I think a lot of you are overthinking this severely.
You have a source:
- something needs to be recorded, a voice, an instrument, whatever. You record it with the correct gain and have a new a new audiotrack in your daw. > done
- you use a VSTi or external instrument via midi, you set the desired volume via the gear or plugin
- you load a sample or earlier wave, these are either already normalled to 0dB, or you can do that yourself if you wish (or not)
All of the above has nothing to do with mixing, it's just taking care of optimal signal to noise ratios
Now, if you have collected multiple channels in one of the above ways, you'll notice that you will exceed maximum headroom and things sound like *flower*, regardless of headroom you will also find that not changing volume of some instruments, your mix will sound like *flower*, so you need to setup up the volume relationships between the channels.
This means you start with your first channel, leaving some headroom, not peaking at zero already and you build the relative volume relationships with the other channels from there.
If you left enough headroom, you can finish your mix without clipping, if you didn't leave enough headroom, you turn down the main fader (if no individual channels are clipping) or select all faders and turn them down, leaving the main fader on unity.
The rest is just overcomplicating things. The daw doesn't care whether you take the volume down with the pre-gain, the vsti, the instrument/audio clip, your first insert input, or your channel fader, or your group fader, or your main fader IT DOESN'T MATTER.....
As long as you don't clip badly designed VST(i)'s, you're golden. :lol: but you will hear that as it will distort.
Now go make fabulous music!