|
 |
Hello -
I'm using POVRAY 3.6 and know very little about raytracing.
I'm trying to generate a graphic to illustrate the Riemann surface of
arctan.
I've tried gnuplot and rejected it because I can't figure out to get the
surface transparent in that program.
Instead, I've taken a material definition from Tor Olav Kristensen's
Villarceau Circles that looks pretty much how I'd like and built an
isosurface using it. Kristensen's basic idea was to surround the image
with a very large sphere (blue in his case, white in mine) and use
radiosity to light it.
The problem is that I've getting part of the surface washed out in white,
and I don't know how to fix it (other than by moving the camera to an angle
I don't want).
Rather than attach everything, I'm posting the image and the source here:
http://www.freesoft.org/bbaccala/arctan_surface.html
Can anybody suggest either how to fix the POVRAY file, or perhaps some other
solution entirely to this problem?
Any advice would be greatly appreciated...
Post a reply to this message
|
|
|
|
"Brent Baccala" <cos### [at] freesoft org> wrote:
> Hello -
>
> I'm using POVRAY 3.6 and know very little about raytracing.
>
> I'm trying to generate a graphic to illustrate the Riemann surface of
> arctan.
>
> I've tried gnuplot and rejected it because I can't figure out to get the
> surface transparent in that program.
>
> Instead, I've taken a material definition from Tor Olav Kristensen's
> Villarceau Circles that looks pretty much how I'd like and built an
> isosurface using it. Kristensen's basic idea was to surround the image
> with a very large sphere (blue in his case, white in mine) and use
> radiosity to light it.
>
> The problem is that I've getting part of the surface washed out in white,
> and I don't know how to fix it (other than by moving the camera to an angle
> I don't want).
>
> Rather than attach everything, I'm posting the image and the source here:
>
> http://www.freesoft.org/bbaccala/arctan_surface.html
>
> Can anybody suggest either how to fix the POVRAY file, or perhaps some other
> solution entirely to this problem?
>
> Any advice would be greatly appreciated...
Hello!
Your problem comes from the max_gradient parameter of your isosurfaces. You
did not specify it, so POV-Ray uses the default which is 1.1. This is not
enough. If you have a look at the "Messages" tab of the editor, you will
see a warning for this purpose. Setting max_gradient to 3 seems to be OK.
union {
isosurface {
function { pow((z/10-(atan2(y,x)+pi/2)),2) - .01 }
max_gradient 3
contained_by { box { <0,-200,-100>, <200,200,100> } }
rotate <0, 0, 90>
}
isosurface {
function { pow((z/10-(atan2(y,x)-pi/2)),2) - .01 }
max_gradient 3
contained_by { box { <0,-200,-100>, <200,200,50> } }
rotate <0, 0, -90>
}
isosurface {
function { pow((z/10-(atan2(y,x)-pi*3/2)),2) - .01 }
max_gradient 3
contained_by { box { <0,-200,-100>, <200,200,100> } }
rotate <0, 0, 90>
}
Nota: you can also use the 'evaluate' feature os isosurfaces. See help file.
Regards.
Bruno
PS: Is radiosity necessary here?
Post a reply to this message
|
|
|
|
"Bruno Cabasson" <bru### [at] alcatelaleniaspacefr> wrote:
> Hello!
>
> Your problem comes from the max_gradient parameter of your isosurfaces. You
> did not specify it, so POV-Ray uses the default which is 1.1. This is not
> enough. If you have a look at the "Messages" tab of the editor, you will
> see a warning for this purpose. Setting max_gradient to 3 seems to be OK.
Yes, that fixes it. I had seen the warnings, but since everything looked OK
at the other camera angle, I had just been ignoring them.
> PS: Is radiosity necessary here?
I have no idea. The scene looks pretty much how I like it, though.
Thank you, you were a big help!
Post a reply to this message
|
|
