Today's color printers get better, faster, cheaper, and more capable. Will we each have a color printer on our desk by decade's end?
Michael Weiss
Rapid change is the name of the color-printing game these days. Rising quality,
falling costs
, new and improved printer technologies, and widely available software that can output color are all contributing to the explosive growth in color printing in a wide variety of settings from home to school to office. These trends will continue over the remainder of the decade at an even more rapid rate.
Accurately predicting what will change is a lot easier than correctly guessing when it will happen. New products with new features will bombard the market--some breakthroughs appear just around the corner; others will need four or five years to perfect. Howe
ver, the one constant will be that the large number of major industry participants will keep pushing themselves, and consequently one another, to provide users with new and better color printers. There certainly seems to be an industry consensus that color output will become increasingly important.
A compelling argument can be made that technology generates both user interest and software developer support, especially when delivered to the marketplace by Canon, Epson America, and Hewlett-Packard in the case of ink-jet technology or by HP and Xerox for desktop color laser printers. Other highly respected manufacturers, such as Tektronix and Eastman Kodak, have generated similar support by producing specialty products based on thermal-wax transfer and dye-sublimation color-printing technologies.
Liquid Ink-Jet Printing
The most obvious thing about color liquid ink-jet printing is its enormous and growing popularity. Both HP (in 1993) and Canon (in 1994) have passed the 10 mi
llion mark in ink-jet printer shipments. HP sold 5 million in 1993 alone, and 2 million of those were color printers. For 1994, HP expects 60 percent of the ink-jet printers it sells will be color-capable. Indeed, since HP introduced the DeskJet/DeskWriter 320 and 540 and retired its older models, it no longer ships monochrome-only ink-jet printers. In addition to HP, Canon, and Epson, ink-jet printer suppliers include Apple, DEC, Lexmark, Tektronix, Texas Instruments, and Xerox.
The success of liquid ink-jet is due to the printers' continually decreasing cost; improvements in print quality, performance, and ease of use; and effective marketing. Recent advances in liquid ink-jet printing technology include higher resolution, better ability to print on plain paper, faster speed, surprisingly large improvements in print quality, and the ever lower costs. Lightfastness and waterfastness have also improved.
Printing speed is increasing, as the vendors incorporate more nozzles into their print heads.
The new Canon BJC-4000 has a 128-nozzle head and prints at 5 ppm (pages per minute) in monochrome. Another development on the horizon is a full-width print array. Canon could build that today, as well as a 256-nozzle head, but not at a price that makes them feasible for desktop printing. In five years, this will change.
According to many industry participants, another major breakthrough will be true medium independence--the ability to print on almost anything. The key to this is ink chemistry, something no ink-jet producer will discuss in detail. Ink chemistry is a key area of competitive advantage for HP, Canon, and Epson, whose R&D departments are working on improvements in pigments, lightfastness and waterfastness, and suitability for printing on a wide variety of media. For example, Canon already has a large specialty printer that uses a 256-nozzle head to print at 400 dpi (dots per inch) on a wide variety of textiles. It prints on T-shirts, and the images do not fade; it also prints on bathing su
its, and the ink will not wash off. These advances in ink technology will find their way to the desktop by the end of the decade.
Another area of major improvement that's close at hand is dramatically improved resolution and thus print quality. New printers from Epson and Lexmark entered the market in 1994 that were able to deliver extremely high-quality images because of new printer driver software. Resolution will continue to increase, and this year you will see true 600- by 600-dpi ink-jet printers to replace the so-called 600- by 300-dot-addressable solutions currently on the market. The Epson Color Stylus, used with its software drivers, already claims 720 by 720 dpi.
The next plateau in print quality will be photo-realistic images. In the next few years, units will offer high resolution (600 or 720 dpi) with more depth of color and enhanced drivers to compete with continuous-tone printing from color laser, variable dot-wax transfer, and dye sublimation. Indeed, a number of major vendors be
lieve that the ability to print photo-realistic images at a high speed and keep costs low will be the key to market success at the end of the decade.
Other issues, such as the cost per page, will fall by the wayside. Printing costs will decline with improvements in medium independence, as users will no longer need to purchase special paper. Costs will drop further as competition forces vendors to lower their ink-cartridge prices, which today are still high-profit items.
If liquid ink-jet printers can deliver high quality with robust duty cycles and offer advanced paper handling at speeds comparable to those of laser printers, they will compete aggressively with all color-printing technologies. There are no technical impediments to any of these attributes. In fact, ink-jet printers have a manufacturing cost advantage that should ensure that they remain the lowest-cost color printers in the marketplace. For the foreseeable future, liquid ink-jet will be the color technology to beat.
Solid Ink-Jet Printing
Solid ink-jet, or phase-change, technology offers a good deal of promise as well. This technology uses heat to melt the ink material, which at room temperature has the texture of a hard crayon. Once the ink hits the paper or other medium, it cools almost instantly. This attribute makes solid ink-jet perhaps the most medium-independent color-printing technology that is currently available on the desktop--early suppliers of the technology liked to show samples printed on sandpaper and toilet tissue.
