In article <3a3a48b2@news.povray.org>, "Rune" <run### [at] inamecom> 
wrote:

> But I think in reality fire doesn't absorb light (almost not anyway). The
> reason it is still visible is that it is many times brighter than the
> surrounding objects.

Actually, it does absorb and scatter light, sometimes to a very 
significant degree. Especially yellow flames, such as a candle flame or 
wood fire, where there are many incandescent particles of carbon and 
other smoke particles, and a lot of complex organic substances. It might 
be an interesting experiment to shine laser beams through various types 
of flames to measure the effects...
However, you are right that it usually just overwhelms the background 
more than blocking it. Because of the limited, linear dynamic range, POV 
will just make extremely bright flames look white.


> I think if POV-Ray showed light intensities based on for example an 
> logarithmic function rather than linearly it would be possible to 
> make more realistic scenes, and it would be easier too. I think it 
> would affect more things than you would think; perhaps completely 
> change the way we think of colors in POV-Ray.

This is possible to do as a post_process filter(*), in fact, the 
"curves" filter could be used with a "raw_color" filter to do 
this...then you could set up an include file with macros to imitate the 
response curves of various film types or the human eye. Or maybe a 
"limit_range" post_process that scales colors with components >1 down so 
their brightest component is equal to one...not exactly realistic, but 
maybe useful.

*the post_process patch is in MegaPOV, I have released the code for the 
curves filter in these groups, and the raw_color and limit_range filters 
aren't implemented yet, though they will be easy to do.

-- 
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/

<><

Post a reply to this message