- This topic has 22 replies, 3 voices, and was last updated 2 years, 3 months ago by
.
-
Posts
-
I bought calibrite cc studio (former ColorMunki Photo) today and tried to calibrate my iMac 2019 27-inch display but I am not satisfied with the results. After trying all recommended settings (Single Curve + Matrix with black point compensation) it always ends the same – the shadows are not deep and I feel I lack contrast and saturation in colors and image looks flat (you know it is missing that trademark Apple displays glossy photo look). I am also unable to see or install any corrections – the dropdown menu just isn’t there in Display & instrument section and in Tools menu it is greyed out. Is that a way it’s supposed to be? I am also including the curves screenshot and I am wondering if that Tone Response curve is correct. From my Photoshop experience, creating curve like this makes photo look washed out and that is exactly how my screen looks like. Can anyone provide any assistance with this?
Calibrite ColorChecker Studio on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.I am unable to edit the post so I will just add this here: when I measured the Uncalibrated vs Calibrated display from Tools menu, the Uncalibrated display showed contrast ratio around 1400:1 and calibrated 1100:1, so there is definitely some drop in contrast as well.
iMacs are like big laptops regarding screen. 1 or two channels must be limited to get certain whitepoint.
If you modified 3 the you made it wrong = you are using VCGT to fix brightness instead of monitor/laptop brightness. Use white level as measured and set brightness manually. After calibration brightness will fall a little depening on distance between native whiet and your target white.
Thank you for reply, Vincent. I have attached settings I have been using. I had to manually enter 7100K Whitepoint, otherwise in reports I get Measured vs. assumed target whitepoint ΔE*00 as NOT OK (around 3.00). But anyway, it doesn’t work. Even if I set Whitepoint to As measured, the results are still loss of contrast and large boost in shadows. I even tried to set Tone curve to As measured but with same results.
How do I know that I am losing contrast and shadows? I have also calibrated my inkjet printer with Calibrite ccStudio app and printed the test photo from Lightroom using printer ICC profile created by ccStudio. The photo has been edited on same uncalibrated display. The printed photo has rich blacks, contrast and saturation same as uncalibrated display. Before I calibrated the printer my prints were washed out and desaturated so that means that the calibration of printer was done properly. From my point of view, there is something wrong with calibration of display itself. Now, I am not blaming DisplayCAL because I have exact same results from Calibrite native app. So I assume the calibration device itself is unable to measure the screen properly. I was just hoping that DisplayCAL offers additional settings (such as corrections which I am unable to use) that will help me with this.
Now I am using uncalibrated display + calibrated printer and the prints look the same as on screen. But somehow I think that isn’t correct way. Or maybe I misunderstood what the calibration is about.
Additional proof: when using calibrated ICC profile, when I enter the color R: 5 G: 5 and B: 5 in Photosohp and then make a screenshot, open it and measure it, it measured R: 13, G: 13 and B: 13. Same when I use the Digital color measure app in macOS, it doesn’t measure correct RGB value. Plus the black looks like grey.
Thank you for reply, Vincent. I have attached settings I have been using. I had to manually enter 7100K Whitepoint, otherwise in reports I get Measured vs. assumed target whitepoint ΔE*00 as NOT OK (around 3.00). But anyway, it doesn’t work.
That is not what was happening. Even if you set D50 as target it will modify 2 channels, not 3.
Even if I set Whitepoint to As measured, the results are still loss of contrast and large boost in shadows. I even tried to set Tone curve to As measured but with same results.
You need to stick to defaults due to macos color management engine limitations.
How do I know that I am losing contrast and shadows?
Contrast: by measuring (it’s unavoidable). Shadows: by using Adobe for evaluation since all macos apps relying in apple color management engine are unreliable unless you stick to default profile types. (or just measure if calibrated display & custom display profile match)
I have also calibrated my inkjet printer with Calibrite ccStudio app and printed the test photo from Lightroom using printer ICC profile created by ccStudio. The photo has been edited on same uncalibrated display. The printed photo has rich blacks, contrast and saturation same as uncalibrated display.
No, it cannot happen, by definition. If paper white reflects as much as screen brights (lux/pi = cd/m2 as an staring aproximation) at your current setting and they have the same color, light reflected by black ink will be brighter than black of screen (unless you softproof ink color).
If you softporoof ink color using a macos color management compatible display profile (BPC, infinite contrast fake black), black ink simulation will be fake, simulated in will be lifted (poorer display contrast means more lift)Also if you measure black screen with such cheap spectrophotometer results won’t be reliable. You need a colorimeter, Xrite i1d3 family recommended. If somebody told you that with a cheap i1Studio (or older colormunki photo) you won’t need a colorimeter for display they lied. Same with an i1Pro2/i1Pro3. But don’t worry, your i1Studio can be used to measure (high res mode, 3nm) CCSS or CCMX corrections for your new colorimeter (after you buy it)
Also the HW of such spectro is not able to measure WLED PFS backlight from P3 Apple screens (or any other manufactuer using that tech) although ArgyllCMS 3nm driver (hi res mode) can imporve it, specially for whitepoint.
Also that spectrophotometer won’t be able to capture M1 behavior of paper so beware simulation paper white under certain contitions.Before I calibrated the printer my prints were washed out and desaturated so that means that the calibration of printer was done properly.
You profiled it.
From my point of view, there is something wrong with calibration of display itself. Now, I am not blaming DisplayCAL because I have exact same results from Calibrite native app. So I assume the calibration device itself is unable to measure the screen properly. I was just hoping that DisplayCAL offers additional settings (such as corrections which I am unable to use) that will help me with this.
Now I am using uncalibrated display + calibrated printer and the prints look the same as on screen. But somehow I think that isn’t correct way. Or maybe I misunderstood what the calibration is about.
