Pages: [1]   Go Down

Author Topic: Different Colors on Picture of two matched Monitors and in Reality  (Read 3448 times)

eisbaer

  • Newbie
  • *
  • Offline Offline
  • Posts: 8

Hi,

I took a picture of my two Monitors ( an EIZO CG276, calibrated with Color Navigator to D65 120cd) and a Mac Book Pro Retina (D65,120cd). On the Mac Book I adjusted the calibration with the whitepoint a litte to get rid of the red tint. Now the Monitors almost match perfectly if I look at them (besides the Color Gamut for sure)

Now I took a picture of both Monitors with my Canon 7D and they are worlds apart on the picture although when I look at them they are almost a perfect match.

Why??? Is this because of the different backlights of the EIZO and the Mac and that the CMOS Sensor of my camera interprets them differently as my eye?

How can I take a picture of my monitor where the colors match?? (without tweaking the calibration as mad) Is this in anyway possible?

Thanks
Frank
Logged

Czornyj

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1948
    • zarzadzaniebarwa.pl
Re: Different Colors on Picture of two matched Monitors and in Reality
« Reply #1 on: February 11, 2014, 04:19:48 am »

Why??? Is this because of the different backlights of the EIZO and the Mac and that the CMOS Sensor of my camera interprets them differently as my eye?

How can I take a picture of my monitor where the colors match?? (without tweaking the calibration as mad) Is this in anyway possible?
Yes, it's the difference between spectral sensitivity curve of the camera sensor and the human eye.

Yes, to some degree it's possible - all you need is a multispectral camera (as long as spectral characteristic of your eyes match standard observer functions).

To achieve it with trichromatic sensor you need to create camera profiles for each screen individually, render the RAW 2x with each profile, and combine them in Photoshop.
Logged
Marcin Kałuża | [URL=http://zarzadzaniebarwa

TonyW

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 643
Re: Different Colors on Picture of two matched Monitors and in Reality
« Reply #2 on: February 11, 2014, 05:39:08 am »

Does it make any difference when both screens are square on to the camera i.e. your Macbook is at the same angle horizontal and vertical and a similar height to the Eizo?

If you just want to match how the screens look to your eye then why not cheat a little and apply a curve adjustment layer through a mask in PS for the Macbook screen only?
« Last Edit: February 11, 2014, 05:41:47 am by TonyW »
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Different Colors on Picture of two matched Monitors and in Reality
« Reply #3 on: February 11, 2014, 10:25:55 am »

Now the Monitors almost match perfectly if I look at them (besides the Color Gamut for sure)
That's all that matters.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Fine_Art

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1172
Re: Different Colors on Picture of two matched Monitors and in Reality
« Reply #4 on: February 14, 2014, 12:38:51 am »

That is actually a very good test of how good is the camera color. If it sees things differently you are forced to use camera profiles to try to counter the difference. All the monkeying around may work well for a particular source light, but it is very unlikely it will work across a broad spectrum of lighting. So you would need to do test shots at beginning and end of shoots using IT8 or whatever, then adjust a profile for the shoot, then work on your pictures. No wonder there are so many sliders in the raw converters.

I will have to do this test with my PCs. Nikon and Sony cameras. Thanks for bringing this up.
Logged

D Fosse

  • Guest
Re: Different Colors on Picture of two matched Monitors and in Reality
« Reply #5 on: February 14, 2014, 03:32:08 am »

This is no surprise, or shouldn't be. Creating separate camera profiles for LED and flourescent (and each possible variety thereof) is standard operating procedure. These light sources don't have continuous spectra, but peaks and troughs that the sensor reads very differently from the eye.
Logged

Steve House

  • Full Member
  • ***
  • Offline Offline
  • Posts: 247
Re: Different Colors on Picture of two matched Monitors and in Reality
« Reply #6 on: February 14, 2014, 10:43:48 am »

Tony raises a valid point.  The angle of the screen is quite different for the Mac from the Eizo.  To compare the colours the screens must be parallel to remove the angle of view from the equation.  The fact that you don't see as much difference with the naked eye isn't surprising ... the brain corrects colour perception to what it knows the colours ought to be while the camera sees what's really there.
Logged

Fine_Art

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1172
Re: Different Colors on Picture of two matched Monitors and in Reality
« Reply #7 on: February 14, 2014, 11:37:00 am »

Tony raises a valid point.  The angle of the screen is quite different for the Mac from the Eizo.  To compare the colours the screens must be parallel to remove the angle of view from the equation.  The fact that you don't see as much difference with the naked eye isn't surprising ... the brain corrects colour perception to what it knows the colours ought to be while the camera sees what's really there.

Not when the 2 images are in the same field of view.

The laptop is tilted to the camera. Laptop screens are far more angle sensitive than large LCD screens.
Logged

Czornyj

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1948
    • zarzadzaniebarwa.pl
Re: Different Colors on Picture of two matched Monitors and in Reality
« Reply #8 on: February 14, 2014, 11:53:30 am »

Tony raises a valid point.  The angle of the screen is quite different for the Mac from the Eizo.  To compare the colours the screens must be parallel to remove the angle of view from the equation.  The fact that you don't see as much difference with the naked eye isn't surprising ... the brain corrects colour perception to what it knows the colours ought to be while the camera sees what's really there.

Both displays have IPS type panels with decent viewing angle characteristics, so it's virtually non-issue in this case.

Chromatic adaptation takes some time, so in direct comparison you can see if there's a visual difference or not.

A trichromatic sensor doesn't "see" what colours "really are", that's a pure urban myth! In fact it's virtually impossible to capture "real" colours with such camera, and the more the spectra of light that illuminates the scene differs from the daylight, the less "faithful" colours we get. But even under daylight the captured colours are always more or less distorted, and it's physically impossible it would had been otherwise.
« Last Edit: February 14, 2014, 11:55:28 am by Czornyj »
Logged
Marcin Kałuża | [URL=http://zarzadzaniebarwa

Steve House

  • Full Member
  • ***
  • Offline Offline
  • Posts: 247
Re: Different Colors on Picture of two matched Monitors and in Reality
« Reply #9 on: February 14, 2014, 01:08:55 pm »

...

A trichromatic sensor doesn't "see" what colours "really are", that's a pure urban myth! In fact it's virtually impossible to capture "real" colours with such camera, and the more the spectra of light that illuminates the scene differs from the daylight, the less "faithful" colours we get. But even under daylight the captured colours are always more or less distorted, and it's physically impossible it would had been otherwise.
Didn't mean to imply the camera is capturing real colour - it has its own biases.  But it may "see" distinctions where the eye doesn't because it's performing a purely physical process with no psychology involved, unlike the eye which is doing both capture and interpretation.
Logged

Czornyj

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1948
    • zarzadzaniebarwa.pl
Re: Different Colors on Picture of two matched Monitors and in Reality
« Reply #10 on: February 14, 2014, 01:17:05 pm »

Didn't mean to imply the camera is capturing real colour - it has its own biases.  But it may "see" distinctions where the eye doesn't because it's performing a purely physical process with no psychology involved, unlike the eye which is doing both capture and interpretation.

Problem with trichromatic sensor is that it "sees" colours in a different way than the human eye, so you need a camera profile to correct the differences. Such profile can be "tuned" only to one light source (like daylight), so if the light source spectra is much different than daylight spectra (which is the above case) the captured colours can be distorted quite heavily.
« Last Edit: February 14, 2014, 01:20:53 pm by Czornyj »
Logged
Marcin Kałuża | [URL=http://zarzadzaniebarwa

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Different Colors on Picture of two matched Monitors and in Reality
« Reply #11 on: February 14, 2014, 01:21:08 pm »

Didn't mean to imply the camera is capturing real colour - it has its own biases.  But it may "see" distinctions where the eye doesn't because it's performing a purely physical process with no psychology involved, unlike the eye which is doing both capture and interpretation.
If it 'see's a color we can't, it's not a color. Unless a camera meets the Luther-Ives condition, and I don't know that any do, such cameras exhibit significant observer metamerism with respect to humans. Good article on the subject:
dougkerr.net/pumpkin/articles/Metameric_Error.pdf
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".
Pages: [1]   Go Up