- This topic has 8 replies, 3 voices, and was last updated 6 years, 8 months ago by
.
-
Posts
-
Hi,
I’ve never calibrated monitors before using an instrument (did it manually before). I just bought a i1 display pro and installed displaycal. It downloaded stuff for the instrument and also a correction for my display (Samsung S24A850DW , which is supposed to be 100% srgb, Led IPS (or more precisely PLS))…Maybe its just me who got used to wrongly calibrated monitor, but, i cant help but feel something is off… There seems to be a light greenish/yellowish tint to the whites… After trying multiple different settings, number of patches, speeds, corrections, I just cant get it right… pretty much wasted most of my day trying to figure this out.
When I am doing the Interactive Display Adjustments, I need to tweak the RGB of my monitor and in order to get the bars to line up in the middle, I need to boost the green value much higher than the other colors, for example R54 G68 B53 … it feels wrong to my eye but, I continue the calibration anyways… In the end, when the profile is done, if I compare to my old settings, there is a huge difference… the whites dont look white anymore…
I tried showing my monitor to my friend, first with the displaycal profile, he said everything was fine, but then when I put my old profile and reset the RGB value of my monitor, he also agrees with me that my old manual settings seem more accurate, mainly again because of the whites…
Im wondering if my i1display pro is defective or something?
During the interactive display adjustment, could I set the RGB value so that I get a true white color (regardless of what the program tells me to change) and then do the calibration this way?
Also, I find the target 120cd brightness to be very low… so i aimed for 150cd… is there anything wrong with that? ( having it at 120 makes the white even more ugly)…
Help ?
Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.Hi,
visually, most people tend to prefer colder (bluer) whites because they look brighter, but that doesn’t mean that a blue white is more (or less) correct. Unless you aim to match a specific reference white (e.g. lightbox), there is no need to aim for any specific target whitepoint, and you can as well calibrate to native white (which is what the DisplayCAL defaults do).
Also, if you’re using a colorimeter correction that was contributed by another user, please note its usefulness to your particular situation is up to you to evaluate. It may or may not improve the absolute accuracy of your colorimeter with your display.
During the interactive display adjustment, could I set the RGB value so that I get a true white color (regardless of what the program tells me to change) and then do the calibration this way?
Absolutely.
Also, I find the target 120cd brightness to be very low… so i aimed for 150cd… is there anything wrong with that?
That’s fine as well and actually what you should be doing anyway (try to roughly match ambient light level for reduced eye strain).
there is no need to aim for any specific target whitepoint, and you can as well calibrate to native white (which is what the DisplayCAL defaults do).
Thanks for the reply, just a clarification, what do you mean by native white? would that correspond to a value of 50/50/50 for my in-monitor RGB values?
Also, if I understand correctly, you are saying that I can set the RGB values of my monitor to whatever I like in order to reach a white color that seems “neutral white” to me, and this wont affect the calibration ? Even though the Interactive Display Adjustment is NOT displaying Green check marks? That seems counter intuitive to me, but, just want to be sure… The colors will still be accurate?
Thanks for the reply, just a clarification, what do you mean by native white? would that correspond to a value of 50/50/50 for my in-monitor RGB values?
It corresponds to whatever the monitor comes set to by (factory) default.
Also, if I understand correctly, you are saying that I can set the RGB values of my monitor to whatever I like in order to reach a white color that seems “neutral white” to me, and this wont affect the calibration ?
Yes.
Even though the Interactive Display Adjustment is NOT displaying Green check marks? That seems counter intuitive to me
The delta E values and bars show you how far away the current whitepoint measurement is from the given target whitepoint (if any), or from the daylight or blackbody locus if no specific target is set (“As measured”). Usually, when the whitepoint is on the daylight locus, the human visual system has an easier time of seeing it as white.
The colors will still be accurate?
Yes. White is just the pivot point, the important thing is the relation of colors to white (or rather the neutral axis), because this is how we perceive colors.
Ok, I ran another calibration overnight with the same setting, except this time I adjusted the RGB value of my monitor to something that I consider more neutral white.
After the calibration, the results seem ok, it said I had 96% sRGB (which is closer to the 91% I had last time), the average ΔE was something like 0.25 and the Maximum ΔE was about 1 … however if I then go and do a Verification, the measurement report gives me totally different Maximum ΔE of 4.54 !? and the measured vs assumed target is 7.57 ? …
The settings I used for this verification are:
Setting: selected my latest calibration profile.
Testchart or reference: Extended verification testchart
Simulation profile CHECKED with : srgb iec61966-2.1
Apply black output offset (100%) CHECKEDSo I dont know, why is the ΔE from the measurement report so different than the ΔE displayed right after calibration?
I had done verification measurement with my previous calibrations and the max ΔE were much lower in the measurement report (although I didnt like the white color, everything seemed off, too green)…
There’s a difference between the profile self check (which just compares the profiling measurements to what the resulting profile predicts) and a verification using a different set of measurements (and in this case also involving a simulation profile). It is actually quite common that some monitors do not fully encompass the Rec. 709/sRGB gamut. So that explains the larger error for the one particular blue hue (which is simply out of gamut for your display). The assumed whitepoint is always based on a daylight or blackbody locus, but you’re using your own custom white.
ok… so, i tried again… i reset my monitor to default, did the pre-calibration RGB adjustment, this time followed the RGB indicators and I tried to run a ~8000 patch calibration…
After running the verification i see that the nominal tolerance is exceeded… so again something is wrong… i just dont know what… Any idea what settings I have to change?
Might help.
A problem I have had since monitor started having channel adjustments which ran from 0 to 100 rather than 0 to 255 is obtaining the colour temperature I want. So taking the monitor I am currently profiling so far the best I have managed to achieve is 6700K with a delta E at the initial setting stage of 0.1 to 0. This typically take 10 to 15 mins of fiddling with the channels and the brightness. That in my case is currently 122 cd/m^2. These things inter react. I also try and keep the channel setting as high as possible so that the % change per step is as small as it can be.
This may not be an intuitive process. Too high a colour temperature for instance suggests less blue but a slight change to one of the other channels might be the best way to go followed by an adjustment of blue and the other channel. This is why it can take a long time.
It’s also important that any software corrections on the machine are all reset and that the monitor contrast setting is at it’s default value. It may be worth making small changes to that later but in my experience on several monitors there’s no point.
John
