I just tried reencoding an album I had originally encoded in Nero AAC. I tried iTunes AAC, 128kbps, VBR enabled. The gap sounds more or less just like with Nero - it's rather light, and doesn't affect the tempo of the music, but it is definitely always there.

Based on these off the cuff tests, I'm willing to conclude this for good, without going to objective testing - iTunes does not provide a true gapless implementation. Or, if it does, it's broken.

Nevermind

Oh well it's at least better than it was :/

Have you tried with ALAC which damn well ought to be able to be played back gapless, if it can't then they should remove the L from the acronym.

AAC?

He's referring to Apple Lossless.

I've got all my music encoded as Apple Lossless and I'll be darned if I can hear any pauses now.

Between gapless playback, the new cover flip view, and the set-top box that will be coming out in Q1 '07, I'm a happy camper with no plans to miss my Monkey's Audio or FLAC days.

Jay

Great info. Great news. Thanks for the info on 4th G.

FYI: The first time I ran iTunes 7... it walked the entire library and "determined gapless playback information." Yes, it included LAME generated MP3 files. It was fairly slow. I would love to know what the logic is for "determining gapless playback" for various types of files... Did they really use the LAME padding/delay info when available?

If you read part of page 2, you'll see that it appears that iTunes is appending some proprietary info to each file to help iTunes specifically play without gaps.

This is not always true. I have added a bunch of LAME 3.90.3 APS files to the library, and it does not modify the files at all (provided you turn SoundCheck off). It certainly seems like it is using the LAME padding/delay info. I don't believe it is just some sort of silence detector - I played a file with more than a minute of silence at the end, and it plays the full silent part.

That doesn't explain what it does to generate the additional gapless info. This is pre-existing content, not ripped using iTunes. I can also confirm it does not modify the MP3 files... they are all untouched. So, the extra info gets stored in the iTunes database.

I was pointing out that if they removed the L from ALAC you would have AAC.

Ah. Gotcha.

Jay

If anyone gets it working on a 3G, let me know. Thumbs down over here.

The gap in (lossy) audio files is a consequence of padded datas which were added by the encoder in order to fill the last frame (and also the first one I suppose). As a consequence, any clever gapless system based on silence detection shouldn't look for null audio for more than a few miliseconds and therefore will keep some long silent parts of music located at the beginning and the end of each file.

I installed iTunes7, upgraded my 5G iPod to firmware 1.2, ripped a gapless CD with EAC to Lame MP3 -V2 --vbr-new, imported the tracks to iTunes, checked the checkbox for "part of gapless album," and synced to my 5G. Tracks are played gaplessly perfectly!

I set the Gapless option on my entire library of Nero AAC encoded tunes and updated my iPod to the v1.2 firmware--NO GAPLESS (shorter delay between tracks now).

So I figured I'd find an album with a lot of merging tracks and encode it with iTunes' encoder. It plays back in iTunes gaplessly. Hmm... curious. So I synced it with my iPod and, lo and behold, juicy gapless-goodness.

...Apple's trying to make me use their codec... and I've been so loyal to Nero.

I think a new listening test's in order.

But do you need iTunes for the actual iPod to playback gaplessly with the new firmware?

I use Winamp which will write to my iPod, all my MP3s are LAME encoded and gapless so can I just drag these onto my iPod without going through iTunes stupid crap?

What's wrong with iTunes? I've heard it is very convenient (or do you prefer, mmm... Complicated? Sophisticated?).

I've just always used Winamp, I'd be willing to give iTunes a second chance I guess I just don't want to be forced to use iTunes.

Hello everyone,
I've recently had my 60GB photo stolen, so I was considering buying a new thing. This gapless news is great!
Can anybody confirm whether an old ipod video would be gapless-enabled after firmware/itunes update?
And BTW, what about 3rd party ipod managers such as yamipod (having it on your ipod is such a delight....)
In general - does buying a new (6G? 5G+?) ipod makes sense, considering i'm not that much interested in video as such? I insist that if i get a thing that is not (fully) gapless, while the RIGHT STUFF is available, will make me sink into depression for years (well, actually just until the new one gets lost/stolen ;-).
Thanks,
Ivan H

Yes.

