- This topic has 22 replies, 4 voices, and was last updated 6 years, 3 months ago by
.
-
Posts
-
My blacks (#2/#3) are still a little bit “off” – but I don’t know yet how to compensate this.
It’s normal for the black to have slightly higher delta E (although below 2 is still low) because normally for best contrast the black point isn’t touched by the calibration, and you usually don’t notice black having a slightly different hue if the black level is sufficiently low. If you want black point hue correction (sacrificing some contrast), you can enable it in the advanced options of the calibration tab.
Hello
This is my first post. Congratulation to such an excellent software like DisplayCal!
I work with a CS2730 as well. I have done what Pharaday recommended (in his post from 09.02.2017) and got the same results like Chris. DisplayCal is nearly in perfect agreement with ColorNavigator (used probe: i1 Display Pro).
I went a step further and rented a Colormunki Photo to do a matrix correction for (my grading monitor and) the CS2730.
Results:
CS2730 hardware-calibrated / profiled in CN (compensation table: no compensation):
Verification with DisplayCal (no Correction) = Whitepoint 0.17CS2730 hardware-calibrated / profiled in CN (compensation table: no compensation):
Verification with DisplayCal (& selfmade matrix correction) = Whitepoint 8.82CS2730 hardware-calibrated / profiled in CN (compensation table: color management):
Verification with DisplayCal (& selfmade matrix correction) = Whitepoint 2.26So what should I do?
– Distrust the Colormunki Photo and go back to no compensation at all?
– Just rely on ColorNavigator with compensation table put to “color management” and live with less optimal results with the DisplayCal validation (done with selfmade matrix correction)?Maybe I should also mention that I get perfect fitting colors between my grading monitor CS240 and the CS2730 with this set up:
– CS240 via Intensity Pro 4K, “Video 3D LUT for Resolve”-setting in DisplayCal (& selfmade matrix correction for CS240).
– CS2730, “Video 3D LUT for the Resolve”-setting in DisplayCal (changed for the GUI & selfmade matrix correction for CS2730). This on top of “hardware-calibrated / profiled in CN (compensation table: color management)” DisplayCal verification = Whitepoint 0.72
Calibrite Display Pro HL on Amazon Calibrite ColorChecker Studio on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.Hi,
if I’m understanding correctly, you calibrated and profiled the display with ColorNavigator, not using the correction that you created with DisplayCAL (which I think cannot be used with CN to begin with) for verification? That is not going to work. If using a colorimeter correction, you have to use it everywhere (for calibration, profiling, any other measurements).
Hello Florian
Thank you for your replay.
In short I came to the following workflow:
I wanted to do the calibration in hardware, so I did that in ColorNavigator (with its own kind of correction). The correction I created, I only used for the verification (in DisplayCAL). This because I thought that the verification ist only about how close a profile is to universal values, which are independent from how the profile was created. Is this wrong? The verification of that was ok, except whitepoint (2.26)
For the Resolve GUI I used DisplayCAL to create a 3D LUT (with my correction) on top of the ColorNavigator Profile. When I did a verification of that, I got ok results.
For my grading monitor (Eizo CS240) I used DisplayCAL to create a 3D LUT (with the correction I created for this monitor). So this was done 100% with DisplayCAL. The verification of that was ok.
My impression was, that I am not so wrong, because when I compare the colors of my grading monitor with the Resolve GUI, they look very identical.
I hope, I have not misunderstood the whole concept.
This because I thought that the verification ist only about how close a profile is to universal values, which are independent from how the profile was created. Is this wrong?
Kind of. The profile characterizes the monitor, i.e. it translates from device RGB values to the measured CIE values (and vice versa). But when measuring a monitor, the instrument and any applied colorimeter correction of course influence the measured CIE values.
For the Resolve GUI I used DisplayCAL to create a 3D LUT (with my correction) on top of the ColorNavigator Profile. When I did a verification of that, I got ok results.
Yes, because you created a new profile for the 3D LUT.
It’s no problem to verify 3rd party profiles as long as they are ICCv2 profiles not v4 (trying to verify a v4 profile should result in an appropriate error message). But the same settings (monitor/graphics card) need to be used as when the profile was originally created.
So with “the same settings (monitor / graphics card) need to be used” you also mean the the same correction?
At the beginning of this thread Chris wrote about doing verification tests and using different spectral corrections:
“I still can’t get a good verification test unless the card is set to manual at D65. I’ve used spectral corrections for GB-LED, RB-LED, WLED including the CS270 and CS2420 corrections.”
“After many hours of experimenting the best spetral correction for this monitor is the GBr-LED. However the video card temperature setting must be set to manual and 6500K.”
He must still be writing about the verification in DisplayCAL, because there are no detailed spectral corrections in ColorNavigator. There is only the choice for compensation table: “no compensation” or “color management.”
So, if I don’t misunderstand something, I should not have based my own steps on this procedure?
