Dual 1.25 G4 too fast

Discussion in 'General Mac Discussion' started by MegaSignal, Sep 17, 2004.

  1. MegaSignal macrumors 6502

    Oct 20, 2003
    Spent some time in the Apple Store today encoding a 9-minute movie (with iDVD 4.0.1, background encoding enabled) on both a Dual 2.0 as well as a new G5 iMac:

    Dual 2.0 G5 (tower), 1.5G RAM: 9 minutes (1x encoding)

    iMac G5 1.8 20", 512Mb RAM: approx. 14 minutes (0.64x encoding)
    {Both of the above using Panther}

    Dual 1.25 G4 (tower), 1G RAM: 10 minutes (0.9x encoding)
    {My "old" machine, still on Jaguar 10.2.8}

    My question: Why isn't the Dual 2.0 faster? (To say I was a little disappointed is an understatement). =>Just one note: Both of the G5s had the energy saver set to "Automatic" as I was unable to change this setting - the Apple store had locked this feature. Could this be a reason?

    Thanks all in advance for your input...
  2. mikeyredk macrumors 65816

    Mar 13, 2003
    probably ask them to change it for you. they prob will
  3. cubist macrumors 68020

    Jul 4, 2002
    Muncie, Indiana
    Dual CPUs helps you out here a lot, altho the iMac will do better with more RAM. Yes, I'd change the Power Mac to "High". After the Dual 1.25, you don't mind a little fan noise - and why get a tower if not to compute at full speed?

    Your quote, "You can never underestimate the stupidity of the general public.
    Scott Adams (1957 - ), The Dilbert Future", is reminiscent of a quote of H. L. Mencken, "Nobody ever went broke by underestimating the stupidity of the American public".
  4. MegaSignal thread starter macrumors 6502

    Oct 20, 2003
    48 hours

    Give me a couple of days - I'll retry the test with the energy saver set to its highest setting; I'll report back.

    Any other comments in the meantime are most certainly welcome. (I'm trying desperately to justify a $3600 expense).

    Thanks all again.
  5. Sun Baked macrumors G5

    Sun Baked

    May 19, 2002
    All depends on how iDVD does the encoding...

    If it keeps the majority of the movies on the HD, or loads it all into memory before encoding starts.

    If it keeps the majority of the movie on the HD, then all you're testing is the I/O -- since all the machines have near equal HD transfer speeds (the SATA bus may be fast, but actual drive isn't and quite near the ATA drives transfer speed anyways -- lot's of future proofing here on SATA drives.)

    It's really why people do a bunch of stuff to make sure the Photoshop has memory to work with instead of scratch files.
  6. MegaSignal thread starter macrumors 6502

    Oct 20, 2003
    Interesting comment about the SATA drives; I thought that I might see a bump because of these, but right now, I'm not sure where (if anywhere) the bottleneck is. I suppose I would see a slight improvement using a WD Raptor, but who knows?

    I don't think that DVD Studio Pro would be much different as both apps now use the same encoder (Compressor); I believe I read somewhere that Compressor uses the resident HDD as a temporary folder - so keeping one file or the other on your scratch disk would not make any difference.

    I still like the G5s (both the tower and the new iMac version) for nothing more than their superior cooling systems and relative quietness. I just thought that a Dual 2.0 could cut through an encoding process faster.
  7. MegaSignal thread starter macrumors 6502

    Oct 20, 2003
    Store lineup - lots of extra memory installed on display models

    BTW -

    At the Apple Store in Edina, MN, I noticed the following lineup during my test:

    Two iMacs; both the 1.6-17" as well as the 1.8-20" had 512Mb of RAM installed, according to System Profiler.

    Four G5 towers, three 2.0s, and a 1.8; as noted above, one of the 2.0s had 1.5G of RAM installed, while another had 4G (!); System Profiler reported all as having the 970 chip; I forgot the unix commands to determine whether or not these were the newer "fx" (90nm) series...did not spot any 2.5s.

    All computers felt sluggish compared to my Jagged Dual 1.25, taking longer to boot iDVD, exhibiting numerous beach balls, taking a very long time to perform simple cut and pastes in iMovie, and so on. I was once told that it is faily common to have these machines' HDDs re-imaged on an almost daily basis, as over-zealous would-be customers create havoc on these systems. Perhaps this could also point to a reason as to the relative slowness of these systems...

    ...it just seems odd, as a tremendous amount of real-world users are very delighted with the performance of their machines, posted here and on other sites as well.
  8. Mechcozmo macrumors 603


    Jul 17, 2004
    At least every time I walk into thye Apple Store near me, one of the computers is having its hard drive restored. And they never seem quite as fast, I agree. I think it has something to do with permissions, because if I computer is ever slow, after a permission repair, they seem to be fine.

    Oh, and some people just don't get those macs and completely mess them up...I saw some collage kid trying to download viruses from the net to the computer "because it would be cool." Needless to say, it didn't work.
  9. MegaSignal thread starter macrumors 6502

    Oct 20, 2003
    Just got back from the new G5 iMac thread elsewhere on this site; it would appear that setting the Energy Saver to Highest Performance (from "Automatic") will indeed make a rather substantial difference.

    I will attempt to go back to this store, find the 2.0 with the 4G of RAM, ask a sales person to unlock the Energy Saver Preference, repair permissions, and re-run my encoding test of a 9-minute movie.

    ...as long as the store is not over crowded.
  10. MegaSignal thread starter macrumors 6502

    Oct 20, 2003
    7 Minute encode

    I was able to re-test a Dual 2.0 with 4G of RAM on a 9-minute movie after setting the Energy Saver to utilize maximum performance. Using iDVD 4.0.1 with background encoding enabled yielded a 7-minute encoding time. It still did not appear that the CPUs were worked hard enough; the fans remained at their lowest RPM throughout the test.

    This could be due in part that Compressor has not yet been optimized for 64-bit computing - I'm not sure.

    Either way, the 2.0 is a great machine. However, I think that I'll wait until Tiger is out in its second or third iteration prior to ordering one at this point in time.

    Thanks to all who've responded. As always, I appreciate the comments and feedback!

Share This Page