|
 |
I'm trying to render a wall texture using an orthographic camera just in
front of a wall. (see image)
There are some shadows of boxes cast onto the wall.
The round blobs indicate the corners of the wall.
Reading the doc's, it looks like right and up vector set the size of the
camera, however my corners (round blobs) don't end up in the corners of the
rendered image. Can someone help me with this ?
The walls are generated in rhino3d, then exported to pov-ray. I add the
ray-traced textures back into the model, then export it to 3ds, then import
that into blender, then generate a standalone walkthrough using the blender
game engine.
// render options: +w256 +h256 +FN
#declare texture_corner_1 = ....
#declare texture_corner_2 = ....
// claculate camera position and size:
#declare texture_corner_1xz = texture_corner_1*<1,0,1>;
#declare texture_corner_2xz = texture_corner_2*<1,0,1>;
#declare textre_width_mm = vlength(texture_corner_1xz-texture_corner_2xz);
#declare textre_height_mm = vlength(texture_corner_1.y-texture_corner_2.y);
#declare texture_center = (texture_corner_1+texture_corner_2)/2;
#declare texture_normal = -z;
// orthographic camera projection, using parallel camera rays:
camera {
orthographic
location texture_center+texture_normal*5.0 // 5.0mm in front of surface
look_at texture_center+texture_normal*2.5 // 2.5mm in front of surface
right x*textre_width_mm // horizontal size of view
up y*textre_height_mm // vertical size of view
}
sphere{texture_corner_1,150 }
sphere{texture_corner_2,150 }
Post a reply to this message
Attachments:
Download 'wall_17.png' (24 KB)
Preview of image 'wall_17.png'
|
|
com> wrote:
nevermind, got it fixed.
vlength() doesn't work very well when the input isn't a vector...
