There will be more on this soon - I sent a first proper prototype off to be fabbed last week. Principle is that it has a bunch of inputs (6) with different gains and pan positions, and one overall headphone gain knob. So you set up a mix by re-patching.
You can't tweak levels individually, so it forces you to mix elsewhere - with gain or pitch or filtering - changing the sound itself so it works at a level that might be louder or quieter than you might want it.
Just like if you're the drummer or piano player or trumpeter with a real band in a room, you have to learn to play at a level that works, not just as loud or quiet as you want. That's the theory, we'll see if it works.