The current use of Influx in NTMI is linear: interface control outputs are weighted by a matrix, and the resulting output is linearly mapped to process control inputs.
The Influx quark has an example for terrain-like remapping on the influx output, which makes the range of accessible areas in parameter space much wider:
https://github.com/supercollider-quarks/Influx/blob/master/Examples/Remap_Influx_by_envelope.scd
It would make sense to include this option in the NTMI, maybe as 3 steps:
1 - linear mapping as was, maybe with tanh compression
2 - remapping by simple terrain envelope, e.g. with 7 steps
3 - remapping with more complex envelope, e.g. 11 or 13 values.
The current use of Influx in NTMI is linear: interface control outputs are weighted by a matrix, and the resulting output is linearly mapped to process control inputs.
The Influx quark has an example for terrain-like remapping on the influx output, which makes the range of accessible areas in parameter space much wider:
https://github.com/supercollider-quarks/Influx/blob/master/Examples/Remap_Influx_by_envelope.scd
It would make sense to include this option in the NTMI, maybe as 3 steps:
1 - linear mapping as was, maybe with tanh compression
2 - remapping by simple terrain envelope, e.g. with 7 steps
3 - remapping with more complex envelope, e.g. 11 or 13 values.