sorry, like brian i’ve been occupied with lots of non-norns related things in the past weeks.
but in the background we’ve made some infrastructure changes to the norns SC system allowing (among other things) voices in this engine to be processed on separate CPU cores. so we have much more total processing available and i can go ahead and revert some of the corner-cutting changes, meaning better interpolation and more audio-rate params for MLR (better sync, smoother fades.)
adding panning and filtering per-voice in the engine is definitely something i want as well. and it is quite simple. how to control such addditional params from MLR interface is another question that i probably won’t worry about.
(actually the engine does support panning already, now that i think about it - just the mlr script doesn’t provide any controls for it (yet)).
this thing with the channel assignments is pretty baffling. it’s not intentional, and i’m 99% sure that it’s not something the engine itself is doing. seems as though MLR script is somehow setting up the engine to sample from the master outputs (after monitor bussing? really i don’t get it) instead of directly from the inputs. will investigate.
anyways, various engine-level improvements are coming soon.