I'm in the camp that will not normalize as part of the workflow. There
is a time and place for everything and if you plan to have a Mastering
Engineer involved, do not send a normalized file. Letthem do the gain
adjusting. As far as level matching, I would agree that a simple
increase or decrease in gain is the solution for the particular problem
of channel matching L to R, if it is not more than 3dB or so. The reason
for the caveat is that increasing the overall gain also increases the
noise by the same amount and 3dB doubles the overall gain. So, listen
carefully to the result and make sure that it is acceptable.
On the subject of normalizing: iZotope has one of the best normalizing
algorithms IMHO. When you execute the process, the results are usually
quite good because iZotope seems to be better at maintaining the
dynamics better than other software. When I do use the process, because
normalizing reads the waveform peaks (Lou is correct), I will often
reduce the peaks individually if there aren't too many and, you'll get
more bang for your buck.
Corey Bailey Audio Engineering
On 3/4/2018 8:18 PM, Lou Judson wrote:
> Yes, that is good! Normalizing shouldnít alter left/right balance, and you should be able to read the peak level from a whole file, and raise the level by a suitable amount.
> Lou Judson
> Intuitive Audio
> On Mar 4, 2018, at 7:13 PM, Martha Forsyth <[log in to unmask]> wrote:
>> I have been avoiding normalizing - instead boosting by a few dB when necessary. Somehow I'm afraid that normalizing will somehow homogenize the sound - sometimes one channel is lower than the other for realistic reasons (I'm working with field recordings) and I don't want to risk tampering with that.
>> Does this sound like reasonable thinking to you experts?
>> Martha Forsyth
>> On 3/4/2018 7:44 PM, Lou Judson wrote:
>>> So donít normalize!
>>> (Doctor. it hurts when I do this! Well son, donít do that!)
>>> Lou Judson
>>> Intuitive Audio
>>> On Mar 4, 2018, at 4:33 PM, Steven Smolian <[log in to unmask]> wrote:
>>>> Final result.
>>>> I can keep the sound as I go from 32 to 24 to 16, with a slight less overall
>>>> air around the sound at 16. If I normalize, I get the crackles.