Help needed w/ Normalizing

  • Thread starter Thread starter zip
  • Start date Start date
zip

zip

New member
Can anyone help me understand the difference in Peak normalization and Avg RMS power normalization??

I want to normalize a live recording to cut out a few clips but not sure which mode is best.

With the RMS power mode there are attack / delay times and a scan level function...what do those do and how should I set them??

Any help would be greatly appreciated!!

zip
 
Normalization is to make a sound as loud as it can possibly be (without altering it's wave-shape), so Peak Normalization finds the loudest point in the track, and makes everything slightly louder, so that the peak now is as loud as you can go.

RMS normalization is more complex, but neither type of normalization will really fix your problem. Once you've got a clip, you've got a clip!.. Part of your waveform is bigger than what the computer can store, and that's that! You can make the whole wave-form smaller, but it will just lose resolution, and you don't actually get back the data that's missing because of your clip (you'll get the same clipping sounds you had before).

I'm sure that there are tools to help with this though.. hopefully someone here knows the answer.. I would start by looking for pop and hiss reduction tools.

In the future, keep this in mind -- you're better off getting too low a signal on digital equipment, than clipping. It's not like analog.. it doesn't add warmpth or depth when you clip.. it just sounds like a click.

YEUCH.


- Rich
 
Maybe I wasn't very clear about what I did. The original boot was full of midrange with little low and high end. The overall level was a bit low as well. What I did was to run it through multiple EQ passes removing mid and adding bass / treble. After that I ran it through low level compression. The mix now sounds good but I may have gone too far with the EQ causing the sound to be "saturated" but not really clip. There are a few source tape clips I know I have to deal with...

So do I need to EQ it again to bring the levels down a bit or is their an easier way???

Am I making any sense here??

zip
 
You make perfect sense. But first EQ-ing it one way and then another makes no sense. Get the original mix, and run it through EQ and light compression in one go instead, Basically remaster it. Running it multiple times through EQ might give you loads of phase problems, which in fact may be the cause of why you think it doesn't sound good...
 
Usually, most normalization plug-ins give you the option, to compress the wave audio if it exceeds the Odb limit, which usually happens if you are using average rms. However, the best advice i could say is to make sure you are recording with the hottest possible signal so in the end you are using as little normalization as possible. Remember, normalization adds noise so use it sparingly.
 
Just to clarify a point...

Normalization DOES NOT add noise. It does however, increase the apparent noise floor along with any increase in gain applied to the signal.

A minor, but important distinction...

Bruce Valeriani
Blue Bear Sound
 
Thanks for the input...

I thnk I will try a one pass EQ but I am going to make an assumption. Smack me if I am wrong...LOL.

If I "add" the decible levels of the two EQ passes (ie pass #1 at -1.5dB and #2 at 0.3dB) I could set that frequency at -1.2dB and be done - right?? I would of course need to go and do that for each frequency...

I did try a one pass EQ but I thought I got a better sound with 2 passes - but I may have just got lucky. Good thing I saved all the presets under different file names!!

For the record I didn't add to a frequency with one pass and subtract with the second...it was more like I rounded the entire profile (smile) the first time then cut it (like a V) with the second pass. It seemed to work better but maybe not...

Is my assumption correct??

zip
 
Well..... when you mixing down, try to avoid EQ unless U ABSOLUTLEY HAVE TO use it......... because it is a lot better to futz around with mic postitions to get the sound U want then it is to just EQ the hell out of it. I don't think that U should have to take so many passes with the eq.

And doing the EQ in seperate passes will make a big differece in the sound for digital. for example if U take -10 db from 1k, then on the next pass put +10 db back into the 1k freq it will not sound the same as it did when it had no eq on it.

Sabith
 
Have to EQ!! I'm remixing a bootleg CD that's FULL of midrage. I'm pulling it out while adding bass / treble frequencies. You are absolutely right about the multiple pass EQ though... With the dual pass approach the overall output was hotter but distorted during the louder musical passages.

When I "added" the frequency levels and did single pass EQ...it was damn close to 0dB and sounded much clear.

I will remember what you said about avoiding EQ when I do my other recordings and am micing my amps.

Thanks

zip
 
You must log in or register to reply here.
Back
Top