Hi Tracers!

Currently I'm working on a not-too-large project, a view over Earth from
an altitude of about 300 to 600 miles, with the sun not much above the
horizon, just casting the first morning light on the surface before the
viewer (at least finally; at the moment I would even be glad if it
worked correctly with the sun overhead).

As I still have dug somewhat into PoV 3.1, I think it would be doable
only when placing a transparent (hollow!) outer atmosphere sphere around
the solid Earth body (the atmosphere with a radius of 6.528 km compared
to 6.378 km for the planet itself) and then filling this sphere with an
interior/media, setting emission color and scattering etc.

But whatever scattering mode I use, I never get the blue gradient
atmosphere seam along the horizon familiar from real images shot from
orbit...  can anyone here help me further?

See you in Khyberspace!

Yadgar

Post a reply to this message

> 
> But whatever scattering mode I use, I never get the blue gradient
> atmosphere seam along the horizon familiar from real images shot from
> orbit...  can anyone here help me further?
> 

The correct scattering type for a (clear) atmosphere is probaby Raleigh
(type 4). The atmosphere gets thinner at higher altitudes, so perhaps you should
also simulate that (density with "onion" or "spherical" pattern). And for the
amosphere to be blue, you obviously have to tweak the colour of emission,
scattering, etc.

-- 
Margus Ramst

Personal e-mail: mar### [at] peakeduee
TAG (Team Assistance Group) e-mail: mar### [at] tagpovrayorg

Post a reply to this message

Margus Ramst wrote:
> 

> >
> > But whatever scattering mode I use, I never get the blue gradient
> > atmosphere seam along the horizon familiar from real images shot from
> > orbit...  can anyone here help me further?
> >
> 
> The correct scattering type for a (clear) atmosphere is probaby Raleigh
> (type 4). The atmosphere gets thinner at higher altitudes, so perhaps you should
> also simulate that (density with "onion" or "spherical" pattern). And for the
> amosphere to be blue, you obviously have to tweak the colour of emission,
> scattering, etc.

IRL, the blue comes from wavelength-dependent scattering. POV still
doesn't do this (although someone was working on a patch some months back).

Of course, you could try to fake it.

-Xplo

Post a reply to this message

"Xplo Eristotle" <inq### [at] unforgettablecom> wrote in message
news:39CABB99.A5937937@unforgettable.com...
| Margus Ramst wrote:
| >

| > >
| > > But whatever scattering mode I use, I never get the blue gradient
| > > atmosphere seam along the horizon familiar from real images shot from
| > > orbit...  can anyone here help me further?
| > >
| >
| > The correct scattering type for a (clear) atmosphere is probaby Raleigh
| > (type 4). The atmosphere gets thinner at higher altitudes, so perhaps
you should
| > also simulate that (density with "onion" or "spherical" pattern). And
for the
| > amosphere to be blue, you obviously have to tweak the colour of
emission,
| > scattering, etc.
|
| IRL, the blue comes from wavelength-dependent scattering. POV still
| doesn't do this (although someone was working on a patch some months
back).
|
| Of course, you could try to fake it.

Which means use a density_map having a color shift.  I've done that, and it
can work but you might find there is difficulty in getting the color to show
while also thinning the media out.  I used a fairly large blue value
compared to the red and green because it seemed relational.  In other words,
if the red and green were 0.05 and blue were only 0.5 chances are it won't
be very blue compared to if it were 10.  Or if the blend isn't being allowed
for to give blue only zone, i.e. [0 blue 0] [0.1 blue 1] can remain all blue
while fading toward the outer edge but not [0 blue 1] [0.1 rgb 1].

Bob

Post a reply to this message

I seem to remember reading that the atmosphere density drops off
exponentially with a scale of 8.4km, and that air scattering for blue
frequencies is 15 times that for red. These are the values I use for
atmospheres, anyway.

Bye for now,
	Mike Andrews.


