any ideas on how to approach (refractive) dispersion from a bsdf point of view (suitable for PBR, that is). will it help to compute dispersion the old way and simply convert the color to a bsdf and add that to the trace bsdf? mario? jason? stu? symek? cheers