Calibrating with ColorMunki Display Headache

Home Forums Help and Support Calibrating with ColorMunki Display Headache

Viewing 14 posts - 1 through 14 (of 14 total)
  • Author
    Posts
  • #2987

    sjc69
    Participant
    • Offline

    Having awful problems with Colormunki Display and OEM 1.1.0 software in trying to calibrate my brand new 27″ iMac on 10.11.4. I’ve been working with X-Rite, but they can’t solve my issues and now want me to send the device back (at my cost) for analysis. I have a hunch its a software issue, so I’m wondering if DisplayCAL would be worth trying before I go to the expense and hassle of returning the device. Does DisplayCAL work with the Colormunki Display? Do I need to uninstall the OEM 1.1.0 software before installing and calibrating with DisplayCAL? If so, does anyone know how I’d go about that? I’ve searched online and can’t find the answers to these questions. Lastly, does anyone else have experience of problems with the Colormunki Display and latest iMac Retina screens/Mac OS? Sorry if these questions are basic. I’m trying to get some work out the door and I’m at the end of my tether trying to wade through all this stuff :(.

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

    #2989

    Florian Höch
    Administrator
    • Offline

    Hi,

    out of interest, what’s the nature of the problem you’re having?

    If you do want to try DisplayCAL, you don’t have to uninstall the X-Rite software (I think in the past they used a system extension that would grab the device and made it unavailable for 3rd party software, but I believe this is no longer an issue). Just leaving all settings at defaults should produce an acceptable result.

    #2990

    sjc69
    Participant
    • Offline

    Hi Florian. I have a yellow cast and the screen is way too bright. They had me re-run the calibration after installing their default linear profile which made a slight improvement on the yellow cast. They also had me try setting the white luminance at auto, 90 and 120, but all the profiles output almost identical. Flicking between the different profiles calibrated with different settings makes no difference in monitor appearance. I’ve been sending them the profiles and they seem puzzled by it.

    After installing DisplayCAL, should I exit the OEM in the tray? Thanks for your help – much appreciated!

    #2991

    Florian Höch
    Administrator
    • Offline

    I have a yellow cast and the screen is way too bright

    Ok. When trying the calibration and profiling with DisplayCAL, it should offer to import the default X-Rite colorimeter corrections on startup. The right choice for your iMac display is likely “Spectral: White LED” (under “correction”). You can leave the other settings at default. During interactive display adjustment, just set the backlight (brightness) of the iMac to a comfortable level, and as the iMac lacks RGB balance controls, you can ignore the readouts and skip to calibration after adjusting brightness. The rest of the procedure is automatic.

    After installing DisplayCAL, should I exit the OEM in the tray?

    Probably a good idea.

    #2992

    sjc69
    Participant
    • Offline

    Hi Florian. I installed DisplayCAL and ran the calibration. The profile is certainly very different to the one I created using the OEM software. Much less yellow and very little difference to the out of the box iMac profile – would you expect this? I ask because I’m still not seeing on screen what I have printed out. I have a calibrated workflow – brand new Canon Pixma Pro-100 Printer, Canon ink, Red River Paper and ICC profiles, latest Photoshop, Photoshop managing colors, Softproofing, etc, etc, etc. What I see on screen is yellowy/khaki and what I have in my hand is reddish brown. I just don’t know where to go from here. Wondering whether to ditch the Colormunki all together (I’ve only had it 2 months!) and get something else. I think I’m probably going to do that, but thought I’d ask you one last time before going down that route. Thanks again!

    #2993

    Florian Höch
    Administrator
    • Offline

    The profile is certainly very different to the one I created using the OEM software. Much less yellow and very little difference to the out of the box iMac profile – would you expect this?

    Possibly. By default Mac OS X uses a profile created from EDID, which is usually good enough that colors don’t look visibly off, so I wouldn’t expect excessive differences after calibration/profiling.

    Canon Pixma Pro-100 Printer, Canon ink, Red River Paper and ICC profiles

    Hand-measured, or vendor-supplied profiles? (I tend to be very sceptical in regards to the latter)

    what I have in my hand is reddish brown

    Under what illumination?

    #2995

    sjc69
    Participant
    • Offline

    I ran another calibration, just to double check for consistency. I darkened the room (don’t know if that makes a difference) and chose Auto for the correction drop-down. The resulting profile was identical in appearance to the first one. Very similar to the iMac factory profile (just slightly less saturated in my opinion).

