"nemesis" <nam### [at] gmailcom> schrieb im Newsbeitrag 
news:web.4478a5d9a626bc34b6a6018f0@news.povray.org...
> Sven, i hardly see any atmosphere at all.  Guess the sky could be a little
> higher? ;)

Yes, it could, but I wanted to stay as realistic as possible. The thin layer 
of atmosphere should be visible only at close range (see attached photo).


> and i think the correct setting for an atmosphere would be scattering 
> media,
> not emitting.  But it's far slower.

Can you publish here a smaple of scattering media? It would help me a lot! I 
try it now with a sphere and a pigment_map, but have first to figure out how 
to create a spherical pigment_map, as gradients x, y or z are only towards 
one axis. I hardly worked before with pigment_maps.


> and BTW, do we really see our city lights from space?  I thought most
> cientists believe our civilization would go unnoticed by an eventual
> observer at a quick glance of our planet...

Hmm, here you might have a point, but not for sure. So far as I know, you 
might not see individual lights (of course not), but just the biggest 
clusters of lights such created by big cities and maybe reflecting clouds 
above them. But I am not sure. But in any way, lights look good. If only I 
could find a way to manage that they would be only at the nightside of 
earth...

That would help me also for future planet renders...

Greetings,

Sven

Post a reply to this message

Attachments:
Download 'Planet.jpg' (100 KB)

Preview of image 'Planet.jpg'

"nemesis" <nam### [at] gmailcom> wrote:
> Christoph Hormann <chr### [at] gmxde> wrote:
> > Sven Littkowski wrote:
> > > - despite of ambient 0.0 the atmosphere is even visible on night side
> >
> > Obvious with emitting media...
>
> Sven, i hardly see any atmosphere at all.  Guess the sky could be a little
> higher? ;)
>
> and i think the correct setting for an atmosphere would be scattering media,
> not emitting.  But it's far slower.
>
> and BTW, do we really see our city lights from space?  I thought most
> cientists believe our civilization would go unnoticed by an eventual
> observer at a quick glance of our planet...

> and i think the correct setting for an atmosphere would be scattering media,
> not emitting.  But it's far slower.

AH yes, that's it... Emitting lights up at night.... in my atmosphere, I use
"scattering" which produces the most realistic results.  I posted my
atmosphere macro in another forum on this news group, the "rings cutting
the atmosphere" or something like that, when I posted a question on another
problem.

Post a reply to this message

"Sven Littkowski" <sve### [at] jamaica-focuscom> wrote:
> Hi.
>
> Please see here my earth render image. The planet consists of four items:
> - planet surface sphere (bump map and texture)
> - planet lights sphere (semi transparent PNG image at ambient 1.0)

NICE city lights!  However, here's a concern I noticed... it seems that you
have a very cloudy daytime and a very clear nighttime.  That will
definitely be unrealistic.

And yes.... the daytime is so extremely bright, that you won't see anything
on the night side.... but if you concentrate on the night side, then there
will be lights.... which are as bright as the stars.  But on cloudy earth,
only portions will be seen.

Post a reply to this message

EagleSun wrote:
> NICE city lights!  However, here's a concern I noticed... it seems that you
> have a very cloudy daytime and a very clear nighttime.  That will
> definitely be unrealistic.

Floating cities?  Flying cars!

They promised me there would be flying cars!  Where are my flying cars?!

-- 
Tim Cook
http://home.bellsouth.net/p/PWP-empyrean

-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GFA dpu- s: a?-- C++(++++) U P? L E--- W++(+++)>$
N++ o? K- w(+) O? M-(--) V? PS+(+++) PE(--) Y(--)
PGP-(--) t* 5++>+++++ X+ R* tv+ b++(+++) DI
D++(---) G(++) e*>++ h+ !r--- !y--
------END GEEK CODE BLOCK------

Post a reply to this message

Hi Eagle Sun,

about the lights: I also wonder that they are visible. But it is 
explainable. Let me start with some characteristics of my scene:
- The surface of earth consists of two textures and a bump map.
- The lights are actually the second texture of the surface.
- The clouds are another sphere just slightly bigger than the earth sphere.
- The clouds sphere consists of one semi-transparent PNG image map (the 
whiter the more untransparent) and a bump map.
- There is a tiny distance between lights (earth surface) and clouds.

I think, many lights are still visible because the clouds (like in real 
nature) are a bit transparent at their edges and of course between the 
clouds. So I could say, the lights are shining through the clouds in many 
locations. I am just sad that the cloud layer does not blur the lights 
shining through, as real clouds would do. Real clouds would diffuse the 
light shining through and blur it, but also be lightened a bit by the lights 
beyond them.

Questions:

- Who knows any way to limit the lights to only the nightside?
- I need a sphere which functions like fog: a color shift from transparent 
at the outest outside to blue in the center, for an atmosphere.

Greetings,

Sven



"EagleSun" <nomail@nomail> schrieb im Newsbeitrag 
news:web.447a83d9a626bc349756002c0@news.povray.org...
> "Sven Littkowski" <sve### [at] jamaica-focuscom> wrote:
>> Hi.
>>
>> Please see here my earth render image. The planet consists of four items:
>> - planet surface sphere (bump map and texture)
>> - planet lights sphere (semi transparent PNG image at ambient 1.0)
>
> NICE city lights!  However, here's a concern I noticed... it seems that 
> you
> have a very cloudy daytime and a very clear nighttime.  That will
> definitely be unrealistic.
>
> And yes.... the daytime is so extremely bright, that you won't see 
> anything
> on the night side.... but if you concentrate on the night side, then there
> will be lights.... which are as bright as the stars.  But on cloudy earth,
> only portions will be seen.
>
>

Post a reply to this message

"Sven Littkowski" <sve### [at] jamaica-focuscom> wrote:
> - Who knows any way to limit the lights to only the nightside?

You can use a gradient pattern that's orientated to point at your light
source to mix your textures. You have to transform it a bit first, though.
I ended up writing a small scene to check this, so I'll just post it in
povray.text.scene-files (hopefully).

> - I need a sphere which functions like fog: a color shift from transparent
> at the outest outside to blue in the center, for an atmosphere.

That's a job for scattering media, I imagine. Emitting media won't do it.
There have been several examples posted up here, like this one by Tek:

http://news.povray.org/43c8dedf%40news.povray.org

Or this one, by Scott Gammans:

http://news.povray.org/web.3fbd42078ae459e44e0a58990%40news.povray.org

(hopefully those work, if they don't then just search news.povray.org for
"scattering planet atmosphere" keywords or similar).

Tom

Post a reply to this message

Sven,

For illuminated lights in the shadowed portion you could use MegaPOV's
'projection pattern'.  You can find samples here:

http://megapov.inetart.net/samples.html

Post a reply to this message

"Mark Birch" <las### [at] hotmailcom> schrieb im Newsbeitrag 
news:web.447b904ca626bc344daddc090@news.povray.org...
> For illuminated lights in the shadowed portion you could use MegaPOV's
> 'projection pattern'.




Hi Mark,

I just wonder if the current version of MegaPov has all the same features as 
the Beta 13. I hope, the POVgrammers will include all MegaPov features into 
the future regular POV-Rays, as I consider the MegaPov features as quite 
interesting, helpful and usable. I truly hope the glorious programmers will 
take them.

For now, I prefer to use the Beta 13 in order to beta-test it and to get a 
deep feeling of the scene writing. Starting to use now MegaPov, would be 
some circumstance. Despite of that I (like nearly everyone here) like the MP 
features.

Best greetings and thanks for your hint,

Sven

PS:

Mark you - I still do consider your hint. Let me publish here (again) the 
entire scene code so you can see it. Could you please (if you like) alter 
that code so it could function as 'projection pattern' within MegaPov? That 
way, I could learn some different MegaPov programming and could more make my 
mind. Thanks.

---------------------------- 
NOTE: I am using the great universe sky sphere wonderfully done by Nekar 
Xenos
----------------------------

#include "colors.inc"
#include "textures.inc"
#include "shapes.inc"
#include "metals.inc"
#include "glass.inc"
#include "woods.inc"

#declare Earth = texture { pigment { image_map { jpeg "E:\Scripts\Eigene 
Bilder\Povray Textures\Sphere_Earth_01.jpg" map_type 1 } rotate < 0.0, 0.0, 
0.0 > } finish { ambient 0.0 } }
#declare EarthLights = texture { pigment { image_map { png 
"E:\Scripts\Eigene Bilder\Povray Textures\Sphere_Earth_Lights_01.png" 
map_type 1 } rotate < 0.0, 0.0, 0.0 > } finish { ambient 0.25 } }
#declare EarthClouds = texture { pigment { image_map { png 
"E:\Scripts\Eigene Bilder\Povray Textures\Sphere_Earth_Clouds_03.png" 
map_type 1 } rotate < 0.0, 0.0, 0.0 > } finish { ambient 0.0 } }

global_settings
{
 assumed_gamma 2.0
 radiosity {  }  //
}

camera
{
 location < 000.0, 0.0, -300.0 >
 look_at < 000.0, 0.0, 0.0 >
}
/**/
light_source
{
 < 1000.0, 0.0, 250.0 >
 color White *3
}

