I am trying to get some effective dust in a 20*10*20 (xyz, pov-ray
coordinate system) room, I want to be able to see the light reflecting off
of the dust, especially light which is coming from a stained glass window
(i.e. coloured light).

Any help with this would be greatly appreciated.  Code would be especially
useful.  If any more informaton about the scene is required please ask.
(Also render time isn't critical so I don't mind if the dust is complex to
render - I have an Athlon XP1700 with 512Meg of RAM that can be left running
overnight if nessecairy).

Thanks in advance.  Alastair.

Post a reply to this message

You'd have to use a hollow box container,
but make sure you have no coincident surfaces
with the actual room itself.

Then you'd use
interior{media{scattering{X,Y} intervals A samples 1,B method C}}

where X is the scattering-type (I guess 3 or 4 would do fine),
Y is the rgb-amount of scattering (which is somewhat dependant
on your lightsources, and how thick the dust should be)
Intervals and Samples are typical media-settings, and I'd
say you should use at least 20 intervals with 1 to 20 samples
to begin with. In that case, don't forget to use method 2. Method
2 would be best in any case, since method 3 doesn't work too
good for such situations as the one you are in, and method
1 is too grainy.

Hope that helps a bit.

(PS: Overnight might not suffice, you know...)

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html

Post a reply to this message

Thanks a lot, but what exactly do you mean by:

"but make sure you have no coincident surfaces with the actual room itself."

If you mean nothing should intersect with either the box or the outside line
of the box this could create a problem as there are several objects within
the room.

And thanks again.

Post a reply to this message

I*'ve have tried several permutations of your settings (the renders do not
take long at 320*240) but to no avail.

An example of the settings I've tried is:

box { <-10, 0, -10>, <9,10,15>
           pigment { color rgbf <1,1,1,1> }
           interior { media {   scattering {3, 0.02}
                                         intervals 20
                                         samples 1,10
                                         method 2
                                    }
                         }
           hollow
    }

I can't see any change except for the scene being darker, my light sources
are:

light_source { <0, 10, 0> color White
               spotlight radius 25 falloff 35 tightness 10 point_at <0, 2,
0>
               media_attenuation on
             }
light_source { <1000, 1000, 750> color rgb <1.0, 1.0, 1.0>
                media_attenuation on
               }
light_source { <1.9, 3, 13> color rgb <1.0, 0.9, 0.9>
                fade_distance 2.5 fade_power 0.
                media_interaction off
             }
light_source { <5, 5, 7.99> color rgb <1.0, 0.9, 0.9>
               fade_distance 1.5 fade_power 1
               media_interaction off
             }
light_source { <7.89, 5, -2> color rgb <1.0, 0.9, 0.9>
               fade_distance 2 fade_power
               media_interaction off
             }

if I set the final three lights to media_attenuation on (opposed to
media_interaction off) then these get huges pure white halos that dominate
most of the scene.  For the first two lights do i need to set
media_interaction on or anything?  or is media_attenuation on sufficient?

Post a reply to this message

That might be tricky, because you will end up having to "punch out"
surfaces in your "dust box".  Otherwise, you will get speckle artifacts all
over the place.  Maybe was could say it's just big dust?

Cheers!

Chip Shults
My robotics, space and CGI web page - http://home.cfl.rr.com/aichip

Post a reply to this message

So, first of all, media_attenuation only takes
care that light diminishes when traveling through
media or fog.
Media_interaction actually sets if the light is visible
in the scattering-media.

Another thing to notice is that realistic light-sources
should use a fade_power of 2. To get extra-realism,
use a low fade_distance settings (perhaps around .1)
but higher rgb values, like 7 to 12 or such.

In your case you might try other scattering types.
Type 1 is the most predictable and may already
solve what you want.
Another thing to notice is extinction inside the
scattering statement. It takes care that lighting
of media-particles diminishes the more the light
has to travel through the media. Setting it to
0 might also help to increase the brightness
if the dust.

And for the coincident surfaces, I meant that you
should take care that the walls of the media-container
and the actual walls of the room don't overlap.
When both have the same dimensions, it could
happen that you get some media artefacts on the
walls.


--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html

Post a reply to this message

"Tim Nikias" <tim### [at] gmxde> wrote in message
news:3C80CED1.BA22EC07@gmx.de...
><snip>

> Method
> 2 would be best in any case, since method 3 doesn't work too
> good for such situations as the one you are in, <snip>

Can you explain further? I thought Method 3 was the best (smoothest), and
thus ideal for this.

All the best,

Andy Cocker

Post a reply to this message

In article <3c95cf69@news.povray.org>,
 "Andrew Cocker" <big### [at] mariner9fsnetcouk> wrote:

> Can you explain further? I thought Method 3 was the best (smoothest), and
> thus ideal for this.

Method 3 is basically like method 2, but it will take additional samples 
where needed, so it gives the most accurate results for the amount of 
work. I have no idea why he said it isn't good for this kind of 
situation.

-- 
Christopher James Huff <chr### [at] maccom>
POV-Ray TAG e-mail: chr### [at] tagpovrayorg
TAG web site: http://tag.povray.org/

Post a reply to this message