Pardon me for being intrusive . . . but I wanted to share an alternate
solution to this problem which occured to me. It is a bit cheaper in parsing
time (though not much) and loop cycles.


#declare Radius = 11;
#declare N = 2500;
#declare C = <0,8,0>;
#declare s = seed(pi);
#while (N>0)
  sphere {
    <0, 0, 0>
    0.25
    pigment { rgb <1,1,1>} finish { ambient 0.2 diffuse 0.6 }
    #local D =  pow(rand(s), 1/3)*Radius;        // cube root accounts for
increase in rad. wrt vol. Changing this can change the radial distr. of
spheres.
    #local R = < rand(s)*360, rand(s)*360, rand(s)*360,>;
    translate x*D
    rotate 45
    rotate R
    translate C
   }
   #declare N=N-1;
 #end

. . . for what it's worth. :-)

Abe

Post a reply to this message