> Should we let [x] = dual osc (I'll admit I hadn't really considered this, but it is damn tempting) I'm sonically capable of approaching "massive" territory... hell, I could even coax some duophonics... maybe. I'd likely be looking for a high resolution converter and would be more concerned with multiple cv than gates. All sequencing duties would remain in the MPC. My guess is that this would tend to lead me to a more "traditional" approach (at least as far as my historical way-of-doing-things is concerned). I'm also wondering at this point if I shouldn't just grab an Animal panel.
If you are planning on going for a "voice-only" panel (i.e. no sequencer), I'd just get an Animal, unless you *really* want /N comp or a TGO. Better density that a dual-osc + creature IMHO, and almost the same price.
> Should we let [x] = seq-a (this is really where I started, thanks in part to kkonkkrete's youtube vid), while I may still do some note for note sequencing from the MPC, I'd probably spend a whole lot more of my time programming gates/triggers to control some seq-a functions. My convertor would now need more gates than CV (and from my preliminary research, it seems it would be a custom or DIY solution).
This really depends on what you want to use your system for. I thought I really needed a sequencer, because I wanted a completely stand-alone 1-panel unit, which could do melodic things and that I could play while sitting on the sofa. The Seq-A can certainly do a lot more than just pitched sequences. On the other hand, the deeper I get into Serge synthesis, the more I realise that I often want more "voice" --- many of my patches use everything except the sequencer sub-unit of the Seq-A M-odule. You can do a huge amount without a sequencer. In some ways, I wish I had got an Animal and a TKB. On the other hand, the creature on its own is SO amazing that pretty much any pairing and you'll end up with a superb little system.