re: starting with a sequencer vs. other approaches - I think it comes down to what sort of music/work you want to make. If you want to make music with a strong sequenced melody element, then starting with a sequencer (and something that makes sound, of course) does make sense to me, and using a modular to do it still allows as much (or as little) chance/unpredictability to be added in, and in a hands on/tangible way. I don’t have a grid, but if I did, I’d definitely start with Ansible. The various CV/gate outputs could act as modulation sources to use in many other ways besides melodies, too.
On the topic of 16-step limitations, this is a cool example of how one can mix up a simple melody with some basic modular elements (in this case, a CV sequence, track (sample) & hold, something to trigger the S&H, and two VCOs. It’s all based on a linear 16-step sequence, but is a lot more interesting. You could do the same thing with any two VCOs, any sequencer, something that spits out gates at different timings than the main sequence (both of those could come from Ansible), and any track & hold or sample & hold (would give a slightly different result, but same idea) module.
Hopefully this isn’t to situation-specific, just meant as an idea of how a simple sequence can be changed dramatically with some simple additional elements. Add in some way to transpose the sequence, such as using a precision adder to combine two or more sequences (both from Kria, say), and you could have a crazy amount of variation still within a 16-step maximum, and potentially still with a compact setup.
Do you already have any sort of drum machine with some sort of analog sync? Since your examples include some drum elements, I’ve had a lot of fun with a Volca Beats synced to my modular (via Pamela’s Workout in my case, either Pam’s or Volca can be master).
