In article <38E930DF.34C53EA2@netvision.net.il>, Lewis 
<nle### [at] netvisionnetil> wrote:

> Nice!
> Can you explain?

Ehh, not really, I just chose what seemed to be right, but I will try:

You don't need to use the ratio keyword, as I understand it, it only 
affects spotlights and cylinder lights. Since you will have rays of 
light coming through the clouds, you need to raise the samples to get 
the smaller details.

You had separate objects for the cloud layer and the media container. 
This was just unnecessary and harder to adjust, so I put them both in 
one container. The only times you would want the media container to 
extend outside the cloud layer is when you are attempting a complete 
simulation of the atmosphere, and then you would probably want a 
separate media for the outer part.

I added double_illuminate to the cloud/media container because the light 
is on the other side of it, this illuminates the visible side of the 
cloud texture. You can adjust the falloff of the illumination by 
changing the brilliance value in the finish.

I also changed the color_map on the clouds to use filtering only(I 
rarely use transmit, filter seems more realistic for nearly everything).

I moved the light_source so it was shining toward the camera, this makes 
the media show up better(since you aren't using an isometric scattering 
model). This is what produces that nice glow where the "sun" is.

Other than that, I just tuned the density and extinction values to look 
right.

-- 
Christopher James Huff - Personal e-mail: chr### [at] yahoocom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Web page: http://chrishuff.dhs.org/

Post a reply to this message