Yves,
Of course you don’t have any need to simulate a CMYK device on your RGB printer.
Then start with a set of “known” Lab targets.
However you want to create the patches is your choice.
You could use InDesign? To generate a bunch of “colored squares”, large enough
for reading with your instrument? Encoded in CIE Lab? InDesign natively
supports CIE Lab space.
That way, you have ultimate control of which Lab colors you want to test the
accuracy of.
Then export to PDF, with no conversion.
Then import into Photoshop, if you prefer to continue the process from within
Photoshop.
Then convert AbsCol using your printer profile, from Lab to RGB.
Print. Measure and compare.
You could also use your original RGB target, assign it your choice of RGB
profile such as sRGB or AdobeRGB, convert AbsCol to your printer profile.
Print. Measure and compare.
/ Roger
From: argyllcms-bounce@xxxxxxxxxxxxx <argyllcms-bounce@xxxxxxxxxxxxx> On Behalf
Of Yves Gauvreau
Sent: April 29, 2022 9:11 AM
To: argyllcms@xxxxxxxxxxxxx
Subject: [argyllcms] Re: Print evaluation?
Roger,
I have no need for "proofing" especially if it means simulating a CYMK printer
on an RGB printer.
The patches they suggest we use are created to be printed on specific papers on
a CMYK printer or press, yes I know we could "proof" that to my RGB printer.
Why go through all this work around instead of going directly to what I want to
evaluate, an RGB printer printing an RGB image on the paper I want as
accurately as possible.
I want to know if it prints a black and white as black and white and as neutral
as the inks and the paper can achieve.
I want to be able to exploit as much of the gamut as is possible with my
papers. All the papers I used can exceed both Adobe RGB and sRGB colorspace in
some areas and yes the inverse is true as well.
Clipping may be acceptable to some, but I found this unacceptable and I prefer
a more elegant and refined solution. I think gamut mapping on an image by image
basis is the best approach.
I think every one will agree, there is no point to test a printer with values
that were used to create the profile. Furthermore, I would also think there is
no point on using out of gamut values either. Then, what should I use to make a
“good” set of patches, suggestions?
Let's say I use the criteria Roger suggest, 1.5 average and maximum 5 DE2000.
Would you suggest something specific to evaluate the neutrality and or the
chromaticity independently?
Any other point I should look for?
/Yves
Le 4/29/2022 à 8:12 AM, graxx@xxxxxxxxxxxx <mailto:graxx@xxxxxxxxxxxx> a écrit
:
IDEAlliance recommends DeltaE 2000 with some specific “ideal” numbers.
You could use the same kind of criteria.
For proofing, for instance, the average of the IT8.7/4 1617 patches has to be
within 1.5 and the max was below 5. I sometimes profile my Epson printer as an
RGB output device and when I use IDEAlliance criteria for proofing, I get the
expected values.
Depends what you want to proof.
/ Roger
From: argyllcms-bounce@xxxxxxxxxxxxx <mailto:argyllcms-bounce@xxxxxxxxxxxxx>
<mailto:argyllcms-bounce@xxxxxxxxxxxxx> <argyllcms-bounce@xxxxxxxxxxxxx> On
Behalf Of Yves Gauvreau
Sent: April 29, 2022 7:10 AM
To: argyllcms@xxxxxxxxxxxxx <mailto:argyllcms@xxxxxxxxxxxxx>
Subject: [argyllcms] Print evaluation?
Hi,
are there any standards to evaluate a RGB printer specifically? Something like
Ideal Alliance certification stuff maybe or whatever but for an RGB printer.
~Yves
