1-Gamut coverage. Gamut "edges" will set a minimum dE for each colorspace to simulate. Usually for wide gamuts most of colorspace edges to simulate are not printable so it does not matter. For example AdobeRGB green is not printable, but cyan should be printable.
2-Profile accuracy. It means how good profile is capturing actual monitor behaviour.
That second one "could" be improved with a Profile only with DispcalGUI. Read its documentation for "profile only" (without calibrating) tasks. Even a DCCS "table profile" 400 patches should be more accurate than a DCCS matrix one.
1- Results between 95-99% are OK. 99% AdobeRGB intersection is for the panel in its native state.
2- Do not think in % terms because it is useless. Run a comparison between AdobeRGB and your profiles, there should (WILL) be a very small dE between profile green 0,255,0 and AdobeRGB green, so it's fine. Your eyes do not see % coverages but color differences (deltaE2000)
Your reports are very good and DCCS gives you a very good neutral grey (better than my unit, but my AMD GPU fix that issue) and expected "after DCCS" contrast, so it's all OK
Yes, validate profile against a reference ("measurement report"). Two ways:
-validate calibration against AdobeRGB profile, not useful since if you need AdobeRGB support you need color managed app... so it's useless unless you need to validate grey ramp calibration in a fast way.
-validate profile simulating AdobeRGB, that means how that profile will perform in color managed apps showing AdobeRGB images.
Be careful with gamma simulation options, a true 2.2 power gamma means infinite contrast (0cd/m2 brightness in black). Read DispcalGUI documentation.
That setup with "use smilation profile as target" is "CASE 1" in my previous post. Without that checkbox is "CASE 2", which is the color managed approach.
You used to check against a true 2.2 power gamma that your monitor CANNOT do, because it has not infinite contrast. I warned you prevoiusly against doing that... Select unmodified.
yumichan
3 Apprentice
•
738 Posts
1
October 7th, 2015 02:00
Color managed performance needs two things:
1-Gamut coverage. Gamut "edges" will set a minimum dE for each colorspace to simulate. Usually for wide gamuts most of colorspace edges to simulate are not printable so it does not matter. For example AdobeRGB green is not printable, but cyan should be printable.
2-Profile accuracy. It means how good profile is capturing actual monitor behaviour.
That second one "could" be improved with a Profile only with DispcalGUI. Read its documentation for "profile only" (without calibrating) tasks. Even a DCCS "table profile" 400 patches should be more accurate than a DCCS matrix one.
yumichan
3 Apprentice
•
738 Posts
1
October 2nd, 2015 00:00
1- Results between 95-99% are OK. 99% AdobeRGB intersection is for the panel in its native state.
2- Do not think in % terms because it is useless. Run a comparison between AdobeRGB and your profiles, there should (WILL) be a very small dE between profile green 0,255,0 and AdobeRGB green, so it's fine. Your eyes do not see % coverages but color differences (deltaE2000)
Your reports are very good and DCCS gives you a very good neutral grey (better than my unit, but my AMD GPU fix that issue) and expected "after DCCS" contrast, so it's all OK
t6ho
1 Rookie
•
14 Posts
1
October 2nd, 2015 11:00
yumichan, when you say run a comparison, is there a report on dispcalGUI to do this with?
yumichan
3 Apprentice
•
738 Posts
1
October 5th, 2015 01:00
Yes, validate profile against a reference ("measurement report"). Two ways:
-validate calibration against AdobeRGB profile, not useful since if you need AdobeRGB support you need color managed app... so it's useless unless you need to validate grey ramp calibration in a fast way.
-validate profile simulating AdobeRGB, that means how that profile will perform in color managed apps showing AdobeRGB images.
Be careful with gamma simulation options, a true 2.2 power gamma means infinite contrast (0cd/m2 brightness in black). Read DispcalGUI documentation.
t6ho
1 Rookie
•
14 Posts
1
October 5th, 2015 14:00
This is what I did:
And this is what I got:
https://www.dropbox.com/s/97wzjzu17rnjdv7/Measurement%20Report%203.0.4.3%20%E2%80%94%20DELL%20U2413%20%40%200%2C%200%2C%201920x1200%20%E2%80%94%202015-10-05%2016-04.html?dl=0
Does it seem right, yumichan?
yumichan
3 Apprentice
•
738 Posts
1
October 6th, 2015 01:00
That setup with "use smilation profile as target" is "CASE 1" in my previous post. Without that checkbox is "CASE 2", which is the color managed approach.
You used to check against a true 2.2 power gamma that your monitor CANNOT do, because it has not infinite contrast. I warned you prevoiusly against doing that... Select unmodified.
t6ho
1 Rookie
•
14 Posts
1
October 6th, 2015 12:00
Okay.
This is what I did and got:
https://www.dropbox.com/s/ttnh2itorexohv4/Measurement%20Report%203.0.4.3%20%E2%80%94%20DELL%20U2413%20%40%200%2C%200%2C%201920x1200%20%E2%80%94%202015-10-06%2013-42.html?dl=0
All seems good now, right?
t6ho
1 Rookie
•
14 Posts
1
October 6th, 2015 13:00
How could the results be improved? Via dispcalGUI calibration?
yumichan
3 Apprentice
•
738 Posts
1
October 6th, 2015 13:00
Could be better but yes.