"Samuel Benge" <stb### [at] hotmailcom> wrote:
> "Simon Copar" <inf### [at] gmailcom> wrote:
> >
> > This could actually be simulated as a thin hollow shell filled with absorbing
> > media, but that's overkill for this purpose.
>
> Probably. You might try the aoi pattern if you haven't done so already.
>
> pigment{
>   pigment_pattern{
>     aoi
>     color_map{[.5 rgb 1][1 rgb 0]}
>   }
>   poly_wave falloff_val
> }

Right, didn't think of that. The poly_wave does not approximate the realistic
angular dependence that well (it starts rising too quickly, makes the almost
perpendicular regions too opaque), but I can just sample the curve by myself
with more entries of the pigment_map (or just push 0.5 to something like 0.75).

This will make the images so much better.

Still, built-in method would be nice to have.

Post a reply to this message