Emitters + CoronaLights = noisy light from CoronaLights

[Polymax]

Why CorobaLights getting a lot noisier when used together with large emitter?

[maru]

Tried setting Lights Samples Multiplier to higher value? (if it doesn't work then try MUCH higher like 32)

I think it's some kind of adaptive light sampling that in this situation is focused more on indirect lighting.

[Polymax]

(if it doesn't work then try MUCH higher like 32)

Yes, i know, it's work, but very long!

[Polymax]

small correction, noise adds parameter "directionality"

[Ondra]

yes, I know about these defficiencies, I'll fix them in the summer.

[Polymax]

and another:
"Directly visible" parameter in the CoronaLight does not work properly if there is emitter.
1. CoronaLight ceases to give light about yourself
2. More noisy
3. Less intensity

[Ondra]

this is a feature. What I should do is separate "directly visible" and "visible in reflection" checkboxes, then you would uncheck only the first one and it would work ok

[Polymax]

ok, it will be fine!

[iliyang]

this is a feature. What I should do is separate "directly visible" and "visible in reflection" checkboxes, then you would uncheck only the first one and it would work ok

In either case, the intensity of the light shouldn't change. Only it's visibility and potentially the noise. That is, the "visible (in reflection)" setting should be accounted for in the MIS weights. Even with the current implementation the intensity shouldn't change depending on the setting.

[Stan_But]

anymore testing about noise and visibility of lights + emitters

http://3dcenter.ru/forum/index.php?showtopic=113708&st=1575&gopid=1476881&#entry1476881

may be it would be useful

one more thing: looks like increasing count of emitter's faces we get more noise. Solution is to make more simple geometry for emitters. Daily build 18.04.13

[Ondra]

"Directly visible" no longer influences intensity