Benchmark: XP32 vs XP64 - Is there really any benefit to x64 yet?

Discussion in 'Windows 64bit' started by Will, Nov 20, 2005.

  1. Will

    Will Guest

    Hi,

    I've run a 3D benchmark on my PC, first with XP 32 and then with x64. I used
    3DMark 2005 to run the test.

    The overall scores are as follows

    Windows XP 32: 5630
    Windows XP 64: 5525


    I was shocked that the performance is acctually worse on x64.

    So this has left me scratching my head thinking what are the real benefits
    of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with my 2GB.

    How long will it be before developers other than MSFT start writing 64bit
    code? 18 months - longer?


    Are we running Windows XP 64 just for the sake of it?


    Will
     
    Will, Nov 20, 2005
    #1
    1. Advertising

  2. Will

    McGrandpa Guest

    "Will" <> wrote in message
    news:43805865$0$203$...
    > Hi,
    >
    > I've run a 3D benchmark on my PC, first with XP 32 and then with x64. I
    > used 3DMark 2005 to run the test.
    >
    > The overall scores are as follows
    >
    > Windows XP 32: 5630
    > Windows XP 64: 5525
    >
    >
    > I was shocked that the performance is acctually worse on x64.
    >
    > So this has left me scratching my head thinking what are the real benefits
    > of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with my
    > 2GB.
    >
    > How long will it be before developers other than MSFT start writing 64bit
    > code? 18 months - longer?
    >
    >
    > Are we running Windows XP 64 just for the sake of it?
    >
    >
    > Will
    >

    Yaknow, I tried the same thing you did there. And for my expectations, I
    was very disappointed... in Futuremarks software. I'm running an A64 X2
    4800+ in this machine, and in the old one I have a P4 3.0E (Prescott, 1m L2,
    800 fsb). And it gets better 3DM05 scores than the X2 in XP Pro 32 bit.
    Score in x64 is several hundred points less for the X2! Wonder why....
    single core at 2.4gHz, HT is off, and it's 32 bit software. The X2 slaps
    the P4 silly in everything I run. I think I will ignore those benchmarks
    and just enjoy using one of the fastest CPU's on the planet.
    McG.
     
    McGrandpa, Nov 20, 2005
    #2
    1. Advertising

  3. On Sun, 20 Nov 2005 11:05:10 -0000, Will <> wrote:

    > Hi,
    >
    > I've run a 3D benchmark on my PC, first with XP 32 and then with x64. I
    > used
    > 3DMark 2005 to run the test.
    >
    > The overall scores are as follows
    >
    > Windows XP 32: 5630
    > Windows XP 64: 5525
    >
    >
    > I was shocked that the performance is acctually worse on x64.
    >
    > So this has left me scratching my head thinking what are the real
    > benefits
    > of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with my
    > 2GB.
    >
    > How long will it be before developers other than MSFT start writing 64bit
    > code? 18 months - longer?
    >
    >
    > Are we running Windows XP 64 just for the sake of it?
    >
    >
    > Will
    >
    >


    WOW64 must involve some overhead, so until your benchmark becomes 64bit,
    and aware of the extra registers etc, it's going to be marginally slower.
     
    Mark Gillespie, Nov 20, 2005
    #3
  4. Will

    Will Guest

    "Mark Gillespie" <> wrote in message
    news:eek:p.s0jfg606j2q9yy@xpdesktop64...
    > On Sun, 20 Nov 2005 11:05:10 -0000, Will <> wrote:
    >
    >> Hi,
    >>
    >> I've run a 3D benchmark on my PC, first with XP 32 and then with x64. I
    >> used
    >> 3DMark 2005 to run the test.
    >>
    >> The overall scores are as follows
    >>
    >> Windows XP 32: 5630
    >> Windows XP 64: 5525
    >>
    >>
    >> I was shocked that the performance is acctually worse on x64.
    >>
    >> So this has left me scratching my head thinking what are the real
    >> benefits
    >> of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with my
    >> 2GB.
    >>
    >> How long will it be before developers other than MSFT start writing 64bit
    >> code? 18 months - longer?
    >>
    >>
    >> Are we running Windows XP 64 just for the sake of it?
    >>
    >>
    >> Will
    >>
    >>

    >
    > WOW64 must involve some overhead, so until your benchmark becomes 64bit,
    > and aware of the extra registers etc, it's going to be marginally slower.


    Hi Mark,

    Sure, that makes sense.

    So we should assume to expect lower performance on x64 in general for apps
    which are not 64bit. At this time I'd say thats > 95% of apps.

    So I think my point still stands - XP32 is faster (at least for now).


    Will
     
    Will, Nov 20, 2005
    #4
  5. In this case the benchmark (SiSoft) is a 64 bit program. At least it has
    been COMPILED to run on a 64 bit processor. Whether the applets in the
    software have been tweaked to take advantage of 64 bit processing is a
    matter for smarter minds than mine ! I have a 3800+ with a X800 series video
    card modded with an ATI Silencer 5 and a WD740 so as Will says, I will sit
    back and enjoy one of the fastest machines around !

    >
    > WOW64 must involve some overhead, so until your benchmark becomes 64bit,
    > and aware of the extra registers etc, it's going to be marginally slower.
     
