Indirect lighting on translucent particles

Hi there,

I was testing indirect lighting on particles (with opaque and translucent materials).
When I use opaque material it’s ok, but with translucent material “ILCQ volume” parameter seems not taken into account and all particles have the same color.

I was wondering if it’s the expected result?
Is there a way or a planned feature to improve the translucent particles reaction regarding indirect lighting?

Thanks.

With opaque material

With translucent material

With animation :

Opaque particles

107326-opaque_particles.gif

Translucent particles

107327-translucent_particles.gif

Hi Sylvain,

Since we use a deferred renderer, translucent materials use the translucent light volume, which is a low-resolution approximation of the scene lighting;could you try checking r.TranslucentLightingVolumeDim and see what the dimensions of volume are, as well as adjusting it higher? Also, r.TranslucencyVolumeBlur may have an effect on the visual outcome. Please let us know if increasing the size of the volume helps with your problem.

Thanks,
Olaf

Hi Olaf

Thank you for your answer.

Base on what you said.
I’ve looked at the “r.TranslucentLightingVolumeDim” settings and here they are:

r.TranslucentLightingVolumeDim = 64

r.TranslucencyVolumeBlur = 1

I 've tried to increase those values but I didn’t see any improvement.

To be clearer, here’s a short description of the setup that I use on the ScreenCapture/GIF :

  • I only use 2 baked static lights: one green and one red (no dynamic lights).

  • I use only one particle system (with only one emitter) and the “Indirect Lighting Cache Quality” setting is set to “ILCQ volume” but it seems react like I have set “ILCQ Point”.

I feel that there is only one sample for the entire particle system because all particles have the same color changing at the same time from red to green when I move the Particle system.

Thanks

Hi Sylvain,

Ah, I misunderstood the question originally. I just reproduced this in a few different configurations. Yes, it appears that it’s only using the actor position when it should be using the particle or vertex world position (depending on lighting mode). I believe this is a bug - I’ll look into this a little further and will open a ticket shortly.

Thanks,
Olaf