You may have noticed that several projector makers now publish a controversial new specification known as Color Light Output (CLO), along with the traditional ANSI Lumen ratings on their spec sheets.

Why do we need two different ways to measure a projector's brightness? In this article we'll explore how Color Light Output differs from ANSI lumens, and what it means to you as the projector buyer.


Does Color Light Output (CLO) matter?

Advocates of the new CLO spec argue that since three-chip projectors and single-chip projectors create white and color values differently, the ANSI lumen spec is not a valid way to compare their brightness. Certainly, if a 3LCD projector and a DLP projector both measure 3000 ANSI lumens (which is a measure of white brightness only), that means they can both project a white test pattern of equal brightness. But, they point out, who watches a white test pattern? Isn't it more important to know how bright projectors are when displaying full color images? And though the 3LCD and DLP both produce 3000 ANSI lumens of white, the color images on the DLP projector will often be dimmer than they are on the 3LCD. The CLO spec, it is argued, takes color brightness into account and gives buyers more info about the projectors they are evaluating.

"Not so fast, there bub," say those who object to the CLO spec. Though CLO measures color brightness, it does not take into account color accuracy. In order to get the highest ANSI lumen and CLO ratings out of a 3LCD projector, you must run all three chips wide open. And if the UHP lamp behind them has a green bias as they typically do, then the white light on the screen will have a green tint. So you may have a projector rated at 3000 ANSI lumens of white light and 3000 lumens of Color Light Output, but the picture looks bad anyway because the color balance is way off. Since the CLO spec does not address color balance, it gives buyers nothing new or important about the projectors they are evaluating.

Meanwhile, DLP engineers can compensate for green lamp bias with a larger red filter, so the white light from DLP is often a cleaner, more neutral white than you'd normally get from a 3LCD projector with all chips wide open. And if you try to calibrate out the green bias on the 3LCD its lumen output drops, sometimes a lot, and the ANSI lumen and CLO specs may no longer be relevant. So how does the introduction of a new brightness spec that is flawed in the same way the ANSI lumen spec is solve anyone's problem?

The thing is, people on both sides of the CLO debate have excellent points. But it is much easier to understand the controversy when you can actually see the differences in actual images rather than just talk about the concepts. So that's what we will show you in this article.


Definitions: ANSI Lumens vs Color Light Output

ANSI lumens: The ANSI lumen spec measures the brightest white that the projector can produce. It is measured by taking meter readings on a projected 100% white test pattern. The number you derive from the readings, say 3000 ANSI lumens, is the maximum brightness of white that the projector is capable of.

Color Light Output ("CLO"): The CLO method is similar, but instead of using a 100% white test pattern, one uses red, green and blue test patterns instead. Separate meter readings are taken for red, green, and blue, then added together. This time if we end up with 3000 lumens, that is the maximum brightness of color that the projector is capable of. So it is called color brightness, or color light output.

Why do different technologies create color differently?

All three-chip projectors, whether DLP, 3LCD, or LCoS, have three independent color channels for red, green, and blue. In order to produce white, all three color channels are turned fully on. So by definition, maximum color light output and maximum white light output are always the same. If a projector measures 3000 ANSI lumens, it will also produce 3000 lumens of Color Light Output.

However, on most single-chip DLP projectors made for business or classroom use, color is defined with a spinning color wheel that has not only red, green, and blue filters, but also a white filter and perhaps some secondary filters like cyan or yellow. The white filter allows white light from the lamp to bypass the color filters and boost the brilliance of the image. By doing this you end up with a very bright white as measured by the ANSI lumen method. But if you measure red, green, and blue independently and add them up like the CLO spec calls for, they don't add up to the total white value. So on a single-chip DLP projector with a white filter in the wheel, the CLO number would be less, often much less, than the ANSI lumen number. Since there is no marketing advantage to the makers of single-chip DLP projectors to quote CLO numbers, they never do.

The big question is...what difference does it make?


Executive Summary

For those who don't have time
to read the whole article...

The test shoots in this article illustrate that the ANSI lumen spec and the new Color Light Output spec can both be misleading, but for different reasons.

The ANSI lumen spec is often misleading because it does not take either color brightness or color accuracy into account. Knowing that a projector is rated at, say, 3000 ANSI lumens does not tell us anything about how it will perform with full color images. The picture's whites may be bright, but the colors may be dim, or poorly saturated or wildly inaccurate. The picture may have an objectionable tint. Most commercial/business class projectors (regardless of the technology used) have flaws of this nature that the ANSI lumen ratings ignore. Once you adjust the picture controls to get rid of the color flaws, a projector can end up putting out much less light than its specs would indicate.

The new Color Light Output spec suffers from similar limitations: it does not take color accuracy into account, and it reveals nothing about whether the projector is calibrated to retain highlight detail. So a projector may have two specs that say "ANSI Lumens: 3000 and Color Light Output: 3000" and still end up giving you a picture you would not want to watch. Once the projector is calibrated for best picture quality, it may put out far less light than its ANSI lumen / CLO specs would indicate.

Despite its flaws, the CLO spec does point to a real difference between three-chip projectors vs. single-chip DLP projectors. These two different technologies often do produce pictures that contain quite different amounts of color information. When a 3LCD projector and a DLP projector with a white filter have the same ANSI lumen rating, the DLP may produce noticeably dimmer full color images. To that degree, the "white only" ANSI lumen spec can be viewed as misleading and biased in favor of single-chip DLP projectors. In theory, the CLO spec is intended to remedy this bias.

