"Bald Eagle" <cre### [at] netscapenet> wrote:
> And here you go:
>
> http://www.philiplaven.com/Publications/AO-42-03-p436.pdf
>
> http://www.philiplaven.com/mieplot.htm

http://www.philiplaven.com/p4.html

Post a reply to this message

"Bald Eagle" <cre### [at] netscapenet> wrote:
> Alain Martel <kua### [at] videotronca> wrote:
>
> > Tried projected_through with a filtering pattern creating a gradient. It
> > don't work. You get a warning about projected_through requiring a
> > texture-less object, and the pigment or pattern is ignored.
>
> Yeah - RTFM shows that it doesn't do what I had thought it did.
>
> https://wiki.povray.org/content/Reference:Light_Source#Projected_Through
>
> Projected Through
> You can use projected_through with any type of light source. Any object can be
> used, provided it has been declared beforehand. Projecting a light through an
> object can be thought of as the opposite of shadowing, in that only the light
> rays that hit the projected through object will contribute to the scene. This
> also works with area lights producing spots of light with soft edges. Any
> objects between the light and the projected through object will not cast
> shadows, additionally any surface within the projected through object will not
> cast shadows. Any textures or interiors on the object will be stripped and the
> object will not show up in the scene.
>
> The syntax is as follows:
>
> light_source {
>   LOCATION_VECTOR, COLOR
>   [LIGHT_SOURCE_ITEMS...]
>   projected_through { OBJECT }
>   }
>
>
> Maybe use a disk with a cylindrical pattern and a gradient, or a sphere with an
> onion and a gradient.
>
Here's something I tried.

Post a reply to this message

Attachments:
Download 'lightplay.jpg' (19 KB)

Preview of image 'lightplay.jpg'

I haven't done any more work on the diffraction stuff, but I did happen across
an interesting parametric equation describing the real shape of a falling
raindrop.

r = Sigma [A * ( 1 + Cn * cos(Cn * Theta)]
with a table of 8 C values for droplet radii ranging from 0 to 3mm

Had a devil of a time working that all out so that the values get linearly
interpolated for any given raindrop size.
Then I:
plotted it out with spheres,
messed up the parametric {},
tried to convert the parametric into an isosurface,
tried to render a sor {};
fixed my parametric {},
finally used a lathe {}, and added an ior of 1.33

I generated 100 raindrops of each size and randomly translated them.

- BW

Post a reply to this message

Attachments:
Download 'raindrop.png' (931 KB)

Preview of image 'raindrop.png'

Cousin Ricky <ric### [at] yahoocom> wrote:
> Many years ago, alphaQuad aka Melody issued a glory ping pong challenge,
> but was quite unsuccessful.

I had some very nice exchanges with the friendly Philip Laven, and we talked
about how simulating the optics was "fiendishly complicated", and how here on
the forum we'd like to merely simulate the effect as best we could without
approximating the underlying plane wave interference.

So, for the moment, I noticed that POV-Ray has an inbuilt rainbow atmospheric
effect.

https://wiki.povray.org/content/Reference:Rainbow

So I took the image from
http://www.philiplaven.com/Mie-r10-175-180-0p01-300col.jpg
(at http://www.philiplaven.com/p2c1.html)
Which already has the rgb values for the glory laid out in a linear form, and
plugged that (as a cropped version) as an image_map into a pigment and then into
a function, and used a loop to define my color map entries using the 0-1 index
as an argument to the function.
Then I used that color_map for the rainbow, tweaked some parameters, stuck a
plane with a cloud pigment behind it, and there we go.

Important note: the rainbow must make use of a mechanism very much like media,
since the plane needed to be hollow to see the rainbow. (One can also just flip
the plane normal using the inverse keyword)

I'm trying to figure out how to cast a shadow _through_ scattering media to
simulate the Brocken Spectre, but haven't figured that part out yet.

I might need to alter the color_map function to change the fading effect, but it
looks passable for now.

- BW

Post a reply to this message

Attachments:
Download 'glory2.png' (219 KB)

Preview of image 'glory2.png'

OK, I got the shadow, I just need to figure out how to get that LONG shadow that
goes all the way from the camera and out to the "orthographic shadow".

- BW

Post a reply to this message

Attachments:
Download 'glory2.png' (266 KB)

Preview of image 'glory2.png'

On 2/20/25 19:07, Bald Eagle wrote:
> I'm trying to figure out how to cast a shadow_through_ scattering media to
> simulate the Brocken Spectre, but haven't figured that part out yet.

Playing with an idea.

The attached image uses multiple isosurfaces as colored dust. I used 
yuqk's new f_popnrm_rnoise() to break shapes up rather than my usual 
method with extreme turbulence blowing isosurfaces apart.

It kinda of works as far as I've gotten...

The thought - still untested - is given these particles exist, it should 
be I can project light from the backside 'through' them and into 
scattering media between the camera and the particles.

The effect counts on fairly heavy AA (it's effectively part of the 
sampling / color-blending). With media too - this method might get 
really slow, but we'll see.

Off to do some RL stuff, but I'll try and get back and try the projected 
light / media part of it soon.

Bill P.

Post a reply to this message

Attachments:
Download 'try00.jpg' (66 KB)

Preview of image 'try00.jpg'

William F Pokorny <ano### [at] anonymousorg> wrote:
> On 2/20/25 19:07, Bald Eagle wrote:
> > I'm trying to figure out how to cast a shadow_through_ scattering media to
> > simulate the Brocken Spectre, but haven't figured that part out yet.
>
> Playing with an idea.

Nice.

However for the glory, Philip will be quite cross is you treat it like a
diffraction effect.
It's not - it's a plane wave interference effect.

The colors that you get are therefore sort of reversed from a regular rainbow.

Image_map attached.

- BW

Post a reply to this message

Attachments:
Download 'mie-r10-175-180-0p01-300col.jpg' (7 KB)

Preview of image 'mie-r10-175-180-0p01-300col.jpg'

On 2/26/25 18:08, Bald Eagle wrote:
> The colors that you get are therefore sort of reversed from a regular rainbow.
> 
> Image_map attached.

Thanks, Bill, for posting the image_map image with the colors.

Attached is an image where the light source was moved behind the plane 
with the wrinkles pattern clouds. I used the outer most of the six, 
color, isosurface dust, rings as the projected through shape. The light 
rays magically pop through the blocking plane and shoot out of a 
gazillion dust particles on the other side!

It is slow. Right at two hours on my old Intel i3 CPU. It's not all due 
scattering media; My leaving the plane's pigment as a wrinkles pattern 
was costly. A stored image mapped to the plane would be faster.

More work would be needed to make it look somewhat realistic to any 
given real world phenomenon. The image is generated 'looking the wrong 
way', for example, but the method has potential for faking effects like 
Glories I think.

Bill P.

Post a reply to this message

Attachments:
Download 'try01.jpg' (54 KB)

Preview of image 'try01.jpg'

William F Pokorny <ano### [at] anonymousorg> wrote:

> Thanks, Bill, for posting the image_map image with the colors.

Of course.  Many thanks to Philip Laven for making the output of his Mie theory
program available.

> The image is generated 'looking the wrong
> way', for example, but the method has potential for faking effects like
> Glories I think.

Clever way to do it, and yes, that's slow!
Are you using no_image for your shadow-casting object?

I'm also curious why "the wrong way" works, and I can't create a good shadow as
shown by the scattering media documentation page.  Do we not see shadows when
the camera is in line with the light source?

Might need to do an animation of a light source revolving around a ring of
columns . . .


-BW

Post a reply to this message

On 2/27/25 06:25, Bald Eagle wrote:
> Are you using no_image for your shadow-casting object?

No. It's completely black because there is no direct light or radiosity. 
It's sitting between the camera and the isosurface dust in the wrong way 
configuration; we are seeing the completely shadowed side.

Aside: I am using 'no_shadow' for the dust rings because I don't want 
them to burn time self shadowing.

> I'm also curious why "the wrong way" works, and I can't create a good shadow as
> shown by the scattering media documentation page.  Do we not see shadows when
> the camera is in line with the light source?

Attached is an image flipped the right way around where everything is 
projected onto a white plane (this the original form for the idea). The 
shadowing object is now just behind the camera - as is the light 
projected through dust.

The original idea was to have a colored light source for each ring, and 
to project the image onto some "clouds." This requires 6 light sources 
in my current set up - hence no AA was used in the image attached!

It's enough to show the right way around works too, though need to look 
for other ways to represent the colors (filter in front of the dust 
maybe, to get to one light source and one isosurface dust shape).

Aside 2: I did have to increase my method 3 media samples by about 3x in 
the right-way direction for the media to look smooth. I 'think' because 
the light intensity is spreading more in the right-way-around / truly 
projecting onto something, direction.

Bill P.

Post a reply to this message

Attachments:
Download 'try02.jpg' (96 KB)

Preview of image 'try02.jpg'