[Radiance-general] clouds textures

[Greg Ward]

Hi Bob,

You need to use a mapping that indexes off the ray directions rather  
than locations, which are meaningless for sources (as you  
discovered).  The fisheye.cal file included in the ray/lib directory  
should be a good starting point:

{
         Calculate coordinates for a 180 degree fisheye lens.
         Assume view direction is (0,1,0), view up (0,0,1), (1,0,0)  
right.

         4/30/90
}

fish_u = .5 + Dx/fish_Rxz * fish_Ry;
fish_v = .5 + Dz/fish_Rxz * fish_Ry;

fish_Rxz = sqrt(Dx*Dx + Dz*Dz);
fish_Ry = Acos(Dy) / PI;

----------
You may need to apply a rotation to this if your z-axis points up, as  
is usually the case.  Something like:

void colorpict cloud_pat
9 red green blue clouds.pic fisheye.cal fish_u fish_v -rx 90
0
0

cloud_pat brightfunc skyfunc
2 skybr skybright.cal
0
7 1 3.58e+01 3.07e+01 1.55e+00 0.054270 -0.269929 0.961350

-------
In this example, I have added your cloud pattern to a sky created by  
gensky.  If you have an HDR image of the sky you're working from, you  
may just be able to go with that.

Additional resources:

Mark Stock's Radiance Miscellany page:
	http://mark.technolope.org/pages/radmisc.html

Link to Simon Crone's Radiance skies collection:
	http://radsite.lbl.gov/radiance/pub/libraries/

-Greg

> From: bob coyne <coyne at semanticlight.com>
> Date: January 31, 2006 9:48:36 AM PST
>
> What's the best way to get a sky with a cloud texture map in  
> Radiance (assuming
> the cloud texture map is already "sphericized")?  I've tried  
> applying a COLORPICT pattern onto
> the sky SOURCE  but it doesn't work (I get black).  I assume this  
> is because the UV
> coordinates aren't available in the same way (with Lu, Lv) that  
> works with mesh objects.
>
> I've also tried making a polygonal mesh sky dome. That's easily  
> textured but creates other problems
> (shadows and other lighting effects on the sky) .
>
> Thanks for any help or guidance on how to do this sort of thing!
>
> - Bob
>