Calibration doesn’t look good.

Home Forums Help and Support Calibration doesn’t look good.

Viewing 10 posts - 1 through 10 (of 10 total)
  • Author
    Posts
  • #25002

    Steve Moore-Vale
    Participant
    • Offline

    Hi. I have an eizo cs230 coloredge monitor and wasn’t happy with the spyder software calibration so switched to DisplayCal. The first calibration was great in that the rgb bars showed me I had a green tint and I was able to fix that but ever since then when I work on photos in Lightroom I notice I’m smooth backgrounds at very low iso settings the smooth areas look broken up and full of artefacts when zoomed in to 100%. They should still look smooth at that zoom level as they certainly have done before. I am wondering if my calibration has something to do with this. I have looked at the same photo at the same zoom level on my phone and it looks ok compared to the eizo. Any help would be greatly appreciated. I am now going to work but if needed I can go into more detail later about what calibration settings I used.

    #25004

    Vincent
    Participant
    • Offline

    Spyders are not accurate, that will be my 1st guess.

    Since LR has dithering those artifacts look like GPU calibration artifacts and not caused by colormanagement. If DisplayCAL loader (windows) does not fix it you’ll need a GPU suited for calibration (high bitdepth LUTs + dithering at output, like AMDs), and you’ll need a realiable colorimeter too, like Xrite’s i1d3 family (i1displaypro, etc).

    Also CS230 should have HW calibration with ColorNavigator (CN), so buy an i1Displaypro, use CN and see how it looks. Since calibration is stored in monitor, no need for new GPU. You can validate results with DisplayCAL (WHite LED IPS spectral correction fro i1dispalyrpo colorimeter)

    • This reply was modified 3 weeks, 5 days ago by Vincent.

    i1Display Pro on Amazon  
    Disclosure: As an Amazon Associate I earn from qualifying purchases.

    #25007

    Steve Moore-Vale
    Participant
    • Offline

    Thank you for your reply Vincent. I’ve seen a lot of posts that say Spyders aren’t accurate. I use an Nvidia RTX 2060 founders edition GPU.  So if I’m reading your reply correctly you’re suggesting that I get an i1 pro and use color navigator with my existing nvidia  gpu rather than use display cal. Is that right?  Can you clarify what you mean by validating using display cal? Will the Argyll cms not interfere with color navigator? Thank you.

    #25008

    Vincent
    Participant
    • Offline

    If you want to test if you are experiencing issues related to GPU calibration., save this to your computerand open it with MS paint, if it is not smooth (steps) there is a calibration issue usually caused by GPU (HW or driver). If it is smooth but there are colorations spuder seems the most problable source of errors.
    It needs to be tested in MS paint to check it without color management.

    http://www.lagom.nl/lcd-test/img/gradient-h.png

    Regarding ArgyllCMS vs CN, as long as you do not install a profile with DisplayCAL , its tray calibration loader won’t be there so there is no conflict. There are no driver conflicts because i1displaypro colorimeter (not i1 pro, an spectro) does not need drivers since it’s a HID device, like a mouse. Do not install argyll drivers for it, they are not needed.

    #25009

    Steve Moore-Vale
    Participant
    • Offline

    Many thanks for your continued support Vincent.  Thanks for the picture. I will try that when I’m home from work. It’s interesting you mention coloration as when I was working with some files in Lightroom the other day they looked like they had severe color noise when zoomed in to 100% but it wouldn’t go away when I increased the color noise slider. I have heard spyder a aren’t great. Plus I have the express  4 version which has got  to be a very basic unit.  So if I’m understanding you correctly my best bet is to get an i1display pro and then use color navigator to calibrate using the i1. As I said, first and foremost i will check that image in paint and go from there.

    #25012

    Steve Moore-Vale
    Participant
    • Offline

    I have now looked at the picture in Microsoft Paint. The gradient is smooth with no steps/banding visible and I can’t see any coloration to it either.  Would you still suggest I try an i1 just mainly because it’s a far superior unit and will suit my needs better?

    #25015

    Vincent
    Participant
    • Offline

    I have now looked at the picture in Microsoft Paint. The gradient is smooth with no steps/banding visible and I can’t see any coloration to it either.  Would you still suggest I try an i1 just mainly because it’s a far superior unit and will suit my needs better?

    Then it does not look like a calibration issue.
    In theory LR uses dithering so it can deal with not idealized profiles and still be smooth & neutral (10bit test images for example)… but just in case redo your DisplayCAL+spyder calibration using single curve+matrix profile and black point compensation (profile tab) using the same settings as you used last time, and check if then it is gone.
    I theory LR show not show issues with 3xTRC profiles even if there are small drifts between TRCs (I’ve used them!).. but just check if a more idealized profile solves your issue at the expense of some color accuracy.

    PS: I mean “develop” module in LR, that’s where dithering is used. Other parts may show truncation errors (bands/colorarion) like any other colormanaged app without dithered outputs.

    #25017

    Steve Moore-Vale
    Participant
    • Offline

    Thank you Vincent. I will redo the calibration with single curve and matrix and black point compensation ticked and will see what happens then. I do find very often in loupe view in the library module that the photos look super horrible like they are low bit depth and then when I click on it to zoom in it corrects itself and fixes that issue.  I may buy an i1 and use color navigator in the future even if what I’m doing now does solve my issue. I don’t like using what I know is a substandard colorimeter which is what I have. Thanks again for your help.

    #25220

    uran0s
    Participant
    • Offline

    I was visualizing the range of ashes you had talked about previously, and it looks really bad, even in photoshop despite improving a bit.

    I tried it on macbookpro and it is perfect from the start, without calibration.

    Out of curiosity, I connected the BenQ monitor to my girlfriend’s PC, which has Windows 10, and to my amazement, the ash range test is perfect.

    I also noticed on Mac pro, that from DisplayPort to Hdmi cable makes a difference in colors, HDMI skin tones are more real, and in DP they are green tones.

    Even if you connect the BenQ to the Macbookpro via an HDMI port, the gray scale is incorrect, it only looks good on the macbookpro monitor itself. But is it perfect on windows?

    Any solution? Will this go there with calibration only?

    I’m really sad, I had no idea that the Mac was having these problems.

    #25221

    Vincent
    Participant
    • Offline

    Maybe macbook is outputing an unexpected pixel format for that display. There shoud be a info/diag tool on macOS suystem utilies to check pixel format something like ARGB888 or ARGB2101010)
    Some apple laptops have the same misbehavior with Dell displays in the past. That is related to macOS video card driver, not to calibration.

    White should look with the same color in windows or in macOS. No green tones.

    • This reply was modified 1 week, 3 days ago by Vincent.
    • This reply was modified 1 week, 3 days ago by Vincent.
Viewing 10 posts - 1 through 10 (of 10 total)

You must be logged in to reply to this topic.

Log in or Register

Display Calibration and Characterization powered by ArgyllCMS