    Norman Brooks, Nov 20, 2005
    #5
  6. Will

    Will Guest

    "Norman Brooks" <> wrote in message
    news:OvD$...
    > In this case the benchmark (SiSoft) is a 64 bit program. At least it has
    > been COMPILED to run on a 64 bit processor. Whether the applets in the
    > software have been tweaked to take advantage of 64 bit processing is a
    > matter for smarter minds than mine ! I have a 3800+ with a X800 series
    > video card modded with an ATI Silencer 5 and a WD740 so as Will says, I
    > will sit back and enjoy one of the fastest machines around !
    >


    Hey I didn't say that :) But i agree, nice to have a fast PC - I've just
    ordered the XFX NVIDIA 7800 512mb so I'm hoping that will help to pump up
    the 3D score.
     
    Will, Nov 20, 2005
    #6
  7. Re: Benchmark: XP32 vs XP64 - Is there really any benefit to x64 y

    With Gainward 7800GTX 256 MB I scored 8008 in 3DMarks 2005. Let me know if it
    is worth to upgrade to a 512 MB.

    http://web.telia.com/~u86514936/3DMark0305.PNG

    --
    Asus P5WD2 Premium (955X Express)
    P4 3.73 GHz EE
    RAPTOR 2 x 74GB in RAID 0
    DDR2 PC8000 2GB
    GeForce 7800GTX
    Samsung 24" TFT
    Enermax 660W
    Plextor PX-712SA


    "Will" wrote:

    >
    > "Norman Brooks" <> wrote in message
    > news:OvD$...
    > > In this case the benchmark (SiSoft) is a 64 bit program. At least it has
    > > been COMPILED to run on a 64 bit processor. Whether the applets in the
    > > software have been tweaked to take advantage of 64 bit processing is a
    > > matter for smarter minds than mine ! I have a 3800+ with a X800 series
    > > video card modded with an ATI Silencer 5 and a WD740 so as Will says, I
    > > will sit back and enjoy one of the fastest machines around !
    > >

    >
    > Hey I didn't say that :) But i agree, nice to have a fast PC - I've just
    > ordered the XFX NVIDIA 7800 512mb so I'm hoping that will help to pump up
    > the 3D score.
    >
    >
    >
     
    =?Utf-8?B?TWFyYXRvbm1hbm5lbg==?=, Nov 20, 2005
    #7
  8. Will

    Will Guest

    Re: Benchmark: XP32 vs XP64 - Is there really any benefit to x64 y

    Sure thing, i will post the result. But you'll have to watch this thread as
    it won't arrive until 28th november.

    I expect (and hope) there to be a decent increase in performance from my
    current 6800 Ultra.


    Will

    "Maratonmannen" <> wrote in message
    news:...
    > With Gainward 7800GTX 256 MB I scored 8008 in 3DMarks 2005. Let me know if
    > it
    > is worth to upgrade to a 512 MB.
    >
    > http://web.telia.com/~u86514936/3DMark0305.PNG
    >
    > --
    > Asus P5WD2 Premium (955X Express)
    > P4 3.73 GHz EE
    > RAPTOR 2 x 74GB in RAID 0
    > DDR2 PC8000 2GB
    > GeForce 7800GTX
    > Samsung 24" TFT
    > Enermax 660W
    > Plextor PX-712SA
    >
    >
    > "Will" wrote:
    >
    >>
    >> "Norman Brooks" <> wrote in message
    >> news:OvD$...
    >> > In this case the benchmark (SiSoft) is a 64 bit program. At least it
    >> > has
    >> > been COMPILED to run on a 64 bit processor. Whether the applets in the
    >> > software have been tweaked to take advantage of 64 bit processing is a
    >> > matter for smarter minds than mine ! I have a 3800+ with a X800 series
    >> > video card modded with an ATI Silencer 5 and a WD740 so as Will says, I
    >> > will sit back and enjoy one of the fastest machines around !
    >> >

    >>
    >> Hey I didn't say that :) But i agree, nice to have a fast PC - I've
    >> just
    >> ordered the XFX NVIDIA 7800 512mb so I'm hoping that will help to pump up
    >> the 3D score.
    >>
    >>
    >>
     
    Will, Nov 20, 2005
    #8
  9. I'd have to say that it simply shows that benchmarks are interesting, but
    don't reflect what I see on my machines. Not suprising. Now it's certainly
    true that I don't think MS Word runs one bit faster in x64 Edition. But
    then, I didn't expect it to. And since my overall machine seems to run
    somewhat faster, I'm happy. As we see new applications that are native
    64-bit, we'll see the whole thing shift. I'm not surprised that it's not
    materially different yet (and I'd have to say that the scores are pretty
    close) from a benchmark standpoint. It _is_ after all the same chip. And
    there's not inherent reason we'd expect x64 to run significantly
    differently. What will change, though, is as native apps take advantage of
    the greater memory address space, more and bigger registers, and other
    enhancements, we'll see significant speed improvements in those apps.

