Hi all,
I noticed that when I use interpolation (both 2 or 4) with map_type 1, the
interpolation algorithm mixes the values of the top and bottom edges of the
image (that should be two completely different points). In the following script,
the "test.bmp" is a simple bitmap with top half white and bottom half black. If
you deactivate the interpolation option, the spheres are rendered correctly,
while if you activate interpolation the white half sphere shows a small black
spot in the center, and the black half sphere shows a small white spot in the
center. In the script I render two spheres to see the effect on top and bottom
simultaneously.

In my opinion, the algorithm should avoid the mixing between top and bottom
edges of the image.

Thanks,
igmar

------------------------

// Camera ########################
camera {
    location <1, .8, -7>
    look_at 0
    angle 50
}

// Ambient ########################
plane { y, -1.5 pigment { rgb 0.5*<.5,1,.8> } }

background {  .7*<1,1,1> }

light_source { <0, 2, 0>, color rgb 1*<1, 1, 1> shadowless }

// image_map texture ########################
#declare texmap = texture {
        pigment {
             image_map {
                bmp "test.bmp"
                map_type 1
                //interpolate 2
                filter all .2
        }
    }
}


// Objects ########################
sphere {
    <0,0,0>, 1
    texture {texmap}
    rotate -x*210
    hollow
    translate -x*1
}

sphere {
    <0,0,0>, 1
    texture {texmap}
    rotate -x*50
    hollow
    translate x*1
}

Post a reply to this message