Yes, something like this, for example, so that if you’re holding down a long note, it won’t be stolen as you’re playing others but instead they will take an available voice. And if there is no available voice then the algorithm should steal the note that voice that was least recently activated. (This is what I find intuitive and how it works on my poly synths at least)
The other issue is that if more than one voice are allocated to the same button you get strange behavoiur, I’m not sure how to tackle this.
Even with these imperfections, using the grid ops to operate just type like this feels really great!
This makes complete sense, but does it matter? It seems to function fine the other way too
Amazing! I’d missed these (though I was also quite proud of my mathematical work around!)