SOUND ON SOUND

[AMusicPlayerPlus]

Hi,

At which level it should be recorded each track?

I'm doing a mix. It consists of 5 tracks. Each track consists of some VST/Instrument/Voice. At which level it should be recorded each track to get it not too loud and not to loose "juice"? What is best practice? Please advise.

Thank you!

[Mike Stranks]

Each track at somewhere between -12dBFS and -6dBFS. Keep it like that throughout the mix and only normalise the master out to -1dBFS when you have completed the mix and are 'printing' the stereo file.

[AMusicPlayerPlus]

Each track at somewhere between -12dBFS and -6dBFS. Keep it like that throughout the mix and only normalise the master out to -1dBFS when you have completed the mix and are 'printing' the stereo file.

Many Thanks!

[James Perrett]

I would say that the levels that Mike mentions should be peak levels - average levels should be down at around -20dBFS.

generally speaking louder closer to the top of the green edge is better because you can always turn stuff down later to get a better sounding mix.

[Luke W]

generally speaking louder closer to the top of the green edge is better because you can always turn stuff down later to get a better sounding mix.

I've got to disagree to an extent there. With 24 bit recording you can track rather low and still be miles away from the noise floor, making bringing the level up later very simple with no audible disadvantage. Working with less headroom makes clipping much more likely, especially when working with live sources/microphones, and that's a far bigger problem to deal with than adding some gain to a lower level signal.

I know you're not suggesting lighting the meters up to the top at all times, but thought it'd be worth a mention as someone less experienced may simply read it as "the louder the better".

[Arpangel]

Depends what you’re recording, a dynamic acoustic instrument or a backing track on a keyboard, radically different things.
As a rule of thumb, for most things, I aim for an average level around -12 but for our piano and marimba stuff, maybe down to around -20 or lower.
If you have a high track count, maybe 50 tracks or more, you just have to optimise levels at mix-down, I always use 24-bit, as it gives a bit more leeway regarding pushing levels later on.

[Moroccomoose]

I signed up to this and follow it pretty much to the letter...
https://www.soundonsound.com/techniques/gain-staging-your-daw-software

When mixing this is my method:

Set the meters to be green up to -18, then yellow up to -10 and red between -10 and 0.

On individual tracks, keep everything green, perhaps nudging into yellow and do not allow red! Obviously doesn't apply to busses and master, but this will prevent clipping and help keep things manageable.

Keep the faders on unity and set the levels using the preceeding signal chain gain structure...or use the pregain level if your set up has one.

Once this is all set up, then use your faders to mix.

If it seems quiet, turn up your monitors! Once the mix is finished, render it then process it to increase the volume in a separate project.

For me this works really well.

Stu.