POV-Ray : Newsgroups : povray.advanced-users : Moons and planets in daylight Server Time
26 Nov 2024 15:37:26 EST (-0500)
  Moons and planets in daylight (Message 1 to 5 of 5)  
From: John Bindas
Subject: Moons and planets in daylight
Date: 27 Feb 2000 13:15:49
Message: <38b969d5@news.povray.org>
The last few weeks, on my way to work, I noticed the Moon in the morning sky
and it reminded me of a problem I had months ago doing a sci-fi scene. When
you look up at the crescent Moon in daylight, you don't see the unlit part
of the Moon. You just see the lit crescent in a blue sky. This effect is a
common sci-fi art theme, an alien landscape with two moons and a Saturn-like
planet in the sky.

What works great for a night-time or a space scene, an ambient of 0 does not
work for a daylight scene. When I try to duplicate the real world effect in
POV, I get both the lit and unlit portions of the Moon visible against the
sky. If I drop the ambient down to 0, the unlit portion is black. If I match
the ambient to the color of the sky sphere (or some fraction of it like <.2,
.6, .8>*.10), the unlit portion is visible. I have tried things like
"negative" spotlights aimed at the unlit portion. Something like:

light_source {
 <0, 10, 0>
 color rgb <-.8, -.4, -.2>
 spotlight
 radius 15
 falloff 20
 tightness 10
 point_at <0, 0, 0>
}

With simple white test spheres, this works. However, I spent a lot of time
tinkering with the "negative" light color and the ambient statement. And
then I found with any complex texture, it doesn't work. My best success has
been with a constant fog with a distance set far enough away not to
interfere the majority of the scene. The unlit portion of the Moon is
obscured, but not completely invisible. To make the unlit portion invisible
or mostly invisible, the lit portion becomes too "washed out". I haven't
tried CSG, and I have a feeling this may the best solution. Anyway, this
hasn't bothered me enough to make me to break out my old math book and
calculate the correct angle of the CSG difference to the light source(s). It
bugs me just enough to ask this group if anyone out there has any hints or
tricks for a raytracing a moon or planet in daylight?

---
jb


Post a reply to this message

From: Margus Ramst
Subject: Re: Moons and planets in daylight
Date: 27 Feb 2000 15:18:26
Message: <38B9870C.1D61CDDD@peak.edu.ee>
You have to remember that the moon is outside the atmosphere, so you can't
really do this with a blue (infinite) sky_sphere.
I have achieved great results simulating the atmosphere with coloured fog or
media in a finite, spherical container. Now, the blackness of space and the
black (shadowed) part of the moon are not visible.
If the viewpoint & ground is now positioned in realistic proportion to the
atmosphere, you also get the natural darkening of the sky towards the zenith.
If you want, I can look around for an example scene.

Margus

John Bindas wrote:
> 
> The last few weeks, on my way to work, I noticed the Moon in the morning sky
> and it reminded me of a problem I had months ago doing a sci-fi scene. When
> you look up at the crescent Moon in daylight, you don't see the unlit part
> of the Moon. You just see the lit crescent in a blue sky. This effect is a
> common sci-fi art theme, an alien landscape with two moons and a Saturn-like
> planet in the sky.
>


Post a reply to this message

From: Bob Hughes
Subject: Re: Moons and planets in daylight
Date: 27 Feb 2000 15:33:41
Message: <38b98a25@news.povray.org>
Well you about covered them all I think.  Nothing good enough huh?  Could be
said of lots of things I suppose  :-)
I've done the CSG bit too and didn't like it at all.  I've always wanted some of
the shadow side to show up some, as if lit by reflected planet-light or
secondary star or something.  Either that or as though seen at twilight or
through a thin atmosphere.  Then those fog and ambient methods usually do okay.
What's needed for unseen shadows of beyond-atmosphere objects in the daylight
sky would be a new type of thing where a light source would only make the shone
upon parts of objects visible and yet not allow for tracing of any unlit parts
(now I'm wondering if there already is).
Anyway, a work-around would be to use huge specular highlighting (roughness 1)
on a fully clear object.  That way only the lit side shows.  Just a problem of
the possibility of seeing the inside of such a object if the camera were placed
wrong.  In fact the highlighting might show through from behind as well.
A lot of people have done this sort of thing so maybe there are answers, but I'm
think mostly the answer is to tweak the fog, ambient, and background.

Bob

"John Bindas" <big### [at] ixnetcomcom> wrote in message
news:38b969d5@news.povray.org...
| The last few weeks, on my way to work, I noticed the Moon in the morning sky
| and it reminded me of a problem I had months ago doing a sci-fi scene. When
| you look up at the crescent Moon in daylight, you don't see the unlit part
| of the Moon. You just see the lit crescent in a blue sky. This effect is a
| common sci-fi art theme, an alien landscape with two moons and a Saturn-like
| planet in the sky.
|
| What works great for a night-time or a space scene, an ambient of 0 does not
| work for a daylight scene. When I try to duplicate the real world effect in
| POV, I get both the lit and unlit portions of the Moon visible against the
| sky. If I drop the ambient down to 0, the unlit portion is black. If I match
| the ambient to the color of the sky sphere (or some fraction of it like <.2,
| .6, .8>*.10), the unlit portion is visible. I have tried things like
| "negative" spotlights aimed at the unlit portion. Something like:
|
| light_source {
|  <0, 10, 0>
|  color rgb <-.8, -.4, -.2>
|  spotlight
|  radius 15
|  falloff 20
|  tightness 10
|  point_at <0, 0, 0>
| }
|
| With simple white test spheres, this works. However, I spent a lot of time
| tinkering with the "negative" light color and the ambient statement. And
| then I found with any complex texture, it doesn't work. My best success has
| been with a constant fog with a distance set far enough away not to
| interfere the majority of the scene. The unlit portion of the Moon is
| obscured, but not completely invisible. To make the unlit portion invisible
| or mostly invisible, the lit portion becomes too "washed out". I haven't
| tried CSG, and I have a feeling this may the best solution. Anyway, this
| hasn't bothered me enough to make me to break out my old math book and
| calculate the correct angle of the CSG difference to the light source(s). It
| bugs me just enough to ask this group if anyone out there has any hints or
| tricks for a raytracing a moon or planet in daylight?
|
| ---
| jb
|
|


Post a reply to this message

From: Peter Popov
Subject: Re: Moons and planets in daylight
Date: 27 Feb 2000 16:19:20
Message: <bpG5OPDoZCz8C2ymHY9UyLB4pCpF@4ax.com>
Before I start digging around in my archives, have you tried matching
the ambient to the color of the sky, giving it a bump map and a
somewhat high diffuse value?

I remember asnwering the exact same question some time ago, I just
don't remember if this is the exact same answer I gave back then :)


Peter Popov
pet### [at] usanet
ICQ: 15002700


Post a reply to this message

From: Nieminen Juha
Subject: Re: Moons and planets in daylight
Date: 28 Feb 2000 03:10:48
Message: <38ba2d87@news.povray.org>
Try making the sky_sphere also black and add a fog of certain height
which simulates the atmosphere.

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):5;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.