Apple Releases iTunes 5

[SiliconAddict]

WHY DO FOLDERS NOT CARRY OVER TO iPODS? WHAT THE HELL'S THE POINT WITHOUT iPOD SUPPORT!?

GAH.

Maybe it’s the fact that the iPod's current firmware was designed around iTunes 4.x. Maybe? Maybe? You think? Huh? Keep your panties on. It will probably get an update just like iPods got an update when they introduced Podcasting.

Oh and PS.......STOP SHOUTING! ITS REALLY FREAKING ANNOYING.

 

[runtimelogic]

iTunes Modernization / Gapless Playback

Oh Jeez. Still Carbon. They'll have to convert it to Cocoa to compile it for Intel. I was hoping v5 would have been it.

Then they can update their playback engine to support gapless playback while they are at it. Good to see that so many other people on this forum feel the pain caused by the lack of this feature in iTunes. About 30% of my music collection are continuous mix recordings.

Seriously -- my senior project in college was writing an audio playback application for live performace. It was far from elegant, but it supported the equivalent of gapless playback.

Of course the playback engine was designed for playing multiple audio sources simultaneously and "mixing" them together into an audio output stream, but in my eyes the solution for iTunes is just a simplified version where there are only two audio sources. The problem of AAC / MP3 frames that someone mentioned earlier could be solved by writing a byte marker for the final frame of each track to the file during importing. It would indicate the end of the track and allow for gapless playback. Trying to "detect" the end of a track is a bad idea since many recordings include intentional silence which should be preserved to reamin true to the recording.

That my 2 cents on the issue anyhow.

 

[runtimelogic]

Usability and Multitasking

Oh, and about the look and feel of iTunes 5: Aside from the stylistic implications which will differ with taste, I'd like to add that I don't like that the usability of iTunes 5 seems to have suffered.

The lack of spacing between the source list and the edge of the window (there used to be a nice frame there) makes it too easy to accidentally click outside of the app window when selecting a source. Also this is poor design visually since the contents of iTunes is not clearly seprated from windows appearing behind it.

Another thing I have noticed is that when I'm playing an album from my library and I browse to another location in the library iTunes plays a track from the currely viewed area and doesn't continue to play the album like it did in version 4. I'm a bit confused about what is happening exactly here so if someone can clarify how this is supposed to work under 5, that would be very helpful. All I know is it's behaving differently than 4.

 

[bankshot]

Missed this one last week but felt it still warranted a response.

Oh Jeez. Still Carbon. They'll have to convert it to Cocoa to compile it for Intel. I was hoping v5 would have been it.

1. Why would they have to convert to Cocoa to compile for Intel? Are you suggesting that Carbon isn't supported at all on OS X for Intel? That would be absolutely insane, and a good way for Apple to ensure their demise after the switch.

2. iTunes will always be Carbon. Why? Because iTunes for Windows uses a special version of Carbon that was ported to Windows. There's simply no way they're going through all the pain of porting iTunes to Cocoa and porting Cocoa to Windows for zero benefit. Talk about a waste of money.

3. Why do so many people here still assume that Cocoa is so much better than Carbon? Far as I can tell, it's not. From what I hear, Carbon is a fine toolkit, both for porting old MacOS apps, and for writing brand new ones. It's not going away. Cocoa is very nice (I've never written a line of code that uses Carbon, but I've been using Cocoa lately), but I've seen no evidence whatsoever that it's soooo much better than Carbon. I've seen good arguments both ways. There are very few differences that an end-user would even notice between the two (spell checking is the only one that immediately comes to mind). Is it just that Cocoa is "new" (not really if you consider its NeXT heritage) while Carbon is based on the old Mac toolkit (and therefore somehow inferior)?

Anyway I didn't mean to jump on this post in particular, just wanted to lay to rest a couple misconceptions I saw in that little sub-conversation. Well, for the 3 people who actually come back and read this far.

 

[NicP]

Oh, and about the look and feel of iTunes 5: Aside from the stylistic implications which will differ with taste, I'd like to add that I don't like that the usability of iTunes 5 seems to have suffered.

