Final Master Levels

A

adam79

New member
Is it possible to get a recording as loud as pro mixes without some kind of brickwall limiter on the Master Bus? Should tracks naturally get louder when EQ'd properly? I'm trying to figure out if the whole loudness war thing has to do with just putting a limiter on the Master bus or dialing in the limiter to crush.

Do you guys mix into the limiter or put it on at the end? Also, do you insert it as the first, last or some other position in the chain? I'm using Waves L-3, which limiter do you guys like?
 
It's always last when I do my pseudo-mastering. If there's anything after that, it's just metering, which I'd turn off before bouncing out.

There are only a few genres and delivery formats where I'd imagine you want to "crush" your final audio file, and, at least my sense is there's been some push or dialing back in the loudness wars, but I do know some folks are still producing sausages... I'd get some good, current reference tracks and make sure you're targeting something similar for what and where you're delivering.

I don't use EQ to achieve any target level. I'd say most of the time it's mostly level neutral. Individual track compression, the simple physics of tracks being summed, and more compression on busses probably are the main reasons the mix gets progressively louder and lose dynamic range, intentionally or not. (I wish every plugin had a level matching switch - I still get fooled...)
 
keith.rogers said:
I don't use EQ to achieve any target level. I'd say most of the time it's mostly level neutral. Individual track compression, the simple physics of tracks being summed, and more compression on busses probably are the main reasons the mix gets progressively louder and lose dynamic range, intentionally or not. (I wish every plugin had a level matching switch - I still get fooled...)
Click to expand...

Tracks being summed is when you have analog summing hardware, ya?

I've read tons of stuff about gain staging that recommends the output match the input. This makes it impossible for the mix to get progressively louder. Unless the last fx in the chain doesn't get it's ouput level adjusted.
 
adam79 said:
Tracks being summed is when you have analog summing hardware, ya?

I've read tons of stuff about gain staging that recommends the output match the input. This makes it impossible for the mix to get progressively louder. Unless the last fx in the chain doesn't get it's ouput level adjusted.
Click to expand...
Not summing hardware, just the fact that as you add tracks, you add more energy and it will get louder in those frequencies that overlap and don't have phase issues causing canceling.

I don't have external hardware save the mic preamps in or in front of the interface. When I do see things creeping up a little faster than I like, I will go look at inputs/outputs in the plugins (all ITB), especially the compressor(s), and sometimes make minor tweaks there.
 
alterman said:
I don't know if i'm doing it all right and can't explain it in pro terms, but i'll try it in normal short words.

If i do the right things on each and all individual tracks, at the end i have to do less or sometimes hardly anything on the master.
So i always aim to do the magic in the tracks, and not for the limiter which i try to only use for that slight extra.

What i do? Everything that's possible and needed for each track. And sometimes a track is that good on it's own that it hardly needs anything more.
That can even be a limiter in one or only some track. I do what i need or think will get the sound as i wish, no matter if it's "wrong or right". But i must confess that mostly i do the regular stuff.
Click to expand...

The painfully redundant answer is 'it just depends'.

Yes, it is best to get all tracks in order in regards to eq and compression before the master out bus.

I use a basic limiter on output bus while I am tracking. The stock Cubase one as it does not eat up computer resources so much while I record.

If client wants to save cash and master the output level in my studio, then I go with FabFilter ProL at the last in the chain. Sometimes a Waves API 2500 and a FabFilter ProQ, sometimes not. It just depends.

I always recommend an artist serious about their 'album' use a separate mastering guy. Someone not vested in the whole recording process that has a great room/monitoring, and a non biased opinion of the music.

When you spend hours and hours with the same band tracking, one tends to be a part of the band and the ability to separate what you 'hear' is tainted from what the actual result is. You just get so used to hearing the project that you kind of lose objectivity in my experience.
 
In my 'pseudo-mastering' I put the limiter on to barely do any of its compression - only kicking in on the loudest things hitting it. Output to -0.3dB. Then I put a touch of regular compression on it and bring the output to -0.1dB - something that ashcat suggested here. Still doesn't go into the red on Reaper the way some 'pro tracks do when I insert them, but the volume difference is not noticeable when I play tracks in the car.
 
There is something which works quite different with some limiters. I love the FF Pro-L because it can reduce transients as much as 6db without actually hearing the usual 'pumping'. Many options as to how you use it so it way more complicated than just a 'brick wall' limiter. It eats resources because it 'looks ahead' and grabs them out before you hear them. Not sure that made sense...I haven't quite understood how it works because I don't care. It just sounds really good to me and can push up the perceived loudness without getting shitty.

Hard to explain but all I know is it is great at smoothing out things you wouldn't find possible with other limiters. And no, I am not selling it.... lol
 
It kind of depends on what music you are mastering, if it was house or trance type stuff it would be brick walled to within an inch of its life, if it was a folk Ballard hardly a touch.

