Pages: [1]   Go Down

Author Topic: Calibration issues - CG318  (Read 1148 times)

Cornelius

  • Full Member
  • ***
  • Offline Offline
  • Posts: 102
  • photographer based in switzerland
    • Cornelius Fischer
Calibration issues - CG318
« on: October 15, 2018, 10:17:01 am »

Hi all

I need a little bit help from the color experts here. A couple of days I retired my CG277 and replaced it with a CG318-4K display. A truly wunderfull screen.
Now, I have some serious calibration issues which resulted in a warranty exchange from EIZO. The replacement screen still has the same issue and EIZO tries to give me a workaround instead of fixing the hardware. But let me describe the problem.

When calibrating the CG318 with the internal sensor to a specific white point (x,y) e.g. 5500k, I get a serious magenta tint in greys and whites.
Compared to my other CG277 and CG276 (internal sensor calibrated) and to proof prints on neutral Tecco PM230 paper, the CG318 clearly shows grey tones and colors tinted to magenta.
When calibrating the CG318 with my Color Munki Photo using ColorNavigator software, the result is quite decent and fits to the other 2 EIZO screens and the proof prints.

--> Result of my tests: Calibrating the CG277, CG276, CG318 and a cheap Dell Ultra Sharp to the same target whitepoint, using the ColorMunki for the Dell and the internal sensors of the CGs, all screens locks very similar except of the CG318. On the CG318 greys and colors will have a clear magenta tint.

--> With my understanding of colors, color management and measurement devices, I would clearly say, the internal sensor has a wrong reference calibration.

As this is already my 2nd CG318, EIZO likes for push me for a workaround. So here is what they suggested me today on the phone:
Correlate the built in sensor to an external sensor e.g. my ColorMunki Photo
Use the manual adjustment of the white point. Place the CG277 and CG318 next to each other and manually chance the white point on the CG318 until it fits to the CG277 and use this as target.
Use the ColorMunki photo to measure the paper white (I have norm light) of the target paper and use this as calibration white point target (x,y). Continue the display calibration with the ColorMunki and just do not use the internal sensor.

All these tips a nice, and they all give me decent results. But still, the internal sensor either can not be used, or I need to do a sensor correlation, which is IMO pretty odd to by a 5000$ monitor with a build in calibration sensor, which can not be used due to a obvious miss alignment.


What would you suggest? Should I push EIZO to a another replacement? Should I just use the sensor correlation and live with that (results here are good and the colors are correct with this method) or is my expectation just to high and it is common that sensors can have such a miss alignment and need to be correlated to and external device?

And by the way, does ColorMunki Photo need a recalibration of it's sensor from time to time? Should I contact Xrite and ask if they offer something like that? The Munki is alredy 5 years old and I always have the feeling, my displays have a slite yellowish tint when using the ColorMunki for calibration. OR, should I buy the X-rite i1 Pro2 Spectrophotometer suite which include printer profiling and use this photometer als calibration reference for all screens (we have 3 workplaces) and printers?
« Last Edit: October 15, 2018, 10:21:21 am by Cornelius »
Logged

elliot_n

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1219
Re: Calibration issues - CG318
« Reply #1 on: October 15, 2018, 10:32:42 am »

I recently tested an old i1 Display2 puck against the the built-in sensor of my Eizo CG275. The results of the i1Display were totally out of whack. So, yes, these things deteriorate. (It was ancient - more than 10 years old - so I binned it.)

It seems that you have a lot of measuring devices (four, I think - the built-in calibrators in the 3 Eizo displays, plus a ColorMunki). If they are all giving you good results, except the one in the CG318, I would take some photographs to demonstrate the problem, and then ask Eizo for a replacement (or a refund).

Logged

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2197
Re: Calibration issues - CG318
« Reply #2 on: October 15, 2018, 12:59:08 pm »

When I got my Eizo 318 I saw the same shift measured with an I1Pro 2. In my case the older Eizo 301 became my second monitor.

Further, when the monitors were placed side by side and I adjusted "white" to appear the same on both monitors then measured them both using the I1Pro 2, the xy coordinates differed significantly.

And if I matched xy coordinates, the whites were pretty far off from each other. Looking at the spectrum, there was quite a difference. The newer 318 had much less power in the 420-440nm range compared to the 301.

