Felix Torres
06-09-2006, 03:00 PM
HDTVs want to be big. Just the transition from 4x3 to widescreen is enough to bump screens upwards one size, from 27” to 32”, in order to maintain image height for SD content. The resolution improvements from HD allow, but do not necessarily require, a second size bump going from SD/ED resolutions to 720, and a third bump going to 1080. Economics also factor in; consumer surveys have found, no shock, that consumers are willing to stretch a bit when purchasing a TV and that two-thirds of consumers would pay an extra $300 to get a larger/higher-res display.<br /><br />Given that about two-thirds of existing displays in US households are 27” or smaller, the mainstream HDTV market that will be exploding between now and 2009 can be expected to end at about 42”. The market that begins at 44” and above is an entirely different creature than the market for mainstream TVs. For one thing it isn’t quite as price-sensitive. For another, it is primarily a North American-dominated market, and it is significantly smaller. Finally, where direct-view LCD displays are expected to dominate the under 44” HDTV market (they are, after all, pretty much the only player in those sizes), they will be only one of several contenders in the 40-50 inch market. Because size matters in this market, the manufacturers have already started a wholesale migration to 1080p display resolution as a standard and, as early as the fourth quarter of 2006, over 50% of the units shipping will be built to this resolution. (More on this at a future date.)<br /><!> <br />For the near-term future, the dominant technology in Big-Screen HDTV (50+ inches) is most likely going to be rear projection displays, using one of three types of micro-display imaging chips/panels. Two of them are based on variations of the basic LCD light-valve technology, the third on microscopic arrays of reflective mirrors.<br /> <br /><b>Micro-display Rear Projection System product lineups, models</b><br /><br /><img src="http://www.digitalmediathoughts.com/media/users/634/Part%204%20-%20table%201b.gif" /> <br /><br />The latter technology is called DLP (digital light processor) and it is available to all comers from Texas Instruments. To date, over 75 manufacturers are buying these TI chips to build displays and projectors. Since production volume is relatively high (for a big-screen technology) and the manufacturing processes originated in the semiconductor industry, DLP chips are relatively inexpensive. <br /><br />In the HDTV arena, the primary use of these chips is in table-top rear-projection displays built around a single DLP chip and a rotating color wheel in sizes ranging from 44 inches to 70+ inches. (Some boutique vendors provide triple-chip rear- and front-projection systems with exquisite image quality at premium prices and very low volumes that obviously do not factor in the typical buyer’s decision process.) <br /><br />Because of the reflective nature of their imaging engines, DLP-based rear projection units tend to be very bright and have very high contrast levels. Because they rely on persistence of viewing effects, a small minority of people see rainbow color effects in some scenes. Starting later this year, DLP rear projectors will start transitioning away from color-wheel systems to triple-LED sight sources, which will provide even brighter images, faster start-up time, longer light source life cycles, and no rainbow effect.<br /><br />The strength of the DLP technology is that the mirrors, being very small, can be flipped from a reflective to a non-reflective state very, very fast. This allows for a variety of image presentation tricks that allow for a form of optical anti-aliasing as well as for compositing images at resolutions higher than the chips’ native mirror count. Higher than 1080 native resolutions (1440p, QuadHD) await the arrival of electronics that can generate these images, probably some time in 2007, which should open the door to larger than 75” DLP displays, especially if combined with spherical lenses and advanced optics that allow for 70” units that are less than 10” deep.<br /><br />As flexible and as cheap as DLP systems are, though, they are not the cheapest. That honor falls, for now, to triple-LCD micro-displays, that use three miniature monochrome LCD panels (usually around 3” in size each) in their imaging engines. Because the projected image uses light that goes through the LCD panels, the overall brightness and contrast is lower than the other two kinds of micro-displays and because the LCD panels are so small, the inter-pixel borders that carry the control signals to the individual pixels are visible, unlike with direct-view LCDs. Finally, to date, nobody has announced 1080p 3LCD displays, which suggests that, going forward, the technology is destined for entry-level systems only. Which is not necessarily bad as there are a lot of people who find 3LCD images preferable to those of more expensive systems, mostly because the images are rendered with a natural softness that more closely approximates the look of projected film in a movie theater. It also helps that name-brand 50” 3LCD units can be readily found in the $1500 range and can be expected to hit $999 within the next two to three years.<br /><br />Just because 3LCD micro-displays are the clear 720p entry-level tech for big-screen HDTVs doesn’t mean that DLP will have the mainstream big-screen market all to itself. Because it won’t.<br /><br />The third micro-display technology, LCOS (Liquid Crystal On Silicon; sold under a variety of smoke-n-mirror names such SXRD, ILP, etc) is a reflective LCD technology: typically three inch-sized chips sandwich a monochrome LCD panel atop the electronics that drive the individual pixels. Because the driver wires lie behind the pixels, the inter-pixel barriers can be much thinner than on transmissive LCD panels, resulting in smoother, sharper images. And because the light source is continuously reflected off the LCD surface, brightness and contrast are noticeably better than DLP systems with comparable-power lamps. The result is simply stunning. The best LCOS displays are not cheap but they offer brightness that rivals direct-view LCD at sizes direct-view can’t match at any price. They are, today, quite simply the benchmark for HD image quality. Presumably, the LED-based DLP units coming later this year will even the playing field, but one does have to wonder what an LED-driven LCOS could do, too. Probably allow for 75” LCOS implementations.<br /><PAGEBREAK><br /><b><span>Competition is Good for Consumers, No?</span></b><br />Competition, especially at the lower end of the size spectrum, will be driving down DLP and LCOS prices but not at the rate of the direct view LCD panels. Micro-displays simply don’t have the worldwide volume to keep up with the LCD panel supply chain economics because really large screens are primarily a North American product. European retailers won’t even display them in their showrooms and Asian buyers are focused on the form factor. Manufacturers are turning to advanced optics and industrial design to bring forth newer, thinner rear projection displays that can compete with flat panels in footprint and beat them in weight, power consumption, and image quality. A point that is encouraging these efforts is that 70+ percent of flat panels are never wall-mounted. Which means, that the footprint of the panel stand (9-12 inches) is the rear target, not the thickness of the panel itself. In other packaging areas (bezel width, display weight, power consumption, etc) it is common to see 60” micro-display units that outperform 50” flat panels. (Indeed, some large rear projectors are so light they come with tie-down straps to ensure they don’t tip over.) So there is hope for broader adoption of the technology beyond North America.<br /><br />For now, though, pricing curves look to be less aggressive than the LCD panels, although more aggressive than the larger Plasma panels they will be competing with.<br /><br /><b>3LCD Micro-display Component pricing breakdown, US$</b><br /><br /><img src="http://www.digitalmediathoughts.com/media/users/634/Part%204%20-%20table%202.gif" /> <br /><br />Because of the nature of the technology, rear projection systems tend to scale upwards extremely well, price-wise. The only incremental costs in going from, say, 50” to 60” is a slightly larger cabinet, slightly modified optics, and a brighter light source. The rewards, in contrast, are a much larger profit margin and decreased competition from direct-view technologies at the larger sizes. As a rule of thumb, at most price points at which they compete with panel technologies, rear projection systems offer an extra foot of screen size and often higher resolution, for the same price. For buyers to whom bulk is not that critical, micro-displays are a very attractive high-end option.<br /><br /><b>Micro-display Rear Projection System Projected Pricing, US$</b><br /><br /><img src="http://www.digitalmediathoughts.com/media/users/634/Part%204%20-%20table%203.gif" /><br /><br />While rear-projection system manufacturers are not about to concede any market segment, it is easy to read between the lines of their discussions that they do not intend to bleed to death in fighting for the 44-50” market segment and will be focusing their design efforts, going forward, towards ensuring their continuing domination of the 50”-and-larger market.<br /><br />Which brings up the question of whither Plasma displays?<br />If direct-view LCD is going to dominate the small and medium sizes and rear projection units the high end, where do PDPs fit into the market that is taking shape around us?<br /><br />In a word: niche. Not one, though, rather several.<br /><br /><b>Next up: </b>PDP fans discover life’s a niche.<br /><br /><i>Felix Torres is a dabbler in home entertainment electronics and a survivor of both the home computing wars of the 80's and the multimedia wars of the 90's who is currently most interested in home media networks and the North American transition away from broadcast media. </i>