>In a message dated 6/24/2003 12:49:41 PM Pacific Standard Time,
>> I have an opinion on that: never normalize. Never. No need for it. At
>> the last stage of CD mastering, overall levels might be raised so the
>> loudest sounds are close to max or limited close to max, but that's
>> as far as you want to go. In intermediate steps, you need headroom
>> for processing.
>What about normalizing to 50%. I do this on all my recordings before I sta=
>to edit?. Am I making a mistake? I use CoolEdit.
>John V. Moore Nature Recordings
To me, this raises the question about how much bit depth affects
sound quality and, if so, when is it appropriate to change it.
If a sound is saturated in the field, say above 15%*, I agree with
Dan, no need to normalize just to use the file, you can boost
adequately in the mix or even when burning to CD. If a field
recording has peaks lower than 10%, and I need to equalize it, I go
ahead and bring it up as close as I can to full saturation (90-97%)
at the same time (remembering to bring back speaker playback level).
I don't normalize just to normalize. If I do have to equalize a 15%
saturation file, I do the same because there's always a good chance
I'll be doing some more things to it later on. If a file uses all
the bits, it withstands and "holds" alteration better. Most
noticeably, there is less discrepancy in sound quality between
previewing a change like eq and listening to the file after its been
processed. Those working on computers without sound cards might have
The only thing risky about 97% saturation is blowing your speakers.
As long as a file has to undergo a change, one may as well grab the
bit depth. If I need max volume and it acquires harshness-- then I
try to address this problem with eq at the same time. I routinely
mix many sound files all with peaks at 97% or -.3dB. As long at I
do not boost any of them, there's no over-modulation or distortion.
When multiple generations are involved, the mixes made from files
with good bit depth stand-up better and seem smoother. True also from
unchanged field recordings. For a while, it made me nervous to see
all those big fat mix files, but they all come down when burning to
I mention this because it seems bit depth could be a bigger factor
for people working without an audio card or a cheap audio card where
the digital to analoq (D->A) is marginal to start with. Makes sense
that monitoring would be more accurate with more bits to work with.
*These are only ballparkpercentages because it depends on how loud
the particular element will be used in the final sound file or mix.
University of Wisconsin-Milwaukee