Xbench?

Discussion in 'Mac Apps and Mac App Store' started by mrgreen4242, Jan 1, 2005.

  1. mrgreen4242 macrumors 601

    mrgreen4242

    Joined:
    Feb 10, 2004
    #1
    Ok, is XBench a really poor benchmarking tool? Why do dual G4 1ghz PMs get the same score as G5 iMacs? The G4's even have better GPU's! Anyways, is a G5 iMac really as fast as a Dual G4? Is there a better benchmarking utility?

    Thanks,
    Rob
     
  2. dsharits macrumors 68000

    dsharits

    Joined:
    Jun 19, 2004
    Location:
    The People's Republic of America
    #2
    From waht I've seen, Xbench is pretty accurate. The G4 is about the same as the G5 because it has dual processors, for one thing.

    Daniel
     
  3. mrgreen4242 thread starter macrumors 601

    mrgreen4242

    Joined:
    Feb 10, 2004
    #3
    I think I wasn't clear. I would have expected the DP G4 1ghz to be faster than an iMac G5 1.6ghz... even the faster G4 DP machines are only a little faster than the G5 iMacs... I was wondering if the G5 is really that fast or if Xbench unfairly measures G5s?

    Rob
     
  4. cluthz macrumors 68040

    cluthz

    Joined:
    Jun 15, 2004
    Location:
    Norway
    #4

    xbench is poor. Dual 1GHz G4 is MUCH slower than a G5 in most tasks.
    Also the standard Radeon 9000 in the G4 is NOT faster than the FX5200.

    xbench isn't a very good benchmarking tool

    I can't see how you get the same result with a G4 1GHz x2 as the iMac.
    A Dual 1GHz should be around 120-130 points, while my new iMac gets 165-170 points
     
  5. locovaca macrumors regular

    Joined:
    May 14, 2002
    Location:
    Iowa
    #5
    The problem is that you can't just look at the one number and say, "Oh, this is faster." Any benchmark on which you rely on a single, all-encompassing number is a bad one.

    However, XBench does give you the break-out of the numbers. That's what makes it a good benchmark- but you gotta compare the individual numbers, not the total score.
     
  6. mrgreen4242 thread starter macrumors 601

    mrgreen4242

    Joined:
    Feb 10, 2004
    #6
    I was looking at systems that have upgraded GPUs... from GeForce4Ti's to ATI 9X00's, most of which are generally faster than the 5200ultra (although they don't all support the range of features that the 5200 does).

    What type of iMac do you have? I was comparing 1.6ghz iMacs with under a gig of ram to dual 933 and 1ghz G4s with between 512mb and 1.5gb's of ram. The both seem to fall in the 120-140 range for overall performance.

    That is a very god point. Benchmarks tend to be usefull when you are comparing a specific task to the same task on a different machine. I suppose what I was asking is if Xbench was accurate in it's reporting. Sounds liek most people seem to think it is.

    So, maybe a better question is what results in Xbencg are important to the day to day performance of typical apps? What number should one look at if you wanted to know which machine would perform better with iMovie or iDVD, for example?

    Thanks for the replies, very interesting.

    Rob
     
  7. cluthz macrumors 68040

    cluthz

    Joined:
    Jun 15, 2004
    Location:
    Norway
    #7
    The GeForce 4Ti was wicked fast when it came, but it isn't much faster than the FX5200Ultra, on the other hand, the Radeon 9700pro is a different story...

    I've got a BTO 20-inch (1.8GHz, 1024MB matched pair, 250GB HD).
    One of my best friends have a MDD Dual 1.0GHz with Radeon 9000 PRO and 1.5GB RAM, and the iMac smokes that machine.

    I've got max 172 points on xbench (it varies from run to run, but avg is 165-170)

    Look at these tests, the iMac crushes a Dual 1.42 G4 (with Radeon 9700PRO) in 4 out of 6 tests...
    http://www.barefeats.com/imacg5.html
     

Share This Page