The lack of spacing between the source list and the edge of the window (there used to be a nice frame there) makes it too easy to accidentally click outside of the app window when selecting a source. Also this is poor design visually since the contents of iTunes is not clearly seprated from windows appearing behind it.

Another thing I have noticed is that when I'm playing an album from my library and I browse to another location in the library iTunes plays a track from the currely viewed area and doesn't continue to play the album like it did in version 4. I'm a bit confused about what is happening exactly here so if someone can clarify how this is supposed to work under 5, that would be very helpful. All I know is it's behaving differently than 4.

yes i think there should be more spacing around the outside and also between the scrolling boxes

 

[NicP]

Then they can update their playback engine to support gapless playback while they are at it. Good to see that so many other people on this forum feel the pain caused by the lack of this feature in iTunes. About 30% of my music collection are continuous mix recordings.

Seriously -- my senior project in college was writing an audio playback application for live performace. It was far from elegant, but it supported the equivalent of gapless playback.

Of course the playback engine was designed for playing multiple audio sources simultaneously and "mixing" them together into an audio output stream, but in my eyes the solution for iTunes is just a simplified version where there are only two audio sources. The problem of AAC / MP3 frames that someone mentioned earlier could be solved by writing a byte marker for the final frame of each track to the file during importing. It would indicate the end of the track and allow for gapless playback. Trying to "detect" the end of a track is a bad idea since many recordings include intentional silence which should be preserved to reamin true to the recording.

That my 2 cents on the issue anyhow.

yes it sucks! 90% of my music is continuous mixes!

trying to detect the end of the track is ok, so long as you only look in the last frame, the worst that can happen is 1 frame of intentional silence removed

 

[runtimelogic]

yes it sucks! 90% of my music is continuous mixes!

trying to detect the end of the track is ok, so long as you only look in the last frame, the worst that can happen is 1 frame of intentional silence removed

I'm not familiar with the MP3 standard at that low of a level. I'm guessing the frame size is variable though. Do you happen to know what the biggest frame size is that it supports (IE. what is the biggest duration we could expect to lose with that method?)

Maybe we can solve the problem in this thread and then Apple can just have someone implement it. If iTunes was open source like Darwin, I'd be happy to volunteer my time to implement this feature.

 

[NicP]

I'm not familiar with the MP3 standard at that low of a level. I'm guessing the frame size is variable though. Do you happen to know what the biggest frame size is that it supports (IE. what is the biggest duration we could expect to lose with that method?)

Maybe we can solve the problem in this thread and then Apple can just have someone implement it. If iTunes was open source like Darwin, I'd be happy to volunteer my time to implement this feature.

the reason why itunes is not gapless is because of itunes, not the files it plays

the formats are either gapless or can easily be made gapless, for example ogg stores where the music ends in the file so the trailing silence can be omitted from playback

its part of apples playback engine that is the problem, which we cannot fix

 

[wdlove]

I downloaded iTunes 5 last evening 12.6 MB. Also iTunes Phone Driver 1.1 380 KB, not really sure what it's for though.

Only notice slight changes in the appearance of the site.

 

[weldon]

I'm not familiar with the MP3 standard at that low of a level. I'm guessing the frame size is variable though. Do you happen to know what the biggest frame size is that it supports (IE. what is the biggest duration we could expect to lose with that method?)

MP3 has a variable length frame, but length is determined by bits per second, not time duration. MP3 uses a fixed frame duration (notice I didn't say size) that equates to about 26ms of audio in each frame. In other words, about 38 frames per second. The last frame can have padding bits set to indicate that there is no audio after a certain point.

The problem isn't the 5-20 ms of dead air in the last frame. The problem is in the way iTunes handles playback of queued MP3 files.

 

[runtimelogic]

MP3 has a variable length frame, but length is determined by bits per second, not time duration. MP3 uses a fixed frame duration (notice I didn't say size) that equates to about 26ms of audio in each frame. In other words, about 38 frames per second. The last frame can have padding bits set to indicate that there is no audio after a certain point.

The problem isn't the 5-20 ms of dead air in the last frame. The problem is in the way iTunes handles playback of queued MP3 files.

OK, thanks -- good to understand that about the MP3 format.