[argyllcms] Re: Colprof feature request: amount of black pointhue correction control

Nikolay Pokhilchenko wrote:
> Take a look to .ti3 attached.

I find for instance this huge divergence "interesting":

$ colprof -v -qh -S sRGB.icm LSM2484

$ xicclu -s255 -fif -ir LSM2484.icc
0 0 0
0.000000 0.000000 0.000000 [Lab] -> Lut -> 0.000000 0.000000 0.000000
[RGB] (clip)

$ xicclu -s255 -fb -ir LSM2484.icc
0 0 0
0.000000 0.000000 0.000000 [Lab] -> Lut -> 30.935831 23.114864 0.000001
[RGB]

$ xicclu -s255 -ff -ir LSM2484.icc
0 0 0
0.000000 0.000000 0.000000 [RGB] -> Lut -> 12.068596 -0.064363 -8.293796
[Lab]
30.935831 23.114864 0.000001
30.935831 23.114864 0.000001 [RGB] -> Lut -> 19.881831 0.965791
-1.891457 [Lab]

=> that's a difference of about 10 delta E !


[Btw, looking at the measurements, I find this RGB printer driver not
too well-behaved in dark regions - possibly the generation of CMYK
separations from RGB is not optimally tuned for the used inks/paper. It
seems to put down too much cyan and magenta ink (and/or not enough
yellow ink) for RGB=[0,0,0]. Just wondering, is this the driver provided
by the printer vendor, using original inks/paper from the printer
vendor? Did you tweak any driver settings? (e.g. increase cyan/magenta
density)]

Regards,
Gerhard

Other related posts: