PhotonGI cache
Re: PhotonGI cache
Dade, how is glossines threshold handled when there is a glossines map used??
Re: PhotonGI cache
What do you mean with glossiness map ? A texture applied to glossy UV roughness ? In that case, the average texture value is used for the comparison with the threshold.lacilaci wrote: Tue Feb 12, 2019 6:52 am Dade, how is glossines threshold handled when there is a glossines map used??
Re: PhotonGI cache
Isn't that a side effect of caching being a biased Solution ?But I still get weake light in GI (using same depth for photons as for GI)

Re: PhotonGI cache
Cache entries are still created only on "DIFFUSE | REFLECT" and "GLOSSY | REFLECT" materials. I will start to lift that limitation only after everything is more stable and all major problem has been fixed. We are pretty near to that point.
Re: PhotonGI cache
It is a factor but the difference is still a bit too high, there may be a bug. I will check after I have tried the same solution with caustic cache.Sharlybg wrote: Tue Feb 12, 2019 9:40 amIsn't that a side effect of caching being a biased Solution ?But I still get weake light in GI (using same depth for photons as for GI)![]()
Re: PhotonGI cache
I'm also getting strong cache samples near corners (larger lookup radius helps in some cases but there is often few samples that are too bright).Dade wrote: Tue Feb 12, 2019 9:42 amIt is a factor but the difference is still a bit too high, there may be a bug. I will check after I have tried the same solution with caustic cache.Sharlybg wrote: Tue Feb 12, 2019 9:40 amIsn't that a side effect of caching being a biased Solution ?But I still get weake light in GI (using same depth for photons as for GI)![]()
It reminds me of the problem you had before, where you had strong fireflies near corners, but now it seems there are often strong cache samples near corners while more away from corner they are weaker than pathtracing.
Re: PhotonGI cache
here's how that bleeding looks. I'm getting this everywhere (probably related to glossy reflection ?)
Last edited by lacilaci on Tue Feb 12, 2019 11:19 am, edited 1 time in total.
Re: PhotonGI cache
BTW on the topic of caustics... could reflective caustics be handled similar as gi (cache not directly visible) and specular caustics separately(directly visible)?
Re: PhotonGI cache
My Blender 2.79b says the scene has been written with 2.80 and crasheslacilaci wrote: Tue Feb 12, 2019 10:18 am here's how that bleeding looks. I'm getting this everywhere (probably related to glossy reflection ?)

Re: PhotonGI cache
I'm sorry, wrong file...
here: