Re: [tim-janik/beast] BlepSynth initial implementation (#142)

[Tim Janik <notifications github com>]

Great!

Looks like a really cool device. I've taken your advice and moved the sources into devices/blepsynth/ and adjusted the commit messages and sources. Under devices/, no dedicated Makefile.mk is needed, the *.cc files are picked up automatically.

I've also cleaned up the includes (e.g. we always use our internal assert_return()), implemented NOTE_ON, NOTE_OFF and ALL_NOTES_OFF.

To address some open issues in the sources:

TODO: unison_voices property should be an integer property, range 1-16, default 1"

• What is missing here? You can just set min/max and step=1. Later we can add a hint that a number-input might be better than a knob here.

TODO: cutoff property should have logarithmic scaling

• I need the exact mappings here, e.g. minimum, center, maximum, e.g. 32.7, 523.25, 8372.02?
  These are C values taken from: https://www.inspiredacoustics.com/en/MIDI_note_numbers_and_center_frequencies
• Or a function with input range, i.e. the above is roughly: 32.7 * 2**x

TODO: we need non-linear translation between percent and time/level

• Same here, does the discussion we previously had on panning help here?

TODO: replace this with true midi input

• Done, can we get rid of the Note toggles now?

TODO: should be easier to get choice value

• Yes, I agree, we should make a topic on choice -> audio value mapping.

TODO: under some conditions we could enable SSE in LadderVCF (alignment and block_size) *

• Anything needed for that from my part? We already utility SIMD and auto-vectorization in the production build.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or unsubscribe.