[Openicc] review of new wide gamut editing space

[Chris Murphy]

On Mar 22, 2012, at 5:14 AM, Kai-Uwe Behrmann wrote:
> 
> The 32+bpc float support group is my home since many years :-)
> However, at some point we have to satisfy window managers and toolkits, which resist on 8-bpc pipelines. Lets see which approach fits them best.

An 8bpc pipeline that will be compositing in 8bpc? 

I guess you'd need some discerning eyes and perhaps some excessive compositing in 8bpc ECI RGB to see if it's a problem solved by switching to sRGB. The 1.8 gamma function defined TRC is better for 8bpc invertibility, I think. That could outweigh the drawback of extra gamut with the lower bitdepth.


> Meanwhile, I will drop creating a new colour space. Either digikam people find a existing distributable ProPhotoRGB profile or we will build one after the spec for our photo communities.

I will be happy to eat crow, while not wiping the egg off my face, if someone demonstrates device and image encoding TRC of L* has any efficacy, let alone meaningful efficacy in the real world.

Chris Murphy