On Wednesday 24 September 2008, Chris Moller wrote:
> Maciej Pilichowski wrote:
> > So there is obstacle, indeed. And till now I have no idea how to solve
> > this.
> Might be a bit of a wild branch on this topic, but I'm wondering if all
> the effects described here could be implemented as a /subtractive/ process.
> I haven't looked at gimp's implementation, but generally speaking colors
> combine as Cr = Ce * ( 1- o) + Cb * o, where Cr is the resultant color
> (clamped full saturation of each of it's components), Ce is the existing
> color, Cb is the brush color, and a is the opacity.  What if, instead,
> you made Cr = Ce * (1  - o)  - Cb * o, with Cr clamped at 0?  All this
> doesn't fiddle with the alpha channel at all, but might get the desired
> effect.

Isn't that pretty much what the 'subtract' mode does?  And yes, it doesn't 
involve the alpha channel.


Attachment: signature.asc
Description: This is a digitally signed message part.

Gimp-developer mailing list

Reply via email to