Those who do not support the CLO spec say that its only real purpose is to point out a particular weakness in DLP technology without drawing attention to related flaws in 3LCD technology--most commonly, the unattractive blue/greenish tint you get when the 3LCD projector is in its brightest operating modes. And from a practical perspective, since most vendors will not publish CLO specs, they cannot be used as points of comparison in the same way that ANSI lumen numbers typically are. So in the end, detractors insist that CLO should be seen for what it is--a marketing tool used by 3LCD promoters rather than an official spec that has been accepted and embraced by the industry.

Nevertheless, though the new CLO spec is indeed a marketing-driven venture by those who promote 3LCD technology, the fundamental point that CLO makes is valid: the traditional ANSI lumen spec utterly fails to provide buyers with an accurate apples-to-apples comparison of how bright projectors really are when they are used to display full color images.

Setting up the Test

To see how DLP and 3LCD projectors can differ in their display of color, we selected two WXGA projectors (one 3LCD and one DLP) built for larger conference room applications. They had similar ANSI lumen specs above 5000 lumens, and both were rated at 2000:1 contrast. Since the goal of this study is to explore the weaknesses of 3LCD and DLP technologies as they relate to ANSI lumen and CLO specs, by prior agreement with those who supplied us the test units we will not disclose the models used. We are grateful to those who contributed the equipment, and for their willingness to support this study.

To guarantee screen neutrality we used an 11-foot wide Studiotek 100 supplied by Stewart Filmscreen. Our thanks to the folks at Stewart for providing this product. The Studiotek 100 is a perfectly neutral white, 1.0 gain screen, and it is ideal for this type of comparative evaluation. We placed the projectors on racks and projected two 5-foot wide images onto this screen so we could take simultaneous screen shots of both.

Two Tests at Different Calibrations

In the projector world, when you maximize light output you compromise picture quality. Color balance goes out the window, details in highlights can often disappear, black levels are not so black, colors get not so saturated. For many reasons the picture does not look as good as it would if you reduced the light output of the projector to bring these factors into balance. But ANSI lumen and CLO specs are frequently based on the projector's maximum light output, regardless of what the picture looks like, in order to get the biggest numbers for marketing reasons.

By the way, this is not the vendors' fault. The problem exists due to the fact that the ANSI lumen spec has no color balance standard--it doesn't care what the picture looks like, only how bright it is. Meanwhile, professional buyers for corporations, governments and school districts routinely issue Requests for Quote that specify lumen requirements but never stipulate color balance requirements. If vendors are to survive in this world they need to play the cards they've been dealt. At the present time, that means they must market products with lumen ratings that are not based on ideal color calibrations.

SO...given that this is the case, in this 3LCD vs. DLP test we set up the projectors two ways and took side-by-side comparison shots in each:

Test # 1. Brightest Mode, No Calibration: In the first series of test shots we put the projectors into their factory defaults for their brightest operating modes. This shows what they do out of the box with no adjustments, just selecting their brightest configurations. Set up this way, we get the maximum lumen output along with some obviously degraded picture quality.

Test # 2. Reduced Brightness, Improved Picture Quality: In the second series of test shots we adjusted color balance and contrast on both projectors to produce the brightest pictures possible consistent with a more neutral color balance and a sufficient rendering of detail in the highlights. The light output was reduced, color quality on both units improved, and the pictures looked much better.





TEST # 1:
PROJECTORS SET TO MAXIMUM
LIGHT OUTPUT w/ NO CALIBRATION


IMPORTANT NOTE: The results of this test are valid only for these two particular models, not all LCD and DLP projectors in general. These two test units show image characteristics that are common in commercial/business projectors but not home theater. Furthermore, business-class DLP projectors can be designed with a variety of color wheel configurations that will produce different results. Typically, home theater DLP and 3LCD projectors are optimized for maximum color quality, not maximum light output, and the results herein do not apply.

1. Monoscope Test Pattern

LCD vs DLP, Test Pattern 13LCD Projector                                                             DLP Projector

These pictures show how these two projectors look in their brightest pre-programmed operating modes, typically called either Dynamic or Presentation on most projectors. Settings are in factory defaults, so this is the way they perform out of the box.

The most obvious difference is that the LCD has a decidedly blue/green cast to the image, whereas the DLP is comparatively neutral. The DLP is actually slightly warmer in color temperature than neutral, but its white is much closer to pure white than we get with the LCD. Its gray values appear gray, without any apparent color tint.

ANSI Lumens: By official ANSI lumen standards the DLP projector is the brighter of the two. Its official rating is 1000 lumens higher than that of the LCD. Our meters indicate that the LCD model is producing 4545 ANSI lumens in this mode, and the DLP is at 5800, or 28% brighter on white measurements.

Color Light Output: As per the previous discussion, the CLO number on an LCD projector will always equal the ANSI lumen value. In this case, the sum total of red, green, and blue luminance values adds up to 100% of white (or what passes for white) on the LCD. On the other hand the CLO number on this DLP projector in this maximum brightness operating mode measures about 1300 lumens of color light output, or just 22% of white.

2. Resolution Test Pattern

LCD vs DLP Test Pattern 2 3LCD Projector                                                             DLP Projector

2. This is a resolution test pattern, but for our purposes here it reveals more about the color bias on the LCD. Notice that it is decidedly blue in the lower to mid-tone values, shifting to cyan as it gets brighter, then to green in the highlights. This chart is supposed to be neutral gray and white, pretty much as it appears on the DLP. What this chart reveals is that any picture being shown on the LCD projector (in this uncalibrated bright mode) will have a cool bias. Mid-tone gray objects will always appear blue. Whites will have a greenish tint. Skies and blue water will appear bluer than they are in the original source, and skin tones will look sickly.

The color bias on the LCD projector can be calibrated out, but the brightness of the picture will be reduced. We will see that in the second set of test shots later in this article.

3. Color Bar Test Pattern

LCD vs DLP Color Bars, uncalibrated 3LCD Projector                                                             DLP Projector

3. On this standard color bar test pattern, compare how these two projectors are displaying it with the ideal pattern below. Neither projector is displaying it correctly, but the errors are quite different. And of the two, from a color accuracy perspective, the 3LCD is doing a better job than the DLP. Its red bar in particular is much closer to accurate than the DLP's dim, brownish tone. Cyan is also reasonably accurate on the LCD compared to the teal on the DLP. On the other hand, the LCD suffers from a lack of saturation--none of the colors are as rich and vibrant as they should be. Blue and magenta in particular are downright pastel.

ColorBars

The DLP projector's interpretation of the color bar pattern is quite poor. Red is brown, magenta is purple, cyan is teal, and with the notable exception of white, they are all dull.

The errors on these projectors can be reduced or in some cases corrected entirely with some attention to calibration of contrast and color balance, but only at the expense of lumen output.

Leaving aside color accuracy for the moment, it is easy to see from the screen shot that the LCD rendering of this pattern is substantially brighter than that of the DLP. This goes directly to the point that advocates of the CLO spec try to make: This DLP projector measures 28% brighter on a white pattern, but the LCD is able to generate substantially brighter color values. So if you intend to project full color subject matter, the ANSI lumen spec is not a valid measure of brightness between LCD and DLP projectors (that is, single-chip DLP projectors with white filters in their color wheels).


What difference does it really make? Let's take a look...

4. Wine Glasses

LCD Projector                                                             DLP Projector

4. A couple of wine glasses show how these errors translate into real life. Here, the red color of the wine is obviously more vibrant on the LCD. The dimmer brownish red we saw in the color bars above manifests itself here with a relatively weak rendering of the wine. On the other hand, contrast is much better on the DLP. The wineglass is much better defined against the brighter white background. On the LCD you see the undesirable tonal shift in white, and the glass merges with the background--the highlight detail is missing. This can be fixed on the 3LCD projector by reducing contrast at the expense of overall image brightness.

5. Night Scene

LCD Projector                                                             DLP Projector

5. A shot of a city at night reveals the latent advantage in contrast and black level that DLP projectors often have over their LCD counterparts. Though the image is much brighter on the LCD, it is bluish and dull compared to the DLP projector which presents this particular scene much more successfully despite a bias toward green. This can also be fixed with an adjustment on the LCD, but again at the expense of image brightness.

In these tests we tended to notice more digital noise on the 3LCD projector, and it was quite abundant in dark scenes like this one in which it is visible even in the small screen shot.

6. Desktop, No Color

3LCD Projector                                                             DLP Projector

6. Here is a scene with very little color content. The DLP projector does a much better job with this image than does the LCD. It is more neutral in color and higher in contrast. For any type of subject matter that has little color the DLP (in its brightest dynamic/presentation mode) will have a latent advantage. That includes everything from financial spreadsheets and text documents to black and white photography.


7. Pencils

3LCD Projector                                                             DLP Projector

7. This image is a mix of gray and colored elements. The background is actually middle gray, and the DLP projector presents it fairly accurately while the LCD interprets it as blue. The colored pencils are brighter on the LCD and their color tone is closer to accurate, but they lack saturation and are too pastel.

On the DLP projector, while it gets the gray background correct, its color shows greater inaccuracies. The yellow pencil appears as olive green and the orange appears to be a deep, solid brown. But the white highlights on the glass are correct on the DLP while appearing blue-green on the LCD. The LCD picture is brighter, the DLP picture is higher in contrast. Neither image is a faithful reproduction of the source signal.

As an aside, on some of these screen shots you will notice some brightness uniformity issues, and it is most obvious in this image--the LCD is darker on the right, and the DLP is darker on the left. This is due to some vignetting in the wide angle camera lens used to make the screen shot, not to any uniformity problem with the projectors.

8. Balls of Yarn

LCD Projector                                                             DLP Projector

8. An example of high color saturation. The flaws we have seen in previous images appear here as well. The LCD colors shift toward undersaturated pastel, the highlights are blown out, and the image has a bluish tint overall. Meanwhile, the DLP colors are much more saturated, but dimmer with significant inaccuracies---yellow appears orange, cyan appears green, and red appears dark plum. The DLP does a much better job holding detail in the highlights, but the picture is relatively dim.

The LCD projector is certainly much brighter, and many might say more successful. With a bit of calibration to reduce the overdriven highlights, the LCD will look a great deal better. Promoters of the CLO spec are correct to point to the inadequacy of the ANSI lumen spec which rates the DLP projector 1000 lumens brighter than this LCD. Clearly with full color images the LCD is able to put a brighter image on the screen, and in many cases it will be the more compelling image.

9. Bell Peppers

3LCD Projector                                                             DLP Projector

9. This is a good example of disaster on both projectors. The LCD projector is blowing out highlight detail, and colors are not as saturated as they should be. However, there is no question that the LCD's picture is much brighter than the DLP. The saturated red areas measure a whopping four times brighter.

This configuration on the DLP projector has the "Brilliant Color" control at its maximum setting, which means it is pumping out as much white light as possible. Bright whites against subdued color create increased contrast. The green stems are slightly darker than ideal, but there is much more balanced detail in them than is apparent on the LCD. The big problem is that the reds really get hammered in this image.

Both of these images have obvious flaws. Reasonable people may differ on which one is the more successful rendering of the subject, but everyone would agree that neither one is great.

10. Pastels

3LCD Projector                                                             DLP Projector

10. In this scene of pastel colors the DLP image appears to be higher in color saturation than the LCD. Most would agree that the DLP's rendering is the more successful of the two. In this situation the DLP's subdued color brightness renders the subjects darker, which contributes to the impression of higher saturation and contrast. However, there is significant color shift. The ball of yarn in the middle is supposed to be blue as it is on the LCD, but it is teal. The yellows show an olive green bias as well.

Meanwhile, the overdriven contrast on the LCD reduces saturation, and the blue-green bias interferes with color interpretation. These flaws combine to produce a decidedly poor result. Both of these flaws can be substantially reduced as we will see in Test # 2, but getting rid of the blue-green tint and reducing contrast to bring back highlight detail will reduce the total lumen output of the LCD.

11. Maiko, Kyoto

3LCD Projector                                                             DLP Projector

11. The DLP projector has a clearly superior image with this subject. The tonal values in the face appear to be reasonably accurate, whereas the LCD renders the young lady's face with very little detail and a decidedly green bias. The red tones on the DLP are much darker than they should be, as this is a bright red kimono and bright red lipstick. But the combination of good contrast, neutral gray tones and poor color accuracy combine to create an image that, while not accurate in some respects, is certainly more attractive.

11. Black and White Photography

LCD Projector                                                             DLP Projector

11. I do black and white photography in my spare time--this is my photo of Mesquite Dunes in Death Valley. When it comes to presenting my work on the large screen, the DLP projector has two big advantages. First, the neutral gray scale is vital. The LCD's blue tint can be largely calibrated out, but once that is done the projector is about 60% of the white brightness of the DLP projector.

Second, this DLP projector has an advantage over the LCD in "ANSI contrast," which is the range between black and white in a given frame. What this translates to is more image detail being retained in both the highlights and the shadows/dark areas. You would not know this from the specs since both of these projectors are rated at 2000:1 Full On/Off Contrast. Unfortunately, few vendors publish ANSI contrast specs. But for reasons of both gray scale neutrality and inherent contrast, the DLP is the better of these two commercial projectors for the display of b/w photography.

12. PowerPoint Presentation I

3LCD Projector                                                             DLP Projector

12. This image reveals how key weaknesses of both projectors can affect PowerPoint presentations when they are run at maximum lumen output. The LCD's tinted image reduces contrast and makes it harder to read than it otherwise would be.

However, the DLP projector has a big problem here as well. At first glance the DLP appears to be the more successful of the two images. But if you look closely on the 3LCD image you can see that the bars in the graph have numeric data on them. The LCD's much brighter bar color makes this data visible, whereas the DLP's very dark coloring on the bars pretty much eliminates any ability to see and read the data.

If you are preparing a PowerPoint presentation using full color graphics, be aware that the colors you see on your computer screen can be radically reinterpreted when they are displayed on commercial/business class LCD or DLP projectors in their brightest modes. Your presentation should be previewed on the projector itself to make sure you don't end up with problems like those in this image.

13. PowerPoint Presentation II

3LCD Projector                                                             DLP Projector

13. In this full color PowerPoint slide the vibrant color on the 3LCD projector makes the overall picture brighter and the text easier to read. Especially in ambient light, the extra color brightness helps. Keep in mind that this DLP projector as it is currently calibrated measures 28% brighter on an ANSI lumen basis. But that is for white only. In this image the LCD's yellow is 80% brighter than that of the DLP. The only element that is brighter on the DLP's image is the white text "Usability."

With this type of subject matter the DLP's much darker rendering of yellow reduces the contrast between the background and the text. Even though the DLP has an intrinsic advantage in contrast with text documents, spreadsheets, and other non-color subject matter, it can show much lower contrast when the projected image has a lot of color as it does here.


This series of screen shots illustrates the various problems we might encounter when running projectors in their brightest, maximum lumen settings. What happens when we apply some color and contrast adjustments to our projectors while sacrificing some brightness? Let's go on to Test # 2 and take a look...





TEST # 2:
PROJECTORS REDUCED in BRIGHTNESS
with IMPROVED COLOR AND CONTRAST


IMPORTANT NOTE: The results of this test are valid only for these two particular models, not all LCD and DLP projectors in general. These two test units show image characteristics that are common in commercial/business projectors but not home theater. Furthermore, business-class DLP projectors can be designed with a variety of color wheel configurations that will produce different results. Typically, home theater DLP and 3LCD projectors are optimized for maximum color quality, not maximum light output, and the results herein do not apply.

Objectives:

In this second series of shots we see what happens when we calibrate both projectors for better color balance and contrast, and we set them to an equal ANSI lumen output.

Resetting the LCD projector. The calibration adjustments have taken most of the blue-green bias out of the picture. The contrast control was reduced substantially to allow highlight details to display properly. Brightness was reduced to deepen black levels. All of these changes were at the expense of image brightness. After adjustments the ANSI lumen output (white brightness) dropped about 24% from 4545 lumens to 3450 ANSI lumens.

Resetting the DLP projector. In the first test the DLP was putting out 5800 lumens of white light, and the color brightness was 1300 lumens, or about 22% of white. For this test we reduced the white light output to 3500 ANSI lumens to match the LCD, while keeping color values the same. This was done by reducing the "Brilliant Color" control, in this case cutting it in half. The end result was that color light was still measuring 1300 lumens, but it was now 37% of white instead of 22%. Increasing the percentage of color information relative to white improves overall color performance, as you can see in the following test images.

On a DLP projector is it usually an option to continue reducing the white component in the picture by further reducing the Brilliant Color control, and it can be taken all the way to zero. At zero, all of the light from the lamp is being directed through the color filters and none through the white filter. Accordingly, on this particular DLP projector in this configuration, ANSI white light lumens can be made to equal CLO color lumens, and both are 1300 lumens. The Brilliant Color control on a commercial DLP projector can usually be set anywhere between 0 and 10, and it allows you to seek the optimum trade-off between maximum lumen output (10) and maximum color performance (0) for your most important type of display content.



1. Monoscope Test Pattern

3LCD Projector                                                             DLP Projector

1. After calibration, most of the blue tint that was obvious on the LCD projector in the first test has been eliminated. These are not home theater projectors so a perfectly neutral grayscale is difficult to achieve with the high light output. The objective was to move the LCD as close to a neutral gray scale as possible while still maintaining substantial light output. There is still some subtle blue bias in the mid-tones, but it is not nearly as blue as it was. The green tint in the white is gone. The contrast control was reduced substantially to bring highlight details back into the pictures.

Meanwhile, the DLP projector is now set to an ANSI lumen white output of 3500 lumens to match the 3LCD. The highlights show a slightly warmer than neutral tone, and the lower light levels still contain a bit more green than is ideal. But the two projectors are essentially equal in white light output. Color balance, while not home theater perfect, is well within tolerable limits for commercial applications.

Color Bar Test Pattern

3LCD Projector                                                             DLP Projector

ColorBars

2. With these calibration adjustments, the color bar test patterns look somewhat improved. The LCD projector is slightly more saturated, the DLP is slightly brighter than it was. The most obvious difference between the two projectors is still that the colors on the LCD are brighter across the board. Red in particular remains substantially weaker on the DLP than on the LCD. This weakness in red will show up in a variety of ways in real life images.


So enough of test patterns, let's move on to see what all this looks like in actual photographs...







3. Colored Paper

3LCD Projector                                                             DLP Projector

3. In this high saturation image the LCD projector displays an obviously brighter picture. However, if one focuses on the white strip, the DLP projector appears to be higher in contrast.

The apparent contrast advantage of the DLP comes from two factors we've already seen--the reduced color brightness increases the difference between white highlights and color subject matter, and DLP projectors generally have a greater range of white to black in a given scene than do LCD projectors.

These two projectors are both rated at 2000:1 Full On/Off Contrast. Unfortunately the Full On/Off Contrast spec that vendors typically publish says nothing about the ratio of black to white in a given scene. For that we would need the ANSI Contrast spec. Very few vendors publish ANSI Contrast numbers since they are usually well below 1000:1, and therefore have no marketing sizzle.

4. Pastels

3LCD Projector                                                             DLP Projector

In this scene of pastel colors we see the same trade-off as in the higher saturation scene of colored paper. The LCD projector is once again brighter, and the DLP's image is not as bright but is higher in contrast and has more apparent color saturation. The LCD's blue values are double the brightness of the DLP, and its yellows are 50% brighter. But the lower color brightness on the DLP renders the colors darker, so they look higher in saturation and the picture looks higher in contrast. So here the lower color brightness actually ends up producing a more pleasing, if less accurate image.

Since the white component has been reduced substantially on the DLP, color light as a percentage of white has been increased. This improves color accuracy. In this configuration, the blue ball in the center is now blue rather than teal as it was the first time. Yellow is still off a bit, but these errors are not as pronounced as they were in the first test. Reducing the "Brilliant Color" control tends to improve color on many DLP projectors as it does on this one.

5. Fall Color

3LCD Projector                                                             DLP Projector

5. This is another example of low saturation colors. The LCD projector is rendering the scene with more accurate color and brightness balance. In the actual theater the 3LCD is obviously brighter, though the process of taking a screen shot and reproducing it on a monitor tends to neutralize the apparent difference. But the darkening of the red foliage on the DLP (which is an error) creates higher contrast. Many people might say they prefer the way it is being displayed by the DLP due to the higher contrast.

6. Pencils

3LCD Projector                                                             DLP Projector

6. Recall that prior to calibration the LCD projector was interpreting the background of this image as blue. With the color adjustments, it does a much better job seeing the background as gray, though it retains a more subtle bluish tint.

In this image the LCD has an advantage in color brightness and accuracy. According to the spot meter, the yellow pencil is 50% brighter on the LCD than on the DLP, and it is rendering a reasonably satisfactory yellow. The DLP's yellow has a shift toward olive once again. The orange pencil is double the brightness on the LCD compared to the DLP. Here again the DLP shifts orange toward brown though not as badly as in the first test; in this image the orange pencil still retains an orangish hue, where before it was fully brown.

On the other hand, the DLP projector again shows greater apparent contrast and saturation. The darker colors look higher in saturation, while the white reflection glare on the pencils is the same brightness on both projectors. So the greater range in brightness between the white glare and the darker colors on the DLP contributes to higher contrast.

In short, both of these projectors do better with this image after calibration, but the LCD is ultimately the brighter and more accurate of the two.

7. Bread and Fruit

3LCD Projector                                                             DLP Projector

7. The weakness of red on the DLP projector is quite obvious when subject matter of high-saturation red is a dominant element in the image. All commercial DLP projectors may not manifest a weakness in red to the degree that this one does. But we have seen this on other DLP models and it is not uncommon. If the content you plan to display uses a lot of high saturation red (particularly vital for eye-catching red highlights in digital signage and advertising applications), this is a limitation that should be explored prior to purchasing a projector.

Also notice in this scene that white values (the bread) are about equal in the two images, whereas the color values are quite different in brightness. The LCD does a better job in this scene of producing a more natural balance of illumination.

8. Woman and Bamboo

3LCD Projector                                                             DLP Projector

8. This image illustrates the subtle effects of a weak red channel. The LCD image in this scene is brighter, but only by about 20% so not enough to make a difference in most situations. The larger problem here is accurate flesh tones. Lighter skin tones require a bit of red to make the skin look naturally warm, or pinkish. The weakness in the red channel on the DLP projector shifts the young woman's skin tones toward a less attractive gray-green hue.

If you intend to display portraiture or any other frequent display of flesh tones, it is important to avoid this particular flaw. It can be avoided on most DLP projectors by turning the "Brilliant Color" control off or to very low values, or by choosing a DLP projector that has a much smaller white segment in the color wheel than they often do. You may give up some light output, but the display of pleasing and accurate skin tones is vital when people constitute an important part of the content to be presented.

9. Cityscape, Night Scene

3LCD Projector                                                             DLP Projector

9. Color is never the only issue in picture quality. Other contributing factors include dynamic range, black levels, digital noise, color uniformity, and image sharpness. DLP projectors can often outperform LCD on some or all of these aspects of image quality. This DLP projector outperforms its LCD counterpart in dark scene detail and black level, which is not unusual.

Now that this LCD projector is calibrated, this scene looks far better than it did in its maximum brightness mode. Its black level rivals the DLP without losing much detail. Brightness can be reduced even further for a deeper black at the expense of shadow detail, if that were desired.


Concluding Thoughts

This comparison of two particular DLP and 3LCD projectors shows that the traditional ANSI lumen spec and the new Color Light Output spec can both be highly misleading, but for different reasons.

The ANSI lumen spec is a problem because it does not take either color brightness or color accuracy into account. Knowing that a projector is rated at 3000 ANSI lumens does not tell us anything about color saturation, color accuracy, image tint, or how dim colors might be. Projectors can and often do deliver bright but unattractive pictures at their rated ANSI lumen output. How much light must you sacrifice in order to get a balanced picture? The ANSI lumen spec gives you no idea. So it is really no standard at all.

Into this ANSI lumen mess comes the new CLO spec which suffers from similar limitations that plague the ANSI lumen concept: It does not take color accuracy into account, and it reveals nothing about whether the projector is calibrated to retain highlight detail. So you can have two specs on a product sheet that says "ANSI Lumens: 3000 and Color Light Output: 3000" and still end up with a picture you would not want to watch.

Despite the flaws it has in common with ANSI lumens, the CLO spec does point to a real difference in how color and white light are created on three-chip projectors vs. single-chip DLP projectors. Practically speaking, the CLO spec is used as a marketing tool to highlight what several 3LCD vendors see as a weakness in the design of single-chip DLP projectors. The DLP vendors counter with the claim that the white filter is no weakness at all, but rather the extra white light increases the brilliance and apparent contrast of the picture and thus should be viewed as a benefit.

Setting aside the marketing rhetoric, this study shows that the color limitations in single-chip DLP projectors can be real. To that degree the proponents of CLO make a valid point. Buyers should be aware that an LCD and a DLP projector with the same ANSI lumen rating may not deliver the same brightness of color images. However, the fact that CLO is not based on any color balanced standard leaves it open to much of the same criticism that ANSI lumens suffers. In this test, our LCD projector lost 24% of its brightness after adjustments to remove the blue/green tint and to regain highlight detail. So if a projector is rated at 4000 Color Light Output lumens, but only produces 3040 lumens once the picture actually looks good, that is something buyers should know as well. But the CLO spec remains silent on this issue.

Specsmanship is and always has been a marketing game, at least where brightness and contrast specs are concerned, so the controversy will continue. One thing is certain--the debate will raise awareness of not only the inadequacy of the ANSI lumen spec, but the inadequacy of all attempts to quantify a projector's brightness without a more comprehensive and commonly agreed upon set of standards.

Comments (25) Post a Comment
Gary Hatch Posted Sep 19, 2013 1:23 PM PST
What would a Dila/LCoS (JVC/SONY) projector fall in this mix?
Evan Powell (Editor) Posted Sep 19, 2013 2:04 PM PST
JVC's D-ILA and Sony's SXRD are both versions of LCoS. The projectors they make with this technology are three-chip units, so white light and color light values will always be the same on them. Sony quotes the CLO spec and JVC does not.
Darin Posted Oct 30, 2013 10:51 AM PST
Definitely some interesting results and comments. Thanks. I'm wondering if for Powerpoints in Test 1 you looked at anything like pie charts or line graphs that use multiple colors. Given the results in the Color Bar Test Pattern it might be interesting to see how Powerpoint presentations that are trying to differentiate information by color would tend to fare on each in a room with lots of room lighting on.
TimN Posted Feb 25, 2014 9:59 AM PST
It would be interesting to take this same evaluation and conduct it a year later or with say a 1000 hours on the projectors. Change the lamp and run the same tests again.

I'm guessing you would not get the same results on the LCD projector due to color decay from the organic compound in the LCD panels (chips). You could have a severely yellowish image from color decay and still meet the CLO ANSI lumen specifications.

This is the big advantage in my opinion for DLP projectors over LCD projectors. Same color from one year to the next regardless of how many times the lamp is changed.
chris Posted May 5, 2014 9:40 AM PST
Many LCD chips today are inorganic, so I don't know how relevant that is today, and I would also figure a true head-to-head test would be difficult as there are so many other factors, whether testing with brand new units, or older units with thousands of hours on them. For instance, it's been my opinion that most entry-level projector that use lesser DLP chips don't look as good (overall) as their LCD counter parts. Consider the Epson 5030UB and/or 6030UB and the Panasonic PT-AE8000U when compared to anything buy BenQ, Optoma, Vivitek, etc. for the same/ similar money. There are longevity issues, maintenance issues, and other set-up parameters to discuss as well, but in general... simply talking to the light output and it's perception in the image quality of the projector... I think LCD wins every time over DLP. When you start climbing the food chain a bit and start considering "better" LCOS units and DLP units, this changes some, but the bulk of the projectors purchased for home use are well under $5k and for those buyers it's hard to surpass the overall image quality of the LCD projectors offered in this pricing category.
Lee Posted May 12, 2014 12:26 PM PST
Chris, I don't know where you are getting your information. First - no one knows exactly which LCD models have inorganic LCD chips, and many brands don't disclose this- so why you would you say that the #1 subject on the Projector Central forum is not relevant to a consumer anymore? Secondly, on what basis are you making the claim that DLP projectors use "lower quality parts"? Based on the Amazon top seller list today, the top four 1080p projectors are DLP, before a LCD projector is ranked. I have a hard time believing that they have "lower quality parts", when there are also sub $1k 1080p LCD projectors as well. I think the article shows that the image quality of a projector is not defined by a single spec, and this is where reviews by both this site and consumers really help consumer's decision making on a projector - regardless of the technology.
Michael Miller Posted Jun 4, 2014 11:25 PM PST
And if you run the test after 3 years your DLP projector will have lovely sparkles all over it from the failed DMD chip.
Joshua Posted Jun 11, 2014 10:36 AM PST
Hey all, I just purchased a 730hd Epson with lCD and this is my first purchase of a projector. I'm wondering if I made a mistake for 2 reasons. First, it's 720p and does not do 3d. Second, will the LCD remain viable over time and product the best image? I'm considering swapping it out for Optoma DH1011 or a ViewSonic PJD7820HD.

Question is, will the DLP look as good as the 3 LCD from the Epson?
Bob Posted Jun 23, 2014 2:03 PM PST
Projection technology has been around for a long time now - both DLP and LCD projected images look great. I feel confident recommending both technologies without hesitation. If a customer is not happy with an image after adjustment, most projector manufacturers will assist customers or swap out the projectors.

The only companies really disputing this fact are the manufacturers of LCD panels and DLP chips. The DLP camp has not run a negative campaign for years, but this year at INFOCOMM, an LCD manufacturer had a side by side demo calling out the benefits of CLO. In the demo, both projectors were displaying images in high bright mode and both images were over-driven, and were not really acceptable, but the DLP image looked worse. When the settings were changed to a more user friendly mode, both projectors looked pretty good.

It seems that some projector manufacturers are determined to shift market share from one display panel technology to another. This might be a great idea in a growing market but the PJ market is relatively flat.

These component manufacturers should focus on finding ways to grow the PJ segment instead of shifting share.

At INFOCOMM there were several amazing LARGE screen displays that called out the benefits of projectors, including edge blending, LED, laser, interactivity, and digital signage. The industry needs to focus on the big picture and not components.
Bob Posted Jun 23, 2014 2:20 PM PST
All the projectors you are considering have their merits. First consider your requirements - how much ambient light is in the room, what type of screen are you using, what size screen and what will you be watching?

For a dedicated home theater you don't really need 3000+ lumens unless you will be watching a lot of 3D content....most people don't after watching a few movies in 3D.

I do recommend moving up to 1080p, if within your budget. You may find that 3000+ lumens is too bright for a dedicated home theater. Also, look for a projector designed for home theater - for instance a color wheel with a white segment may focus on brightness instead of saturated colors.
Bob Posted Jun 23, 2014 2:21 PM PST
This could happen but is extremely rare.
AV_Integrated Posted Jul 16, 2014 8:34 AM PST
It would be excellent to have you guys run through this test again using three of the most popular 'cheap' home theater models. The Optoma HD25-LV (or similar) with a 2x, 6 segment (RGBCWG or RGBCYM) color wheel, the BenQ W1070 with the 6x RGB/RGB color wheel, and a Epson 2030 or 3020 model.

There's a lot of talk out there about the brightness specification of the Optoma models, but it seems that real world viewing won't actually allow it to deliver the same brightness, post calibration, or with normal use, that the W1070 is capable of delivering, and the cheaper Epson model may not be able to match either in regards to contrast, but may do very well elsewhere.

While dedicated home theater users may typically not care, for those with family room environments, non-dedicated spaces, or larger screens, the importance of getting post-calibration real world color brightness figures really seems like an important specification that can't be found anywhere else.

Similarly, it would be great for all reviews to include color wheel information and color wheel speed at differing input frequencies as most projectors don't have the same color wheel speed for 60hz content as they do for 24hz content.

Thanks for a very interesting read, even if I am getting to it a bit late!
Home Theater Install Posted Jul 24, 2014 8:59 AM PST
AV_Integrated: I agree with your comment regarding RGB/RGB color wheel providing most accurate color display. I believe the Optoma HD30B is the model you need to compare to the W1070.
David Posted Sep 5, 2014 12:48 PM PST
One of the most comprehensive, well explained articles explaining (and showing) the differences in DLP and LCD color and brightness quality. Seeing the images side by side really helps...
Michael Posted Nov 11, 2014 2:47 AM PST
An excellent article and fair test which should help buyers understand what questions to ask.

I think a second article revolving around the various light sources would be useful as we now have the traditional hot lamp, LED, LED/Laser Hybrid and Laser engines; and again the technology is generally being sold on brightness (lumens).

There will undoubtably be differences particularly in colour fidelity and full on/off contrast between the light sources; and although LED for instance has a low lumens count this may not be the important factor for some buyers, such as artists using pico projector for installations. We know that the cinema projectors are governed by standards and there’s no reason why we shouldn’t be provided expectations of the colour fidelity over the lifetime of the light source.

Possibly a more difficult test to create but I think it would be very useful for the art, education, business and home cinema buyers to understand which light source works best for their application.
CAW Posted Apr 22, 2015 4:44 PM PST
This article was the most informative I've read so far. Being not very experienced in projector purchasing, the information given here answered so many questions. Thus helping me to make a very informed projector purchase. The actual visual tests were great. Thank you very much.
devx Posted Jul 8, 2015 4:30 AM PST
what's the difference between LED brightness and DLP brightness? I see a lot of LED projectors with 400-500 lumens of brightness, does this mean these projectors will be 5 times less bright than an average DLP projector with 2500 lumens. For example, the Optoma ML550 (LED-500lumens) and Optoma HD141x (DLP-3000 lumens), what will be the difference in terms of brightness.?
Chief Master Posted Dec 16, 2015 12:59 PM PST
This article was very informative. Thank you Evan
Michael Lang Posted Mar 28, 2016 11:07 AM PST
Evan,

One of the best articles I have ever read comparing DLP with LCD. Where ddi you get the images
Gene Houghton Posted Apr 9, 2016 7:01 AM PST
Great article. I would be interested to see how the comparison would be on a .6 screen.
Rammy Leeds Posted Jun 12, 2016 11:13 AM PST
I see a lot of pico projectors now saying 1800 lcd lumens. But the manufacturer wrote that this is equivalent to 250 ANSI lumens. Why do they do this? Why not stick to ANSI as a measure of brightness? More importantly is there a conversion chart so I can see roughly what these LCD quoted lumen projectors are in terms of comparison to those that write their lumens in ANSI format. Yes there are other pico projectors that just state lumens in ANSI format only. How can I compare these pico projectors with each other if they are using different units of measurements for their brightness. I understand some pico may be using different technology eg lcd vs a halogen lamp.

I've currently an optima pico projector at 100 ANSI lumen. I bought a 3000 LCD lumens pico projector by another manufacturer but how will it roughly compare to the optoma one?

Please help someone
Ted Nunn Posted Oct 5, 2016 12:56 PM PST
Great article. It is nice to see a lack of bias. To fully cover the situation, I would love to see you cover the home theater/entertainment projector technology issue. These two camps can BOTH produce excellent projectors, but the potential customer must endure the fighting and misinformation that is put forward. The same issues that you addressed, along with the Rainbow Effect, are bantered back and forth by both manufacturers and consumers. It would be nice for an EXPERT to share the TRUTH that lies somewhere in the middle of it all. It is no wonder that people back off concerning projectors, when this is a very exciting way to watch entertainment. There should be less mystery and more solid answers for inquiring customers.
Michael Posted Nov 18, 2018 5:31 AM PST
As a DLP user for over 15 years, I believed this was a much superior technology compared to LCD.

Recently I managed to purchase a 2nd hand fairly cheap Epson 3LCD projector. I based the purchase solely on price as I felt it was a good deal.

The result is that i am utterly gobsmacked at the image it produced. It is extremely vivid and TV like compared to a DLP. I cannot believe what I have been missing all these years.

This debate truly is significant and I commend ProjectorCentral discussing this, although it has been some years.

CLO is real. DLP ANSI brightness does not compare to 3LCD ANSI brightness. Comparing the tw now, the DLP is at least half as bright as the LCD even though they both have ANSI rating.

I don't think I will ever go back to DLP after seeing the difference with my own eyes. I urge you all to do your research on both technologies when buying a projector as there are major differences for projectors in the same price range.
Josefine Walther Posted Jan 20, 2020 8:37 AM PST
With projectors that use the 1-chip DLP system, colors are projected sequentially (in other words, your eye is not receiving color information continuously), which can result in much lower color light output, in relation to the white light output. To compensate for this, 1-chip DLP projectors often add a white segment to the color wheel in order to boost While Brightness, but the fact remains that the degree of Color Brightness is less than the White Brightness. This difference is usually not stated by the manufacturer in their projector specifications. What you most often see is a single Lumens output specification, rather can one that lists two lumens specifications, one for WLO (White Light Output) and one for CLO (Color Light Output), which provides a more accurate profile of how much Color Brightness you can perceive.
MARK Posted Jul 5, 2020 1:21 PM PST
Read every word and comment. This article had great information, notes and picture examples which helped all of us see the same results. I learned a lot in this article and wanted to Thank You for taking the time to compile all this information. Keep Up The Great Subjects and Articles. THANKS !

Post a comment

 
Enter the numbers as they appear to the left