Firstly, I'd like to say that it has been great to take part in this challenge. Personally, it has been a great opportunity to return to making stuff after months of no hardware projects. Also, it has been very helpful to get myself with the Bela platform and BeagleBone Black, and this will be incredibly useful for future projects. And if that wasn't enough, I have made huge progress on Kazumi, which already had a past before this challenge and now has a future beyond it. That said, I am aware my two last posts are after the deadline. I would have loved it to make it in time, but as we all know, things go wrong all the time... and I should have planned better. I close now with a few points about What I have learned and Next steps.
What I have learned
Even with an ultra-low latency board, capacitive touch over I2C is always going to be slower than the analog inputs.
A basic understanding of Karplus-Strong string synthesis
SLS printing is the way to go for strong and accurate parts
Fusion 360 is great for my needs (basic solid and surface modeling)
There's lots of stuff I can't think of right now so I'll try and come to edit.
Lots of Pure Data programming
Doing some testing with other musicians
Design a soft case to carry it in
Find a better speaker and adapt the top part to fit it.
Experiment with the multiplexer capelet to use different variable resistors and sensors to control parameters
Send control messages via OSC from a mobile device and eventually design an app to do this.
Try making a wooden base using a CNC machine.
Design a capacitive touch capelet (or a board that goes next to it if it's not compatible with the multiplexer capelet) with four MPR121s that provides 48 electrodes to replace the hand-made one.