> 
> Hi Tracers!
> 
> Currently I'm working on a not-too-large project, a view over Earth from
> an altitude of about 300 to 600 miles, with the sun not much above the
> horizon, just casting the first morning light on the surface before the
> viewer (at least finally; at the moment I would even be glad if it
> worked correctly with the sun overhead).
> 
> As I still have dug somewhat into PoV 3.1, I think it would be doable
> only when placing a transparent (hollow!) outer atmosphere sphere around
> the solid Earth body (the atmosphere with a radius of 6.528 km compared
> to 6.378 km for the planet itself) and then filling this sphere with an
> interior/media, setting emission color and scattering etc.
> 
> But whatever scattering mode I use, I never get the blue gradient
> atmosphere seam along the horizon familiar from real images shot from
> orbit...  can anyone here help me further?
> 
> See you in Khyberspace!
> 
> Yadgar

Post a reply to this message

Hi Margus and all others!

Margus Ramst wrote:

> The correct scattering type for a (clear) atmosphere is probaby Raleigh
> (type 4). The atmosphere gets thinner at higher altitudes, so perhaps you should
> also simulate that (density with "onion" or "spherical" pattern). And for the
> amosphere to be blue, you obviously have to tweak the colour of emission,
> scattering, etc.

Now I've tried interior with ior=1, in "media" I used rayleigh scattering with a
medium blue (rgb 0, 0.5, 1), the density is modified by "spherical", with a density
map:

[0 rgb <0,0,1>]
[0.95 rgb<0, 0.5, 1>]
[1 rgb<0,0,0>]

I thought this should result in an atmospheric fringe starting with a medium to
bright blue at the Earth's surface, outwards gradually fading to black. The bright
blue near the surface in fact fades, but into blue instead of black, and the blue
has a sharp border with black space!

I found out that the latter is because of the selected scattering color... is there
way to modify also the scattering so that it gets weaker towards the surface of the
atmospheric sphere?

For a better explanation here the POV code:

//  $MRY$: 'Air'




#declare Air =

   material  // Air

   {

      texture

      {

         pigment

         {

            color rgbft <1.0, 1.0, 1.0, 1.0, 1.0>

         }

      }

      interior

      {

         ior 1

         media

         {

            scattering {4, rgb<.0, .5, 1>}
            density
            {
               spherical

               density_map

               {
                  [0.0   rgb <0,0,1>]
                  [0.95  rgb<0,0.5,1>]
                  [1.0   rgb<0,0,0>]
               }

            }

         }

      }

   }


See you back in orbit ;-)!

Yadgar

Post a reply to this message

> 
> Now I've tried interior with ior=1, in "media" I used rayleigh scattering with a
> medium blue (rgb 0, 0.5, 1), the density is modified by "spherical", with a density
> map:
> 
> [0 rgb <0,0,1>]
> [0.95 rgb<0, 0.5, 1>]
> [1 rgb<0,0,0>]
> 

The color of your media is easier to control if you set either the scattering
color or color_map entries to grayscale, because the two are multiplied to get
the final density/color.
The sharp border is there because you have your map entries backwards: with
spherical, 1 is at the center and 0 is at the perimeter, whereas you seem to
have assumed the opposite.

-- 
Margus Ramst

Personal e-mail: mar### [at] peakeduee
TAG (Team Assistance Group) e-mail: mar### [at] tagpovrayorg

Post a reply to this message

> 
> Hi Tracers!
> 
> Currently I'm working on a not-too-large project, a view over Earth from
> an altitude of about 300 to 600 miles, with the sun not much above the
> horizon, just casting the first morning light on the surface before the
> viewer (at least finally; at the moment I would even be glad if it
> worked correctly with the sun overhead).
> 
> As I still have dug somewhat into PoV 3.1, I think it would be doable
> only when placing a transparent (hollow!) outer atmosphere sphere around
> the solid Earth body (the atmosphere with a radius of 6.528 km compared
> to 6.378 km for the planet itself) and then filling this sphere with an
> interior/media, setting emission color and scattering etc.
> 
> But whatever scattering mode I use, I never get the blue gradient
> atmosphere seam along the horizon familiar from real images shot from
> orbit...  can anyone here help me further?

	It does not scatter by color. 

	But if you want to play with a good model, you need several layers of
atmosphere with different scattering by density. And the color
scattering when it exists is by density so you may be able to fake it by
making the upper layer a forced bluish scatter. 

-- 
The last stand of the Brunen-Ghee.

	-- The Iron Webmaster, 5

Post a reply to this message