To get the solar BSDF of micro-prism film, we have got the geometry using 3D laser scanner, and the first thing I thought of was using genBSDF command to generate the BSDF.However, I feel something wrong that the characteristic dimension (L) of micro-prism film is about 20μm, and the wavelength of infrared wave is larger than 0.78μm (λ), is ray-traced method applicable to this condition(λ/L < 25)?
I believe that most prismatic films are scaled so as to minimize diffraction effects, but they will still have some color fringing and slight directional preferences related to wavelength. A simulation in genBSDF would not capture this behavior as you surmise, and genBSDF usually doesn’t include color, anyway.
In any case, your question about how to import BSDF data is a good one, but not an easy one to answer. It is possible, but direct specification requires that you use either the Klems or Shirley-Chiu sampling directions, similar to what is done in src/util/genBSDF.pl. If your sampling pattern is different, you can try feeding the pabopto2bsdf tool with an appropriate set of hemisphere sampling files, as if you were working from goniophometer data, then process the output via bsdf2klems or bsdf2ttree as we do for our BSDF measurements.
Doing any of the above requires detailed knowledge that is not available in any cohesive set of documents, which makes it a real challenge. My answer would have to span several pages to get you started on just one approach, and I don’t know which approach is best for you.
