Normalizing vs Compression or Limiting

[O.G. Killa]

Originally Posted by barismanco123

Maybe I have not understood here but you said "If you look at normalizing as the process of making your loudest peak, your loudest possible peak".

The loudest peak will always be the loudest peak and for that reason normailizing on that basis isnt really necersary, no?Or maybe I am not understanding?

I probably should also mention that you do have a point... it is really useless to normalize your files to 0dB peak. But at one time, people thought it was better to "use all your bits" in a digital recording, so they would try to make their signal as loud as possible without altering the dynamic range. But, in this day and age we know better, or do we? LOL

But!!! normalizing to a specific median level can be sometimes useful, as I've mentioned in some of the examples I've posted.

And I should also point out... Normalizing is not a very commonly used tool for every day music production. In the last 10 years I've probably used it maybe 6 times (when I was working on making a sample library) and that's it! Oh and I've used it a couple times on Voice Overs for radio spots!

[O.G. Killa]

I guess I should add one last comment that might help...

There is almost not difference between the gain plugin and the normalize plugin... they both do the EXACT same thing... BUT!!! With the gain plugin it will turn the whole file up or down by the amount you specify (+3dB or -6dB, etc). The Normalize plugin will turn the whole file up or down to get it's peak (or RMS) to match the level you set in the plugin window.

The big difference is realized when you have a bunch of files in your protools session and you move from one to the next. With the gain plugin, if you put -2dB in the GAIN window, each file you hit apply to will be turned down by 2dB.

But what if you don't want to bring EVERY file down by 2dB. What if you want every file to have a peak level of -2dB? Using the Gain plugin you would have to analyze each file, then compute how far away it's peak is from -2dB, then you'd have to apply that much gain/attenuation to make the file have a peak of -2dB...then you'd have to do this to the next file, and the next file, and so on...

But with Normalize, if you put -2dB in the Normalize window, you can process every file. It kind of automates the process for you. Some of the files might get turned down and some might get turned up, but afterward each file will have a peak level of -2dB. You could almost call it an "intuitive", or smart version of the Gain plugin.

Does that help explain kind of what it is?

[DrFord]

actually I was going to rename my production company "Love Ewe Productions." It's jsut soooo inspiring. Plus I'm a baaaaaaaaaaaaaaa d boy!

[barismanco123]

All of this is a big help - espeacially the use of attenuating a file/signal to match RMS levels in mastering.

I understand now why it´is a relatively unused process in many peoples workflow.

It also seems like something that when used should be used as a fine tune adjuster and not as an RMS at 0.0 DB type tool.

I feel its like a gain rider but in a more automated/rapid respect.

If I ever master I do use the gain plug sometimes to increase volume without using FX.

In other words I understand it better and for that thankyou gentlemen.

Baris

[O.G. Killa]

Originally Posted by barismanco123

All of this is a big help - espeacially the use of attenuating a file/signal to match RMS levels in mastering.

I understand now why it´is a relatively unused process in many peoples workflow.

It also seems like something that when used should be used as a fine tune adjuster and not as an RMS at 0.0 DB type tool.

I feel its like a gain rider but in a more automated/rapid respect.

If I ever master I do use the gain plug sometimes to increase volume without using FX.

In other words I understand it better and for that thankyou gentlemen.

Baris

Exactly!!! I think you got it. And also, for other people reading, if you set your Normalize plugin to 0dB RMS it would be completely distorted...why? Because Normalize doesn't compressor or limit at all... it would just turn the volume up. And if your RMS level was, say -14dB but your peak was -1dB... setting it to 0dB RMS means you are adding 14dB of gain to the whole file. Doh! Lots of crackling, clipping, ugliness.

[O.G. Killa]

Originally Posted by barismanco123

espeacially the use of attenuating a file/signal to match RMS levels in mastering.

Load some of your favorite albums from CD into your daw and analyze their peak and RMS levels. You'll probably notice that all their RMS levels are very close and they are all seriously brick-wall limited, but their peak level might differ as much as 2 or 3dB from song to song. The mastering engineer usually isn't changing the "peak" output setting on his limiter, he's leveling it set to -0.2 or whatever for each song... so that change in peak is him/her hormalizing the RMS levels after they have "mastered" the audio.

[DrFord]

You know I have to admit that I have really never used the RMS normalizing before in the mastering process. I mean, I'm no mastering engineer. In fact, I try to use multiband compression, but I can't really tell if I made it better or worse. But I can make one hell of a song, that's why I'm a producer and I let the professionals do the mastering.

In fact, I think that is one of the most important things a "producer" can understand about their role, that they aren't always the best person for the job, and delegation is the sign of a good captain. A good mastering engineer is usually an older, trained, specialized engineer with special equipment sometimes self designed, in a highly tuned room. If I could get Lord Algea or Maserati to mix my records I would be a happy camper.

All that being said I am going to start RMS peak checking my mixes against my target mixes to see if I have the same energy. Honestly, some of my favorite mixes (the entire Tool AEnima album for example) aren't neccesarily the loudest albums, but they sound amazing.

[O.G. Killa]

And that is usually the case. The CDs that aren't the loudest usually sound best because their dynamic range (and subsequently, their fidelity) is more intact.

I saw in interview with Andy Johns once... and in the middle of the interview he said something in passing that really caught my attention. He was talking about how the only thing in a "rock" mix that gives you any sense of fidelity are the drums, most of the time. Because the drums have very pronounced transients and are much more dynamic than say, a guitarist playing a G power chord for 2 bars, the drums end up dictating whether the recording sounds good or not.

Now obviously this isn't the case all the time. But it brings up a good point... What makes us feel like something is of higher fidelity than something else? I've never done a formal study on it, but I would say from my own experience the dynamics of a recording do. While yes, comparing two signals, the louder one will always inherently sound better... but what happens when you make them the same average level... which one will sound better then? This really started to come together for me when I started recording and mixing a lot of movie trailer music. That stuff is Soooo dynamic. And to make it sound really good you HAVE to make it very dynamic.

I also remember someone telling me once that Jimmy Page's idea for Led Zepplin was a rock band that was very dynamic. They would get really loud, and then really soft, and then really loud, and then really soft... I'm not sure if he ever really had that intention, but when you listen to a lot of their music that concept holds true. In a lot of their songs the alternate from loud to soft. Think of Black Dog, I'm Gonna Leave You, Stairway to heaven, and on and on and on... So for me, it's no coincidence that A LOT of people also consider those albums great sounding and amazingly engineered albums... My guess is people feel that way because the songs are so dynamic. It's just food for thought I guess...

I think as the loudness wars hits it's climax and things cannot get any louder, we might see a shift to "dynamic" music.

Also... as a person learns how to mix, it is actually good practice to make a mental note of RMS levels (like Dr ford mentioned). If your mixed and mastered version ends up having the same RMS level as your favorite track, you can then REALLY pull your mix apart when comparing and really see what you did wrong. Maybe you have your drums a little lower than your guitars compared to your favorite sounding song... or maybe your intro is too quiet compared to the rest of the song and you need to put some more automation into the mix to balance the intro with the rest of the song... and so on and so forth...

Over time you will get acclimated to the way your listening environment sounds and you will have "taught" yourself how to mix in that listening environment. That's also why a lot of the "famous" mixers only mix out of one studio. They've become acclimated to that room. If they mixed in another room it might take them longer to mix a song since they aren't familiar with the sound of the room.