[argyllcms] Re: ArgyllCMS: shadow details and blue color issues. What I'm doing incorrectly?

  • From: "Xavi" <aruiz@xxxxxxxxxxx>
  • To: <argyllcms@xxxxxxxxxxxxx>
  • Date: Tue, 16 Oct 2012 14:17:01 +0200

(Sorry for the recent incorrectly sent email, due to alteration of the
subject, if it has arrived)

Dear Graeme.

I enjoy reading your replies although I discover that my knowledge of color
management is very basic.

All the images that I use to test the printer profiles, are mainly labelled
with the AdobeRGB and the others with the sRGB color spaces. Then, seems
that this is not the reason of color change.

As I commented to you,  I have changed all the cartridges of my Pro9000 to
use another brand of ink, to have more chances to discover what is
happening. Then I can't use at this moment, the printer with the ink of the
example of ICM that you used in your icclu example

I must to use the MP540 printer with the
_ACPU.icm". Sorry to the extent name, but I have created a lot of profiles
and the name was growing trying to identify them.

I have do the same example, than you show me, using this ICM:

.\icclu -fb -ip
50.000000 0.000000 0.000000 [Lab] -> Lut -> 0.563818 0.500477 0.630465 [RGB]

.\icclu -ff -ir
0.563818 0.500477 0.630465
0.563818 0.500477 0.630465 [RGB] -> Lut -> 52.109114 0.326121 -0.977612

I don't know if the reading with an spectro some patches, is a good way to
show you what I obtain on prints but I have print a simulation of the
patches of a colorchecker card and I have read the results with a spectro.

I give you the link of a RAR file that contains a XLS file with a table with
the results (also included the Icm and the ti3 files). 
I have read only 7 colors randomly but if you think that's a good idea to
read the 24 patches, I will do a table with all the colors of the
ColorChecker card and their deltaE.

All these patches are printed without the BPC option, as you explained ( I
understand that ALWAYS I must to print without BPC when I use profiles made
with ArgyllCMS). The ICM was created with the option -qm.

The first group of values are the "real" Lab values of the color checker
card. The second group are the Lab values printing with the relative
colorimetric intent and the third group are the Lab values when I print
using the Perceptual intent. You will see a column, on the right of each
group, that's the calculated deltaE 76. The two prints were made using the

The average deltaE value, of the 7 patches, when I print with the relative
colorimetric intent is 6,12 with a peak of 8. The average value deltaE
value, when i print using the perceptual intent is 10,70 with a peak of 15.

Applyiing the icclu example that you show me to the values of blue color

.\icclu -fb -ip
28.87   14.81   -50.15
28.870000 14.810000 -50.150000 [Lab] -> Lut -> 0.358232 0.399245 0.879326

.\icclu -ff -ir
0.358232 0.399245 0.879326
0.358232 0.399245 0.879326 [RGB] -> Lut -> 39.085966 11.557139 -49.964686

I must obtain, on paper and with perceptual intent, a patch with 39.09 11.56
-49.96 Lab values, but I read a patch with 39,35        4,45    -48,45 Lab
values. That's a deltaE76 of 11,77.

I don't know if these are acceptable values on prints, but seems to show
very big deltaE values. But, as commented before, I don't know if the method
applied to check the final results, and to verify the ICM profile, has sense
or not.

Graeme, excuse that I disturb you but  I don't know another way to tray to
understand what is happening. The only way that I am trying to solve it is,
trying different parameters on colprof, print patches with different media
settings, read them and for the moment, only time, paper and ink wasted.

Thank you, once more for your help.

Best Regards,

-----Mensaje original-----
De: argyllcms-bounce@xxxxxxxxxxxxx [mailto:argyllcms-bounce@xxxxxxxxxxxxx]
En nombre de Graeme Gill Enviado el: lunes, 15 de octubre de 2012 8:48
Para: argyllcms@xxxxxxxxxxxxx
Asunto: [argyllcms] Re: ArgyllCMS: shadow details and blue color issues.
What I'm doing incorrectly?

Xavi wrote:
> On my case the softproof and the print are very very similar. On 
> prints, using perceptual intent, I see the blue color as a light 
> purple color, as I see on the screen when I softproof with Photoshop 
> CS6. Also checked on prints with 2 models of spectros using spot read.

        I'm guessing that the image you are using is from the DQ-Tool
Monitor Reference File by Fujifilm Europe GmbH / Jens Rubb.
If so, it's an unlabelled (ie. no colorspace) RGB file.
The color of the "blue" fan bland is something like .09, .14, .55 RGB, so it
is not a pure RGB blue, but is tending a little towards purple in the
original, and all four of the soft proofs hint at this.

> Also the overall color is changed to a magenta cast, when I use 
> perceptual intent. Using the saturation intent the blue color is less 
> purple color and is more vivid but skin colors, for example, are more 
> reddish and excessive saturated.

I can't verify this - the profile through perceptual seems very self
consistent and neutral, ie:

icclu -fb -ip Argyllv140_396parxes48Gris3Blanc.icm
50.000000 0.000000 0.000000 [Lab] -> Lut -> 0.345472 0.388520 0.513978 [RGB]

icclu -ff -ir Argyllv140_396parxes48Gris3Blanc.icm
0.345472 0.388520 0.513978 [RGB] -> Lut -> 52.334969 -0.072152 -0.770024

< 1 delta E hue & saturation error seems reasonable through a perceptual B2A

It tends towards cyan as it nears the black point, due to Argyll's "extend
and bend" neutral mapping.

> I have created a collage with 3 cropped photos. Please, tell me if you 
> see the reddish/magenta cast on color skin. And also, if you agree 
> with me that the blue color of the sky, seems a blue/purple very light and
not saturated:
> http://i436.photobucket.com/albums/qq82/aaruizz/prcptualvsRC_zps09f2fe
> 42.jpg
> On the other hand I see very natural colors, on the right image, 
> softproofed using the same profile, but with the colorimetric intent.

You soft proof does look somewhat reddish - but then I'm not sure how your
softproof workflow is functioning. For instance, you are using BPC which
could be having an effect on the neutrals, given the very non-neutral black
point of your colorspace - ie. Argyll maintains neutrality down the neutral
axis until it nears the black point, at which stage it becomes cyan tint, to
match the black point. If the BPC then maps the cyan black to neutral with
the mapping being in a straight line to the white point, then the neutral
mid tones will get shoved in a megenta direction.

Note that typical Argyll workflows won't use BPC and are not created to
anticipate the effects of applying BPC, because you can do a more
sophisticated mapping using the collink gamut mapping machinery. My typical
softproof workflow is as follows:

# Create proofing transform from output to AdobeRGB collink -v -ila -G -qh
-cpp -dmt Argyllv140_396parxes48Gris3Blanc.icm AdobeRGB1998.icm

# Create proof image back in AdobeRGB space cctiff
Argyllv140_396parxes48Gris3Blanc_v.icm skin_sky_p.jpg skin_sky_pv.jpg

The above softproof (not using BPC) doesn't show such a tint.

> I am doing all the profiles with a Canon Pro9000MKII with third party 
> inks (Hobbicolors UW). I will do,  from zero, new profiles for another 
> Canon MP540, that uses only four different color cartridges (CMYK) and 
> also I will start creating profiles, from zero, for the Pro9000MKII 
> but using cartridges refilled with other third party ink (OCP), that 
> uses a very neutral black color but that their gamut is more limited 
> vs Hobbicolors Ultra Wide Gamut inks.

I think a neutral black color will give better visual results, and leave
less room for workflow details to affect the outputt.

Graeme Gill.

Other related posts: