Prepress Tutorial Wide Open Windows

Prepress Tutorial Wide Open Windows PrePress UCR/GCR | Screens | Moiré | Dot Gain | Densitometry UCR/GCR When producing gray values with four ink colors, substantial savings can be realized by replacing portions of the Cyan, Magenta, and Yellow with black ink. Two methods are used to do this, Under Color Removal (UCR), and Gray Component Replacement, (GCR). With UCR, a single black ink is used for any shade of gray produced by C+M+Y in shadow portions of an image. Both images you see here are from Photoshop 5 's CMYK setup. GCR goes a step further and substitutes black for equal components of C+M+Y in any portion of the image. With this 1-for-3 replacement, ink use is reduced by 2/3 for this portion of the image. The black used is less expensive than the replaced colors, easier to recycle, and more plausible to be a recycled ink itself. Less ink coverage allows for easier gray balance, better run stability and color consistency, faster make-ready, more consistent and faster ink drying and faster printing. The images produced show cleaner and more consistent colors, improved detail and sharpness. UCR/GCR | Screens | Moiré | Dot Gain | Densitometry file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (1 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Screens Digital output options There are two basic methods for achieving output for digital images. One is direct digital imaging, in which each pixel of an image corresponds directly to an output device element. The other method involves dithering*, in which the four colors cyan, magenta, yellow and black are composed to give the impression of continuous tone. Each method attempts to produce the appearance of a continuous tone image but all printing processes use some sort of element to produce color and tone. (*Dithering is the technique of arranging pixels in a pattern to reproduce tonal value.) Dither patterns can be ordered or disordered. Ordered dithering produces a pattern that is predetermined and specific. Disordered dithering employs a certain degree of controlled randomness in the dither pattern. The dithering process relies upon the halftone cell. The halftone cell controls the placement of pixels within the cell which in turn simulate color and tone. Some of the available output options are: halftone screening, stochastic screening, continuous tone and contone. Halftone Screening (Amplitude Modulation or AM Screening) For the last 100 years, color printing has been based upon halftone screening. Halftone screening uses halftone cells (which are comprised of different sized dots) arranged in a grid pattern to create the illusion of light and dark areas. This conventional halftoning technique is referred to as amplitude modulation because the size or amplitude of a dot is changed or modulated to create different tonal values. The single dot within the halftone cell grows larger as the tone value becomes darker and smaller as the tone value becomes lighter. The center from one halftone cell to the next is always the same. The spacing of dot placement is controlled by the line screen which is referred to as lines per inch (lpi). The higher the line screen the more continuous an image will appear. For example, halftone dots will be visible with a 60 line screen and invisible to the naked eye at a 150 line screen. The four color process screens (cyan, magenta, yellow and black) are usually rotated at different screen angles. These angle rotations create the traditional rosette pattern which can be seen at low line screens. Repetitive patterns that occur are normally called artifacts. Line screen and angles sometimes create unwanted moir patterns. Most often these moir patterns occur with checked or herringbone patterns that conflict with a screen angle or by screens that are poorly reproduced. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (2 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Laser printers use a matrix of imaging elements to create the halftone dot. To determine the matrix, divide the dots per inch (dpi) of the laser printer by the intended line screen. For example, a 300-dpi printer combined with a 100 line screen would use a matrix of 3 x 3 image elements per halftone dot. The number of image elements per inch that a printer or imagesetter can produce is known as the device resolution. As the number of imaging elements per inch increases so does the quality of the halftone dot. The elements per inch combined with the line screen controls the number of gray levels that can be achieved with an output device. For example, a 1200 dpi laser printer using a 150 line screen would image 150 dots per inch and every inch would contain 64 imaging elements. Therefore each dot is created by a halftone cell that contains 8 x 8 imaging elements. An 8 x 8 cell contains a total of 64 on/off imaging elements. This 8 x 8 cell can potentially produce 64 levels of gray. The pixels per inch required for a bitmap image are dependent on the line screen to be used. The ratio of the bitmap image resolution to the output device is 2:1. The most simple method for determining pixels per inch is simply doubling the line screen. For example, when printing with a 100 line screen, the bitmap image should contain 200 pixels per inch. Users should be aware that increasing the pixel information greater than the 2:1 ratio does not increase the output quality and generally wastes file space and increases RIP (raster image processing) time. Raster Image Processing, also know as RIP or render, refers to the conversion of digital information into physical printed output. Rational and Irrational Screen Angles Rational screen angles contain halftone cells that are always the same size and shape. These halftone cells address tone uniformly across an entire image. The downside of rational angles is that the number of line screens and screen angles are limited by the output resolution. This makes it difficult to avoid moir and artifacts unless large halftone cells are used and printed at low line screens. An example of a rational screen angle is the traditional rosette screen angle pattern. An alternative to rational screening is irrational screening. Irrational screening uses non-uniform halftone cells that are different in shape and size. These non-uniform cells allow any screen angle to be used with any line screen. To present a consistent response to tonal values, predetermined spot functions are assigned to different tones. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (3 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Stochastic Screening (Frequency Modulation, FM Screening) Stochastic or frequency modulated screening uses very small dots of the same size which are placed at random to create color and tone. FM (frequency modulation) dots create tonal value by varying the number or frequency of dots, whereas AM (amplitude modulation) halftone screening varies the dot size to create different tones. Hence the terms frequency, which refers to the number of given dots in an area, and modulated, which refers to the density of the dots relative to the tonal value of the input pixels. Stochastic screening has the ability to adapt to image content. This significantly increases image detail. Stochastic dots are typically 1% to 2% of halftone dot size. The word stochastic was derived from the Greek word stochos meaning to guess and is used to describe processes in which the state of a variable is determined by random factors. FM screening is based on the random placement of dots. As a result line screen, halftone grids, rosette and moir patterns of AM screening disappear. FM screening increases the number of dots to generate dark tones and decreases the number of dots for light areas. If you remember, AM screening increased the dot size for dark areas and decreased the dot size for light areas. The relationship between the bitmap image to the output devices FM screening is 1:1. The FM dot is more closely related to continuous tone than to the standard AM halftone dot which requires a 2:1 ratio. The increased detail available with FM screening carries the added benefit of being able to use bitmap images with as little as a 1:2 ratio when printing with a 600 dpi device. A problem of past FM screening has been that some offset presses and proofing systems have trouble holding this very small stochastic dot. The second generation of FM screening uses a cluster approach which combines very small micro-spots into large micro-dots. This dot-cluster approach was developed to minimize the difficulty of plating and holding these tiny FM dots on conventional offset presses. This approach also reduces graininess in highlight areas. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (4 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Continuous Tone (Direct Digital Imaging, CT, Contone) Continuous tone is defined as output in which the cell is completely filled with color and tone, leaving no white in the cell. Continuous tone printers produce the illusion of a smooth continuous image without using the halftone dot and the primary colors. Continuous tone matches each bitmap pixel with a dot on the output device at a ratio of 1:1, also called direct digital imaging. If you have a color printer that has a resolution of 300 dots per inch (dpi) it would output one inch of bitmapped data at 300 dpi. For instance, a bitmap image at 300 pixels per inch (ppi) that is 1200 x 1500 pixels would print at a size of 4 x 5 inches. Should the image resolution be changed to150 ppi, pixelization would occur and the image output would loose its continuous tone appearance. (Pixelization occurs whenever the image resolution is less than the output devices full resolution.) An example of continuous tone output is that from a dye sublimation printer. Continuous tone output can also be achieved without this 1:1 ratio by using line screens to achieve the additional gray levels. The introduction of high resolution color laser printers has brought the ability to render multi-bit pixels on laser engines. These engines rely upon screening implementations and very small dots to achieve a continuous tone simulation. Near photographic quality can be achieved using this method. FM screening as well as AM screening may be utilized for producing images. Resolution Resolution is a way of describing images that are composed of pixels. An image appears to be continuous based upon its number of pixels and its resolution. In order for an image to be continuous, one must not be able to see the individual pixels that were used to create it. If resolution of an image is less than required by the output device, individual pixelswill appear as jagged edges. will appear as jagged edges. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (5 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows A printed magazine image will look continuous from reading distance, but when viewed with a magnifying glass will show the individual dots within the halftone. Therefore, distance affects the continuous appearance of an image. Resolution of the output device is tied to the number of elements per inch it can address to produce a dot. As the elements per inch increase, more information is available to produce a better quality dot. The pixels in the example are shown as squares to represent the address grid. Pixels actually produced are usually round or oblong. Meeting resolution requirements of the output device is critical to the quality of the image. The number of pixels an image to be rendered needs is directly proportional to its output. When the resolution of the bitmap image matches the output resolution correctly you will not see the individual pixels. The way in which the pixels of a bitmap image relate to the output device is called the sampling ratio. For example, the ratio of bitmap images to halftone dots is 2:1. Screen Frequencies Imagesetters create halftone screens using screen frequencies, measured in lines per inch (lpi). A screen frequency can be represented by a grid. Each square in this grid is a halftone cell, capable of holding one halftone dot. Think of each halftone screen as a grid that is superimposed on the image recorder resolution grid. Halftone Dots and the Imagesetter Grid file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (6 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows In reality, the imagesetter images at the intersection of the lines on the grid to make a spot. If the illustrations showed that, you could not see where one dot ended and the next began. For illustrative purposes, the graphics in this page show the imagesetter spot as the block created by the grid. Higher screen frequencies produce finer halftone screens. Lower screen frequencies produce coarser halftone screens. Often, frequency is determined by the type of paper used to print the image: newspapers typically use an 85-to-100 lpi screen to print halftones, while magazines using glossy paper need a finer screen and may use 133-to-150 lpi or higher to print halftones. For very high quality promotional materials or fine art reproduction, frequencies of 180-to-200 or more should be used. To convert a photograph into a halftone, the halftone grid is superimposed on an image. Superimposing the Halftone Grid on the Image Each halftone cell is assigned a different sized dot to represent the image data for the cell. When looked at together, the dots resemble the original image. In the superimposed image, some cells would be white, some black, and the rest various shades of gray depending on the size of the halftone dot. If the image was broken into an 8 lpi screen (shown by the halftone grid), the wine glass would be virtually indecipherable. However, when the image is printed using a 100 lpi screen as shown above you can clearly see the wine glass. The size of halftone cells is determined by the interaction of the screen frequency with the image recorder resolution. Each of the halftone cells in the previous illustration is comprised of many imagesetter spots (created by the image recorder laser beam when it is focused on a point of paper or film). Each of the imagesetter spots within a halftone cell can be turned on (producing a color in your final output) or left off (producing white). The combination of imagesetter spots produces a halftone dot of a specific size and shape. To create different shapes, the image recorder turns the imagesetter spots on in different sequences. Each sequence is determined by a mathematical equation called a spot function. A separate spot function exists for each dot shape. Common shapes include round, diamond, square, and elliptical. Virtually all imagesetting systems support PostScript, and the current version of PostScript is Level 2. PostScript Level 2 handles halftone screens a little differently than PostScript Level 1 did. Specifically, screens are defined by halftone dictionaries, which are more efficient at handling the multiple screen definitions. Screens are defined with four factors: frequency, angle, spot function, and threshold matrix, which includes both the dot shape and the sequence for growing the dot with laser spots. The imagesetter accesses the screens stored in the halftone dictionary at RIPing time for each halftone. Some imagesetters create the screens in the halftone dictionary as they are needed; other imagesetters require that the screens be built and stored prior to their use. Storing screens in advance speeds up the imagesetting process because the imagesetter doesn't need to create the screen, however, if a halftone requires a screen that hasn't been built, the job may be aborted or the