"Cousin Ricky" <ric### [at] yahoocom> wrote:
> On Tue, 24 Feb 2009 06:16:06 -0400, clipka <nomail@nomail> wrote:
> > There are times when I wish for a more realistic (viz: more "channels")
> > color
> > model in POV...
> >*sigh*
>
> I get that longing every time I use fade_color or absorption.  Unless the
> channels are balanced (not a realistic condition), these effects will
> approach the most dominant primary color.
>
> I first noticed this in simulating sea water: in the shallows, the color
> looked surprisingly realistic, but as it got deeper, the sea turned a rich
> phosphor blue.  I later noticed a similar problem with colored grass:  as
> I attempted to saturate some colors that were washed-out, but of the
> desired hue, golden-yellow became phosphor vermillion, and chartreuse
> became phosphor green.
>
> The capability of extra channels, or even a continuous spectrum, would be
> welcome. However, I don't think this feature would be practical beyond a
> few applications, such as transparent materials.
>
> I created the attached images a few years ago to get a feel for the
> problem, although I didn't feel that anything could be easily done about
> it.  I figured that when it became important enough for me, I'd poll here
> for ideas.  The images use media { absorption } and fade_color,
> respectively.
>

I recently came up with a method (actually two) that takes an rgb color and
progressively changes its color saturation anywhere between 100% and 0%.  Based
on POV's gray formula found in section "3.3.3.4  HF_Gray_16" in the docs. I
don't know if it has any relevance to the situations mentioned here, or how it
might be 'variably' applied across z-space, but I should probably post it
anyway. (Can ground fog take a color_map or pigment_map to change its color
across z-space?  I've never tried that.)

Ken W.

Post a reply to this message