On Fri, Mar 1, 2013 at 5:57 AM, Nicolas Robidoux <nicolas.robidoux@gmail.com > wrote: > 1) It would appear that regardless of whether the gradient is >>computed<< > in pre- or non-pre, the >>specification<< of the starting, ending and stop > colors should be in non-pre. > yes, that is the case. Colors are always specified in non-premultiplied. > > 2) Maybe there should be some way to specify whether the gradient should > be computed in linear light, in perceptual light, or taking the current > color space as is. > yes. It would be nice if we could specify that gradients are interpolated in an Lab (or other non-sRGB) colorspace. > > No need to correct me. I know that the probability that I am missing > something (including "everybody knows this") is high. > No worries. People in general don't know how gradients and especially colorspaces work so it's good that you bring this up.
Received on Friday, 1 March 2013 17:38:59 UTC