Listening Test

  • Thread starter Thread starter NL5
  • Start date Start date
Not to mention there are at least two different MP3 codecs, each of which does something diferent at each bit rate and affects different program content differently. If one wants to test MP3 codecs, that should be a different test altogether, but when it comes to testing stuff like dither, MP3 should be left out of the process and testing altogether. There's no way otherwise to know with subtle sounds how much of what you're hearing is an artifact of the MP3 process and how much is something that came before it.

G.
 
I did a wee test as I thought the difference between 16 bit wave and 320kbps MP3 was very small.

Heres 2 samples.

The 1st is the difference between a 16 bit wave file and the same file at 320kbps MP3 using Lame MP3 encoder.
http://www.serotoneband.co.uk/files/misc/Difference between 16bit and 320 MP3.wav

The 2nd is the original 16 bit wave file.
http://www.serotoneband.co.uk/files/misc/Original 16 bit file.wav

I was actually kind of shocked to hear how much 320kbps MP3 takes from the original 16 bit wave file!

Eck
 
I was actually kind of shocked to hear how much 320kbps MP3 takes from the original 16 bit wave file!
If you actually want to hear what's missing, run this test:

1. Load the unaltered WAV file into track 1 of your editor.
2. Load the MP3 into track 2 of your editor.
3. Invert the polarity of the MP3 track (invert the entire waveform).
4. Sum the two tracks together.

The remaining summed sound will be everything that was in the original WAV that was lost in the MP3 encoding.

G.
 
If you actually want to hear what's missing, run this test:

1. Load the unaltered WAV file into track 1 of your editor.
2. Load the MP3 into track 2 of your editor.
3. Invert the polarity of the MP3 track (invert the entire waveform).
4. Sum the two tracks together.

The remaining summed sound will be everything that was in the original WAV that was lost in the MP3 encoding.

G.

Thats exactly what I did in my last post.
Did you not read the full post? tut tut :D

ok maybe I didn't word it too well but I couldn't think of a better way to word my last post without explaining exactly what I meant.

Eck
 
Last edited:
My fault. I misread it. You wrote it fine, I just forgot to take the needle out of my arm before posting :rolleyes:

G.
 
Did anyone take a listen to the short samples?
Anyone feel that it sounds like the 1st sample has a lot more in it than it should for the difference between MP3 at 320 and 16 bit wave?

Eck
 
Did anyone take a listen to the short samples?
Anyone feel that it sounds like the 1st sample has a lot more in it than it should for the difference between MP3 at 320 and 16 bit wave?

Eck
I gave it a listen, Eck, and frankly, no, I'm not really suprised (though I have run these kinds of tests before, so I kind of knew what to expect.)

G.
 
I gave it a listen, Eck, and frankly, no, I'm not really suprised (though I have run these kinds of tests before, so I kind of knew what to expect.)

G.

Cool, suppose thats the first time Ive really heard anything like that so it was a bit of a surprise!
Thanks fro listening.

Eck
 
I think the real issue of this and related threads is how much of this muck is masked or buried under the signal to be audible (not that it doesn't exist). As I mentioned elsewhere, some ears are more prone to being disturbed by noise levels, others by how much audio is compressed, others by unbalanced EQ, etc. It's part of what makes audio production so subjective.
 
Back
Top