Which approach is correct for leaving headroom for mastering

[Franco]

Mix to what sounds good to you and bring down the master fader until the signal no longer hits 0dBFS (get as close to peaking at "0" without going over).

[kingtoad]

Theoretically and conceptually, B is definitely optimal and the best way to work. In practice, doing A is unlikely to cause major problems in a modern, floating point DAW, but some plugins still use fixed point maths so it's best to keep levels nice and low.

[Upright]

Originally Posted by kingtoad

Theoretically and conceptually, B is definitely optimal and the best way to work. In practice, doing A is unlikely to cause major problems in a modern, floating point DAW, but some plugins still use fixed point maths so it's best to keep levels nice and low.

Thank you, this is exactly what I was wondering.

[MASSIVE Master]

If you calibrate your monitoring chain reasonably well, you'll probably find that you just don't clip your mixes anyway.

[Upright]

Originally Posted by MASSIVE Master

If you calibrate your monitoring chain reasonably well, you'll probably find that you just don't clip your mixes anyway.

Some are telling me that you'd be mastering my audio in a 32-bit float environment anyway...so as long as the audio is below zero and I export at 24-bit there should be plenty of room to work.

[MASSIVE Master]

True for the most part. Still, calibrating your monitoring chain (calibrating everything you have for that matter) is always a good thing to do.

It's "proper form" (for lack of a better term) for everything.

[peelproductions]

One thing I was taught a few years back that made a HUGE difference to my mixes, and there ability to be mastered well, was the use of the trim plugin in my DAW of choice.
The first thing I always, always do before mixing is set all faders to unity and mute.
I then un-mute each one individually and adjust the pre fader gain using the trim plugin so each track has a peak level of around -12db.
I find that doing this my mixes have a better perceived headroom and clarity and any EQ and compression used later seems to sit better in the track, and I never have to worry about levels.
The mix buss in the daws seems to work better at this level, sounds mental but it happens.
Try it,
JohnG

[Lagerfeldt]

As already mentioned you'll probably find no difference in sound between the two methods, but it depends on your plug-ins. So the safest method and the best workflow is B.

This PDF explains the subject in details and also deals with some of the myths:

http://www.popmusic.dk/download/pdf/...ital-audio.pdf

[RichArien]

The right way not only for mastering but also for your mix is B (except for the -6db goal, as you said at first and people has told here that doesn't matter that much)
if you mix everything hot everytime you add a new element or effect to the mix you will be likely hitting the roof, and you will have to turn down the master again and again and again, and this will give you more problems than staying at a healthy level so you can add all of those without having to constantly re-calibrate your master level, or at least not doing it all the time.

[SASMastering]

It would be a good plan to consider your recording and mixing gain structure from the very outset. This article should be useful, it provides insight into gain structure and relates it to both the history of console mixing/recording metering and use of a DAW (and it's meters) at 24 bit.

Gain Structure

It is a very cautious article meant for 'knowing the rules' and ideal for beginners, once you have comprehended them fully you are free to bend the rules knowing the potential compromises.

cheers

[brethes]

Method be is the one you should get into then habit of using, as you can still get distortion from clipping with method A.

[wado1942]

Massive said it best. Turn up your monitors so you'll never be tempted to reach 0dB. I learned that a long time ago an have never looked at the digital peak meters since. My mixes got a lot better as well because I'm concentrating on how it sounds rather than what the levels are. If you're setting up your mix, then backing off the levels to something safer, your mixing choices have already been compromised by poor monitoring technique.

[SASMastering]

I believe record/mix gain structure is primary and monitoring level is incidental/preferential relative to the gain structure you have chosen.

[fastlanestoner]

I'm a big fan of fader resolution. I would leave the master fader at unity and balance the mix to proper levels

[JozTheShark]

Originally Posted by fastlanestoner

I'm a big fan of fader resolution. I would leave the master fader at unity and balance the mix to proper levels

I never touch the master fader, it's not like we have no control over what we send to it.

[john john]

If you pull down the master fader, you may not be clipping the stereo output anymore but the front end of your mix bus is still receiving the full wrath of your clipping signal. If your not clipping, then it's not that big of a deal... if you're mastering engineer wants a quieter mix, he'll turn it down

In the 24bit realm, there's no advantage to mixing hot (especially if you're going out of the box for any processing). I think most people who mix hot are either scared of a noise floor (which they shouldn't be) or they don't like to pull the faders down too far.

[Bob Olhsson]

The master fader is an issue with some analog consoles but not others. As far as I understand, it's never an issue in a DAW.

[RichArien]

Originally Posted by john john

If you pull down the master fader, you may not be clipping the stereo output anymore but the front end of your mix bus is still receiving the full wrath of your clipping signal.
.

Not true on a DAW.. but still.. I think we can all agree that mixing with a healthy amount of headroom is advised in any case.

[deft_bonz]

Originally Posted by kingtoad

Theoretically and conceptually, B is definitely optimal and the best way to work. In practice, doing A is unlikely to cause major problems in a modern, floating point DAW, but some plugins still use fixed point maths so it's best to keep levels nice and low.

Originally Posted by MASSIVE Master

If you calibrate your monitoring chain reasonably well, you'll probably find that you just don't clip your mixes anyway.

B

That's how I work at least during mixing. Never tested if it sounds different, but it made mixing easier and better. I work at -12 ITB, then I hardly have to move the master fader, it stays at 0 and levels are almost always around -10 to -6

[john john]

Originally Posted by RichArien

Not true on a DAW.. but still.. I think we can all agree that mixing with a healthy amount of headroom is advised in any case.

Ahhhhh, didn't know that! Even did a little more research to confirm. Interesting, thanks!

[Bouncer]

Nothing pleases me more than a premaster that has no headroom... so I send it back asking for headroom and the client puts a limiter on the master channel to bring it down a few dB.

I especially love it when the client claims not to have used any compression or limiting but the wave form is a solid block of wood.

[huejahfink]

Originally Posted by Bouncer

I especially love it when the client claims not to have used any compression or limiting but the wave form is a solid block of wood.

Great! Straight digital clipping instead. ;-)

Sent from my LT26i using Tapatalk