[Openicc] L* was: MS on HDR CMS :)

[Chris Murphy]

On Feb 15, 2008, at 4:26 AM, Jan-Peter Homann wrote:
> Assumptions about the design of colormanagement workflows need  
> always to be tested in practice. The idea behind L* workingspace  
> and monitor calibration is to use the same lightness scaling as in  
> the profile connection space L*a*b*.

Why is that even relevant? Why is it important? Why does it matter?  
It has no correlation to how our devices behave.

> To analyze the differences from L* to Gamma 1,8 and 2,2 just do  
> following tests:
>
> 1) Convert a scale with 256 steps of L*a*b* gray to RGB- 
> workingspaces with Gamma of 1,8 and 2,2 and look about the  
> resulting steps in RGB
> 2) Convert a scale with 256 steps of RGB Gamma 1,8 and 2,2 to  
> L*a*b*and  look about the resulting steps in L*a*b*

Yes, and? The point of color management is that as a user I don't  
need to care about ambiguous RGB values. Am I seeing on-screen what I  
want and is that being translated effectively to print? That's what I  
care about. I care about results.

> As the ICC colormanagement is based on L*a*b* as profile connection  
> space, I think it is good strategy just to test if a L* RGB  
> workingspace and monitor calibration has some advantages.

Certainly we should work on improving workflow. Certainly we should  
test. Testing should have occurred before ECI-RGB was changed, a  
final release was produced, and it was made a new ISO standard color  
space. If we're still testing the concept let alone the released  
color space, the process has been done exactly ass backwards.

I think there is an ECI list, I should probably move this discussion  
over there as it appears to be of limited interest here.

Chris Murphy