- This topic has 7 replies, 2 voices, and was last updated 6 years, 12 months ago by
.
-
Posts
-
I’ve got i1Display pro and I need to match two monitors. I’ve spent over two weeks to do so first with i1profiler than installed DisplayCAL but still have some color cast on one or another.
video card: Nvidia GeForce8400 GS
Monitors: Samsung S27 D390H and Acer S27 1HL Dbid
I am totally new to all this but so far there are some strange things I realised.
I thought that my i1Display would need calibration and I would be asked to do this before I use it first time. Nothing like that happened.
After calibration with the same black point and color temperature Samsung has definitely grater contrast and red cast mostly visible on bright grey areas while Acer goes more into green but contrast is visibly lower. Also in a few occasions when I was adjusting red color it looked like program or device did not react on changes – I moved red to maximum and control window was still showing that it was not enough. Is it possible that the device is not working properly? I tried to use black point as measured but it did not work.
Is there any way to do this calibration in easy way and get at least acceptable results?
Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.Hi,
I thought that my i1Display would need calibration and I would be asked to do this before I use it first time. Nothing like that happened.
The i1 Display Pro is a colorimeter, it does not need calibration.
After calibration with the same black point and color temperature Samsung has definitely grater contrast […]
Note that if you want to match the black levels you need to use the lowest one that both monitors can achieve.
and red cast mostly visible on bright grey areas while Acer goes more into green
To get a good whitepoint match you should adjust the whitepoint visually. Pick one of the two monitors as reference, display a white patch (e.g. using DisplayCAL’s visual whitepoint editor), then adjust the other one to match (again, use the visual whitepoint editor, see the documentation for details).
Also in a few occasions when I was adjusting red color it looked like program or device did not react on changes – I moved red to maximum and control window was still showing that it was not enough.
The red channel is already clipping on your monitor, you need to reduce the other channels.
Is there any way to do this calibration in easy way and get at least acceptable results?
You should leave everything at defaults and only set the same black level (e.g. 0.1 cd/m2, which should be achievable by both monitors) and white level target on the “Calibration” tab, as well as match the white visually (see above).
Hi Florian
Thanks for quick respond. I really appreciate it.
As I mentioned I’ve already tried to use visual whitepoint editor but it is too complicated for me. When I try to move it to other monitor it looks like my profile for it goes disabled and I can’t see this white color I would like to get from it. It is pretty cool feature but probably for someone with greater experience than mine.
I saved verification charts and ran report on calibrated device after a few calibrations and here is what I could not get about i1profiler and I can’t get about DisplayCAL. Your advice is:
You should leave everything at defaults and only set the same black level (e.g. 0.1 cd/m2, which should be achievable by both monitors)
And that’s what I thought but Acer seems to have black point little bit lower than 0.1 and Samsung less than 0.2. Because I want white level at 160 cd/m2 and understand that some monitors may go brighter by sacrificing black point I set 0.25 cd/m2 for both. Now, as long as Samsung ends up with something close to it e.g. 0.2315 cd/m², Acer always go much lower like 0.1931 cd/m². Also measured luminance is lover than requested e.g. 151 and 145 respectively. Now I could rise luminance easily but during calibration program ask me specifically to lower it and reading shows that I have something between 160 and 164 ( with about 30% of native luminance it’s not easy to get it perfect on OSD and would be even harder when I go to recommended 120 cd/m2). I would expect that program follows my instruction and gives me whatever possible between 0.25 and 160 but instead I get lower black point but also lower luminance. To me it looks like program always try to get to the lowest possible black point ignoring my settings. Again I could live with this as long as both profiles end up close enough. I will try again do as you’ve said and record it because on the very beginning I tried default option but I can’t remember what was wrong and did not take notes.
Anyway my point here is that I spent reasonable money on equipment which was suppose to be more accurate than my eyes and believe me or not I can match those two monitors better with it’s video car interface ‘by eye’ than by using sophisticated equipment and software. So I think that I am doing something wrong but still don’t know what.
Thanks for your effort
As I mentioned I’ve already tried to use visual whitepoint editor but it is too complicated for me. When I try to move it to other monitor it looks like my profile for it goes disabled
Sure, the calibration will be temporarily disabled so the white can be freely adjusted.
Now, as long as Samsung ends up with something close to it e.g. 0.2315 cd/m², Acer always go much lower like 0.1931 cd/m²
Does the Acer have any “dynamic” features like dynamic dimming/contrast etc.? If so those would need to be disabled.
Now I could rise luminance easily but during calibration program ask me specifically to lower it
If you haven’t set a white level target, DisplayCAL will show the initial reading and the difference of the current reading towards it. This is purely informational. If you have set a white level target and white point target, it is also important to leave enough headroom for the actual calibration – if you have already dialed in the monitor to 160, there’s no headroom (assuming the white point doesn’t match and needs to be adjusted by calibration).
I can match those two monitors better with it’s video car interface ‘by eye’
That’s what I suggested, match the monitor white by eye, but don’t use the video card (software) controls, use the monitor (hardware) controls instead. As you won’t be using the visual whitepoint editor, calibrate with whitepoint set to “As measured”.
Hi Florian
Thanks for your advice.
You wrote:
Sure, the calibration will be temporarily disabled so the white can be freely adjusted.
It does not make to much sense here. I calibrated one display and I am happy with the result. The logic way is to use it’s white balance as a reference for the second one. Obviously profile for this device must be loaded otherwise there is no reference point available. I tried to go around and set a pure white wallpaper on both monitors than opened editor but measured a white color from wallpaper instead of editor field. Did not work but was kind of educational. After profiling Samsung (reference) showed luminance at 159.5 and Acer 157.92.
Anyway, with white point 120 cd/m2 and black point 0.1 I got something looking ‘nearly right’. To get there I needed to use low calibration speed and 1553 patches. By the way does number of patches have any influence on profile file size?
It looks like I am not the only person having problems with Acer and green cast so I would like to tray something else but I will start new topic to ask about it.
Does the Acer have any “dynamic” features like dynamic dimming/contrast etc.?
No, it does not. But it has some mysterious option called DDC/CI which is not very well explained by producer. I tried to switch it on and off after profiling but did not see any difference.
Last question on this topic:
I always set whitepoint at 6500K sometimes using keyboard and sometimes using drop down list. In a few occasions report shows ‘assumed target white point’ at 6400K. Why is that?
It does not make to much sense here. I calibrated one display and I am happy with the result. The logic way is to use it’s white balance as a reference for the second one.
Then you have to drag the whitepoint editor only to the display you want to adjust (i.e. to the Acer in this case), as it will only reset calibration on the particular display it is on.
Anyway, with white point 120 cd/m2 and black point 0.1 I got something looking ‘nearly right’. To get there I needed to use low calibration speed and 1553 patches.
I usually do not recommend going below “medium” calibration speed because of diminishing returns. Amount of profiling patches have no influence on calibration, but affect the profile accuracy.
By the way does number of patches have any influence on profile file size?
Yes, because the measurement data is embedded in the profile.
But it has some mysterious option called DDC/CI which is not very well explained by producer.
DDC/CI, Display Data Channel Command Interface is a standard that defines an interface for sending commands to the monitor, which can be used by (e.g.) calibration software to automatically adjust the screen. The problem is that most monitors only support a subset of commands, so you’d have to figure out how to drive each individual monitor by trial-and-error (due to often lacking documentation) and of course you need physical access in the first place. Therefore, no generic implementation is available. DisplayCAL does not support DDC/CI.
I always set whitepoint at 6500K sometimes using keyboard and sometimes using drop down list. In a few occasions report shows ‘assumed target white point’ at 6400K. Why is that?
Because the assumed target is calculated directly from the measured XYZ coordinates and rounded to a multiple of 50. A fluctuation of +- several dozen K from measurement to measurement is a normal fluctuation for consumer devices. Furthermore, correlated color temperature is not a visually significant measure of color difference, for that you should look at delta E.
Hi Florian
Thanks for explanation.
It looks like some other people had similar problem with pair Acer – i1 display pro because I found colorimeter correction file here:
I had some problem with installing it because when I went to: tools – correction – import colorimeter correction.. I could not see the file I saved on my disc. Finaly I clicked on globe icon next to correction and program found it for me. Funny thou it looks like exactly the same ccmx file.
With this correction I got the best results so far at least in numbers:
Measured vs. assumed target whitepoint ΔE*00 0.1
Average ΔE*00 0.18
Maximum ΔE*00 0.81
RGB gray balance 0.49However in report on calibrated device I got this:
15:06:52,911 DisplayCAL: Starting interaction with subprocess
15:06:52,915 Setting up the instrument
15:06:54,119 inst_creat_disptype_list can’t find cbid 0 for ‘C:/ProgramData/ArgyllCMS/Xrite
15:06:54,119 i1 DisplayPro, ColorMunki Display & Acer S271HL (GretagMacbeth i1 Pro).ccmx’
15:06:54,121 inst_creat_disptype_list can’t find cbid 0 for ‘C:/ProgramData/ArgyllCMS/Xrite
15:06:54,121 i1 DisplayPro, ColorMunki Display & Acer S271HL (GretagMacbeth i1 Pro).ccmx’Does it mean that this correction was not used or perhaps it was but did not work well?
I went to: tools – correction – import colorimeter correction..
It’s called “import colorimeter correction from other software” for a reason – there’s no need to import CCMX or CCSS format files as they are already in the correct format.
inst_creat_disptype_list can’t find cbid 0[…] Does it mean that this correction was not used or perhaps it was but did not work well?No, that message is purely informational. It just means the correction file was created a while ago, and does not contain a DISPLAY_TYPE_BASE_ID (cbid) entry.
