So Apple Lossless isn't doing it for me

[Galley]

1. Variable bitrates depend on the complexity of the music. I have lossless files of acoustic blues songs with bitrates less than 400Kbps.

2. Hidden bonus tracks can give you unusually low bitrates. As an example, I recently ripped a track that have about 4 minutes of silence in it. The resulting bitrate was 500Kbps. Once I edited the track's stop time and re-imported it, the bitrate was 850Kbps.

 

[Teej guy]

OP: why are you using VBR if you want Lossless?? that doesnt make any sense to me. didnt even know you could use VBR with Lossless!!!

Lossless codecs use VBR, there is no CBR mode.

 

[DoFoT9]

Lossless codecs use VBR, there is no CBR mode.

well that is just OUTRAGEOUS!!!

its LOSSLESS! not LOSSY!

*stops using lossless*

im sticking to FLAC and AIFF

EDIT: wait are you sure?? are you sure it just doesnt use compression on the files whilst maintaining the 'lossless' quality?

 

[Teej guy]

well that is just OUTRAGEOUS!!!

its LOSSLESS! not LOSSY!

*stops using lossless*

im sticking to FLAC and AIFF

EDIT: wait are you sure?? are you sure it just doesnt use compression on the files whilst maintaining the 'lossless' quality?

...have you even read the thread?

Lossless codecs don't use bits in the same way a lossy codec does. A lossy codec uses psycho-acoustic modelling to put together something that sounds transparent to the source.

A lossless codec is basically just compression, but using various methods depending on the codec. However, one thing all lossless codecs share is that they are lossless (how many times do we need to say this?)

That's why the thread starter's example of an 18 minute track is 48MB or something. Much of that is a silent gap between the end of the track and a bonus track. The FLAC doesn't need to spend bits on digital silence, so it doesn't. Variable bit-rate. It uses more or less bits depending on how many are needed to reconstruct the source material perfectly/bit-for-bit.

 

[DoFoT9]

...have you even read the thread?

Lossless codecs don't use bits in the same way a lossy codec does. A lossy codec uses psycho-acoustic modelling to put together something that sounds transparent to the source.

A lossless codec is basically just compression, but using various methods depending on the codec. However, one thing all lossless codecs share is that they are lossless (how many times do we need to say this?)

That's why the thread starter's example of an 18 minute track is 48MB or something. Much of that is a silent gap between the end of the track and a bonus track. The FLAC doesn't need to spend bits on digital silence, so it doesn't. Variable bit-rate. It uses more or less bits depending on how many are needed to reconstruct the source material perfectly/bit-for-bit.

ok i see now. lossless is lossless 😛

i dont get why it is VBR but, i thought it would be a constant quality, because its lossless regardless of the fact that if there is a whole orchestra or silence playing.. *confused*

 

[TuffLuffJimmy]

ok i see now. lossless is lossless 😛

i dont get why it is VBR but, i thought it would be a constant quality, because its lossless regardless of the fact that if there is a whole orchestra or silence playing.. *confused*

If it were a constant bit rate and lossless then it wouldn't be a smaller file size than the original file. 🙄

 

[DoFoT9]

If it were a constant bit rate and lossless then it wouldn't be a smaller file size than the original file. 🙄

then i dont see how it still classifies as lossless.

 

[TuffLuffJimmy]

then i dont see how it still classifies as lossless.

because the information that is stored in a AIFF file (let's just say this is the original) is analogue. A compressed Apple Lossless file takes that information and expresses it more efficiently with digital expressions that mean what the analogue signal puts out. At least, that's the best I can explain it. Please, for the love of God: quit thinking of it in bit rates!!!

 

[DoFoT9]

because the information that is stored in a AIFF file (let's just say this is the original) is analogue. A compressed Apple Lossless file takes that information and expresses it more efficiently with digital expressions that mean what the analogue signal puts out. At least, that's the best I can explain it. Please, for the love of God: quit thinking of it in bit rates!!!

hhmm ok i get that first part - the file is converted into digital binary makes sense, but what about the sample rates, modulation and things like that? wouldnt they cause some sort of audio loss (not that we would notice, of course).

and oh, im not thinking in bitrates, thats just an average indicator which nobody cares about 🙄

 

[TuffLuffJimmy]

hhmm ok i get that first part - the file is converted into digital binary makes sense, but what about the sample rates, modulation and things like that? wouldnt they cause some sort of audio loss (not that we would notice, of course).

and oh, im not thinking in bitrates, thats just an average indicator which nobody cares about 🙄

lossless |ˈlôsləs; ˈläs-|
adjective
having or involving no dissipation of electrical or electromagnetic energy.
• Computing of or relating to data compression without loss of information.

