Ask the Expert Question-and-Answer Archive

by Ron Joseph

February, 2005

Converting RAL and Pantone colors to CIE Color Coordinates

Q. I need to know what is the (x,y) value of CIE 1931 diagram of the RAL 7024 color. Could you give me some further information about this issue, please ?

For information I found the following information: RAL 7024 has the following CMYK values: 80 60 50 40. Is it correct? (corresponding to pantone 432C).

Thanks a lot for your time.

A. There is no short or simple answers to this question. Here is why:

1. The "xy" coordinates for a printed or painted reflective color patch are calculated from the spectrum of the light (illuminant) used to look at them. There is a set of coordinates for each illuminant; a tungsten bulb will give you different coordinates than the ones obtained with daylight, and also different than the ones obtained with a fluorescent fixture.

2. CMYK values are determined from a mix of conditions. The actual CMYK transform profiles used vary by country (often Euroscale in Europe, SWOP in the USA), by process (newsprint, glossy), by user preferences, and by the rendering intent (Perceptual, relative Colorimetric, etc.) in which the transform was based. There is no way to convert back to colorimetric coordinates unless you know all these variables. L*a*b* data (CIELAB), associated with info on the illuminant and Observer (i.e. CIE 1931-2 degrees or CIE 1964-10 degrees) used for measurement, is always preferable.

3. RAL 7024 (Graphitgrau) is a Classic RAL color. The CMYK data you provided and Pantone 432C are certainly related, a dark-bluish-green grey, but not close. The Pantone 432C L*a*b* coordinates for Illuminant D50 (2 deg. Observer) are (32,0 / -2,4 / -7,4). The L*a*b* D50 coordinates for CMYK (80-60-50-40) based on Euroscale Coated v2, Relative Colorimetric, and Black-Point Compensation, is (27 / 0 / -7). There is a large difference of 5 (5%) in L* (perceived brightness). Other conversion settings will give more or less variance from these values.

The corresponding D50 "xy" coordinates are somewhat similar (0,311 / 0,334 for the Pantone chip, and 0,317 / 0,328 based on the CMYK data). This means that the two data sets have similar hue (tint) and chroma (saturation). However, as mentioned previously, these coordinates can be completely different when computed for other illuminants (D65, C, etc.).

In conclusion, "xy" data is a moving target. With the info provided, I can get a relatively precise "xy" data set for a given illuminant, but the perceived brightness is imprecise. A more precise answer for the "xy" coordinates would require the illuminant specified for viewing, and L*a*b* coordinates of the reference color (with illuminant and Observer info). An even better match could be obtained by scanning a small sample of the required paint color (AND knowing the Illuminant); doing so would prevent any metamerism effect due to the different colorants used in the "reference" patches and the actual paint.

We could perform this analysis if need be.

Best regards,

Danny Pascale