    Then I printed the image that I’m having problems with on Canon Pro Luster paper (given what you said about third party profiles). I think there is an issue with the Red River Profiles because compared to the Canon paper, the print out is very red. I’ll take that up with their support dept tomorrow. But, the Canon printout was still nothing like I see on screen. I’m comparing under normal lighting conditions. By normal, I mean a small studio with good daylight and supporting LV spotlights (not directed at the screen). I flicked through my Display profiles and found the closest match on screen to my printout is sRGB IEC61966-2.1 which is the same profile I have assigned to the image.

    I’ve been at this for 9 hours now, and I’m no nearer sorting out my calibration issues, so I think a large glass of wine is in order at this point … LOL!

    I don’t know what the issue is … Colormunki, ICC profiles, or it could well be something I’m doing fundamentally wrong (this I well accept and worry about :-/).

    #2998

    Florian Höch
    Administrator
    • Offline

    I printed the image that I’m having problems with on Canon Pro Luster paper (given what you said about third party profiles)

    The general problem with any vendor printer (or rather, printer+paper+ink combination) profiles, which includes the Canon supplied ones, is that they are at best a snapshot of the batch they used at the time with a specific single (if you’re lucky, an average) printer that matches your model, and you have no real way to evaluate how accurate those profiles are for your specific printer unless you have the equipment (spectrometer) to measure it (at which point you could also create printer profiles yourself).

    In a print-to-screen matching situation, the monitor is usually the lesser concern when it’s properly calibrated and profiled (although for best possible visual match the document white when soft-proofing should produce a close match to a white sheet of the paper you’re using when illuminated by the light you view the prints under).

    There’s also things that can go wrong in the print pipeline even if the printer and monitor profiles are accurate, print settings and buggy printer drivers being a possible culprit for example.

    In your shoes I would first make sure that your print settings are correct (=seek support from Adobe if you print through Photoshop and possibly Canon), and when you have verified that and still get a vastly different result to what you see on screen, try to find a printer profiling service where you can send a printed target that they make available to get a custom-tailored profile for your specific printer+paper+ink combination measured from the exact batch of ink/paper and printer you’re using.

    #3000

    sjc69
    Participant
    • Offline

    Florian, thank you so much for taking the time to provide me with this information! I’ve ordered a Spyder5Elite Studio kit that includes printer + paper + ink profiling equipment. Everything you said makes sense to me about the canned profiles. I’ll also work with Adobe to ensure I have everything set up correctly. I did speak to Canon and they were very helpful, but their solution was to use their plug-in for Photoshop. It’s a nifty little plug-in for those not wanting to go down the calibration route, but its a bit of a hack for those that do. I read on Amazon reviews that DisplayCAL works very well with the Spyder5Elite :).

    #3019

    sjc69
    Participant
    • Offline

    Here’s my update … the good news is I have been seriously cramming on color management issues so I’m much more familiar with the terminology and diagnostics to try and work out what’s going wrong with my workflow. Bad news, still no nearer sorting it out!

    Firstly, I cannot get a consistent monitor calibration using either the Spyder5Elite or the Colormunki Display. OEM software profiles differ very much from those produced using DisplayCAL. The DisplayCAL profiles come out almost identical to the factory iMac profile. Using the OEM software, I have issues with the ambient light monitor on the Colormunki – regardless of whether I have my blinds drawn and lights off or blinds open and lights on, Colormunki tells me my ambient light is very low and recommends cd/m2 of 80. The Spyder5Elite recommends 120 cd/m2 with my blinds drawn, lights off and 200 cd/m2 with my blinds open and lights on. However, the Spyder5Elite resulting profiles I get are too warm (pinky/yellow). I now have many profiles, all different. Thing is, how do I know which of these profiles is correct (if any). I need a starting point and I understand monitor calibration is where the starting point should be?

    I created a custom paper profile using the Spyder5Print and the largest target available (including grays). The resulting profile is wildly off. PDI test file in TIFF format displays about 30% out of gamut and the colors must be massively out of gamut because the color shift is crazy. Of course, the custom paper profile I made doesn’t bear any resemblance in soft proofing to any of the monitor profiles I made.

    For the life of me, I cannot understand what is going on here … I have been down so many roads, spent so much time, money and effort and I’m no where nearer to getting a color managed workflow it seems. To title this discussion ‘a headache’ is becoming a massive understatement :(.

    #3020

    Florian Höch
    Administrator
    • Offline

    OEM software profiles differ very much from those produced using DisplayCAL

    The difference usually comes from different whitepoint, measurement mode and/or colorimeter correction settings. Make sure they are all the same in each software when doing comparisons.

    In terms of print-to-screen matching, it helps to choose a whitepoint that makes the monitor white produce a close match to the actual viewing illuminant that prints are viewed under.

    Using the OEM software, I have issues with the ambient light monitor on the Colormunki

    It is better to not rely on any ambient light compensation functionality. For most consistent results, you want to control the ambient light, not the other way around. This usually also means eliminating or at least minimizing the effects of natural daylight, and have wall paint that is spectrally flat enough (e.g. no colored walls).

    Thing is, how do I know which of these profiles is correct (if any)

    They are probably all correct with regards to the respective settings they were created with (see above). You can get a colorimetric evaluation of monitor profiles via measurements by using the verification functionality in DisplayCAL, for example (the Spyder5 software may offer something similar).

    Of course, the custom paper profile I made doesn’t bear any resemblance in soft proofing to any of the monitor profiles I made.

    First step is to make sure that your monitor white closely matches your viewing illumination (see above).

    When creating printer profiles, make sure to print the targets in a consistent manner (e.g. depending on the software you use, this could mean disabling any printer and/or application level color management).
    The prints need to have dried enough before you measure them (can range anywhere from roughly half an hour to half a day depending on printer, inks and paper).

    Then, when you print using your custom printer profile, make sure to be consistent with the settings (use the same settings as when printing the targets, except use the printer profile you created). A way to verify the print path is working correctly in terms of color management is via measurements (the SpyderPrint package should offers this if it’s worth its salt).

    Sofproofing also needs to be setup correctly. I.e. in Photoshop:
    “Device to simulate” -> Select your custom printer profile here.
    “Preserve numbers” -> Needs to be OFF.
    “Rendering intent” and “Black point compensation” -> choose the same settings here that you use when printing.
    “Simulate paper color” -> If you have matched your monitor whitepoint to the viewing illuminant (my recommendation above), this should be ON. If you have matched your monitor whitepoint to the appearance of a white sheet of paper under your viewing illuminant, this setting should be OFF.
    “Simulate black ink” -> should usually be ON.

    #3025

    sjc69
    Participant
    • Offline

    Florian, again, thank you! I suppose its confusing for a beginner that Colormunki Display offer the ambient light measurement as part of the calibration process, but it doesn’t actually work! Datacolor have been great on the support front so far (very different from my experience with Xrite). The’ve supplied me with a comprehensive e-book on Color Management so I’m going off to cram-up on that for the next few days. Then, take a deep breath, and start again :D.

    #3038

    sjc69
    Participant
    • Offline

    Hi Florian. I take on board what you say about matching the white point to the viewing illuminant, but I really struggle with judging that so I’d like to be able to get at least some indication from the sensors on the colorimeters (even if its just a guide). I have both my Colormunki and Spyder5 sat alongside one another in the same place on my desk by my monitor. When I measure the white point color temperature for the Colormunki under the Calibration panel in DisplayCAL, it measures 4130 and I already know from studying color management that it can’t be right. I’d have to be sat in a pretty dim room for that measurement to be right and I’m not! I can’t measure the white point color temperature for the Spyder5 in DisplayCAL because I get an error message that its monochromatic only, but I can measure it in the OEM software and it measures at 6500. This mirrors what I am experiencing in the OEM software when I compare them side by side. Is there any explanation for this other than the sensor is faulty on the Colormunki?

    #3039

    Florian Höch
    Administrator
    • Offline

    When I measure the white point color temperature for the Colormunki under the Calibration panel in DisplayCAL, it measures 4130

    How did you measure? Gray card under the illuminant (I guess the spotlights?) and point the ColorMunki with diffusor attached onto the lighted gray card? I see no reason why you couldn’t have some trust in the 4130K reading in that case, unless you know for a fact that the illuminant should be rated at a higher correlated color temperature (e.g. from its datasheet).
    But note that you shouldn’t go blindly by the numbers anyway: For best visual match, don’t set a whitepoint target, and adjust the whitepoint purely visually.

    I can’t measure the white point color temperature for the Spyder5 in DisplayCAL because I get an error message that its monochromatic only, but I can measure it in the OEM software and it measures at 6500.

    It is true that the ambient light sensor in the Spyder5 is monochromatic. The Spyder5 software probably returns a 6500K reading regardless of actual illuminant chromaticity.

Viewing 14 posts - 1 through 14 (of 14 total)

You must be logged in to reply to this topic.

Log in or Register

Display Calibration and Characterization powered by ArgyllCMS