What does the term colorspace mean? It is the term applied to the way color is described, and there is more than one way to describe color. Colorspace can also be used to describe the different colors used to render color gamut. Color spaces include the descriptions of color such as RGB and CMYK as well as types of RGB descriptions: Lab, HSB, Index, Multichannel, and YCC. This dialog is intended to be a layman’s description of color space; refer to any number of qualified texts for technically accurate descriptions of color. X-Rite’s "The Color Guide and Glossary" is excellent, and free; look them up on the world wide web for additional information.
RGB color is what we use to capture color. CMYK is what we use to render color on the printed page. Why the two different color spaces? Because R+G+B=white and C+M+Y=black. RGB refers to colored light energy and CMY refers to pigments or colorants we use to render images to reflective targets.
RGB color is typically read in 0-256, where 0 is black and 256 is white. There seems to be no real reason why this numbering system is in place other than the system is based on factoring the number 2, which computers count very efficiently. RGB is considered color that we can capture and is a smaller color space than what human vision can see. RGB is also considered transmissive and emissive (in addition to reflective) which allows a density range considerably in excess of what can be printed. Continuous tone color transparency film uses the RGB color space as do monitors, digital cameras, and scanners.
Profiled RGB is a calibrated RGB. It is a color space that you will hear more of in the next year or so. Profiled RGB is a color space that is defined by colors on a target that have known descriptions. Currently the target for defining Profiled RGB is the IT-8 target, which most of you know as the Kodak Q-60. The Q-60 has color patches on it that have known, spectrophotometer validated descriptions. Profiled RGB color is RGB color that is software corrected, so that if a Q-60 were in the scene, the Q-60 would render correctly with respect to the known values for all of the patches. There is a problem with this colorspace: the problem is the limited color gamut of the target. Since the IT-8 target is manufactured with traditional silver halide dyes (ektachrome and ektacolor print media) the colors in the target do not comprise a large color gamut; we’re limited to the colors that can be rendered by silver based photographic materials. Recently, the Macbeth ColorChecker has been added by some software profiling applications--Colorblind and Kodak come to mind.
CMYK is a color space we use to render RGB or sRGB images onto a reflective target and we usually measure CMY in percentages of 0-100. Because the target does not allow light to come through the rendering, it is necessarily smaller in density range. That makes some RGB colors unavailable to the CMYK rendering, especially those that are low valued and highly saturated. CMY is a subtractive color space that is complementary to RGB, its colors come from the overlap of the RGB colors (see illustration above). We use this color space because as suspension based colorants the addition of C+M+Y equals black (almost). In our environment, CMY is used for inks and dye sublimation ribbons. Because C+M+Y in 100% quantities equals almost black (actually a very dark brown), we add a little black ink to the lowest values to bump those values to neutral lows and black.
Lab is a type of CIE* color space description. It is used to describe RGB or sRGB where the luminance (density) is described separately from the chrominance (color). This separation is particularly efficient for adjusting color and for targets of different density ranges. L is the luminance portion of the file Photoshop calls this channel Lightness). a is the channel whose color description describes from red to green and b is the channel whose description of color runs from blue to yellow. Color adjustments are more efficient in Lab because we can change the color without affecting the density or contrast. We can also change the contrast or density without affecting the color. For the same reason why Beta videotape really was better than VHS, Lab is better than RGB. Lab is Photoshop’s native colorspace and it allows several advantages over the RGB description. Jpeg for example uses Lab to compress the a and b channels while leaving the luminance largely intact.
Because we’ll probably not use HSB, YCC, Index, or Mulitchannel descriptions of RGB we’ll leave them for some other time. If you are interested, see the Photoshop manual for additional information.
MegaVision has pioneered a calibrated colorspace that uses the Macbeth Colorchecker for its CIE coordinates. We call this calibrated colorspace CalRGB. The Colorchecker has a larger gamut than the IT-8, largely because its colors are rendered with the silkscreen process rather than using silver halides with dye couplers and organic dyes. MegaVision’s T2 makes CalRGB through the Convert button in the Function window. We believe enough in this colorspace that we’ve chosen it as our default colorspace on our single pass cameras, the S2 and the S3.
Colorspace is going to be more important as applications mimic Photoshop 5.0’s ability to bring our files into a Working Space for profiled export to device specific targets. We’ll update this important change in the way we work with files in the near future. Stay tuned to this topic for additional information as it becomes available.
* CIE: Commission Internationale de l’Eclairage. The French body that first investigated how humans see color and how color could be most accurately described and measured.