Basic Concepts in Color Calibration

While I wrote this article a few years ago for a class I was teaching, I have found people still refer to it. Thus, I am including it and updating it. Of all the How to Use Graphics articles I am including on this web site, this one had to be changed the most because monitors, programs, and video card technology have improved radically in the filed of color as well as in the field of software color calibration.

Color management is probably one of the most frustrating areas for many people. Since I do not work on Macintosh computers, I will not address them in this discussion nor will I go through a Ahow to calibrate your system as a whole@ because each system is different and each user=s goals are different, also. I will, however, share some of my own experiences on various systems and programs. Hopefully this will be helpful.

While I know this will annoy some authors, I have read many articles on color calibration and, perhaps, in a prepress house, they are valid. However, I have found them extremely confusing and necessarily tedious. I have been using Epson Color Stylus Photo printers for years. I have an Alps printer, as well, that I use for certain effects. In the past, I never had trouble with my outcome if the particular graphic program either could be manipulated manually. Recently, with the advent of Photoshop 6, I have not had any trouble at all just leaving the settings generic.

The three components that should agree with each other are the input devices such as scanners and digital cameras, output devices such as printers, and of course, the monitor. The monitor can be calibrated with a hardware calibrator or visually or through pre-packaged software, or hardware created profiles. Basically,  input and output devices can be calibrated through profiles or visually. Each of these components will be addressed individually.

Years ago, when I first started in this field, I wanted to know what standard I needed to use in order to calibrate my monitor. I called service bureaus and pre-press houses and they told me that I needed to know the final output device. When I told them I didn=t know what I would be using, they basically told me that they couldn=t help me. Thus, I designed my own system which I have found works. I work in RGB space and do not worry about CMY color separations. Since I am not designing for others, client=s colors are not an aspect that concerns me nor do I have to match a particular output device that is connected to someone else=s machine. Most professional level graphic programs allow one to select printers=s profiles and a service bureau can further instruct the designer on what color modifications they must make.

When calibrating a monitor, the following specifications can be gotten from the manufacturer: The color temperature of the picture tube, the x and y values for the RGB values of the picture tube=s phosphors, and the gamma of the monitor. Some monitors already have a profile integrated into the program. While this sounds complicated, it is and it isn=t. While I hate generalizations, I have found that monitors today display color fairly accurately provided that there is a good video card in the computer. Also, monitors and/or video cards usually have ways to calibrate the screen image. A good monitor and the video card do not usually cause the problems in calibrating one=s monitor. Usually the cause is the individual program that is being used and/or lack of understanding of how color works on the part of the user. A few graphic software versions ago, one could calibrate one=s monitor visually within the particular graphic program. One could change the values of blue, red, cyan, etc. within the program itself. (This would not cause a global change as does calibration using monitor or video card calibration.) One could either compare this to an outside source or to the output of one=s printer depending on the software program. When I wrote this article, initially, I found the following statement to be true that too many programs come with built in profiles for the users to select to theoretically calibrate their system, and in my personal opinion, that is where the problems occurred. However program versions since the year 2000 have been much better in the field of color calibration. The default settings are usually pretty good for the home user.

Other conditions, also, influence color visualization and, thus, output. Firstly, assuming that one has two identical systems with equal room illumination, no two monitors will show colors the same. Secondly, as monitors age, the phosphors change, the picture tube gets duller and colors change. None of this is a large problem if one has calibration hardware or if it is built into the video card or monitor itself.

Scanners, printers, and digital cameras don=t usually come with advanced software programs for calibrations. Some flatbed scanners do and some programs have setups for scanner calibration. Many printers have a minimal set of options to change color output.

Another aspect that must be taken into consideration is the light under which one is working and the light under which the picture will be viewed when printed as hard copy. When transmitting images over the web, different browsers, computer systems, etc. become a factor, also. Many graphic software and web image generating programs allow the user to preview how the image will look under different conditions such as under a Windows or a Macintosh system environment or when using different color palettes.

My goal is to be able to use any printer regardless of its particular quirks. I need to know that the color on my screen will not only print as seen, but is, also, "correct" and can match some type of color standard. For this purpose I have chosen to use Trumatch color swatches as my standard. However, Pantone swatches or any others for which a CMYK color formula is available will work fine. One can purchase these at graphic supply store, through mail order catalogs, and at some service bureaus. A package costs around $65.00 to $85.00

The following method I have found has worked for me as I have changed computer systems, programs, video cards, printers, and monitors.

1) I make boxes on my screen and fill them with 100% colors for Cyan, Magenta, and Yellow. I have coated swatches from Trumatch that correspond. (I can also do this from the color library in Photoshop since all the swatches are numbered and have equivalents in the Photoshop library.) The key is to have an outside source. I work in RGB space not CMYK color space. In other words, I set my mode to RGB.

2) Next, I make boxes and fill them with Red, Green, and Blue. I also take these formulae from Trumatch swatches.

3) I next make a series of boxes and fill them at 10% intervals starting with 100% Black and ending with 0% Black so that the first box is filled with 100% Black, the next with 90% Black, etc. For these fills, I make sure that Cyan = 0, Magenta = 0, and Yellow = 0. The only variable manipulated is K or Black.

4) I, then, make more boxes and fill them with color blends, once again, taken from my swatches.

If the colors on the screen match the Trumatch swatches that I have used taking into account ambient (room) lighting and the difference between the monitor's RGB color space and the printer's inks that use the CMYK color model, I, then, print my results on the paper type I will be using for the majority of my work using the corresponding printer's setting. Often, at this point my job is done. If the colors on my screen do not match, I try to correct them. How one does that depends on the program being used and the hardware.

If I can manually tweak, color value, monitor gamma, etc., I usually can achieve a good result. The biggest problem occurs when all one has to work with are prepackaged hardware profiles. What I have discovered is that I eventually can find a combination of hardware that works even if I do not own the hardware. For example, in one major program, I had to invoke the profile of a separation printer that I did not even own so that the colors on my screen matched my outside source and the printed output matched this outside standard as well. Again, this did not occur with the programs that I used that were released since 2000.

I review software for major manufacturers, and, thus, have used most of the major products over the past few years. The system I have outlined has worked on most of the major software image editing, painting, and drawing programs.

Earlier I mentioned ambient room light and how it can affect the viewed hard copy. I work under daylight fluorescent lights. I, also, often view my work under tungsten light. The color values of my work will appear different. The critical factor is not a number, but to know that there is a difference and to understand this difference when working.

To see my work, I usually set it under a tungsten light near a window. This gives me a combination of light since I do not know how the work will always be viewed.

Related Articles: