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: Question about MP3 decoding (Read 730 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Question about MP3 decoding

I don't use MP3 much, but I was wondering about the output stream.
Does it in fact decode to 16/44 PCM as input to the DAC regardless of whether it is VBR, 320, 128 or whatever?

Since DACs take a PCM bitstream (or in some cases DSD), I would assume that must be the case, right?



Re: Question about MP3 decoding

Reply #1
For bit depth, the decoder has to do with this. Theoretically there is no limit. I'd assume that most decode to 16 bits per sample.

For sample rate, the streams themselves have the sample rate data.

Re: Question about MP3 decoding

Reply #2
It depends on the decoder.   If you open in Audacity you get 32-bit floating point.  Audacity uses this format by default and the clue that it's not converted to regular (integer) 16 or 24-bit first, is that lots of MP3s happily go over 0dB.*

The sample rate will remain unchanged.   I have quite a few 48kHz MP3s that came from video.




* One side-effect of the lossy compression is that the wave shape changes, making some peaks higher and some lower.   The MP3 often ends-up with peaks around +1dB, even if it was ripped from CD which is limited to 0dB.

Re: Question about MP3 decoding

Reply #3
I don't use MP3 much, but I was wondering about the output stream.
Does it in fact decode to 16/44 PCM as input to the DAC

It seems you are thinking not of conversion to file, but of decoding for immediate playback. Then your player likely has a volume control between the "decode" and the output. And not unlikely, that works in floating-point. So then the player could as well make sure it is decoded to float.
Only later in the process it is sent to the DAC to become something you can actually listen to.


And here is a year 2000 test for > 16 bit accuracy: http://mp3decoders.mp3-tech.org/24bit.html

... and a test from who-knows-when: https://www.underbit.com/resources/mpeg/audio/compliance

Re: Question about MP3 decoding

Reply #4

Quote
It seems you are thinking not of conversion to file, but of decoding for immediate playback. Then your player likely has a volume control between the "decode" and the output. And not unlikely, that works in floating-point. So then the player could as well make sure it is decoded to float.
Only later in the process it is sent to the DAC to become something you can actually listen to.
yes exactly. Thanks for the info.