Everybody wants a monster-size HDTV, so it came as no surprise when readers wrote in asking for head-to-head comparisons of the big-screen HDTV technologies. The first round came with “Plasma vs. LCD” (February/March, available on the S&V Web site). After reading our unflinching test of those popular types of hang-on-the-wall TVs, you asked to see LCD take on DLP. Well, here ya go: a battle of the big-screen, rear-projection TV (RPTV) behemoths.
As with our plasma vs. LCD face-off, we sought two same-size sets so that screen area wouldn’t be a factor in the comparison. A pair of 50-inch contenders quickly came into view: Samsung’s HL-P5085W DLP ($3,900) and Hitachi’s 50VS810 LCD ($4,200). The biggest difference between the sets, aside from their display technologies, is that the Hitachi is a fully integrated, digital cable-ready HDTV while the Samsung is an HDTV monitor that needs a separate tuner to receive digital broadcasts. Other than that, both are among the spiffiest-looking and most feature-laden TVs in each manufacturer’s line.
One key similarity between the two sets is the native resolution of their display chips: 1,280 horizontal x 720 vertical pixels. But LCD and DLP RPTVs use very different processes to display pictures. Most DLP (Digital Light Processing) sets have a single chip and a rotating filter that chops white light from a lamp into a sequence of red, green, and blue beams. The beams are reflected from the chip, which contains hundreds of thousands of tiny mirrors. These mirrors pivot thousands of times a second to control the brightness of the pixels and are synchronized to display a red, green, or blue projection that the eye blends into a full-color image.
In an LCD (liquid-crystal display) projection TV, light from a single lamp is directed to a trio of miniature LCD panels that process the red, green, and blue light components separately. The pixels in each panel contain a liquid-crystal material that regulates the amount of light passing through them by twisting and untwisting in response to electrical voltages. After exiting the LCD panels, the three colored beams are combined by a prism and projected onto the screen by a lens.
Just watch a basketball game, concert, or movie in high-definition on a DLP or LCD TV, and you’ll see how much more crisp and detailed their pictures look compared with old-school CRT (cathode-ray tube) rear-projection sets. But detail is only one part of the video-quality equation. Equally important is how well they handle things like color, contrast, and the reproduction of deep blacks — areas where the CRT traditionally excelled. It’s even more important to know how they compare with each other.
To find out, we set up the Samsung DLP and Hitachi LCD side by side and performed an exhaustive comparison. Besides myself, the judges included technical editor David Ranada (DR) and contributing technical editor David Katzmaier (DK), with each of us making his evaluations alone and independently of the others. Each set was calibrated to the best of our ability to conform to the established standards for both regular and high-def programs. In a perfect world, this would have resulted in exactly matched color, brightness, and contrast levels, but given how differently each manufacturer adjusted its TV at the factory, the match wasn’t perfect (see “Lab Setup,” facing page, for details). It was close enough, though, for us to feel comfortable that our test results are valid.
Our procedure was simple: Standard- and high-def programs and test patterns were split using a Gefen 1:2 HDMI splitter or a Key Digital HD View 6 component-video distribution amplifier and viewed on both TVs simultaneously. We ranked each TV’s contrast, color, picture detail, and picture uniformity on a scale from 0 to 5.
Contrast describes the range of tonal steps between the deepest shadows and brightest highlights. It also covers fine points like the set’s ability to maintain solid shadows through changes in overall picture brightness. Color examines how well the set displays a full range of colors at different brightness levels, as well as its overall color intensity. And picture detail gauges the image’s sharpness, while picture uniformity judges its consistency across the screen.
As with the results of our plasma vs. LCD comparison, our judgments here aren’t necessarily representative of all models on the market. But by choosing high-end DLP and LCD RPTVs from reputable makers, we’re giving you a good idea of what to expect when shopping for a big-screen HDTV.
Of all the factors that contribute to a good-looking TV picture, contrast is probably the most crucial because it can affect other important things like color and detail. Our assessment of contrast — an admittedly broad category — took into account not only the TV’s ability to cleanly reproduce a full range of steps between the blackest black and the whitest white, but also the relative evenness of those steps. We also looked at the TV’s shadow depth and detail and its overall brightness level. Finally, we kept an eye on how well it maintained those qualities as the picture shifted between bright and dark scenes.
DR, DK, and I all judged that the Samsung DLP had considerably better shadow depth than the Hitachi LCD — blacks on the DLP simply looked, well, blacker. We could clearly see this in a scene from the Ed Wood DVD (it’s a black-and-white movie) that takes place in a dark bar. Both TVs revealed plenty of fine detail in the pictures and signs hanging on the dank walls — DR concluded that the LCD was better than the DLP at shadow detail — but the LCD’s inability to render a truly deep black flattened its image somewhat. Ultimately, we all agreed that the DLP created a deeper, more three-dimensional picture.
We also all saw shifts in the LCD’s overall contrast. This was most obvious on test patterns, but we could also see it in some movie clips. In a bright mountain scene from The Lord of the Rings: The Fellowship of the Ring on DVD, fine shadings in the snow blended into the surrounding white on the LCD, but the DLP displayed a broad range of subtle tones.
I was irked by the DLP’s tendency to emphasize white areas of the picture, which gave everything an “enhanced” look — the video equivalent of listening to music with your receiver’s treble control turned up too high. To me, picture highlights looked much smoother on the LCD, particularly in bright and high-contrast shots. But “enhanced” is in the eye of the beholder, I guess, because both DK and preferred the DLP’s punchy picture. DK, for instance, liked the greater level of “pop” in dark DVD movies like Pirates of the Caribbean. Ultimately, though, the DLP gave the LCD a sound whomping when it came to contrast.
COLOR
While the finer points of contrast can be hard to quantify, it’s easy to measure a TV’s color performance using test patterns. That explains why DK, DR, and I all had basically the same judgment of the DLP and LCD color rendition. We focused mainly on each set’s ability to maintain consistent color at various brightness levels. (This is actually called “grayscale uniformity” since you test to see how well the TV maintains a specific color of gray.) We also paid close attention to the sets’ overall color saturation and balance, which can mean the difference between seeing rich, realistic turf on a baseball field or a pale approximation that suggests the field’s in bad need of watering.
The DLP did better than the LCD at maintaining consistent color as the image brightness varied. On test patterns, it showed an even ramp of mid-gray tones between black and white, while the LCD showed slight green or red tints at different steps — an effect we could occasionally see in movie clips. DK pointed out that in a high-def D-VHS tape of Terminator 2, the actors’ skin had a pink-red bias on the LCD but looked completely natural on the DLP. Both DR and DK also noted a blue cast to shadows in some of the dark clips we watched on the LCD set, like the bar scene from Ed Wood.
But the LCD’s color-shifting grayscale wasn’t a problem on most programs. And it clearly did a better job than the DLP with color saturation and balance. On almost everything we watched, from T2 to The Fellowship of the Ring to high-def documentaries on PBS, the LCD displayed a richer, more enticing color range. Maybe the most dramatic example of this came when we watched a scene from The Matrix Revolutions DVD. While rain falls on the actors, a bright green glow suffuses them from below. To my eyes, this special effect barely registered on the DLP but came across dramatically on the LCD. DK, meanwhile, described the LCD’s green as being “more toxic, as it should be.” Tallying up the positive and negative points of color performance, the LCD won by a nose.
PICTURE DETAIL
Both the Samsung DLP and Hitachi LCD share the same 720p (progressive-scan) native HDTV display format, so you’d think they’d show the same level of detail, right? Wrong. The number of pixels on a TV’s display chip doesn’t always directly correspond to actual picture resolution. And the processing some TV makers use to heighten the apparent sharpness of pictures only complicates the issue. This processing sometimes adds a harsh, artificial look that makes you wonder why they went through the trouble in the first place.
Resolution test patterns showed that the DLP handled sharpness with high-def signals better than the LCD did. While the DLP came up slightly short of full 720p resolution, it bested the LCD on this test. And with 1080i (interlaced) signals, the LCD looked even softer. Interestingly, the LCD set’s resolution notably improved when we used its HDMI input instead of its component-video input — a testament to the potential for better image quality offered by digital video connections.
When we finally put aside test patterns and started watching movies, the LCD acquitted itself well. Watching the high-def tape of T2 via both sets’ HDMI inputs, DR observed that both TVs did equally well with finely detailed textures like human hair. DK was more critical of the LCD’s overall detail. He noted that in a scene from Master and Commander, the ropes connecting the ship’s sails to its mast — a detail that was clearly visible to him on the DLP — were obscured on the LCD. My impression of the two TVs put me somewhere in the middle. While the DLP revealed more detail on a number of programs, its excessive edge enhancement gave it a noisy, “processed” look. On many levels, I found the LCD’s smooth, noise-free images more pleasing. With the final scores in for picture detail, the DLP took the lead by a full point.
This might seem like an “everything including the kitchen sink” category, but picture uniformity covers important aspects of a TV’s performance that can fly under the radar when you’re in the store drooling over its high-def images. A world of 24/7 high-def programming is definitely coming, but in the meantime we need our HDTVs to do a solid job of “upconverting” the lower-res standard-def pictures from DVD players, VCRs, standard-def cable boxes, satellite receivers, and hard-disk recorders. Picture uniformity also includes screen uniformity, or the evenness of the image at various points across the screen; viewing angle, which is how well a set retains its brightness and color from off-center seats; and whether or not it introduces “artifacts” — blemishes in the picture that may be peculiar to a particular technology.
Both sets delivered solid-looking pictures with standard-def (480i) sources. But pictures on the DLP looked soft viewed side by side with the LCD. Going straight to the first battle scene in Master and Commander — a killer test for false contours (which show up mostly as coarse, unnatural patches or bands in shadowy scenes) — both DK and I noted that the pictures looked smoother on the LCD as the ship sailed through the fog. The DLP’s picture looked patchy in comparison.
Single-chip DLP systems that use a rapidly spinning filter wheel to create full-color images are prone to something called “rainbows” — quick flashes of colored light on the edge of sharp transitions from black to white. While each of us saw several rainbows, they weren’t frequent enough to be distracting (and some viewers may never notice them at all).
LCD, meanwhile, is prone to the “screen-door effect,” a gridlike texture that appears mostly in light images on flat areas of color. It occurs when the LCD panel’s pixel structure is magnified by the TV’s projection lens, and it will be less noticeable if you sit farther away from the TV. Screen-door effect can also be an issue with DLP TVs, but it tends to be a bigger problem with LCD because LCD panels must run their electrical connections in the spaces between the pixels, resulting in wider spacing than their DLP cousins of similar size and pixel count. At our 8-foot viewing distance — a good span for this screen size — it wasn’t apparent on the Samsung. But DK and I could see a faint grid at various times on the Hitachi.
Programs with poor video quality on analog cable channels can sometimes trip up an otherwise impressive TV. In a billiards match on ESPN, the green surface of a table looked patchy and soft on the DLP compared with the LCD, which did a better job with “noisy” programs. But with higher-quality sources, the DLP’s picture showed excellent uniformity, with consistent brightness and color at all points across its screen. The LCD’s image looked brighter at the center of the screen than at the sides and corners, an effect known as hotspotting. It got worse when the image was viewed from off center, but it was primarily visible on test patterns and difficult to detect with movies. The DLP was uniformly bright up to 30° off center.
While you should keep an eye out for the LCD screen-door effect and DLP rainbows when shopping for a big-screen TV, other performance aspects like upconversion and viewing angle vary from set to set, so we can’t stamp them as DLP- or LCD-related issues. When all the aspects of picture uniformity were taken into account, the result was a draw.
Which is the better big-screen TV choice: DLP or LCD? Our final tally finds DLP with a half-point lead, but the differences between the TVs weren’t as dramatic here as they were in our plasma-vs.-LCD face-off, which was a clean sweep for plasma over flat-panel LCD. DLP ran circles around LCD when it came to contrast. It also beat LCD by a comfortable margin with picture detail. But LCD slightly outperformed DLP in terms of color performance, and both did equally well with picture uniformity. So while DLP wins this battle, LCD came in a close second — so close that both technologies deserve a careful look when you’re out shopping for your next TV.
DLP vs. LCD: Just the FactsNow that you know the photo-finish results of our DLP/LCD matchup, here are some more things you should know about each technology before buying a TV. First, most DLP and LCD rear-projection TVs (RPTVs) have a native resolution of 1,280 x 720 pixels, which matches the 720p HDTV format. A few newer — and more expensive — DLP sets bump that number up to 1,920 x 1,080 pixels, which is enough to fully accommodate the 1080i HDTV format used by most broadcasters. No 1,920 x 1,080-res LCD RPTVs have been announced.
The single-chip configuration used for DLP RPTVs — the source of “rainbow” effects — isn’t likely to change anytime soon. Some DLP front projectors use three chips, but all these cost $30,000 or more. On the other hand, all LCD projectors, both front and rear models, use a three-panel configuration to display images.
