Right now in daily builds it looks like this:
luxcore:
e-cycles:
the e-cycles doesn't use portal but luxcore does use vis. map... This is because in such simple scenario portal for e-cycles doesn't make any difference.
What I find very interesting is that sidewalls and ceiling is much cleaner in e-cycles, but back wall is cleaner in luxcore.
Another important thing is that I used spp as halt factor for both renderers so each is 100 samples(I do know it's not the same thing for every renderer)
But still e-cycles was done in 6 seconds while luxcore 12. While I learned by now that luxcore gets better than ecycles as complexity of the scene grows, it is still beyond me how noisy soft shadows are in luxcore.
This noise in shadows gets impossible to clean in some cases:
I assume however that atm the new system isn't enabled by default...?
Env. Light Visibility Cache
Re: Env. Light Visibility Cache
The code is currently on the "envmap_cache" branch (https://github.com/LuxCoreRender/LuxCor ... vmap_cache) so it is not part of the daily build. I have still to write several parts to be able to render any scene, support OpenCL, etc. I will merge the branch when the CPU code is done so it can be tested.
It will available as an option like visibility map (but not enabled by default).
It will available as an option like visibility map (but not enabled by default).
Re: Env. Light Visibility Cache
This could be super striking to have guided method over Env + Area + Meshlight + point & spotlight MIXED with Photon GI cache will rock like Sun CoreThis method is very similar to a guided method. It "guides" only for env. light sources but it could "guides" for any kind of incoming light (direct and indirect).
Re: Env. Light Visibility Cache
Yeah I actually wonder, could this help sampling heavy meshlights?
Re: Env. Light Visibility Cache
Not this specific cache but the guided generic extension aiming for any kind of light source, it would.
However it must be clear that it is pretty much "Path+GPU+PhotonGI" Vs "BiDir+CPU+Guided sampling" (i.e. guided sampling make most sense to accelerate BIDir and if it is worth doing, it is quite debatable compared "Path+GPU+PhotonGI").
Re: Env. Light Visibility Cache
Do you think Path+PhotonGI caustics will ever be able to fully substitute BiDir? (glossy, volumes, etc.)
Re: Env. Light Visibility Cache
And some sort of light grouping and approximation isn't possible? With hi-poly meshes that have same material..
Re: Env. Light Visibility Cache
So do you mean there is no gain in speed by adding guiding method on top of PhotonGI ?However it must be clear that it is pretty much "Path+GPU+PhotonGI" Vs "BiDir+CPU+Guided sampling" (i.e. guided sampling make most sense to accelerate BIDir and if it is worth doing, it is quite debatable compared "Path+GPU+PhotonGI").