iTunes 2 is wayyy too processor hungry

24

Comments

  • Reply 21 of 76
    [quote]Originally posted by applenut:

    <strong>



    duh. but when it takes 25-30 percent on a G4 there is something wrong.</strong><hr></blockquote>



    The processor in the original post is a 333MHz G3 in an iMac. I could explain to you the problems that the iMac has with its lack of memory bandwidth causing CPU bottlenecks, but I doubt that you'd see it as anything other than another thing to bash Apple for.



    On my Pismo 400, iTunes 2 takes 15%-20% CPU time with EQ and Sound Enhancer off. SE takes up around 1% more time, and EQ sucks up 3%-10%. There are occasional 10%-15% spikes, which are caused by buffering more data into memory.



    XMMS and MacCast are both front-ends for independent codecs, so it's not surprising that both take little CPU time; the CPU usage occurs in kernelspace. iTunes, on the other hand, uses its own codec. It uses Fraunhofer, which is effecient for a MP3 codec.



    Audion, for what it's worth, sucks up almost twice the CPU time that iTunes does...Audion uses the LAME codec, which results in better sound quality, but is less effecient.



    So, if Apple wanted to improve iTunes' CPU usage, they'd have to come up with a codec that is more effecient than Fraunhofer, or they'd have to integrate the codec into the OS as a kmod. The former would be difficult and rather pointless, and the latter would basically do nothing.
  • Reply 22 of 76
    applenutapplenut Posts: 5,768member
    [quote]Originally posted by Morte:

    <strong>



    The processor in the original post is a 333MHz G3 in an iMac. I could explain to you the problems that the iMac has with its lack of memory bandwidth causing CPU bottlenecks, but I doubt that you'd see it as anything other than another thing to bash Apple for.



    On my Pismo 400, iTunes 2 takes 15%-20% CPU time with EQ and Sound Enhancer off. SE takes up around 1% more time, and EQ sucks up 3%-10%. There are occasional 10%-15% spikes, which are caused by buffering more data into memory.



    XMMS and MacCast are both front-ends for independent codecs, so it's not surprising that both take little CPU time; the CPU usage occurs in kernelspace. iTunes, on the other hand, uses its own codec. It uses Fraunhofer, which is effecient for a MP3 codec.



    Audion, for what it's worth, sucks up almost twice the CPU time that iTunes does...Audion uses the LAME codec, which results in better sound quality, but is less effecient.



    So, if Apple wanted to improve iTunes' CPU usage, they'd have to come up with a codec that is more effecient than Fraunhofer, or they'd have to integrate the codec into the OS as a kmod. The former would be difficult and rather pointless, and the latter would basically do nothing.</strong><hr></blockquote>





    bottlenecks my ass. my Performa can play 8 MP3s at a time without a hitch under BeOS. 160 Mhz may I add.



    iTunes is poorly programmed. spew all the worthless shit you want to about how its not apple's fault its the codecs fault. it's all BS. Apple's codec doesn't even sound good and it sucks up that much.
  • Reply 23 of 76
    sinewavesinewave Posts: 1,074member
    [quote]Originally posted by applenut:

    <strong>



    duh. but when it takes 25-30 percent on a G4 there is something wrong.</strong><hr></blockquote>



    It takes 25 up on my 350 G3.
  • Reply 24 of 76
    sinewavesinewave Posts: 1,074member
    Its all BS! Its all BS! Its all BS! I don't have a reason as to why to explain it's BS.. but it's BS!



  • Reply 25 of 76
    Morte, I don't follow your explanation. Also, I really doubt there are kernel modules for mp3 playback. That (playing mp3s) is strictly user space. The kernel merely allows the software to 'talk' to the hardware, and vice versa. CPU usage is independant from this.



    Clarification: Web browsers require cpu usage, but there are no kernel mods for mozilla.



    [ 11-19-2001: Message edited by: stimuli ]</p>
  • Reply 26 of 76
    [quote]Originally posted by stimuli:

    <strong>Morte, I don't follow your explanation. Also, I really doubt there are kernel modules for mp3 playback. That (playing mp3s) is strictly user space. The kernel merely allows the software to 'talk' to the hardware, and vice versa. CPU usage is independant from this.

    </strong><hr></blockquote>



    Well, it wasn't a very good explanation, because I oversimplified, or maybe I didn't simplify enough. Or something. <img src="graemlins/bugeye.gif" border="0" alt="[Skeptical]" />



    Anyway, let's take it from the bottom...I'll probably screw this up somewhere, but it's Monday, so I have a fresh week's alottment of mistakes to work with:



    A vastly simplified view of how a MP3 player such as XMMS or Maccast plays a MP3 would be as follows. The player makes a call to a sound library/framework (LibC5(?) and Quicktime, respectively), the library interfaces with the sound driver (usually a kmod/kext, unless the sound chip is actually supported by the kernel, as is the case with the occasional chip and Linux 2.4 ), the driver with the kernel, and the kernel with the hardware.



