Hello!

Can't implement a parabolic mirror.
The scene is as follows: a parallel stream of light, a parabolic mirror (white,
in the example it is blue-red), in the focus of which is a white plate. The
camera shoots this plate from the back. I expect that the mirror will reflect
the light and the plate will be illuminated from back, but it is fully dark, and
this does not correspond to physics .. What to do?

"cmd": ["povray","$file", "+D", "+P", "+H400", "+W400", "+Fng", "+A", "+AM2",
"+Q11"],

file.pov (after test, place camera at <0,0,0>..):

#version 3.7;
global_settings{assumed_gamma 1.0}
//global_settings {ambient_light 0*<1,1,1>}
global_settings { max_trace_level 5}
#include "colors.inc"
#include "consts.inc"
#declare T=0.1;

camera {
 //orthographic
 location 2*z+2*x // must look from 'o'
 right 5*T*x
 up 5*T*y
 sky y
 look_at 0.5*z
}

// perfect parabolic mirror
parametric {
 function { u },
 function { v },
 function { u*u+v*v }
 <-1,-1>,<1,1>
 contained_by {box {<-1,-1,0>, <1,1,0.5>}}
 pigment {
  function {z}
  color_map {
   [0 color Blue]
   [1 color Red]
  }
 }
 finish{ ambient 0
   diffuse 0.5  // phong 0.5
   reflection 1
 }
}

// target
polygon { 5, <-T/2,-T/2>, <-T/2,T/2>, <T/2,T/2>, <T/2,-T/2>, <-T/2,-T/2>
  pigment{color White}
  finish { ambient 0 diffuse 1 reflection 0}
  translate 0.5*z
}

light_source {5*z color 1*White parallel point_at o} // looks_like{sphere {o,2
pigment{color Orange}}}

Post a reply to this message

Le 15/08/2020 à 13:59, ttyUSB0 a écrit :
> Hello!
> 
> Can't implement a parabolic mirror.
> The scene is as follows: a parallel stream of light, a parabolic mirror (white,
> in the example it is blue-red), in the focus of which is a white plate. The
> camera shoots this plate from the back. I expect that the mirror will reflect
> the light and the plate will be illuminated from back, but it is fully dark, and
> this does not correspond to physics .. What to do?
> 
> "cmd": ["povray","$file", "+D", "+P", "+H400", "+W400", "+Fng", "+A", "+AM2",
> "+Q11"],
> 

you need photons !

see scenes/advanced/optics.pov

Post a reply to this message

Le_Forgeron <jgr### [at] freefr> wrote:

> you need photons !

Yes, apparently one does.

I added
 photons{
  target
  reflection on
  refraction off
  }

after the finish block, and

 photons {
  count 20000
  autostop 0
  jitter 0.4
 }

to the global_settings block, and I got a faint square.
bumping up the light_source to rgb 10 made it more visible, so presumably using
a higher count or spacing will yield a similar effect, though at a slower render
speed.

Post a reply to this message

Thanks to everyone!
I read the Photons section and understood where to dig.
Why is the light reflected so weakly? The count parameter does not affect the
illumination.

Post a reply to this message

"ttyUSB0" <nomail@nomail> wrote:
> Thanks to everyone!
> I read the Photons section and understood where to dig.
> Why is the light reflected so weakly? The count parameter does not affect the
> illumination.

That's a good question, but not really ever using photons except in the odd
scene here and there, I really couldn't tell you.

What I did before to get the polygon to show up (I used a box, to view it from
the side when I was understanding the overall scene geometry) was crank the
light_source up to intensity 10.

If you keep the light_source at 1, you can change the finish block of the
polygon to something like diffuse 100 and that will brighten it up. If you layer
a box and a polygon or triangles, you can make the side facing the light_source
have a "normal" diffuse value, and then only give the mirror-facing side the
huge diffuse value.

You might also experiment with texture and interior_texture to do the same thing
with a single 2D object.

This guy seems to be an ace - maybe contact him with a question.   I'm sure
you'll both have a lot in common!   :)

https://www.youtube.com/watch?v=xjMjWtntm9k

Post a reply to this message

Le 2020-08-16 à 06:24, ttyUSB0 a écrit :
> Thanks to everyone!
> I read the Photons section and understood where to dig.
> Why is the light reflected so weakly? The count parameter does not affect the
> illumination.
> 
> 
The count or spacing parameter have no effect on the intensity of the 
caustic, only the intensity of the light source.
High count, or small spacing, will result in finer smoother results. The 
reason is simple : The amount of light, or it's intensity, reaching the 
surface is evenly distributed among all the photons shot toward that 
surface.

Next contributing factor : The effective reflectivity of the surface. As 
your reflection value is 1, you can't get more without breaking physics. 
In fact, a value of 1 is not quite realistic as there are no reflector 
capable of reflecting 100% of the incident light.

Finally is the position of your target screen relative to the focal 
point. At the focal point you should have a very small, but very bright, 
spot. Moving in or out, the illuminated spot get broader, but also dimmer.

Post a reply to this message

On 8/16/20 6:24 AM, ttyUSB0 wrote:
> Thanks to everyone!
> I read the Photons section and understood where to dig.
> Why is the light reflected so weakly? The count parameter does not affect the
> illumination.
> 
Seems like you are making progress, but your post reminded me of the 
following thread:

http://news.povray.org/povray.binaries.images/thread/%3C428a5544%40news.povray.org%3E/

which might be a useful reference. Warp posted source code for the 
images. It was probably written for 3.6, but IIRC v3.7 without a version 
keywords defaults to v3.6 behavior.

Bill P.

Post a reply to this message