http://www.winosi.onlinehome.de


Tim Cook wrote:

> Hmm just getting to an internet connection where I can read the
> POV NGs after a few days...

> on the way up to Illinois for vacation, the reference to
> the experiment used to test whether light is particles or
> waves struck me as being interesting.
> 
> Question: Are POV photons waves, or particles? (or neither) ;)
> 
> As I don't currently have access to POV 3.5 b, I must ask
> y'all to try this for me.
> 
> Construct a photon-using scene as follows:
> 
> area   wall w/1 slit in centre
> light  |        |         |
> *      |        |         |  white screen
>        |        |         |
>                 wall w/2 slits dividing wall into 3 equal parts
> 
> I predict the following possible outcomes of this render:
> 1) the white screen will not be illuminated at all
> 2) the white screen will have an interference pattern on it
> 3) the white screen will have two lines with fuzzy edges
> 4) the white screen will have two lines with sharp edges
> 
> Another question: is polarisation possible in POV?
> 
> - Tim Cook
> http://empyrean.scifi-fantasy.com

Post a reply to this message

On Sat, 12 Jan 2002 16:17:00 -0500, Christopher James Huff wrote:
> POV-Ray is not designed for physically correct simulation of optical 
> effects, it took a long time just for dispersion and photon mapping to 
> be added. Effects like diffraction are often difficult to implement, 
> very slow to compute, and would contribute little or nothing to the 
> majority of scenes.

Polarization wouldn't be much harder to deal with than dispersion, though
birefringence might make it a little more interesting.

Note, too, that dispersion isn't perfect either.  For one thing, it 
assumes that the function relating IOR to wavelength is very simple
for the visible spectrum.  For another, since it's being forced to 
work with tristimulus values (RGB) instead of spectral power distributions,
it has to infer some sort of tristimulus->SPD conversion where none 
exists.  If we provided a way to specify colors as SPDs this wouldn't 
be such an issue, of course, but we haven't done that.  (For more on
tristimulus values, SPDs, and the pitfalls of conversion, check out 
the Color FAQ at http://www.inforamp.net/~poynton/PDFs/ColorFAQ.pdf )

Daren Wilson's original dispersion patch had some support for something
approaching SPDs in the form of 12-component colors, but that support
didn't make it into 3.5.  On the other hand, his RGB->SPD conversion was
much less realistic than the one I threw together for 3.5, so hopefully
users will feel that it's balanced out at least a little.

--
#macro R(L P)sphere{L __}cylinder{L P __}#end#macro P(_1)union{R(z+_ z)R(-z _-z)
R(_-z*3_+z)torus{1__ clipped_by{plane{_ 0}}}translate z+_1}#end#macro S(_)9-(_1-
_)*(_1-_)#end#macro Z(_1 _ __)union{P(_)P(-_)R(y-z-1_)translate.1*_1-y*8pigment{
rgb<S(7)S(5)S(3)>}}#if(_1)Z(_1-__,_,__)#end#end Z(10x*-2,.2)camera{rotate x*90}

Post a reply to this message

In article <slr### [at] fwicom>,
 Ron Parker <ron### [at] povrayorg> wrote:

> Polarization wouldn't be much harder to deal with than dispersion, though
> birefringence might make it a little more interesting.

Wouldn't you have to keep track of the amount of light polarized along 
each plane? Probably with some kind of spline relating intensity of 
light to angle of polarization? A lot more data for each ray...

-- 
 -- 
Christopher James Huff <chr### [at] maccom>

Post a reply to this message

On Mon, 14 Jan 2002 13:41:47 -0500, Christopher James Huff wrote:
> In article <slr### [at] fwicom>,
>  Ron Parker <ron### [at] povrayorg> wrote:
> 
>> Polarization wouldn't be much harder to deal with than dispersion, though
>> birefringence might make it a little more interesting.
> 
> Wouldn't you have to keep track of the amount of light polarized along 
> each plane? Probably with some kind of spline relating intensity of 
> light to angle of polarization? A lot more data for each ray...

Compared to what we keep track of for dispersion?  Not really.

-- 
#macro R(L P)sphere{L F}cylinder{L P F}#end#macro P(V)merge{R(z+a z)R(-z a-z)R(a
-z-z-z a+z)torus{1F clipped_by{plane{a 0}}}translate V}#end#macro Z(a F T)merge{
P(z+a)P(z-a)R(-z-z-x a)pigment{rgbf 1}hollow interior{media{emission 3-T}}}#end 
Z(-x-x.2x)camera{location z*-10rotate x*90normal{bumps.02scale.05}}

Post a reply to this message

Ron Parker wrote:
> Christopher James Huff wrote:
> > Wouldn't you have to keep track of the amount of light polarized
> > along each plane? Probably with some kind of spline relating
> > intensity of light to angle of polarization? A lot more data
> > for each ray...
> Compared to what we keep track of for dispersion?  Not really.