    Now, since the library/framework that the MP3 players use is not unique to the player, and are loaded as a part of the system, the CPU time taken by the decoding and playback isn't included in the player's process. I can't think of anywhere else that the CPU usage could be slotted in than kernelspace (which I'm using as a kludgy term for system processes in general).



    iTunes, on the other hand, has a sound library that is unique to it, so the CPU time that it uses is put under the iTunes process.



    That, at least, is how I understand it. I could be utterly wrong. Wouldn't be the first or last time. <img src="graemlins/smokin.gif" border="0" alt="[Chilling]" />



    In any case, MP3 playing is going to take a certain amount of processing power, and that figure is going to be pretty static because no codec is going to improve on the effeciency of the algorithm, and the algorithm isn't a very good one compared to the newer sound compression formats. The only way that Apple could make iTunes consume less CPU time would be to include support for a more effecient format, such as Ogg Vorbis. It would be awesome if Apple were to start pushing Ogg.
  • Reply 27 of 76
    Ah so!



    but...



    I was led to believe iTunes is based on Jeffrey Robbins' SoundJamMP software... why on earth would any MP3 player have it's own library instead of making calls to Quicktime??!!!



    [quote]iTunes, on the other hand, has a sound library that is unique to it

    <hr></blockquote>



    R U sure? Is Apple on drugs?



    Apple-&gt; <img src="graemlins/smokin.gif" border="0" alt="[Chilling]" /> &lt;- smoking drugs
  • Reply 28 of 76
    [quote]Originally posted by Morte:

    <strong>



    Audion, for what it's worth, sucks up almost twice the CPU time that iTunes does...Audion uses the LAME codec, which results in better sound quality, but is less effecient.</strong><hr></blockquote>



    Try using Audion 2.6.1, the latest version. CPU usage is fantastic. Right now, with the EQ and scrolling titles off, but the sound maximizer on, it is using 8% of my cpu.



    Turning on scrolling titles makes cpu usage jump to the 20s because of an Apple bug that will hopefully be fixed soon.
  • Reply 29 of 76
    torifiletorifile Posts: 4,024member
    [quote]Originally posted by Uh huh:

    <strong>



    Try using Audion 2.6.1, the latest version. CPU usage is fantastic. Right now, with the EQ and scrolling titles off, but the sound maximizer on, it is using 8% of my cpu.



    Turning on scrolling titles makes cpu usage jump to the 20s because of an Apple bug that will hopefully be fixed soon.</strong><hr></blockquote>



    Uh huh,

    Maybe you hit on something here: scrolling titles in iTunes is always on. Maybe that's what's causing the spike in cpu usage. With iTunes not playing, but open, cpu usage is 0.0%, as it should be. With it playing, and the window as small as it can go, it's at 18%. With it small, but so that I can see the scrolling title, it's at about 30%. What's that mean? The scrolling titles seem to make up about an extra 12% of cpu time. That's my non-scientific assessment of the situation.



    But, does anyone else find it odd that it uses the same processor time on almost all processors? One would think that it would take more cpu on a 333mhz g3, than on a 800mhz g4, but that doesn't seem to be the case. What's up with that?
  • Reply 30 of 76
    applenutapplenut Posts: 5,768member
    [quote]Originally posted by Sinewave:

    <strong>Its all BS! Its all BS! Its all BS! I don't have a reason as to why to explain it's BS.. but it's BS!



    </strong><hr></blockquote>



    yea, alright



    Test system: PM G4/400 AGP

    Test Song: Cloud 9- George Harrison

    iTunes w/equalizer- 23-25 percent processor usage.

    iTunes wo/equalizer- 18-21 percent processor usage

    Finder playing MP3- 5-7 percent processor usage.





    Happy now



    yea, it's not iTunes fault at all, of course not
  • Reply 31 of 76
    applenutapplenut Posts: 5,768member
    [quote]Originally posted by torifile:

    <strong>



    But, does anyone else find it odd that it uses the same processor time on almost all processors? One would think that it would take more cpu on a 333mhz g3, than on a 800mhz g4, but that doesn't seem to be the case. What's up with that?</strong><hr></blockquote>



    yea, but sinewave seems to think there is nothing wrong with that at all. MP3 decoding is so hard it should be like that
  • Reply 32 of 76
    sinewavesinewave Posts: 1,074member
    [quote]Originally posted by Uh huh:

    <strong>



    Try using Audion 2.6.1, the latest version. CPU usage is fantastic. Right now, with the EQ and scrolling titles off, but the sound maximizer on, it is using 8% of my cpu.



