I was wondering if anyone could direct me toward a scene with a 
realistic daytime sky, clouds, haze and sun that I could reuse in my own 
scene (I have already determined units of measurement for my scene).

Thanks!

-Mike

Post a reply to this message

SharkD wrote:
> I was wondering if anyone could direct me toward a scene with a 
> realistic daytime sky, clouds, haze and sun that I could reuse in my own 
> scene (I have already determined units of measurement for my scene).
> 
> Thanks!
> 
> -Mike

Such things are very hard to port between scenes due to scale.  At 
least, assuming you're using media for the abovementioned effects.


...Chambers

Post a reply to this message

SharkD a écrit :
> I was wondering if anyone could direct me toward a scene with a 
> realistic daytime sky, clouds, haze and sun that I could reuse in my own 
> scene (I have already determined units of measurement for my scene).
> 
> Thanks!
> 
> -Mike

Most of those are hard to port, especialy if the scale of the scenes 
differs.

The daytime sky can be done with a sky_sphere to provide the backdrop 
for the other elements. Usualy, all you need is a nice gradient.

The haze and clouds are harder to port, and, in some case, it's not 
possible.
Realistic clouds often require the use of media. Same for the haze.
If you change the scale, you need to change the density, and possibly 
the pattern(s) used.
If your POV units represent a smaller unit (feet to meter for example) 
that the original, the media will get to thick.

The easiest part, is the Sun. Look at the lightsys and sunpos.


Alain

Post a reply to this message

With a bit of tooling around I managed to make it work. I created a 
panorama using it that you can find here:

http://isometricland.com/gearhead/ghpanotour.htm

-Mike

Post a reply to this message

I'm trying to scale the volumetric clouds (in particular the flat "cloud 
bank") found on this page:

http://www.geocities.com/evilsnack/tut01.htm

The Povray docs only briefly mention what you need to do in order to 
scale interior media. The example in the docs is for an emission media 
with no explicitly defined density map. The clouds script however is a 
scattering media with a density map (see below). Have I done things 
correctly?




#local Inc_Scale_Amount = Width;
#local Dec_Scale_Amount = sqrt(3) / vlength(Inc_Scale_Amount);

#local pigCloudBank = pigment
{
	average
	pigment_map {[1 planar][1 granite scale 50]}
}

box
{
	<-200,-2,-200>,<200,2,200>
	hollow
	texture {pigment {rgbt 1}}
	interior
	{
		media
		{
			scattering {1, 0.5 * Dec_Scale_Amount}
			method 3
//				samples 30, 100
			intervals 1
			density
			{
				pigment_pattern {pigCloudBank}
				density_map {[.6 rgb 0][.61 rgb 1]}
			}
		}
	}
	scale Inc_Scale_Amount
	translate y * Width * 8
}

Post a reply to this message

SharkD schrieb:
> #local Inc_Scale_Amount = Width;
> #local Dec_Scale_Amount = sqrt(3) / vlength(Inc_Scale_Amount);

What is "Width"?

It should be a scalar, otherwise you'll be running into trouble, as 
you'd need an anisotropic media density (i.e. different density 
depending on the orientation of the traversing ray) for a proper job.

Then if it is a scalar, the above values should actually be:

   #local Inc_Scale_Amount = Width;
   #local Dec_Scale_Amount = 1 / Width;

If you absolutely need it to be a vector, you'll probably want the 
"thickness" to remain the same, i.e. retain the effective absorption in 
Y direction; you'd achieve this by defining the values as:

   #local Inc_Scale_Amount = Width;
   #local Dec_Scale_Amount = 1 / Width.y;

Post a reply to this message

clipka wrote:
> SharkD schrieb:
>> #local Inc_Scale_Amount = Width;
>> #local Dec_Scale_Amount = sqrt(3) / vlength(Inc_Scale_Amount);
> 
> What is "Width"?
> 
> It should be a scalar, otherwise you'll be running into trouble, as 
> you'd need an anisotropic media density (i.e. different density 
> depending on the orientation of the traversing ray) for a proper job.

Good catch! Yes, "Width" is a scalar. However, even after fixing this 
the clouds don't look quite like the image on the website I linked to. 
The clouds are supposed to be dark underneath, but aren't when I render 
the scene.

light_source
{
	<+3000, +6000, -7000,> * Width
	rgb 2.0
	parallel
}

The location of the light source is well above the clouds. But, the 
light source is twice as bright. How might I modify the media to 
compensate? I suppose I could contain the clouds in a light group with a 
dimmer light source, but I'd rather avoid this since it affects things 
like shadows.

-Mike

Post a reply to this message

SharkD wrote:
> Good catch! Yes, "Width" is a scalar. However, even after fixing this 
> the clouds don't look quite like the image on the website I linked to. 
> The clouds are supposed to be dark underneath, but aren't when I render 
> the scene.

Also, when I do diminish the light source, the clouds still become less 
"dark" the farther away they are in the scene. I might expect this from 
a point light source due to the angle of the light, but my light source 
is parallel so it should be immune.

-Mike

Post a reply to this message

SharkD <mik### [at] gmailcom> wrote:
 The clouds are supposed to be dark underneath, but aren't when I render
> the scene.

> The location of the light source is well above the clouds. But, the
> light source is twice as bright. How might I modify the media to
> compensate? I suppose I could contain the clouds in a light group with a
> dimmer light source, but I'd rather avoid this since it affects things
> like shadows.
>
> -Mike

You can try increasing the extinction value in the scattering block.  By
default, extinction is 1.0, but you can increase it to something higher to
increase the amount of light absorbed by the media without actually thickening
the media.  By doing this, though, your clouds may have a very strong color
contrast with white at the tops and a very dark bottom.  You can control the
contrast of the clouds by balancing the scattering media with absorption and
emission medias.


-Reactor

Post a reply to this message

Reactor wrote:
> You can try increasing the extinction value in the scattering block.  By
> default, extinction is 1.0, but you can increase it to something higher to
> increase the amount of light absorbed by the media without actually thickening
> the media.  By doing this, though, your clouds may have a very strong color
> contrast with white at the tops and a very dark bottom.  You can control the
> contrast of the clouds by balancing the scattering media with absorption and
> emission medias.
> 
> 
> -Reactor

Thanks! Any tips on the increasing brightness problem I described in my 
other post?

-Mike

Post a reply to this message