create less lights for the scene than for the grid?

[hammerbchen]

Hi,

While reading new feature list from here:
https://docs.chaosgroup.com/display/PHX3MAX/3.00.01

I don't understand this one:
"Using grid-based self-shadowing, create less lights for the scene than for the grid"

Could someone explain to me?


Chen

[Svetlin.Nikolov]

Hey,

Phoenix generates lights at render time for the fire and those work to illuminate the smoke of the simulator and the scene around it even without using GI. The perfect-looking case would be if there was one light per each fire voxel, but of course this would render forever. You have the Light Resolution which creates the lights from a low-res fire grid in order to reduce their huge number. However, the smoke can use grid-based self-illumination which is way faster than the ray-traced one, so it might work with more lights, but the scene around the volumetric would render quite slow, especially when there are complex geometries. This is why, when you set the Light Resolution to 5%, this applies for the volumetric itself, but for the scene it would actually be e.g. less than 1%.

Eventually we have to add a new light type to V-Ray that would render properly and fast with glossy surfaces and would use V-Ray's probabilistic lights feature, but until then these lights are managed by Phoenix and unless we copy the rendering core into Phoenix as well, they will have some limitation and would render slower..

I hope this helps Cheers!

[hammerbchen]

Hi Svetlin,

Thanks for your detail explanation. It is hard to imagine the benefit of this feature by reading the line, a side by side comparison might help. Will try to make one myself.

Cheers,