Colormunki/ArgyllCMS profile lacking dark tones (compared to ccStudio)

nertog

Print Addict
Joined
Jun 30, 2009
Messages
156
Reaction score
82
Points
161
Location
Regensburg, Germany
Printer Model
Epson R3000 - PX720 - P50
Do you have any documented evidence for that ?
No, I cannot make a comparison with commercial software from X-rite or Datacolor as I do not own either. What I do see is that Argyll creates colorimetric profiles which give you the darkest neutral black possible with a particular printer-ink-paper combo. Deviating from neutral (and thus from the "rules" of the colorimetric rendering intent) might result in darker blacks, of course.

Now, with the perceptual rendering intent all bets are off as there is no well described standard to follow.

@OP: have you tried using the -S flag with a small percentage of gamut compression (1% or so) instead of a reference ICC profile? This might give you deeper blacks without hurting the overall look of the image too much...
 

Ink stained Fingers

Printer VIP
Platinum Printer Member
Joined
Dec 27, 2014
Messages
5,856
Reaction score
6,982
Points
363
Location
Germany
Printer Model
L805, WF2010, ET8550
The black level as stored in the profile data is the actual value as measured during the scanning process, it's a kind of raw data, and this black is not neutral but depends on the ink/paper combination actually measured.
The neutral black is the point where the vertical neutral gray/L* axis (a=b=0) cuts through the surface of the gamut volume as shown in this crop of a profile display

Blackpoint.png

and this neutral black point is typically higher than the abs. black point - the lowest tip of the gamut volume. Lots of users don't differentiate between these different blackpoint types. Various profiling programs let you adjust the actual value for the blackpoint - somewhere between the actual and the neutral blackpoint. This does not just affects the neutral blackpoint but the overall position of the (neutral) gray axis. I don't see ArgyllCMS to offer such blackpoint adjustment.
And to summarize all this you either use ArgyllCMS as is or use a profiling package which does it differently.
 

lmylm

Printing Apprentice
Joined
Feb 17, 2024
Messages
14
Reaction score
9
Points
10
Printer Model
Inkjet
@nertog

have you tried using the -S flag with a small percentage of gamut compression (1% or so) instead of a reference ICC profile? This might give you deeper blacks without hurting the overall look of the image too much...

The colprof documentation shows:
Code:
-s src.icm|cperc   Apply gamut mapping      to output profile perceptual B2A table for given source, or      compression percentage
-S src.icm|experc  Apply gamut mapping      to output profile perceptual and saturation B2A table, or      expansion percentage

Does that mean for gamut compression (rather than expansion) I would need to use -s (rather than -S)? In any case: how do I use this parameter with a percentage? Like “-s1” for 1%? I did play around with this parameter before but wasn’t able to see a difference in the profile using macOS’s ColorSync Utility. So I assumed I used it the wrong way but couldn’t find out how to use it correctly so far.

@Ink stained Fingers

Thank you. This “lifted” black point is what I referred to in my original post.
Screenshot 2024-02-27 at 20.59.28.png

I assumed there were settings/parameters/options for ArgyllCMS that can “drag” the black point down so the profile would become usable. I think that is the adjustment you are talking about. I really wonder how it is possible that this doesn’t exist and still everyone seems to be happy with ArgyllCMS.
 

nertog

Print Addict
Joined
Jun 30, 2009
Messages
156
Reaction score
82
Points
161
Location
Regensburg, Germany
Printer Model
Epson R3000 - PX720 - P50
@nertog



The colprof documentation shows:
Code:
-s src.icm|cperc   Apply gamut mapping      to output profile perceptual B2A table for given source, or      compression percentage
-S src.icm|experc  Apply gamut mapping      to output profile perceptual and saturation B2A table, or      expansion percentage

Does that mean for gamut compression (rather than expansion) I would need to use -s (rather than -S)? In any case: how do I use this parameter with a percentage? Like “-s1” for 1%? I did play around with this parameter before but wasn’t able to see a difference in the profile using macOS’s ColorSync Utility. So I assumed I used it the wrong way but couldn’t find out how to use it correctly so far.
This is likely a typo, both options should do gamut compression. -S1 is correct and works on both the perceptual and saturation table. The gamut in Colorsync is a plot of your measured patches and will not change with different settings in Colprof.

To check the actual behavior of your ICC profile you can use Gamutvision (free) and have a look at the RGB and simulated neutral density curves, just like @Ink stained Fingers did in some of the earlier posts in this thread.

