In part II of this series, we explored sharpness differences between LCOS and DLP technologies.  In that article we also speculated that there appeared to be some unique changes to the JVC RS35 that improved sharpness and perhaps reduced chromatic artifacts as well. In this article we take a closer examination of the apparent differences that we saw between the RS20 and the RS35.

If you will recall, we noted that the RS20 edge pixels were diminished both in light output and geometry, and we suggested that this may be due to field fringing or other panel effects rather than lens effects.  The RS35 on the other hand showed significant improvements to these edge pixels, but with one caveat – a transition dependence where black to white edge pixels (when moving from left to right) seemed to show the improvements while the white to black edge pixels (also left to right) seemed unchanged from what we saw with the RS20.  We also saw chromatic artifacts on the RS20 edge pixels that were not present in the RS35, and we speculated that perhaps this was also related to the diminished edge pixels rather than the conventionally accepted causes of convergence and chromatic aberration.

As we noted at the beginning of part II, there are always variances among units and so it is always best to take such findings with a grain of salt.  What adds credence to any speculation however, is repeated confirmation of the same behavior with other units.  In this thread at AVS Forum, some members have commented on the same behavior on their RS35s and one member, Randal Feemster (Cam Man) graciously forwarded this high quality photo closeup of another RS35 owned by Erik Todoroff (HogPilot).

Photo 1 - RS35 screenshot

In it we can see the same transition effects that we discussed earlier, black to white transitions (from left to right) are well defined, but white to black (also moving from left to right) transitions show a large loss of pixel area.  We also see a lack of the color artifacts from left to right that we saw with the RS20.  So this photo provides additional support of the behavior that we noted in part II.  We also hope to perform detailed MTF measurements on another RS35 soon to help further reinforce or dispel the findings that we have developed so far.

This photo also shows transition effects happening as one goes from bottom to top, but with red color effects occurring on the top most edge pixels that are also diminished the most.  This is similar to what we saw with vertical line pairs with the RS20 in part II and as we noted in part II, a quick look at the RS35 line scans with white horizontal line pairs didn’t seem to show the same improvements that we were seeing with vertical line pairs.  Incidentally, the photo above also appears to show blue misconvergence (seen at the bottom of the text) which is not related to the edge color effects that we noted in Part II.

This article will focus on the left to right transition improvements that we’ve seen on the RS35.  We will do a similar treatment of bottom to top transitions as we’ve done here and if anything noteworthy is found, we’ll post a follow up article.

RS20 Examined

First let’s take a look at the RS20.  If you recall, the photo below is from a vertical, 5 pixel wide line pair:

As we can see from this color photo, the right pixel in the 5 pixel line pair is lower in intensity and thinner.  Chromatic artifacts exist on both edge pixels.  Conventional wisdom would say that this is probably due to convergence errors from a misaligment of the three panels or chromatic aberrations in the lens.  Let’s take a more detailed look though and see if those issues hold or if there is something else going on.

By taking separate red, green, and blue images of this pattern with a stationary line camera, we can see the performance of each individual panel.  This was done below with white (all three panels) at the top and the other three colors aligned vertically below it so that we can easily check for convergence errors.  Convergence errors will show up as a misalignment between the group of “on” pixels.  The on pixels in one line scan image will be shifted either to the left or right and will not line up with the other panels if misconvergence is present.  Using this method we can determine the degree of misalignment very precisely and thereby get some clues about the chromatic effects that we were seeing with the RS20.   We can also use this method to dig deeper into the diminished pixels that we saw in Part II with white.  We can for example see if the diminished edge pixels are due to one specific panel or multiple panels.  The 5 pixel line scans are shown below.

RS20 - Aligned Panel Line Scans (click to enlarge)

From the line scans above we can see some surprising results.  The biggest surprise is that the symmetric diminishing of the edge pixels that we saw with white is not actually symmetric when we examine each color individually.  The red panel shows a problem with dark to light transitions (left to right) but is okay with light to dark transitions (also from left to right) and the situation is reversed for the green panel.  We can see nearly spot on convergence between red and green, although blue is slightly misconverged.  We can also see that the blue panel is generally not very well delineated, perhaps due to less than ideal blue focus (note: in Table I of Part II we showed that Blue MTF on the RS20 was reduced about 20% compared to the Red and Green panels).

Next lets examine the line scan images from a color perspective.  The center points of each pixel seem to be roughly 175 camera pixels apart which would put the center of the left most edge pixel for both red and green at about line 1475 (based off of the camera pixel location).  Looking vertically along this location at each line scan, it’s clear that the left edge pixel is diminished more in red than any other color.  If we look at the color photo we see a cyan coloration.  With a video RGB color system, any time that we reduce the amount of red from a white color mix, we will end up with green and blue which is the secondary color Cyan.  Which just happens to match the same color of the chromatic artifact we see from the RS20 color photo.  The lack of misconvergence or bleeding of other colors due to chromatic aberration is a clear indication that the color artifact that we are seeing in the photo above is primarily caused by the diminution of the red edge pixel.

Next lets take a look at the right edge pixel from the photo.  This edge pixel follows the bright to dark transition (again left to right).  Looking at the line scans we see that red is comparatively well maintained but green and blue are diminished now.  Not surprisingly we will get a lack of green and blue in the RGB mix which gives us red.  Looking at the photo, that is exactly the color that we see with this pixel.  Since Green is the dominant color in Rec. 709 and Rec. 601, it’s also not surprising that we see a loss of brightness as well.

What is unclear is why the left Red pixel is diminished while the right Red pixel seems to be less affected.  Similarly, it’s also unclear why the left green pixel is not as affected as much as the right edge pixel.  Unfortunately this lack of symmetry complicates the speculation that we had earlier that field fringing was the primary cause of the diminished pixels as it seems as though this would cause both pixels to be diminished equally.

In summary, we have seen good indications that the color effects that we are seeing on this particular, well converged, RS20 are directly related to the diminished edge pixels that we discussed in Part II rather than the commonly accepted notions of misconvergence or lens chromatic aberration.  Unfortunately we also have new questions about what causes the diminished edge pixels as it doesn’t seem to be symmetric within a given color and the left to right transition is different between Red and Green panels.  Something new to ponder…

RS35 Examined

For now let’s put our pondering about the RS20 on hold and examine the RS35 using the same techniques described above.  The photo that we saw in Part II is included below.  Note the lack of color artifacts, although as we noted the right edge pixel is diminished in both size and light intensity.

Photo 7 - 5 white pixels (RS35)

Line scans from each individual panel were captured as we did with the RS20 and the results are shown below:

RS35 - Aligned Panel Scans (Click to Enlarge)

Here we see that the Red and Green panel alignment is also excellent and no convergence errors can be seen.  As we saw with the RS20 there is a slight amount of blue misconvergence, but not a significant problem.  Next let’s take a look at the Red panel.  With the RS20 we saw that the left Red pixel was diminished which caused the Cyan tint that we noted with the left pixel.  In Part II, we openly speculated that the RS35 might use some sort of compensation on some transition pixels which mitigates the fringing effects that create the diminished edge pixels.  Looking at the Red Panel we see something very interesting.  The first pixel (the one circled) is boosted and is noticeably brighter than the other pixels in the group by roughly 10%.   This despite the fact that it’s an edge pixel where we expect some at least some degree of diminished light intensity.  This boosting of light intensity is something that we hadn’t seen before with any other line scans on the RS20.  In fact all of the edge pixels on all of the RS20 showed some degree of diminished light intensity.   So this seems to be a strong indication that some sort of boost/compensation is being applied in the panel as it’s highly unlikely that lens or optical effects could cause this sort of spike.  The net affects from this red spike are two fold:  1) The pixel intensity and geometry are preserved  and 2) There is no longer a shortage of Red in the color spectrum for this edge pixel so we no longer see chromatic problems.  Looking closely at these line scans compared to the RS20 we see similar but more subtle boosting/compensation going on with the first (left most pixel in the line pair) pixel in the Green and Blue Panels as well.

Incidentally, we also see much better definition with the Blue panel which confirms the MTF measurements that we took in Part II where we noted in Table 1 that the Blue MTF of the RS35 was improved by 20% over the RS20.

Now lets examine the right edge pixels on the RS35.  Looking vertically along the line scans at this pixel position we see that all of the edge pixels show diminished light output and geometry which explains why this pixel shows up as being roughly 1/2 the size of the other pixels.  On the RS20 we noted that only the right-most Red pixel in the group was fully formed which led to a red tint with this pixel.  Curiously, this red pixel is diminished on the RS35 and now all 3 panels are back in color balance so we no longer see the same color artifacts that we saw on the right edge of the RS20 even though the pixel is still not as bright and fully formed as it should be.   Overall, the diminished right pixel seems to be a “win” because of the lack of color artifacting, so perhaps this was an intentional change implemented by the JVC engineers.  Unfortunately there is no way to know the “whys” although we do have a clear indication of improvements to the RS35 panel over the RS20, improvements that are clearly transition dependent.

Summary

Putting things together we have more information and perhaps a few answers, although we have also raised a few more questions in the process.  Our speculation that the RS35 is using some form of boost/compensation in order to deal with fringing effects seems to have been given a lot of credence.  The fact that the RS35 Red left edge pixel, after a “off” to “on” transition, is not diminished as it was with the RS20 and in fact is noticeably brighter than the other pixels in the group seems to be a clear indication of that.  We have also seen the other edge pixels in a off to on transition maintain their light intensity better than we saw with the RS20.

Curiously, the rightmost red pixel on the RS35 has dropped a notch in brightness and delineation, although, overall this seems like a win as it keeps the colors balanced and prevents the distracting color artifacts that we saw with the RS20.   We’ve also shown that what looked like convergence or chromatic aberration errors on the R20 was in fact related to the edge pixel issues that we discussed in Part II.  These issues are likely panel related rather than lens related and they seem to have been corrected with the RS35, at least in the vertical direction.  On the downside, we have new questions about what causes the fringing effects on the edge pixels, as the RS20 red and green panels do not seem to be symmetric in this respect.