    --
    Charlie.
    http://msmvps.com/xperts64

    Will wrote:
    > Hi,
    >
    > I've run a 3D benchmark on my PC, first with XP 32 and then with x64. I
    > used 3DMark 2005 to run the test.
    >
    > The overall scores are as follows
    >
    > Windows XP 32: 5630
    > Windows XP 64: 5525
    >
    >
    > I was shocked that the performance is acctually worse on x64.
    >
    > So this has left me scratching my head thinking what are the real benefits
    > of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with my
    > 2GB.
    > How long will it be before developers other than MSFT start writing 64bit
    > code? 18 months - longer?
    >
    >
    > Are we running Windows XP 64 just for the sake of it?
    >
    >
    > Will
     
    Charlie Russel - MVP, Nov 20, 2005
    #9
  10. Will

    Dennis Pack Guest

    Charlie:
    Before x64 was released using CPP builds on a dual boot machine
    benchmark tests available at the time were very close (within about 100
    points) and the video drivers available in x64 were very slow compared to
    the x86 drivers available. But that was comparing apples to apples. At the
    same time rendering a video clip on the same machine took 71 minutes in x86
    and 51 minutes with x64. I gave up on benchmark tests because were slanted
    towards different areas, some AMD vs P4 or ATI vs NVidia.


    "Charlie Russel - MVP" <> wrote in message
    news:...
    > I'd have to say that it simply shows that benchmarks are interesting, but
    > don't reflect what I see on my machines. Not suprising. Now it's certainly
    > true that I don't think MS Word runs one bit faster in x64 Edition. But
    > then, I didn't expect it to. And since my overall machine seems to run
    > somewhat faster, I'm happy. As we see new applications that are native
    > 64-bit, we'll see the whole thing shift. I'm not surprised that it's not
    > materially different yet (and I'd have to say that the scores are pretty
    > close) from a benchmark standpoint. It _is_ after all the same chip. And
    > there's not inherent reason we'd expect x64 to run significantly
    > differently. What will change, though, is as native apps take advantage of
    > the greater memory address space, more and bigger registers, and other
    > enhancements, we'll see significant speed improvements in those apps.
    >
    > --
    > Charlie.
    > http://msmvps.com/xperts64
    >
    > Will wrote:
    >> Hi,
    >>
    >> I've run a 3D benchmark on my PC, first with XP 32 and then with x64. I
    >> used 3DMark 2005 to run the test.
    >>
    >> The overall scores are as follows
    >>
    >> Windows XP 32: 5630
    >> Windows XP 64: 5525
    >>
    >>
    >> I was shocked that the performance is acctually worse on x64.
    >>
    >> So this has left me scratching my head thinking what are the real
    >> benefits
    >> of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with my
    >> 2GB.
    >> How long will it be before developers other than MSFT start writing 64bit
    >> code? 18 months - longer?
    >>
    >>
    >> Are we running Windows XP 64 just for the sake of it?
    >>
    >>
    >> Will

    >
    >
     
    Dennis Pack, Nov 20, 2005
    #10
  11. Yes. Different benchmarks have wildly different results on the same two
    machines, supposedly testing similar things. I use this Ferrari every day.
    At first, I used it in 32-bit some of the time, and 64-bit some of the time.
    By the time I'd had it 3 weeks, I used it in x64 Edition 95% of the time.
    And now I haven't booted into 32-bit XP in weeks. That tells me all I need
    to know about how fast the OS is, and how compatible with my program mix. :)

    --
    Charlie.
    http://msmvps.com/xperts64

    Dennis Pack wrote:
    > Charlie:
    > Before x64 was released using CPP builds on a dual boot machine
    > benchmark tests available at the time were very close (within about 100
    > points) and the video drivers available in x64 were very slow compared to
    > the x86 drivers available. But that was comparing apples to apples. At the
    > same time rendering a video clip on the same machine took 71 minutes in
    > x86 and 51 minutes with x64. I gave up on benchmark tests because were
    > slanted towards different areas, some AMD vs P4 or ATI vs NVidia.
    >
    >
    > "Charlie Russel - MVP" <> wrote in message
    > news:...
    >> I'd have to say that it simply shows that benchmarks are interesting, but
    >> don't reflect what I see on my machines. Not suprising. Now it's
    >> certainly true that I don't think MS Word runs one bit faster in x64
    >> Edition. But then, I didn't expect it to. And since my overall machine
    >> seems to run somewhat faster, I'm happy. As we see new applications that
    >> are native 64-bit, we'll see the whole thing shift. I'm not surprised
    >> that it's not materially different yet (and I'd have to say that the
    >> scores are pretty close) from a benchmark standpoint. It _is_ after all
    >> the same chip. And there's not inherent reason we'd expect x64 to run
    >> significantly differently. What will change, though, is as native apps
    >> take advantage of the greater memory address space, more and bigger
    >> registers, and other enhancements, we'll see significant speed
    >> improvements in those apps. --
    >> Charlie.
    >> http://msmvps.com/xperts64
    >>
    >> Will wrote:
    >>> Hi,
    >>>
    >>> I've run a 3D benchmark on my PC, first with XP 32 and then with x64. I
    >>> used 3DMark 2005 to run the test.
    >>>
    >>> The overall scores are as follows
    >>>
    >>> Windows XP 32: 5630
    >>> Windows XP 64: 5525
    >>>
    >>>
    >>> I was shocked that the performance is acctually worse on x64.
    >>>
    >>> So this has left me scratching my head thinking what are the real
    >>> benefits
    >>> of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with my
    >>> 2GB.
    >>> How long will it be before developers other than MSFT start writing
    >>> 64bit code? 18 months - longer?
    >>>
    >>>
    >>> Are we running Windows XP 64 just for the sake of it?
    >>>
    >>>
    >>> Will
     
    Charlie Russel - MVP, Nov 20, 2005
    #11
  12. Will

    Dennis Pack Guest

    Charlie:

    I agree fully. The only time I boot x86 is to verify that it's
    up to date, because I have one old DOS program that I use twice a year and
    don't want to give up.