Okay okay... great Apple... but one very interesting question:
Battery life = stuttering problem?

On a similar note, what about 4G grayscale?

Unfortunately, none of the 4G Pods

THIS is why I use rockbox.

I roughly summarize as follows:
iTunes7 support true gapless playback for MP3 encoded by LAME or AAC by iTunes7, and quasi-gapless playback for lossy encoded by the other encoders (including neroAacEnc).
Activate "Gapless Album" only when you'd like to avoid crossfade playback. Gapless playback is always on by default.

Visible gap will expain this briefly.


Current iPod series that support gapless playback are 5G and newest iPod suite(?). Old nano don't. 4G?

Any info to supplement?

It works with my 4th Generation iPod Photo (later versions were called iPod Color)

The Big Question still is: Does it use Lame gapless info?

It's obvious it doesn't use the Lame header to store the result of iTunes' own analysis because the average user will have more than just Lame encoded MP3's. It could still read the Lame header though and use that info for gapless playback (and copy it to the ID3 field as well, where it stores the result of its own analysis).

If I drag an album encoded with Lame 3.97b2 into iTunes the gapless analysis goes extremely fast. Could iTunes make use of the Lame header instead of analysing the file?

Is there a way to test this? Or perhaps someone knows the answer already?

Edit: haregoo, you actually say it does use the Lame header supplied data?

Which firmware & model number ? I have an M9830B running 1.2.1 and it does not work. No one on the relevant Apple Discussions thread has had any luck either.

Absolutely.
Is there a way to know exact same length as original other than using LAME tag or something similar?

Not that I know of, but are you sure it's the exact same length or could it be a very good estimate?

You know iTunes support encode. Decoding to wav is also supported.

It took me a while to understand what you meant by this but I see your point now!

Thanks haregoo! Well, it appears Apple did a rather good job on this one. Perhaps they'll introduce reading the Nero encoder info in a later version, that wouldn't hurt.

I have the original iPod Photo 60 GB, which is M9586X with firmware 1.2.1

The files that have worked gapless are AAC/M4A files encoded with various versions of iTunes/Quicktime. Some as old as iTunes 4.7 (Quicktime 6.5.2), but most iTunes 6 (Quicktime 7.x).

I've checked it on more albums including studio recordings, and live albums, and it has worked everytime.

You could create a copy of an MP3 file with altered enc delay/padding settings (via Foobar) and see how it affects iTunes' output.

It's strange that an older model should be upgradeable and the later ones not. I'm sure there'll be plenty of discussion of this here and there but I suppose your case makes the possibility of a suitable upgrade for later 4G's realistic, if unlikely.

I read that thread over at the official support forums, I don't think people are turning gapless on. They are all assuming that the initial scan of your iPod makes it work. But you have to then select all the tracks in gapless albums, right click to bring up the Multiple Item Information screen, then turn on Gapless Album at the bottom right hand corner. By checking the box, and selecting YES.

I read a lot of posts complaining but I can't see any that follow those instructions.

You're right, that would be a good test. Can't do it though because I haven't got a Windows box around at the moment, hence no Foobar...

The question seems answered by haregoo though, it uses the Lame header info.

I followed, I'm complaining and I want my wife's 5G.

Isn't that dangerous

No, she'll have to sleep sometime and I'll take my chance.

It is a bloated CPU hog POS

So I've heard reports by a lot of people saying that gapless works perfectly on all tracks, but no matter what I do, I always get a mild tick. Has it worked perfectly for anybody yet?

Axon: Have you tried ripping a disc with iTunes?

I'm getting intermittent gapless playback. Dark Side of the Moon (transcoded from Musepack to Nero AAC) has, for me, a click between the first two tracks, and none between the second two. It seems to be doing gap analysis or something rather than parsing gapless data with Nero AAC.

I ripped a CD with iTunes to ALAC and it played back perfectly.

On an album cut from mp3/cue using pcutmp3, I got perfect gapless playback in iTunes, but not on my 5G iPod. Hm.

It's definitely much better than truncating the last couple seconds of a track like previous versions did.

