Page 1 of 1

Polarisation and Brewster's angle

Posted: Mon Dec 11, 2006 4:21 am
by OnoSendai

Posted: Mon Dec 11, 2006 4:47 am
by u3dreal
cool stuff going on there :D


u3dreal ;)

Posted: Mon Dec 11, 2006 5:02 am
by manitwo
very cool ono !

Posted: Mon Dec 11, 2006 5:05 am
by Kosmokrator
yes very cool but what is exactly pollarization of light?i mean what we can archieve with this method?less noise?more quality?more acuracy of rendering?

Posted: Mon Dec 11, 2006 5:07 am
by OnoSendai
Kosmo: it's needed to achieve full realism with glass, sky, water etc..
No, I'm not going to back that statement up :)

Posted: Mon Dec 11, 2006 6:12 am
by zsouthboy
For example, for those of you wondering what this is for, a photographer can put a polarizer filter on his/her lens to filter reflections in a pond (the surface of the water *disappears* if you turn it just right), or enhance the blueness of the sky in a landscape picture.

It's just another step towards Indigo emulating reality exactly. :)

For example, look at the picture on this page: ... ype=search

Posted: Mon Dec 11, 2006 6:18 am
by OnoSendai
altho i didn't add polarisation so that we could all use polarising filters, tho that might be nice.
I added it so that stuff like glasses of water, the sky reflected in a pane of glass etc.. will be more realistic.

Posted: Mon Dec 11, 2006 7:40 am
by atmmatt
zsouth, yikes that's a gigantic filter!

I use polarizers in astronomy quite a bit. If you take two polarized filters and rotate them together you can knock down the incoming light to something under 5% (I think). Also if you have the same two polarizing filters and look through them at glass (with a laptop screen behind) you can see strain and stress points inside the glass. Mostly use the polarizers to look at the moon and not get blinded from the bigger telescopes.

p.s. ono, will the polarization be a property of the glass material or something we can add and control?


Posted: Mon Dec 11, 2006 7:54 am
by oodmb
it would be nice if it was something that we could control in the material as compared to only the camera, i know that certain materials reflect light from different angles better than others and certain materials polerize light themselves, i also think some materials have changing polarization of light, for example thin layers of oil on water. however, it would be nice it this wasnt a necessary statement in the material code as many people would have absolutly no idea how to get this effect working or looking nice.

it would also be cool if you could make this effect effectable by a texture.

this seems to be the biggest advance of indigo compared to biased renderers like yafray.

Posted: Mon Dec 11, 2006 10:41 am
by CTZn
Indeed, what is this effect based on ? IOR, surface roughness, ... ? Just curious...

Posted: Mon Dec 11, 2006 11:51 pm
by Maluminas
Is it cpable of calculating polarized light rotation by opticaly active materials? It can be used for interesting effects. For example some opticaly active biomolecules rotate the polarized light plane differently at different wavelenghts so you can "select" the wavelenght that goes throught. A simple setup to see this would be a lamp, two filters and a glass of corn syrup (glucose and fructose are optically active). Place the glass of syrup between the filters and place the lamp behind this sandwich. Rotate only one filter slowly and see the color of the syrup change in the colors of the rainbow because when you rotate the filter you "select" the color that passes throught at this exact angle. And when the axis of the two filters are perpandicular and no light should go throught, the will actually be some light going throught because the sugar dephases it and it no longer is polarized perpendicular to the second filter.

Did that make any sense? :lol: Its not necessairy for realism it would just be cool if it did it :D

Posted: Tue Dec 12, 2006 12:52 am
by CTZn
It made sense to me, but I think I'm gonna stop corn right now, sounds psychedelic :lol: I think OnoSendai's point was a bit more basic, while it would already make reflections and refractions even better !
OnoSendai wrote:i didn't add polarisation so that we could all use polarising filters, tho that might be nice.
Steps, one by one ;)

Posted: Tue Dec 12, 2006 11:20 pm
by OnoSendai
The polarisation effects on reflection won't be configurable. Rather they arise naturally from the Fresnel equations for reflection from a dielectric or metal, which differ for the components of the light in the different polarisation directions.

Maluminas: I'm not gonna try and do anything that complicated :)