Calibration of DisplayP3, AdobeRGB and Rec709 – am I right?

Home Forums Help and Support Calibration of DisplayP3, AdobeRGB and Rec709 – am I right?

Viewing 12 posts - 1 through 12 (of 12 total)
  • Author
    Posts
  • #33173

    Mikael
    Participant
    • Offline

    I’m running a Macbook Pro + an Eizo CG729X as my primary display. The Mac display is DisplayP3 while the EIZO covers a bunch of color spaces and can be changed using the on-display controls.

    I’ve successfully managed to get a good Rec709 workflow on both of the displays, for my semi-pro setup it’s enough. I basically setup both of the displays the same way (whitepoint, whitelevel, blacklevel and tonal curve) with the simulation profile set to Rec709 ITU-R BT. 709. I load the .icc’s on the OS level and activate the “Use Mac display color profiles for viewers” in Resolve to match the displays. I feed Resolve with Rec709 data.

    So … I need to fill in the gaps and make sure my learnings are correct so far:

    – I can never change the actual color space used in the display with an .icc. The Eizo color space can be changed using the physical on-display controls and then calibrated properly using DisplayCal.

    – If I want to match two displays I can match the Calibration settings and validate the calibration with the Simulation profile dropdown to understand how much gamut coverage there is, correct?

    – If I want to run Lightroom with ProPhotoRGB it’s correct to calibrate the DisplayP3 Mac display as DisplayP3 + AdobeRGB on the Eizo (largest color gamut available on-display controls) and then Lightroom will show me proper colors as it uses OS level .icc profiles? I match the Calibration settings for both displays using different colors spaces and they will show the same result as long as the source data fits within both of the color space gamuts? Is this the preferred workflow?

    In the end I want to secure AdobeRGB and make sure I can maximize the capabilities of my printer.

    Thanks,

    Mikael

    #33180

    Vincent
    Participant
    • Offline

    – I can never change the actual color space used in the display with an .icc. The Eizo color space can be changed using the physical on-display controls and then calibrated properly using DisplayCal.

    Use ColorNavigator (CN) for general purpose calibration, then validate with DisplayCAL. If you own an i1displaypro read Midas’ thread (CS2731) about how to correct CN7 to use the proper WLED PFS spectral correction if default generic matrix for that model does not works as intended.
    An i1Pro”X” (spectrophotometer cannot read that displayproperly with CN although with ArgyllCMS and high res it may get a more accurate read, you really need an i1d3 colorimeter or better with those WLED PFS displays.

    If that EDR stuff looks too much complicated for you: CN, native gamut, then on after calibration whitepoint tweak make it mach DisplayCAL readings for D65 with the proper CCSS correction (HP Z24x)

    – If I want to match two displays I can match the Calibration settings and validate the calibration with the Simulation profile dropdown to understand how much gamut coverage there is, correct?

    Again, set P3 gamut coordinates in CN and limit CG-X native colospace. Same warings about CN7 as explained above. As Eizo’s target whietpoint you can use native WP in mac or D65 for both. You can use CN “after calibration” controls to fix whitepoint visually.

    – If I want to run Lightroom with ProPhotoRGB it’s correct to calibrate the DisplayP3 Mac display as DisplayP3 + AdobeRGB on the Eizo (largest color gamut available on-display controls) and then Lightroom will show me proper colors as it uses OS level .icc profiles? I match the Calibration settings for both displays using different colors spaces and they will show the same result as long as the source data fits within both of the color space gamuts? Is this the preferred workflow?

    No. Eizo should be set to native gamut (~sRGB/AdobeRGB blue, ~AdobeRGB green, ~P3 red). Match white & brightness in both displays (numerically or using one as reference) and once they are calibrated & profile. you are done.

    In the end I want to secure AdobeRGB and make sure I can maximize the capabilities of my printer.

    Then you do not want to limit your Eizo Display to AdobeRGB.

    • This reply was modified 2 years, 3 months ago by Vincent.

    Calibrite Display Pro HL on Amazon  
    Disclosure: As an Amazon Associate I earn from qualifying purchases.

    #33201

    Mikael
    Participant
    • Offline

    Thanks Vincent, excellent! will give this a try tomorrow when I’m back in the studio.

    #33203

    Mikael
    Participant
    • Offline

    Hey again, trying to pick up the important learnings here. Need to understand things bottom-up 🙂

    1.) Rec709 for both displays for post-production workflow

    > If that EDR stuff looks too much complicated for you: CN, native gamut, then on after calibration whitepoint tweak make it mach DisplayCAL readings for D65 with the proper CCSS correction (HP Z24x)

    In ColorNavigator7 (CN) how can I match the whitepoint, there is Native, Standard value and Color Temperature options. What is the DisplayCAL readings for D65 with CCSS correction (HP Z24x), is this done using DisplayCAL?