/mnt, iTunes doesn't seem to use that much more CPU than foobar2000 during playback. It is an elegant, easy-to-use interface that simplifies everything yet remains reasonably powerful. It is certainly no "POS". Just because you have a different preference does not invalidate the reasons that other people have for choosing iTunes. Please troll Apple products somewhere else.

TheQat, nice album!

Running out of the house this morning, I randomly added a known "gapless" album (Global Communication's 76 14) into my "daily plan" playlist, which I sync every morning while guzzling coffee, brushing my teeth and letting the dogs out for their morning stretch/piddle. I didn't remember anything about how I'd encoded it -- I thought that I'd done it with my "usual defaults" (i.e. iTunes AAC 192kbps w/VBR enabled) but it turns out I'd encoded this particular album with Lame 3.93, preset-standard.

The output of id3info:

*** Tag information for 01 4 02.mp3
=== TIT2 (Title/songname/content description): 4 02
=== TPE1 (Lead performer(s)/Soloist(s)): Global Communication
=== TALB (Album/Movie/Show title): 76 14
=== COMM (Comments): ()[XXX]: Encoded by LAME version 3.93 (http://www.mp3dev.org/)
=== COMM (Comments): ()[eng]: Encoded by LAME version 3.93 (http://www.mp3dev.org/)
=== TCON (Content type): Electronica/Dance
=== TRCK (Track number/Position in set): 1/10
=== COMM (Comments): (iTunNORM)[eng]: 0000030D 00000375 00000A7F 00000A46 0002943B 0001A52B 00006EF2 0000613D 0002BA97 0001D500
=== APIC (Attached picture): ()[, 0]: image/png, 146802 bytes
=== TYER (Year): 1994
=== COMM (Comments): (iTunPGAP)[eng]: 1
*** mp3 info

I can't hear any audible gaps with a 5G iPod w. 1.2 firmware.

Okay, I would just like to summarize:

1. Finally Apple has fixed the gaping bug in the iPods that prevented them from playing an album as it was intended. By most accounts this works well for most people but isn't perfect for all.

2. They have released some simple games for the iPod at 5 USD a pop.

Woohoo! No wonder everyone is getting so excited! That is soo cool!


But hang on....

For a while now you have been able to install Rockbox on the thing and get true gapless playback, support for a much larger range of formats (including ogg!), a whole bunch of free games including the ones Apple are now trying to sell like Tetris and Asteroids, as well as ones Apple will never touch like Doom and chess. Add to that Rockbox's support for themes and the ability to use your iPod without having to go near the bloated beast iTunes and I can't see what the hell everyone is getting excited about.

The *only* reason to use the retail OS on an iPod is for playing apple video files; and if you use Rockbox and want to do that then it takes something like 10 seconds to boot back into the retail OS to watch the file.

So Apple have finally fixed a major outstanding bug in their OS (in a rather convoluted fashion, by the sound of it), and added the ability to *buy* games for it. They still have a long way to go before they can claim to provide the best iPod operating system.

Now what you made my do I have to install itunes again and make a screenshot with the default settings to prove my point. Using about 6 - 8 and sometimes 13 percent of CPU power just to decode a MP3 while other players like Wimap, foobar2000 and WMP rarely use about 1 or 2 percent of the CPU. I am not trolling but it is abit unaceptable for music player to use that much resource .

A screenshot of iTunes 7 (with default settings) + Task Mannger playing a gapless album:

Doesn't running RockBox on a 5g cut your battery life in half (or worse)? Seems like there's at least one more reason to run the retail firmware...

Unless you're one of the 95% of iPod owners for whom the Rockbox Manual might as well be written in the Indus Valley script

Task Manager's "cpu usage" report is quite unreliable. "Unaceptable(sic)" is quite a subjective quality. iTunes provides much more functionality than just MP3 decoding. Granted, its MP3 decoder uses more CPU than foobar2000 or Winamp's. It still less than 90% and multitasks just fine.

The whole topic is completely off-topic. If you want to bitch about iTunes having a slow MP3 decoder (despite the irrelevance of its slowness, especially on modern multi-core systems), the place to do so would be the iTunes 7 thread.