The improvements on the RS35 edge pixels that we have seen with the left edge pixels directly result in improved sharpness and a lessening of color artifacts and likely explain at least some of the 9% higher MTF results that we measured with the RS35 over the RS20 with vertical line pairs.  As we have seen, these improvements only seem to apply to the left edge pixels that follow a off to on transition.  It’s unknown why similar improvements have not been made to the right edge pixels, but it gives us something to look forward to with future products.  As we saw in Part II, LCOS is capable of high inter-pixel contrast and if the edge effects that we’ve discussed here can be further improved,  we should see continued sharpness improvements in LCOS displays.

November 30, 2009

A few months ago in this article, we took a close look at the sharpness differences between DLP and LCOS video projectors and in addition, we examined the differences between a projector that used a .65″ DC3 DMD (the Optoma 8200) and another that used the .95″ DC3 DMD (the Planar 8150). That article provided a good basis for general sharpness comparisons, but it stopped short of assigning an objective metric in order to fully describe the sharpness differences between these projectors. It was apparent from that article that the Planar 8150 was the sharpest projector of the 3, but readers were left to wonder how much sharper, 10%, 20% 80%? If the sharpness differences could be expressed as a single objective number then the comparisons could be made much easily. Fortunately, there is such a metric called MTF (modulation transfer function) which is well known in optical design and it is often used and discussed in camera and lens comparisons.

In this article, we remedy the shortcomings of the previous article and we apply considerable effort at measuring the MTF performance of a digital projector so that we can provide an objective, numeric, sharpness metric for this and subsequent articles. By measuring MTF, we are providing the first independent, detailed look at the MTF performance of a digital projector.  In doing so, we have found a wealth of information that can be used for comparisons between projectors and display technologies. Most importantly, we have found that it can also provide valuable information about how the sharpness of a given projector can be affected by real-world constraints and modes such as throw distances, lens shift, iris settings, lamp output, etc. Information that owners can directly apply towards maximizing the image and performance of their projector.

MTF (Modulation Transfer Function) Overview

MTF is the definitive tool used to measure and specify the sharpness of optical systems.  Unfortunately, providing a detailed explanation of MTF theory is well beyond the scope of this article and it’s something that is well covered in text books and articles on the Internet (as an example here is a good MTF introduction). Fortunately, when boiled down to its essence, MTF is actually a very simple concept that describes how well an optical system can resolve black and white images. Typically a pair of white lines on a black background (or black lines on a white background) are used and the thickness and distance of the lines are varied, which is used to glean information about the response or optical performance of the system as a function of frequency. The idea being that thick lines (low frequency) are much easier for an optical system to resolve than thin (high frequency) lines. The spacing of the line pairs is referred to as the “spatial frequency” and is usually expressed as line pairs per millimeter (lp/mm). After being passed through the optical system, the contrast of a specific group of line pairs (with a specific spatial frequency) is measured and the MTF is determined using this formula:

MTF = (Imax – Imin) / (Imax + Imin) Where Imin and Imax refer to the Intensity of light for the black (Imin) and white (Imax) portions of the line pairs – as measured after passing through the optical system.

As the spacing and thickness of the line pairs is reduced, the resultant MTF changes in response to each of these frequencies and a plot of MTF vs spatial frequency is created. When the contrast differences are high, the MTF will approach 1 (100%) and when the contrast differences are low the MTF will be low. At higher frequencies, the MTF is reduced and there becomes a point where the line pairs at this high spatial frequency can no longer be resolved and this is referred to as the limiting frequency or the Nyquist Limit. With digital 1080p projectors, the highest frequency that can be represented is single pixel on and off line pairs, and these 960 vertical line pairs (or 540 horizontal line pairs) are the limiting frequency for digital projectors. In theory it’s possible for a digital projector to have a nyquist limit below the 1080p on/off pattern but in actual practice all digital projectors seem to be able to easily resolve this test pattern. What is of most interest then is what is the measured MTF at this Nyquist Limit and if it is low what is the shape of the MTF curve at other frequencies.

In practice, two different methods are usually used to determine a MTF plot, one uses a slit or knife edge measurement to collect line spread data which can then be used in a Fourier Transform to calculate the MTF at all frequencies. This method has the benefit of being fast, but the math is complex (although tools exist for quickly performing the calculations – the FFT) and the results are relatively general because they rely on extrapolations of calculations rather than precise measurements at each specific spatial frequency. The other method that is used involves a series of line pair test patterns to measure the MTF directly and precisely for each frequency, but the downside to this method is that it is more laborious and requires many measurements and a series of test patterns. Being a glutton for punishment and after investing in a relatively expensive CCD line camera, I decided to put it to good use and follow this latter approach for this article, although subsequent articles may use the other approach (and possibly compare the results of both approaches).

It should also be mentioned that the white and black line pairs are traditionally created as a sinusoidal test pattern where the intensity of white gradually changes from black to white rather than as a step function using one level of full black and another level of full white. Using sinusoidal patterns fits well with the mathematics that are used in the Fourier calculations. Unfortunately, with the advent of digital displays, sinusoidal patterns are impossible to render at high frequencies because there aren’t enough pixels in the line pair to provide the smooth transitions between the black and white peaks. This is a problem faced by anyone applying MTF to a digital device where the pixels themselves are used to provide the stimulus. This means that the Fourier calculations used to determine MTF can break down somewhat, although in practice using on and off test patterns does not usually change the results substantially. Even still, this is another reason why we chose to measure MTF directly. It should also be noted that at larger frequencies where enough pixel resolution was available to create a smooth transition between peaks, both types of test patterns were used and the results obtained were almost identical.

MTF examples – CRT and JVC QX-1 projector

So lets take a look at an example of an MTF plot. Below is a hypothetical representation of the MTF plot of a CRT projector, the vertical line represents the maximum spatial frequency for 1920×1080 HD content and as we can see the MTF of our hypothetical CRT is 30% at this limit. Because our hypothetical CRT is an analog device it may be possible to drive it to higher frequencies than 1080p and as we do so we can see that the MTF drops off rapidly before reaching a limiting frequency (the nyquist limit) which is typically quoted as being when the MTF reaches 5%.

Hypothetical Example - CRT MTF

Next lets take a look at an actual test case of a digital projector (JVC QX-1) which was published in this SPIE white paper. According to the article the plot below was obtained from a high resolution line camera using a camera and techniques very similar to those used in this and subsequent articles.

Actual MTF Data - JVC QX1 Projector

Unlike our hypothetical CRT example, here we have a digital projector with a fixed pixel count (2048×1536 in this case). We can see that the MTF is much higher than the previous CRT example and this applies to all frequencies up to the nyquist limit (the vertical line). The key difference in the case of the QX-1 projector compared to the CRT is the QX-1 has a fixed pixel count which fundamentally limits the spatial frequency (nyquist limit) to no more than the one pixel on and one pixel off line pair pattern which is indicated by the vertical line. With this particular projector, the pixel count is higher than 1920×1080 so the frequency limit will in turn be higher. This projector is therefore capable of reaching higher spatial frequencies than those possible with 1080p source content. The white paper mentions that the MTF at the Nyquist Limit is greater than 75%, but the reader should realize that with slightly lower spatial frequencies (i.e. those used in 1920×1080 content), the MTF will be even higher than that shown by the nyquist limit. Incidentally, the MTF plot shown here continues beyond the Nyquist Limit, but the reader should realize that this represents theoretical optical performance derived from the fourier calculation and this performance will never be realized unless the pixel count of the microchip is increased or the size of the microchip is reduced. Fundamentally the MTF of this particular projector with the given microchip ends at the Nyquist Limit and all of our MTF plots will also end at this limit.

What MTF doesn’t tell us

As we will see, we have taken all sorts of MTF measurements – measurement in the horizontal and vertical planes, in individual RGB colors, as a function of light intensity and with a variety of iris settings, projector throws, lens shifts and projector modes. Despite all of these measurements there are still many measurements that we haven’t performed. For example we are not measuring the sharpness across the field and in particular at the edges. Without these measurements we can not determine coma and other optical effects that may be present. It isn’t the goal of this article to completely characterize the optical performance of projectors and fortunately, we don’t have to in order to glean important aspects of the sharpness performance of a projector. Similarly, it is well known that optical quality may vary across units and it is not our intention to make the claim that these results will hold for all projectors of a given brand or model. In this respect, these measurements are no different from any other metric such as color accuracy or contrast performance that are subject to the same unit to unit vagueries and variances and yet these other metrics are still widely reported in magazines and projector reviews.

It is also worth noting that what is being measured is the overall system MTF that includes the chip and optical system. The individual components including the projector lens will have higher MTF if measured by themselves.

Edit: It should also be obvious to the reader that sharpness is only one of many attributes that contribute to good image quality. Contrast, greyscale reproduction, brightness, color accuracy, feature sets, throw, price, etc. will all play a role in the purchase of a projector and sharpness is only one of those factors. Ultimately, the relative merit of sharpness over any other factor is an individual decision and an article like this will only help to provide insight into the sharpness factor alone.

Enter The Samsung SP-A900B JKP Special Edition Projector

Now that we’ve presented the basics on what MTF is and how to interpret the charts, lets move on to the meat of this article which is a detailed examination of the MTF of the Samsung SP-A900B projector. This projector was chosen as the first projector for this MTF series specifically because it has outstanding full field optics and is a 1-chip design with no convergence issues and little chromatic aberration. As such it sets the bar very high and gives us an idea what top-tier projectors are capable of delivering as far as sharpness. The projector used was also not a cherry picked review sample, but instead purchased and shipped at random, so hopefully the results should be close to what a buyer might receive, unit to unit variances not withstanding.

First let’s take a look at what the pixel data looks like from a high resolution CCD line camera. The image below is data taken from the Samsung projector for a single pixel on and single pixel off pattern. As we can this this provides very precise details of the light intensity for both the bright and dark pixels. As can be see the contrast differences between light and dark pixels is only around 15:1 even with this very sharp projector. The camera is capable of resolving inter-pixel contrast differences of well over 300:1, so the results can be very precise and not limited by the accuracy of the equipment. The reader should note that the x-axis is labeled “pixels” and this refers to the CCD camera pixels rather than the projector pixels. Here we have oversampled and imaged 18 projector pixels (9 on and 9 off) with 3000 camera pixels in order to achieve precise results. It should also be noted that this line scan is imaged directly off of the projector so that camera optics or screen effects are not an issue.

One Pixel on, One Pixel off Linescan - Samsung A900B (Click to Enlarge)

So taking the line scan contrast information above, we can calculate the MTF for this frequency (40 lp/mm) and do it similarly for other frequencies using on/off test patterns of various pixel widths. For the chart below we used 7 different test patterns that varied the line width and spacing from one to 7 pixels. If we plot this data we get the MTF curve for the Samsung, which is shown below. As can be seen below the Samsung has outstanding sharpness across all frequencies and with single pixel on/off patterns (the nyquist limit) it achieves a whooping 88% MTF.

Samsung MTF Plot - Vertical Line Pairs

The spatial frequency in the plot above was determined by dividing the line pairs by the chip size (.95″) which is traditionally how projector MTF data is presented. Since chip sizes vary and we would like to compare the results from many chips, the data was recharted using spatial frequencies determined by the equivalent of an 8′x4.5′ screen. This is the same way that the QX-1 data was presented in the JVC white paper mentioned above with the exception that they utilized the equivalent of a 12′ wide screen rather than an 8′ wide screen. Despite the common practice of expressing spatial frequency based on chip size, it is very convenient and desirable to plot the data based on screen size because to the user this is all that matters – in other words, what is the sharpness on a given screen independent of chip size and other factors.

At this point however, the reader should realize that larger chips like the .95″ DC4 DMD have an inherent resolution advantage over a smaller chip like the .65″ DC3 DMD if the pixel count is the same. This is because in real terms, the spatial frequency in line pairs per millimeter of the smaller chip will be higher. Another way to look at it – if the pixel pitch remained constant, but the smaller chip were increased in size to equal the larger chip, it would have a higher pixel count and the nyquist limit would be higher. This is an important consideration that the reader should keep in mind when drawing conclusions about technology differences (LCOS vs DLP for example) based on MTF numbers alone when those chips have dissimilar sizes (.95″ for DLP for example, .7″ for DILA, .6″ for SXRD). In other words, the reasons for a lower MTF may have more to do with the chip size itself (or something else like the optics) than anything to do with differences in the technology.

The plot above shows MTF with spatial frequency represented by line pairs on the equivalent of an 8′ wide screen. Note that the shape of the curve has remained unchanged as has the MTF at the nyquist limit, but now we have a uniform way of comparing MTF across chips that vary in size, which is something that we will do with subsequent articles.

Since the MTF curve above ended up being so high and relatively uninteresting across the other frequencies, all of the subsequent Samsung measurements will all be done at the Nyquist limit using single on/off pixel patterns.

MTF Horizontal vs Vertical Line Pairs

Now lets take a look at what happens if we use horizontal vs vertical on/off line pairs. This is a useful measurement for 3-panel projectors as the convergence in the horizontal and vertical directions may not be the same.

• Horizontal MTF: 88%

• Vertical MTF: 88%

In other words the MTF was identical in both directions. Incidentally it’s worth pointing out that the spatial frequency is the same in both directions even though there are more vertical lines than horizontal line pairs and this is because the density of lines is the same in both directions.

MTF of Individual Colors – R,G, B and White

Now let’s measure the MTF of each individual color. This will be an interesting test in future articles on 3-chip projectors because it can tell us how much the convergence of the 3 panels may be holding back the MTF of a projector. On a 1-chip (and also a 3-chip display), it also tells us which color has the best focus and if the optical system is limited because of the inability to focus one or more colors.

• Red MTF: 83%

• Blue MTF: 91%

• Green MTF: 93%.

Since green is the dominant color in both Rec. 709 and 601 color standards, it is the critical color to optimize and the Samsung does an outstanding job with both this color and also blue. Red is the second most dominant color and here we see a drop-off in performance, which if corrected without harming the other colors, could allow the Samsung to hit MTF values in the low 90s.

It should also be mentioned that these measurements were obtained after optimizing the focus for white. While it’s possible to refocus when measuring each individual color which would have provided higher results for each color, this has little value because it will have harmed the focus for white and it is the best MTF for all colors displayed simultaneously that we are interested in. It should be noted however, that if the display suffers from color bleeding from misconvergence or chromatic aberration then it might be beneficial to maximize the focus (i.e. minimize the pixel spot size) of the problem color at the expense of less overall sharpness, but this situation was not the case with the Samsung and was not necessary.

Other than providing MTF for each individual color, there were no additional measurements for chromatic aberration. Measurements of pixel spot size is something else that can readily be performed with the equipment and techniques used in this article and perhaps we will include these measurements in subsequent articles.

MTF vs Lamp Setting, projector modes and iris settings.

MTF was measured for both the bright and theater modes of the Samsung and found to be:

• Bright mode MTF: 84%

• Theater mode MTF: 88%

Only a small loss in sharpness was found between lamp modes. Next let’s put the iris into manual mode and measure it at the smallest aperture position (iris =100).

• MTF with Iris at largest aperture (Manual Iris mode): 88%

• MTF with Iris at smallest aperture (Manual Iris mode): 76%

The lens was refocused in this iris setting with the thought that perhaps the iris movement caused the lens to slightly defocus, but all focus attempts did not improve on this MTF. Owners should be aware that using the projector in this mode can cause a reduction in sharpness. This is particularly important if they have reduced the iris aperture so as to throttle back the light output when attempting to compare the performance of this projector against other projectors that have less light output.

MTF vs White Intensity

So far all of the line pairs utilized 100% peak whites. Out of curiousity, let’s see if varying the intensity of white (gray) will significantly change the MTF.

• MTF – 100% white: 88%

• MTF – 60% white: 87%

• MTF – 40% white: 86%

In short negligible differences in MTF are found when ramping down the white peaks from 100% to 40%. The reader might wonder why the MTF is unchanged since after all, MTF is primarily a contrast measurement and we’ve just drastically changed the brightness of the white pixels. The answer of course is that the black values representing the off pixels has changed also and by about the same amount which leaves the MTF unchanged.

MTF vs Throw Distance

MTF was also measured at maximum and minimum throw distances. Initially the MTF was seen to change significantly, but this is because the focus shifted. After refocusing the difference between longest and shortest throws was unchanged.

• MTF – Longest throw: 88%

• MTF – Shortest throw: 88%

MTF vs Lens Shift

All of the measurements performed so far unless otherwise noted were made close to the center of the image and with a neutral vertical lens shift (the Samsung has no horizontal lens shift adjustment). With the Samsung optical design, using a neutral lens shift causes the exit light of the projected image to remain centered in the lens and it falls on an area that is much smaller than the diameter of the lens. When lens shift is added, the exit light is shifted from center and falls closer to the edge of the lens which is why a large diameter lens is needed. Adjusting the lens shift from maximum top to maximum bottom lens shift effectively sweeps the projected image from the top of the lens through center and then to the bottom of the lens. In measuring MTF vs lens shift, the camera was at all times positioned at the center of the image while the image itself was lowered. MTF vs lens shift was measured in only one direction (down) and this would be similar in orientation to someone mounting the projector high on a shelf and using lens shift to lower the image onto a screen.

It’s been widely reported and generally assumed that using lens shift degrades the sharpness of an image and is generally something to be avoided. With some projectors it’s been reported that using some lens shift can raise ANSI contrast somewhat because reflected light from the lens elements will be off-axis and will not project directly back into the optical path. Before doing doing this measurement the assumption was that this measurement would show a degradation in sharpness and the tradeoff/downside of using lens shift. What was found was very surprising….

• MTF – Centered Lens Shift: 88%

• MTF – 1/2 Maximum Lens Shift: 89%

• MTF – Maximum lens shift: 91%

These results run counter to conventional wisdom and were very surprising. Increasing lens shift actually made slight improvements in sharpness to the center of the image on this particular projector! The reasons for this are unclear and may be related to the same reasons that ANSI contrast can be improved with lens shift or perhaps it was simply because this particular lens just happened to have a sweeter focus at the edge of the lens than at the center, or perhaps it’s a combination of both factors.

Edit: It was pointed out to me by a Samsung owner that the extreme edge of the field will distort when maximum lens shift is applied even though the center (the area being measured) may still have a sharp focus. As mentioned earlier in the article, only a few pixels are being measured and there can be deviations across the full field so please keep this in mind with all of these numbers.

MTF and Intra-Image Contrast

There has been discussion about the perceptions of both intra-image contrast and color with high MTF. Unfortunately such discussions go beyond the scope of this article. However, there were some interesting discoveries relating to intra-image contrast that came out of this detailed look at MTF that are worth discussing.

