|
|
> I would like some help getting a cube to look like jelly. I have tried using
> something that looks like glass and lowering the reflectivity but I can't get it
> right. I am after a translucent whiteish jelly cube, I would like to be able to
> partially see through it but for it to be obvious against a background. Could
> you give me some guidance on how to achieve this or if anyone has made jelly or
> a jellyfish etc some sample code.
>
> Thanks for any help
>
>
You got the first part right by playing with the ior that should be
between that of pure water (1.3) and that of glass (1.5).
If you want it to have a coloration that depends on it's thickness, you
need to add "interior pigmentation".
The way to do that is to use fade_color, fade_distance and fade_power.
fade_color is the actual coloration of the interior of the object.
fade_distance controls the distance it takes for the color to affect the
coloration by 50%.
fade_power affect the linearity of the fading. You should use 1 (linear
fading) or 1001 (exponential fading) to get a physicaly correct effect,
but any value can be used.
There are samples scenes that illustrate fade_color. Light will NOT
illuminate that interior.
You can also use some scattering media. It will cause your object's
interior to become foggy. Usefull if you want to see "beams" of light
through your object, or have the interior affect and be affected by
incomming light.
If you use media, the documentation is somewhat misleading by
emphasising a sampling method (method 1) that is NOT the default one
(the default is method 3).
The intervals value should remains at the default value of 1.
You adjust the quality of the sampling using samples alone. samples MUST
be set to at least 3 (default value), in some cases, you may need a
value of 100 or even more. In your case, a value between 3 and 20 may be
correct.
You can use 2 values for samples, but the second one is always ignored.
You can take a look at the madia based sample files that comes with the
distribution.
Alain
Post a reply to this message
|
|