What order do I normalize, compresss, EQ etc?

T

trackstarz

Guest
In what should should I begin normalizing, compressing, EQ, and adding effects to my vocals?

I'm using Adobe Audition for an multitrack/editor and Reason 2.5 for beats.
 
eq first to bring up the parts you need then apply compression to smooth everything out then if you have to normalize but do realize that normalizing brings up noise so you may have to eq again to cut out ground rumbe , mic noise , hum , hiss or what not i recomend something like the waves l3 multimaximizer over normalizing a track for a more polished effect though.
 
The typical signal chain on a mixer is is input (pre-amp), insert, then eq.

In this case, normalize, compress, then eq.

Your levels need not be too high to begin with, the normalizing stage is up to you to device on what you need.

Typically you compress before eq'ing. Ideally, you'd use a compressor that you are familiar with. So you'd know what sound the compressor will give you. Then you'd use the eq to further shape the sound of the compressor.

If you eq and then compress, the compressor will probably cloud your eq'ing work with its own character. Then you'd have to use another eq after the compressor again. There may be cases when you may want this if you have eq's that have different characters but usually 1 eq will do the trick.
 
Pretty much the only rule is to normalize last. You can EQ befire or after Compresion. there is no right or wrong, they sound different. Let your ears be the judge. And what works on one song might not work on another.

Normalizing should be the final step though. You normalize (or use some sort of digital limiter like an L2) to make sure your tracks are as loud as possible. Normalizing brings your loudest peak in your audio right up to digital zero. If you EQ after this and end up boosting frequency's there is a really good chance you will clip. If you cut frequencys or compress you will be lowering your signal, and therefore it won't be as loud as it could.
 
JaretLeash said:
Pretty much the only rule is to normalize last. You can EQ befire or after Compresion. there is no right or wrong, they sound different. Let your ears be the judge. And what works on one song might not work on another.

Normalizing should be the final step though. You normalize (or use some sort of digital limiter like an L2) to make sure your tracks are as loud as possible. Normalizing brings your loudest peak in your audio right up to digital zero. If you EQ after this and end up boosting frequency's there is a really good chance you will clip. If you cut frequencys or compress you will be lowering your signal, and therefore it won't be as loud as it could.
Click to expand...


Just a couple of things, here...

If you are going to normalize your audio, you should absolutely do it BEFORE you compress... otherwise it will make your compression settings irrelevant... your compresser will react differently to your normalized audio... a compressor reacts to the levels in your audio which will have changed after normalization.

And for the EQ/normalization scenario... whether you EQ before or after normalization, the same results will occur... UNLESS you are EQ'ing, not on an insert plugin (I am assuming a digital audio environment) but rather applying those EQ changes permanently to your audio through bouncing the track with the EQ or rendering the EQ with the audio or using an audiosuite EQ or however you may do it, there will no difference.

If you are in fact re-rendering the audio with the EQ applied, it will have a bearing on the normalization... but to use EQ that way is rather unusual.


and the other thing is that normalization is different from limiting
 
normalizing is (in my opinion) what you do when you have not set the levels correctly in the recording stage of the project.
get a good hot line in and make sure that your not cliping (going over 0dB). Cause when you normalize, your turning the volume up of the whole wave. If you turn the wave volume up you turn the floor noise up too. hiss hiss!!!
 
I have to disagree on the "when" to normalize thing...

(*) Keep in mind that I NEVER normalize anything. I have yet to find an occasion that it's worth it - Except for dance academy complilations. And even then I still avoid it 90% of the time. But that's for another thread...

If the level before compression isn't that hot, lower the threshold. That's what it's there for. Requantizing by normalizing is going to be just another additional step that will either make NO difference (IF you're lucky) or much more likely, add even more quantization noise.

Having that gain change occur during the make-up gain stage during compression takes steps away from the process. Bringing a signal UP in volume (normalizing) just to bring it DOWN in volume (compression) just to bring it back UP in volume (make-up gain) isn't a quiet or efficient way to go about things...

Plus, let's keep in mind that if you're doing this (compression pass) digitally, if you normalize before compression, you've just used up ALL of your headroom - Before the signal even gets to the compressor.

As far as EQ'ing before or after normalizing, it's again the difference of starting WITH headroom or without.

In my book, no 24-bit digital signal EVER goes above -6dBfs or so at any time until it's very final rendering (during the mastering phase). Headroom is good room. Headroom on quality analog gear is "loose" to a point. Headroom in digital is a hard and unforgiving ceiling.
 
Last edited by a moderator:
MASSIVE Mastering said:
I have to disagree on the "when" to normalize thing...

(*) Keep in mind that I NEVER normalize anything. I have yet to find an occasion that it's worth it - Except for dance academy complilations. And even then I still avoid it 90% of the time. But that's for another thread...

If the level before compression isn't that hot, lower the threshold. That's what it's there for. Requantizing by normalizing is going to be just another additional step that will either make NO difference (IF you're lucky) or much more likely, add even more quantization noise.

Having that gain change occur during the make-up gain stage during compression takes steps away from the process. Bringing a signal UP in volume (normalizing) just to bring it DOWN in volume (compression) just to bring it back UP in volume (make-up gain) isn't a quiet or efficient way to go about things...

Plus, let's keep in mind that if you're doing this (compression pass) digitally, if you normalize before compression, you've just used up ALL of your headroom - Before the signal even gets to the compressor.
Click to expand...


I do agree with most of what you are saying...

I myself never normalize anything under normal circumstances...

...and I agree that you should simply lower the threshhold on the compressor rather than normalizing to bring the level up only to bring it back down.


(In my post, I simply answered a question without giving my personal opinion on the value of normalizing... just to make it clear that I am not saying you should be normalizing all your audio)



...oh, and having just re-read your post... were you saying "when" to normalize as in "In what situation should you normalize" or as in "should you normalize before or after the compressor"?

My post was in regard to "should you normalize before or after the compressor"

assuming that is the question (which is what I thought it was)... and assuming this person wants to normalize for whatever reason whether it provides any value or not, and he wants to compress, and wants to know in what order to do them...

... I still say you would need to normalize before the compression otherwise you will just need to change your settings anyway.

UNLESS you compress your audio then commit that compression to tape (note: when I say "to tape" I mean any recording medium you may be using... just needed to add that in since some people seem not to understand that :) ) before you normalize, so you are normalizing the compressed signal... but I don't really know why somebody would do that.
 
My point is to compress first and make-up the gain. Normalizing to go up just to go down is adding uneccessary steps and quantization noise - AND using up all the available headroom before the compressor.

Lowering the threshold keeps the headroom and adds no extra steps.
 
MASSIVE Mastering said:
My point is to compress first and make-up the gain. Normalizing to go up just to go down is adding uneccessary steps and quantization noise - AND using up all the available headroom before the compressor.

Lowering the threshold keeps the headroom and adds no extra steps.
Click to expand...


I understand that and agree with what you are saying in the sense that it is not necessarily advisable to normalize your audio only to bring it back down again... but you cannot normalize after a compressor... normalizing is something applied to the audio itself, so it is, by its nature, always before a compressor on an insert with regard to the signal flow (unless you commit your compressor to tape first).

since, no matter when you normalize, it will be before the compressor with regard to signal flow-- if you are going to normalize for whatever reason (the question is not whether to normalize or not... the question is: if you are going to normalize, AT WHAT POINT IN TIME SHOULD YOU DO IT? before you compress or after you compress?), you should normalize before you compress simply because if you normalize after you have gotten your compressor settings, you will just have to re-set the compressor anyway because the level of the audio has changed.

...and as for when or where in the signal flow should you place normalizing and compression... well, there is not really any choice due to the nature of "normalizing" and how it is applied to audio (unless you first commit your compressor to tape first... which I would not personally advise doing)

Normalizing will always be first in the signal flow.


Am I understanding you correctly to say that your answer in your previous post is saying: You should not normalize at all because there is no point in raising the level just to lower it again and no reason to lose that headroom.


Maybe I am not reading you correctly... it sounds to me like you are saying something analogous to:

QUESTION: "Where can I get some candy"

ANSWER: "Don't eat candy because it rots your teeth."

--and even though it may be bad for your teeth and you should not necessarily eat it, that is not the point... the point is that this guy wants candy and wants to know where can he get some regardless of whether it is good for you or not.
 
But why dont you just re-record the audio at a good level and then you have no less job later and no additional floor noise?????
 
MASSIVE Mastering said:
I have to disagree on the "when" to normalize thing...

(*) Keep in mind that I NEVER normalize anything. I have yet to find an occasion that it's worth it - Except for dance academy complilations. And even then I still avoid it 90% of the time. But that's for another thread...

If the level before compression isn't that hot, lower the threshold. That's what it's there for. Requantizing by normalizing is going to be just another additional step that will either make NO difference (IF you're lucky) or much more likely, add even more quantization noise.

Having that gain change occur during the make-up gain stage during compression takes steps away from the process. Bringing a signal UP in volume (normalizing) just to bring it DOWN in volume (compression) just to bring it back UP in volume (make-up gain) isn't a quiet or efficient way to go about things...

Plus, let's keep in mind that if you're doing this (compression pass) digitally, if you normalize before compression, you've just used up ALL of your headroom - Before the signal even gets to the compressor.

As far as EQ'ing before or after normalizing, it's again the difference of starting WITH headroom or without.

In my book, no 24-bit digital signal EVER goes above -6dBfs or so at any time until it's very final rendering (during the mastering phase). Headroom is good room. Headroom on quality analog gear is "loose" to a point. Headroom in digital is a hard and unforgiving ceiling.
Click to expand...
im impressed. a giant among men!
 
Are you saying that I'm fat?!?

Not that you'd be far off... :D
 
Wow - I never even answered the original question...

Rule of thumb - Corrective EQ before compression - You don't want an overtone to trigger the compressor if that overtone isn't even going to be there later - Plus, try EQ'ing that overtone (or hiss, or rumble, or anything else that's easy to correct in a steady-state) after compression - It's nearly impossible.

Shaping EQ after compression.
 
MASSIVE Mastering said:
Wow - I never even answered the original question...

Rule of thumb - Corrective EQ before compression - You don't want an overtone to trigger the compressor if that overtone isn't even going to be there later - Plus, try EQ'ing that overtone (or hiss, or rumble, or anything else that's easy to correct in a steady-state) after compression - It's nearly impossible.

Shaping EQ after compression.
Click to expand...
I definitely agree with that, it just makes complete logical sense. And for those who perhaps don`t get the terminology...

Corrective EQ = Cuts

Shaping EQ = Boosts
 
Not completely necessarily - Corrective EQ can involve boosts, and shaping EQ can certainly (in my case almost always) involve cuts.

But the addition / subtraction analogy sticks pretty well - When you're correcting, you're doing something to subtract from problems and when you're shaping, you're adding character. :D
 
You must log in or register to reply here.
Back
Top