Strange thoughts pop into my head, and was wondering about the phenomenon where
when you get a piece of paper wet or greasy, you can have it look dark or light
depending upon its orientation.
I've also noticed that on engraved reflective surfaces - you can get an
inversion, so to speak.

I looked around a bit, and found some things to read about this:
https://www.reddit.com/r/askscience/comments/2fzrlg/why_does_paper_become_translucent_when_it_gets_wet/

What I was wondering is:
1.  Does this have some sort of Name?   "The XYZ effect"?

2. Is there anything more to this?   I mean that in the sense of - has anyone
studied this and found unanticipated ways to control and/or tweak this in
interesting and unexpected ways?

3.  Has anyone to date modeled this with POV-Ray, perhaps with an animation?


Thanks!

Post a reply to this message

Am 29.02.2016 um 02:33 schrieb Bald Eagle:
> Strange thoughts pop into my head, and was wondering about the phenomenon where
> when you get a piece of paper wet or greasy, you can have it look dark or light
> depending upon its orientation.

That's because when wet or greasy, the paper's diffuse reflectivity is
reduced, but the specular reflectivity is increased. Specular
reflectivity is highly angle-dependent -- even if you make sure that you
look at the surface from the direction "opposite" to the incoming light,
specular reflection is strongest at shallow angles. (See "Fresnel
equations".)

> I've also noticed that on engraved reflective surfaces - you can get an
> inversion, so to speak.

This is also related to the angle-dependent nature of specular reflections.

> 3.  Has anyone to date modeled this with POV-Ray, perhaps with an animation?

There are a few things at work here:

- Translucency: Wet paper is translucent (dry paper is, too, but to a
much lesser degree). You can model this either by using a solid shape
(e.g. a thin box) with media or SSLT, or by using a non-solid shape
(e.g. a pair of triangles) with POV-Ray 3.7's backside illumination
feature (specify a second parameter to "diffuse" to specify how much a
light behind the surface should illuminate the front side).

- Angle-dependent specular reflection: If the object is supposed to be
made of metal, use "metallic" in both the reflection block and in the
finish block itself. If the object is supposed to be made of anything
else, specify an interior and ior, use "fresnel" in the reflection
block, and if you use POV-Ray 3.7.1 also use "fresnel" in the finish
block itself.

- Angle-dependent diffuse reflection: Yes, this exists, and it behaves
just the opposite of the specular reflection. Think of it as specular
reflection "stealing" light from the diffuse reflection. To model this,
you need to use either an angle-of-incidence ("aoi") patterned texture,
or POV-Ray 3.7.1 and "fresnel" in the finish block itself.

- Etchings: I guess surface normal perturbations should do the trick.

- Rough specular reflections: You may want the specular reflections to
be somewhat blurry, especially in case of paper. You can achieve this by
using very small-scale random surface normals (e.g. bumps) and some
means of oversampling (e.g. using many averaged textures with different
surface normals, a slight bit of focal blur, or high-quality anti-aliasing).

Post a reply to this message

On 29-2-2016 5:56, clipka wrote:
> Am 29.02.2016 um 02:33 schrieb Bald Eagle:
>> 3.  Has anyone to date modeled this with POV-Ray, perhaps with an animation?
>
[...]

> - Angle-dependent specular reflection: If the object is supposed to be
> made of metal, use "metallic" in both the reflection block and in the
> finish block itself. If the object is supposed to be made of anything
> else, specify an interior and ior, use "fresnel" in the reflection
> block, and if you use POV-Ray 3.7.1 also use "fresnel" in the finish
> block itself.
>
> - Angle-dependent diffuse reflection: Yes, this exists, and it behaves
> just the opposite of the specular reflection. Think of it as specular
> reflection "stealing" light from the diffuse reflection. To model this,
> you need to use either an angle-of-incidence ("aoi") patterned texture,
> or POV-Ray 3.7.1 and "fresnel" in the finish block itself.
>

This is the principle of 'damask' metal or cloth. For the latter, I have 
used one image_map (p_map17: texture of the cloth) and another one 
(p_map18: black/white copy), and the following code (directly taken from 
a Poser/Poseray model):

#declare Damask_Red_1_=
#declare F1=finish{specular 0.2431373
                    roughness 0.00286312
                    phong 0 phong_size 0
                    diffuse 0.6
                    reflection{0 } conserve_energy
                    }
#declare F2=finish{specular 0
                    roughness 0.00286312
                    phong 0 phong_size 0
                    ambient rgb <0.00,0.00,0.00>
                    diffuse 0.6
                    reflection{0 } conserve_energy
                    }
material{
          texture{uv_mapping
                  pigment_pattern{p_map18 }
                  texture_map{
                      [0
                         pigment{p_map17 }
                         finish{F2}
                      ]
                      [1
                         pigment{p_map17 }
                         finish{F1}
                      ]
                      }}
}

The same can be done I guess to simulate Damascene swords of course.

-- 
Thomas

Post a reply to this message

clipka <ano### [at] anonymousorg> wrote:

[ a lot of stuff! ]

Thanks for the prompt and VERY thorough reply.

IRL, specular reflection yields linearly polarized light, correct?
So a polarizing film ought to squelch the tone-reversal effect to some extent...

Post a reply to this message

> clipka <ano### [at] anonymousorg> wrote:
>
> [ a lot of stuff! ]
>
> Thanks for the prompt and VERY thorough reply.
>
> IRL, specular reflection yields linearly polarized light, correct?
> So a polarizing film ought to squelch the tone-reversal effect to some extent...
>
>
>
>
>
>
At shallow angles, specular reflection is polarised, but that 
polarisation dimishes to none as you approach perpendicularity.
Presently, POV-Ray don't compute polarisation in any way. It also don't 
support birefringeance nor light interference like you see on a CD or 
vinil record.


Alain

Post a reply to this message