Just to mention (again) that absolute levels do not actually matter until right when you are rendering the real final distribution file. Up to that point what you should really be looking at is the Dynamic Range of the signal. This is the difference between the highest peak - the single loudest sample coming out - and the integrated level whether you're talking LUFS or RMS.
When you see things like loudness recommendations from streaming services or whatever, it's generally assumed that your file will peak at or near 0dbFS and then the loudness target basically IS your DR. If your file does not peak at 0dbFS, you can't really say how it compares or what you might need to do to meet that target.
Maybe you're integrating to -18 but peaking at -4. That file doesn't need any limiting to meet the -14db target for YouTube. Just turn it up 4db. Maybe you're integrating to -12, but peaking at +2. Again, you're actually hitting the target, but you need to turn it down - just static flat attenuation.
I'm not sure I'm really making the point the way that I want to, but if you're asking something like "My mix integrates to -18, but I want it to be at -14, what should I do?", my answer starts with the question "Where is it peaking?" because until I know that, I have no way of even starting to tell you what you need.
|