Variance of White

The first is that it’s been an open question with many display types whether the light intensity of white pixels vary significantly as the area of a white region is made smaller and smaller. With reflective displays like LCOS and DLP, the large white regions in a test pattern such as the ANSI contrast pattern are pretty much unchanged throughout the rectangle (excepting relatively issues such as optical rolloff or shading uniformity). But what happens when the white regions are made very small? In my own experiments with low APL intra-image contrast, I’ve found that the white regions maintain their white points very well until the white regions become too small to be measured reliably with a light meter. If we keep making the rectangles smaller, what happens to the white level once we get down to clusters of a few pixels and even the individual pixel level? By employing the techniques used in this article it’s possible to measure the white level for a white field all the way down to a single pixel and the results are startling. With the Samsung DLP projector, single pixel white levels were 94% of what they are with full field white. In fact here is the data:

With a LCOS display (JVC RS35) this same experiment was repeated and single pixel white levels were maintained at 79% of the full field value and well into the 90% range with two pixels. So with both of these reflective technologies, there is only a small rolloff in white level as white regions are decreased to a few pixels.

This in turn makes it very easy to measure intra-image contrast as it is only the blacks that are significantly changing and need to be measured. As an example, say we take something like the ANSI contrast test pattern displayed on a reflective DLP or LCOS projector and shrink the white rectangles down to something very small and on the order of a few pixels. When we measure the black area (not in the immediate region of the small white regions) we will find almost no washout from the white pixels and we will end up measuring very close to the black level floor of the projector. Since we know now that the white levels in the small cluster of pixels are unchanged, we will have measured intra-image contrast that is very close if not equal to the native on/off contrast of the projector, which is a startling result. To be sure there are few film scenes that contain single pixels of full white in a perfectly black field but this does have important implications in very dark scenes. As an example picture a space scene with only bright stars on a black field. We now know that in such a scene the stars will be rendered by a DLP or LCOS projector at close to what is called for by the luma value of those white pixels in conjunction with the display gamma. Since the small amount of display luminance will cause little washout (the stars are small) and the black level will be very low (maybe not at the black level floor of the projector but at some low value determined by film black and how the film was transferred and encoded to video). Bottom line, the the intra-image contrast will be relatively predictable and it will be much higher than most people may have previously thought. This realization helps to validate the importance of native sequential (on/off) contrast as a performance metric in dark scenes which is something that is often overlooked.

Perceptions of Sharpness

While we are avoiding the topic of the perceptions of contrast and color due to sharpness, it is worth spending some time to discuss the perception of sharpness itself. In particular, the visibility of the pixel grid surrounding a pixel will be perceived by the HVS (Human Visual System) and will have some impact, for better or worse on the perception of sharpness. From an MTF perspective, all that matters in determining sharpness is the light intensity of on pixels and off pixels. The definition of the pixels and whether they appear as round, square or something else has no bearing and does not change the MTF results. In particular, the fill ratio and the screen door effect (SDE) which results from a visible pixel grid has no bearing on MTF results so long as it doesn’t affect the light intensity of the on or off-state pixels. To a human, however, the visibility of the pixel grid can provide the perception of improved sharpness because the eye is seeing fine details that are smaller than the pixel itself (the pixel grid). At the subconscious level, the presence and perception of this fine detail may act to increase the perception of sharpness. For the same reason, the presence of film grain in movies has sometimes been cited as providing a heightened perception of sharpness, even if this perception of sharpness does not translate into seeing more true detail in the movie itself. Both SDE and film grain then may be perceived in ways that are similar to the perceptions of edge enhancement, which is a well known technique that increases the perception of sharpness even though it provides no fundamental improvement in resolution to the content itself (and in some cases can be detrimental to true detail).

With those thoughts in mind, it’s worth looking at the pixel line scans of the Samsung DLP and examine the visible pixel grid structure from 5 “on” pixels and compare them to a similar scan from a JVC RS20 projector. As we will see in subsequent articles, the JVC RS20 projector has lower MTF and less pixel grid structure and it’s anyone’s guess how much the increased perception of sharpness of the Samsung over the JVC is based on the MTF improvement alone or the combination of the higher MTF and more perceptible pixel grid.

In the image above for the Samsung A900, the large peak represents 5 “on” pixels surrounded by 5 “off” pixels and the pixel grid can be seen as the valleys between the peaks (“fingers”) representing the 5 on pixels. By comparison, the 6 “on” pixels on the JVC RS20 as seen below are relatively flat because the pixel grid is less pronounced.

(Note:   The image below was revised 1/7/2010) with improved RS20 data.  As can be seen, the edge pixels on the RS20 are diminished and this is covered in parts II and parts III of this series of articles on MTF.

Line Scan - JVC RS20 6 on (white) pixels and 6 off (black) pixels

Summary

Independent measurements of MTF have been completely missing in digital projector reviews even though they are commonly used in reviews of other products (DSLRs, lenses, etc.). Hopefully this article shows that that there is a wealth of information that can be gleaned by closely examining MTF in digital projectors. Measuring and comparing MTF values is very useful in technology comparisons, product comparisons and most importantly, we’ve shown that it provides sharpness information on the various modes and iris settings of a specific display.  This information can be used by an owner of a display to maximize the setup of their projector for optimum viewing results. Videovantage is a hobbyist blog and we’re proud to have provided the first independent, detailed, glimpse of MTF and the inner workings of a DLP projector at the pixel level.

As we have seen, the Samsung SP-A900B is an outstanding projector from a sharpness perspective and is among the best in this respect.   It would be interesting to compare this projector to other .95″ DLP projectors like the Marantz 11S2 and the Planar 8150 and perhaps we will do that in the future. Sharpness aside, we’ve also had an opportunity to view the Samsung with various content and hope to do a full review of the projector soon. As you might have guessed from some of the data shown above, we also have performed MTF measurements on several JVC projectors including the RS35 and we will be posting these results in subsequent articles in this series. Thanks for taking the time to read through this article and we hope that you have enjoyed what you have read and hopefully learned something new in the process. Thanks to image sharpness aficionado, Mark Haflich at Soundworks Audio and Video, Kensington MD for his help as a sounding board and motivation for finishing this article.

Note:  Part II of this series of articles on MTF compares DLP with 3 successive generations of DILA projectors and can be found Here.

Nov 26, 2009

The introduction of high performance video processing chips that feature per-pixel motion adaptive de-interlacing, high quality scaling and a host of other features has transformed the high-end consumer video experience.  The performance of HQV, VXP and ABT video processing chips has become well known and the use of these chips has gone from the realm of dedicated $2000 set top boxes to being used in high volume displays, receivers and players.  The use of these powerful chips, while no longer a novelty, has been restricted to mostly the  consumer market.  With the release of a couple of products by Gefen that are geared towards post production and the professional broadcast industry, this is about to change.  Gefen has introduced two new products that include the use of the Sigma Designs (formerly Gennum) VXP chips.  Both of these products are designed to facilitate HDMI to HD-SDI connectivity,  while also providing quality scaling, de-interlacing, denoise and detail enhance filters.  This article focuses on the HDMI to HD-SDI product called the, “EXT-HDMI-2-HDSDIS”, which lists for $1499.  Another similar product called the “EXT-HDSDI-2-HDMIS” is also available which converts HD-SDI to HDMI and also includes VXP processing.  This unit is available for $1299 but was not covered in this article.

For this article, we took a close look at the Gefen and the first thing that is apparent is that it’s meant for serious broadcast quality usage.  It’s a rugged unit that comes with the hardware needed for rack mounting and there is a lot of attention to detail to ensure that it works reliably in this environment.  As an example, the unit is shipped with both a DC power cord and a HDMI cable and both have integrated fasteners designed to keep the cables from being easily disconnected.  As a nice touch, both are blue and match the color scheme of the enclosure.

The unit also comes bundled with an easy to use remote control (shown at the left) along with two batteries (one for a replacement backup).   The remote is essential as there are no front panel switches or displays and all settings are accessed through the remote.

In providing conversion between HDMI and HD-SDI, Gefen has targeted a relatively small, niche market where there is a need but few products to satisfy this need.  Other less expensive HDMI/HD-SDI converters exist (most notably the HI5 and HA5 adapters by AJA), but these support only a few video formats and are missing key formats like 480p and 1080p60.  By contrast, the Gefen supports these and a host of other formats via pass through mode and will also up and downscale to other formats and resolutions.  Frame rate control is also performed and the unit can do reverse 2-3 pull-down in order to generate native 24hz frame rates from 60hz telecined content.   Also included are picture adjustments for denoise, detail enhance, color and gamma adjustments.

Feature Set

In addition to basic HDMI to HD-SDI conversion, the Gefen adapter contains a host of features, including:

• 10-bit resolution for greater precision and dynamic range
• Proprietary 10-bit motion adaptive video de-interlacing with edge interpolation for HD/SD formats
• Support for all HDMI 1.2 audio formats
• Advanced noise reduction and enhancement.
• Max. active image size of 2048 samples x 2048 lines PBP processing for various combinations of video and graphics with alpha blending.
• Fully integrated sprite based multi-plane OSD controller.
• Frame rate conversion to/from any refresh rate.
• Pattern mode w/ color bars & coarse hatch patterns.
• Color correction.
• Noise Reduction.
• Detail Enhancement.
• Brightness Adjustment.
• Gamma Selection.
• Aspect Ratio Select.
• Custom Timing output mode.
• French/English Menu set.
• RS-232 upgreadeable firmware.

A Closer look

Let’s take the Gefen for a spin and look at some of the OSD menus and examine some of the features in detail.

Main OSD Menu (click to enlarge)

The menu above is the main menu that comes up when the menu button on the remote is pressed.  As mentioned there are no front panel indications and all interactions are handled via menu overlays that are inserted into the video stream.  So a person needs an HD-SDI monitor or similar equipment in order to navigate through the menus.

At the bottom of the main menu, one can see both the input and output video formats  (1080p24 in this case).

Output Format Menu (click to enlarge)

The menu above shows the available output video format options. As can be seen the choices are extensive and include most if not all of the video formats that one would want.

Link Options (click to enlarge)

The link configuration menu to the left allows the user to select between RGB and YCbCr (4:4:4 or 4:2:2) colorspace options.

Not shown is the Genlock Reference menu which allows the user to lock the output timing based on the input signal or based on an external clock provided at the rear of the unit.

Next let’s take a look at the available input formats from the menu below.  The unit can auto detect the timing being used or it can be forced to lock onto the various video and computer formats shown on the two screens below.

Input Video Formats (click to enlarge)

Input Computer Formats (click to enlarge)

Next let’s take a look at the menus used to control the various picture settings

Color Controls (click to enlarge)

Here we see the basic brightness and contrast settings.  Independent adjustments for R,G or B are a nice touch.

Gamma adjustments (Click to enlarge)

Here we see provisions for basic gamma adjustments.

Detail Enhance Controls

Here are the detail enhance settings (more on that later).

Noise Reduction Menu (click to enlarge)

Summary

Overall, the Gefen works very well in performing HDMI to HD-SDI conversion.   Scaling between various input and output video formats worked fine and without a hitch.  There was an initial concern that it might be difficult to sync the device properly since there is no built-in display, but Gefen thought about this and provided a video output format toggle on the remote so that the user can get it out of whatever mode it may be in and force it to use one of the common video formats (480i, 576i, 720p and 1080i).  In this way the user can quickly get the unit sync’ed up to a display and back into the OSD menu.  The unit also allows the user to specify a power on video format so that no user intervention is required on powerup even with rare video formats.

The only issues encountered were with features that are probably low on the priority list for a product such as this.  The filter settings for detail enhance did not seem to work as well as other VXP implementations that we’ve tried.  Similarly, 24p reverse 2-3 pulldown seemed to suffer from occasional motion studders.

Overall, the unit worked very well and if one wants HDMI to HD-SDI conversion across many different video formats or support for uncommon video formats such as 2Kp24, the Gefen is the only game in town.

If you’re a Videophile, you’re already aware of the importance of a good display calibration, and you may have invested the time and money into doing the job yourself.  If you have done the calibration yourself, you may have come to the realization that many calibration programs are written by programmers rather than calibrators and these programs can be hard to work with or missing in key features.  For these reasons, I was very interested in hearing the news that Tom Huffman was releasing his own calibration software.  Tom is well known and respected in the AVS Forum and video calibration communities and he has authored some key posts at AVS Forum (including this must read on Color Management Systems). Being an experienced calibrator, Tom knows what he wants to see in calibration software and after using the product to calibrate several different displays, I’m happy to report that Tom has accomplished what he has set out to do.  ChromaPure is an excellent video calibration tool and as we will see, it has some key features that sets it apart from the competition.

Quick Summary

Before getting into the nitty gritty calibration details, let’s talk about the key feature set of the program.  (Also please note that v1.1 of the software was released just as this article was being completed and the features of this version will be discussed towards the end of this article).

• Very nice and easy to use CMS calibration menu with separate % error indicators for HS and L for all RGBCYM primaries and secondaries.  Each color also displays the color difference (dE) between the selected standard and the measured value. This feature alone makes the program a standout.  If a full blown CMS is not available, the program also has a nice color decoding menu as a rough way to optimize the gamut with the tint/color controls.
• Use of color difference (dE) error indications throughout the measurement menus (greyscale, gamut, etc.), which makes it easy to decide when it’s safe to stop tweaking or when more calibration time is necessary.
• User selectable color difference conventions (CIELUV, CIELAB or CIE94).
• User selectable colorspace choices (Rec.709, SMPTE-C, EBU, DCI).
• Meter offset feature, which makes it very easy to train one meter from another.  This can also be used to calibrate directly from the projector while compensating for room and diffuser effects  (first measure the projector off of the screen without the diffuser and use this offset information while performing subsequent measurements directly off the projector).
• The program supports the following probe types:  X-Rite Display 2 Colorimeter, X-Rite Chroma 5 Colorimeter and X-Rite i1Pro SpectroPhotometer
• Optimized Workflow - Leave it up to an experienced calibrator like Tom to get the calibration steps ordered correctly.
• Very intuitive and easy to use GUI - The calibration steps are ordered from top to bottom with each step represented by an icon that activates a menu tab.
• All of the standard calibration metrics are well supported (White Point, Greyscale tracking, and Gamma).
• Very detailed and comprehensive context sensitive help menus.

Licensing/Installation/Bundling

The ChromaPure software can be purchased separately or sold as a bundle with several different probe types.  Installing ChromaPure is a breeze and it’s possible to install it on multiple computers so long as the same probe is used.  Unlike other programs, ChromaPure does not lock the meter in order to prevent its use in other programs.  If you buy a meter as a bundle with ChromaPure, you’re free to use it with other programs and if you use a meter that isn’t bundled with ChromaPure, it will work fine so long as the software can read the serial number on the probe so that ChromaPure can be licensed properly.

So how does the software perform?

After installing the software and hooking it up to an X-Rite Chroma 5 Colorimeter, the first thing that was readily apparent is the maturity and stability of the instrumentation interface.  With one minor exception that I’ll talk about later, the probe readings were very solid, consistent, reproducible and there were no issues at all with the probe interface.  As a first release of the software, I expected the probe to behave similar to other programs that I’ve used in the past which is to say a little flaky.  Instead the probe interface is rock solid.

The other thing that is readily apparent is the well thought out and easy to use workflow.  The software layout is perfect for the typical calibration.  If one goes through the menus from the topmost icon buttons (pre-calibration steps) through the middle icons (calibration steps) to the bottom (post-calibration steps), the results will work very well.   As a case in point - one display that I calibrated was a JVC RS20, which offers a very good CMS (with v1.1 of the JVC firmware), the RS20 also has an 11-step greyscale adjustment where both the greyscale and the gamma can be adjusted.  The greyscale is adjusted using separate R, G and B settings while the gamma is adjusted using the white (RGB) setting.  The ChromaPure work-flow defaults to setting the gamma first, followed by the greyscale color balance and I think this makes much more sense than the other way around (because the gamma adjustment is usually a more coarse adjustment while the greyscale is more of a fine tuning adjustment).  Doing it in this fashion helps to minimize the amount of time spent doing unnecessary calibration, although a person is free to move the calibration steps around in a different order if they wish.

So let’s take a close look and examine the menus along the way.

The Option menu is not too unusual with the exception that it provides support for various dE models and another nicety that allows the user to select the number of measurements and report either the Mean (average) or Median.  Here is a link that describes the differences between the two and it’s easy to see why I prefer the Median method if the readings are not consistent and there are some outliers in the data set.  In actual practice the measurements with a C-5 were consistent so both methods delivered about the same results.

The offset menu is shown below and as previously mentioned this is a nice feature that allows someone to automatically keep and maintain an offset between two different probes.  This is particularly useful when one probe is not very accurate but has better dark performance, in this case the less accurate meter can be “trained” by the more accurate meter with a bright test pattern and then used for dark, low IRE measurements.  I also found it very useful as away to correct for diffuser, screen and room color errors.  In this situation the same meter (C-5) was measured off the screen without a diffuser (as the reference meter), the diffuser was then added and while using the same test pattern a second reading was done at the projector (as the field meter).   I noticed a 5 dE (CIE94) color error from just the diffuser alone being added (both measurements off of the screen), so you can see why it’s important to factor out these contributions to color error.

Options Menu (click to enlarge)

Pre-Calibration

Next up is the pre-calibration Grayscale adjustment (see below) which tells us what to fix and where.  This also gives us a baseline for the final improvements to gamma and greyscale.  As can be seen from the screenshot below, the overall menu layout is excellent and it’s a nice touch that the gamma and greyscale RGB balance are all presented on the same screen.

Greyscale pre-calibration (click to enlarge)

The Color Gamut is also measured as part of the Pre-Calibration steps.  As can be seen the x,y locations of RGBCYMW are displayed relative to the chosen colorspace (Rec. 709, EBU, etc.).  The RGBCYM luminance information is also shown which is a critical component that is often left off of other calibration menus when measuring the gamut.  The color difference dE for each primary and secondary is also shown.

Gamut (click to enlarge)

Calibration

Now we can move on to the actual calibration.  The White Point menu is shown below.  (Side note:  The user should be careful to ensure that the Y reading is well above the minimum sensitivity of the probe and Chromapure includes a Y reading for this purpose, something that was added at my request early in the Chromapure development phase).  Note that the measurement smoothing settings that we saw earlier in the options menu are repeated here as an override for this particular menu.   If the user has a display where the gain and offset settings are buried in the service menu, this is the time when they need to make those adjustments to get the coarse D65 color temp adjusted before final tuning with the greyscale and gamma adjustments.

White Balance (click to enlarge)

The Color Decoding menu was skipped because the display being used (JVC RS20) uses a CMS which is the more precise and therefore preferred way to set the color gamut.  The CMS menu is shown below.  The separate indicators of Hue, Saturation and Lightness along with the dE for each color makes using the CMS a breeze.  This feature is a blessing for anyone with a display that has a CMS and one of the standout features of ChromaPure.  Note that the colors are designed to be worked through in the order seen in the menu (top to bottom, left to right) and the color selection automatically advances.   Once a color has been measured it’s possible to go back and redo it, but other than that the order must be followed.  I would have preferred to go through the colors in any order (based on the sequence of test patterns being used) and I’ve communicated this desire to Tom, so hopefully this will be added in a future release.  Tom has mentioned that test patterns are available for free that follow the order being used in the program so that is another option as well.

CMS Menu (click to enlarge)

After getting the gamut dialed in, the gamma adjustment comes next.  The continuous measurement option makes setting the gamma a very quick operation that yielded very accurate results.   The menu below shows the results after adjusting for a gamma of 2.2.  This also shows one weakness in the initial release which is support for only 9 data points along the greyscale.  Supporting the 5% grey point is very important in getting shadow detail correct and lack of an adjustment point in this region is a shortcoming.  After discussing this with Tom, I was pleased to see that support for a gamma adjustment in this region was made a priority in v1.1 and it is currently available.  Some devices including the JVC RS20 also include support for a 95% gamma adjustment and I would like to see this added in future releases (along with support for 20 greyscale steps at 5% increments across the full greyscale).

