Here is a very small scene which shows some odd artifacts. There should 
be no coincident surfaces here - or is there? You can see surprisingly 
more artifacts by removing the double slash before the second box. What 
causes these and how can I get rid of them?

You can see the artifacts in the shadow of the superellipsoid.


light_source { <0,2,2> 1}

camera {
     location <0, 0.1, -0.5 >
}

union {
     box {<-0.25,-0.02,-0.25>,<0.25,0.0,0.25>}
     // box {<1,1,1>,<2,2,2>}

     superellipsoid { <0.4,0.4>
                       scale <0.22,0.015,0.22>
                       }
     pigment{ rgb 1}
}

Severi Salminen

Post a reply to this message

On Wed, 10 Dec 2003 13:46:08 +0200, Severi Salminen 
<sev### [at] NOT_THISsibafi> wrote:

> Here is a very small scene which shows some odd artifacts. There should 
> be no coincident surfaces here - or is there? You can see surprisingly 
> more artifacts by removing the double slash before the second box. What 
> causes these and how can I get rid of them?
>
> You can see the artifacts in the shadow of the superellipsoid.
>
>
> light_source { <0,2,2> 1}
>
> camera {
>      location <0, 0.1, -0.5 >
> }
>
> union {
>      box {<-0.25,-0.02,-0.25>,<0.25,0.0,0.25>}
>      // box {<1,1,1>,<2,2,2>}
>
>      superellipsoid { <0.4,0.4>
>                        scale <0.22,0.015,0.22>
>                        }
>      pigment{ rgb 1}
> }
>
> Severi Salminen

Looks like inaccuracy in the root solver, scaling the whole scene by 10 
removes the artifacts.
Translating the superellipsoid slightly in the y-direction solves also the 
problem.
The sturmian root solver could also help, but it can't be used (stated in 
the doc 6.5.1.11  Superquadric Ellipsoid).

-- 
light_source{20*y,1}#macro _(M,X,Y,P)#macro L(N,D)#if(N)#declare 
P=P+D;box{-
0.5,0.5translate z*mod(9*P.gray,4)pigment{rgb P}rotate 45*x+clock*y 
translate
P}L(N-1,D)#end#end#if(M)L(mod(M,8)<mod(X,3)mod(Y,3)1>-1)_(div(M,8)div(X,3)div
(Y,3)P)#end#end _(2301603551,12850,60365,20*z-5*x)plane{y,-9pigment{rgb 1}}

Post a reply to this message

> Looks like inaccuracy in the root solver, scaling the whole scene by 10 
> removes the artifacts.
> Translating the superellipsoid slightly in the y-direction solves also 
> the problem.
> The sturmian root solver could also help, but it can't be used (stated 
> in the doc 6.5.1.11  Superquadric Ellipsoid).

Ok, thanks for your observations. In the meantime I'll just wait a 
128bit version of POV-Ray :)

Severi Salminen

Post a reply to this message

emp### [at] cchutfi news:oprzyrtszv0x9foi@news.povray.org

> The sturmian root solver could also help, but it can't be used (stated
> in the doc 6.5.1.11  Superquadric Ellipsoid).


Btw, I suggest to create a macro that will i.e. generate smooth-mesh 
imitating superquadric ellipsoids.

Why? I think that this object implementation is broken:
1) it renders terrible slow (about 10-20 times slower then i.e. torus) 
especialy when pov must calculate intersections with it
2) it very often had problems/artifacts like mentioned above
3) both things get worse when using supperellipsoid in CSG

-- 
#macro g(U,V)(.4*abs(sin(9*sqrt(pow(x-U,2)+pow(y-V,2))))*pow(1-min(1,(sqrt(
pow(x-U,2)+pow(y-V,2))*.3)),2)+.9)#end#macro p(c)#if(c>1)#local l=mod(c,100
);g(2*div(l,10)-8,2*mod(l,10)-8)*p(div(c,100))#else 1#end#end light_source{
y 2}sphere{z*20 9pigment{function{p(26252423)*p(36455644)*p(66656463)}}}//M

Post a reply to this message

> Btw, I suggest to create a macro that will i.e. generate smooth-mesh 
> imitating superquadric ellipsoids.
> 
> Why? I think that this object implementation is broken:
> 1) it renders terrible slow (about 10-20 times slower then i.e. torus) 
> especialy when pov must calculate intersections with it
> 2) it very often had problems/artifacts like mentioned above
> 3) both things get worse when using supperellipsoid in CSG
> 

Thanks for the suggestion! I'm still learning and don't know anything 
about smooth meshes. But they might indeed work better. And I tried to 
model a pillow using superellipsoids and it didn't work...

Severi Salminen

Post a reply to this message

sev### [at] NOT_THISsibafi news:3fd75455$1@news.povray.org

> about smooth meshes. But they might indeed work better. And I tried to 
> model a pillow using superellipsoids and it didn't work...

I think that there was ready isosurface function generating a pillow, named 
like maybe f_pillow ? try to search in help.

-- 
#macro g(U,V)(.4*abs(sin(9*sqrt(pow(x-U,2)+pow(y-V,2))))*pow(1-min(1,(sqrt(
pow(x-U,2)+pow(y-V,2))*.3)),2)+.9)#end#macro p(c)#if(c>1)#local l=mod(c,100
);g(2*div(l,10)-8,2*mod(l,10)-8)*p(div(c,100))#else 1#end#end light_source{
y 2}sphere{z*20 9pigment{function{p(26252423)*p(36455644)*p(66656463)}}}//M

Post a reply to this message

Rafal 'Raf256' Maj wrote:
> [...]
> 
> Why? I think that this object implementation is broken:
> 1) it renders terrible slow (about 10-20 times slower then i.e. torus) 
> especialy when pov must calculate intersections with it
> 2) it very often had problems/artifacts like mentioned above
> 3) both things get worse when using supperellipsoid in CSG

Please don't pretend there is a bug if you don't have serious arguments. 
    Something isn't broken just because it does not work as you expect 
it. The three points you listed are completely natural if you know how 
superellipsoids work.  If you don't like them - don't use them (and as 
you can read up in the newsgroup archives there are good reasons for 
using an isosurface version instead.

Christoph

-- 
POV-Ray tutorials, include files, Sim-POV,
HCR-Edit and more: http://www.tu-bs.de/~y0013390/
Last updated 25 Oct. 2003 _____./\/^>_*_<^\/\.______

Post a reply to this message

chr### [at] gmxde news:rd2### [at] tritonimagicode

>     Something isn't broken just because it does not work as you expect 
> it. The three points you listed are completely natural if you know how 
> superellipsoids work.  If you don't like them - don't use them (and as 

Yes, this was my point - I realy do not like them :)
I agree there probably isn't a bug.

-- 
#macro g(U,V)(.4*abs(sin(9*sqrt(pow(x-U,2)+pow(y-V,2))))*pow(1-min(1,(sqrt(
pow(x-U,2)+pow(y-V,2))*.3)),2)+.9)#end#macro p(c)#if(c>1)#local l=mod(c,100
);g(2*div(l,10)-8,2*mod(l,10)-8)*p(div(c,100))#else 1#end#end light_source{
y 2}sphere{z*20 9pigment{function{p(26252423)*p(36455644)*p(66656463)}}}//M

Post a reply to this message