View attachment 15972
I assumed there were settings/parameters/options for ArgyllCMS that can “drag” the black point down so the profile would become usable. I think that is the adjustment you are talking about. I really wonder how it is possible that this doesn’t exist and still everyone seems to be happy with ArgyllCMS.
The ccStudio plot makes no sense. It shows a combo that can reach absolute black on paper (L=0). Those gamut plots should faithfully represent the patches measured by your spectro... and you can check this: does the Lab-value of the darkest point in the plot correspond to the black patches you´ve printed on the target sheet (Spotread to verify)?
 

Ink stained Fingers

Printer VIP
Platinum Printer Member
Joined
Dec 27, 2014
Messages
5,856
Reaction score
6,982
Points
363
Location
Germany
Printer Model
L805, WF2010, ET8550
I think lthis case is closed for me at this point, we have discussed the differences in quite some detail, and it is now up to you - @lmylm - to make it working for you and what you expect from a profile - within the constraints of the available programs.
 

lmylm

Printing Apprentice
Joined
Feb 17, 2024
Messages
14
Reaction score
9
Points
10
Printer Model
Inkjet
@nertog
The ccStudio plot makes no sense. It shows a combo that can reach absolute black on paper (L=0). Those gamut plots should faithfully represent the patches measured by your spectro... and you can check this: does the Lab-value of the darkest point in the plot correspond to the black patches you´ve printed on the target sheet (Spotread to verify)
I can see how the ccStudio plot doesn’t make sense for showing the displayable gamut of my printer/paper combination. Obviously it can not print absolute black. Yet I get darker blacks using the ccStudio profile.
I create profiles not to find out about the paper gamut but to get nice looking prints. For this I seem to have to use ccStudio.
Apparently it is not possible to reproduce a similar behavior to ccStudio with ArgyllCMS (stretching the black point to absolute black, which seems to make the printer print darker darks).
Because of this I will have to use ccStudio with all its downsides (waste more paper, not being able to share the profiles). At least I get correct prints.
Thank you so much for your help anyways. I will still have to check out the -S/-s parameters. Maybe this does some magic. 🙏😊


@Ink stained Fingers
I think lthis case is closed for me at this point, we have discussed the differences in quite some detail, and it is now up to you - @lmylm - to make it working for you and what you expect from a profile - within the constraints of the available programs.
Thank you so much for all your help. I really appreciate the insight I got from your messages. Obviously with the gained knowledge I still only skim the surface and there are still some areas in this topic that confuse me. I will dig deeper though, my profiling journey has only just begun!
In the meantime I got a Windows environment running and installed Gamutvision which will help me a lot completing my knowledge to my satisfactory. I will start by finding out what the “simulated neutral density curves” @nertog mentioned are about.
Thanks again @Ink stained Fingers! 🙂

For now I feel like what I wish for doesn’t exist. As soon as I find more energy to dive deeper, I might have more questions. And who knows, maybe one day I can contribute my own PrinterKnowledge. Thank you all for being so supportive!
Also, thank you @pharmacist 😁
 
Last edited:

Graeme Gill

Getting Fingers Dirty
Joined
Nov 13, 2014
Messages
29
Reaction score
30
Points
48
Location
Melbourne Australia
Printer Model
Epson R1800
I don't see ArgyllCMS to offer such blackpoint adjustment.
Except that it provides exactly that for perceptual rendering. Setup your perceptual table for your source colorspace, and it will attempt to map your black to the lowest possible neutralish black that the printer provides.

(If I remember correctly) the previous discussion and investigation revealed that some software gets confused when it attempts to do its own black point mapping, making a very poor job of it. So you need to make sure that such BPC is turned off if you want to get the best out of the the ArgyllCMS perceptual gamut mapping.
 

Ink stained Fingers

Printer VIP
Platinum Printer Member
Joined
Dec 27, 2014
Messages
5,856
Reaction score
6,982
Points
363
Location
Germany
Printer Model
L805, WF2010, ET8550
So you need to make sure that such BPC is turned off if you want to get the best out of the the ArgyllCMS perceptual gamut mapping.
There is a problem for quite a number of users of the popular Lightroom package - BPC is always on in Lightroom for the perceptual rendering intent , you find quite some postings to this problem in luminous-landscape.com or dpreview.com and at plenty other postings like here

https://www.colourphil.co.uk/lightroom-cc-print.shtml
copied from here in the 'Rendering Intent' section

'Black Point Compensation: In Lightroom it can't be deselected, as it's always 'ON', which is highly desirable.'

Or here
https://forum.luminous-landscape.com/index.php?topic=97346.0


Or here

https://www.dpreview.com/forums/post/64887619


Adobe is describing the BPC functionality here

https://www.color.org/adobebpc.pdf

and BPC in connection with the rendering intents in section 6.2

but this does not explain in any detail how this is implemented in actual Adobe software
 
Last edited:
Top