output may not be what you expect. A big advantage of the halftone dictionary approach is that the threshold matrix can be file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (7 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows shared between frequencies and resolutions. Screens may differ by frequency and resolution, but certain proportions remain the same. For instance, the threshold matrix for a 100-line screen at 1200 dpi is identical to that of a 200-line screen at 2400 dpi. Spot Functions The shape of a halftone dot can make a difference in the appearance of an image. Usually a particular dot shape is chosen to make an image appear more natural, for instance to avoid a tonal jump in flesh tones. Other times a dot shape is chosen for special effects, to make the image appear less natural. In all cases, the correct dot choice is important for the proper rendering of an image. A dot shape's performance is particularly noticeable in the following areas: q q q Highlights The very lightest highlights should be tonally smooth and be able to reliably hold a 2% to 3% dot. The resulting highlight modeling gives extra vitality to the image. A well-behaved highlight dot should provide enough area for the ink to adhere to the paper and to give smooth tones. Midtones Around the 50% area, certain kinds of dots (Euclidean, for example) become large enough to begin touching. This area should be smooth, without a tonal jump. If this jump occurs -- especially noticeable in Caucasian flesh tones -- shadows look harsh, and gradients in skin tone, instead of being smooth and delicate, look rough. Other kinds of dots -- elliptical, round, square -- do not touch at the 50% area. Shadows The screen should hold a small negative dot without plugging, or filling with ink, on press. The ability to hold a fine shadow dot brings extra detail to the shadows, and greater realism to the image. Dot shape is not just important in high-quality halftones. When printing in newspapers and shoppers, such factors as rough paper stock and less expensive presses can result in higher dot gain. A well-behaved dot and the proper precompensation for dot gain can result in a clean, well-rendered image, instead of one that is muddy or ill-defined. In PostScript imaging, the shape of the dot is determined by a PostScript operation called a spot function. The RIP calculates the dot shape from the spot function. Spot function names can be confusing. For example, there are two types of square spot functions. In one of these, the halftone dots are shaped like squares all the way through the tint scale. In the other, the halftone dots start out shaped like circles, grow to square shapes in the midtones, and then become circular again. In addition, vendors use different spot functions to create their halftone dots. Not everyone's round or square dot is going to grow in exactly the same way. Round The round dot is a true circle and maintains its round shape as it grows. When the adjacent spots begin to touch, at around the 78% area, the white spaces in between look like pointy, concave diamonds. Up to the 78% area, round dots have the advantages of maximum compactness and minimum edge. Maximum compactness helps tiny halftone dots adhere to paper. Minimum edge helps minimize dot gain, since dot gain is an edge effect (dots expand outward from their file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (8 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows edges). The fact that the tone jump doesn't occur until the 78% area means that the tone jump is in a rather dark area anyway, so it may not be noticeable. Plus, when total ink coverage is controlled to 240% - 300%, the jump will probably not be seen. This illustration shows sample dots at 00 angle. The blue lines show the cell boundaries. Sample Round Dots (00 Angle) A disadvantage to the round dot occurs in shadow areas. The negative space between the round dots in shadows is a pointy, concave diamond, as shown in the 78% dots in the above illustration. The four narrow cusps and long edges can plug easily, so round-dot shadows can be tough to control on press. Inverted Round This dot has the same shape and shape advantages as the round dot, but it is a white dot on a black background. The inverted round dot is an excellent choice for high-quality, lowkey images because it renders shadows well. Instead of an edge-intensive, minimum-volume pointy diamond in the shadows, the inverted round dot puts a round white dot there. So your shadows get the advantages of the round dot -- maximum compactness and minimum edge -and the integrity of the dot is maintained. Since the inverted round dot is the inverse of the round dot, the tonal jump occurs at about the 22% area. Given the inverted round dot's function in low-key images, where highlights may be nonexistent or of lesser concern, the possibility of a tonal shift is less of an issue. This illustration shows sample dots at 00 angle. The blue lines show the cell boundaries. Sample Inverted Round Dots (00 Angle) Square file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (9 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows The square dot is truly square and maintains this shape as it grows. It