I think I've been coming to this conclusion for a while.
Currently, we provide a fancy spline that calculates the actual achromatic response, based on the original white point, but many of these spaces are actually converting colors based on an adapting luminance and background luminance. The achromatic response where chroma is zero doesn't look the same as when you discount illuminance, which is exactly what you expect.
I'm thinking that all the fancy overhead to preserve this true achromatic line is wasted effort. Much like RYB where we return its achromatic colors, maybe we should do the same for all these spaces. It simplifies the code, probably makes it faster, and probably gives results not too different than what people would expect. I imagine most probably can't tell that HCT white is slightly less white due to accounting for adapting luminance. It also makes it much easier for people to quickly crate a variant and change the background and adapting luminance.
I think I've been coming to this conclusion for a while.
Currently, we provide a fancy spline that calculates the actual achromatic response, based on the original white point, but many of these spaces are actually converting colors based on an adapting luminance and background luminance. The achromatic response where chroma is zero doesn't look the same as when you discount illuminance, which is exactly what you expect.
I'm thinking that all the fancy overhead to preserve this true achromatic line is wasted effort. Much like RYB where we return its achromatic colors, maybe we should do the same for all these spaces. It simplifies the code, probably makes it faster, and probably gives results not too different than what people would expect. I imagine most probably can't tell that HCT white is slightly less white due to accounting for adapting luminance. It also makes it much easier for people to quickly crate a variant and change the background and adapting luminance.