guys guitar doesn't count it's like the doctor who immortal time traveler guy it just shows up everywhere and you're like yup this is normal

ok so the broader thought experiment here is i'm really enjoying learning the mandolin, and i'm really enjoying building and exploring mollytime, and i'm feeling a gap where i'm not really good enough to play music with people but it is nevertheless an unmet need but i could maybe use a sequencer to practice with but mollytime mostly sounds like snes sounds which makes it a somewhat odd pairing so like how do i make that not sound ironic or like i didn't buy enough daw plugins to sound good lol

first idea: mic the mandolin and obliterate the sound with filters so it matches the chiptunes better. I think this is a non-starter though because it's a pretty instrument and it feels wrong to remove that from the sound.

second idea: keep the sound of the instrument the same, but also maybe explore an additive approach involving delay loops and/or granular synthesis stuff. especially if it can put the different voices can sorta bleed into the same conductive media so to speak. maybe worth exploring

third idea: same as the first idea but backwards, leave the mandolin alone, figure out how to mutate the synth sounds into something that feels more acousticy without necessarily trying to emulate any particular traditional instrument. i have no idea how that would work

fourth idea: lie to the audience and say I was just practicing the mandolin next to the hole in reality in the corner of my apartment and hope none of you know that the hole in reality in the corner of my apartment doesn't sound anything like that

fifth idea: use a fun narrative to make it work. maybe do an album where it starts off with a song or two of, say, bluegrass music, but then the mandolinist gets abducted by aliens. main problem here is scope creep

sixth idea: add some basic sampling capabilities to mollytime and build instruments out of field recordings instead of just analog synthesis

@aeva
You should do this anyway. Sampling is never not useful.

@MissAemilia it's a planned feature, i've just been putting it off features with significant ui elements for tech debt reasons

@aeva
Fair and reasonable. UI is a bitch and a nightmare and thats why I avoid writing them in general.

@MissAemilia oh no it's not that; *mine* is specifically a nightmare because it's largely still the same prototype code i wrote last summer in a fugue of transcendent lucidity that allowed me to bang out the basic ui in a few weeks. that also means if i want to change how something renders on all of the major screens i have to modify all of the places where i copy and pasted that drawing routine because i was following the DDRY principle of antiabstraction (don't DRY (don't repeat yourself))

@MissAemilia I genuinely love writing UI code it's really fun, I just wasn't sure what my needs were going to be for the project up front, so I prioritized building things quickly so I could figure out what they were going to be

@aeva You just use plain SDL drawing for UI, right? I seem to recall that last time I looked at the repo.

@MissAemilia yeah, i'm doing everything with their basic 2D renderer and the sdl font library. everything that isn't text is drawn with single color polygons, even the sweet gradient illusion background. i want to transition that to using ES2 directly instead so i can do shaders and stuff easier

@MissAemilia describing everything parametrically was a good idea because it makes resolution independence easy and it also makes it fast to iterate on things. i want to take it a step further so that it can also be the basis for a theming system

@MissAemilia oh the other reason why i want to rewrite the UI is so I can yeet python entirely, which will greatly simplify the build system and make distribution much simpler