The QuBit Nebulae is, to be fair, an actual computer inside a rack module. The Raspberry Pi inside is crazy powerful compared to the microcontrollers inside the trilogy modules, and runs CSound and PD because it’s actually running Linux, first of all. And there’s an immediacy that might be lost when you make something that feels like ‘an instrument’ be more like ‘a comptuer’ internally: that module actually takes time to boot up (though I believe they’ve worked hard at minimising it).
The microcontrollers in the trilogy modules and the Teletype aren’t powerful enough to run an OS layer, really: they need to be coded against the metal. So you can’t run a high-level interpreter on them.
It might be possible to write code to generate appropriate C from higher-level code, on the computer side, before flashing it, but that’s not exactly simple either.
This is basically the hard thing about microcontrollers; they’re pretty powerful, especially the high-end Arm stuff (an M4 with floating point does all manner of impressive things in many modules out there right now) but there’s no easy way to simplify the work of getting code on there - unless somebody else does a lot of work first.
I forget if the Nebulae can be a USB host, now you mention it.
