Re: feTurbulence returns "biased" values?

Don't worry, I figured it out myself :) (I forgot about the alpha 
channel...)

Jasper van de Gronde wrote:
> I was looking at the distribution of the color values produced by 
> feTurbulence (type=fractalNoise) and found that they are centered around 
> 0.75 (approximately) instead of 0.5. This seems to be true for Inkscape 
> as well as Batik. I've also checked the implementation in Inkscape 
> against the code in the specification, and it seems to match.
> 
> Every non-SVG example of Perlin noise I've been able to findproduces 
> values centered around 0.5, which seems more logical. Any idea where the 
> discrepancy comes from? Is this by design? Am I missing something?
> 
> 
> 

Received on Monday, 26 April 2010 12:15:05 UTC