[argyllcms] Aw: Re: RGB black is not clean

  • From: 0815xl@xxxxxx
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Thu, 21 Oct 2021 19:51:27 +0200

  Ok, I have now try two RGB profiles with only 100 fields. Once standard printcal and once -xm95 -xc95. Standard has in photoshop a black with 3 2 -3 and the other 7 8 -3. That it is getting brighter was clear, but it has become less neutral.   Gesendet: Montag, 18. Oktober 2021 um 21:44 Uhr
Von: 0815xl@xxxxxx
An: argyllcms@xxxxxxxxxxxxx
Betreff: [argyllcms] Aw: Re: RGB black is not clean Hello and thanks for your reply!   I test now printcal with -xm 95 -xc 95 and these curves cut at 95%. So it look like it works. I think its better to make a small profile with 100 patches for testing the result on black?   You mean the raw raster values sent to the printer ?   I´m not sure. I just convert sRGB black in photoshop absolut colorimetric to in my printer colorspace an read the value with the pipette.   Thanks and best regards, Roman     Gesendet: Montag, 18. Oktober 2021 um 06:47 Uhr
Von: "Graeme Gill" <graeme@xxxxxxxxxxxxx>
An: argyllcms@xxxxxxxxxxxxx
Betreff: [argyllcms] Re: RGB black is not clean 0815xl@xxxxxx wrote:

Hi,

> my cmyk profile has a neutral black in the aF rendering intent (7,5 -0,6 -0,3).
> Unfortunately not the RGB profile, its black (aF) is deeper but not neutral (4,6
> 3,8 -4,0).
> Would more yellow fix the error? What can I do to fix it?

if you are using printcal to create the linearization, then take
a look at the maximum device target values you are using.
(Actually I can't remember if it' s the maximum or minimum values
you should try changing for a print RGB space. You might
have to experiment and find out.)
You can probably change them to adjust the black point hue
to some degree, but usually you will want to keep one
channel at 100% while reducing one or two of the others.
There is a tradeoff between the L* value and how neutral it will be.

> I have now viewed black from cmyk and rgb profile in photoshop.

You mean the raw raster values sent to the printer ?

> If I convert sRGB into my printer color space absolute colorimetric. I get 5 3 -4
> (RGB) and 7 0 -2 (CMYK).
> You can clearly see that cmyk is better/more neutral.

But not quite as dark.

> With RGB there is still plenty of reserve, why it cannot reach 7 0 -2?

It depends on what the linearization is doing to your values, and
how you printer is converting from RGB to CMYK.

Cheers,
Graeme Gill.

 

Other related posts: