|
 |
On 02/01/2013 06:02 PM, Christian Froeschlin wrote:
> I tend to forget how they work together, and then I rediscover that
> the explanation given in the radiosity reference (3.3.4) regarding media
> is rather terse ("Radiosity estimation can be affected by media" for
> the media parameter).
>
> There are different effects possible and it is not quite clear what
> happens here. Can media get illuminated by radiosity? Does media reduce
> the amount of light received by objects in the vicinity of an emitting
> object? Does emissive media contribute light to radiosity calculations?
> (the latter question is answered in the media documentation for
> emission). Which effect is not possible, which is always
> present, which is controlled by the parameter?
>
> Media with photons and radiosity with normals are explained in
> more detail in the tutorial section (2.3.7.7 / 2.3.8.4) that would
> probably be a good place for some detail on radiosity + media
> (with links from the radiosity and media reference?)
I'd agree, the reference passages you pointed out are indeed lacking,
and some of the topics are expanded adequately in the tutorial.
Expanding the missing topic (radiosity & media) seems to be the
challenge ... I'm open to suggestions ;-)
Post a reply to this message
|
|
|
|
"Kenneth" <kdw### [at] gmail com> wrote:
> Christian Froeschlin <chr### [at] chrfrde> wrote:
> > I tend to forget how they work together...
> >
> > There are different effects possible and it is not quite clear what
> > happens here. Can media get illuminated by radiosity? Does media reduce
> > the amount of light received by objects in the vicinity of an emitting
> > object? Does emissive media contribute light to radiosity calculations?
>
> There *was*/is a problem in POV-Ray with rad and media (in the current stable
> v3.62 release and earlier) which sort of contributes to the 'lack of clarity' in
> the docs. For object media (a sphere with a transparent pigment, for example),
> any other objects or parts thereof that are inside the media sphere do not
> receive radiosity lighting *from the media*--neither scattering nor emission
> types--but objects OUTSIDE it do. ('Atmospheric' media doesn't have this
> problem.) But it takes a rather particular/specialized test scene to show the
> effect; it's usually not grossly obvious, due to other scene objects also
> contributing rad light.
It does not work for me. What am I missing? My scene (stripped down to the bare
minimum) is below. The function {} might seem strange - it is generated
externally, designed to cover a certain object with fire (object not included).
I expected the floor to be at least faintly illuminated by the fire, but it
isn't.
My version is RC6 compiled as arch linux package (RC7 is not yet available in
the package form).
#version 3.7;
#declare CAM=<0,-10,1.5>;
camera {
location CAM
look_at <0,0,1.7>
right x
up (image_height/image_width)*z
sky z
}
global_settings {
max_trace_level 10
assumed_gamma 2.2
radiosity {
brightness 1
count 180
pretrace_start 0.08
pretrace_end 0.01
recursion_limit 2
media on
}
}
plane {
z,0
texture {
pigment {
wrinkles
color_map {
[0 color rgb <0.35,0.2,0.1>]
[0.5 color rgb <0.25,0.2,0.2>]
[0.8 color rgb <0.09,0.075,0.075>]
}
turbulence 0.3
scale <10,0.3,1>
}
normal {
bumps 0.1
scale <0.2,0.5,1>
}
finish {
ambient 1
diffuse 1.5
specular 0.3
roughness 0.02
}
rotate 37
}
}
cylinder {
<0,0,0.85>,<0,0,3.2>,1.9
interior {
media {
intervals 2
emission 1
density {
function {
max(0,exp(-3*(z-(1.122814)))-0.5*exp(-6*(z-(1.122814)))
-8*(x-(-0.483531))*(x-(-0.483531))-8*(y-(-1.046504))*(y-(-1.046504)) )
+max(0,exp(-3*(z-(1.126421)))-0.5*exp(-6*(z-(1.126421)))
-8*(x-(-0.367037))*(x-(-0.367037))-8*(y-(-0.759095))*(y-(-0.759095)) )
+max(0,exp(-3*(z-(1.139738)))-0.5*exp(-6*(z-(1.139738)))
-8*(x-(-0.581356))*(x-(-0.581356))-8*(y-(-0.585519))*(y-(-0.585519)) )
+max(0,exp(-3*(z-(1.179992)))-0.5*exp(-6*(z-(1.179992)))
-8*(x-(-0.596242))*(x-(-0.596242))-8*(y-(-0.749582))*(y-(-0.749582)) )
+max(0,exp(-3*(z-(1.213622)))-0.5*exp(-6*(z-(1.213622)))
-8*(x-(-0.358601))*(x-(-0.358601))-8*(y-(-0.733758))*(y-(-0.733758)) )
+max(0,exp(-3*(z-(1.229194)))-0.5*exp(-6*(z-(1.229194)))
-8*(x-(-0.873088))*(x-(-0.873088))-8*(y-(-1.011451))*(y-(-1.011451)) )
+max(0,exp(-3*(z-(1.243289)))-0.5*exp(-6*(z-(1.243289)))
-8*(x-(-0.496820))*(x-(-0.496820))-8*(y-(-0.302525))*(y-(-0.302525)) )
+max(0,exp(-3*(z-(1.244577)))-0.5*exp(-6*(z-(1.244577)))
-8*(x-(-0.989911))*(x-(-0.989911))-8*(y-(-0.673642))*(y-(-0.673642)) )
+max(0,exp(-3*(z-(1.250126)))-0.5*exp(-6*(z-(1.250126)))
-8*(x-(-0.588272))*(x-(-0.588272))-8*(y-(0.369979))*(y-(0.369979)) )
}
color_map {
[0.2 color rgb 0]
[0.5 color rgb <0.3,0,0>]
[0.8 color rgb <1,1,0>]
[0.9 color rgb 2*<1,1,1>]
}
warp {
turbulence 0.44
lambda 2
omega 0.7
octaves 3
}
}
}
}
texture {
pigment { color rgbft 1 }
}
hollow
}
Post a reply to this message
|
|