Gamma Menu (click to enlarge)

Post Calibration

After the main calibration steps have been completed, it’s time to measure the greyscale again.  This is the same step as was done previously, if everything is okay a user can advance to the next step.  If there are still issues (color or gamma), it’s possible to go back and touch up one or more regions.  As is usually the case with calibration, all settings tend to be related so that modifications in one area will make changes elsewhere and the post calibration measurements are needed in order in order to determine if one is done or if more work is needed (lather, rinse, repeat).

Post-Cal Greyscale (Click to enlarge)

Next up is the color gamut measurement again.  This step is needed in order to determine where the xyY points ended up at for our primaries and secondaries.   If the greyscale, gamma and color gamut all measured well during post calibration then we are done, except for saving our data and generating a report.  Otherwise, if something has shifted during the adjustment phase it’s time to go back and do another iteration.

Report Generation

The report generation uses Microsoft Excel which provided the only real snag that I encountered with the product.  The laptop that I use for calibration has Open Office Calc loaded rather than Excel and the report generation is incompatible with calc.  For now if you use ChromaPure you’ll want to have Excel already loaded.  Tom has said that later versions of the product will incorporate custom plotting and reporting software so that Excel is not needed.  The reports that are generated are excellent however and I’ve included a sample file in .pdf format.  It’s worth mentioning that there is a workaround if Excel is not load, simply export the data, copy it to a machine with ChromaPure AND Excel installed and then import the data and then press the calibration report icon.

Here is a sample report generated in PDF Format that was taken from an actual calibration of a JVC RS20 that was used for this article.  As you can see it’s possible to obtain excellent results from ChromaPure and if the display has a full greyscale and CMS adjustment as the JVC has, ChromaPure is an essential tool.

Suggestions/Issues

The only real issue that I had when running ChromaPure happened while I was training a QA engineer on calibrating a 50″ Panasonic Plasma and had to go back and forth between menu’s and measurements while explaining how the calibration process works.  As mentioned, ChromaPure allows the calibrator to freely move back and forth across the menu’s in an ad-hoc fashion and apparently I had taken some measurements with the continuous mode enabled in in one menu and then attempted another measurement in another menu while the continuous mode was still enabled.  Subsequent measurements in the new menu caused the software operation to lag and eventually become unstable and crash.  I reported this to Tom and it was fixed with the release of v1.1.  Other than that the program worked exceptionally well.

As previously mentioned, one other feature that I would like to see added is measurement support for all 5% greyscale steps which can provide a better perspective on greyscale accuracy, particularly if the greyscale has sharp spikes and dips.   Most displays do not provide calibration points with this degree of precision which is why it was left out of ChromaPure, but measuring the greyscale with this degree of precision even if one can’t directly modify the greyscale is still useful.  This is particularly true if there are greyscale spikes or dips that are so pronounced that a person may be willing to tradeoff accuracy in the adjacent 10% steps (or full greyscale if 10% adjustments are not available).   In conversations with Tom, full 5% support is targeted for the 2.0 release so this feature should be available soon, in the meantime the program is very useful as is.

Another feature that I would like to see added is color and greyscale measurements that are completely random access.  The program auto advances in a prescribed order and allows the user to go back and redo steps, but initially the program forces the calibrator to go through the colors in a prescribed fashion as the later steps/colors are grayed out.  It seems easy enough for the software to support auto advance while also allowing the user to jump ahead to whatever color/greyscale step that they want to start with.

New V1.1 features

As mentioned, a new version of the program was released after the initial cut of this review was completed.  The new version added some important new features including:

• Support for DTP-94 colorimeter
• Support for CIE2000 color difference model
• XML support for session exports.
• Handling of situations where continuous mode is enabled in one menu and a single measurement is attempted in another menu.

What’s next

Tom says that many more features are planned for later releases including:

• Calibration automation by building support for external signal generators.
• Support for additional high-end meters.
• Dropping the Excel reporting entirely and adding integrated reporting with pdf output (you mention this in the review).

Tom says that he hopes to have these features implemented before the end of the year.  He also hopes to release Version 2 with new modules and a richer feature set in the first or second quarter of 2010.

Summary

I asked Tom what his key goals were when he set off to develop ChromaPure.  He said that he wanted to offer a calibration program that incorporated the features that he wanted to use, along with a well thought out workflow.  I’m pleased to report that ChromaPure has delivered on those goals and more.   ChromaPure is an excellent calibration program that is well worth the very reasonable $200 purchase price.  The software is easy to use, well thought out and is very stable.  Surprisingly so for an early release.  The CMS support is a standout feature that is not available in competing products and by itself is worth the price of the software.  When combined with all of the other features like multiple colorspace support and multiple color difference model support, the program is a compelling bargain.  I found that the program yielded excellent results while significantly speeding up my calibration times.  After using the program, I’ve decided to use ChromaPure for all of my display calibration needs.  Future display reviews on this blog will be done using ChromaPure, so expect the graphs from ChromaPure to show up often.

The original Silicon Optics Realta HQV chipset set the bar in quality de-interlacing when it was released 4+ years ago.  It brought $50K Teranex level performance and per-pixel motion adaptive de-interlacing into an economical single chip form factor.  It’s been widely used since then and incorporated in many products including the contrast leading JVC RS20 projector.  IDT purchased Silicon Optics and the HQV property last year and it’s been an open question if IDT would continue development of the HQV architecture.  This question is made all the more interesting in light of the heated competition it’s been getting from the Sigma VXP (formerly Gennum) chipset.  The question of where HQV is headed was answered today when IDT announced the new HQV Vida chipset.  Below are some of the key details from todays press release (which can be found here):

“The Vida processor enhances image detail and quality with four-field motion adaptive de-interlacing, multi cadence tracking, expanded 12-bit color processing and detail enhancement. The result transforms standard-definition sources to HD quality and makes HD look even more detailed. Moreover, the device also provides real-time clean up of highly compressed video, reducing compression artifacts of block and mosquito noise from lower-quality sources.

July 18, 2009

Microsoft Silverlight 3 was released last week and this seems like a good time to write an article about the technology.  This article will provide a quick overview of adaptive streaming which is a key feature of Silverlight.  Microsoft’s Silverlight is well positioned to compete with Adobe Flash which is the other dominant adaptive streaming technology.  In particular, this article will discuss what adaptive streaming means to the videophile community.  If time permits, I may do a second and more in depth article on the technology itself.

Let me start by saying that I hate to come out and say it, but I have this recurring perception of Microsoft as a tech giant that is more interested in strip mining customers for every last dollar than doing something truly innovative (Vista comes to mind).   Sometimes change for the sake of change can be used as an excuse to generate profits without providing true value to the customers.  Every once in awhile though, something comes out of Microsoft that shakes this negative perception and makes me realize that they actually do have some smart people in Redmond (although having met Stacey Spears I already knew that).  Microsoft Silverlight is one such product.

Prior State of the Art in Video Streaming

First, let’s discuss traditional video streaming and examine the issues and see why something better is needed.  Traditional streaming video has used a “Progressive download” format where a client requests a video and the server satisfies this request by dumping a a lot of data in file transfer fashion over the network which the client queues up in a play-out buffer.  Once the play-out buffer has filled up sufficiently to ensure a few seconds of uninterrupted playback, the video is displayed.   This is a very common approach that is still in widespread use today.  Unfortunately this ”dump and play” format has a lot of drawbacks that I’ll highlight below:

• Dump and Play places large unrestricted bandwidth demands on a network.   The dump process will use as much bandwidth as is available so that it can transfer as much data to the client as quickly as it can.  This is great for downloading files, but if a user is going to be downloading and watching say a 2 hr video it might be better to throttle (control) the download and spread it out over time and give some of the network bandwidth to other users.  In some cases, throttling mechanisms exist, but it requires equipment,  planning, usage policies, etc., and all of this “OA&M” adds cost to the content providers.
• If the user terminates playback prematurely, the play-out buffer is wasted which in turn means wasted bandwidth that could have been allocated to other users.
• Random Access is a cumbersome process with the dump and play format because a user may not have yet downloaded the new playback location.  If this is the case, then satisfying this location request means tossing the old play-out buffer and streaming from the new location which also means wasted bandwidth. It can also mean long delays while the new play-out buffer is filled.
• Streamed files are difficult to cache and current systems do not scale well.
• Older streaming technology isn’t web friendly and it often uses custom ports that aren’t well known, which means that they may be blocked by firewalls.
• TCP is often used as the transport, which while reliable, can add retransmission delays if data is lost.  These retransmission delays can be detrimental to the user experience if the user wants smooth uninterrupted playback (although at the possible expense of additional video noise represented by the missing data).
• In addition to the retransmission issue, TCP is also not an ideal transport from a video streaming perspective because it incorporates a congestion control mechanism called the, “Nagle Algorithm”.  This means that data is initially sent slowly using a reduced window size which is gradually ramped up over time so long as the data is received okay on the other end.  For video streaming we need to send the data quickly, as needed and on demand and without outmoded bandwidth restrictions.

To solve the dump and play problems described above, various improvements were attempted such as using RTSP as a protocol and download bandwidth thinning.  All of these new features improved things for both the user and the content data networks (CDN), but it despite these improvements however, serious problems remained.

