1
$\begingroup$

I implemented a MicrofacetTransmission in nori. But it can not pass all Chi2test when alpha is large(>= 0.4 or ?),part of chi2test is passed.And I can not found the problem.Can someone help me? Nori framework. Code is below: sample function

pdf function rendered image

$\endgroup$
4
  • $\begingroup$ I suggest adding some example pictures. It may help uncovering the problem faster. $\endgroup$
    – ivokabel
    Commented Dec 29, 2021 at 20:33
  • $\begingroup$ @ivokabel Well, I added a rendered image. $\endgroup$
    – bin guo
    Commented Dec 30, 2021 at 3:23
  • $\begingroup$ Not exactly sure what hypothesis are you testing in your Chi2 test, but it seems that you use the basic microfacet model which only deals with directly scattered ligh and neglects multiple scattering events on the surface. This is not a big deal for low roughnes parameters, but at higher values you get noticeable loss of energy. That's probably why the sphere is so dark. This is pretty much expected behaviour and if you want something better you'll need to implement some kind of energy reinjection method. $\endgroup$
    – ivokabel
    Commented Jan 1, 2022 at 13:59
  • $\begingroup$ null hypothesis $\endgroup$
    – bin guo
    Commented Jan 10, 2022 at 12:52

0