Behaviour of approx. indirect lighting in 2.5 beta

Hey,

curious I follow the development of the new Blender 2.5 and first i was quite fascinated by the new indirect lighting thingy, which replaces the old radiosity feature, i guess.

First i’m a little bit confused by the fact, IL is only available in approximate lighting mode.
In my opinion it’s a great improvement of the vertex-based radiosity-lighing, but can IL really only be done with approximated results. Ok, i’m not really into all those how-to-calculate-lighting programming things, but could there be a raytraced version at all, or is it just a question of coding it?

So here is a little example of my work. I was playing around with IL and thought like “Wow, that can’t be right.”
Decreasing the number of bounces of course seems to reduce the incorrect light at the left, but isn’t that completely against the idea of indirect lighting? I know, that the aim of IL is basically to make surfaces emit light and avoid point light sources. But if the emitted light (in the IL bounce-around process) does not really care if there’s a huge wall or not, it is quite annoying while rendering a scene, which at least should have physically correct lighting. Isn’t that the final goal of all our lighting calculations?

Is there any workaround to make the left “room” completely dark, while keeping the benefits of using indirect lighting?

Best regards,
Jilo

Attachments



IL reautayce is udner renderbanch so i think you need to find a renderbranch version to avhe this beginning to work but it is still a WIP and not yet into 2.5
but hopefully should come soon

and even the approxiaamte is still under developement but giving some good results compared to 2.4X

find the IL in News and discussion forum there are some example of it

here is one example
http://blenderartists.org/forum/showthread.php?p=1692275&highlight=mushroom#post1692275

salutations

Thanks, i’ll have a look at graphicall.org