Enter Smooth Streaming (aka Silverlight)

Adaptive Streaming and in particular Microsoft’s Silverlight incorporate several clever features designed to improve life for both the user and the CDNs.  Rather than look at video streaming as a large file transfer with local playback at the client, adaptive streaming is designed to break up the video into small manageable chunks and send the client only the few seconds of video that is currently needed.  This helps to ensure a fast and responsive user experience and it eliminates wasted download data and therefore wasted bandwidth.  The first question that may come to mind though is, “well that sounds great, but what happens if the network is slow”.  This is where the “Adaptive” in adaptive streaming comes in.   The video being downloaded isn’t just one encoded stream but is actually encoded in multiple formats from higher quality (higher encoder bit rates and resolutions) to lower quality (lower bit rates and resolution) and each is available as a separate stream.   The client is then free to request video from any of the available streams.

If the client is not able to receive the video chunks in a timely fashion (fast enough to ensure uninterrupted playback), it can request a stream where the video chunks are smaller and hopefully, easier for the network to send.  It’s important to realize that the smaller amounts of data don’t represent shorter intervals of time, instead they represent video data that has been encoded at a lower resolution or a lower bit rate or more typically, a combination of both.   For a user with a fast connection for example, chunks of video data all encoded at high resolution and high bit rates may be sent while a user with a slower connection will get video data of a lesser quality.  In both cases however, the user experience in terms of responsiveness should be similar.

Adaptively switching between streams seems easy enough in theory, but MPEG technology was built on the premise of video frames being encoded in a form that represents differences between past and future frames.  Randomly picking a location from a stream that starts on say a P or B frame will result in a messy image that may not even be recognizable.  To fix this situation, each chunk from all of the streams must start at the beginning of a GOP and each GOP must be closed and not have references to other GOPs.  This allows the adaptive switching to be seamless and the only artifacts that the user will see is a difference in the quality of the video.

Since data is being sent as small chunks or fragments, Microsoft has opted to use a new file format based on a Mpeg-4 container format which is the first new file format used by Microsoft since ASP.  Since each chunk of video data is small and completely self contained, the client need only rely on it’s file-name as a way to reference its stream location and quality which makes the files easily cacheable by the CDN and therefore easily shared by other clients who may at some point want to download the same file.

Silverlight makes full benefit of the fact that these files can be cached by using HTTP as the download mechanism for fetching these files.  In addition to smooth streaming, using standard HTTP commands to fetch these files is the other key benefit of Silverlight.   Suddenly, we have an architecture that is completely compatible with the world wide web and we’ve gone from an older style video streaming architecture with heavyweight file transfer operations,  custom servers, bandwidth issues, etc. to something that pulls data on demand from the client side via HTTP and is therefore very representative and compatible with typical web and HTTP operations.

Silverlight Benefits

So let’s summarize the benefits that we’ve talked about so far:

• User interactions are quick and responsive.
• Bandwidth usage is more predictable.
• More users can be supported over a given bandwidth because every user consumes only the bandwidth that they are currently viewing rather than downloading large chunks to be played later.
• HTTP based access plays well with the Internet (no firewall problems).  This also facilitates video playback from a browser (note: up until version 3, Silverlight was a browser plug-in technology, but version 3 supports desktop playback as well).
• Small files are completely cacheable so that the “edge caching” that is already  in place by CDNs can be used.
• No specialty Infrastructure is required – HTTP Internet equipment will do the job.
• Since a file isn’t transferred in it’s entirety, the only portions of a video that may be resident on the client is the small chunk being played, so in theory, copyright protection is improved.

In addition, Silverlight 3 also has some other key features that we haven’t yet discussed, including:

• Improved Digital Rights Management (DRM).
• A rich multimedia toolkit that allows videos to be interactive.
• A 3D graphical toolkit that allows for transformations, animations, etc.

So what does this mean to the Videophile Community?

So far Silverlight seems like a huge positive and a very clever solution to a problem that has been in need of a fix for quite sometime.  Because of these advantages, it is very likely that this technology will be rapidly adopted.  Here is the rub though.  There are really two use case scenarios that need to be addressed.  The first use case represents viewers who are surfing the web and viewing something like youtube for example.  In this scenario the user wants fast response and they may be heavily interacting with the stream - Starting, stopping, replaying segments, skipping ahead, etc.   In this scenario where fast response and quick playback is the norm, Silverlight is a huge boon to both the users and the CDNs.  But let’s examine the use case represented by the videophile who wants to stream and view a movie in its entirety.  In this scenario, user interactions will be very limited and the viewer is most interested in viewing their movie with as high of quality as possible.  In this scenario quick playback will take a back seat to video quality and the user may be willing to queue up the video 30 minutes or an hour before hand, if this means that the quality of the video will be the best that it can be.  In this scenario, all forms of adaptive streaming that rely on transient switching to a lower quality stream may come up short.

Unfortunately, the videophile community is small and it’s unlikely that the second use case described above will weigh much on the minds of content providers.  In the spectrum of video quality, this means that most of these adaptive streaming tehnologies will fall well below the current king of consumer video quality the Blu-Ray Disc.  It’s likely to fall well below even broadcast HD quality.  The videophile community has few options other than to vote with your pocketbook and use your collective purchasing power to favor those content providers who are streaming movies with the best video quality.  Hopefully at some point videophiles can have the best of both worlds.  Fast, responsive, streaming video when surfing the web and high quality streaming video when watching an uninterupted 2 hr movie.

July 08, 2009

On Monday, the VideoLan Project, developers of the popular streaming media software player VLC  shipped a major new release, version 1.0.0.  VLC is well known in the open source community.  What started as a student project has grown into one of the better known and successful open source projects.  VLC supports many different video formats and streaming protocols and does so on many platforms including Linux, Windows and Mac OS X.  According to the VideoLan Project website, the key features of VLC 1.0.0 are:

• Free, Open Source and Cross-Platform
• Independent of systems codecs to support most video types
• Live recording
• Instant Pausing and Frame-By-Frame Support
• Finer speed controls
• New HD Codecs (AES3, Dolby Digital Plus, TrueHD, Blu-Ray Linear PCM, Real Video 3.0 and 4.0,…)
• New formats (Raw Dirac, M2TS,…) and major improvements in many formats…
• New Dirac encoder and MP3 fixed-point encoder
• Video scaling in fullscreen
• RTSP Trickplay support
• Zipped file playback
• Customizable toolbars
• Easier encoding GUI in Qt interface
• Better integration in Gtk environments
• MTP devices on Linux
• AirTunes streaming
• New skin for the skins2 interface

They say a picture is worth a  thousand words so here is a diagram that shows the various ways that VLC (and an associated video server called VLS) can be used for video streaming:

VideoLan Streaming diagram

 For more details including the free download link head over to the VideoLan Group

 The much anticipated Oppo BDP-83 combo blu-ray player with SACD and DVD-Audio has officially shipped! 

Oppo’s players are well known in the videophile community and are highly regarded for their ability to play and upconvert SD content with very few issues.  The Oppo DV-983H DVD player for example was the first player to ever score a perfect 100 on the rigorous stress test used at, “The Secrets of Home Theater and High Fidelity” website.  

One of the reasons for the huge interest in this product stems in part from the high visibility that this product received during the late stages of development and testing.  Oppo released hundreds of units to randomly selected users as part of an “early adopter program”, the early adopters then helped to test the product and provide descriptions of bugs.  The comments from the users was posted publicly and discussed in a thread on AVS Forum.  The product was not released for sale until a majority of users in the EAP agreed that it was ready for prime time.  I had hopes of being one of those selected during the EAP program, but unfortunately luck was not with me as I wasn’t one of those selected so I did not receive my unit until a few days ago.   

Another reason for the interest in this product is because of it’s relatively low price tag of $500 which is a bargain considering that the only other universal SACD, DVD-A and BD player on the market (the Denon DVD-A1UDCI) has a MSRP of $4,500.

After just a few days of using the BDP-83 a few things have become apparent:

• This player is very fast to load and play and is among the fastest BD players.
• The quality of deinterlacing and upconverting is very good as it uses the well known and mature anchor bay technology chipset.
• The remote is excellent with backlighting and many direct keys.
• Since it supports DVD-A, SACD and BD formats, it’s a perfect match for older pre-pros that lack HDMI audio input connectors but have a single block of 7.1/5.1 analog inputs like my venerable Lexicon MC-12B. 

So far the video performance seems excellent, but I’m still in the process of evaluating the audio performance.  The guys at The Secrets website have posted an outstanding review however, so you can read the details here.   In addition there are several useful threads over at AVS and they can be found below:

Official BDP-83 Owner’s Thread

BDP-83 vs other BD Player Thread

The unit can be purchased direct from Oppo here

June 21, 2009

Frequent debate and speculation has centered around the sharpness differences between DLP and LCOS display technologies.  Unfortunately these comparisons are complicated by factors such as 3-chip panel convergence errors, lens quality and chip size.  Often, these debates have compared images from single chip DLPs with no convergence errors, larger chip sizes (.95″ for most commercial DMDs) and expensive optics.  By comparison, LCOS chips tend to be smaller (.7″ JVC or .6″ Sony), with less than perfect panel registration (owing to their 3-chip design) and optics befitting their lower price points. 

With the recent rollout of the smaller .65″ TI DC3 DMD which has been incorporated into Optoma’s new HD8200 , we can equalize one of the variables in the sharpness equation and compare images from similar sized chips directly.  Unfortunately MTF(sys) = MTF (lens) + MTF (chip) so without knowing the MTF (sharpness) of the lens it’s impossible to say conclusively that sharpness differences in an image comparison are solely due to differences in the technology.  But this is an interesting comparison nonetheless and it does provide one data point even if the results are not conclusive.  

For this comparison we’re going to examine the sharpness of 3 projectors.  For DLP projectors we are going to examine images from the new Optoma HD8200 with the new .65″ DC3 and also the Planar 8150 which has the older and larger .95″ DC3 DMD.  For LCOS we will examine a JVC RS20 with very good convergence between red and green but with blue off by approximately half a pixel.  Since humans are much less sensitive to blue and it is weighed much less than the other two colors (~11%), the blue alignment issue is less significant than if one of the other two colors were misaligned. 

First Up – Optoma 8200 vs JVC RS20

First let’s take a look at images of both projectors from normal viewing distance using the Sony Blu-Ray Easter Egg resolution test pattern:  

Optoma HD8200 (click to enlarge)

JVC RS20 (click to enlarge)

 On initial examination, both images look very similar and seem to resolve this pattern well.  The biggest difference in the images is the cyan and magenta tint on the RS20 that affects some of the areas of the test pattern, particularly the zone plate portion of the test pattern which is the area at the corners with radiating lines looking like spokes in a wheel.  This is most likely due to color bleeding from the blue misconvergence.  Unfortunately there is also a slight tint to the white in these photos that wasn’t in the actual images and was introduced by the camera.  These particular photos were taken with a Nikon Coolpix camera, but later photos in this post are taken with increasingly higher quality cameras (a Panasonic Lumix DSLR and later a Canon 50D DSLR). 

Optoma (click to enlarge)

RS20 (click to enlarge)

Examining these two images closely, both projectors look remarkably similar as far as resolution, although the RS20 seems to have a slight advantage in detail and sharpness, with the drawback of blue color bleeding from the misconvergence of blue.   

RS20 vs Optoma – a more rigorous look

Next, let’s take a closer look at the pixel definition of both projectors.  In this example we’ll use a test pattern consisting of single pixel alternating on/off columns.  What makes this pattern interesting  is that this represents the maximum spatial frequency of a digital display and this is how MTF is traditionally measured on a digital microdisplay.  A detailed discussion of MTF is beyond the scope of this post and rather than get into the details here I’ll just mention a few points that are important to keep in mind.

The formula for MTF is:  MTF = Lmax – Lmin / (Lmax + Lmin).  

(edit: 6/30/09 – The general form for MTF is complex, but this formula works for a specific spatial frequency.  A discussion of this can be found here: http://www.revisemri.com/questions/equip_qa/measuring_mtf )

MTF is a contrast measurement relating the luminance from the peak white (Lmax) from the “on” pixels to the darkest blacks in the valley between the on pixels (Lmin).  Classic MTF measurements use a sinusoidal waveform of varying frequencies from low frequency to high frequency and MTF is calculated for each of these frequencies and plotted on a graph with the Y-axis being MTF as a percentage from 0 to 100% and the X-axis being the frequency.  Because the equation is normalized, MTF graphs typically start off at a high percentage (100%) and then roll off at the highest frequencies.  With a digital panel, the alternating on/off pixel pattern provides the highest spatial frequency and therefore the MTF will be at it’s lowest point on the curve with this pattern.  For this reason, this alternating pattern is the most useful pattern used for comparison purposes and is often referred to as the Nyquist cut-off frequency. 

For these next comparisons, we’ll convert the images to a greyscale, which makes it easier to compare relative luminance differences.  Any color details including the blue misconvergence of the RS20 is not thrown away from these photos but rather is added to the luminance values in the greyscale.  I should also mention that all of the photos in this post were taken directly off of a Stewart Studiotek screen, so any small “sparklies” and texture that you see are due to the screen.  As before, the camera used, camera distance and camera settings are all the same between this next comparison with the only difference being a slight change of exposure to equalize the brightness differences between projectors. 

Optoma (Click to enlarge)

RS20 click to enlarge)

With the single alternating pixel pattern above, it appears that the RS20 seems to have a sharpness advantage.   Individual pixels however, seem to have more definition on the Optoma.  In particular the vertical lines on the RS20 seem to merge together.  But, overall however, the pixel brighness of the peaks is brighter on the RS20 and the blacks in the valley (off pixels) are darker.  Using the MTF equation above would indicate that the RS20 has higher MTF at the 1080p Nyquist Limit.    This can be more easily seen when we superimpose portions of both images in the photo below.

Optoma (top) JVC RS20 (bottom) superimposed

Unfortunately a digital camera captures only relative luminance differences in an image and not absolute luminance, so without more information (like the camera gamma), it’s probably not possible to calculate MTF directly.  We can, however, determine relative luminance differences and this in turn tells us which projector has the higher MTF even if we do not know what this number is exactly. 

Since we are using a digital camera to measure sharpness differences we can also go a step further and rather than visually inspect just a few pixelsm we can look at the histogram distribution of all of the pixels of each image.  These are shown below and were obtained from Photoshop: 

Optoma histogram

RS20 histogram

In these histograms, the X-axis represents pixel brightness with dark at the left and white at the right.  The Y-axis represents the total number of pixels in the image that share the same brightness.  Ideally, with a perfect display, one with high MTF, we would see two vertical spikes, one representing bright (on) pixels and another spike of black (off) pixels and with little spreading between them.  The important thing (from a resolution standpoint) when viewing these histograms is to have as much distance separation as possible between the spikes.  Looking at the images above, we see that the RS20 clearly has more separation between the spikes even though the bright spike representing the spread of “on” pixels is broader and less defined than the dark (off pixels) spikes. 

Next Up – Planar 8150 vs JVC RS20

 The next comparison is between the same RS20 and the Planar 8150 which uses the larger and more common .95″ DC3 DMD.   For this comparison we will use a Canon 50D camera which yields a little bit more resolution in the photographs and  higher pixel density (15.1M Pixels). 

Let’s start off by examining real world images as we did with the Optoma and RS20 comparison.  The images below are from a dialog box in the Windows desktop: 

Planar (click to enlarge)

RS20 desktop (click to enlarge)

From these photos the Planar is clearly sharper than the RS20 and by inference also sharper than the Optoma HD8200, although the pixel grid and SDE (screen door effect) is also much more noticeable.   We can see the SDE differences even by looking at a solid white field which is shown below:

Planar White (click to enlarge)

RS20 white (click to enlarge)

RS20 vs Planar a more rigorous look

The first comparison will be of the now familiar, single pixel, on/off, “multiburst” test pattern.  As we did earlier, we will convert these images to a greyscale without throwing away the contribution to Luma by whatever colors are found in the images (including the blue misconvergence of the RS20).  As was done before, the camera position and zoom settings haven’t changed between these photos and the only difference is a slight change in exposure, to correct for the brightness difference between projectors.  It should also be mentioned that slight exposure differences will not affect the relative differences between white and dark and only serve to shift the whole curve towards bright or dark without affecting the shape of the curve and therefore the peaks needed for relative MTF comparison.   Therefore, so long as the exposure isn’t overexposed or underexposed (which would clip the whites or the blacks), this technique will work well and we can use small exposure changes to correct for projector brightnesses.

Planar (Click to enlarge)

RS20 (click to enlarge)

 Next, let’s take an even closer look at each of the test patterns from above.  Note:  The images below were captured by optically zooming the lens and weren’t resized in photoshop: 

Planar closeup (click to enlarge)

RS20 Closeup (click to enlarge)

Planar and RS20 closeups superimposed (Planar on the bottom)

 

Planar RGB Histogram

 

Next is the JVC RGB histogram.   We can see how well the green and red histograms overlap.  The blue histogram, however at least on this particular projector, shows less sharpness in blue compared to the other colors. 

 

 

RS20 Histogram (RGB)

 

.65″ DC3 vs .95″ DC3 DMD

 The data for one projector using a .65″ DC3 and another using a .95″ DC3 have been posted above in conjunction with the RS20 being used as a comparison for each.  It’s left as an exercise to the reader to directly compare the two.   As was mentioned earlier, it is impossible to know if the large differences in sharpness that we are seeing is due to differences in the lens or due to differences in the chip sizes because we are seeing the combination of both.  I will say though that the lens in the Optoma did not seem to be particularly bad.  It did not suffer from too much chromatic aberration for example, and it seemed to be fairly even across the field, although it was a little less so in this respect than the lens on the RS0 or the Planar. 

 Summary

In this post we examined the sharpness differences of an interesting mix of 3 projectors -  Two DLPs using two different DMD sizes (.65″ and .95″) and one LCOS with a .7″ DILA panel.  All three projectors resolved 1080p test patterns well and none seemed to suffer any deficiencies in sharpness with test patterns or video images.  The two projectors that were the most similar in terms of resolution and SDE also happened to have similar chip sizes.  Most surprisingly these same two projectors  (The Optoma HD8200 and the JVC RS20) were the most dissimilar in terms of display technology with one being single chip DLP based and the other being a 3-chip LCOS design.   By comparison, the Planar 8150 was significantly sharper than the other two when compared with test patterns.  With video content the sharpness advantage of the 8150 was also noticeable although perhaps not as striking.  SDE on the 8150 was also much more noticeable unfortunately.  Color bleeding on the RS20 due to it’s 3-chip design was also apparent.  One thing that is clear from these comparisons is that there is much more to sharpness than the panel technology alone.  We have shown for example how a 3-panel LCOS design can be just as sharp or sharper than a given 1-chip DLP design.

  

In this post we also showed how easily the CCD in a ubiquitous DSLR can be used to determine relative sharpness differences.   We used two DSLRs, a less expensive Panasonic Lumix and a more expensive Canon 50D and so long as the projector pixel’s were oversampled with the camera pixels, the results seemed to track very well even with different positions, lens and camera settings. 

 

There is an implication in this post that the panel size alone can have a large impact on image resolution.  More data needs to be collected to prove this implication, but if true it means that the quality of optics as chip sizes are reduced will be increasingly important.  Many manufacturers have shown a desire to reduce chip sizes as a way to lower costs, and we have seen TI do this with the .65″ DC3 DMD used in this comparison, we’ve also read published reports that JVC will go to a .45″ 4th generation 1920×1080 display chip.  As these new, smaller chips are released it will be interesting to see how much image sharpness will be affected.  It may become the case that any cost savings due to smaller chip sizes will be eclipsed by the increased costs required by new demands on the optical and lens system.