I could probably live with this limitation, based on how I use my DN. I allocate one voice to the drum track and try to avoid voices suddenly cutting one another off (which sounds like crap). I am willing to impose that limitation on the drum track because I need the other 7 voices elsewhere, and because I’m not trying to make epic drum parts. If my DN had a 9th voice, I might allocate a second voice to the drums.
I suppose if you have a monophonic voice on the kick drum, and the successive kicks are far enough apart, you don’t have to worry about overlap. A little bit more difficult when successive sounds are close together.
But, what about creating a legato melody? On the piano, legato can be achieved by creating some overlap between successive notes. In other words, more than one note may be sounding at a time.
I am curious how users who’ve been consciously or unconsciously using polyphony/voice-sharing on their other devices…are going to react to this limitation of the ST. One obvious change comes to mind, the need to be more disciplined about avoiding sudden cutoffs on the tails of sounds.
Others have pointed out that polyphony can be created by stacking monophonic voices. For me (with my beloved Digitone as my frame of reference) I consider this to be a tedious PITA on the ST. I would have to copy the settings of one voice into another, assuming I want the same sound or timbre from each voice. And then, if I decided to change the sound, I’d have to repeat the process. Then there is the filtering and balancing of the chord-members. All that switching between voices. What a nightmare.
That leaves the chord machine. I am waiting for a demonstration of what is really possible. For example, what is the workflow for arranging a Bach chorale on the ST? Early adopters of gear like to tell you how you can do anything on that gear. But what about something in particular? In other words, how a machine approaches a well-defined outcome is more instructive than playing random-turn-the-knobs on all its features.