The normalization is song-by-song to my knowledge. It doesn't normalize the audio within the song itself. (Over)compression is a key part of the aesthetic of a lot of pop hits today, and while in the face of normalization you won't be able to make your song overall louder than the next, it won't necessarily stop you from creating a "wall of sound" waveform if that's the sound you're looking for.
Absolutely, and nor should it. But what it should mean is that if you make your song mostly quiet, your peaks will be louder than a wall of sound.
Wait, what does that mean for somehow who records music as a hobby?
If my recording is all the time at around -10dbRMS, it will sound as loud to the listener as if it was on -15dbRMS(itunes)?
Also, how exactly does it sound when most of my recording is at -10dbRMS and a couple peaks are at -5dbRMS?
> itunes and youtube normalize everything to -15dbRMS and -13dbRMS respectively.
Are you using "normalize" in the audio production sense, as in adding or subtracting a constant gain to an entire track? I'm pretty sure that iTunes doesn't do this, at least not by default.
http://productionadvice.co.uk/lufs-dbfs-rms/