yeah, i think software control of buck and boost converters is a mainstream thing to do at this point, and if you think about it, that's what you're doing every time you control a motor with a pwm signal, even if the motor's windings are the only inductor. building discrete dc-dc converters turns out to be the main purpose for the 184-picosecond-resolution high-resolution timer included in chips like the stm32g484. i was reading this appnote yesterday: https://www.st.com/resource/en/application_note/an4539-hrtim...
incidentally, even the old attiny chips that don't have adcs at all do have analog comparators, so you can do tricks like that with them too
i'd be happier with the on-chip op-amps if they also included on-chip analog multiplexers, resistors, and capacitors sufficient to build something like a sallen–key filter in software, but so far only the cypress psoc line seems to be doing that, and it's probably a lot of chip real estate to sacrifice for an inferior version of a couple of 0.3¢ external components
incidentally, even the old attiny chips that don't have adcs at all do have analog comparators, so you can do tricks like that with them too
i'd be happier with the on-chip op-amps if they also included on-chip analog multiplexers, resistors, and capacitors sufficient to build something like a sallen–key filter in software, but so far only the cypress psoc line seems to be doing that, and it's probably a lot of chip real estate to sacrifice for an inferior version of a couple of 0.3¢ external components
i'm not an ee either, i just play one in ngspice