is aligned with the screen angle and does not start to touch until near the 100% area, when individual pixels begin to fill in. The square dot is usually used for special effects, especially with coarse screens. The crossed lines that result from its negative spaces result in a crosshatch look. This illustration shows sample dots at 00 angle. The blue lines show the cell boundaries. Sample Square Dots (00 Angle) Inverted Square Sometimes called a "crossed-line screen" for its special effects use, the inverted square starts out as crossed black lines in the highlights. These lines grow thicker as density increases, forming white squares on a black ground. The lines grow smoothly, and there is no tone scale jump since there is nothing to plug up. This illustration shows sample dots at 00 angle. The blue lines show the cell boundaries. Sample Inverted Square Dots (00 Angle) Diamond The diamond dot is a square placed at 450 to the screen angle. It grows to form a checkerboard at 50%. The diamond dot is symmetrical above and below 50%, so it doesn't need an inverted counterpart; it is its own inverse. This illustration shows sample dots at 00 angle. The blue lines show the cell boundaries. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (10 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Sample Diamond Dots (00 Angle) Line The line spot function places lines at the angle requested. As density increases, the lines get thicker. This illustration shows sample dots at 00 angle. The blue lines show the cell boundaries. Sample Line Dots (00 Angle) The line spot function is used for special effects and works best with a nonstandard set of angles. For instance, the following set of angles works well: Magenta = 450, Yellow = 1350, Cyan = 1050, and Black = 1650. Ellipticity The Euclidean dot's abrupt tone shift can be reduced by using an elliptical dot. An elliptical dot produces elongated midtone dots that join two opposite corners at somewhat below 50% and the other opposite corners at somewhat above 50%. This produces two smaller jumps in the tone scale rather than one larger jump. Typically the corners of the elliptical dot touch at approximately 40% and 60%. Because only two corners join between elliptical dots, the tone gradation across the 50% dot is smooth. Consequently, the elliptical dot is frequently used in images with soft midtone vignetting (flesh tones, for example). Elliptical dots typically perform well in midtones but not as well in shadows. After the long axes of the dots touch (midtones), the rows of dots begin to look like footballs placed end to end. When the short axes begin to touch (shadows), the resulting negative spaces are pointy and concave. These typically tend to fill easily with ink, plugging up the shadows. This illustration shows how ellipticity affects dot shape. The samples show 0.8 ellipticity applied to the round and euclidean dots at 00. The blue lines show the cell boundaries. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (11 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Sample Elliptical Dots (00 Angle) Ellipticity does not affect the line spot function. When using the square spot function, ellipticity alters the shape of the corners as the spot grows. However, as with no ellipticity applied to the line spot function, the edges in both directions don't touch until just about 100%, depending on the resolution and line screen in use. The inverse square spot function works inversely -- the edges in both directions don't touch until just about 0%. A Gallery of Tints The following illustrations show the four-color tints produced by each spot function. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (12 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Sample Tints These tints were created with cyan at 150, magenta at 750, and black at 450. The yellow plate has been omitted because it would be virtually invisible. UCR/GCR | Screens | Moiré | Dot Gain | Densitometry Moiré When overlayed, the dots on the four films (cyan, magenta, yellow, and black) produce a pattern.The only acceptable pattern is the rosette. Rosettes are pleasing to the eye and when generated properly generally do not detract from the images they recreate. To form a rosette, the four halftone screens must be placed at different angles. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (13 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Screen Angles The conventional angles, as shown in the illustration, are: q q q q Black: 450 Magenta: 750 Cyan: 150 or 1050 Yellow: 00 or 900 Advanced screening technologies can generate two types of rosettes: open (also called clear-centered) or closed (also called dot-centered) rosettes. Open rosettes, because they have more tolerance for errors that might be introduced at the press, produce better results for most types of images. Closed (left) and Open (right) Rosettes The extra room in the open rosette provides more tolerance for small press shifts (a shift of half a dot's size can cause moiri). In addition, the open centers help avoid ink trapping. Incorrect overlapping of dots and ink contamination problems are more prevalent in closed rosettes. For images with a lot of shadow detail, closed rosettes, which are less visible, help provide more detail in the shadows, although image data in the highlight areas may be reduced. In the color separation process, dot patterns can cause problems. If the screen angles are not precise, independent patterns created by the combination of two or more screen grids, called moiri may interfere with the image. This illustration shows a moiri pattern created by the combination of the magenta and black plates. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (14 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Example of Moiré Advanced screening technologies have been developed to combat moiré. UCR/GCR | Screens | Moiré | Dot Gain | Densitometry Dot Gain One of the side effects of printing is dot gain. Dot gain is the effect that occurs when a larger-than-specified dot appears on the final printed piece. For example, you might specify a 10 percent dot, but when you measure the final printed piece (or the imagesetter film), the resulting dot is 15 or 20%. Dot gain is a normal and expected phenomenon of the printing process. Variations in the amount of dot gain occur because of differences in papers and inks, and compensation for dot gain normally occurs in the process of color separating a scanned image. The SWOP (Specification for Web Offset Publications) standard even specifies dot gain within a certain range for advertising materials so that ad agencies can be sure product shots will look the same when ads are printed in various publications. Dot gain is a simple concept to understand. The paper absorbs the ink and the ink spreads from the image. Depending on the absorbency of the paper, the ink may spread a little or a lot. That's dot gain. Each place where ink is put on paper, the ink spreads. When the ink spreads, the resulting dot size is larger than the specified dot size. A 15% dot may end up looking like a 17% dot. While this change may be insignificant by itself, when you combine the four pages of a color separation, dot gain can change the color of the image, usually degrading the image quality. Since the color produced by a halftone screen depends on dot size, dot gain (or less frequently dot loss) can change the color intensity of the printed piece. How much dot gain an image experiences depends on many factors, including: q q Line screen Resolution file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (15 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows q q Paper Press Typically, the more dots you print (higher line screens at higher resolutions typically product more halftone dots), the greater the dot gain. Dot Gain and FM Screening When using FM screening, you need to pay special attention to dot gain. Your production system must be carefully maintained and calibrated to ensure quality output. Differences between batches of photomedia or the strength of the development chemistry can introduce unpredictable variations in screening output. FM screens use microdots instead of halftone dots. Microdots are tiny dots, sometimes the size of a single etter pixel. Because FM screening uses such tiny microdots, it uses a lot of them. A group of smaller dots have more total edge area that an equivalently sized larger dot. Dot gain is an edge effect; the more edges, the more dot gain. (The edge effect also applies to AM screening. Fine screens, with smaller dots, usually have greater dot gain than coarse screens.) This illustration compares the dot gain of uncompensated FM screens to the typical dot gain from conventional screening process.> Dot Gain Curve If you carefully control your production environment and include dot gain compensation in your screen sets, your output will appear exactly as you intend it to appear. UCR/GCR | Screens | Moiré | Dot Gain | Densitometry file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (16 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Densitometer Spectrophotometers and Colorimeters are commonly used in the paint industry and in the manufacture of inks. They are not generally used for everyday measurement or control of color in the printing industry. We work directly with Red, Green, and Blue values in the scanner or color monitor. Likewise, we work with Cyan, Magenta, and Yellow values on film or press. The instrument used to directly measure these values is the densitometer. The densitometer measures either reflectance or transmission of light. If film is being measured, a transmission densitometer is used. The amount of light transmitted through a small section of film is divided by the amount of light transmitted to the light sensor without film to give percent transmission. A reflection densitometer is used to measure reflection copy, proofs, or press sheets. Here, light is reflected off the sample instead of transmitted through it. Density Our eyes are not equally sensitive to equal changes of light from light to dark. We are much more sensitive to small changes in light areas than dark areas. For this reason, the densitometer converts percent transmission (or reflection) into a logarithmic scale known as density. Density values more closely correspond to the eye's sensitivity. The table compares percent transmission (or reflection) to density. You can't produce consistent, high-quality color unless your output density is in specification. Variations in Dmax (maximum density) alter tints and produce color shifts. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (17 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Dmax and Dmin (minimum density, or the amount of light absorbed by the paper or film) values allow you to monitor the consistency of the entire production system. You should measure positive and negative dot areas and tints to check the accuracy of film percents and to calibrate the imagesetter. If image density is too great, then the blacks and dark colors will indeed be very dark, but the tints and halftones will be too dark and the highlight areas will be much darker than they should be. The whole image will have a muddy appearance, and will likely have color shifts. Conversely, if the image is lighter than it should be, the blacks look gray upon reproduction, the tints and halftones are not accurate, and the white or highlight areas bleed past their borders. The whole image looks washed out. This illustration shows an example of an image where the density has been adjusted to show the difference density can make. High vs. Low Density Transmission or Reflection Transmission or Reflection 100.000% 50.000% 10.000% 1.000% .100% .010% .001% Density 0.00 0.30 1.00 2.00 3.00 4.00 5.00 Measuring Color with a Densitometer To measure color with a densitometer, color filters are introduced into the optical system. The color filters used are generally the same Red, Green, and Blue filters that are used in color separation. On press, the press operator will use the color filter that complements the ink color being measured. That is; the Blue filter for measuring Yellow ink, the Red filter for measuring Cyan ink, and the Green filter for measuring Magenta ink. To measure and define any color in color space, readings are taken through all three file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (18 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows filters. Inks Red Cyan Magenta Yellow 1.40 09 02 .41 1.35 .06 Filters Green .11 .53 1.04 Blue Sample density readings from solid ink patches on press sheet These values can then be used to help correlate the color separation process to specific ink/paper/press conditions. Densitometry Densitometry helps you find a balance for accurate tone reproduction. You can use a densitometer to measure: q q q q The The The The original image film output from the etter proofs press run You can then use these density measurements to adjust your system. For example, if you find the density of the film is too great, you can adjust the exposure value for the image recorder. You can also use density readings from the film to determine if the photomedia processor needs adjustment. Once you've calibrated your etting system, you can measure the results of a test page run on your press. This density tells you the total dot gain. You can use this information to build transfer functions. You can also use the density readings to build dot gain compensation into the screen sets. Halftone may contain several different shades of gray within the same image. (Color are just four grayscale printed using different color inks, so this theory also applies to printing in color). Areas quite close to each other in the image may have completely different gray values. For example, an image of an eye has a very dark area (the pupil), surrounded by a lighter area (the iris), surrounded by a white area. Because tones vary within a halftone image, you cannot take reasonable density measurements from it. That is why you see density color bars and tint grayscales in the cropping area of the page. If the density is correct for those areas, it will be correct for the on the page. Grayscale Density Control Sample file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (19 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Densitometers You measure image density with either a transmission or a reflection densitometer. Transmission densitometers measure the amount of light passing through film. Reflection densitometers measure the amount of light bouncing back from the surface of the printed page. Both types of densitometers measure the image density by calculating how well the black area absorbs light. Reflection Densitometer You use a reflection densitometer to measure density on paper, whether RC (resin-coated) paper imaged in the etter, or the final printed page. A reflection densitometer shines a known amount of light onto the image on paper. Some of the light is absorbed by the ink on the paper; the rest of the light bounces off of the image back into a photocell (a device that measures the quantity of light) in the densitometer. Technically, the light must be at a 45° angle and the photocell at a 90° angle above the paper. The densitometer subtracts the amount of light returned from the amount of light shone to determine how much was absorbed. The amount absorbed gives the density reading. On paper, solid black areas generally reflect between 5% and 1% of the light. In other words, the ink absorbs between 95% and 99% of the light. These dot percentage values correspond to optical