clipka nous illumina en ce 2009-04-01 04:52 -->
> Ive <"ive### [at] lilysoftorg"> wrote:
>> For "radiosity-lit-only" scenes (or MCPov for that matter) it would even
>> mean "specular 0" and "phong 0".
> 
> That's not necessary, as without classic light sources these values don't have
> any effect anyway (well, maybe a *tiny* bit more computing time).
> 
> They should definitely be non-zero for materials to work in non-radiosity-lit
> scenes, as they will have to simulate highlights from the classic lights, which
> we can't see otherwise.
> 
> 
> Ideally, there should be *some* connection between these values and the
> reflectivity of the material (and likewise between the roughness or
> corresponding phong parameter, and the blurriness of reflectivity in case
> micronormals are used), though I'm not sure how exactly these should be
> related.
> 
> 
Any bluring will have an effect on the tightness of the highlights. The tighter 
the highlight, the more visible the effect.

Low reflectivity goes with weak highlights. Ether use a multiple of the 
reflection value or something like the square root of the reflection value. The 
highlights tend to look as if it increase faster than the perceived reflection 
for low reflection.

-- 
Alain
-------------------------------------------------
You know you've been raytracing too long when you have ever gotten in a flame 
war over various rendering softwares.
Stephan Ahonen

Post a reply to this message