Other advantages of this type of printer include the production of vibrant colors that are especially good on transparencies, waterfastness, and true clean-hands installation and resupply. Image quality is excellent, although not currently up to the level of dye sublimation. Tektronix, the leading supplier of solid ink-jet printers, has developed techniques that make monochrome text and graphics quality quite good. On the negative side, solid ink-jet printers print slowl
y, and they are expensive to buy and operate. Also, the color is not very lightfast; it tends to fade quickly when exposed to sunlight.
Tektronix has made significant R&D investments in the technology and is clearly aware of the current shortfalls. There is every reason to believe that its next generation of printers will be at least twice as fast as the current Phaser 300i, which doubled the speed of the first-generation Phaser III PXi.
There is nothing inherent in the solid ink-jet technology that would inhibit it from being competitive with desktop color laser solutions. A. J. Rogers, strategic marketing manager of the Tektronix Color Printing and Imaging division, likes to say that "it has the simplicity of liquid ink-jet without the drying-time problem." Solid ink-jet printing is already fully medium-independent and is more reliable and easier to use today than color laser printing. Further, the lightfast problem seen in Tektronix offerings can be overcome: Brother offers a solid ink-jet pr
inter that uses a pigmented ink that is much more resistant to fading.
Major solid ink-jet improvements need to occur in two areas: performance and price. Speed is the easier of the two to achieve. The next generation of solid ink-jet products, due out this year, is likely to comprise true page printers with the ability to print color at least as fast as current color laser printers. Price breakthroughs will be contingent on higher sales volumes that would dramatically reduce manufacturing costs. To do this, solid ink-jet printers will have to keep pace with the print-quality improvements expected for both liquid ink-jet and color laser technologies.
Solid ink-jet color-print quality is already good for graphics and will improve with further development. However, text quality must be improved, especially in monochrome, to compete successfully with laser printers. It seems clear that color laser printers will improve to match monochrome laser quality with increased resolution, moving from 300 to
600 to 1200 dpi using smaller toner particles and resolution-improvement techniques. Solid ink-jet devices will have to match the higher resolution and better text imaging. There are no basic technical impediments to this, but print quality comparable to today's monochrome laser printers is probably two generations of solid ink-jet printers away.
Solid ink-jet/phase-change technology shows good promise for the end of the decade. By then, these printers are likely to have a small footprint, speeds that match desktop color lasers, and prices in the $2000 to $3000 range. With those specs, the technology should remain attractive to many.
Color Laser Printing
Color laser printing is currently the hottest printer topic. At this writing, four vendors were supplying desktop color laser printers: QMS, Xerox, HP, and, just about to appear, Tektronix. They offer speed, adequate text quality and color that is good enough for general-office printing, and address-networking capabilities
. As with all new technologies, they still command a price premium, but prices are dropping quickly. The 12-ppm monochrome HP LaserJet 4 Plus sells for $1839, and the color LaserJet costs $7295.
Unfortunately, there are a number of difficulties with color laser printing today. First, it is far from clean: Toner installation is a messy procedure. Moreover, some units require surprisingly frequent intervention by users to service components. As a consequence, they are significantly harder to live with than other color-printing technologies or monochrome lasers.
Another disturbing issue, hopefully temporary, is that the QMS and Xerox models (both using Hitachi's color-laser engine), appear to be mechanically fragile. Users report needing higher-than-normal numbers of service calls to keep these printers working properly. The HP Color LaserJet does not appear to suffer from this problem. The Tektronix offering is unknown because it was unavailable at press time.
Perhaps the most serious probl
em with color laser printers is lower print quality than what users are likely to expect. The QMS, Xerox, and HP devices don't support continuous tone reproduction; they must create combination colors using halftone (dot-pattern) screens. The new Tektronix desktop laser printer does use continuous tones; its overall quality should compare favorably to the imaging capabilities of the larger and more expensive Canon and Xerox machines. Note, however, that text quality for desktop color laser units, such as the HP Color LaserJet, does not compare with today's office standards set by 600-dpi monochrome printers, although it may be good enough for most purposes.
Because color laser printers are still in their infancy, today's difficulties should not be viewed as a failure of the technology. Over the next five years, all vendors' color laser printers will become simpler and more reliable. The next-generation devices should be capable of clean-hands installation and require minimal user intervention. Quality
will go up quickly, as will speed.
Today, you have to compromise to get affordable quality color. By the end of the decade, you won't have to. The one open issue is cost. If color laser-printer prices reach the $2000 to $3000 range, they will likely find their use in higher-end desktop environments, while comparable ink-jet printers will occupy the low end and midrange.
Thermal-Wax Transfer and Dye Sublimation
Compared with ink-jet or laser printers, today's thermal-wax transfer and dye-sublimation printing technologies are more specialized (i.e., less versatile). Each has clear technical strengths and weaknesses. Both have good color-imaging characteristics. TTP (thermal-transfer printing) produces excellent transparencies, while dye sublimation is the clear leader in producing photo-realistic images.
Perhaps the greatest drawbacks of both technologies are their cost and their requirements for special media. Both are quite expensive on a cost-per-page basis, and dy
e-sublimation units are expensive to purchase. Further, the cost per page for TTP is fixed--the cost per print is unaffected by the amount of color on a page because the process uses a full block of transfer ribbon for each page printed. These ribbons are expensive, their pricing largely controlled by patent-licensing arrangements.
Achieving best quality from TTP requires special medium. The slick paper that gives thermal transfer an optimum base for printing also makes it impossible to integrate transparently into a larger document. Attempts at plain-paper solutions have resulted in lower print quality.
Dye sublimation has similar drawbacks. As with thermal-wax transfer, ribbon use is fixed and expensive. The one bright light here is that dye-sublimation ribbons are not bound by the same proprietary pricing constraints as thermal wax, so you are likely to see a dramatic cost-per-page reduction. The process will continue to command a premium over other types of color printing, however, because o
f paper costs. The paper needed for dye sublimation is an expensive, photographic-grade stock.
You can also expect dye-sublimation mechanisms to become significantly less expensive in the next few years. As host processors become more powerful, some printer manufacturers will provide "dumb" dye-sublimation printing devices with radically reduced prices. They might be slower, but they will also be much less expensive, while still providing excellent imaging capabilities.
Other advances will also be made. The new Eastman Kodak 8600 dye-sublimation printer provides much better lightfastness than has been previously achieved. Additional improvements in image quality also can be made, although devices are likely to remain at the 300-dpi level, simply because higher resolution is not particularly important for continuous-tone rendering.
The performance of thermal transfer and dye sublimation differ considerably. Dye sublimation is a slow technology, while some of the fastest color printers on t
he market are TTP units. In extensive benchmark testing across technology lines, the Tektronix Phaser 220e and 220i achieve better throughput than any other color device tested--this includes liquid-ink, solid-ink, dye-sublimation, other thermal-transfer products, and some color laser printers.
Although both processes can produce generally good images, there are still significant image-quality problems. Both experience problems producing text in small point sizes with good character shapes and smooth edges. In addition, both have trouble with fine-line detail, especially hairlines printed perpendicular to the paper path. These problems are inherent in the technology because it is difficult to quickly heat and cool individual pixels to produce fine levels of detail under certain circumstances.
Considerable creative effort has also been invested in developing machines that can serve new users and new applications where the inherent drawbacks are less significant. Fargo and Seiko Instruments have d
ual-technology devices that can produce thermal-wax transfer and dye-sublimation output, and a variety of vendors have introduced large-format printers in the 24- by 40-inch class.
More improvements are on the way. In the next 12 to 24 months, true 600-dpi thermal-transfer printers will be available. Another 1995 print-quality improvement will be the use of variable-dot printing, where dots of varying sizes will replace today's dithering techniques that incorporate same-size dots. The process will require more memory in the printer but will produce print quality midway between traditional thermal transfer and dye sublimation. Products using these new technologies will be faster than dye-sublimation printers, perhaps down to only 2 minutes per page. Media costs will also be lower.
From a marketing perspective, dye-sublimation and thermal-wax transfer technologies will remain competitive through the 1990s: Their excellent print quality will ensure their use in specialty markets. But beyond that, t
hey will be under increasing pressure from the other color-printing technologies in terms of both image quality and cost. And they've already lost the speed battle.
Closing the Gap
All the major desktop color-printing technologies have a mix of strengths and weaknesses. The suppliers are aware of all the shortfalls and are working to correct them as quickly as possible. By the end of the decade, users should expect fast, clean, easy-to-use, and inexpensive color printers. Color quality will be high, and output will be much more waterproof and lightfast than it is today.
What is absolutely not clear at this point is which--or whether--one of these technologies will emerge as clearly superior to the others. It seems more likely that all the printing technologies will converge and deliver similar capabilities to users. Printer prices will drop, as will cost per page. Each technology will still have advantages and disadvantages, but they will be small and may matter only in ce
rtain applications. In the end, technology itself will probably not be the principal determinant. For color printing, says Ron Patten, senior product-marketing manager at Seiko Instruments, "success will be determined by marketing, not by technology."
illustration_link (11 Kbytes)
Build Yourself a Color Transform
illustration_link (44 Kbytes)
This cutout is an example of a 3-D, nonlinear color transform. By cutting out these two shapes and folding them into cubes, the complexity of a color transform may be more fully appreciated.
The cube on the left is divided into planes of constant value, as shown by the black lines on its surfaces.
After color transformation to the cube on the right, the cut
ting planes become curved cutting surfaces.
This specific color transformation was designed to match a printer to a monitor and was made using the Candela Color Management System.
As simple as cut and paste!
Michael Weiss is president of MWA Consulting, Inc., of Palo Alto, California, an industry-analyst organization that focuses on printing and imaging issues. You can contact him on the Internet at
mwac@applelink.apple.com
or on BIX c/o "editors."
Final Output
