Dell UP2718Q calibration for AdobeRGB

Home Forums Help and Support Dell UP2718Q calibration for AdobeRGB

Viewing 6 posts - 46 through 51 (of 51 total)
  • Author
    Posts
  • #22715

    Wire
    Participant
    • Offline

    Ya so that slight gray tracking mismatch is not a big deal

    See attached.

    Photoshop, Adobe RGB images

    One is factory Adobe RGB reference mode, Adobe RGB display profile CG unity.

    The other is native mode, DisplayCal custom GC cal and profile. I tweaked white of this one to agree with Dell’s factory reference white. Again I don’t know if Dell or corrected DTP94 is correct so it’s academic.

    OMG the gray tracking is just a disast… Uhhh… Perfect!

    So which one is which?

    If you are sitting in front of displays, side by side, there is one aspect that will tell the custom cal from the factory reference mode: It’s the pure red patch, not gray tracking. But even though the difference is A/B quite visible, so you could use it to pick one from other, try telling by eye which one would be truer?

    The attachment is straight out of phone, un-retouched and un-graded in any way. Just stripped of metaadat.

    /wire

    Attachments:
    You must be logged in to view attached files.
    #22721

    asdfage wegagag
    Participant
    • Offline

    @ Vincent, what is the most affordable Spectro you’d recommend for Monitors/ TVs/ Oleds

    #22746

    Vincent
    Participant
    • Offline

    Ya so that slight gray tracking mismatch is not a big deal

    See attached.

    Not small, it’s big… if your DTP94 is accurate. If it is not… then things on spot (like whitepoint) or report are off by the same amount.
    It’s easy for a colorimeter to be off on actual wp coordinates…but seeing the whole range of greys off from each other? Unlikely

    Also range issues cannot be captured in a image with so much noise. Range is past 2 on both monitors and that is bad QC, barely usable as it is.

    Since you have DIsplayCAL and a supposed to be custom correction for colorimeter… it’s strage/weird/nonsesne to do not correct it if you GPU has not banding issues. I do not remember your computer HW regading that question. i1d3 slowest calibration can be done in 30mins ans last for several months on a LED monitor. Medium speed on a DTP94 should be time affordable and you’ll get rid of that bad calibrated grey ramp that you have on your two displays.
    Unless you have some mac mini with its limited iGPU… it’s a nonsense to do not correct it while wasting posts here.

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

    #22753

    Vincent
    Participant
    • Offline

    @ Vincent, what is the most affordable Spectro you’d recommend for Monitors/ TVs/ Oleds

    Triluminos, WLED PFD phosphor backlights have a very significative spectral signature in red channel due to their phosphor. You can use DIsplayCAL “(i)” button in a i1d3 spectral correction combo to see spectral power distribution plot.

    Hence… if you use Xrite spectros which are the most affordable (i1Studio for example) you’ll need to use high res mode (3nm) if you want to capture that.
    It limits the software available to that kind of spectros to DIsplayCAL/HCFR and other ArgyllCMS dependent software that may appear in the future.
    AFAIK Calman and that other stuff will measure at 10nm… which could lead to not so accurate measurements on those TVs/Monitors. This can be a problem if your TV/monitor has some kind of HW calibration and you need to use some of these closed/paid software to write it’s internal LUTs.
    Some of them like Calman allow you to write your own correction matrix manually (or offset or whatever way they name it)… so it’s possible to measure and correct a colorimeter like a i1d3 or much more expensive devices like a Klein with DisplayCAL and an i1Studio, then copy manually CCMX correction to Calman and calibrate internal LUTs with Calman while measuring with corrected colorimeter.
    Other vendor HW cal solutions like those from Dell or Benq do not allow that. NEC & EIzo allow colorimeter correction but from their own measurements (10nm with an i1Pro2 for example), so IDNK right now if you can “hack” and write it manually. If you own one of this, contact vendor support (do not ask about hack… ask about enter manually custom matrices/offset for colorimeter)

    So as summary, an i1Studio may do the job while working on 3nm mode with ArgyllCMS software. The use its measuremensts to correct a colorimeter like an i1d3 or better. Use colorimeter for calibration/profilling.
    If you need to measure that kind of “difficult” backlight (narrow SPD) with closed software because of HW calibration you may need other devices … and they are MORE expensive (by a huge amount).
    I would choose an i1pro2 basic+i1dispalypro, you may pay the cost gap by profilling printers with ArgyllCMS (FWA). Also I’m not sure about i1Studio support for illumination verification software like CT&A, check their website if you need that functionality.

    • This reply was modified 4 years, 1 month ago by Vincent.
    #22767

    Wire
    Participant
    • Offline

    @rest-of-world Based on my data-points the conclusion is “Use Dell UP out-of-box Adobe RGB and sRGB modes in good health”

    Regardless of points about DTP94 instrument, or slight verification report issues, or the relevance of a factory certification, or noise in the photos I supplied, a UP2516D custom calibration side-by-side with the builtin Adobe RGB mode, shows very little difference between these two alignments.

    The photo I submitted is in good faith, it isn’t hiding anything in the noise.

    And the 2 dE tracking error in verificationreport for factory mode is not readily visible.

    Color looks right, no visible banding, crossovers or other weirdness—8-bit data path can work fine.

    And if you want better, you can create a custom cal and bring reported tracking well under 1 dE, as a verification report I posted earlier in thread shows.


    @charlesss
    if you have free 15 mins to help add data points for UP2718Q, you say you have known excellent instruments, you can run an Adobe RGB simulation verification pass on small number of patches, it will take 5–10 minutes, and post the verification report. To do so

    1. set display control to color space > Adobe RGB
    2. choose DCal verification tab, “simulate”, “use simulation profile as display profile” “black offset 100% (default)
    3. Run report

    /wire

    #22781

    Vincent
    Participant
    • Offline

    The photo I submitted is in good faith, it isn’t hiding anything in the noise.

    I know. I just pointed that you try to use a common camera as some kind of colorimeter trying to capture narrow spectral distribution from those UP25 or 27.

    Also a lot of people could atttach some 10x15cm print test with a gradient in not so good light conditions that looks neutral thanks to noise but upon visual inspection shows typical a* variation in CMYK without greys photo inkjets for home, confirmed by spectrophotomerter analysis (that web you linked before is a joke for monitor reviews but has a very nice B&W printing test that I do recommend). That situation is extremely close to the one we have here.

    And the 2 dE tracking error in verificationreport for factory mode is not readily visible.

    1.5x from DUCCS is cleary visible in all units I’ve seen, so… range over 2 and individual greys with high dC errors seems… very unlikely to not be noticed visually if you work with B&W.

    And if you want better, you can create a custom cal and bring reported tracking well under 1 dE, as a verification report I posted earlier in thread shows.

    I hope you do so, it would be a pitty to work in a monitor wit a bad factory calibration when you have tools to correct it… unless GPU hardware limits you.

    @charlesss if you have free 15 mins to help add data points for UP2718Q, you say you have known excellent instruments, you can run an Adobe RGB simulation verification pass on small number of patches, it will take 5–10 minutes, and post the verification report. To do so

    1. set display control to color space > Adobe RGB
    2. choose DCal verification tab, “simulate”, “use simulation profile as display profile” “black offset 100% (default)
    3. Run report

    If he has 20 min it’s more productive to use DUCCS with an i1displaypro, get a Native or AdobeRGB calibration in CAL1, then verify both with some under 100 verification patch. Verifcation itself is less than 3min with huge testcharts and that colorimeter… and if he owns Dell with HW cal or plan to own one of these or CGs or PAs or CS, I would say that it’s a mandatory investment IMHO.

    Range issues can be partially corrected in GIMP/PS if he chooses “table” profile in DUCCS (3xTRC), and corrected in Lightroom CaptureOne (dithering) with the same profile type in almost every GPU HW (further GPU calibration corrections may not be GPU independent)

Viewing 6 posts - 46 through 51 (of 51 total)

You must be logged in to reply to this topic.

Log in or Register

Display Calibration and Characterization powered by ArgyllCMS