How to get neutral B&W with Colormunki + ArgyllCMS

[Ink stained Fingers]

the image is in .jpg format, and the compression algorithm groups pixels into one color which may take away some of the gray levels you are apparently missing. I would recommend to do a test with a gray scale patch, a continous one, in .tiff format without compression artifacts and check to which gray level you can differentiate them in the printout and to look for some banding in the dark gray area which may point to a profile inconsistency, and to do that check of a printout with/without black point compensation and rel. col. rendering intent.

 

[RogerB]

the image is in .jpg format, and the compression algorithm groups pixels into one color which may take away some of the gray levels you are apparently missing. I would recommend to do a test with a gray scale patch, a continous one, in .tiff format without compression artifacts and check to which gray level you can differentiate them in the printout and to look for some banding in the dark gray area which may point to a profile inconsistency, and to do that check of a printout with/without black point compensation and rel. col. rendering intent.

The jpeg image seems to have rasonable gradation in the region that the OP showed, but the L- values are very low - between 1 and about 6. This is obviously a difficult tonal range to reproduce without a good profile and with BPC. I agree that the OP should use a greyscale ramp to examine the response in this region, with different rendering intents. A ramp that runs from 30, 30, 30 to 0, 0, 0 (RGB values) would be good for this.

 

[FLMK]

Hi

I use an Epson Premium SemiGloss Photo Paper and Sudhaus ink. My printer is a Canon IX6540 and I profiled my printer with a Colormunki Photo.

I took this chart to test my shade of grey:

Here is the result.

For me, my profile is "reasonable"... There's no break in the shades of greys.

When I print in photoshop, I use black point compensation... May the problem be here?

Thanks for your help.

 

[Ink stained Fingers]

There is no problem with black point compensation, it's your decision what you do with dark pixels which are below the black point of the paper or close to it - either truncating them into paper black without any differentiation or using the smoothed S-kurve at the bottom end.
There are quite some articles about BPC in the internet:

http://www.gamutvision.com/docs/blackpoint.html
and with more theory behind it

http://www.color.org/WP40-Black_Point_Compensation_2010-07-27.pdf
https://a248.e.akamai.net/f/1953/89...am/Adobe/en/devnet/photoshop/sdk/AdobeBPC.pdf

But the essence is that the profile cannot differentiate very dark tones, close to the black point of the paper, as good as in the middle part. If you want to see more details at the dark end you would need to enhance them e.g. with a HDR like increase of the local contrast before print. You have to work against the shallow tone curve there.

For me, my profile is "reasonable"... There's no break in the shades of greys.

Yes, there is no obvious flaw in the dark grays

 

[FLMK]

Too bad...

I have an other question.

Do you know a free software that show the gammut in 3D of a profile (for windows)?

Thanks in advance.

Bye.

 

[The Hat]

@FLMK, if you profiled your inks before making this halftone chart then clearly something is very wrong, because the profile is way off.

The original Kodak chart is 4 times smaller than your posted chart and when your tone chart is reduced and put together with the original chart, your black is completely green, what happened to your profile ?

 

[Ink stained Fingers]

you can try iccview.de, an online ICC profile viewer. If you want to know more about profile assessment you may look to the gamutvision program

 

[FLMK]

The Hat: It's 4 times bigger because I scanned it in 800 dpi (to see every details)...
My chart isn't as green as what you see... My scanner has changed a little bit my chart.
But to be fair, it looks like it's not totally neutral... But it's subjective.
I think it's because of the artificial light of my room...
I've printed the wolf picture in black and white (with a chocolate turn), and it's OK. I don't have a green tint...

 

[The Hat]

OK I can understand that but you should have also scanned the Kodak chart at the same time that would have been the best way of comparing both uploaded halftones.

This is what I see..

 

[FLMK]

I don't have the kodak chart...
I've downloaded it on the web...
So, it will be difficult to scan the original chart...
The aim of my scan was to show all grey shades... and to prove that I don't have breaks in the shades.


The funny part of the story is that sometimes, when I look my printed B&W kodak chart, it's a little bit greenish, and sometimes, it's not.

I think it depends of the light or something like that (or what I looked at before)...

But I can say that my printed B&W pictures are not greenish...