The Logic Off Topic list

--- In logic-ot@y..., "Gilles Ruppert" <gilles@l...> wrote:
> Sorry to tell you Sascha: wrong, see my other post. if you don't believe me
> check out Ken C Pohlmann, Principles of Digital Audio. You might not hear
> the mistake, but it comes up & you will see it when you analyse it. More
> steps & you will start to hear the mistake (I speak about A LOT MORE
> PROCESSES).  It becomes eventually audible in Fade outs or low volume
> passages.

It is true that you should avoid normalizing, when afterwards you are going to apply A LOT of processing. But I wonder if it's not the processing that will mangle the original sound more than the error due to calculations.
 
> OK. Imagine this: you have a 3 bit soundfile, which gives you 8 steps. The
> volume of your sine tone is "7". You reduce the volume by -6 dB (normalizing
> does nothing else than changing volume), though half as loud. 7/2 = 3.5, but
> how will it be showed, as 3.5 does not exist? It is either 3 or 4 in the
> digital domain & voila our rounding error & quantization noise. Well with 16
> bit & 24 bit, the resulting noise is of course of much lower volume, because
> the steps are far smaller. Of course you are also working with a complex
> wavefile & the result will be different for every cycle & at very low
> volume, but believe me, it has a reason why Mastering Engineers do not
> normalize!

This is a wrongly explained example that is used to justify to not use normalization. When normalizing, floating point calculus is used, and not the way you explain it above. When using floating point calculus, the mistakes will be much lower.

One of the reasons why not to use normalize is mentioned on the digido.com site. Here is a quote from one of the articles:

"The Myth of "Normalization"
Digital audio editing programs have a feature called "Normalization",
a semi-automatic method of adjusting levels. The engineer selects
all the segments (songs), and the computer grinds away, searching
for the highest peak on the album. Then the computer adjusts the
level of all the material until the highest peak reaches 0 dBFS.
This is not a serious problem esthetically, as long as all the
songs have been raised or lowered by the same amount. But it is
also possible to select each song and "normalize" it individually. Since the ear responds to average levels, and normalization
measures peak levels, the result can totally distort musical values.
A compressed ballad will end up louder than a rock piece! In short,
normalization should not be used to regulate song levels in an
album. There's no substitute for the human ear."

It's a much more pratical reason not to normalize.
Hope this clears s things up a bit. Mastering is still an art on itself, just like mixing is, or composing. Tell me about it, I've been trying to master a track of mine for a couple of months; still don't like the result, but I'm getting there, :-).
Yoonchi.