I have a lot of ideas for a '1.2' version of six sines. Not 2.0. Still compatible. But expanded. Here's my rough list
- Reset Controls when automated for clap edge for things like sample rate
- Clamp targets
- LFO -> M etc... depth as an additive and attenuation target
- Pink, White, Tilt (with N for tilt) DONE
- Old BaconPaul Chip LFSR 'semi-tuned' (with N for 'sequence')
- Can we smooth MPE pitch bend with a lagger? (Or do we)
- Unison retuning slider could cubic rescale
- Mod target sliders scale matches display rescaling based on target (could be tricky)
- Now we have DES move MPE to the instance
- if you load a stream which has it from the clap edge set the instance to that
- if you load a patch from an sxsnp ignore the patch setting
- move the mpe enabled param to be instance based (leave the param in to not break stuff but basically dont use it to drive mpe)
- Clap Wrapper Standalone upgrades
- windows ui open isn't right
- jack on linux
- basically in mpe mode interpolate between keys for mts tuning in voice.cpp
- Implement phase remap done
- Implement phase formant sweep + window style done
The ratio editor is a knob. That's clumsy. segmented control will be better. Want modes - like segmented float,
- Add a simple built in spectrum and scope
- Make the upsample / downsample path have more control and better viz
- Specifically have the pipeline be
- Run at internal sample rate to make high bit rate clean signal
- Soft drive saturator (use the cubic clamp form) with optional small drive
- Add a cytomic SVF at 16khz / root 2 resonance
- ZOH decimate the signal to 32khz at upper rate
- Do a bitheight redicution (round(f * (1<<b)) / (1<<b))
- Add an optional highpass at 5-15hz to kill DC
- Downsample with strategy
- Each of those steps except the first and last is optional.
- basically all options to make it less 'clean' which all toggle at internal block
- Names
- LFOs and Envelopes per voice with power siwtch
- In-use display
- Make sure everything is themed DONE
- Write a jucegui theme editor which allows constaints to used subset I guess DONE
- Add an audio input and allow operator 1 to take the audio input
- Tricky part: have to upsample from audio input rate to six sines rate, so there will be some input latency.
- Mostly useful for really some FSU FM work.
- Add an author field for goodness sake and show it in the UI somwher
- that crazy idea kisney and i chatted about
- more t/k
- Allow a user to load a wav and take L channel (if stereo) or mono as a single cycle wavetable.
- Probably allow CSV of floats also
- Gotta upsample to 4096 and take derivs to make it work like tables
- Gotta stream into patch
- Since we want sharing these are probably 'patch level' slots which the node adddresses
- Really worried about aliasing here.
- Note Expressions?
- Clap Polymod?
- Definite maybe at best for these tho
- Remember that super cool paper on all pass networks as phase modulation?
- Well make that a mode if I can.
- Not sure if this fits in the architecture properly - might have to do all-pass mods at the end after the signal isotherwise made etc