Jaws
gDoc
Metamerism is defined in simplest terms as the phenomenon whereby two colors match when viewed under one light source but fail to match each other under a different one. A specific color that predictably changes its perceived hue under these different light sources is called a metameric color.
Metamerism is a known and measurable issue for production proofing and printing. In some workflows it is especially a concern, for example in color critical printing such as skin colors.
Reducing metamerism in the color workflow
Research has shown that one can calculate the degree of metamerism that might be present in a color workflow. Global Graphics’ scientists have done research to determine the severity of metamerism and have solutions within their RIP technology to address it.
By properly engineering the color setup, knowing what settings to tweak in the Harlequin RIP, this known phenomenon can be controlled or at least limited.
Color matching on proofs
The problem is that proof prints are designed to match the result that will be obtained on a future printing press run but typically use different inks or dyes.
Today’s printers include many output options such as inkjet - dye, pigment and solvent inks, dye-sublimation, toner-based and even wax or solid ink devices, and so have very different ink sets in terms of their spectral characteristics.
Metamerism can be a problem here if proofs are carefully matched in a viewing booth but then appear to be different when viewed under another a light source.
Consistent lighting
A real world classical example of this was a magazine editor outside of New York who was sent proofs by courier twice a day for a series of travel magazines. Typically proofs sent in the morning were approved 95% of the time whereas proofs sent with the afternoon courier were rejected 30% of the time.
Most often the editor just viewed the proofs in her office, despite having a D50 viewing booth in the production department the floor below.
It transpired that the indirect sunlight in the morning was a reasonable approximation to D50 but the afternoon light was reflecting off the red brick wall of the next building which resulted in a warmer color.
The solution was to purchase a small viewing booth for her office that used the same bulbs as the viewing booth in the pre-press production facility where the proofs were made.
Testing for Metamerism
There are 3 components that go into the calculation of XYZ values for a given color. These XYZ values are known as CIE Tristimulus Values representing the reflectance response for a color sample.
The product is then multiplied by the CIE Standard Observer (an average of color sensitivities of test subjects with normal color vision to a range of presented color stimuli) resulting in the CIE XYZ values for the given sample.
Measure against the color target
A test experiment can then be run whereby a press sheet test form with a color target with x number of patches is printed and measured and the same target is then printed on the proofing system and again measured.
These two data sets can then be analyzed to determine and look for metameric colors for the chosen illuminate D50 but also by replacing the illuminant in the calculation of the CIE XYZ values for a given color sample can also see what will happen to any color should the illuminant change, in essence being able to predict metamerism for any number of illuminant changes.
Fig1 shows is the range of color errors for a press sheet that was printed using a moderate amount of GCR and an inkjet proofing system using pigmented inks changing the illuminant from D50 to F8.
If we accept that for most colors (excluding neutrals for which a tighter criteria is warranted) Delta E*76 of 2 will be a noticeable difference then for this example 30% of the colors for the press and 50% of the colors for the proof will show color differences. This suggests something needs to be done to control this variable if possible.
Secondly, this chart does not show where in the color space the errors are occurring. What if all of the errors were in say the blue region? Fig 2 shows a more complete picture of what colors are being the most affected by changing illuminants in this example comparing D50 to an actual light box viewing illuminant. Note the Y axis C* is describing the color change removing the lightness component. The important thing to note is that all colors do not respond the same to an illuminant change.
Color management
Research has shown that if a given proofing workflow does require that the proofing device be different from the press then metamerism is a real concern. The best approach to limiting this known effect is to use color management and to use the maximum amount of black ink as possible on press and on the proofer. The way this is done for the proofing workflow in the Harlequin RIP is via the rendering intents.
Based on the nature of the screening, the number of black inks, the resolution selected for the proofer, and the quality of the printer profile being used, in many cases these settings can be set quite high. Black Threshold is set to reflect when black ink can be used to make up a color on color scale from 0-100.
Obviously for fair-colored skin using black ink in the shadows around the eyes and neck may be visible so the black threshold would need to be adjusted to minimize this issue but a few tests can fix this. Black Generation is similar to UCR/GCR but in a color managed workflow uses profile data to determine the amount of black to use.
Being able to use Heavy or Maximum here greatly reduces Metamerism for many inkjet ink sets so is recommended.
