I’ve done some experiments, and I’ve concluded that completely mixing the filament is unnecessary, and that with the rudimentary experiments I’ve been able to indicate I can get a color resolution less than 1mm within a reasonable nozzle size, no chamber necessary.