What exactly does mastering to -12LUFS mean, referring to integrated, short-term and momentary LUFS?

[Edwin]

Hi Everyone,

The title would be a basic question if I was just writing music for online streaming, but I can't find important details about this in relation to writing library music:

-When you are asked to master something to -12LUFS, does that refer to integrated, short-term, or momentary LUFS, or all three? I understand that if you just ask Ozone to master to -12LUFS it will use integrated, but that sometimes leaves you with a big dynamic range and that leads to further questions:

-If "mastering to -12LUFS" includes short-term/ momentary LUFS: let's say I have a solo classical piano piece spanning a large dynamic range that for one section sits around -22LUFS short-term, and at its loudest section reaches about -10LUFS short-term. Should I automate volume so that both sections are sitting around -12LUFS? Or is it best to leave some dynamic range in place, like the quiet section is -14LUFS while the loud section is -10LUFS? Or do people not bother with volume automation and instead they do this by slamming against a brick wall limiter in the -8LUFS section while bringing up the -20LUFS section?

-When pushing percussive sounds (e.g. piano) hard against a limiter or compressor, the sound becomes unnatural. You can hear the compression working, and the result is an unnatural ADSR for the piano notes where the attack is reduced, the decay kicks in too late and the sustain is longer. If you push the limiter really hard, there will be distortion. If my classical piano piece has strong transients, for example a quiet accompaniment with a prominent melody in octaves, do publishers/ libraries/ editors care that the momentary LUFS is going to be bouncing between -10LUFS and -20LUFS constantly? Or would they normally prefer some or a lot of this audible compressing/ limiting?

Thanks for any help!

[AlanHall]

Good question! I for one will be standing by to watch the discussion.

My only contact with this is from using a final limiter that read out LUFS in momentary and short term (NOT integrated). I'd master so that the short term peaked at or a little over -14 LUFS, and get screener feedback that my levels were too low. So I added ADPTR Metric AB to my monitor bus and found out that short term can go way over -14 before the integrated measurement arrives at -14. Yes, I integrated over the whole track, the integrated level stabilizes after some few dozen seconds of playtime. Louder, and for the music I was mastering (droney and sustained music) did get caught by the limiter in spots. Previously I tried to prevent the peaks from getting clipped.

[cosmicdolphin]

The title would be a basic question if I was just writing music for online streaming, but I can't find important details about this in relation to writing library music:

The same guidlines apply for library music as they would do for streaming.

-When you are asked to master something to -12LUFS, does that refer to integrated, short-term, or momentary LUFS, or all three? I understand that if you just ask Ozone to master to -12LUFS it will use integrated, but that sometimes leaves you with a big dynamic range and that leads to further questions:

Integrated

let's say I have a solo classical piano piece spanning a large dynamic range that for one section sits around -22LUFS short-term, and at its loudest section reaches about -10LUFS short-term. Should I automate volume so that both sections are sitting around -12LUFS? Or is it best to leave some dynamic range in place, like the quiet section is -14LUFS while the loud section is -10LUFS? Or do people not bother with volume automation and instead they do this by slamming against a brick wall limiter in the -8LUFS section while bringing up the -20LUFS section?

Leave some dynamic range but not too much. It's contextual, you'd expect mre in a classical piano track than a pop song

-When pushing percussive sounds (e.g. piano) hard against a limiter or compressor, the sound becomes unnatural. You can hear the compression working, and the result is an unnatural ADSR for the piano notes where the attack is reduced, the decay kicks in too late and the sustain is longer. If you push the limiter really hard, there will be distortion. If my classical piano piece has strong transients, for example a quiet accompaniment with a prominent melody in octaves, do publishers/ libraries/ editors care that the momentary LUFS is going to be bouncing between -10LUFS and -20LUFS constantly? Or would they normally prefer some or a lot of this audible compressing/ limiting?

Whatever sounds best is what they will want.

[Edwin]

Thank you for the answers

So basically they really do mean -12LUFS integrated, however there should be some dynamic range within that depending on the genre. The amount of dynamic range is contextual, and using compression/ limiting to flatten the dynamic range shouldn't negatively impact the sound of the piece.

So in the case of the hypothetical classical piano piece, I might opt for a short term LUFS range of about 5 LUFS, perhaps ranging from -10 to -15LUFS. I'd imagine that would be best achieved with volume automation across sections, and some fairly transparent multiband compression to control transients, transparent limiting, and perhaps some 'exciters' or saturation.

[AlanHall]

FYI: In yesterday's Taxi TV, Library owner Steve Barden mentioned that he expects a dynamic range of 4 to 8 dB for typical cues. He explained that too great a DR would allow the soft passages to be completely masked by the dialog.