Trying to get 3 displays to agree on whats white

[Olaf Arlt]

Hello everybody,

after using the not really cheap monitor Eizo EV2436W for some time at home and using an older Eizo S1931 as a second screen, I now had the opportunity to take another Eizo EV2436W home from my company. I had originally bought the Eizo because it had good color fastness and -reproduction in tests at that time.

I didn’t find it unusual that the Eizo S1931 had a visibly different white point and always looked slightly red/blueish compared to the more expensive EV2436W. But I was very surprised when even the two identical models EV2436W displayed the white background of e.g. Microsoft Word or Excel directly next to each other in very different ways.

So I thought that I maybe should calibrate the monitors with a colorimeter to solve the issue. First I bought the cheaper Spyder 5. But neither with the included Spyder 5 Express software nor with the comparatively more professional looking DisplayCal, did I manage that all three monitors showed a similar white. At least with the two identical EV2436W I had hoped to get rid of the color difference. With the manual visual calibration tool of Windows without a colorimeter, I had managed to do this in a more or less satisfactory way!

After some googling it turned out that many photographers and other professional image editors prefer the i1 Display Pro from x-rite to the Spyder because of better results. So I thought to myself, “ok, if you buy the cheapest device, you can’t expect perfect results” (although the colorimeter EX3 sold by Eizo itself is in the end only a rebranded Spyder 5!) and ordered the i1 Display Pro from amazon.

But the result was unfortunately again: neither the included software by x-rite nor the use of DisplayCal led to a fairly similar color representation on the three monitors. If I move an empty Word window with a large white background between the two screens, the white area is still displayed very differently.

I have now spent about 3 days with monitor calibration and in the meantime also learned a lot (e.g. that one should not set the calibration software to “native” or “as measured” as the color temperature target value, because then it’s no wonder if each monitor comes up with something different) and now calibrated all monitors to the generally recommended 6500K with 2.2 gamma and 250 cd/m2 white luminance and 0,2000 black luminance as target values.

But what irritates me strongly is the fact that the software supplied with the i1 has given me other specifications than DisplayCal when calibrating the monitor hardware (i.e. setting the gain values for red, green and blue in the monitor menu). When using the same colorimeter! I had assumed that for the same measuring instrument, pure white without color cast must always look the same. How can it be that here, for example, for the red channel, sometimes a value of 90% and sometimes a value of 96% is obtained to have the calibration tool report the measurement as optimal for balanced RGB?

I also found it strange that if the green bar was displayed as too weak and I lowered the “red” and “blue” settings in the Gain menu accordingly, it looked as if both “red” and “blue” were still a bit too strong. But then it was enough to reduce either “red” or “blue” by 1% and all three color bars were suddenly in the optimal range. So apparently it was left to my taste to decide if I want the monitor to show a tad more red or blue to have the colorimeter certify pure white. It can’t be an issue of the sensitivity of the colorimeter, because with every single keystroke on the gain control of the monitor the respective bar went up or down a bit immediately.

After some googling I came across an article like this:
https://support.datacolor.com/index.php?/Knowledgebase/Article/View/226/0/i-just-calibrated-two-monitors-to-the-same-gamma-and-white-point-but-when-i-compare-them-they-are-not-the-same-why-is-that

So according to that article, it doesn’t make sense to expect two monitors to look the same, because our eyes adapt very quickly to the environment, but that’s of course not happening with directly adjacent objects. Nevertheless, I wonder what is the point of a calibration, if in the end the displays show still different white tones with different color casts. Using the last color profiles created including hardware calibration, my middle monitor shows a slight green tint, my left identical monitor a slight blue/red tint and my right monitor (S1931) also a slightly stronger blue/red tint.

Which monitor should I use as a reference when editing photos? Theoretically it shouldn’t matter, because all three are perfectly calibrated according to the calibration software.

Additionally I stumbled over articles claiming a purely optical calibration without any colorimeter would suffice in most cases anyway. Which makes me ask if I should send the two colorimeters back to amazon, because it seems to be a waste of money, if you can achieve good results with a purely optical calibration as well. And above all I had managed that my monitors all displayed the same white without color cast in a direct comparison.

Do I expect too much? Did I miss something essential?
I would be very happy about tips and help from professionals, because after a lot of googling, the willingness to invest in good hardware and the creation of countless monitor profiles under Windows 7 and Windows 10 (I have a dual boot system) I am unfortunately at the end of my rope.

Just for information: all monitors are equipped with the official Eizo driver in the device manager, so they are called EV2436W or S1931. Before the calibration I did a total reset of all monitors and switched off monitor features like an automatic brightness adjustment depending on the room lighting. One EV2436W has 2214 hours usage time, the other 4927 hours. With the latter, the colorimeters show a little less brightness, so I included a target of 250 cd/m2 to “downgrade” the brighter Eizo display to the capabilities of the “older” Eizo. That helped reduce the tint a little more, but the difference is still significant and irritating.

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

[Author]

Posts