Reviewing the literature, I concluded that my color vision likely had much less sensitivity to the shorter wavelengths. Probably due to age and known yellowing. CIE matching functions were made with fairly young volunteers in their 20's.  Even that aside, there is considerable variation in matching whites that have different spectra yet the same xy.

However, I preferred the D50 "white" of the 318 over that of the 301 but also wanted to use the built in colorimeter in the 318.

What I wound up doing was setting the xy on the 318 to match my GMB viewing box with a really neutral matte paper. Next I adjusted the white of the 301 to visually match the 318. Then I measured boththe visually matching white xy's. They were WAY OFF. However, since I now had the xy coordinates that produced visually matching whites on the two monitors I just calibrated the 301 with the spectro and used its xy values to set its white point.

Now they both match, the 318 can use its internal colorimeter, and they both match my GMB box.
Logged

daicehawk

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 88
Re: Calibration issues - CG318
« Reply #3 on: October 16, 2018, 06:19:20 pm »

Eizos internal colorimeters are a laugh. Used one in a wide gamut model i cant recall now, wide gamut, color edge series. The grey axis was ok, but the blue was cyan. Changed to i1pro and off it went perfect. So I would not exchange mails about crooks and workarounds and just take a compensation i1pro software-free worth from Eizo. That said, that display was great in terms of gamut and colors.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Calibration issues - CG318
« Reply #4 on: October 16, 2018, 06:58:24 pm »

Eizos internal colorimeters are a laugh. Used one in a wide gamut model i cant recall now, wide gamut, color edge series. The grey axis was ok, but the blue was cyan. Changed to i1pro and off it went perfect. So I would not exchange mails about crooks and workarounds and just take a compensation i1pro software-free worth from Eizo. That said, that display was great in terms of gamut and colors.
Not really conclusive. I'd expect differing instruments to 'require' different settings to produce the same results. Now if you can't rid of the cyan bias no matter the settings and you test this with more than one sample of the Eizo instrument (because it's possible that one unit went south, filters aged whatever), it's an interesting observation but I wouldn’t state that all Eizo internal Colorimeters are a laugh. That's possible, you need more data.
Also, the i1Pro, depending on which product you're referring to (hate X-rite's silly names) might be a Spectrophotometer which would behave differently too


Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Cornelius

  • Full Member
  • ***
  • Offline Offline
  • Posts: 102
  • photographer based in switzerland
    • Cornelius Fischer
Re: Calibration issues - CG318
« Reply #5 on: October 29, 2018, 04:02:24 pm »

Thanks for your inputs.

In the meantime, I actually could calibrate my CG318 to a neutral white/grey, based on my reference prints. This also fits quite good to standard iPhone and MacBook Pro screens/colors. So I'm happy with that. Screen shows what comes out of the printer.

The only thing I'm quite pissed off, when I enter x,y target values based on my norm light tubes and using the build in sensor and as comparison the ColorMunki, I can not get really precise results. Sorry, but is a sensor not made for exactly this? Entering a specific white point and then running a calibration which should fit exactly to this specific target x,y ??
Logged

smthopr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 612
    • Bruce Alan Greene Cinematography
Re: Calibration issues - CG318
« Reply #6 on: October 30, 2018, 12:38:22 pm »

a couple years ago, at NAB convention, Eizo was introducing the new monitors with the new display with the better contrast.  They were showing some models with built-in calibration, but the Eizo representative told me that it would be better to use an i1 Display Pro probe than the built in probe.  I don't know why he said this, but he did.

For what it's worth, I have a CX271 display whose built-in probe is used only for maintaining white balance, but not real calibration.  I never use it as it seems off to me.  The CG2770 with the same display tech has a built in probe that I tested on a loaner display and seemed to work pretty well, matching my i1 Display Pro device.

There was recently some discussion on the Liftgammagain.com forum about the backlight type in these new displays.  I think this is the issue you may be seeing.

https://liftgammagain.com/forum/index.php?threads/cg247x-backlight-type.11552/page-2

This discussion is a bit technical though...
Logged
Bruce Alan Greene
www.brucealangreene.com
Pages: [1]   Go Up