Display calibration consistency between old and new technology.

[Tim Lookingbill]

I have limited access to a wide range of displays of varying gamuts with improved technologies as well as different brands new and old of display calibration packages like the Spyder, ColorMunki and i1Display to test this out.


I was curious to know from those with access to these items old and new if they found a consistency in the appearance of color managed images on displays of various age, gamut and quality calibrated with new and old calibration packages within different brands. Has anyone done a thorough study on this?

The reason I ask is from all the posts I've come across lately stating how off color looks in some form or fashion using the newer displays like those with LED backlights and/or with wider gamuts using old or new calibration packages. Most base this "off" look by comparing their older displays with smaller gamuts and outdated or soon to be outdated calibration devices and software. It makes me wonder about whether these manufacturers of calibration packages are able to maintain a consistency as technologies improves or if they're actually changing/controlling how color should look on a calibrated display.

Some user reports state they have to calibrate to 5000K to get a screen to print match while others using different devices and packages have to calibrate to 6500K. What's going here? On my old 2004 G5 iMac I have to calibrate to 6500K using the original i1Display because if I calibrated to 5000K I'ld have quite a bit of editing to perform on all my images. See below what 5000K does to color on my G5 iMac.

How big of an issue has this become and is it really noticeable between old and new devices?

[attachment=19833:D65vsD50need_edits.jpg]

Thanks for the input.

[Wayne Fox]

I have limited access to a wide range of displays of varying gamuts with improved technologies as well as different brands new and old of display calibration packages like the Spyder, ColorMunki and i1Display to test this out.


I was curious to know from those with access to these items old and new if they found a consistency in the appearance of color managed images on displays of various age, gamut and quality calibrated with new and old calibration packages within different brands. Has anyone done a thorough study on this?

The reason I ask is from all the posts I've come across lately stating how off color looks in some form or fashion using the newer displays like those with LED backlights and/or with wider gamuts using old or new calibration packages. Most base this "off" look by comparing their older displays with smaller gamuts and outdated or soon to be outdated calibration devices and software. It makes me wonder about whether these manufacturers of calibration packages are able to maintain a consistency as technologies improves or if they're actually changing/controlling how color should look on a calibrated display.

Some user reports state they have to calibrate to 5000K to get a screen to print match while others using different devices and packages have to calibrate to 6500K. What's going here? On my old 2004 G5 iMac I have to calibrate to 6500K using the original i1Display because if I calibrated to 5000K I'ld have quite a bit of editing to perform on all my images. See below what 5000K does to color on my G5 iMac.

How big of an issue has this become and is it really noticeable between old and new devices?

[attachment=19833:D65vsD50need_edits.jpg]

Thanks for the input.

I believe some older devices cannot correctly profile some of the newer displays, such as LED backlight displays, for example Spyder 2. However, I have and do use frequently a great variety of displays from Apple to Eizo and i have not seen any issues getting decent consistency.  These are all calibrated with a i1 Pro, however I have tested a colormunki and it works just fine.

If someone is claiming they need to use a white point of 5000k it most likely would be an adjustment to match the light source they use when viewing the prints, not because of the devices or displays.  Personally I use a custom white point of 6100k because it seems to more closely match my prints.  But just like the brightness of a display cannot be a hard number that everyone uses but must be adjusted by each person based on their individual viewing conditions, the white point at times may need to be slightly tweaked.

I would suggest to a person requiring a white point of 5000k that perhaps they might want to take a look at their viewing lights ...

[digitaldog]

Some user reports state they have to calibrate to 5000K to get a screen to print match while others using different devices and packages have to calibrate to 6500K. What's going here?

The illuminant under which they are viewing the print?

[Tim Lookingbill]

Thanks for the confirmation on consistency with your setup, Wayne.

This thread:

http://photo.net/digital-darkroom-forum/00VcF8

at Photo.Net on the Spyder 3 Elite on Snow Leopard had one of the posters, Brian M, indicating he had to set the WB to what the ambient light sensor said which was 5000K which happen to give him exact screen to print matches. I don't get that on my system with the original i1Display which doesn't have an ambient light sensor.

This and other threads on color behavior discrepancies and/or errors on newer packages and displays got me started wondering if these manufacturers are tweaking the appearance of the numbers with each upgrade and brand name. Other package's ambient sensor could be seeing something completely different and still call it 5000K or any Kelvin number for that matter.

Just wondering if photographers are all on the same page with the appearance of their calibrated displays considering these discrepancies. If they're different, just wondering how different.

Over at Best Buy several months ago I calibrated about 4 different Mac systems on display using OS X's eyeball calibrator on two Mac laptops and two iMacs and the images in the default photo gallery showed identical results on each image across all four Macs. No discrepancies.

I really don't get this having to set your display to 5000K. It must look really different than what I get and I have the Solux D50 lamp to compare my prints against.

[digitaldog]

The ambient sensors don’t really do squat, at least in terms of those two packages. Well they tell you if its too high. It doesn’t and can’t tell you what WP to calibrate to, based on so many factors in differing displays PLUS the undefined viewing conditions. Hopefully some group of settings provides the much sought after screen to print match.

Just Normlicht has a booth, software, Colorimeter, display combo that all talks to each other. I don’t know if it works or not (it should).

[Tim Lookingbill]

The illuminant under which they are viewing the print?

Do you mean the illuminant provided by a D50 light source like the Solux? Because that's the only light that's available that actually mimics the spectral reflectance characteristics offered by direct sunlight.

The Solux will make different white objects reflect a different cast depending on their spectral makeup on their surfaces. For instance I have a white painted wall that looks yellow under these lights because I'm assuming from the paint's sulphur content while Epson Glossy and Matte Inkjet paper will look much more neutral next to the white wall. The inkjet paper looks like it has optical brightners even though it doesn't.

I don't think the math is included in calculating and compensating this kind of illuminant effect written into a custom display profile which I do see affecting the perception of a match.

[digitaldog]

Could be a Solux (and they have various CCT options), a GTI booth, who knows. It has to be defined. They are all different. So how can any one WP be right even if the values correlated with the illuminant (which they often don’t).

[Tim Lookingbill]

Could be a Solux (and they have various CCT options), a GTI booth, who knows. It has to be defined. They are all different. So how can any one WP be right even if the values correlated with the illuminant (which they often don’t).

Going by the image I posted how many color hues have you seen of D50 or 5000K that look more neutral than my 5000K shown on my iMac in that image?

Do you find they vary by a wide degree?

[digitaldog]

See the lines running perpendicular to the black body curve? That’s the range of colors in for example, 5000K from magenta to green:

[Tim Lookingbill]

That's a very wide degree.

Thanks for the explanation.