DLP and LCD RPTVs cost about the same, starting at around $2,800 to $3,000 for a 42- to 52-inch set and moving up to $6,000 to $7,000 for a 67- to 70-inch HDTV. The price for a specific model will vary according to its features — things like a built-in high-def tuner, a CableCARD slot, and a wide range of video inputs. With both TV types, you’ll need to replace the projection lamp after about 3,000 hours. Lamps usually cost $200 to $400. —A.G.
| In the Lab | ||
| DLP | LCD | |
| Color Temperature (before/after calibration) | ||
| Low Window | 7,855/ 6,570 K |
6,149/ 6,501 K |
| High Window | 7,840/ 6,395 K |
6,192/ 6,523 K |
| Brightness | 99 ftL | 76.2 ftL |
| Both the Samsung HL-P5085W DLP and Hitachi 50VS810 LCD HDTVs underwent grayscale calibration before testing to improve their color performance. After calibration, grayscale tracking was excellent on the Samsung, varying by an average of 84 K, and good on the Hitachi, varying by an average of 164 K. Both sets could deliver extremely bright images, but the Samsung’s post-adjustment light output measured considerably higher than the Hitachi’s. To even the playing field for a side-by-side comparison, the DLP set’s brightness was toned down. One reason for the differences in color that we observed was the Samsung’s substantial color-decoder error: –20% red on the Avia setup DVD’s test pattern. On the same pattern, the Hitachi displayed only a slight error (–5% green) that was easily fixed via the TV’s user controls. —A.G. | ||
Our three video experts summarize their impressions
David Ranada
These sets produced much more closely matched images than the ones in our plasma/LCD comparison. The most significant concern with the DLP was the rainbow effect, but I never saw it on movies. While both sets could have done a better job of producing darker blacks, the LCD’s black was fairly “bright” and bluish to boot, giving the darker areas in black-and-white films a distinctly blue tint. To me, the LCD also had a “dirty-screen” effect (technically, fixed-pattern noise) — unrelated to the “screen-door” effect mentioned elsewhere — that could be seen not only in basic test patterns but also in all areas of smooth texture that move during a scene, such as in animated features like Tron and The Incredibles. So while the casual observer would be impressed at the similarity between the sets, the DLP had fewer problems with critical program material and is the one I preferred.
Al Griffin
In the interest of full disclosure, I have to admit to walking into this test with a bias for DLP rear-projection TVs. Although I’d never actually reviewed a rear-projection LCD set, I’d seen plenty of them in stores and at trade shows, and their pictures always looked pale and fuzzy. But the Hitachi LCD surprised me. Its color was rich and natural, lending realism to live-action movies and visual punch to animated movies like Toy Story. Compared with the “enhanced,” and occasionally patchy, look of the DLP image, the LCD picture was impressively smooth and clean. The area where the DLP clearly beat LCD — and it’s a key one — was contrast. As a CRT aficionado who only recently abandoned tubes for the new digital displays for home viewing, I prefer the DLP’s deep, CRT-like blacks and more realistic shadow detail. I also found this DLP’s rainbows less off-putting than the LCD’s seemingly inescapable screen-door effect. I’m sure I’d find things to like with a big-screen LCD TV in my home theater, but given the choice, I’d buy a DLP.
David Katzmaier
I definitely prefer the DLP’s picture. Its significantly deeper blacks lent plenty of pop to both bright and dark scenes, especially compared with the LCD. While the LCD had excellent color accuracy, I’m convinced that the DLP’s relatively poor performance in these areas was the fault of this particular TV and not of the DLP technology. I was surprised by the significant increase in detail when watching both DVDs and HDTV on the Samsung. While the LCD’s softer look was readily apparent in side-by-side comparisons, it was probably a problem with this particular TV and not with the LCD technology. The DLP did reveal more video noise, which seemed in most cases to be present in the source as opposed to a fault of the technology. But I don't find this a huge distraction, especially when you’re sitting a reasonable distance from the TV. And I’d be willing to trade a little noise for more picture detail. Yes, I did see lots of rainbows in the DLP (more than usual since I was constantly flicking my eyes back and forth from the DLP to the LCD). But I found the LCD’s screen-door effect even more distracting, especially in pans over bright areas where the grid seemed to sweep across the image.
If you liked this article, you might enjoy the following:
DLP vs. LCoS
Plasma vs. LCD
Back to Homepage
What's New on S&V