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: MP3 encoding mystery size (Read 8261 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

MP3 encoding mystery size

These two files http://screencast.com/t/RTvuewPD3m7 have apparently the same content but one is over three times the size of the other. Can anyone suggest why?

MP3 encoding mystery size

Reply #2 –
Any sensible answers, please?

MP3 encoding mystery size

Reply #3 –
Quote from: chrisjj on
Any sensible answers, please?

If you had performed any but the most cursory of tests before asking others to do your work for you I believe it would have been discovered that one of the files is corrupt and does not contain what it purports to contain.
Creature of habit.

MP3 encoding mystery size

Reply #4 –
Embedded album art?
--------------------

MP3 encoding mystery size

Reply #5 –
Quote from: chrisjj on
Any sensible answers, please?


It was a perfectly sensible answer, just because you neglected to do even the most basic research does not make it a daft answer.

MP3 encoding mystery size

Reply #9 –
Quote from: Soap on
If you had performed any but the most cursory of tests before asking others to do your work for you

You apparently missed the screen shot.

Quote from: Soap on
I believe it would have been discovered that one of the files is corrupt and does not contain what it purports to contain.

You're mistaken. Both contain substantially what they purport to contain.

MP3 encoding mystery size

Reply #10 –
Quote from: capma on
Embedded album art?

Thanks for the suggestions, but MP3tag shows none.

The screen shot shows slightly different lengths. FB2K shows equal lengths, but at the odd value 42:03. As does Explorer. FB2K plays both apparently fine.

MP3 encoding mystery size

Reply #11 –
Quote from: chrisjj on
Quote from: Soap on
If you had performed any but the most cursory of tests before asking others to do your work for you

You apparently missed the screen shot.

I saw it.  Do the math.  Your two tools disagree, one (or both) are wrong.  Both files are not the claimed bitrate and the claimed length.  You have a corrupt file, either at the start (affecting header) or the end (affecting length).
Creature of habit.

MP3 encoding mystery size

Reply #14 –
According to the upper part of the picture, the 1st file has duration 0h 42m 02s. According to the left part -- 3h 04m 45s.

MP3 encoding mystery size

Reply #15 –
I would use command-line mp3 decoder with verbose error messages (I use madplay, but maybe lame is verbose enough too, there's also mpg123) to try to decode them both to .wav and then would look at output of decoder for errors or at the length of resulting .wav files.

MP3 encoding mystery size

Reply #16 –
Perhaps one is MP3-HD and one is not?  Might be a reasonable explanation if you bought it from a store that encodes in that format.

MP3 encoding mystery size

Reply #17 –
Quote from: benski on
Perhaps one is MP3-HD and one is not?  Might be a reasonable explanation if you bought it from a store that encodes in that format.

While that's a good idea as to why one MP3 is 3x the size bitrate*length suggests it should be in general, this is an audiobook.  I'll eat my hat if it's MP3HD
Creature of habit.

MP3 encoding mystery size

Reply #18 –
Try using mp3val and it should tell you what's wrong with each of the files. You can then use the option to fix the problem.

After doing that for the both files, compare the repaired versions...
lame -V 0

MP3 encoding mystery size

Reply #19 –
the left EncSpot window indicates 1914 SyncErrors and clearly shows differing run lengths and number of frames, so why exactly should those files be identical in size ???

MP3 encoding mystery size

Reply #20 –
Quote from: Maggi on
the left EncSpot window indicates 1914 SyncErrors and clearly shows differing run lengths and number of frames, so why exactly should those files be identical in size ???

Because OP is claiming they are the same content therefore the same length (or close enough). 
Creature of habit.

MP3 encoding mystery size

Reply #21 –
I see ... 

MP3 encoding mystery size

Reply #22 –
Quote from: Maggi on
the left EncSpot window indicates 1914 SyncErrors and clearly shows differing run lengths and number of frames, so why exactly should those files be identical in size ???

I think Maggi hit the nail on the head.  I've been toying very little with the MP3 format, so what I know is that the MP3 sync header (0xFF, 0xE0 (masked with same)) (i.e. 11 byte-aligned bits having the value 1) determines the start of a frame.  Anything else can be inside the file (tags, garbage, whatever) and it'll be ignored by the player until it hits a sync header.  So, 1914 sync errors means that there's over a thousand places in that file where non-mp3 frame data could reside.  I would guess that most if it is garbage and a very small amount is probably ID tag information.  My guess is that this is what's taking up the extra space.  If you want more, see here.

Just on an aside, ID3v2 tag information can be littered throughout a file (for chapters, etc) but I'm not sure how many applications will find it beyond the first tag.

If you're really curious, I've been working on an application to thoroughly scan an MP3 file for tags and other "garbage data" just for my own fun.  If you can push this file to me somehow, I'll be glad to take a look.  It seems it'd be a good stress test of my application.

MP3 encoding mystery size

Reply #23 –
This is not a good way to ask the internet to do things for you.
 

MP3 encoding mystery size

Reply #24 –
Quote from: DreamWarrior on
If you can push this file to me somehow, I'll be glad to take a look.  It seems it'd be a good stress test of my application.


Thanks. You have a private message.