BRDF related question...

hi all,
i am trying to play a little with brdf lighting,
so i’ve read a nice paper brought by nvidia,
explaining a bit of the maths underneath this tecnique.
I am a bit confused thou:
when indexing the drdf textures the nvidia guy
uses the light-ray/eye-ray in BTN coordinates space.
Why is that? why can’t i use the simple usual
object space coordinates?

thanks to you all…

the G.

The BRDF is defined to be in tangent-space, so your vectors must be there too. Take a look at the cube-map version of that paper.