[argyllcms] Re: ArgyllCMS: shadow details and blue color issues. What I'm doing incorrectly?
- From: "Xavi" <aruiz@xxxxxxxxxxx>
- To: <argyllcms@xxxxxxxxxxxxx>
- Date: Sun, 14 Oct 2012 02:08:41 +0200
Hello Graeme.
As I commented to you, on my last mail, I have do a profile using the
sRGB_IEC61966-2-1.icc instead of the habitual AdobeRGB.icc, but the blue
color, the yellows, the color skins, are different to these same colors
using the relative colorimetric. The color that changes more is the blue
color.
I have do from zero (new ti1, ti2, TIFF and ti3 files from 396 patches) some
new profiles for a different printer, with a different ink, a MP540 with
Image Specialists ink. Also I have create some new profiles with my main
photographic printer, a Canon Pro9000MKII but witha different ink, OCP ink
instead of Hobbicolors ink. OCP has a more neutral black ink.
The results on both cases are exactly the same that what I commented to you
in my first email.
When I use the relative colorimetric the colors are very very near of colors
that I see on my screen, but the black is not a deeper black and the shadow
details are not showed.
A sample of what I'm trying to explain:
http://i436.photobucket.com/albums/qq82/aaruizz/shadows_zps16bf851d.jpg
On the otrher hand, the shadow details, using the perceptual intent are very
defined, but the colors are not the same than the original image, specially
the blue color. Please, in this sample, see how the blue color changes to a
more light and slightly purple color, the yellow seems more orange color and
the green has a lue tint.
http://i436.photobucket.com/albums/qq82/aaruizz/colors_zps5dd0fb3d.jpg
With another printer and another ink I showed how the perceptual intent
modifies me the colors on softprooft and on prints. Here you can see blue
color and skin color:
http://i436.photobucket.com/albums/qq82/aaruizz/prcptualvsRC_zps2bfd4faf.jpg
Viewing this results, I have created a lot of new profiles, using different
options on the colprof application.
I started with the -q option. I obtain a much deeper black color if I use
the option -qm, instead -qh option or the -qu, but then the gradations are
rough. If I use the option -qh or -qu the black is more a dark grey than a
black.
After, I created more profiles with different options. I tested all the -t
and -T options. I have received errors when I tried to use this options: -t
a -T a -taw -Taw -t al -T al and the application hang up when I used the "T
aa -T r -T la".
All the other -t and -T options worked correctly.
But with any of these I obtaines deeper black colors on relative
colorimetric nor similar colors to rel. Color. Using perceptual intent.
The only option that gives similar blue and yellow, colors using perceptual
intent was the "-t aa" but the profile needs to increase the contrast and
remove brightness but is the only option than shows very similar colors to
the original image and to the rel. Color.
Please Graeme, do you imagine any option to use to increase the contrast of
a profile created with the option -t aa?
Do you know any solution to the poor rendering on shadow details and to the
alteration of the colors when I use the perceptual intent?
If you think that it could be interessant, I can send you the new ti3 files
from the MP540 or from the 9000MKII with the new ink.
To test different options, I am creating profiles with 396 patches,
including 64 grays and 3 white. The problem could be an excess, or few gray
patches? Creating a target of a total of 396 patches, wich is the maximum
number of gray patches you recommend? An the minimum?
Thank you for your appreciate help !
Best Regards,
Xavi.
-----Mensaje original-----
De: Xavi [mailto:aruiz@xxxxxxxxxxx]
Enviado el: viernes, 12 de octubre de 2012 1:53
Para: 'argyllcms@xxxxxxxxxxxxx'
Asunto: RE: [argyllcms] Re: ArgyllCMS: shadow details and blue color issues.
What I'm doing incorrectly?
Hello Graeme.
Thank you, once more, for your reply.
Yes. Some days ago I made a profile using the prophoto.icc instead of the
Adobergb.icc and I see some differences but not important.
Now I am making a profile using the sRGB_IEC61966-2-1.icc, to check if ther
are changes on the intensity of blue color.
Also I have profiled, with your soft, my Canon MP540 and I am letting to dry
the patches printed with a different ink with a very neutral black ink.
Only to test, and to learn, I have made the same profile with the option -q
m, and later with the -q h and -q u option. The results are a surprise for
me. I will show you a comparison with PS CS6 softproof.
I prefer to show you, at the same time, the results of the MP540, the
pro9000MKII with another ink, and the results of creating a profile with
different -q options.
Thank you for your help.
Best Regards,
Xavi.
-----Mensaje original-----
De: argyllcms-bounce@xxxxxxxxxxxxx [mailto:argyllcms-bounce@xxxxxxxxxxxxx]
En nombre de Graeme Gill Enviado el: jueves, 11 de octubre de 2012 1:02
Para: argyllcms@xxxxxxxxxxxxx
Asunto: [argyllcms] Re: ArgyllCMS: shadow details and blue color issues.
What I'm doing incorrectly?
Graeme Gill wrote:
> Note though that it depends to some degree what you are looking for.
> The nature of a print device is that there is a severe trade-off
> between saturation and lightness in blue. The Perceptual intent (by
> the tuning I've chosen) is lighter and less saturated than the
> clipping of the colorimetric intent, and (to my taste) is visually
> closer to the very intense and light color of the source AdobeRGB blue
> (Helmholtz-Kohlraush effect). The colorimetric reproduction is certainly
more saturated, but is overly dark in my opinion.
Note another difference between these intents: The perceptual intent is
gamut mapped for Adobe1998. This means that it is allowing space for a pure
Adobe1998 blue to be mapped within the gamut of the output device. I suspect
that the profile you are comparing it with is not actually doing source to
destination gamut mapping for its perceptual intent, and therefore is not
compressing the saturation of the blue sufficiently to accommodate a pure
blue input, and hence for less saturated input values, is producing a more
saturated output.
Graeme Gill.
Other related posts:
