Calibrated two monitor, same model, different results.

Home Forums General Discussion Calibrated two monitor, same model, different results.

Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • #31889

    Dylanear
    Participant
    • Offline

    I have two LG 31UN97-B monitors. I have a i1Display Pro and calibrated using DisplayCal (using sRGB). After calibrating them both they look ok, better than not calibrated, but one is warmer than the other? It’s not dramatic, but it’s a bigger difference than I’d expect for two monitor of the exact same type having both just been calibrated minutes apart. The second one I recently got from a friend, so they have different histories, ages, I could see them having a little different color, but this is why I calibrate in the first place?

    Will a better, more sensitive calibration sensor do a better job? Will those lose precision over time? My puck is probably 6 years old by now? I’ve never noticed any issues with it, but I’ve always been calibrating the same monitor with nothing else to compare it to (other than my laptop without as large/good color gamut, so I don’t expect a perfect match and I do get an ok match).

    Thoughts? Suggestions?

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

    #31890

    Dylanear
    Participant
    • Offline

    Well, it won’t let me edit that original post, which is very unfortunate since I gave the wrong model name. The monitors in question are 31MU97-B. Not MU, not UN.

    https://www.lg.com/us/monitors/lg-31MU97-B-4k-ips-led-monitor

    #31892

    Vincent
    Participant
    • Offline

    I have two LG 31UN97-B monitors. I have a i1Display Pro and calibrated using DisplayCal (using sRGB). After calibrating them both they look ok, better than not calibrated, but one is warmer than the other? It’s not dramatic, but it’s a bigger difference than I’d expect for two monitor of the exact same type having both just been calibrated minutes apart. The second one I recently got from a friend, so they have different histories, ages, I could see them having a little different color, but this is why I calibrate in the first place?

    Very likely to be caused by uniformity issues. That monitor is (was, not sold anymore AFAIK) one of the worst widegamut models.

    Color coordinates were colorimeter read were +- equal but due overall bad color uniformity across screen you precieve a color cast… because ouside that center measurement point there is a color cast.

    Uncorrectable, maybe that LG hase some color uniformity feature but it will destroy contrast.

    A way to partially correct that overal color cast is to use visual witepoint editor on the less white screen, but if you do that you cannot use absolute colorimetric in LUT3Ds for that display.

    #31896

    Dylanear
    Participant
    • Offline

    The model was one of the first affordable wide gamut LCDs, it’s pretty ancient, I believe it was released in 2016? Hasn’t been made in years, I think there’s been at least two if not three models since in that LG lineage and yes, they don’t have great uniformity, but no, that doesn’t explain what I’m seeing.  The uniformity issues with the model are mostly brightness to my eyes, hue and saturation are reasonably uniform. You can see the color cast difference across the screens, The entire left screen is a tad bit warmer,  the entire right screen is a little cooler. But this is visible in the mid greys, not the edges of the gamut, no matter the monitors deficiencies why wouldn’t the colorimeter, if accurate, not detect this difference and adjust the LUT accordingly?

    Back in the CRT days calibrations would match much worse monitor performance/variations. At least for a few hours/days until the infernal, but glorious tubes would drift.

    • This reply was modified 2 weeks, 1 day ago by Dylanear.
    #31898

    Vincent
    Participant
    • Offline

    Measure it, as said before with such bad monitor* color cast is caused by color uniformity. Colorimeter only reads where it is placed, usually in the center. You see a fullscreen color as a whole, not in center and human vision is easily tricked.

    It was a GB-LED so reading spectral power distribution with an spectrophotometer in the center may reveal production defects/variation in some phosphors or main LED wavelength although it is unlikely. If that happens you need a custom correction for each monitor… and you’ll need an i1pro in high res mode (or equivalent/better device) to check this.
    If you cannot access one of these devoces, do as instructed above regarding visual whitepoint to overcome overall color cast issue (because poor color uniformity ), then do not use absolute colorimetric intents when creating a LUT3D in the “tinted” monitor.

    *)Given a price, you can always get a good monitor with a smaller size, usually Eizo CS line in 24″ or 27″. In those years CS240 was about 600 euro, dC<2 across screen even without uniformity compensation.

