I would like to know if developers are going to implement other SSS algorithms or somehow improve current profile. I appreciate their work, but I still want better SSS.
What exactly do you mean by “better” SSS? Faster rendering? Higher accuracy? More features? Easier control? …?
I mean higher accuracy and more realistic behavior. Let me show some examples in my next post.
Still, random walk would be a great addition.
Quick rundown on the benefits of random walk SSS vs the current implementation?
TBH I think there are more pressing things that need work over SSS improvements (tone mapping!)
You’re absolutely right, there’re more important things to work on right now. Random Walk SSS should be treated as a suggestion for the future versions.
The first and foremost advance of RWSSS over Gaussian is its accuracy, especially on models with thin parts.
I think, implementation of RWSSS is just a matter of time.
A great implementation and explanation done by the redshift dev team on random walk SSS are worth looking into. Maybe the corona team will do something analogous for the future release version.
Just saw these examples on Twitter from another Redshift user Saul Espinosa on X: "More Random Walk experiments using @redshift3d, renders blazing fast only 26s, no denoising and samples great. 🔥 You can read more about it here https://t.co/j2hE6nws9k #Maxon #Redshift3d #cgi #SSS #RandomWalk #digitalart #3d #vfx https://t.co/5U8FHBgiqw" / X