Panther Speed Increase Thread

Discussion in 'Mac OS X 10.3 (Panther) Discussion' started by lsutiger, Oct 24, 2003.

  1. lsutiger macrumors member

    Jul 31, 2003
    I'm interested to learn what everyone thinks of the speed improvement from Jaguar to Panther. Please post what machine you have and if you notice any significant speed or "snappiness" (SP?).

    I haven't received mine yet so I can't comment as of now.

  2. iEric macrumors 6502a

    Jun 26, 2003

    Machine Model: Power Mac G4
    CPU Type: PowerPC G4 (2.1)
    Number Of CPUs: 2
    CPU Speed: 1 GHz
    L2 Cache (per CPU): 256 KB
    L3 Cache (per CPU): 1 MB
    Memory: 1.25 GB
    Bus Speed: 167 MHz
    Boot ROM Version: 4.48f2

    I do see a speed increase in start up times. It takes about 1/2 the time for it to start up.
    It also speeds up apps by 25%. I love Panther.

    I think that Panther uses 2 processors more than Jaguar did - cause when I had Jaguar and I was on a PB 1.0 Ghz, it felt the same as mine...But now with panther, i can actualyl feel the speed :D..But I'm not sure about the Jag/Panther processor theory..i'm just speculating...
  3. madforrit macrumors regular

    Jul 23, 2002
    Berkeley, CA
    Unfortunately, Barefeats ( reports no increase in performance (not startup time) on the G5s.


    But that's ok. Panther is just too cool.
  4. iEric macrumors 6502a

    Jun 26, 2003
    Psh, like the G5s really need speed improvements :) :rolleyes: :D
  5. shadowfax macrumors 603


    Sep 6, 2002
    Houston, TX
    WHO THE ****ING **** would ever shut down their G5?


    i don't understand what you are saying. are you talking about startup time or performance other than startup?

    keep in mind, people, this is unix. it's designed to start up really slow, because it's not designed to start up more than a few times a year, if that. if any of you shut down your computer at night, i pity you, fool! learn to use sleep.
  6. macrumors12345 macrumors 6502

    Mar 1, 2003
    That is because Barefeats is primarily testing programs that call very few System APIs (during the benchmarks), so *of course* the benchmarks didn't speed up! They are still running the exact same code, so how could they??

    A better test of Panther would be if they had tested programs that actually spend most of their time in APIs. I hate to say it, because in general it's a lousy benchmark, but in this particular case XBench is actually not a bad thing to run. Another good example would be something like MS Word, which spends a lot of time drawing text and UI elements (both of these things have been sped up A LOT in Panther). But of course it is hard to benchmark how "fast" Word runs (other than maybe something like scrolling through a very long document), which is why you won't see any of these types of benchmarks on the BareFeats site.

Share This Page