Chromaticity Overview

The color gamut perceived by a person with normal vision (the 1931 CIE Standard Observer) is shown in Figure 1. The diagram and underlying mathematics were updated in 1960 and 1976; however, the NTSC television system is based on the 1931 specifications.

Figure 1 - CIE 1931 chromaticity diagram showing various color regions

Color perception was measured by viewing combinations of the three standard CIE (International Commission on Illumination or Commission Internationale de I’Eclairage) primary colors: red with a 700-nm wavelength, green at 546.1 nm, and blue at 435.8 nm. These primary colors, and the other spectrally pure colors resulting from mixing of the primary colors, are located along the curved outer boundary line (called the spectrum locus), shown in Figure 1.

The ends of the spectrum locus (at red and blue) are connected by a straight line that represents the purples, which are combinations of red and blue. The area within this closed boundary contains all the colors that can be generated by mixing light of different colors. The closer a color is to the boundary, the more saturated it is. Colors within the boundary are perceived as becoming more pastel as the center of the diagram (white) is approached. Each point on the diagram, representing a unique color, may be identified by its x and y coordinates.

In the CIE system, the intensities of red, green, and blue are transformed into what are called the tristimulus values, which are represented by the capital letters X, Y, and Z. These values represent the relative quantities of the primary colors. The coordinate axes of Figure 1 are derived from the tristimulus values:

x	= X/(X + Y + Z)
	= red/(red + green + blue)
y	= Y/(X + Y + Z)
	= green/(red + green + blue)
z	= Z/(X + Y + Z)
	= blue/(red + green + blue)

The coordinates x, y, and z are called chromaticity coordinates, and they always add up to 1. As a result, z can always be expressed in terms of x and y, which means that only x and y are required to specify any color, and the diagram can be two-dimensional.

Typically, a source or display specifies three (x, y) coordinates to define the three primary colors it uses. The triangle formed by the three (x, y) coordinates encloses the gamut of colors that the source or display can reproduce. This is shown in Figure 2, which compares the color gamuts of NTSC, PAL, HDTV and typical inks and dyes. Note that no set of three colors can generate all possible colors, which is why television pictures are never completely accurate.

Figure 2 - CIE 1931 chromaticity diagram showing various color gamuts

In addition, a source or display usually specifies the (x, y) coordinate of the white color used, since pure white is not usually captured or reproduced. White is defined as the color captured or produced when all three primary signals are equal, and it has a subtle shade of color to it. Note that luminance, or brightness information, is not included in the standard CIE 1931 chromaticity diagram, but is an axis that is orthogonal to the (x, y) plane. The lighter a color is, the more restricted the chromaticity range is.

The chromaticities and reference white (CIE illuminate C) for the 1953 NTSC standard (and assumed by the ITU-R BT.601 equations) are:

R:	xr = 0.67	yr = 0.33
G:	xg = 0.21	yg = 0.71
B:	xb = 0.14	yb = 0.08
white:	xw = 0.3101	yw = 0.3162

Modern NTSC, 480i and 480p video systems use a different set of RGB phosphors, resulting in slightly different chromaticities of the RGB primaries and reference white (CIE illuminate D65):

R:	xr = 0.630	yr = 0.340
G:	xg = 0.310	yg = 0.595
B:	xb = 0.155	yb = 0.070
white:	xw = 0.3127	yw = 0.3290

The chromaticities and reference white (CIE illuminate D65) for PAL, SECAM, 576i and 576p video systems are:

R:	xr = 0.64	yr = 0.33
G:	xg = 0.29	yg = 0.60
B:	xb = 0.15	yb = 0.06
white:	xw = 0.3127	yw = 0.3290

The chromaticities and reference white (CIE illuminate D65) for HDTV are based on ITU-R BT.709:

R:	xr = 0.64	yr = 0.33
G:	xg = 0.30	yg = 0.60
B:	xb = 0.15	yb = 0.06
white:	xw = 0.3127	yw = 0.3290

As previously illustrated, the color primaries and reference whites of the modern SDTV and HDTV standards are very close, and would not be an issue in most cases.

Unfortunately, the BT.601 equations, commonly used for SDTV (NTSC, PAL, 480i, 480p, 576i and 576p) content and displays, are based on the color primaries and reference white of the old 1953 NTSC specification. As a result, minor color errors can occur when displaying SDTV programs on a HDTV, or displaying HDTV programs on a SDTV, as illustrated below.

To compensate for the color errors due to chromaticity differences between standard-definition (SD) and high-definition (HD) content and displays, our solutions incorporate the ability to do "chromaticity correction" for each video or graphics plane individually, and each video output port individually. This technology ensures the picture is displayed as accurately as possible, regardless of the type of source or display.