density readings between 1.3 and 2.0. The paper's surface texture scatters and reflects light, some of which finds its way into our eyes, the same value as measured by the densitometer. Such scattering and surface reflection substantially reduces the density of on paper. That is why a black area on paper measures around 2.0, while a black area on film measures around 4.0. Reflection Densitometer Transmission Densitometer To measure density on film or transparencies, you need a transmission densitometer. A transmission densitometer shines a known amount of light through an image on film. The exposed area of the film (the image) absorbs some light, and the rest passes through the film into a photocell. The densitometer uses the difference in the amount shone on the image and the amount registered by the photocell to determine how much of the light was absorbed, resulting in the image density. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (20 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Transmission Densitometer Density Readings The description of densitometer readings include both dot percent and optical density readings. The dot percent value ideally corresponds with the tint percent assigned to that area. For example, a 15% tint should have a dot percent reading of 15%. The optical density reading measures the amount of light transmitted on a logarithmic scale. You don't need to understand logarithms to work with density values; the densitometer does all the calculations. Logarithms are used to make the density values smaller and easier to work with. Some densitometers provide readings in percent values that correspond to tint values. Most densitometers provide an optical density reading, a number from 0.01 to 4.0 or higher, depending on the ability of the densitometer. A reading of white could be 0.02 (even white, on paper or film, absorbs some light; this value is also known as Dmin); a reading of black could be 3.9 to 4.2 for film, or 1.8 to 2.0 for paper (the reading of a 100% black area is known as Dmax). A solid 100% black tint on film should only reflect 0.01% of the light shone on it (absorbing 99.99%). This translates to a density of 4.0. As a comparison, this illustration shows how tint percent values translate into optical density readings taken from paper. This graph assumes that each percent value requested is the actual output dot percent. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (21 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Optical Density vs. Dot Percent Density Measurements> As you can see, even though the optical density readings progress evenly, the dot percentage values do not (look at the difference in the numbers along the dot percent scale). The y-axis, the dot percentage values, is not linear. That is, for each unit of the axis, the value goes up much more than one unit. The first unit (from 0.00 to .025 optical density) covers from 0% to 44% of the dot percentage range. The last unit in the graph (from 1.75 to 2.00 optical density) covers from 97% to 99% of the dot percent range. The actual scale used is a logarithmic scale. This way of calculating densities allows more intermediate values through the range of dot percents (above 90%) where dot gain has the greatest affect. Many densitometers available today provide both types of readings. Calibrating the Densitometer Densitometers, both reflection and transmission, must be calibrated as frequently as specified by the manufacturer, if not more often. Your densitometer kit should include a calibration strip. This strip of film or photopaper has a black area and several gray shaded areas. Each area has a number next to it (or below it, depending on your model). This number is the reading you should get when you measure the area with your densitometer. file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (22 of 23) [26-12-2002 11:37:17] Prepress Tutorial Wide Open Windows Sample Densitometer Calibration Strip Follow the manufacturer's instructions included with the instrument to calibrate your densitometer. If your densitometer is not frequently calibrated, or is not properly calibrated, it will give wrong density readings and you may be unpleasantly surprised at your output. When you calibrate the densitometer, it compares its reading with what the reading should be, and then compensates for any differences. Calibrating your densitometer once a day is a good idea. You should also take care of the calibration strip; do not touch the areas to be measured. Oil and fingerprints can subtly alter the density readings. Do not leave the calibration strip in a lighted area as the ink may fade -- any calibration made from that strip would then be useless. You should always use the same densitometer to calibrate your production system. Individual densitometers may give slightly different readings from the same sample. In addition, different models of densitometers vary in sensitivity. To ensure consistency, always use the same densitometer, frequently calibrated, to measure output from your production system. UCR/GCR | Screens | Moiré | Dot Gain | Densitometry file:///D|/Mijn%20documenten/Wide%20Open%20Windows/Course/prepress.htm (23 of 23) [26-12-2002 11:37:17]