I do usually have a limiter as the final item in the chain, but usually to just grab the odd peak maybe a dB or so.

Alan
 
witzendoz said:
It kind of depends on what music you are mastering, if it was house or trance type stuff it would be brick walled to within an inch of its life, if it was a folk Ballard hardly a touch.

I do usually have a limiter as the final item in the chain, but usually to just grab the odd peak maybe a dB or so.

Alan
Click to expand...

For sure! I forget to mention that as I mostly do hard rock genre. Some country, Hip Hop and random solo artist stuff. Acoustic songs and, well everything is different...

So back to the: painfully redundant answer is 'it just depends'.

I remember years ago looking for the magic combination that makes everything sound great. Still chasing the dragon...
 
jimmys69 said:
I remember years ago looking for the magic combination that makes everything sound great. Still chasing the dragon...
Click to expand...

I must change the way I do things almost weekly, funny thing is that if I listened back to it 3 months later I could not tell you what I did on that particular recording. Just shows there are several ways to skin a cat (sorry PETA)

Alan
 
witzendoz said:
I must change the way I do things almost weekly, funny thing is that if I listened back to it 3 months later I could not tell you what I did on that particular recording. Just shows there are several ways to skin a cat (sorry PETA)

Alan
Click to expand...

In the same boat.

And I am not a cat fan so... I would not skin one per say, but they are not my favorite creatures...

Neither are the hip hop guys. But sometimes my family and dog need food. So I skin a cat...

And sorry PETA... That was a funny snowflake link. Wow!
 
That was some funny shit! :laughings:

I had to check out some of his other videos....the "male snowflake" one was also hilarious. :D
 
mjbphotos said:
...something that ashcat suggested here.
Click to expand...
I appreciate the namedrop and I'm glad I could help.

It never made sense to me, though, to try to do anything to peaks before you adjust for average level. I think it's a holdover from analog workflow, but even then it didn't make a whole lot of sense. I put my lookahead RMS leveler first and then the saturation that I use to control peaks.

Actually, my mastering chain is:

ReaEQ to shelf down the low end and maybe a couple other things for "preemphasis" to sort of shape and focus the action of the compressor.

ReaComp as a long slow RMS leveler with lookahead set to the very lowest ratio with the knee cranked way up and threshold set so that it pretty much never gets to out of the knee. It's almost always doing something but never really doing much.

ReaEQ for deemphasis (shelf the low end back up) and any final spectrum shaping I might want.

Saturation to round off the very tops of the very loud things and make sure that whatever happens it will never ever actually be as loud as X. I don't trust limiters with time constants. They can't promise you a ceiling unless they also add saturation or clipping.

In the end it's not much more than pounding it to tape and letting it distort if it wants to, but it works.

BUT I'm always mastering things that I have also mixed. Usually I do end up slapping that chain on the actual mix project toward the end and kind of finishing the mix through it. I do my best to control the dynamics at every stage from performance to track level compression or saturation to group/bus level so that at the Master is just about some glue and some "aberrant" peaks from when everything just happens to push the same direction at the same time. If it's a one off, I'm done. Render it, see where the peaks fall, adjust the master fader to make that -0.6dbFS, render final. If I'm compiling an album I usually bypass the Master chain, render to floating point so that I don't have to care about absolute levels, and bring it into an album master session with about the same chain.

When you're mastering for other people, it's sometimes more about rescue and repair and that's when you usually do have to do a lot more than just "slam it to tape". I'm kind of glad I don't have to do that kind of thing, but I'd be happy to if there was money on the table.
 
There's a ton of ways to get where you want it to go.

If I'm doing a serious project, then all tracks get their individual control tweaked to the point of it not banging a play-back master too hard. Then I break everything up into aux masters ....subs....stems...whatever you want to call it. Alike noises go together although NOT ALWAYS....So a sub for vocals, bass, guitars,keys,drum,other stuff...These are Aux masters in ProTools and everything is bussed accordingly. I will add compression or a limiter to each of thses auxes until they behave and the master 2 buss gets filled like I want it .....On the master fader I now add my limiter. Ahead of that ..possibly..an EQ, then all the auxes and the master fader all get bussed to another aux which is bussed to a print track. This is where I derive my master from. Each step imparts some control and keeps the master 2 buss from collapsing.

Sometimes, a mix will be good but not have any attitude or isn't gelling as well as it could. I'll record to my print track and then run this out through a stereo compressor and then back in to another track. Usually only a 1.5 to 1 ratio and unity on the output is needed. This newly created track I can go back to digital mastering software that I like and get it to sit comfortably at whatever level I choose....be it smashed to crap of just tickling -.02.

Most of the time I send my print tracks to my mastering engineer. He's got way more horsepower and things to choose from than I ever will.
 
You must log in or register to reply here.
Back
Top