Scintilla wrote:No, that is NOT normalization, that is <i>dynamic range compression.</i>
Normalization scales everything evenly to fit a given peak sample value -- it either makes everything louder or makes everything softer. It is not for evening out levels between parts of a track.
My post was also probably equally confusing to some as well so I shoudl clarify. I also mentioned dynamic range in one of my posts but once again to re-iterate Scintilla...
Normalization does not affect dynamic range. In layman's terms probably the easiest way to explain it is you are setting a new maximum loudness level for your audio file and based on the loudest signal in your audio everything it either brought up to the new max or lowered.
Example: You set new loudness max = -3 dB. Your current audio loudness max is -0.3 dB. Now your entire audio clip is lowered -2.7 dB. The differentials in loudness between parts of the same clip have not been altered. The only difference is the ceiling for loudest possible signal has changed.
Why do this? So that you don't have jarring maximum loudness differences between tracks (which is what happens with AMV submissions). I have had entries with peak of like -0.1 dB and then another we want to run right after it with a peak of -14.0 dB. And both are loud rock music. You can obviously see the problem.
Why did I incorrectly allude that dynamic range can still be lost when adjusting the audio? Because typically if there is clipping you are also applying DRC or some other filter to compensate for things if you are an amateur like myself who is just running some basic sound plugin.