    2.) Matching P3 for both Displays for P3 workflow

    > Again, set P3 gamut coordinates in CN and limit CG-X native colospace. Same warings about CN7 as explained above. As Eizo’s target whietpoint you can use native WP in mac or D65 for both. You can use CN “after calibration” controls to fix whitepoint visually.

    What do you mean by “limit CG-X native colourspace”?

    3.) Matching AdobeRGB, ProPhotoRGB for photo editing and printing workflow

    > No. Eizo should be set to native gamut (~sRGB/AdobeRGB blue, ~AdobeRGB green, ~P3 red). Match white & brightness in both displays (numerically or using one as reference) and once they are calibrated & profile. you are done.

    I’m the stupid one here but I’m trying to understand the workflow, if I want to match Rec709, P3 or the extended ProPhotoRGB colorspace I do:

    Example Rec709:
    1.) Eizo => Set to Native color space
    2.) Set whitepoint (eg D65 for Rec709) => Use “Manual adjustment …” to measure use DisplayCAL and the “Calibration tab” and the measure tools (with Correction “LCD PDF…Z24x G2” applied).
    3.) Gamma (eg 2.4 for Rec709)
    4.) Done

    I then simply match the Mac using it’s native P3 gamut and match the settings above?

    So no matter if I work in Rec709, P3 or extended ProPhotoRGB the calibration workflow would be the same. One thing though, would it be a good idea to limit the Eizo to Mac P3 gamut to make sure both display capabilities are the same?

    Why is the Eizo CN a better tool than DisplayCAL, I could run the built-in calibrator or the i1Display 3 device.

    Thanks for your patience, I’m not a tech expert here just trying to pick how to do this the right way.

    Mikael

    #33204

    Vincent
    Participant
    • Offline
    1. After calibration there should be a visual macthing whitepoint editor, a 2D plane with a* and b* axis. Adjust till DisplayCAL reading with HP Z24x correction hits D65. It should be on CN manual, look for that CIE a*b* 2D Grid.
    2. To simulate using its internal LUTs a smaller colorspace like AdobeRGB (reduce native red), P3 (same on green) or sRGB (reduce both)
    3. If you want to match Rec709 on all apps or OSes no mater if they are color managed or not, use CN to limit colorspace to Rec709, set D65 and 2.4 or 2.2 gamma.
      If you are in a color managed enviroment, like Photoshop, Capture1, Lightrorom or even macOS desktop, you do not need to match Rec709 at all, not even need to match gamma, color management engine of each app will do it for you, that’s why it’s “color managed”

    CN is better because it will use internal hardware to store calibration and allows you to limit colorspace at your will. DisplayCAL only calibrates grey on GPU (which could casue banding on nvidias and intels), and then makes a profile (which is the only thing Photoshop and other app care about).
    Well, it’s “almost better” than DisplayCAL, because CN cannot measure your display with i1d3 with 100% accuracy due to wrong EDR corrections bundled in software. Read Midas’ thread about CS2731 for more info…. or complain to Eizo so they solve it (…some day), or try con contact him on LiftGammaGain so you can get the forged EDR that Eizo refuses to give to its customers.

    #33205

    Mikael
    Participant
    • Offline

    Ok, got it. Will do some reading and tests and report back.

    Super thanks again for being so helpful ????????

    #33220

    Mikael
    Participant
    • Offline

    Hey again!
    Think I got it know 🙂
    I agree with what you’re saying and that is my plan. I’m currently using apps that uses MacOS Color Management (MCM) and relying on the native colourspaces to do the work on both Displays. I quickly ran the Calibrate steps on the Eizo (master) and then double checking the D65 output using the whitepoint tools in the Calibration section of DisplayCAL, the chromaticity coordinates matches exactly. To get Rec709 or any other colorspace using MCM to match on both displays I do:

    Eizo:
    Native gamut
    Brightness: 100 cd/m2
    Whitepoint: D65
    Gamma: 2.2

    Mac:
    DisplayCAL
    Rec709 pre-defined settings (match above)
    Calibrate and apply the profile on the OS level

    Finally, to work in DCI-P3 I would basically just tweak the brightness, whitepoint and gamma accordingly and keep the native gamuts and let the MCM take care of the rest? To work in P3-D65 I would simply match D65 plus the DCI-P3 specification for brightness and gamma, right?

    So million dollar question, when using something like ProPhotoRGB and Lightroom – this is an extreme case, in theory would it be possible to end up with more colors (out of gamut) on my Eizo or Mac and get a mismatch? is it a good idea to limit the Eizo to P3 to always make sure both displays never mismatch IF colors appear out of gamut in the color management pipeline? Most likely a stupid question as AdobeRGB is probably the widest colourspace you get out of the RAW debayer but anyway.