    "Charlie Russel - MVP" <> wrote in message
    news:%230ovY$...
    > Yes. Different benchmarks have wildly different results on the same two
    > machines, supposedly testing similar things. I use this Ferrari every day.
    > At first, I used it in 32-bit some of the time, and 64-bit some of the
    > time. By the time I'd had it 3 weeks, I used it in x64 Edition 95% of the
    > time. And now I haven't booted into 32-bit XP in weeks. That tells me all
    > I need to know about how fast the OS is, and how compatible with my
    > program mix. :)
    >
    > --
    > Charlie.
    > http://msmvps.com/xperts64
    >
    > Dennis Pack wrote:
    >> Charlie:
    >> Before x64 was released using CPP builds on a dual boot
    >> machine
    >> benchmark tests available at the time were very close (within about 100
    >> points) and the video drivers available in x64 were very slow compared to
    >> the x86 drivers available. But that was comparing apples to apples. At
    >> the
    >> same time rendering a video clip on the same machine took 71 minutes in
    >> x86 and 51 minutes with x64. I gave up on benchmark tests because were
    >> slanted towards different areas, some AMD vs P4 or ATI vs NVidia.
    >>
    >>
    >> "Charlie Russel - MVP" <> wrote in message
    >> news:...
    >>> I'd have to say that it simply shows that benchmarks are interesting,
    >>> but
    >>> don't reflect what I see on my machines. Not suprising. Now it's
    >>> certainly true that I don't think MS Word runs one bit faster in x64
    >>> Edition. But then, I didn't expect it to. And since my overall machine
    >>> seems to run somewhat faster, I'm happy. As we see new applications that
    >>> are native 64-bit, we'll see the whole thing shift. I'm not surprised
    >>> that it's not materially different yet (and I'd have to say that the
    >>> scores are pretty close) from a benchmark standpoint. It _is_ after all
    >>> the same chip. And there's not inherent reason we'd expect x64 to run
    >>> significantly differently. What will change, though, is as native apps
    >>> take advantage of the greater memory address space, more and bigger
    >>> registers, and other enhancements, we'll see significant speed
    >>> improvements in those apps. --
    >>> Charlie.
    >>> http://msmvps.com/xperts64
    >>>
    >>> Will wrote:
    >>>> Hi,
    >>>>
    >>>> I've run a 3D benchmark on my PC, first with XP 32 and then with x64. I
    >>>> used 3DMark 2005 to run the test.
    >>>>
    >>>> The overall scores are as follows
    >>>>
    >>>> Windows XP 32: 5630
    >>>> Windows XP 64: 5525
    >>>>
    >>>>
    >>>> I was shocked that the performance is acctually worse on x64.
    >>>>
    >>>> So this has left me scratching my head thinking what are the real
    >>>> benefits
    >>>> of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with my
    >>>> 2GB.
    >>>> How long will it be before developers other than MSFT start writing
    >>>> 64bit code? 18 months - longer?
    >>>>
    >>>>
    >>>> Are we running Windows XP 64 just for the sake of it?
    >>>>
    >>>>
    >>>> Will

    >
    >
     
    Dennis Pack, Nov 20, 2005
    #12
  13. Yup. Run the DOS program in a virtual machine. Perfect solution for
    something you use this rarely.

    --
    Charlie.
    http://msmvps.com/xperts64

    Dennis Pack wrote:
    > Charlie:
    >
    > I agree fully. The only time I boot x86 is to verify that it's
    > up to date, because I have one old DOS program that I use twice a year and
    > don't want to give up.
    >
    >
    >
    >
    >
    > "Charlie Russel - MVP" <> wrote in message
    > news:%230ovY$...
    >> Yes. Different benchmarks have wildly different results on the same two
    >> machines, supposedly testing similar things. I use this Ferrari every
    >> day. At first, I used it in 32-bit some of the time, and 64-bit some of
    >> the time. By the time I'd had it 3 weeks, I used it in x64 Edition 95%
    >> of the time. And now I haven't booted into 32-bit XP in weeks. That
    >> tells me all I need to know about how fast the OS is, and how compatible
    >> with my program mix. :)
    >>
    >> --
    >> Charlie.
    >> http://msmvps.com/xperts64
    >>
    >> Dennis Pack wrote:
    >>> Charlie:
    >>> Before x64 was released using CPP builds on a dual boot
    >>> machine
    >>> benchmark tests available at the time were very close (within about 100
    >>> points) and the video drivers available in x64 were very slow compared
    >>> to the x86 drivers available. But that was comparing apples to apples.
    >>> At the
    >>> same time rendering a video clip on the same machine took 71 minutes in
    >>> x86 and 51 minutes with x64. I gave up on benchmark tests because were
    >>> slanted towards different areas, some AMD vs P4 or ATI vs NVidia.
    >>>
    >>>
    >>> "Charlie Russel - MVP" <> wrote in
    >>> message news:...
    >>>> I'd have to say that it simply shows that benchmarks are interesting,
    >>>> but
    >>>> don't reflect what I see on my machines. Not suprising. Now it's
    >>>> certainly true that I don't think MS Word runs one bit faster in x64
    >>>> Edition. But then, I didn't expect it to. And since my overall machine
    >>>> seems to run somewhat faster, I'm happy. As we see new applications
    >>>> that are native 64-bit, we'll see the whole thing shift. I'm not
    >>>> surprised that it's not materially different yet (and I'd have to say
    >>>> that the scores are pretty close) from a benchmark standpoint. It _is_
    >>>> after all the same chip. And there's not inherent reason we'd expect
    >>>> x64 to run significantly differently. What will change, though, is as
    >>>> native apps take advantage of the greater memory address space, more
    >>>> and bigger registers, and other enhancements, we'll see significant
    >>>> speed improvements in those apps. --
    >>>> Charlie.
    >>>> http://msmvps.com/xperts64
    >>>>
    >>>> Will wrote:
    >>>>> Hi,
    >>>>>
    >>>>> I've run a 3D benchmark on my PC, first with XP 32 and then with x64.
    >>>>> I used 3DMark 2005 to run the test.
    >>>>>
    >>>>> The overall scores are as follows
    >>>>>
    >>>>> Windows XP 32: 5630
    >>>>> Windows XP 64: 5525
    >>>>>
    >>>>>
    >>>>> I was shocked that the performance is acctually worse on x64.
    >>>>>
    >>>>> So this has left me scratching my head thinking what are the real
    >>>>> benefits
    >>>>> of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with
    >>>>> my 2GB.
    >>>>> How long will it be before developers other than MSFT start writing
    >>>>> 64bit code? 18 months - longer?
    >>>>>
    >>>>>
    >>>>> Are we running Windows XP 64 just for the sake of it?
    >>>>>
    >>>>>
    >>>>> Will
     
    Charlie Russel - MVP, Nov 20, 2005
    #13
  14. Exactly. The proof is in real world usage; stability and that je ne sais
    quoi pas feeling that it runs better. Yeah, maybe I'm just deluding myself
    because I had to pay for 64 bit, but I only boot into XP Pro to make sure
    it's still there...


    "Charlie Russel - MVP" <> wrote in message
    news:%230ovY$...
    > Yes. Different benchmarks have wildly different results on the same two
    > machines, supposedly testing similar things. I use this Ferrari every day.
    > At first, I used it in 32-bit some of the time, and 64-bit some of the

    time.
    > By the time I'd had it 3 weeks, I used it in x64 Edition 95% of the time.
    > And now I haven't booted into 32-bit XP in weeks. That tells me all I need
    > to know about how fast the OS is, and how compatible with my program mix.

    :)
    >
    >
     
    Dennis Gordon, Nov 20, 2005
    #14
  15. Will

    Will Guest

    As an interesting FYI - I read somewhere that that the background wallpaper
    can have a detrimental effect on the graphics card.

    In my x86 system, wallpaper was turned off.

    I have subsequently disabled wallpaper and ran 3DMark 2005 again.

    Resulted in a score of 5619 !!



    "Will" <> wrote in message
    news:43805865$0$203$...
    > Hi,
    >
    > I've run a 3D benchmark on my PC, first with XP 32 and then with x64. I
    > used 3DMark 2005 to run the test.
    >
    > The overall scores are as follows
    >
    > Windows XP 32: 5630
    > Windows XP 64: 5525
    >
    >
    > I was shocked that the performance is acctually worse on x64.
    >
    > So this has left me scratching my head thinking what are the real benefits
    > of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with my
    > 2GB.
    >
    > How long will it be before developers other than MSFT start writing 64bit
    > code? 18 months - longer?
    >
    >
    > Are we running Windows XP 64 just for the sake of it?
    >
    >
    > Will
    >
     
    Will, Nov 20, 2005
    #15
  16. Will

    Dennis Pack Guest

    I've never tried VM yet. But it's a good project when I get the time.


