PowerMac G4 - Most Powerful Personal Computer of it's day, am I missing something?

Discussion in 'PowerPC Macs' started by MrCheeto, Mar 6, 2010.

  1. macrumors 68030

    MrCheeto

    Joined:
    Nov 2, 2008
    #1
    What is the deal with my experience with the G4? I'm not expecting it to decode the human genome in seven seconds and do backflips, but if I remember correct, when Jobs introduced the G4 at the turn of the century, he and the audience were going absolute NUTS over its performance.

    Specifically, I remember him encoding a video in real time! I'm sorry, I've tried every size and format, but a twenty-eight minute video that takes two and a half hours to export is FUH-HAR from real time.

    In the Xbench tests, the machine sure comes alive, performing all those floating point operations it was designed to, but in this instance it just doesn't make sense to me.

    What am I missing? Is it simply an OS 9 thing? Were they using some super efficient top secret software? How do I encode video in real time?

    What is the fastest most efficient way to change a media file's format?
     
  2. Moderator emeritus

    robbieduncan

    Joined:
    Jul 24, 2002
    Location:
    London
    #2
    He encoded a video using as-then state of the art codecs. They were much, much less CPU intensive to encode than current codecs. If you want real-time encoding either get hardware that is current or use codecs from 10 years ago and accept the lower quality and larger filesizes.
     
  3. thread starter macrumors 68030

    MrCheeto

    Joined:
    Nov 2, 2008
    #3
    That's what I was hoping to hear.

    What filetypes do you think the Mission Impossible Trailer could have been in in 1999? It was good quality, probably TV quality although widescreen.

    Current hardware? My MacBook Pro encodes in 3x real time with almost every file format, but that's irrelevant to this situation.
     
  4. macrumors 68030

    slughead

    Joined:
    Apr 28, 2004
    #4
    Apple? Fudge benchmarks? Nevar!
     
  5. thread starter macrumors 68030

    MrCheeto

    Joined:
    Nov 2, 2008
    #5
    I don't get it...
     
  6. macrumors 6502a

    reel2reel

    Joined:
    Jul 24, 2009
    #6
    Probably Sorenson 3 or 4. I remember using Squeeze back then, which became a progressively more lame piece of software until (I think) it disappeared from the face of the earth.
     
  7. thread starter macrumors 68030

    MrCheeto

    Joined:
    Nov 2, 2008
    #7
    SORENSON?! Now that's a name I've not heard in a looong time, a long time.

    Hell, I completely forgot about it!

    Worth a shot! I'll see what they offer for OS X! Thanks!
     
  8. macrumors 68030

    slughead

    Joined:
    Apr 28, 2004
    #8
    Apple, particularly in the G4 days, fudged their stats and demos to stupidly insane extent.

    I remember people benching the G4 iMac were getting numbers so wildly different from Apple's (presented at the keynote, btw), Apple.com literally changed them without a word!

    I remember the G4 supposedly having all these "flaups" and there was this "megahertz myth." Apple dumped full page propaganda in tech mags about it. While they had a point, at the end of the day, the x86 processors were cheaper and faster at nearly everything. Then Apple switched to IBM (better but ran incredibly hot [see liquid cooled powermac]), and you know the rest.
     
  9. thread starter macrumors 68030

    MrCheeto

    Joined:
    Nov 2, 2008
    #9
    I don't know about that, in my experience my 1ghz Pentium III's were only so far ahead of my G3 iMac, and the iMac ALWAYS performed smoother, system-wise. Even after using Linux on the Pentiums.

    As well, my 2005 2ghz AMD machine is a crapper, whereas the 2003 PowerMac G4 1.25ghz is just a zippin' along...

    If their scores are so off, how do they get away with it? And where the hell are they getting these numbers? And wasn't Steve using a REAL machine in real time to demonstrate the capabilities?
     
  10. macrumors G5

    Consultant

    Joined:
    Jun 27, 2007
    #10
    Got anything to actually support what you are saying?
     
  11. macrumors 6502a

    reel2reel

    Joined:
    Jul 24, 2009
    #11
  12. macrumors 68030

    slughead

    Joined:
    Apr 28, 2004
    #12
    They didn't... Like I said, other benchmarkers like Barefeats.com called them out on it.

    The whole deal is right here:

    http://www.barefeats.com/imacg5b.html

    edit: They even barred him from the apple store

    Mostly I'm talking about their "numbers" and not demos. However, as discussed earlier, demos depend a lot on which software you choose to demonstrate.

    G5 benchmarks example

    It's really common in the tech industry to rig benchmarks. I remember ATI had a driver out that would detect if "Quake3.exe" were running, and then alter the settings without the user knowing it. If you changed it to "Quack3.exe", the settings would be as selected.


    http://techreport.com/articles.x/3089/1
     
  13. thread starter macrumors 68030

    MrCheeto

    Joined:
    Nov 2, 2008
    #13
    And for ONLY $499!!!

    Expletives...well, at least it still rocks the Quake III!
     

Share This Page