    Thanks again for securing my workflow,

    Mikael

    #33221

    Vincent
    Participant
    • Offline

     

    Mac:
    DisplayCAL
    Rec709 pre-defined settings (match above)
    Calibrate and apply the profile on the OS level

    No. Set 2.2. MacOS is color managed, hence your desired TRC (“gamma”) is irrelevant. So choose 2.2 taht should be close enough to mac embeded screen TRC.

    Also since Apple color management engine is extremely limited & faulty stick to default profile types. Photoshop uses its own engine not as limited as apple’s.

    Finally, to work in DCI-P3 I would basically just tweak the brightness, whitepoint and gamma accordingly and keep the native gamuts and let the MCM take care of the rest? To work in P3-D65 I would simply match D65 plus the DCI-P3 specification for brightness and gamma, right?

    for working with DCI-P3 as long as you are color managed gamma does not matter. Video player should take care of it if it is color managed and configured to use it.

    So million dollar question, when using something like ProPhotoRGB and Lightroom – this is an extreme case, in theory would it be possible to end up with more colors (out of gamut) on my Eizo or Mac and get a mismatch?

    It would be the opposite. Mac screen won’t be able to show printable cyan-greens present in AdobeRGB or eciRGBv2. Hence there will be more out of display gamut colors on mac screen than in  EIzo calibrated to native gamut. EIzo colorspace is bigger =  union of AdobeRGB and DisplayP3 (sRGB blue, adobergeb green, p3 red, more or less)

    Easy to spot them, just  open an AdobeRGB image (or whatever colorspace) in photoshop, then use softproof simulation on each display colorspace (ICC profile). DO not check preserve RGB numbers. Then enable out of gamut waring, do not care about near black colors, just the saturated ones.

    is it a good idea to limit the Eizo to P3 to always make sure both displays never mismatch IF colors appear out of gamut in the color management pipeline?

    I’ll say no, I’ll keep native.

    Most likely a stupid question as AdobeRGB is probably the widest colourspace you get out of the RAW debayer but anyway.

    No, flowers can have reds outside AdobeRGB, unlikely to be printable but can happen (roses). Same for plastic toys or textiles regarding cyans-greens out of gamut in mac display.

    • This reply was modified 2 years, 3 months ago by Vincent.
    • This reply was modified 2 years, 3 months ago by Vincent.
    #33225

    Mikael
    Participant
    • Offline

    This is great! Thanks again Vincent for explaining the process step by step.

    #33226

    Mikael
    Participant
    • Offline

    One more thing, my Mac display is slightly warmer than my Eizo. After calibration with the built-in sensor I get about 6352K (Reference Daylight) when measured using the chromaticity coordinates + Visual white point editor in DisplayCAL.  The Eizo should be 6500K and the Mac should also be around 6500K.

    The Correction dropdown seems to be at play here. Found a colorimeter correction online for the Mac display and using I’m using the HP Dreamcolor for the Eizo. My gut feeling says that the Eizo is right and the Mac is warmer.

    What could possibly affect this in DisplayCAL? Is the Visual white pointer editor the only way to measure whitepoint temperature?

    Mikael

    #33227

    Vincent
    Participant
    • Offline

    CCSS corrections for both screens and an i1d3 colorimeter are bundled with DisplayCAL:
    https://displaycal.net/i1d3
    -Eizo is closer to HP Z24x
    -Macs with P3 display can use the MacBookProRetina2016.ccss
    Both WLED PFS but with slightly different green channel.

    Slight differences in WP even with a lab grade measurement device can be caused by observer metameric failure, due to narrow spectral peaks in WLED PFS red channel.
    If you want CN to measure properly with your i1d3 colorimeter you need to use Midas’ approach (after calibration fine tune in Eizos a*b* editor or full EDR replacement). I won’t trust blindly the integrated colorimeter in CG-X

    #33255

    Mikael
    Participant
    • Offline

    Now, the color temperature seems to be in place. I also ran Photoshop and did a soft proof. Very interesting to learn more about this. Both Photoshop and Resolve play nice on the dual display setup. For some reason Lightroom + “Secondary display” feature does not work as expected, the preview on the Mac display is way more contrasty and warm than GUI window on the EIZO.  Same thing with Photoshop vs. Mac Preview app when showing the images in ProPhotoRGB on the same Mac display. Attached is an extreme version as the sRGB conversion when doing a screenshot of the wide gamut renderings gives a weird result, but anyway it’s sort of like that.

    So in short, it works and it’s not a CN or DisplayCAL issue (my guess) it must be on the software level. Resolve + clean video feed uses MCM and Lightroom main window looks ok on both display, Photoshop runs as expected if I duplicate the image on drag them side by side. The problem is Preview app + Lightrooms second display feature. They seem to be using the Preview color pipeline somehow, strange.

    Attachments:
    You must be logged in to view attached files.
Viewing 12 posts - 1 through 12 (of 12 total)

You must be logged in to reply to this topic.

Log in or Register

Display Calibration and Characterization powered by ArgyllCMS