Thinking of proximity compensation as a multidimensional vector space

For example:


When using a cardioid mic there is an optimum distance from mic to mouth for a natural non bass heavy brain of listener response. This distance may be 1-3ft. If a Pleiades gentle slope hush pass is used it may be 3in to 1ft.


Other parameter is the kind of song or voice register etc.


Other parameter may be the type of mic.


So we have a vector space of for example such elements:


(mic type, singing song register, Pleiades low cut filter cutoff and slope) <--> mouth to mic distance


The optimum distance may vary at different parts of a song.