I'm getting several odd glitches in a couple of scenes which use paralell
cylindrical lights to create lasers. Both scenes use photon mapping to make
the beams reflect off mirrors.

The first scene has two lasers (blue and green) reflecting off mirrors which
then hit a flat white test surface. The weird thing is that instead of
getting a green and a blue spot, I get two large cyan ones. I tried
commenting out the mirror in front of the blue laser, and I got the
expected blue spot, but the green one still produced the large cyan spot.
Even weirder, I seem to have two extra cyan beams which go and hit the
walls of the room and come at right angles to the direction of the original
beams.

The second scene is a sort of hologram setup. The first problem here is an
extra beam appearing near the beam splitter, which dissappears if I comment
out the mirror into the scene from it. All the mirrors are angles at 45
degrees to the incoming beams, so I can't understand why this beam is
appearing. The second probem is that the reflected beams in the distance
don't show up in the media which is used to make the beams visible. They
still produce the expected spot of light on the test surface though.

Finally, one glitch which is common to both scenes is a myterious black disc
which appears at the location of the light source. The weirdest thing about
this disc, is that it can be seen through solid objects. I think this also
occurs with regular cylindrical lights, could this be a bug in cylindrical
lights?

Anyway, thanks for taking the time to read such a long post, and hopefully
someone can help solve this.

Rohan _e_ii

Post a reply to this message

I think I have experienced a problem similar to this myself, I was pointing
3 cylinder lights (red,green and blue; in parallel) at a reflective box.
And catching their reflected light on another box.  Instead on 3 spots
(r,g,b) of light, I got 1 spot of white light, that appeared to be 'chopped
off'

If I replaced the cylinder lights with spotlights I got the expected
results.

I posted here, a while back about this, asking whether it was a known bug.
Someone sorta verified this, so I just assumed it a bug and left it at that.

As for the other problems you are experiencing, I can't help.


Rohan Bernett wrote:
>The first scene has two lasers (blue and green) reflecting off mirrors which
>then hit a flat white test surface. The weird thing is that instead of
>getting a green and a blue spot, I get two large cyan ones. I tried
>commenting out the mirror in front of the blue laser, and I got the
>expected blue spot, but the green one still produced the large cyan spot.
>Even weirder, I seem to have two extra cyan beams which go and hit the
>walls of the room and come at right angles to the direction of the original
>beams.

Post a reply to this message

It's nice to see that I'm not the only one with this problem. Could somebody
out there tell me whether this glitch can be fixed in the scene, or is it a
bug in the lighting and photons in POV-Ray?

Rohan _e_ii

Post a reply to this message

Rohan Bernett wrote:
> 
> It's nice to see that I'm not the only one with this problem. Could somebody
> out there tell me whether this glitch can be fixed in the scene, or is it a
> bug in the lighting and photons in POV-Ray?

A short coded example of the problem would really help.

-- 
Ken Tyler

Post a reply to this message

Here's what I think must be Bruce's, changed a little.

// begin scene code

#declare Spot=no; // yes for spotlights, no for cylinder lights

#version 3.5;

global_settings
{
photons
        {
        spacing 0.01
        }
}

// reflected lights

// when using spots, the results are what I expect,
//    3 distinct lit areas on the left
// when using cylinder, photons from all three
//     lights accumulate at one area

#macro spot_or_cylinder()
 #if (Spot=yes)
 spotlight radius 1 falloff 1.5 tightness 10 // parallel
 #else
 cylinder radius 1 falloff 1.5 tightness 10 // parallel
 #end
#end

light_source
{
0,
color blue 3
spot_or_cylinder()
point_at z
photons { reflection on refraction on }
translate <0.5, 0.25, -3>
}

light_source
{
0,
color green 3
spot_or_cylinder()
point_at z
photons { reflection on refraction on }
translate <0.5, 0.5, -3>
}

light_source
{
0,
color red 3
spot_or_cylinder()
point_at z
photons { reflection on refraction on }
translate <0.5, 0.75, -3>
}


