I want to model a system of mirrors using blender, hopefully depicted below where E is the “Eye”, A & B are mirrors and “O” is the object. The first thing that comes to mind is a solution using environment mapping or using multiple camera’s and textures. It would be nice if the solution is as efficient and hassle free. Any ideas?
The object controlling the envmap (usually an Empty) as to be in the exact simetric position from the camera, using the mirror as reflection axis.
That is, if you have some thing like this:
C
|
|
|
/—E
where / is a mirror at a 45 degree angle, C is the camera and E is the empty.
There is no easy way to do this by hand (unless your mirror surface is one of the major planes (XoY, XoZ, YoZ). There’s also a Python script to automaticly place the empty. I think it was done by Jean-Marie Soler (jms in this forum).
Ummm, I am working in a plane axis with the mirror at 45deg so it’s easy enough to set up and I am now getting a reflection but multiple images (2). It does not seem to matter what mapping option I choose (flat/cube). Any more ideas?
Dug out some mirroring tests I did a long while ago. The best effect I get is to shift the mirror planes center and name the object itself as the target (instead of the recommended empty). Catch is that the results only seem good for shallow angles, and much less than 45deg Maybe this is why multiple camera and texturing jump to mind. Also suspect that a script may help. All suggestions welcome!
Yep, like I said I do get reflections, correct placing/scale but only if the mirror suface is head-on to the camera (haven’t checked precisely but estimate +/- 30deg).