Hello all!

I've been working on my "Worldbowl" image again,
and am trying to update it with my most recent macros,
making things look better and more detailed, in order to
achieve a very pleasing and realistic image which I might
put on Zazzle. In my oppinion, its one of my images which
actually have the option to be made something very
special...

Well, to get to the point, I have come to the conclusion that
some nice raindrops on the window would look very nice
and add some tiny detail to the otherwise boring outside
view.
Right now, I'm just placing blob-components (more or less
random, I'm working on that part), but the main problem I have
is the refraction due to the water. Normally, Water has an ior
or 1.33 and glass one of 1.45 or so. In order to get rid of the
coincident surface problem, I'm just clipping the blob with a box.
This leaves an open shell on the glass surface.
I'm one of those ulra-realistic guys, so I'd like to know either
how I can either calculate an ior for the water to simulate the
refraction taking place between water and glass surfaces.
In this case, its not a big problem, I could just play with different
ior settings for the water till I get a look I like.
But let's not take a glass window, but one for drinking. In that case,
an "open shell" water would be a disc in the glass to simulate the
water surface. But I want to simulate the correct refraction taking place
when the viewing rays passes through the glass into the water.

Sounds pretty complicated and I'm not too sure if I make myself
clear. The other way round:
Most of the time, us POVers leave a little gap between two transparent
surfaces in order to avoid coincident surfaces. But when refracting rays,
the effect taking place is different as opposed to two surfaces touching
each other.

Any ideas?

Thanks in advance,
Tim

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

Post a reply to this message

Hm. Seems like I'm not raising much attention with this question,
eh?

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

Post a reply to this message

3e2356df$1@news.povray.org...
> Hm. Seems like I'm not raising much attention with this question,
> eh?
>

Can't help, I've had this little problem for years and I was never able to
solve it properly. Nothing I tried seemed to work as it should.  See :
http://www.oyonale.com/ldc/images/abeillesdetail2.jpg

The main issue, though, is less the surface problem than the distribution of
the droplets themselves. Ideally, their position should be the product of a
dynamic process, so that simple random positionning always seems wrong.
Perhaps Rune's particle system could help, I don't know. Blobs doesn't
behave like real water either, which is another problem.

G.

--
**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters

Post a reply to this message

In article <3e1c7b0b@news.povray.org>, "Tim Nikias" <tim### [at] gmxde> 
wrote:

> Most of the time, us POVers leave a little gap between two transparent
> surfaces in order to avoid coincident surfaces. But when refracting rays,
> the effect taking place is different as opposed to two surfaces touching
> each other.

Overlap them a bit. POV does things so that refraction is handled 
correctly in this case. The gap method simulates an air gap.

Designing the textures properly also helps. I usually use perfectly 
transparent textures and use interior attenuation or media to provide 
any coloration.

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: chr### [at] tagpovrayorg
http://tag.povray.org/

Post a reply to this message

> Any ideas?
Hi Tim!

Maybe, you can tinker with clipped_by, to clip the surface of the drop which
would contact the window. I don't know wether the ray remembers the ior of
the object it currently is in or not. But if it's so, the ray would go from
1.33 straight into 1.45. which should be right...

Hope could help
Kalle

Post a reply to this message

On Tue, 14 Jan 2003 07:10:48 +0100, "KalleK" <kal### [at] gmxde> wrote:

>Maybe, you can tinker with clipped_by, to clip the surface of the drop which
>would contact the window. I don't know wether the ray remembers the ior of
>the object it currently is in or not. But if it's so, the ray would go from
>1.33 straight into 1.45. which should be right...

Hey Kalle, that's *extremely* neat - did you try it? It does sound
plausible...


Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] vipbg
TAG      e-mail : pet### [at] tagpovrayorg

Post a reply to this message

That's what I'm currently doing, but I'm not sure
if it works as we think it does. As far as I understand
the raytracing technique, POV will stay with ior 1.45 from
the glass until it hits the surface of the water, and gets
refracted the way it should when exiting a 1.33 ior to
1.00 ior (water to air). But I'm not too sure about that,
and also no quiet sure how I should test that...

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

>
> >Maybe, you can tinker with clipped_by, to clip the surface of the drop
which
> >would contact the window. I don't know wether the ray remembers the ior
of
> >the object it currently is in or not. But if it's so, the ray would go
from
> >1.33 straight into 1.45. which should be right...
>
> Hey Kalle, that's *extremely* neat - did you try it? It does sound
> plausible...
>
>

Post a reply to this message

It "does things"? What does it do? Just curious.
Though doing that with blobs is still kinda difficult...
Instead of clipping, I'd have to difference it away
position it slightly inside the window... Pretty tracing
intensive, hu? I'll try and look what happens. Thanks
for the suggestion though,

Tim

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

>
> > Most of the time, us POVers leave a little gap between two transparent
> > surfaces in order to avoid coincident surfaces. But when refracting
rays,
> > the effect taking place is different as opposed to two surfaces touching
> > each other.
>
> Overlap them a bit. POV does things so that refraction is handled
> correctly in this case. The gap method simulates an air gap.
>
> Designing the textures properly also helps. I usually use perfectly
> transparent textures and use interior attenuation or media to provide
> any coloration.
>

Post a reply to this message

You're definitely right about placing the droplets.
I've written a small include, which scatters droplets
randomly across an area (the whole thing was designed
so far only for windows), and also creates "trails":
At one spot, it starts, and moves downward (a little
randomly moving towards right or left), and leaves
very small droplets on the way, and at the end, a big droplet.
Just like those drops that get big enough to flow down.

I've uploaded a zoomed-version on the window,
http://www.digitaltwilight.de/no_lights/rain_1.jpg
Its 100kb, but shows how the results look so far. I'm
fiddling a bit with size and numbers to see if I can
get something near-realistic with this method.

I guess Rune's Particle System won't be of use, cause
it doesn't have particle-particle-interaction, and that
would be what droplets on windows are screaming for.
Perhaps I'll try to create an include file for windows, but
doing it with curved survaces and other objects... Way
too difficult. Got to have a running particle-system to do
that...
But near-realistic isn't too bad. Its all about the looks.
How did you place those droplets of yours? Using trace?
Though some look like the gravitational-pull is actually
affecting the droplet. How'd you do that?

Regards,
Tim

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

>
> Can't help, I've had this little problem for years and I was never able to
> solve it properly. Nothing I tried seemed to work as it should.  See :
> http://www.oyonale.com/ldc/images/abeillesdetail2.jpg
>
> The main issue, though, is less the surface problem than the distribution
of
> the droplets themselves. Ideally, their position should be the product of
a
> dynamic process, so that simple random positionning always seems wrong.
> Perhaps Rune's particle system could help, I don't know. Blobs doesn't
> behave like real water either, which is another problem.
>
> G.
>
> --
> **********************
> http://www.oyonale.com
> **********************
> - Graphic experiments
> - POV-Ray and Poser computer images
> - Posters
>
>
>

Post a reply to this message

Using google this came up when looking for
"water droplets simulation":

http://www.cs.unc.edu/~lin/COMP259/LEC/29b.ppt

Its pretty interesting and perhaps, when I get the time to
understand it, I'll implement it. Could be useful, at least
for meshes (and windows can easily be done with
meshes).


--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

Post a reply to this message