    "Charlie Russel - MVP" <> wrote in message
    news:%...
    > Yup. Run the DOS program in a virtual machine. Perfect solution for
    > something you use this rarely.
    >
    > --
    > Charlie.
    > http://msmvps.com/xperts64
    >
    > Dennis Pack wrote:
    >> Charlie:
    >>
    >> I agree fully. The only time I boot x86 is to verify that it's
    >> up to date, because I have one old DOS program that I use twice a year
    >> and
    >> don't want to give up.
    >>
    >>
    >>
    >>
    >>
    >> "Charlie Russel - MVP" <> wrote in message
    >> news:%230ovY$...
    >>> Yes. Different benchmarks have wildly different results on the same two
    >>> machines, supposedly testing similar things. I use this Ferrari every
    >>> day. At first, I used it in 32-bit some of the time, and 64-bit some of
    >>> the time. By the time I'd had it 3 weeks, I used it in x64 Edition 95%
    >>> of the time. And now I haven't booted into 32-bit XP in weeks. That
    >>> tells me all I need to know about how fast the OS is, and how compatible
    >>> with my program mix. :)
    >>>
    >>> --
    >>> Charlie.
    >>> http://msmvps.com/xperts64
    >>>
    >>> Dennis Pack wrote:
    >>>> Charlie:
    >>>> Before x64 was released using CPP builds on a dual boot
    >>>> machine
    >>>> benchmark tests available at the time were very close (within about 100
    >>>> points) and the video drivers available in x64 were very slow compared
    >>>> to the x86 drivers available. But that was comparing apples to apples.
    >>>> At the
    >>>> same time rendering a video clip on the same machine took 71 minutes in
    >>>> x86 and 51 minutes with x64. I gave up on benchmark tests because were
    >>>> slanted towards different areas, some AMD vs P4 or ATI vs NVidia.
    >>>>
    >>>>
    >>>> "Charlie Russel - MVP" <> wrote in
    >>>> message news:...
    >>>>> I'd have to say that it simply shows that benchmarks are interesting,
    >>>>> but
    >>>>> don't reflect what I see on my machines. Not suprising. Now it's
    >>>>> certainly true that I don't think MS Word runs one bit faster in x64
    >>>>> Edition. But then, I didn't expect it to. And since my overall machine
    >>>>> seems to run somewhat faster, I'm happy. As we see new applications
    >>>>> that are native 64-bit, we'll see the whole thing shift. I'm not
    >>>>> surprised that it's not materially different yet (and I'd have to say
    >>>>> that the scores are pretty close) from a benchmark standpoint. It _is_
    >>>>> after all the same chip. And there's not inherent reason we'd expect
    >>>>> x64 to run significantly differently. What will change, though, is as
    >>>>> native apps take advantage of the greater memory address space, more
    >>>>> and bigger registers, and other enhancements, we'll see significant
    >>>>> speed improvements in those apps. --
    >>>>> Charlie.
    >>>>> http://msmvps.com/xperts64
    >>>>>
    >>>>> Will wrote:
    >>>>>> Hi,
    >>>>>>
    >>>>>> I've run a 3D benchmark on my PC, first with XP 32 and then with x64.
    >>>>>> I used 3DMark 2005 to run the test.
    >>>>>>
    >>>>>> The overall scores are as follows
    >>>>>>
    >>>>>> Windows XP 32: 5630
    >>>>>> Windows XP 64: 5525
    >>>>>>
    >>>>>>
    >>>>>> I was shocked that the performance is acctually worse on x64.
    >>>>>>
    >>>>>> So this has left me scratching my head thinking what are the real
    >>>>>> benefits
    >>>>>> of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with
    >>>>>> my 2GB.
    >>>>>> How long will it be before developers other than MSFT start writing
    >>>>>> 64bit code? 18 months - longer?
    >>>>>>
    >>>>>>
    >>>>>> Are we running Windows XP 64 just for the sake of it?
    >>>>>>
    >>>>>>
    >>>>>> Will

    >
    >
     
    Dennis Pack, Nov 20, 2005
    #16
  17. New version of Virtual Server Standard Edition is $99 USD list price. Tough
    to beat at that price, and it runs just fine on x64 Edition. (note that
    Virtual PC does NOT run on x64 Edition, however.)

