Back From the Brink

A made-for-TV technology is finally ready for prime time

3 min read

Samuel K. Moore is IEEE Spectrum’s semiconductor editor.

It was supposed to be easy. Marry two mature technologies--liquid-crystal displays and silicon complementary metal oxide semiconductor devices--and you'd get instant entry to the market for wide-screen, high-definition televisions. Yet the hybrid technology, known as liquid crystal on silicon (LCoS), long seemed a dead end, and some of its proponents threw in the towel. Even mighty Intel Corp. gave up on it last year.

Yet for a handful of companies that stayed the course, the technology is finally on solid ground. The key to success, it seems, was persistence.

"We've come out of a long dark tunnel, and now we have something to sell," says Vincent F. Sollitto Jr., president and CEO of Brillian Corp., Tempe, Ariz., which has begun making high-end LCoS TV sets after many years of development.

And Brillian is not alone. The leader in the technology, Victor Co. of Japan (JVC), plans to sell 350 000 TV sets this year containing LCoS chips made in part by Aurora Systems. Sony is to release its version in the second half of the year containing its own chips. And LG Electronics has committed to launch one as well, using chips from SpatiaLight, which opened a plant on 10 March in Korea to supply them. A competitor, MicroDisplay, is said to be shipping chips and optics for sets.

The market research firm DisplaySearch, in Austin, Texas, says revenues from LCoS sets increased nearly fivefold last year, to US $52.8 million, and predicts that the technology's share of the high-definition television market will continue to grow steadily.

LCoS is one of three so-called microdisplay approaches for rear-projection television. In most models, it works by splitting a light into red, green, and blue beams and reflecting them off three silicon chips to which a liquid-crystal layer has been bonded [see illustration, ].The chips control the on-off action of the liquid crystal to form an image. Then the light beams recombine through a set of optics to project the image onto a screen.

One of the main advantages of rear projection is that, unlike the case with LCDs or plasma displays, you don't have to pay disproportionately more to make a larger display. Plasma and LCD manufacturers must construct more pixel elements on bigger sheets of glass, but for rear projection television, you just magnify the image.

According to market research firm Insight Media, in Norwalk, Conn., the price of a 50-inch-diagonal rear-projection television is about the same as that of a 30-inch LCD model of the same resolution. And it is the market for 50-inch and larger screens that rear-projection TV makers are targeting. LCoS competes with transmissive micro-LCDs (which project an image through an LCD array) and with Texas Instruments' reflective chip (which directs light with micromirrors).

Why did so many firms, large and small, see their hopes for LCoS dashed? In many cases, they found the devices tempting because they could build them on paid-for equipment developed for earlier generations of chips. Then they were surprised by poor optical quality and bad yield rates for the assembled systems.

"No one could ship on either the first or the second design," says Mary Lou Jepsen, a 10-year veteran of LCoS development, both at Intel and at MicroDisplay Corp., a company in San Pablo, Calif., that she cofounded. Solving design problems was like "peeling an onion," she says: there was always a deeper flaw underneath the one just tackled. The survivors have gone through a dozen or so designs, says Jepsen, who now heads JOE Inc., a Sausalito, Calif., consulting firm.

Even the companies that are now succeeding seemed stymied as recently as last fall. In September, for instance, Brillian lost a major contract to supply the U.S. retailer Sears, Roebuck & Co. with chips for HDTV sets because JDS Uniphase, which was to make the accompanying optics, failed to deliver. SpatiaLight Inc., Novato, Calif., announced deals with seven Chinese TV makers in the last two years, but none of them has yet sold any LCoS televisions. "China has been a real disappointment for us," says Theodore H. Banzhaf, SpatiaLight's executive vice president. Even JVC, the LCoS TV leader, hit a rough patch. In March, the company recalled two models made between May and November 2004 because of a fire hazard from a bad electrical connection.

As for the big-league dropouts, Intel and Philips, the problem seems to have been a lack of perseverance. After missing its chip-delivery targets, Intel officially abandoned LCoS in October. Philips' withdrawal, which came in the same month, was even more surprising, as the company was already shipping sets incorporating the technology. And that further tarnished the technology's reputation. "The ripples from the two high-profile crash-and-burns are still out there," says Chris Chinnock, president of Insight Media.

This article is for IEEE members only. Join IEEE to access our full archive.

Join the world’s largest professional organization devoted to engineering and applied sciences and get access to all of Spectrum’s articles, podcasts, and special reports. Learn more →

If you're already an IEEE member, please sign in to continue reading.

Membership includes:

  • Get unlimited access to IEEE Spectrum content
  • Follow your favorite topics to create a personalized feed of IEEE Spectrum content
  • Save Spectrum articles to read later
  • Network with other technology professionals
  • Establish a professional profile
  • Create a group to share and collaborate on projects
  • Discover IEEE events and activities
  • Join and participate in discussions