>> I hope you like it. :-)

> 
> Nice.
> 
> Why is it such red ?


It looks not so bad on my office LCD. At home it looks more violet and
veries a lot when changing the viewing angle to the screen. Very bad.
You can give me a better rgb value or I try to render with Pov3.7 that
calculates colors differently (physically more correct?), and – even
more important – uses all 4 CPU cores in parallel! :-D

 
> 2 points:
> * there is more past the 2.0 limit (looking at 3, 4 and 5 for instance,
> probably not with 1/4 step). Maybe you can use some log/exp scale
> instead of a linear one. (1.0 is a must have, 0.5 also, as well as 2.0,
> other values might be something along k = 1+log/b (x) ; your choice of
> base b.) Central point might remains around 1.0;

I'll try tomorrow.

> * might have some issues of self-shading or computation of intersection
> with bounding box or similar for n=0 & e >= 1 (notice the bottom of the
> objects on last line)


All superellipsoids are completely inside of a bounding box of
<-1,-1,-1>,<+1,+1,+1>. So I didn't get what you want.

Lars R.

Post a reply to this message