    --
    Charlie.
    http://msmvps.com/xperts64

    Dennis Pack wrote:
    > I've never tried VM yet. But it's a good project when I get the time.
    >
    >
    > "Charlie Russel - MVP" <> wrote in message
    > news:%...
    >> Yup. Run the DOS program in a virtual machine. Perfect solution for
    >> something you use this rarely.
    >>
    >> --
    >> Charlie.
    >> http://msmvps.com/xperts64
    >>
    >> Dennis Pack wrote:
    >>> Charlie:
    >>>
    >>> I agree fully. The only time I boot x86 is to verify that
    >>> it's up to date, because I have one old DOS program that I use twice a
    >>> year and
    >>> don't want to give up.
    >>>
    >>>
    >>>
    >>>
    >>>
    >>> "Charlie Russel - MVP" <> wrote in
    >>> message news:%230ovY$...
    >>>> Yes. Different benchmarks have wildly different results on the same two
    >>>> machines, supposedly testing similar things. I use this Ferrari every
    >>>> day. At first, I used it in 32-bit some of the time, and 64-bit some of
    >>>> the time. By the time I'd had it 3 weeks, I used it in x64 Edition 95%
    >>>> of the time. And now I haven't booted into 32-bit XP in weeks. That
    >>>> tells me all I need to know about how fast the OS is, and how
    >>>> compatible with my program mix. :)
    >>>>
    >>>> --
    >>>> Charlie.
    >>>> http://msmvps.com/xperts64
    >>>>
    >>>> Dennis Pack wrote:
    >>>>> Charlie:
    >>>>> Before x64 was released using CPP builds on a dual boot
    >>>>> machine
    >>>>> benchmark tests available at the time were very close (within about
    >>>>> 100 points) and the video drivers available in x64 were very slow
    >>>>> compared to the x86 drivers available. But that was comparing apples
    >>>>> to apples. At the
    >>>>> same time rendering a video clip on the same machine took 71 minutes
    >>>>> in x86 and 51 minutes with x64. I gave up on benchmark tests because
    >>>>> were slanted towards different areas, some AMD vs P4 or ATI vs NVidia.
    >>>>>
    >>>>>
    >>>>> "Charlie Russel - MVP" <> wrote in
    >>>>> message news:...
    >>>>>> I'd have to say that it simply shows that benchmarks are interesting,
    >>>>>> but
    >>>>>> don't reflect what I see on my machines. Not suprising. Now it's
    >>>>>> certainly true that I don't think MS Word runs one bit faster in x64
    >>>>>> Edition. But then, I didn't expect it to. And since my overall
    >>>>>> machine seems to run somewhat faster, I'm happy. As we see new
    >>>>>> applications that are native 64-bit, we'll see the whole thing
    >>>>>> shift. I'm not surprised that it's not materially different yet (and
    >>>>>> I'd have to say that the scores are pretty close) from a benchmark
    >>>>>> standpoint. It _is_ after all the same chip. And there's not
    >>>>>> inherent reason we'd expect x64 to run significantly differently.
    >>>>>> What will change, though, is as native apps take advantage of the
    >>>>>> greater memory address space, more and bigger registers, and other
    >>>>>> enhancements, we'll see significant speed improvements in those
    >>>>>> apps. --
    >>>>>> Charlie.
    >>>>>> http://msmvps.com/xperts64
    >>>>>>
    >>>>>> Will wrote:
    >>>>>>> Hi,
    >>>>>>>
    >>>>>>> I've run a 3D benchmark on my PC, first with XP 32 and then with
    >>>>>>> x64. I used 3DMark 2005 to run the test.
    >>>>>>>
    >>>>>>> The overall scores are as follows
    >>>>>>>
    >>>>>>> Windows XP 32: 5630
    >>>>>>> Windows XP 64: 5525
    >>>>>>>
    >>>>>>>
    >>>>>>> I was shocked that the performance is acctually worse on x64.
    >>>>>>>
    >>>>>>> So this has left me scratching my head thinking what are the real
    >>>>>>> benefits
    >>>>>>> of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with
    >>>>>>> my 2GB.
    >>>>>>> How long will it be before developers other than MSFT start writing
    >>>>>>> 64bit code? 18 months - longer?
    >>>>>>>
    >>>>>>>
    >>>>>>> Are we running Windows XP 64 just for the sake of it?
    >>>>>>>
    >>>>>>>
    >>>>>>> Will
     
    Charlie Russel - MVP, Nov 20, 2005
    #17
  18. Will

    Dennis Pack Guest

    Thank You.


    "Charlie Russel - MVP" <> wrote in message
    news:...
    > New version of Virtual Server Standard Edition is $99 USD list price.
    > Tough to beat at that price, and it runs just fine on x64 Edition. (note
    > that Virtual PC does NOT run on x64 Edition, however.)
    >
    > --
    > Charlie.
    > http://msmvps.com/xperts64
    >
    > Dennis Pack wrote:
    >> I've never tried VM yet. But it's a good project when I get the time.
    >>
    >>
    >> "Charlie Russel - MVP" <> wrote in message
    >> news:%...
    >>> Yup. Run the DOS program in a virtual machine. Perfect solution for
    >>> something you use this rarely.
    >>>
    >>> --
    >>> Charlie.
    >>> http://msmvps.com/xperts64
    >>>
    >>> Dennis Pack wrote:
    >>>> Charlie:
    >>>>
    >>>> I agree fully. The only time I boot x86 is to verify that
    >>>> it's up to date, because I have one old DOS program that I use twice a
    >>>> year and
    >>>> don't want to give up.
    >>>>
    >>>>
    >>>>
    >>>>
    >>>>
    >>>> "Charlie Russel - MVP" <> wrote in
    >>>> message news:%230ovY$...
    >>>>> Yes. Different benchmarks have wildly different results on the same
    >>>>> two
    >>>>> machines, supposedly testing similar things. I use this Ferrari every
    >>>>> day. At first, I used it in 32-bit some of the time, and 64-bit some
    >>>>> of
    >>>>> the time. By the time I'd had it 3 weeks, I used it in x64 Edition 95%
    >>>>> of the time. And now I haven't booted into 32-bit XP in weeks. That
    >>>>> tells me all I need to know about how fast the OS is, and how
    >>>>> compatible with my program mix. :)
    >>>>>
    >>>>> --
    >>>>> Charlie.
    >>>>> http://msmvps.com/xperts64
    >>>>>
    >>>>> Dennis Pack wrote:
    >>>>>> Charlie:
    >>>>>> Before x64 was released using CPP builds on a dual boot
    >>>>>> machine
    >>>>>> benchmark tests available at the time were very close (within about
    >>>>>> 100 points) and the video drivers available in x64 were very slow
    >>>>>> compared to the x86 drivers available. But that was comparing apples
    >>>>>> to apples. At the
    >>>>>> same time rendering a video clip on the same machine took 71 minutes
    >>>>>> in x86 and 51 minutes with x64. I gave up on benchmark tests because
    >>>>>> were slanted towards different areas, some AMD vs P4 or ATI vs
    >>>>>> NVidia.
    >>>>>>
    >>>>>>
    >>>>>> "Charlie Russel - MVP" <> wrote in
    >>>>>> message news:...
    >>>>>>> I'd have to say that it simply shows that benchmarks are
    >>>>>>> interesting,
    >>>>>>> but
    >>>>>>> don't reflect what I see on my machines. Not suprising. Now it's
    >>>>>>> certainly true that I don't think MS Word runs one bit faster in x64
    >>>>>>> Edition. But then, I didn't expect it to. And since my overall
    >>>>>>> machine seems to run somewhat faster, I'm happy. As we see new
    >>>>>>> applications that are native 64-bit, we'll see the whole thing
    >>>>>>> shift. I'm not surprised that it's not materially different yet (and
    >>>>>>> I'd have to say that the scores are pretty close) from a benchmark
    >>>>>>> standpoint. It _is_ after all the same chip. And there's not
    >>>>>>> inherent reason we'd expect x64 to run significantly differently.
    >>>>>>> What will change, though, is as native apps take advantage of the
    >>>>>>> greater memory address space, more and bigger registers, and other
    >>>>>>> enhancements, we'll see significant speed improvements in those
    >>>>>>> apps. --
    >>>>>>> Charlie.
    >>>>>>> http://msmvps.com/xperts64
    >>>>>>>
    >>>>>>> Will wrote:
    >>>>>>>> Hi,
    >>>>>>>>
    >>>>>>>> I've run a 3D benchmark on my PC, first with XP 32 and then with
    >>>>>>>> x64. I used 3DMark 2005 to run the test.
    >>>>>>>>
    >>>>>>>> The overall scores are as follows
    >>>>>>>>
    >>>>>>>> Windows XP 32: 5630
    >>>>>>>> Windows XP 64: 5525
    >>>>>>>>
    >>>>>>>>
    >>>>>>>> I was shocked that the performance is acctually worse on x64.
    >>>>>>>>
    >>>>>>>> So this has left me scratching my head thinking what are the real
    >>>>>>>> benefits
    >>>>>>>> of XP x64? Yes, we can stick in tons of RAM but I'm quite happy
    >>>>>>>> with
    >>>>>>>> my 2GB.
    >>>>>>>> How long will it be before developers other than MSFT start writing
    >>>>>>>> 64bit code? 18 months - longer?
    >>>>>>>>
    >>>>>>>>
    >>>>>>>> Are we running Windows XP 64 just for the sake of it?
    >>>>>>>>
    >>>>>>>>
    >>>>>>>> Will