Explanations above.
Also check calibration curves after you set white level (cd/m2) “as measured” & calibrated. Whatever whitepoint you put it must modify no more than 2 channels. It is expected and unavoidable, it’s like a laptop and correcting white point means limiting max ouput of 1 or 2 channels.
Setting 7100K daylight target won’t avoid VCGT white correction since even if your dispaly has a native whietpoint in isotherm 7100K CDT or CCT, it may be moved (it is moved by your report 3dE from daylight curve) towards pink or green (although is was measured with such innacurate spectrophotometer without even using high resolution 3nm mode, so it may be better or worse).TL;DR
I my 1st message I thought you were talking about an i1Display Studio and Calibrite software, not an i1Studio spectrophotometer. My fault.
Your HW is very limited measuring white color or black color & brightness of any modern display. White can improve with 3nm mode (high res) in upper right combo. Black.. unlikely. Plan to add to your gear an i1d3 colorimeter.
Also your OS limits the accuracy of display profiles unless you give up on Apple apps and uses Adobe suite alone for image visualization + some potential menu bars artifacts when using a more detailed profile.iMac screens should be good behaved close to D65: minor WP adjustment in 2 channels measured with an i1d3 (+correction) using “faked” idealized matrix profiles supported by apple should be OK.
If you want to compare against printed copy you’ll need proper light source and use softproof… and your i1Studio may not be good enough for that task (depending on UV content)Additional proof: when using calibrated ICC profile, when I enter the color R: 5 G: 5 and B: 5 in Photosohp and then make a screenshot, open it and measure it, it measured R: 13, G: 13 and B: 13. Same when I use the Digital color measure app in macOS, it doesn’t measure correct RGB value. Plus the black looks like grey.
That means nothing. Color management to rescue!
RGB numbers in image colorspace do not need to match display colorspace.
if you want to make a reliable comparision use icclu/xicclu tools from ArgyllCMS (commandline) + spotread.Thanks Vincent for thorough explanation. So basically my spectrometer is only useful for my printer. I need to buy colorimeter for precise adjustment of the iMac screen.
It is a combination of factors…
Make calibration curves look 0=0 and 100% input => max of 2 channels limited. Default settings in displaycal for MacOS should improve 1st message screenshot of calibration curves.Vincent, would you be able to explain it to me how to set it up properly like I am 5 years old? I bought ColorChecker Display Pro (former i1 Display Pro) and I am getting exact same results as with ColorChecker Studio. The Calibration curve looks better but the Tone curve still has bump in blacks and the shadows are still brightened too much. I use MacBook Pro Retina 2016 Spectral correction, Whitepoint, White and Black level As Measured and Tone Curve Gamma 2.2 Relative. I am thinking of leaving the display uncalibrated at all because as I wrote before, the prints are almost identical match to screen then.
Edit: In fact when I run Verification on uncalibrated monitor, I have everything in green numbers OK except from Measured vs. assumed target whitepoint ΔE*00 which is around 7 but the color accuracy is measured OK and green.
It was explained in my previous messages. Calibration curves must have 0in => 0out. 255in => 255out in at least 1 channel.
You must ensure this. If that is not happen you have a misconfiguration or incompatiobility with ArgyllCMS, make sure Argyll is lates version (go to argyllweb) and Displaycal profile config is at default.
Your first message have that curve, then you posted it no more…If that is OK, you get that TRC (last attachement) and that TRC matches calibrated display behavior (report), all is OK. Resulting weird beahvior is result of faulty color management engine from Apple. Adobe suite should work without issue since it uses its own engine.
Regarding “no calibration” just verify factory calibration against default profile, although macOS seems limited too. There was a thread from a week ago or so. Search forum. Simulate DisplayP3 and use it as display profile should work for factory calibration. Verify hat DisplayP3 profile has same TRC as default display profile from Apple.
It does not matter your visual reports regarding print to screen matching. If you use full softproof black will be lifted. If you do not use softproof at all, your tests are useless. This is valid for custom display profile from DisplayCAL or factory calibration using default screen profile bundled in your mac.
If I calibrate the display using default DisplayCAL settings for macOS and then I open the photo in Safari or Preview or Photoshop or Lightroom, it always looks the same – the shadows are too bright and contrast is lost.
Contrast cannot be lost if calibration curves have 0 output for 0 input… which are missing since several post.
Also on a pure power law gamma 2.2, when showing sRGB images color managed, very dark greys (not black) will be lifted vs non color managed… because it is the way to show them. Why? Because sRGB TRC is not 2.2. So the only thing that matters is if calibrated display matches custom profile TRC and/or if uncalibrated display matches default profile TRC… which are the requested test you did not make.
I am attaching everything I was able to make reports of.
calibrated_* – Calibrated with default settings for macOS
uncalibrated_* – Default uncalibrated macOS profile
software_calibrated_* – adjustments made with software calibration options in macOS
Uncalibrated monitor has smaller “hump” in the blacks than the calibrated one. The best measurement report is obviously for calibrated device.
I understand also what you mean with your last post about lifting very dark greys. However, when I print photo on my profiled printer with ICC profile made by i1 Studio the photo is darker and richer than when I display it in Photoshop or in Lightroom on calibrated display. It just looks better to my eyes the way I intended it to look in the first place. So if I would made adjustments on calibrated display to have it dark and rich, it would come out of printer superdark and contrasty.
Furthermore, most of my photos are displayed online for clients to view on their uncalibrated screens and I also need to take into account that their dark greys are therefore deeper and richer so if I would be doing adjustements on calibrated display I would have to boost darks so hard that it would look afwful on uncalibrated display. I am trying to find the best middle way.