    Turning on scrolling titles makes cpu usage jump to the 20s because of an Apple bug that will hopefully be fixed soon.</strong><hr></blockquote>



    Funny I HAVE used the latest version and it DOES use just as much CPU time as iTunes. Do you have a secret version that isn't available to the public?
  • Reply 33 of 76
    [quote]Originally posted by Sinewave:

    <strong>



    Funny I HAVE used the latest version and it DOES use just as much CPU time as iTunes. Do you have a secret version that isn't available to the public?</strong><hr></blockquote>



    If you are using Audion X 2.6.1 under Mac OS X 10.1 and it is taking up more than 20% of your CPU, go to preferences:interface, select MP3s/Files and make sure the track info does not scroll. Audion's implementation of Quartz prevents just a portion of a 'face' to be updated at any time, so with scrolling titles on, Audion has to redraw the entire face as the titles scroll.



    If Audion is still using more than 20% of the CPU then something must be wrong. CPU usage was the major difference between Audion 2.6 and Audion 2.6.1. 2.6 used lots of CPU, 2.6.1 uses between 7-10% for me.



    On the same computer, iTunes 2.0.2 uses between 14 and 19% of my CPU.
  • Reply 34 of 76
    sinewavesinewave Posts: 1,074member
    My roomate had Winamp running in 2k on his 400mhz P2. It was taking about 5 to 55% of his CPU.
  • Reply 35 of 76
    cosmocosmo Posts: 662member
    well i've done some more tests and attempted to optimize both itunes and Audion and, as it turns out they seem to use close to the same amount of processor time in X (10.1.1) (on my machine at least).



    itunes takes around 18-23% and Audion uses around 17-23%



    The only major difference is that iTunes spikes at around 45% just after a new song starts and Audion spikes at no more 30% when a new song starts.



    Audion 2.6.1 hidden in back ground, one player window and one playlist window open



    iTunes 2.0.2 hidden in back ground with the smallest player window closed (simply hiding caused for more cpu usage, but closing the smallest player window then hiding itunes provided me with the least cpu usage).



    I still wish it were around 10-15% with spikes of no more than 25% but its not...too bad i won't be getting a new machine until September 2003.
  • Reply 36 of 76
    sinewavesinewave Posts: 1,074member
    Ok of the three we tested. Audion, iTunes, WinAmp all use about the same processor time.



    Oh But it's just a iTunes thing!



  • Reply 37 of 76
    [quote]Originally posted by stimuli:

    <strong>Ah so!



    but...



    I was led to believe iTunes is based on Jeffrey Robbins' SoundJamMP software... why on earth would any MP3 player have it's own library instead of making calls to Quicktime??!!!





    R U sure? Is Apple on drugs?



    Apple-&gt; <img src="graemlins/smokin.gif" border="0" alt="[Chilling]" /> &lt;- smoking drugs</strong><hr></blockquote>



    heh. Given the Apple employees I've known, I wouldn't be surprised if there's a bit of hashish in every block of tofu served in the cafeteria.



    Anyway, Soundjam/iTunes uses the Fraunhofer codec, which is the original MP3 codec. It's much more effecient than QT's MP3 decoder.
  • Reply 38 of 76
    applenutapplenut Posts: 5,768member
    [quote]Originally posted by Sinewave:

    <strong>Ok of the three we tested. Audion, iTunes, WinAmp all use about the same processor time.



    Oh But it's just a iTunes thing!



    </strong><hr></blockquote>



    how'd you compare WinAMP on different hardware to Audion and iTunes? and its strange how you use 3 MP3 players NOT known for their efficiency.



    And its funny how you never responded to my "benchmarks" showing the Finder to use significantly less CPU than all of em
  • Reply 39 of 76
    sinewavesinewave Posts: 1,074member
    [quote]Originally posted by applenut:

    <strong>



    how'd you compare WinAMP on different hardware to Audion and iTunes? and its strange how you use 3 MP3 players NOT known for their efficiency.

    </strong><hr></blockquote>

    Excuse me.. I was just pointing out that the three MAJOR mp3 players take up the same amount of CPU time.. and it wasn't just a iTunes thing.

    <strong> [quote]

    And its funny how you never responded to my "benchmarks" showing the Finder to use significantly less CPU than all of em</strong><hr></blockquote>

    Finder playing mp3? I didn't know the finder played mp3s.



    [ 11-20-2001: Message edited by: Sinewave ]</p>
  • Reply 40 of 76
    cosmocosmo Posts: 662member
    the finder plays mp3s while in column mode.



    I tested this and foudn the finder to use almost exactally the same amount of processor time as itunes, however it didn't tend to spike, but it also doesn't move onto the next song.



    I will try some of the more optomized mp3 players when i find them (to versiontracker i go).



    How did you compare winamp to the other to mac mp3 players? Top gives you the cpu % for it to be a fair comparison it would have to be done with the same processor.
Sign In or Register to comment.