Topic: normalizing to 0db vs -0.1db

[rhinowing]

I've always normalized my shows to 0db, but I notice a lot of people normalize their shows to -0.1db or -0.3db. Just wondering what the reasoning is behind this? Can my recording actually clip on the peaks that hit 0db?

[yug du nord]

I've always normalized my shows to 0db, but I notice a lot of people normalize their shows to -0.1db or -0.3db. Just wondering what the reasoning is behind this? Can my recording actually clip on the peaks that hit 0db?

As long as the original recording isn't clipped, I think you're fine.

[DSatz]

The analog output circuitry of some playback equipment--perhaps more than you would expect, and including some expensive, highly-touted equipment--can overload briefly even when signal peaks are somewhat below 0 dBFS. By leaving just a small amount of headroom (a dB or two), you can reduce both the likelihood and the severity of any such distortion without sacrificing a quiet recording.

Part of this problem is due to the way digital anti-aliasing filters are often implemented (they don't necessarily have unity gain), and part of it is a very interesting phenomenon which can cause the signal within a sampling interval to exceed the peak amplitude of a 0 dBFS sinusoid even in a unity gain arrangement.

When you have a 90+ dB dynamic range to work with, there's simply no need to push that last dB or two.

--best regards

[yug du nord]

^^^Corrected by our local guru...  disregard my earlier post...  DSatz is Dman!!!

[guysonic]

Great advice.  For the reasons stated I've been normalizing to -0.1 to -0.5 db for the past few years.

[rhinowing]

thanks very much DSatz

[JasonSobel]

and part of it is a very interesting phenomenon which can cause the signal within a sampling interval to exceed the peak amplitude of a 0 dBFS sinusoid even in a unity gain arrangement.

For anyone who's interested, there's a good discussion of this phenomenon in the AD2K manual (AKA the Benchmark AD2402-96).  Starting on page 24:
http://www.benchmarkmedia.com/manuals/ad2402-96-manual.pdf

[trajhip2000]

and if you're normalizing before you do further signal processing - like resampling or dithering from 24 to 16 bits - then you also should be leaving some room, since both of those processes can raise your levels 0.1 to 0.2 dB. I usually leave a margin of -0.8 dB (somewhat arbitrary, it could just as easily be -1 dB).

Steve

[cfox]

I have never normalized a recording...

[sygdwm]

I have never normalized a recording...

ditto. i do, however, raise gain sometimes.

[dactylus]

I have never normalized a recording...

ditto. i do, however, raise gain sometimes.

I think that they're talking about raisng the gain and not raising the level of all the inherent frequencies in the recording to the same level, thus destroying the dynamics. 

There is mucho confusion out there on this issue and for that reason I have NEVER normalized a recording either...

 

[boojum]

Samplitude defines their normalize function as,


"You can use this function to change the amplitude of the sample data. It will be amplified in such a way that the highest amplitude that appears in the range is set to 100% (or another value between 1% and 400%) of the value range. Here the maximum is first communicated and calculated with the selected percentage value. Then all values are weighted at the new factor.

Use this function to fully modulate samples or overmodulate targeted audio segments. Please note that the noise level also rises when normalizing.

This function is especially useful when converting from high to lower bit rates. This guarantees that the otherwise low dynamic range is used to its full extent at lower rates."

[Teen Wolf Blitzer]

Ya I have met a few folks who use that RMS normalize thing that jacks up the whole waveform to 0.  Looks and sounds like crud.  I normalize to -0.02 dB. Meaning the highest point of my recording wherever it is in the show is
-0.02 dB.  I do each pair... meaning mics and board in most cases..and then I mix. 

But a question for you all that neither normalize or apply gain in post.  Aren't your recordings quiet?  I hate having to crank up the volume.  I run my 744 so that the levels are around the second red on the meters.  Whenever I get home however the levels need gain or normalization.  They are low in the spectrum.  That's why I do it.

I have never normalized a recording...

ditto. i do, however, raise gain sometimes.

I think that they're talking about raisng the gain and not raising the level of all the inherent frequencies in the recording to the same level, thus destroying the dynamics. 

There is mucho confusion out there on this issue and for that reason I have NEVER normalized a recording either...

 

[boyacrobat]

amp and volume for me.
dynamics is everything.

if i have to, its by hand, manually , track by track.
i oversee the whole process visually each track, matching crowd noise only.

still i find it a waste of time, amp and volume all thats required.

less is more.


g

[taylordb]

This has been a very interesting read.  I usually amplify (not normalize) in post, but always went to the 0.0 dB benchmark.  From now on it will be to -.02 dB.

[newplanet7]

Normalize peak not rms.

[easy jim]

Ya I have met a few folks who use that RMS normalize thing that jacks up the whole waveform to 0.  Looks and sounds like crud.  I normalize to -0.02 dB. Meaning the highest point of my recording wherever it is in the show is
-0.02 dB.  I do each pair... meaning mics and board in most cases..and then I mix. 

But a question for you all that neither normalize or apply gain in post.  Aren't your recordings quiet?  I hate having to crank up the volume.  I run my 744 so that the levels are around the second red on the meters.  Whenever I get home however the levels need gain or normalization.  They are low in the spectrum.  That's why I do it.

I have never normalized a recording...

ditto. i do, however, raise gain sometimes.

I think that they're talking about raisng the gain and not raising the level of all the inherent frequencies in the recording to the same level, thus destroying the dynamics. 

There is mucho confusion out there on this issue and for that reason I have NEVER normalized a recording either...

 

I (peak) normalize in post by raising the gain to -0.2 dBFS, generally on the master bus/fader unless it is a single stereo to which I am doing no additional processing (i.e., EQ/HPF, etc.).  When I'm mixing from more than one stereo feed in post, I (peak) normalize the final mix by raising the gain to -0.2 dBFS as the final step with a 'trim' plug-in at the end of the chain on the master bus/fader track. 

When it is a 'post-matrix' mix, I like to use some light dynamic compression and EQ to sculpt and carve a bit on the raw resultant mix which is often really raw to my ears; extra dynamic by comparison to the mic source from the summing of all the tracks, and often less frequency balanced as well due to the summing of particular frequencies that might be emphasized in one or both sources due to the room or the engineer's mix.  For post-matrix recordings, I almost always like the mix better - and find it more car stereo/ipod-friendly - when it has undergone some light dynamic compression and a high pass filter to cut out excessive low end typical of live shows. 

That being said, however, I view my goals in creating a post-matrix a lot differently than optimizing the headroom for playback of an ambient stereo recording.

[su6oxone]

This has been a very interesting read.  I usually amplify (not normalize) in post, but always went to the 0.0 dB benchmark.  From now on it will be to -.02 dB.

Sounds like people are normalizing to a peak of anywhere from -0.02dB to -2dB... wish there was a consensus on a safe level to normalize to.