Getting the Most from Simulated-Process-Color Reproduction
Coudray discusses the influences that ink values, software, and other variables have on the creation of effective simulated-process separations.
Simulated-process reproduction offers so many advantages compared to other methods, particularly the fundamentally unstable four-color process, that there is really not much of an argument anymore for using it in favor of other techniques. But that doesn’t mean that simulated process isn’t without its challenges and limitations. This month I’d like to look closely at some of the fundamental issues printers face when they choose to use this approach to full-color reproduction.
Is it the seps?
At the very front of the process is the question of whether your simulated-process separations are adequate or not. Like traditional four-color process, any halftone method is a simulation of what the eye sees and the brain processes. Simulated-process or nChannel separations, where n represents the number of printed colors, physically limits us to replicating tone and color variation through the use of various dot sizes or the number of dots in a give area. There are dozens of ways to make nChannel separations, and they vary from terrible to incredible.
Most of the common approaches use Adobe Photoshop and involve RGB to CMYK conversion and then manipulation of both the RGB and CMYK channel data through scripting, color-range extraction, channel masking, layer masking, and channel calculations. Lesser approaches, such as inversions, difference calculation, and filter enhancement, also may be used.
The more sophisticated methods involve nChannel or nColor profiles, in-RIP separation of an RGB image, or separation through dedicated software that’s designed specifically for the purpose of nChannel color mapping. These methods deliver extremely accurate and excellent separations, but they are considerably more expensive than using Photoshop or a Photoshop plug-in. All of the Photoshop approaches have one thing in common: they’re pretty much guess work based on the skill of the operator. Don’t get me wrong—some amazing separators out there can work Photoshop magic with these methods. But that still does not change the fact they are using their experience and judgment to create grayscale channel data that will deliver a good to great nChannel result.
It’s not my intention to analyze the various techniques, but rather to focus on what happens after the seps are delivered and the job is on press. Press operators resort to all kinds of gyrations to try and find a way of making underperforming separations work. Typical manipulations include changing print order, mesh count, ink colors, squeegee durometer, pressure, angle, hit sequence, and print speed. This time-consuming, hit-or-miss approach mostly yields mediocre results—or worse, the job is pulled and sent back to the art department for new seps, screens, and round two on press. It’s a frustrating and expensive waste of time.
There are two primary approaches to simulated-process separation. The first involves the premise that ink colors are opaque and will not optically mix, but instead rely on physical ink mixing to form secondary or tertiary colors. This approach tends to be very forgiving on press as there is no requirement to control pressure and only limited requirement for control of dot gain. The only gradation that occurs comes from the physical mixing of the ink dots. The dots often completely disappear, which gives the impression of near-continuous tone and makes it considerably easier to get good—and even outstanding—results under almost any printing condition.
The trade-off in this case is that you need as many printheads as you can get your hands on. It’s rare for this type of separation to work with fewer than ten colors, and most outstanding work is accomplished with 13-16 colors. The opaque version of simulated process is little more than a modified posterization, where tonal transitions are softened with some halftone blending. Close examination of the final print almost always reveals some obvious tonal banding and issues with holding fine detail.
The second type of nChannel separation is more complicated. It’s based on the assumption that inks do mix both optically and physically to deliver a much broader range of tone and color. With this approach it’s possible to deliver incredible detail, tone, and color. The color gamut is much, much greater than any CMYK gamut, and is typically accomplished with fewer than eight colors.
As with the first type, there are compromises. In the second case, the compromise comes in the form of increased press control. Dot gain is just as much a factor as with any halftone-based printing method. Additionally, color choice for the selected inks is very important. This is one of the main reasons why simulated process almost always relies on the use of some color-mixing system, typically Pantone, to designate the ink colors as accurately as possible.
Why your choice of ink is so important
Ink selection has much more of a bearing on your success than you might imagine. With the exception of dot gain it’s the most influential factor in final color reproduction. You’ll have an easier time diagnosing separation problems when you understand how the ink works in conjunction with the separation. The root issue is whether the inks you use are capable of delivering the color reproduction based on how you created the separations.
The goal of any separation is to reproduce the broadest range of tone and color with the fewest ink colors/printheads as possible. Traditional four-color process is the lowest common denominator for reasonable to excellent full-color reproduction. It works because ink transparency is fundamental to four-color-process printing. This transparency allows for excellent physical and optical mixing of the halftone dots. The problem is controlling the thickness of the ink layer. Variation in thickness changes the color, which means print pressures and platen height are absolutely critical in order to maintain consistency.
Other printing processes, such as inkjet, toner, traditional offset, and flexography, involve very thin ink films—often just 2-3 microns; however, screen printing deposits anywhere from six to 20 times the thickness of the other processes. This dramatically greater ink thickness is the cause of huge color variations and is a limiting factor for the screen-printing process. This sensitivity to ink-film thickness is one of the reasons simulated-process printing evolved.
Transparency, translucency, and opacity
On the one extreme we have perfect ink transparency in the form of four-color process; on the other, the very limited color mixing of 100%-opaque inks that require many, many printheads. Achieving perfect transparency or opacity is not possible, but you can design and control a wide range of translucent colors through simulated process—the perfect compromise between the extremes.
The object of simulated process is to choose ink colors that are transparent enough to optically and physically blend but opaque enough to eliminate the need for critical pressure and platen-height controls. Therein lies the basic problem you must address at the color-separation stage and on press.
The natural optical characteristics of each pigment is unique. Some are more transparent and some are very opaque in their natural form. The more transparent a color, the better it will mix with other colors. One of the measures of this characteristic is the tinctorial value of the pigment, which refers to the ability of a color to stand up to dilution with white without losing its color strength. The more transparent a color, the lower the tinctorial value. This is very important when choosing our ink colors.
One of the fundamental flaws within the Pantone Matching System is that it doesn’t address the tinctorial aspect of each pigment. As proof, simply look at the book and you’ll see the huge jump in tone value between Reflex Blue and 279 or between 293 and 292. This indicates a very low tinctorial value of the blues. Compare the same tonal separation between 1925 and 1915. In both formulas, 12 parts of trans white were added to four parts of color. It is clear the reds can hold up better to the addition white.
Now we come to the very important part. When two colors are mixed together (physically), the more opaque color—the one with higher tinctorial value—dominates the resulting secondary color. In other words, when you mix red and blue the resulting purple will have a noticeable shift to the red side. Because the Pantone System, as well as Photoshop, ignore this important component, our ability to reproduce accurate secondary colors is left entirely to us. Results are approximate at best unless the separator adjusts for the known translucency of each ink. Likewise, if the separator uses a value that does not match the value of the ink on press, the results will be inaccurate.
Photoshop does allow for a limited approach to the problem. When you create new channels and named as spot colors, you can designate the Solidity value. This is close to an opacity value—it does not display quite right based on how the ink actually prints. Acceptable values here are 25-75%. Best mixing results on press occur when the values are set to 25-35%. Beyond 35% you begin to see domination by the higher value.
To accurately determine a color’s value, simply print a solid, 2-in. patch of the color on black substrate. With your background set to black in Photoshop, fill a spot-color channel with the desired ink color and adjust the Solidity until what you see on the screen matches what you printed. This is not perfect, but you will be closer than if you had made no correction at all. By adjusting the Solidity value, the separator can now make additional channel corrections if necessary. Print order on press is also very important in terms of varying tinctorial value of each ink color. Remember that the more opaque ink dominates the transition. Therefore, you should use the more transparent colors to overprint the more opaque colors.