Sipix Imaging, uses a similar technology to produce black, green or blue colors on a white background. Polymer Vision is working with Sipix and E Ink to create a rollable display prototype called the Readius. "We are enabling very small devices with large displays," says Edzer Huitema, program manager at Eindhoven, Netherlands-based Polymer Vision. The 5-in. pull-out display will offer 16 levels of gray. The device refreshes about once per second -- too slow for menu navigation or video but acceptable for portable navigation systems or as a Web news or e-mail reader. Huitema predicts that Sipix will ship the device in the first half of 2007, with color and touch-screen capabilities available by 2010. E Ink is working on a color filter for its technology that it expects to be ready in the same time frame.

NTera is developing a flexible, conformable bistable display based on its nanochromic technology. Instead of moving particles, NTera's RGB display technology determines colors based on each particle's charge. The design uses a 200-dpi passive- matrix transistor array, which is typically slower than active matrix because it updates each row on the screen instead of individual pixels, as active matrix does. However, NTera claims that its technology is fast enough to support video speeds for short intervals. The company says its partners will produce conformable displays early next year for uses such as side displays on cell phones or input tablets. "That is our primary focus," says Alain Briancon, chief technology officer at NTera.

NTera's technology adds another twist: The metal oxide display material is transparent when not charged, laying a foundation for transparent displays. "An overlay on top of a window could be realized," similar to what moviegoers saw in Minority Report, says Briancon. But NTera's current products are still built on glass.

Kent Displays, is developing a bistable LCD based on cholesteric technology. The supertwisted nematic LCDs used today twist constituent liquid-crystal molecules about 270 degrees, says sales and marketing manager Tony Emanuele. Kent's technology twists the molecules 16 or 17 times. When twisted that tightly, the molecules don't readily unwind, making the display bistable.

In the lab, Kent has demonstrated its technology deployed on plastic, paper and even fabric substrates. "The chemical recipe for cholesteric lends itself more readily to plastic substrates," Emanuele says, because it has 1/1,000th the barrier requirement of standard LCDs. Kent's current displays use a relatively slow passive matrix and are best suited for applications such as e-books and signage. The technology includes two-color combinations of either yellow/black or blue/white. Early units are fairly small, at 2.5 by 1.5 in., and offer 100-dpi resolution. Kent is working on a faster, active-matrix display and expects to be able to manufacture its passive-matrix version on flexible substrates by year's end. "Two to three years from now, flexible displays on plastic will be commonplace," Emanuele predicts.

In addition to the technical challenges, flexible displays face several other hurdles before they can become commercially viable. "Making [displays] on glass is hard enough," says iSuppli's Allen. Jet printing requires an entirely different, if potentially cheaper, manufacturing process that is still in the early stages of development. Until volume production is possible, jet-printed displays will be expensive relative to alternative technologies. For example, in the e-paper market, simple paper shelf labels and signage can't be updated electronically, but they're far cheaper, Allen says.

The key, says Jeremy Burroughes, CTO at Cambridge Display Technology, is to find a profitable niche for early designs. "The hurdle is always to find early areas to get into first," he says, "and gradually build up the knowledge and revenue profile to go into more advanced areas."