lawrence D¹Oliveiro

    AD. Guest

    On Fri, 05 Aug 2005 19:02:43 +1200, Lawrence D¹Oliveiro wrote:

    > In article <1123192988.3a5c5960d2f1ff0e54042a8974760884@teranews>,
    > "AD." <> wrote:
    >
    >>On Fri, 05 Aug 2005 09:47:54 +1200, Lawrence D¹Oliveiro wrote:
    >>
    >>>>Running a server. The issues are less apparent for workstation type
    >>>>loads.
    >>>
    >>> I expect this is likely to change in the near future. With the
    >>> increasing emphasis on multicore processors in lieu of further clock
    >>> speed increases, multithreading performance is going to have to become
    >>> important on the desktop too.

    >>
    >>You're right, but I reckon a little hasty. I think it will take a bit
    >>longer for multithreading to sink in as a 'mainstream' performance
    >>concern.

    >
    > "Mainstream" is not "leading-edge". Is 64-bit hardware "mainstream"? Not
    > in the closed-source world, because the software is taking so long to come
    > out. But in the open-source world it's already commonplace, with software
    > to match.


    So when you said that "multithreading performance is going to have to
    become important on the desktop too", you weren't talking about mainstream
    stuff? How could something be important on the desktop without being
    mainstream?

    --
    Cheers
    Anton
     
    AD., Aug 7, 2005
    #10
  11. In article <1123451848.b18d5c8f5d841c974328a04cab714912@teranews>,
    "AD." <> wrote:

    >On Fri, 05 Aug 2005 19:02:43 +1200, Lawrence D¹Oliveiro wrote:
    >
    >> In article <1123192988.3a5c5960d2f1ff0e54042a8974760884@teranews>,
    >> "AD." <> wrote:
    >>
    >>>On Fri, 05 Aug 2005 09:47:54 +1200, Lawrence D¹Oliveiro wrote:
    >>>
    >>>>>Running a server. The issues are less apparent for workstation type
    >>>>>loads.
    >>>>
    >>>> I expect this is likely to change in the near future. With the
    >>>> increasing emphasis on multicore processors in lieu of further clock
    >>>> speed increases, multithreading performance is going to have to become
    >>>> important on the desktop too.
    >>>
    >>>You're right, but I reckon a little hasty. I think it will take a bit
    >>>longer for multithreading to sink in as a 'mainstream' performance
    >>>concern.

    >>
    >> "Mainstream" is not "leading-edge". Is 64-bit hardware "mainstream"? Not
    >> in the closed-source world, because the software is taking so long to come
    >> out. But in the open-source world it's already commonplace, with software
    >> to match.

    >
    >So when you said that "multithreading performance is going to have to
    >become important on the desktop too", you weren't talking about mainstream
    >stuff? How could something be important on the desktop without being
    >mainstream?


    Look at 64-bit. Is it "mainstream" yet? Not really. But is it
    "important"? Most certainly yes.

    The same will be true shortly of multicore.
     
    Lawrence D¹Oliveiro, Aug 20, 2005
    #11
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Thepcman

    why adelphia sucks?

    Thepcman, Feb 6, 2004, in forum: Computer Support
    Replies:
    1
    Views:
    645
  2. janquin

    why norton sucks.????

    janquin, Sep 19, 2004, in forum: Computer Support
    Replies:
    28
    Views:
    11,180
    janquin
    Sep 20, 2004
  3. Gactimus

    Give me 1 good reason why AOL sucks

    Gactimus, Aug 17, 2004, in forum: Digital Photography
    Replies:
    0
    Views:
    324
    Gactimus
    Aug 17, 2004
  4. Charles Bronson aka The Mechanic

    Kodak digital sucks ..... here is why

    Charles Bronson aka The Mechanic, Feb 3, 2005, in forum: Digital Photography
    Replies:
    11
    Views:
    403
    Keith Holley
    Feb 8, 2005
  5. Why Your Life Sucks!

    , Jun 3, 2007, in forum: Computer Support
    Replies:
    3
    Views:
    457
    =?ISO-8859-1?Q?R=F4g=EAr?=
    Jun 3, 2007
Loading...

Share This Page