23-09-2016, 05:41 PM
(23-09-2016, 02:40 PM)kapu Wrote: Ah, ok. I was thinking the purist perspective, where you set the listening level according to lead vocal, but then the instruments in the beginning might get lost in the car background noise or something.
I’m far from the purist. History of rock’n’roll is fight against rules made by audio purists. But I’m willing to listen reasonable arguments. And I’m listening technical details ONLY when I’m mixing, otherwise I’m listening the song, and most often the song for me is the vox. In my case the analytical listening ruins the experience.
(23-09-2016, 02:40 PM)kapu Wrote: About LUFS metering in music production. I personally have come to think the integrated LUFS measurement should not be paid any attention in the mixing or sound engineering process itself. In production, working by ear on some of the commonly recognized calibration level will 'always' result in a good signal-to-noise ratio yet retaining adequate headroom in 24-bit production, from a technical point of view, if that's what you're thinking. Anyway, at least I can't tell a thing about materials overall dynamics even with very good metering. At best make some vague guesses. Of course, when listening music at fixed speaker level, I can tell that this mix is louder than the last one, and meters show that it's 2 dBs louder, buh that's about it. ^_^
I agree about integrated loudness. To use the integrated loudness in during the mixing would be quite impossible, because one has to calculate it from the whole song. I guess it’s something that is essential only to broadcasters and online playback systems to calculate common loudness to playlist songs.
But I do have a growing reference library (currently about 150 hastily chosen songs) in my DAW and I have calibrated each song to -16LUFS integrated loudness. It’s quite handy when all references (from old 50’s rock’n’roll to modern heavy metal) are directly in somewhat same level. And with reference buss fader I can adjust the references to my mixing level if needed.
Personnally I prefer to use RMS metering with needle, just to see I’m somewhere around -18d-20dbs. RMS must be technically quite near to more modern ”Momentary Loudness” metering. They are just two different perspectives to same issue. And for mixing the RMS is surely accurate enough. And I like the slow meter, it suits my mind that ’s becoming slower day by day.
Although my mixes here are quite loud, it’s because I’m sending the pseudo mastered version. (It’s not bad to have and to develop some mastering skills.) During mixing I have plenty of headroom with quite much of dynamics.