What I would've expected is that every attribute generates a different sound per head, currently it doesn't matter which head you put something on, it will always sound the same, far less possibilities.
