[Mike Rivers]

jtougas wrote:
In my pursuit of improving my recordings, I've taken a few of my mixes
to a friend of mine, to see what he thinks, and I mentioned I'd
normalized one or two of the channels in the tracks since they were a
little quieter than the others.

He looked at me with a fairly shocked expression and said "Oh, man,
you should never normalize!"

So I'm asking y'all: why not?

What normalizing does is locates the loudest point in the file (or
segment or track, however you do it) and, if it's not already at the
peak digital level (or whatever point you set if your program gives you
that option), it calculates how far away from maximum level it is, and
just adds that much gain to the whole file. In essence, it's the
equivalent of turning up the playback volume without the listener having
to touch a control on his player.

Normalizing is often frowned upon by pseudo (and some real)
"professionals," not because it does any damage, but because it
shouldn't be necessary because you should have set the record level so
that it reaches the desired playback level. The damage, if any, has
already been done in the recording process, and you're not making
anything any better by normalizing . . . except for one thing - the
listener is insulated from your "mistake" and doesn't have to turn up
the volume to hear it at at the same level as the last thing he listened
to.

There is, however, some validity to not normalizing to full scale (0
dBFS) on general principles. It has to do with the characteristics of
the player. Many (in fact most) D/A converters increase their distortion
in the last 1 dB or so before full scale. However, this is still a
pretty small amount of distortion. If you're inexperienced enough so
that you feel that it's necessary to normalize to get the playback level
up to the point where it won't make the listener want to turn it up,
there are probably worse problems with your recordings.

So, in summary, it's better to do it right so that you aren't tempted to
normalize, but if you missed the mark, most listeners would rather have
a little more noise in the playback than get off the couch and adjust
the volume control. It's kind of the poor-and-sloppy man's mastering.
How effective normalizing is toward making a recording subjectively
louder is a function of the difference between the peak level and the
average level. If there's one drum hit that's just 1 dB below peak
level, but everything else is 15 dB lower, you won't hear a significant
volume change when you normalize, because all that will happen is that
the overall level will ony be boosted by 1 dB. There are less simplistic
ways of making a recording sound louder, by actually increasing the
density, rather than just the peak level, of the recording. That's most
often part of what's called "mastering" today.