Why does these two images look so different?

[mib2berlin]

Welti, here:

http://render.otoy.com/forum/viewtopic.php?f=33&t=24877

Cheers

[FooZe]

Hi Welti,

These images are interesting.
I'm noticing a few things. Have you changed the geometry of the emitter at all for these test?

The first seems to have a ring of light on the back wall - not sure where this is coming from. Also there seems to be no shadow from the glass ball - this seems un-natural to me, pictures of things like a glass full of water in the sunlight on google would show that you might expect some kind of shadow cast by the ball.
The second (non-octane) one seems to have a lot of light on the back wall and floor as if the emitter did not have the tube round it? This also shows no caustics that i can see.

I have given your obj a test in octane with a few notes:
1) I used PMC.
2) The "efficiency" also effects the effective power of an emitter so play with this too.
3) I used RC3 with some caustic blur.
4) There will be extreme contrast between the lit area and the black room. This means you will need to play with the imager settings to get anything usable (gamma, exposure etc).
5) I gave the ball some absorption to color the caustics a little.

Hope that helps.

Cheers
Chris.

[Welti]

Thanks for your request,

i think the light circle in the background of the LuxRender picture is a caustic from the glass sphere.

In the SLG Scene i used no glass sphere but a reflective sphere (my fault, so it is not compareable).

I will try the scene again after i download the RC3 (thanks for the link mib).

Edit: I think this will take for ever to render...


Thanks all for your help and a beautiful evening.