W2000, p4 1GHz, 256 Mb
WinPOV 3.5 RC 5

While trying to find a workaround for reflected shadow problem, I came
across this anomaly with light groups, AFAICT it should not be expected to
function as the appended scene.

light group 1 has a light and box with the box having no image, i.e only
seen in the mirror
light group 2 has a similar light and box  with the light being shadowless
and the box having no reflection, i.e. they are only seen in real, not the
mirror.  when the sphere is added to this light group, it casts no shadow on
the box as expected, but in the mirror it casts a shadow on the reflected
box which, i think, it shouldn't. Am I missing something here?

-tgq


//start
camera{
  up y
  right x*image_width/image_height
  angle 45
  location <-1000,1000,-1000>
  look_at  0
}

background {rgb <0,0,1>}

#macro L1  ()
  0
  rgb 1
  translate <-300,1000,1000>
#end

box{<0,0,-800><10,400,800>
  pigment{rgb 1}
  finish{ reflection{1 0}diffuse 0}
}

light_group {
  light_source{L1()}
  box{<-1000,-10,-1000><1000,0,1000> pigment{rgb 1} no_image}
}

light_group {
  light_source{L1() shadowless}
  box{<-1000,-10,-1000><1000,0,1000> pigment{rgb 1} no_reflection}
  sphere{y*100 100  pigment{rgb <1,0,0>} translate -x*400 no_reflection}
}
//end

Post a reply to this message

"TinCanMan" <Tin### [at] hotmailcom> wrote in message
news:3cf22e8a$1@news.povray.org...
>
> light group 1 has a light and box with the box having no image, i.e only
> seen in the mirror
> light group 2 has a similar light and box  with the light being shadowless
> and the box having no reflection, i.e. they are only seen in real, not the
> mirror.  when the sphere is added to this light group, it casts no shadow
on
> the box as expected, but in the mirror it casts a shadow on the reflected
> box which, i think, it shouldn't. Am I missing something here?

Section 6.5.9.7 states the following:
___________________________
When no_image is used, the object will not be seen by the camera, either
directly or through transparent/refractive objects. However, it will still
cast shadows, and show up in reflections (unless no_reflection and/or
no_shadow is used also).
When no_reflection is used, the object will not show up in reflections. It
will be seen by the camera (and through transparent/refractive objects) and
cast shadows, unless no_image and/or no_shadow is used.
_________________________

I was confused by this. I had used 3 light groups, placing the floor and
mirror apart too. Thinking maybe global_lights could help somehow but it did
not. What's apparently being described is that a shadow cannot be separated
except via no_shadow. Just one of those things you shouldn't read fast and
instead think over a few moments. Simply put, it's saying no_reflection has
no bearing over shadows. It's merely a "object modifier" alone, each of
those three keywords independantly acting upon things.

Yet, I too, was surprised light_groups wasn't able to discern them
separately.

bob h

Post a reply to this message

> Section 6.5.9.7 states the following:
> ___________________________
> When no_image is used, the object will not be seen by the camera, either
> directly or through transparent/refractive objects. However, it will still
> cast shadows, and show up in reflections (unless no_reflection and/or
> no_shadow is used also).
> When no_reflection is used, the object will not show up in reflections. It
> will be seen by the camera (and through transparent/refractive objects)
and
> cast shadows, unless no_image and/or no_shadow is used.
> _________________________
>
> I was confused by this. I had used 3 light groups, placing the floor and
> mirror apart too. Thinking maybe global_lights could help somehow but it
did
> not. What's apparently being described is that a shadow cannot be
separated
> except via no_shadow. Just one of those things you shouldn't read fast and
> instead think over a few moments. Simply put, it's saying no_reflection
has
> no bearing over shadows. It's merely a "object modifier" alone, each of
> those three keywords independantly acting upon things.
>

I see, but the problem I have here is that the no_reflection sphere and the
reflected box are not in the same light group at all.  If you remove the
reflected-only box and remove no_reflection from the box in the same group
as the sphere, then there is no shadow in the reflection.
In 6.5.8 it states "The light_sources in the group can only illuminate the
objects that are in the group." so I don't see how an object in one
light_group can cast a shadow from a light in a different light_group.

-tgq

Post a reply to this message

In article <3cf3707a$1@news.povray.org> , "TinCanMan" 
<Tin### [at] hotmailcom> wrote:

> In 6.5.8 it states "The light_sources in the group can only illuminate the
> objects that are in the group." so I don't see how an object in one
> light_group can cast a shadow from a light in a different light_group.

This has been explained before.  See the thread "light_group question" in
povray.general sometime in late April this year.

    Thorsten

____________________________________________________
Thorsten Froehlich
e-mail: mac### [at] povrayorg

I am a member of the POV-Ray Team.
Visit POV-Ray on the web: http://mac.povray.org

Post a reply to this message

>From: "TinCanMan" <Tin### [at] hotmailcom>
>Newsgroups: povray.beta-test
>References: <3cf22e8a$1@news.povray.org> <3cf34c16@news.povray.org>
>Subject: Re: Global lights, shadowless, etc.
>Date: Thu, 30 May 2002 07:56:31 -0400

Please correct your computer date, it is a few days ahead of the real world
:-)

    Thorsten

Post a reply to this message

> >Date: Thu, 30 May 2002 07:56:31 -0400
>
> Please correct your computer date, it is a few days ahead of the real
world
> :-)
>
>     Thorsten

Hmm don't know how that happened.
Anyone want to know the lotto numbers? ;)

-tgq

Post a reply to this message

"TinCanMan" <Tin### [at] hotmailcom> wrote in message
news:3cf38e7c@news.povray.org...

> Hmm don't know how that happened.
> Anyone want to know the lotto numbers? ;)

... hope they're cheaper than the last guy I bought them off.

Post a reply to this message