I'm currently playing around with procedural clouds, but as soon as I
add radiosity to my image, the undersides of the cloud pick up some of
the green of the grassy field below.  Is there an easy way to get rid of
it?

The code is way too messy and embryonic to post.  I'm just looking for
general ideas.

Thanks in advance.
-- 
Francois Labreque | Unfortunately, there's no such thing as a snooze
    flabreque     | button on a cat who wants breakfast.
        @         |      - Unattributed quote from rec.humor.funny
   videotron.ca

Post a reply to this message

Francois Labreque <fla### [at] videotronca> wrote:

> I'm currently playing around with procedural clouds, but as soon as I
> add radiosity to my image, the undersides of the cloud pick up some of
> the green of the grassy field below.  Is there an easy way to get rid of
> it?

> The code is way too messy and embryonic to post.  I'm just looking for
> general ideas.

> Thanks in advance.

In questions of radiosity, I always go back to the source.  Ie, the real
world.  Why aren't clouds green in the real world?

Well, several answers that I can see (some which might be wrong).  First,
the ground isn't green over large sections, and the gathered light blends
out.  Also, the dynamic range of the sunlight is so much greater than the
ground that it just doesn't matter.  Finally, the dispersion is such that
the green doesn't get 'turned around' and come back to us on the ground.

The first is easy to fix.  Maybe you could add some bits off to the side out
of the view that color correct your image.  The second isn't, because pov's
dynamic range is limited, though perhaps a post-process color correction
could be done.

The final one requires tweaking the dispersion parameters (this assumes
you're using media) of the scattering so that the light doesn't come back.

I don't know if any of this will help, but it might give some ideas.

Geoff

Post a reply to this message

Geoff Wedig wrote:
>
> Also, the dynamic range of the sunlight is so much greater than the
> ground that it just doesn't matter.

I did some research on this and these are close to real settings to use
for clear sky color, sun color and grass. Because of the sun intensity
some kind of compression is required but unfortunately it requires
all kinds of workarounds in megapov. Just lower the  multiplier if it's
difficult to fix otherwise.


SkyColor=rgb<0.235,0.378,1>
SunColor=rgb<1,0.836,0.567>*7.6
Grass=texture{
        pigment{rgb<0.2,1,0.2>}
        finish{diffuse 0.1 ambient 0}
      }

_____________
Kari Kivisalo

Post a reply to this message

Kari Kivisalo wrote:
> 
> Geoff Wedig wrote:
> >
> > Also, the dynamic range of the sunlight is so much greater than the
> > ground that it just doesn't matter.
> 
> I did some research on this and these are close to real settings to use
> for clear sky color, sun color and grass. Because of the sun intensity
> some kind of compression is required but unfortunately it requires
> all kinds of workarounds in megapov. Just lower the  multiplier if it's
> difficult to fix otherwise.
> 
> SkyColor=rgb<0.235,0.378,1>
> SunColor=rgb<1,0.836,0.567>*7.6
> Grass=texture{
>         pigment{rgb<0.2,1,0.2>}
>         finish{diffuse 0.1 ambient 0}
>       }

Thanks.  It looks a lot better.

I already know about http://www.cs.utah.edu/~bes/graphics/spectra/  Do
you know of other places where I might find similar stuff?

-- 
Francois Labreque | Unfortunately, there's no such thing as a snooze
    flabreque     | button on a cat who wants breakfast.
        @         |      - Unattributed quote from rec.humor.funny
   videotron.ca

Post a reply to this message

Francois Labreque wrote:
>
> I already know about http://www.cs.utah.edu/~bes/graphics/spectra/  Do
> you know of other places where I might find similar stuff?

After trying to find some more reflectance and rgb color data I came
to the conclusion that using reference photos and a couple of diffuse
values as guide is enough for povray because it is all done in simple
RGB color anyway.

Reference photos: http://www.corbis.com/professional/

Diffuse guide:
specular aluminum 0.95 
new snow          0.74 
concrete          0.40 
vegetation (mean) 0.25 
asphalt           0.07

_____________
Kari Kivisalo

Post a reply to this message