I recently found a 2nd hand CG279X with less than 150 hours on it for a good price so I decided to pick it up as that’s what I was used to using back when I was staff. We found the built-in calibration probe/process to be sufficient to align the monitor close enough to our calibrated broadcast reference monitors (all rec709). All the machines were Linux.
When I first connected it to my Mac (via usb-c) everything looked ok. I set the monitor to rec as well as the color profile in displays on the mac. So I do what anyone would do and ran the self-calibration on the monitor itself. Everything seemingly goes fine until it jumped back to the OS. All of the sudden everything is super green. The white point is completely off. After some digging I saw I should update the firmware. Done, no change. Then I downloaded Eizo’s ColorNavigator app and ran the calibration again. No matter what I did, the issue persisted. It’s clearly off.
I’ve since reset the monitor and using the non-calibrated rec mode on the monitor itself and it looks better in the sense that it’s not green but I couldn’t tell you if it’s even close to accurate or not.
Has anyone else had any issues with this or any other Eizo with a built-in probe? It really feels like the OS is getting in the way somehow with the color profiles it then assigns. When you run the process via ColorNavigator, it creates all the ICC profiles for the montior + targets so I would imagine it’d just work…but alas, it doesn’t. It could very well be user error as well…but I feel like I going about things in the correct way.
Can anyone offer any insight/suggestions/advice? On Linux it just worked…
I’ve had instances of the Eizo calibrator probe itself malfunctioning in the past. On these monitors I’ve successfully calibrated them using an external probe instead, through the Eizo ColorNavigator software.
I’m on CG2730s which are probably similar. I just ran the calibration on the Mac last night and had no issues. Color Navigator doesn’t run on Linux as far as I know, but in the past I’ve copied the .icc profiles that Color Navigator creates onto the Rocky side of things. In the display settings of Rocky it gives you the path where they are stored.
I wonder if in your case you can do the reverse. If there are good .icc profiles on Linux, copy them to the Mac and then just select them through display settings.
@Dan_Bergo Good to know. I’ve ordered a Spyder so I hope the route your suggesting will do the trick and indeed it’s a bunk calibration probe on the monitor itself.
@allklier Sadly I don’t have access to a Linux box any longer. What version of MacOS are you on? What is your display profile set to? It’s all quite strange so I’m starting to think Dan is right and it’s the probe itself.
I’m still on Ventura (10.13), upgrading later in the summer during a project break. And yes, trying it with an external probe seems a reasonable next step.
In the meantime, you could check the calibration data that Color Navigator came up with and see what it set the white point to?
In the meantime, if you know anyone else with a CG279X you could grab their calibration. Won’t match your hardware to the T, but is probably within 95% of it, as it would be the same hardware. At least as a bridge until your probe arrives.
When I just moved these monitors to the Mac, the default Rec709 setting from MacOS was quite off (dark and desat) until I calibrated them last night.
On one hand I hope it’s the internal probe as then everything would be sorted when I get the external, on the other hand that would suck. It’s part of the reason to get this monitor. Oh well…until the probe gets here…
Have you tried calibration using just the monitor’s menu and controls? Ie, not using Color Navigator at all.
This is what I did with my CG27.. several years ago(in storage now). I believe it’s the model prior to the version you have.
It would self calibrate once a week no need at all for my mac to be on.
The profile MacOS using for the Eizo is really a generic pass/through profile as all calibration changes happen on the monitor itself using LUTs generated by the calibration that work in line with some hardware adjustments it does.
Maybe double check after you’ve calibrated that MacOs is still using the Generic Eizo profile?
Also on the Mac I always calibrated my Eizo to its full range/color gamut with a gamma 2.2 which is the gamma MacOS os is expecting.
I just let the Flame viewport do the needed color management to Rec 709.
I just don’t like how the Flame UI and MacOS looks in Rec 709. Which is it looks like trash… Lol
I also use my Mac to work on my photos and other stuff. So there’s that.
Off topic but since you ordered a Spyder calibrator. I used to work for Datacolor waaay back in the day. Was a part time remote job(before that was a thing), after I had finished college. Very good and underrated hardware.
X-Rite devices are definitely not better.
Datacolor’s main business is in devices for industrial and manufacturing markets. -
I’m not sure if that is accurate as of recent MacOS versions. Mac doesn’t expect a specific gamma, etc. In your system settings you pick from a variety of color profiles that match the monitor you have connected.
And when you run the ColorNavigator, it installs a custom profile inside MacOS that matches your specific monitor + serial.
So that way what the Mac sends and what the monitor expects is no guessing game.
But what I’m saying here pertains to the 2730, it may be different for 2790 if it uses other tech.
And along those lines, validate if Eizo does store it in-monitor. FSI monitors do, because they receive Rec709 signals from unknown sources (computers, playout, cameras). But most computer monitors rely on the host more. It does help if you have different OS and settings, to make sure everyone sends the matching signal.
Oh interesting, that is different from before. I stopped using that monitor in 2021.
The only thing i wonder on those generated profiles is if they still aren’t just basically a pass through?
Not sure why the CG monitors wouldn’t still use the LUTs generated for it’s in the monitor as opposed to the OS.
If you can still calibrate the monitor using it’s menu’s without the need for your Mac, Linux, Windows machine to be on then I would think it’s still LUTs for the Eizo hardware?
@BenV Yeah, I tried the simple route first before resorting to ColorNavigator. I also flicked through all the different display profiles. Each and every one had a similar green tint.
Happy to hear that you dig the Sypder.
Unsure about the LUTs but I see the same as @allklier with regards to a LCC profile being created with the serial number and for each target profile. When you switch the mode via ColorNavigator, it also selects the corresponding profile within the OS.
I don’t think mine have ever calibrated with the system off (because I rarely turn the system off), so not sure how it works in that case.
They could still do a split LUT, one where the OS based LUT handles color space matching, and the internal LUT is just a delta to some previous checkpoint to account for drift. It definitely seems a orchestrated operatin via USB and coordinating.
I just wanted to report back now that I received the Spyder. I re-calibrated via ColorNaviagtor using the Spyder and it looks good to me. Just for giggles I ran the built-in sensor correlation and, according to the software, the two measurement devices are almost the same.
How and why this whole green fiasco happened is beyond me as the built-in probe is working as expected. Regardless, it’s solved now and that’s all that matters. Plus I have a probe I can use elsewhere for less than 200 euros.
