When voice was silent, the exponential decays in the filter unit
were causing the high pass component to eventually denormalize,
causing high CPU loads. The solution is the same as in the delay
unit: add and subtract a small number from the value, causing
essentially a flush to zero.
https://en.wikipedia.org/wiki/Subnormal_numberFixes#68.
The modulated delay time was converted to int with i32.trunc_f32_u.
This throws runtime error if the modulations caused the delaytime
to become negative, because _u implied that it should be unsigned
integer and negative numbers were out of range. Using
i32.trunc_f32_s fixed this.
Quite often the user wants to experiment what particular unit(s) add
to the sound. This commit adds ability to disable any set of units
temporarily, without actually deleting them. Ctrl-D disables and
re-enables the units. Disabled units are considered non-existent in
the patch.
Closes#116.
The Model was getting unmaintanable mess. This is an attempt to refactor/rewrite the Model so that data of certain type is exposed in standardized way, offering certain standard manipulations for that data type, and on the GUI side, certain standard widgets to tied to that data.
This rewrite closes#72, #106 and #120.
text.Shaper is not thread safe, which caused crash when adding
multiple VSTI plugins to a DAW project. This change fixes that
crash. Further refactorings need to consider where that text.Shaper
should actually reside.
The options are stereo, left and right. Similar to oscillator sample settings, if you tamper with these, it starts to show "custom". Used some of the generic features of go1.18, so had to update go.mod to require go1.18.
The "auto" was misleading, as it meant self modulation when targetting a unit within instrument itself and just voice 0 when cross-instrument modulation. This feature changes the "auto" meaning "self" for instruments self-modulating, and "all" voices for cross-instrument modulations. "all" is implemented by compiling a single send into multiple repeated sends, with only the last popping the stack (if necessary).
Closes#107
BREAKING CHANGE: The problem with crush was that it had very few usable values. This changes the crush to map the value nonlinearly, so the crush resolution is bits. Still the upper portion of the values is not very usable (bits 12-24 i.e. hardly any crushing), but at least the lower portion is usable. But now crush resolution has slightly different meaning.
