Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: iPod MP3 Encoder vs Foobar and LAME 3.98.4 (Read 8099 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

My goal is to be able to play gapless MP3s on my iPod nano 3g. Ripping CDs to MP3 using iTunes produces perfect results. However, my preference for archiving and downloading is FLAC, which I then intend to convert to MP3 using Foobar, which is also, in my opinion, better at structuring the directories, filenames, and tag information than is iTunes.

The problem is that MP3s encoded from FLAC using Foobar come out with a two byte difference in filesize, and don't play perfectly in my iPod. The difference is almost undetectable, but nonetheless while gapless MP3s encoded using iTunes play perfectly, the gapless MP3s produced by Foobar produce an ever-so-slight popping noise on track change. It's just audible enough to be completely annoying to me. The pop only happens on my iPod, it does not occur when the files are played in iTunes, Foobar, or Winamp.

Using the Fight Club Soundtrack CD as an example of a gapless album, I'm ripping tracks 1 and 2. I've used MP3tag to strip all metadata from the files and named them exactly the same to minimize the differences in filesize, but the 2 byte discrepancy remains.

Using the Foobar 1.1 and LAME 3.98.2:

Ripping to FLAC and then to MP3 320kbps CBR produces the pop.
Ripping directly to MP3 320kbps CBR produces the pop.
Both methods produce files of exactly the same size.

Using iTunes 9.2.1.5:

Ripping to MP3 320kbps CBR produces perfect transitions.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #1 –
Does iTunes scan for gapless info when you add the converted ones to the library?

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #2 –
Quote from: probedb on
Does iTunes scan for gapless info when you add the converted ones to the library?


I'm not sure. I do know that checking "part of a gapless album" and selecting either "yes" or "no" has no effect either way for any of the above encoding methods. Selecting "yes" doesn't remove the pop for Foobar MP3s and selecting "no" doesn't produce it for iTunes MP3s.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #3 –
That option is misleadingly named. All it does is disable crossfading when checked.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #4 –
So I guess the real question is: what's different about the iTunes MP3 encoding method that Foobar and LAME just aren't doing?

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #6 –
Quote from: Apesbrain on
Somewhat an old thread but may provide a hint:

http://forums.ilounge.com/itunes-mac-pc/17...pless-info.html

Do newer releases of iTunes have a way to force recalculation of gapless info?


Thanks, I'll take a gander. In the meantime, here are some quick stats - Format Info read by Winamp. Both Files were stripped of all tags.



01 - Who Is Tyler Durden.mp3, ripped using Foobar and LAME:

Payload Size: 12128130 bytes
Header found at: 1044 bytes
Encoder Delay: 576, Zero Padding: 744
Length: 303 seconds
MPEG-1 layer 3
320 kbps, 11607 frames
44100 Hz Joint Stereo

CRC: No, Copyrighted: No
Original: Yes, Emphasis: None



01 - Who Is Tyler Durden.mp3, ripped using iTunes

Payload Size: 12129176 bytes
Header found at: 0 bytes
Length: 303 seconds
MPEG-1 layer 3
320 kbps, approx. 1450 frames
44100 Hz Stereo

CRC: No, Copyrighted: No
Original: No, Emphasis: None

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #7 –
It is somewhat difficult to determine if iTunes is scanning files for gapless playback information when adding only a couple to your iTunes library.  Instead, I suggest you add a large amount of tracks (ie more than 20) and see if iTunes scans them.

I have many Lame 3.97, 3.98, and 3.98.4 encoded files/albums that playback gaplessly in the most recent version of iTunes, my iPad, and my 120GB iPod classic..  They are encoded using various settings (-V 2, -V 0, and 320kbps CBR seem to be the common setting) but iTunes scanned all of them for gapless playback info.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #8 –
Quote from: kornchild2002 on
It is somewhat difficult to determine if iTunes is scanning files for gapless playback information when adding only a couple to your iTunes library.  Instead, I suggest you add a large amount of tracks (ie more than 20) and see if iTunes scans them.

I have many Lame 3.97, 3.98, and 3.98.4 encoded files/albums that playback gaplessly in the most recent version of iTunes, my iPad, and my 120GB iPod classic..  They are encoded using various settings (-V 2, -V 0, and 320kbps CBR seem to be the common setting) but iTunes scanned all of them for gapless playback info.


iTunes is indeed scanning files for gapless playback.

The pop I'm hearing is exactly the same as the little pop I hear when seeking to a new point in any track. I get the seek pop with files from both encoding methods.

I thought maybe it was the Joint Stereo, but that wasn't it. My eyes are on the iTunes MP3 header found at 0, while the LAME MP3 header is at 1044...

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #9 –
I would guess the pop you are hearing is caused by fb2k/LAME using proper gapless measures (i.e. store encoder delay and padding to the file), while iTunes do something like crossfading at track boundaries. All because the iPod ifself is unable to play gaplessly. But I'm just guessing, anybody who knows more is welcome to prove me otherwise.

Anyway, you might try Case's Fake Gapless DSP.
Full-quoting makes you scroll past the same junk over and over.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #10 –
Looking at 02 - Homework.mp3 in Goldwave shows that both versions are exactly the same length, but the waveform in the iTunes file starts sooner and ends sooner.

The waveform in the iTunes MP3 starts at 00:00:00.023968214.
The waveform in the LAME MP3 starts at 00:00:00.051179143.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #11 –
Quote from: Yirkha on
I would guess the pop you are hearing is caused by fb2k/LAME using proper gapless measures (i.e. store encoder delay and padding to the file), while iTunes do something like crossfading at track boundaries. All because the iPod ifself is unable to play gaplessly. But I'm just guessing, anybody who knows more is welcome to prove me otherwise.

Anyway, you might try Case's Fake Gapless DSP.


That seems to have done the trick. Thanks for that.

Out of curiosity, do you think that this DSP would have undesirable effects on any not-particularly-gapless tracks?

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #12 –
Alright, now here's something strange.

Moved onto a different album of mine. And One - Agressor.

The transition between tracks 2 and 3 are perfect, once again, with iTunes.

Using Fubar and LAME, with or without the Fake Gapless DSP, it actually cuts off a millisecond of sound so that the transition seems to jump ahead.

Why would the encoding produce different results from different albums? Or is that a silly question?

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #13 –
Gah.

You know what? Screw all this noise. Apparently the best solution is just to scrap MP3 and to get iTunesEncoder and plug it into Foobar.

Encoding to 320 kbps M4A circumvents all my issues. The only catch was that I had to disable Hyperthreading in order to make it work right. Fortunately, I still run on a Pentium 4 so that wasn't exactly a sacrifice.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #14 –
You can also use qtaacenc (I think that is what it is called) for encoding with QuickTime AAC in foobar2000.  iTunes uses QuickTime for encoding and decodnig purposes so it would be the same.  The only difference is that you would have access to more features, particularly true VBR encoding through QuickTime.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #15 –
Quote from: kornchild2002 on
You can also use qtaacenc (I think that is what it is called) for encoding with QuickTime AAC in foobar2000.  iTunes uses QuickTime for encoding and decodnig purposes so it would be the same.  The only difference is that you would have access to more features, particularly true VBR encoding through QuickTime.


Cool, thanks for that. It's good to have alternatives, especially if I ever need hyperthreading again or upgrade to a modern processor, I assume.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #16 –
Quote from: Yirkha on
I would guess the pop you are hearing is caused by fb2k/LAME using proper gapless measures (i.e. store encoder delay and padding to the file), while iTunes do something like crossfading at track boundaries. All because the iPod ifself is unable to play gaplessly. But I'm just guessing, anybody who knows more is welcome to prove me otherwise.
Are iPod/iTunes definitely unable to do true gapless? I thought they might have implemented a way to deal with LAME tags if present, and fudge 'fake' gapless otherwise.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #18 –
Thanks for the link!

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #19 –
Have you tried foo_dop to transfer the files with foobar to the iPod? It says it supports gapless playback for lame encoded files.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #20 –
If Apple's own iTunes, for whatever reason, is not producing proper gapless playback on the iPod, I doubt a third-party component will.

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #21 –
I'd like to see some recorded samples of the iPod output.  Assuming the implementation in the iPod isn't broken, the only explanation I can come up with is that there is a discontinuity between the end of the previous track and the beginning of the current track in amplitude.  There should not be a discontinuity in time.  We discussed this fairly recently, though it wasn't centered around hardware playback.
 

iPod MP3 Encoder vs Foobar and LAME 3.98.4

Reply #22 –
The M4A Solution using iTunesEncoder in Foobar2000 is working perfectly for me, plus it helps me differentiate between files I've encoded directly from FLAC and all of my old MP3s that came from older ripping programs and downloads.

If I get a little free time, I'll make a recording of the output from my iPod that demonstrates the problems that I was getting before. (I got a little "pop" on the Fight Club soundtrack, I got an actual skip from the And One album.)