    • This reply was modified 2 weeks, 1 day ago by Vincent.
    #31900

    Dylanear
    Participant
    • Offline

    These have “good enough” color and I generally enjoy them quite a bit. But I only got the second one recently because it was the exact same model as what I had and really have been craving two 4k monitors after getting used to that at a job last winter/spring, It was from a friend so it was easy, no shipping or haggling and it was quite cheap.  No way in hell I can afford TWO new nice, large, 4k, wide gamut monitors right now. And I will eventually get something more modern, but likely still not too high end. Probably another cheap-ish wide gamut LG, like the dirt cheap 32UN650 or the similar 32UN880 , or if I can swing a bit more maybe a BenQ EW3280U,  ASUS  PA329C .  If you have suggestions for 32ish” 4k monitors on the lower side of the price ranges, I’m all ears. The new one will go in the center and the old ones will make decent side monitors.  I do everything from my day job (3D artist, mostly lighting), personal photography to flight sims on this PC.  three would be fantastic for the flight sims!

    I’m not at the final end of the color chain, I’m more in the middle. So the final grading of my work after it’s composited is by a full time colorist on high end equipment. The color accuracy of these old, mid range (low end for professional use) monitors is functional, no one is complaining about the color of my work, mostly I just have to match existing imagery, so small variations of color in the monitor comes out in the wash. Hey, these old things have better color than the old 27″ Dells my current company offered for work from home. Most companies I’ve been working for lately are just using sRGB or Rec 709, so not especially demanding. It’s been a long long time since I worked at a studio using a more exotic color space. Everything is moving to ACES, but still mostly on sRGB calibrated monitors. With WFH not going away, that may not change for a long time. Maybe someday I’ll be able to afford something that does a credible job of displaying most of ACEScs and Rec2020. I am not holding my breath for that.

    At some point perhaps I’ll try being patient and measuring at multiple points across the monitors rather than just the center. But I see more variation between the two than across either monitor. The uniformity, while not perfect does appear better than the match between the two? And the uniformity is actually not bad if I stand way back, much of the perceived uniformity variation is simply sitting close to two kinda huge monitors and the inherent angle shifts from that geometry.  It’s never going to be perfect with these old beasts, I just want as good a match as I can get within reason.

    #31901

    Vincent
    Participant
    • Offline

    I’ll repeat it again but it is pointless to go on on this threrad since solution has been explained.

    -bad color uniformity, that LG one of the worst models, bad HW cal software too. Almost as bas as Asus 32″ models excluding one of the latest which is very expensive (but the same bad HW cal software).
    If you need 32″ UHD or DCI-4k the cheapest reliable monitor is PA311D, @2.5k€. HW cal software lacks of exact CCSS/EDR correction but whitepoint can be manually tuned with displayCAL on OSD if you need it.

    -visual whitepoint editor to match overall color cast in the whole screen (not only in the center),  then use relative colorimetric/perceptual for LUT3D creation (to prevent that your LUT3D undoes such visual whitepoint)

    • This reply was modified 2 weeks ago by Vincent.
    #31916

    Dylanear
    Participant
    • Offline

    PA311D looks very nice. But it’s hilarious you suggest that’s the only 32″ 4k that could possibly be useful for professional use.

    Continuing the conversation is pointless yes, But thank you for taking the time to reply and offering your perspective.

    • This reply was modified 1 week, 5 days ago by Dylanear.
    #31918

    Vincent
    Participant
    • Offline

    It is not the only one, just the cheaper if you want near AdobeRGB green primary.
    If you give up on AdobeRGB green and your are ok with only near P3 green primary there should be other options.

    PS: try visual whitepoint editor & relative intents on LUT3D.

    • This reply was modified 1 week, 5 days ago by Vincent.
    • This reply was modified 1 week, 5 days ago by Vincent.
Viewing 9 posts - 1 through 9 (of 9 total)

You must be logged in to reply to this topic.

Log in or Register

Display Calibration and Characterization powered by ArgyllCMS