There is no audio loss, even in things that you wouldn't notice (it's all still there)

 

[DoFoT9]

lossless |ˈlôsləs; ˈläs-|
adjective
having or involving no dissipation of electrical or electromagnetic energy.
• Computing of or relating to data compression without loss of information.

There is no audio loss, even in things that you wouldn't notice (it's all still there)

haha ok thanks for the verification. i got ya now. 🙂

 

[rWally]

hhmm ok i get that first part - the file is converted into digital binary makes sense, but what about the sample rates, modulation and things like that? wouldnt they cause some sort of audio loss (not that we would notice, of course).

and oh, im not thinking in bitrates, thats just an average indicator which nobody cares about 🙄

Another way to think about it...

Say you have an uncompressed .wav file. Since it's digital it's made up completely of 0's and 1's. So, say an uncompressed file looks like this:

000110101111111100000

You notice how you get some repeating numbers there? For instance, there are 8 1's in a row. Wouldn't a more efficient use of space be to say '1, repeat 8 times' rather than writing out 1 eight times? This is sort of how compression works. You find patterns and shrink it down to use less space. All the information is there but it just takes up less space on your hard drive.

What an lossless codec does is find ways to shrink down the file size using something like I described above by finding repeating patterns in the file to reduce the file size. All the information is there to bring it back to it's original, full size and both files still contain the same information. The bit rate is calculated by dividing the total size of the file by the length of the song. This is pretty meaningless when you're talking about lossless because all it's telling you is how efficiently the file was compressed and has nothing to do with the actual sound quality of the song. Lower bitrates on a lossless file simply tell you that there were more repeating 0's or 1's (long stretches of silence, etc.) in the file making it easier to compress.

If it were being converted to a lossy codec (i.e. MP3) you're actually getting rid of 0's and 1's that aren't easily heard (super high and low frequency stuff usally) in order to reduce the file size as well as using the tricks I mentioned above. Once you get rid of the 0's and 1's there's no going back though.

Hope I didn't confuse you too much but I tried to simplify it as much as possible.

 

[DoFoT9]

Another way to think about it...

Say you have an uncompressed .wav file. Since it's digital it's made up completely of 0's and 1's. So, say an uncompressed file looks like this:

000110101111111100000

You notice how you get some repeating numbers there? For instance, there are 8 1's in a row. Wouldn't a more efficient use of space be to say '1, repeat 8 times' rather than writing out 1 eight times? This is sort of how compression works. You find patterns and shrink it down to use less space. All the information is there but it just takes up less space on your hard drive.

What an lossless codec does is find ways to shrink down the file size using something like I described above by finding repeating patterns in the file to reduce the file size. All the information is there to bring it back to it's original, full size and both files still contain the same information. The bit rate is calculated by dividing the length of the song by the total size of the file. This is pretty meaningless when you're talking about lossless because all it's telling you is how efficiently the file was compressed and has nothing to do with the actual sound quality of the song. Lower bitrates on a lossless file simply tell you that there were more repeating 0's or 1's (long stretches of silence, etc.) in the file making it easier to compress.

If it were being converted to a lossy codec (i.e. MP3) you're actually getting rid of 0's and 1's that aren't easily heard (super high and low frequency stuff usally) in order to reduce the file size as well as using the tricks I mentioned above. Once you get rid of the 0's and 1's there's no going back though.

Hope I didn't confuse you too much but I tried to simplify it as much as possible.

lol thanks for the explanation - i already knew most of what you said, but the part on the bitrate in lossless files was very interesting thanks for that. (Y) 🙂

this kind of stuff is very easy for me to understand, would you mind going into deeper detail? or do you have any links lol??