Assuming that this question is related to the other one that you posted, in my opinion you don’t really need to go the 5-Phase route. The Three- and Five-Phase method make sense if you have complex fenestrations (like blinds, perforated shades etc) and if there multiple types of those to evaluate parametrically. Since you are dealing with a standard glazing, it would be a lot simpler to just use the standard Daylight Coefficient Method or the 4 Component Method.
To address your queries specifically:
- Creating BSDFs for clear glazing: Since you already have your glazing data in Optics-compatible format, you can create a BSDF for it by using genBSDF. You can export a RAD definition from Optics, then use that material with a simple glazing polygon and follow the genBSDF tutorial to export a XML-format BSDF. For the final step of the 5-Phase method, instead of the Tensor-tree BSDF with proxy-geometry, you can use the standard glazing polygon directly (There might be a better alternative to this, I am just pointing out the way I know how to).
- “Why simulate with eastBlinds.vf ?” from the tutorial: I was just trying to demonstrate the matrix-reuse feature within the multi-phase methods i.e. one only needs to recalculate the View Matrix and all the other matrices can be reused from the previous simulation for the south view. So, you don’t have to simulate multiple views if they aren’t required.
Regards,
Sarith