The code below shows 2 potential bugs affecting transparency.
These have been observed on 3.1 (using material_map) and 3.5, at least on a
Windows version, running XP.
The first bug makes the transparent part visible. It's apparently linked
with fog turbulence.
The second bug makes the normals behind look weird when the transparency
texture is scaled very unevenly (pure 100% transparent texture is not
affected).

Can other people confirm that these problems exist and that they are really
bugs (they were reported by a Moray/POV-Ray 3.1 user)?


camera {location  -z*35}
light_source {(x+y-z)*50 rgb 1}

fog {
  fog_type   1
  distance   30
  color      rgb 1
// turbulence makes the transparent part visible
  turbulence 0.8
}
sphere{0,8 pigment{rgb x+y} normal{bumps 1}}
union{
    triangle{-x,-x+y,x+y}
    triangle{-x,x,x+y}
    hollow
    texture{
        pigment_pattern{
            bozo color_map{[0.4 rgb 1][0.4 rgb 0]}
        }
        texture_map{
            [0.5 pigment{rgb x} finish{ambient 0 diffuse 1}]
            [0.5 pigment{rgbt 1} finish{ambient 0 diffuse 0}]
        }
// uneven scaling makes the normals behind look weird
        scale <1,1,0.001>
    }
    scale 2
    translate -z*20

}


--
**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters

Post a reply to this message

I can confirm both bugs. Windows XP &  pov3.5

Mick

"Gilles Tran" <git### [at] wanadoofr> wrote in message
news:3d8d447b@news.povray.org...
> The code below shows 2 potential bugs affecting transparency.
> These have been observed on 3.1 (using material_map) and 3.5, at least on
a
> Windows version, running XP.
> The first bug makes the transparent part visible. It's apparently linked
> with fog turbulence.
> The second bug makes the normals behind look weird when the transparency
> texture is scaled very unevenly (pure 100% transparent texture is not
> affected).
>
> Can other people confirm that these problems exist and that they are
really
> bugs (they were reported by a Moray/POV-Ray 3.1 user)?
>
>
> camera {location  -z*35}
> light_source {(x+y-z)*50 rgb 1}
>
> fog {
>   fog_type   1
>   distance   30
>   color      rgb 1
> // turbulence makes the transparent part visible
>   turbulence 0.8
> }
> sphere{0,8 pigment{rgb x+y} normal{bumps 1}}
> union{
>     triangle{-x,-x+y,x+y}
>     triangle{-x,x,x+y}
>     hollow
>     texture{
>         pigment_pattern{
>             bozo color_map{[0.4 rgb 1][0.4 rgb 0]}
>         }
>         texture_map{
>             [0.5 pigment{rgb x} finish{ambient 0 diffuse 1}]
>             [0.5 pigment{rgbt 1} finish{ambient 0 diffuse 0}]
>         }
> // uneven scaling makes the normals behind look weird
>         scale <1,1,0.001>
>     }
>     scale 2
>     translate -z*20
>
> }
>
>
> --
> **********************
> http://www.oyonale.com
> **********************
> - Graphic experiments
> - POV-Ray and Poser computer images
> - Posters
>
>
>
>
>

Post a reply to this message

Gilles Tran wrote:

> The first bug makes the transparent part visible. It's apparently linked
> with fog turbulence.

This has been a known limitation of fog and transparent objects for many years.
It came up several times during the beta of 3.5 and I seem to recall the
general consensus and official response was that media should be used instead.


-- 
Ken Tyler

Post a reply to this message

3D8D9CEB.D3830FD5@pacbell.net...
> This has been a known limitation of fog and transparent objects for many
years.
> It came up several times during the beta of 3.5 and I seem to recall the
> general consensus and official response was that media should be used
instead.

Well thanks, the first bug rang a (dim) bell.
Does the second one deserve a formal bug report ?
It's less a problem than the first one but it's really weird anyway. I
scratched my head for hours figuring it out (the initial problem was
submitted by an user, not by me, so I have to tell him something).

G.


--
**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters

Post a reply to this message

In article <3d8d447b@news.povray.org>,
 "Gilles Tran" <git### [at] wanadoofr> wrote:

> The code below shows 2 potential bugs affecting transparency.
> These have been observed on 3.1 (using material_map) and 3.5, at least on a
> Windows version, running XP.
> The first bug makes the transparent part visible. It's apparently linked
> with fog turbulence.

It isn't a bug, just a limitation of fog. Fog is really a cheat, it uses 
the distance along the ray to figure out how much fog it goes through 
assuming constant or gradient fog, and then uses turbulence with the 
intersection point scaled by the turbulence amount and some other tricks 
to mess with the real fog value. The result is nothing like real fog, 
but usually looks good and is fast to compute.
"Fixing" this would require making fog into a more limited media, if you 
need accurate results in this kind of situation just use scattering 
media.


> The second bug makes the normals behind look weird when the transparency
> texture is scaled very unevenly (pure 100% transparent texture is not
> affected).

I haven't had a chance to render this, but it seems very unlikely that 
there is some problem with scaling of normals, or even scaling of normal 
patterns. Maybe there is a problem with the ray directions not being 
normalized for some reason...

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: chr### [at] tagpovrayorg
http://tag.povray.org/

Post a reply to this message

Christopher James Huff <chr### [at] maccom> wrote:
> It isn't a bug, just a limitation of fog. Fog is really a cheat

  This could be a good proverb for CG: "Don't expect a trick to work like
the real thing." :)

-- 
#macro N(D)#if(D>99)cylinder{M()#local D=div(D,104);M().5,2pigment{rgb M()}}
N(D)#end#end#macro M()<mod(D,13)-6mod(div(D,13)8)-3,10>#end blob{
N(11117333955)N(4254934330)N(3900569407)N(7382340)N(3358)N(970)}//  - Warp -

Post a reply to this message