Now OK ... following to the discussions we had I'd made first
test ... not very satisfiing..

The issue of creating objects with variable optical densities
inside the object.
My first idea was to create different objects inside the
Master-object.

My machine is still on render, but I tried it first with a method
in pure pov.
I'll post it when it's done ... maybe someone of you is faster
...

This was easy to do with iso-surfaces, manipulating the
threshold-value.

this is the current scene, I'm using to test..

-------- var_ior_tst.pov ----------------
//POV ... give Var_ior a 1st try ...
#version unofficial MegaPov 0.5;    // do I need this ??? dont
know .... but, why not ...

global_settings {
        max_trace_level 512       // Important, if you don't
want to see
        max_intersections 512     // black holes ...
        }


////////////////////
// Standard stuff following ...
// Just to see something ..

camera {
 location <-5,25,-30>
 look_at <0,5,0>
 angle 25
        }

light_source {<0,17,13> color rgb <0.5,0.5,0.6>}      // Maybe 6
lightsources are to much ... (aren't they ???)
light_source {<5,17,-10> color rgb <0.6,0.5,0.5>}
light_source {<-15,17,6> color rgb <0.5,0.6,0.45>}
light_source {<0,3,23> color rgb <0.5,0.5,0.6>}
light_source {<50,5,-10> color rgb <0.6,0.5,0.5>}
light_source {<-15,3,-46> color rgb <0.5,0.6,0.45>}


plane { y,0                                    // the typical
checkered plane .... (I can't get without this)
        pigment {
                checker color rgb 1, color rgb 0.5
                }
        }

sphere {0,2.5
        translate <4,1,5>                      // A blue sphere
right behind our object ....
        pigment {color rgb<0,0,1>}
        finish {brilliance 10 phong 5 phong_size 50 reflection
0.3}
        }

sphere {0,1.5
        scale <3,1,1>                          // and a stretched
red one left behind it ...
        rotate y*-30
        translate <-4,1,5>
        pigment {color rgb<1,0.1,0>}
        finish {brilliance 10 phong 5 phong_size 50 reflection
0.3}
        }

//////////////////////////////////////
// Here is the interesting part ...



#declare M_density=0.04;   // How fine should it be ???
Bigger->faster .... Smaller->slower, finer ...
/*
can create a Rendering error: Too many nested refracting objects
...
so keep it big enough ...
*/

// Maybe a union{} of the following will speed this a bit ???
Don't know ..

#declare I=4;
#while (I>0)
#declare I=I-M_density;
isosurface {
        function { sin(x)+sin(y)+sin(z)} //This will make it
really slow ...
        //function { "sphere",<4>}       // I just tested with a
simple sphere ... But I think I need a cray ....

        contained_by {sphere{0,5}}
        threshold -I                     // This is a the
interesting piece ...
        translate <0,5,0>
        pigment {
                checker color rgbt <1,0.9,0,0.9975>, color rgbt
<0.9,1,0,0.9975>
                }
        interior{ior 1+I*0.025}          // And this is the other
part ...
        hollow
        }
#end

//EOF

-----------------------------------------------------------------
Of course, when testing, I use the sphere, not the other function
When setting M_density lower (to create more objects) there comes
the
 - to many nested refracting objects -
message ...
How can I get around this without patching in the source ???

Now I'll try other methods for realizing the main problem, but I
think therefore I have to read through the source (and understand
it) ... Maybe there can something be done in the media-code.

BTW: how can I to a isosurface from a crackle-pattern, or a
marble, or some other patterns ??? And with the use of a
turbulence ???


--

 ,',    Jan Walzer      \V/  http://wa.lzer.net     ,',
',','   student of      >|<  mailto:jan### [at] lzernet ',','
  '   ComputerScience   /A\  +49-177-7403863         '

Post a reply to this message