features to include (largely based on a FR by Pete Dowling):
[ ] access to easing functions
[ ] or attributes for curve segments. these curves should be by default like [scale @classic 0] curves.
[] or optionally there should be the possibility to switch to the curve algo ej put in [function] - seems crazy there is currently no (non-linear) relationship between the stock adsr~ and the stock graphical editor.
[] or automagically link a named function object to a named adsr~ object, like a buffer~ (i often make my own buffer~ write of a custom function then index~ to read, but really not ideal).
[] all segments could optionally also have buffer~s with custom functions available to the individual segments of the envelope.
[] the segments internally could be modular, so it would be possible to use the object to create multi-segment envelopes greater than an adsr, so the user can add hold/sustain sections and additional timed segments at will.
[] the modular would allow for as many a, or d, or r, sections and inserts of any hold / sustain sections.
[] inlets would just grow, attributes referenced via segment index.
[] represent the structure with a dictionary?
[] the above refers to what i’d call a ‘digital’ adsr~ object. i.e., segment end points are known entities and the object can continue to be used to control [thispoly~].
[] however, there could also be an @analog mode (or a new object) where real impulse/decay sections are used (a bit like slide~ i guess).
[] this could either not be able to control thispoly~, or it could use the peter mcculloch / hal chamberlain thing to predict overshoot and curtail, or the simper smoother, or whatever, so it could still be used to control thispoly~.
[] sync inlets / outlets for passing normalised phase or count structures through envelope sections.
features to include (largely based on a FR by Pete Dowling):
[ ] access to easing functions
[ ] or attributes for curve segments. these curves should be by default like [scale @classic 0] curves.
[] or optionally there should be the possibility to switch to the curve algo ej put in [function] - seems crazy there is currently no (non-linear) relationship between the stock adsr~ and the stock graphical editor.
[] or automagically link a named function object to a named adsr~ object, like a buffer~ (i often make my own buffer~ write of a custom function then index~ to read, but really not ideal).
[] all segments could optionally also have buffer~s with custom functions available to the individual segments of the envelope.
[] the segments internally could be modular, so it would be possible to use the object to create multi-segment envelopes greater than an adsr, so the user can add hold/sustain sections and additional timed segments at will.
[] the modular would allow for as many a, or d, or r, sections and inserts of any hold / sustain sections.
[] inlets would just grow, attributes referenced via segment index.
[] represent the structure with a dictionary?
[] the above refers to what i’d call a ‘digital’ adsr~ object. i.e., segment end points are known entities and the object can continue to be used to control [thispoly~].
[] however, there could also be an @analog mode (or a new object) where real impulse/decay sections are used (a bit like slide~ i guess).
[] this could either not be able to control thispoly~, or it could use the peter mcculloch / hal chamberlain thing to predict overshoot and curtail, or the simper smoother, or whatever, so it could still be used to control thispoly~.
[] sync inlets / outlets for passing normalised phase or count structures through envelope sections.