well, I can't actually pinpoint where the "the rest of the owl" transition happened. It's arguably quite early, after the general direction of coloring the center and having helper salt blocks coloring the edge and finding a decent berlo placement so that salt blocks can dupe from berlo in two directions without suppressing for too long. I do feel lucky that the final swings didn't get too problematic - many arm placements felt dangerously close to molecule swings but just worked.