    >
    >
     
    Dennis Pack, Nov 20, 2005
    #18
  19. Will

    Barb Bowman Guest

    did you ever write anything on using Virtual Server on x64? if not,
    great topic...

    On Sun, 20 Nov 2005 15:50:38 -0800, "Charlie Russel - MVP"
    <> wrote:

    >New version of Virtual Server Standard Edition is $99 USD list price. Tough
    >to beat at that price, and it runs just fine on x64 Edition. (note that
    >Virtual PC does NOT run on x64 Edition, however.)

    --

    Barb Bowman
    MS Windows-MVP
    Expert Zone Columnist
    http://www.microsoft.com/windowsxp/expertzone/meetexperts/bowman.mspx
    http://blogs.digitalmediaphile.com/barb/
     
    Barb Bowman, Nov 21, 2005
    #19
  20. Are your scores CPU?

    In that case I scored the following:
    XP 32: 5681
    XP 64: 5834

    --
    Asus P5WD2 Premium (955X Express)
    P4 3.73 GHz EE
    RAPTOR 2 x 74GB in RAID 0
    DDR2 PC8000 2GB
    GeForce 7800GTX
    Samsung 24" TFT
    Enermax 660W
    Plextor PX-712SA


    "Will" wrote:

    > Hi,
    >
    > I've run a 3D benchmark on my PC, first with XP 32 and then with x64. I used
    > 3DMark 2005 to run the test.
    >
    > The overall scores are as follows
    >
    > Windows XP 32: 5630
    > Windows XP 64: 5525
    >
    >
    > I was shocked that the performance is acctually worse on x64.
    >
    > So this has left me scratching my head thinking what are the real benefits
    > of XP x64? Yes, we can stick in tons of RAM but I'm quite happy with my 2GB.
    >
    > How long will it be before developers other than MSFT start writing 64bit
    > code? 18 months - longer?
    >
    >
    > Are we running Windows XP 64 just for the sake of it?
    >
    >
    > Will
    >
    >
    >
     
    =?Utf-8?B?TWFyYXRvbm1hbm5lbg==?=, Nov 21, 2005
    #20
    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. Goody2shoe
    Replies:
    32
    Views:
    1,196
    Goody2shoe
    Jul 11, 2005
  2. =?Utf-8?B?ZWNfbGFja2V5?=

    XP64 & XP32 on the same computer

    =?Utf-8?B?ZWNfbGFja2V5?=, Nov 11, 2005, in forum: Windows 64bit
    Replies:
    1
    Views:
    416
    John Barnes
    Nov 11, 2005
  3. =?Utf-8?B?QmVlcnRqZQ==?=

    Dual boot XP64 and XP32

    =?Utf-8?B?QmVlcnRqZQ==?=, Dec 12, 2005, in forum: Windows 64bit
    Replies:
    13
    Views:
    803
    John Barnes
    Dec 13, 2005
  4. jg
    Replies:
    1
    Views:
    359
    John Barnes
    Feb 23, 2006
  5. gs
    Replies:
    12
    Views:
    2,342
    Charlie Russel - MVP
    Apr 2, 2006
Loading...

Share This Page