// reflecting box
box
{
0, 1
texture
{
pigment { color rgb .5 }
finish { reflection {.6,.9} }
}
rotate y*45

photons
{
target
reflection on
refraction off
collect off
}
}

// refracting box
box
{
0, 1
texture
{
pigment { color rgbf .999 }
finish { diffuse 0 }
}
interior {
        ior 1.5
        }
rotate y*45
translate -.1

photons
{
target
reflection on
refraction on
collect off
}
}

// plain box
box
{
0, 1.5
texture
{
pigment { color rgb 1 }
}
rotate y*225

photons
{
        target
reflection off
refraction off
collect on
}
}

camera { location <0,1,-3> look_at <0,0,0> }

light_source
{
<-3,4,-5> color rgb 1
photons { reflection off refraction off }
}

plane { y,0 texture { pigment { checker rgb 0, rgb 1 } } }

// end scene code

Post a reply to this message

Thats the one. To me it looks like a bug, no idea what it may be
though.

I just recently downloaded the pov source to take a look at, but its
been ages since I've done any C coding... :(



On Mon, 9 Sep 2002 23:41:27 -0500, "hughes, b."
<omn### [at] charternet> wrote:

>Here's what I think must be Bruce's, changed a little.

Post a reply to this message

Ken wrote:
>
>Rohan Bernett wrote:
>>
>> It's nice to see that I'm not the only one with this problem. Could
>> somebody out there tell me whether this glitch can be fixed in the
>> scene, or is it a bug in the lighting and photons in POV-Ray?
>
>A short coded example of the problem would really help.
>
>Ken Tyler
>
I discovered a problem with porting a scene from spotlights to cylinder
lights. What I did was create a screen with a mirror adjacent and
perpendicular to it and aimed two lights at it. One light went direct
and the other reflected off the mirror. It worked exactly as expected
with spotlights, but when I changed the type to cylinder for both
lights the two rays mixed at the center of the screen. (The entire
scene is centered about the origin.) It appears as though the "y" value
of the "point_at" vector is not being handled correctly by POV. I wrote
a simple scene to show this effect and parameterized the relevant
variables. The source code is fully commented and I will send it to
anyone who requests it. The size is 6.24 KB uncompressed, but much of
that is comments.

From my short experiments, I do believe it is a bug in POV-Ray, but
there may be a workaround. I found that sometimes multiplying the "y"
value of the "point_at" declaration by two gave the desired results,
but this did not produce consistent results.

Bob Sorenson

Post a reply to this message

On Sat, 14 Sep 2002 18:07:28 EDT Bob Sorenson wrote:

>The source code is fully commented and I will send it to
>anyone who requests it. The size is 6.24 KB uncompressed, but much of
>that is comments.

  Can the scene be reduced to the basic components which will duplicate
the problem you describe? If so, you can post it here in p.general.
Please read "bug reporting" in the
povray.announce.frequently-asked-questions group, thanks.

-- 
Alan
ako### [at] povrayorg
a k o n g <at> p o v r a y <dot> o r g

Post a reply to this message

Alan Kong wrote:
>On Sat, 14 Sep 2002 18:07:28 EDT Bob Sorenson wrote:
>
>>The source code is fully commented and I will send it to
>>anyone who requests it. The size is 6.24 KB uncompressed, but much of
>>that is comments.
>
>  Can the scene be reduced to the basic components which will duplicate
>the problem you describe? If so, you can post it here in p.general.
>Please read "bug reporting" in the
>povray.announce.frequently-asked-questions group, thanks.
>
>Alan
>ako### [at] povrayorg
>a k o n g <at> p o v r a y <dot> o r g
>
The source code was written to demonstrate the possible bug, is well
commented and easily changed to explore it. I just posted it on the
regular news server in the "windows" area and will email it to anyone
who wants it.

Bob Sorenson

Post a reply to this message

Bob Sorenson wrote:
>The source code was written to demonstrate the possible bug, is well
>commented and easily changed to explore it. I just posted it on the
>regular news server in the "windows" area and will email it to anyone
>who wants it.
>
At the request of someone who appeared to be a moderator of the
povray.windows newsgroup, I canceled the posting.

Post a reply to this message