4
$\begingroup$

I'm confused about IOR in Blender Internal as used in raytraced transparency. To demonstrate my issue, I set up a camera looking through a transparent sheet with IOR>1 (and slightly tinted) at a textured surface. You can clearly see the expected refraction.

But when I flip normals, I get exactly the same render! My question is therefore: How does Blender Internal determine the "in/out"-side if it does not use the normals for that?

Placing the camera inside a properly closed transparent volume with IOR does not change anything.

Clearly, I can invert the IOR to treat camera rays, but I came across this issue when reflections going out and then back into the glass material misbehaved.

Scene Setup Refraction visible

$\endgroup$
3
  • 1
    $\begingroup$ It doesn't use normals to determine where the ray is. It just swaps a traversal flag on ray pass. $\endgroup$
    – user2859
    May 8, 2014 at 8:48
  • $\begingroup$ I think this is the answer, result looks fine to me now, thank you! $\endgroup$
    – user15943
    May 13, 2014 at 6:18
  • $\begingroup$ @user2859 Feel free to add that as a proper answer $\endgroup$
    – gandalf3
    Jul 23, 2014 at 6:22

1 Answer 1

1
$\begingroup$

From user2859's comment:

It doesn't use normals to determine where the ray is. It just swaps a traversal flag on ray pass.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .