Page 1 of 1

Normalization. Explanation and use?

PostPosted: Thu Jan 27, 2005 7:36 pm
by vinylmaster
Hello once again.

Let me get this straight: The normalization of a waveform in Wave Corrector essentially allows the music to have a reasonably consistent audio playback?
The confusion comes in when I go to burn the waveforms onto the CD. I use Nero Express 6. Nero also provides a normalization choice but they also include the ability to increase the disc's volume. (Highlight the tracks you want, click "properties" and then select the "filters" tab. The choices for filtering are "RMS" and "Maximum". I use "RMS" but will change the default from "20" to perhaps "22" or "23".)
Is the process I described similar to increasing the amplitude of the waveform, thus allowing the disc to sound louder? This works for me at work or in the car.

PostPosted: Fri Jan 28, 2005 10:15 am
by Derek
Yes, normalisation can be a little confusing!

You are correct that the idea behind normalisation is to equalise the volume of your recordings. However, it is important to understand how and why this is done.

Recording media have a 'dynamic range' - this is the range between the maximum and minimum signals that can be represented. If you attempt to record above the maximum then the signal is 'clipped' and you will hear distortion. If you go below the minimum then the signal will be lost in the background noise.

Therefore, when you make a recording, the volume should be as high as possible within the dynamic range. That way, the quiet parts of the recording will stay well above the noise floor. A problem arises because when you start your recording you don't know in advance where the loudest part will be and exactly how loud it will be. Therefore you give yourself some headroom to allow for an unexpected loud section. If you give yourself too much headroom then the recording will sound quiet compared to your other recordings. This is where normalisation comes in. After the recording is finished, the program can scan it to find the loudest point and the digitally adjust the volume so that it is a standard level (typically 1dB below clipping level; this will match most commercial CD's).

This is what Nero Express does in its 'Maximum' mode and it is also what Wave Corrector does.

Nero's RMS mode is a little different. This sets the 'average' volume of the recording rather than the peak volume. To understand how this is different consider a recording which is very quiet for most of its length but with a loud crescendo at the end. The RMS normalisation will increase the volume of such a recording because the average volume will be quite low. However, there is a danger if you do this, that the peak at the end will overload and cause distortion. For this reason, it is not recommended.

PostPosted: Sat Jan 29, 2005 12:45 am
by adaywayne
Derek did his usual excellent job of explaining "normalisation". May I be allowed a little addendum?

Normalisation is a very subjective thing. For example, I have just finished transferring a 1958 mono version of the "Messiah" from LP to CD. Now, the Messiah has dozens of "tracks". If I had decided to normalise each track to the same -1dB level then, "Behold The Lamb of God" (at its peak) would have sounded as loud as the "Hallelijah Chorus" (at its peak). Is that what the performers, conductor, and the recording engineer really intended?......I doubt it very much.

So, my solution was to record the whole 2+ hours to one single wave file, and then to "normalise" that wave file to a peak of -1dB. That way, I now have a CD version that is as close as possible to what was intended (dynamics-wise) when the recording was made.

Of course, with today's pop "music", just normalise to +20dB and it will sound just the way you like it!!!!!!!!!
Arnie

Messiah (OT)

PostPosted: Fri Feb 04, 2005 12:54 am
by paul
Did you realize that Handel managed to compose the whole of the Messiah in just 3 weeks!

A follow-up question

PostPosted: Fri Feb 04, 2005 3:44 am
by vinylmaster
Thanks everyone for the advice.
Referring to the comment I made when I burn my CD's in Nero Express (The default level is "20" and I have been raising that to between "21" to "23", depending on the original music) is I generally play the resultant music at work or while driving. It helps somewhere to have a "louder" CD to overcome ambient room or road noise.
I agree that this is not the best thing to do. Some of the albums have an overdriven quality. No harm done, just a bunch of wasted CD-R's. Prior to this year, I didn't know the normalization could be done in Wave Corrector. (So much for the "Master" part of my name, eh?)

Now for the next question: Is it possible to normalize a single track instead of the entire album?

Re: A follow-up question

PostPosted: Fri Feb 04, 2005 4:08 pm
by adaywayne
vinylmaster wrote:Thanks everyone for the advice.
Referring to the comment I made when I burn my CD's in Nero Express (The default level is "20" and I have been raising that to between "21" to "23", depending on the original music) is I generally play the resultant music at work or while driving. It helps somewhere to have a "louder" CD to overcome ambient room or road noise.
I agree that this is not the best thing to do. Some of the albums have an overdriven quality. No harm done, just a bunch of wasted CD-R's. Prior to this year, I didn't know the normalization could be done in Wave Corrector. (So much for the "Master" part of my name, eh?)

Now for the next question: Is it possible to normalize a single track instead of the entire album?


Only if you split the tracks first, I think. Some other programs have the ability to normalise just a "selected" part of a long wave file.
Arnie