There's only 3 more values per ray for a more realistic simulation
of light; amplidude/frequency/rotation around axis of travel...?
Could probably do some interesting things with having materials
change the frequency of the light...

Speaking of something completely different...iridescence.  It's
an interference pattern between two surfaces of oil or similar,
how 'bout trying to do that the hard way in POV?  Should be
doable even in 3.1g...I think I'll play around with that and
such when I get back to my computer.

Post a reply to this message

In article <3C49E792.62CAB2DC@scifi-fantasy.com>,
 Tim Cook <tim### [at] scifi-fantasycom> wrote:

> There's only 3 more values per ray for a more realistic simulation
> of light; amplidude/frequency/rotation around axis of travel...?

I think you would have to store more information, a bunch of "samples" 
around the axis of travel, interpolated with a spline if the 
calculations need it. Ordinary light isn't polarized in a single 
direction, it is a mix of photons polarized in all directions.
Frequency has more to do with color, and is a different issue. POV could 
assume all photons are in the same frequency range independant of 
polarization, and just store weight values specifying the percentage 
polarized at each angle. The more samples, the better the accuracy, but 
slower and using more memory.


> Speaking of something completely different...iridescence.  It's
> an interference pattern between two surfaces of oil or similar,
> how 'bout trying to do that the hard way in POV?  Should be
> doable even in 3.1g...I think I'll play around with that and
> such when I get back to my computer.

There is the "irid" finish feature, which is apparently somewhat based 
on reality, though not a physical simulation...
I don't know what you have in mind though...simulating the waveforms 
with the onion pattern?

-- 
 -- 
Christopher James Huff <chr### [at] maccom>

Post a reply to this message

Christopher James Huff wrote:
> There is the "irid" finish feature, which is apparently somewhat
> based on reality, though not a physical simulation...
> I don't know what you have in mind though...simulating the waveforms
> with the onion pattern?

That or something similar...make two layers of patterned materials
close to each other but slightly misaligned...I've already managed
to achieve anisotropic reflection by actually modelling the finish,
so I wonder what other 'effects' I can do straight cheap-simulation
of.

Post a reply to this message

In article <3C49ED2D.E6961A0E@scifi-fantasy.com>,
 Tim Cook <tim### [at] scifi-fantasycom> wrote:

> That or something similar...make two layers of patterned materials
> close to each other but slightly misaligned...

> I've already managed to achieve anisotropic reflection by actually 
> modelling the finish, so I wonder what other 'effects' I can do 
> straight cheap-simulation of.

Details? I've done it by using averaged textures (I think Ron Parker 
discovered this technique), with each texture identical except for the 
normal...and a slower, older method used a very finely scaled normal and 
high antialiasing.

-- 
 -- 
Christopher James Huff <chr### [at] maccom>

Post a reply to this message

Christopher James Huff wrote:
> Details? I've done it by using averaged textures (I think Ron
> Parker discovered this technique), with each texture identical
> except for the normal...and a slower, older method used a very
> finely scaled normal and high antialiasing.

I was modelling an animation of a record rotating, having created
a series of 15 or so concentric rings out of CSGed cylinders
raised slightly over another thin cylinder.  Each ring was
separated by a distance equal to their width.  The material was
a simple shiny/reflecty grey.  As the 'record' rotated, I
observed the effect caused by the anisotropic reflective
property of brushed metal on, say, the face of a pocketwatch.

...if that isn't what anisotropic reflection is, then Caligari
is mis-labelling it in trueSpace :P

Post a reply to this message

On Sat, 19 Jan 2002 15:39:30 -0600, Tim Cook wrote:
> There's only 3 more values per ray for a more realistic simulation
> of light; amplidude/frequency/rotation around axis of travel...?

You also have to have a value for circular polarization, where the phase
of the horizontally polarized component is delayed by 90 degrees from the
phase of the vertically polarized component.  Also, for the sake of 
minimizing the number of rays we trace, we actually bundle all the 
frequency information together in a spectral power distribution of sorts.

> Could probably do some interesting things with having materials
> change the frequency of the light...

Fluorescence.  Yes, that could be interesting.

> Speaking of something completely different...iridescence.  It's
> an interference pattern between two surfaces of oil or similar,
> how 'bout trying to do that the hard way in POV?

"the hard way" could be harder than you think: it depends on the wave 
nature of light, which POV does not simulate.


-- 
#macro R(L P)sphere{L F}cylinder{L P F}#end#macro P(V)merge{R(z+a z)R(-z a-z)R(a
-z-z-z a+z)torus{1F clipped_by{plane{a 0}}}translate V}#end#macro Z(a F T)merge{
P(z+a)P(z-a)R(-z-z-x a)pigment{rgbf 1}hollow interior{media{emission 3-T}}}#end 
Z(-x-x.2x)camera{location z*-10rotate x*90normal{bumps.02scale.05}}

Post a reply to this message