Greetings.

Who knows how to make with POV-Ray a cloudy planet surface and thin 
atmosphere similar to that of the photo? I am in big need of a macro which 
can do surfaces of planets (spheres). Who can help?

Sven

Post a reply to this message

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

Preview of image 'Planet.jpg'

Sven Littkowski wrote:
> Who can help?

God can (after all, He does raytracing in real time ;) ).

(Oh boy, now I *know* I've been reading these newsgroups too long).



Seriously, what you're asking is something that people have been trying 
to accomplish for years.  Anyway, I'd suggest adding some noise to a 
sphere function and using an isosurface for the planet itself, and 
encase it in a media object for the atmosphere.  I couldn't say about 
the specific settings, though.

...Chambers

Post a reply to this message

Sven Littkowski wrote:
> Greetings.
> 
> Who knows how to make with POV-Ray a cloudy planet surface and thin 
> atmosphere similar to that of the photo? I am in big need of a macro which 
> can do surfaces of planets (spheres). Who can help?
> 
> Sven 
> 
> 
> 
there is not really a readymade solution that can be just handed to you

there have been at least two investigations that I can remember into the 
idea of creating an all in one macro or technique for the rendering of 
an encompassing atmostphere that would produce both sunset and high noon 
effects from the planet surface, and space shot effects too.  Searching 
the images group might turn those up, though they might be more purist 
than you need.  But it is something that POV-Ray should be capable of.

more pragmatically there is a nice summary of cloud techniques by Gilles 
Tran at the bottom of his page
http://www.oyonale.com/ressources/english/sources13.htm
though again his context is more specifically landscapes from the planet 
surface

you already know about Christoph Hormann's work and so probably have 
read that he may be looking into the problem of creating such effects as 
a possible future elaboration of his project

so we are all awaiting with interest the ingenuity that you may bring to 
this ;)

Post a reply to this message

"Ben Chambers" <ben### [at] pacificwebguycom> schrieb im Newsbeitrag 
news:4471f12c@news.povray.org...
> God can (after all, He does raytracing in real time ;) ).
> (Oh boy, now I *know* I've been reading these newsgroups too long).

Ehhh... Yes. You are right... Toooooo long...        :-)

> Anyway, I'd suggest adding some noise to a sphere function and using an 
> isosurface for the planet itself, and encase it in a media object for the 
> atmosphere.

Yes, maybe. I just lack any experience with iso surfaces and media 
objects...
Hmmm, since you talk about God... Do you think you want (surely not, but at 
least I try) to try something like that?
(Oh boy, now I *know* I've been reading these newsgroups too long, as well).

Well-rendered greetings,

Sven

Post a reply to this message

A wah dat? Me and having ingenuity..?

Well, you believe things, I tell you... Ha ha! Well, I am not experienced at 
all with how to make such an atmosphere. Ben Chambers (just above your reply 
to my posting) mentioned a way, but I am not experienced with iso surfaces 
nor with media. His idea sounds interesting, But I cannot work out such new 
things...

I wished I could.

Sven

Post a reply to this message

Sven Littkowski wrote:
> "Ben Chambers" <ben### [at] pacificwebguycom> schrieb im Newsbeitrag 
>>Anyway, I'd suggest adding some noise to a sphere function and using an 
>>isosurface for the planet itself, and encase it in a media object for the 
>>atmosphere.
> 
> 
> Yes, maybe. I just lack any experience with iso surfaces and media 
> objects...

Here's a start.

The function:
#declare fn_planet = function {
	sqrt( pow(x, 2) + pow(y, 2) + pow(z, 2) )
	+ f_snoise3d(x/64, y/64, z/64)*4
	+ f_snoise3d(x/32, y/32, z/32)*2
	+ f_snoise3d(x/16, y/16, z/16)
	+ f_snoise3d(x/8,  y/8,  z/8)/2
	+ f_snoise3d(x/4,  y/4,  z/4)/4
	+ f_snoise3d(x/2,  y/2,  z/2)/8
	+ f_snoise3d(x,    y,    z)/16
	- 100
}

Note: for effect I increased the bump sizes.  For a realistic range of 
heights, IIRC the largest should be *0.4 (not *4), and the scales would 
decrease from there.

And the object itself:
isosurface {
	function { fn_planet(x,y,z) }
	threshold 0
	evaluate 1, 1.2, 0.8
	
	contained_by {sphere {0, 110} }
	
	pigment {color rgb <0.8,.75,.6>}
}

If you want oceans, then this is a crude beginning:
sphere {
	0, 99.75
	pigment {color rgb z}
}

Vary the radius of the sphere to raise or lower the oceans (do it in 
very small increments, though).

I'm still trying out some different media settings...

...Chambers

Post a reply to this message