#macro StarSubPigment(StarSize, StarFrequency)
 crackle
 form < 1.0, 0.0, 0.0 >
 #local Cutoff=StarSize*StarFrequency;
 color_map
 {
  [ Cutoff rgb 1.0 ]
  [ Cutoff rgb 0.0 ]
 }
 scale 1/StarFrequency
#end

// minsize, maxsize = size range of stars (some stars will always be smaller 
due to nature of crackle pattern)
// starfrequency = how many stars
// variationfrequency = how quickly their size changes
#macro StarPigment(MinSize, MaxSize, StarFrequency, VariationFrequency)
 #local NumSizeSteps=8;
 pigment
 {
  pigment_pattern
  {
   bozo
   color_map
   {
    [ 0.0 rgb 0.75 ]
    [ 1.0 rgb 1.00 ]
   }
   scale 1/VariationFrequency
  }
  pigment_map
  {
   #local StepNum=0;
   #while (StepNum<NumSizeSteps)
    #local SizeAmnt=(StepNum+.5)/NumSizeSteps;
    #local PreSize=MinSize+(StepNum/NumSizeSteps)*(MaxSize-MinSize);
    #local PostSize=MinSize+((StepNum+1)/NumSizeSteps)*(MaxSize-MinSize);
    [ SizeAmnt StarSubPigment(PreSize,StarFrequency) ]
    [ SizeAmnt StarSubPigment(PostSize,StarFrequency) ]
    #local StepNum=StepNum+1;
   #end
  }
 }
#end

sky_sphere
{
 StarPigment
 (
  0.000002,  // min star size
  0.000008,  // max star size
  10000,     // star density (higher = more)
  5          // variation frequency (higher = star size changes more over 
nearby parts of the sky)
 )
}

#declare Earth_Surface=sphere
{
 0.0, 1.0
 texture { Earth }
 texture { EarthLights }
 normal
 {
//  density_file df3 "Sphere_Earth_Bumps_01.df3"
//  interpolate 1
  bump_map
  {
   jpeg "Sphere_Earth_Bumps_01.jpg"
   map_type 1
  }
 }
}

#declare Earth_Clouds=sphere
{
 0.0, 1.0005
 texture { EarthClouds }
 normal
 {
//  density_file df3 "Sphere_Earth_Bumps_01.df3"
//  interpolate 1
  bump_map
  {
   jpeg "Sphere_Earth_Clouds_Bumps_03.jpg"
   map_type 1
  }
 }
}

#declare Earth_Atmosphere=sphere
{
 < 0.0, 0.0, 0.0 > 1.0
 pigment
 {
  color rgbt < 0.3, 0.3, 1.0, 1.0 >
 }
 hollow
 interior
 {
  media
  {
   intervals 10
   scattering
   {
    2,
    rgb < 0.3, 0.3, 1.0 >
    extinction 1.0
   }   // 0.75
   density
   {
    spherical
    color_map
    {
     [ 0.000 rgb < 0.0, 0.0, 0.0 > ]
     [ 1.000 rgb < 0.1, 0.1, 0.5 > ]
    }
   }
   samples 1, 10
   confidence 0.9999
   variance 1/1000
   ratio 0.9
  }
 }
 finish { ambient 0.0 diffuse 0.9/**/ }
 scale 1.0075
}

#declare Earth=union
{
 object { Earth_Surface }      //
 object { Earth_Clouds }       //
 object { Earth_Atmosphere }   //
 scale 100.0
}

object { Earth }

Post a reply to this message

Sven Littkowski wrote:
> I just wonder if the current version of MegaPov has all the same features as 
> the Beta 13. I hope, the POVgrammers will include all MegaPov features into 
> the future regular POV-Rays, as I consider the MegaPov features as quite 
> interesting, helpful and usable. I truly hope the glorious programmers will 
> take them.
> 
> For now, I prefer to use the Beta 13 in order to beta-test it and to get a 
> deep feeling of the scene writing. Starting to use now MegaPov, would be 
> some circumstance. Despite of that I (like nearly everyone here) like the MP 
> features.

The only difference between 3.7 and 3.6 (other than the gamma issues 
mentioned) should be speed, and that's mostly true for MP systems.  3.6 
and MegaPOV both have all the same features and SDL syntax that 3.7 
supports.

...Chambers

Post a reply to this message

Thanks, Chambers. I am thinking about it.

Sven

Post a reply to this message