5.1 The total perceived color difference between two non-self-luminous specimens is compared as an equivalent lightness difference between two neutral gray specimens on a gray scale. A fundamental assumption is made that the total color difference can be so evaluated in terms of an equivalent lightness difference. Only the total color differences, that is, a summation of the differences in hue, lightness, and chroma between two specimens is evaluated; this test method is not applicable to the separate precise evaluation of the hue, lightness, and chroma components of color difference.

5.2 The total color difference determined by this test method depends on the degree of uniformity of the specimens and on the sharpness of the dividing line between them. The color difference between specimens having rough or mottled surfaces appears smaller than it would if the specimens had smooth and uniform surfaces. Thus the equivalent CIELAB lightness difference determined for non-uniform specimens will be smaller than for uniform specimens. Likewise, specimens whose dividing line is not sharp will appear to have smaller color differences than those with sharp dividing lines, and for this reason, the equivalent visually observed CIELAB lightness differences will be smaller than the color differences obtained from instrumental measurements.

5.2.1 A physically sharp border between colors differing slightly in the yellow-blue direction in color space appears diffuse. The perceived color difference is noticeably increased by a hairline black separation. This technique imposes a more rigorous test of such small differences.

5.3 In the CIELAB system, a unit of color difference is intended to represent the same visual difference in each of the three attributes; lightness, hue and chroma or alternatively lightness, redness-greenness, yellowness-blueness. It is valid to express color differences that are not simply lightness differences by comparison to a lightness-difference scale.

5.4 Personnel to be employed in the evaluation of color differences with the paired gray scale should be tested for color vision using the procedures in Guide E1499.