Passive 3D Projection – Part 9 Posted on June 29, 2012 By Art Feierman This post is Part 9 in a the series of blogs discussing do-it-yourself (diy) passive 3D projection systems that use two conventional front projectors. This new blog continues the discussion on the use of dual projection system that use color bandpass filtering (i.e., “wavelength multiplexing”), instead of polarization, as the means to separate the right from the left images. In simply terms, wavelength multiplexing, uses two projectors equipped with filters that pass only very narrow bands of colors within the visible spectrum with one of the filters passing a set of colors that is slightly offset from the set of colors passed by the filter being used on the second projector. The viewers then must wear passive 3D glasses whose lenses are filters that pass narrow bands of the visible spectrum that matches those being used on the two projectors. The technique of color bandpass filtering was briefly discussed in Part 1 of this series of blogs on dual projector passive 3D systems and was discussed in some detail in my most recent previous blog (i.e., Part 8 of this series). I suggest you review the earlier blogs of this series in order to have the proper context for this new blog. As I noted in my previous blogs in this series, some of the information used for this discussion is based on information provided by Rob Stewart who has experimented with a number of do-it-yourself dual projector alternatives for passive 3D. Shown below is an illustration (repeated from an earlier blog in this series) showing the functional components that make up a passive 3D projection system. Passive 3D Dual Projector Configuration Before I get into discussing the Omega 3D/Panavision system for “wavelength multiplexing”, as promised for this update, I would like to briefly share some information concerning a possible low cost solution for mounting of external filters (either color bandpass filters or polarizing filter) in front the projector’s lens. This is a follow up to the discussion that was started in Part 4 of this series of blogs. Ideally such filters would be easily moved into or out of the light path of the projector. One hobbyist has reported on the AVS Forum that he has successfully used a low cost bracket assembly that was sold for the purpose of holding a printed circuit board while making repairs (e.g., soldering). The bracket has a clamp for attaching to a table-top or shelf and then an adjustable bracket for holding the PC board. Some such brackets are mechanically configured such that an optical filter can be held by the portion of the bracket intended to hold the PC board and the bracket allows the filter to be moved out of the light path when the filter is not needed (i.e. for 2D projection). Such brackets are available from a number of electronics supply dealers with Panavise being the most widely distributed brand. Also there are some really low cost versions being sold (at the time of the blog) on eBay. Below is an example of a set of two such brackets currently being offered on eBay for less than $10 (plus shipping) under the listing title “2pcs adjustable mounting clip clamp for PCB soldering”. PC Board (filter) Holder/Bracket Omega 3D/Panavision 3D Filter System A color bandpass filtering system was developed a few years ago by Omega Optical and this system was used by Panavision for their proposed approach for presenting 3D in commercial theaters. The filter characteristics were originally optimized for use with commercial DLP, using 3 DMD chips, projectors with Xenon arc lamps (see Part 8 of this series for more info). Unlike the Infitec filter system previously discussed (Part 8), each Omega 3D filter passes five narrowband wavelengths of light spread across the optical spectrum. The exact wavelength of each of these narrow bands is offset between the filters used for the right vs. left images. Omega Filter Bandpass Characteristics The above plot shows both the right and left filter’s bandpass characteristics (i.e., the red and the blue plots) for the Omega 3D filters optimized for use with DLP projectors. Omega offers the kit of filters and glasses with two different sizes of filters and also in versions tailored for DLP vs. LCD/LCoS projectors. The version for use with LCD/LCoS projectors has only recently become available. The basic kit with the smaller size projector filters is being sold on eBay at the THIS LISTING (current at the time of this blog). The kit with a pair of small projector filters and 5 pairs of glasses is approx. $350 while the kit with the larger projector filters and additional pairs of glasses is approx. $850. In March 2012 Omega supplied me with a set of the larger size projector filters and a pair of their 3D glasses for my evaluation. These filters were optimized for use with DLP projectors as they had yet to release the version intended for LCD/LCoS projectors. Pictured below is the pair of Omega filters the 3D glasses. Omega 3D Filters and Glasses The Omega projector filters as well as the lenses of the 3D glasses have a very reflective, mirror like, surface as can clearly be seen in the following photo. Omega 3D Filter (large size) I tested these Omega filters and glasses for a few key characteristics and attempted to identify/verify the capabilities and limitations of this system. The projector I used for this testing was a LCoS based model (JVC DLA-RS40) and although the spectrum of the light output by this projector was not ideal for use with the supplied Omega filters, this should not have any significant impact on the observations discussed below. Light Loss – I used a light meter to measure the light loss through the each filter and also through the corresponding lens of the 3D glasses. I projected a 100% white test pattern and measured the light level (in lux) without the filter and glasses then with the filter placed in front of the projector’s lens and with the one lens of the glasses placed directly in front of the light meter. The light sensitivity curve for the light meter that I used is based on industry standards intended to correspond to the sensitivity of the human eye. The results were approx. an 84% light loss for the right eye and approx. a 78% light loss for the left eye. Note these values are similar to the light loss typical with many active 3D projection systems. However, some active 3D projectors have lower total lumens output when operated in their active 3D mode versus the brightest acceptable 2D mode on the same projector (my JVC is one of those). In the case of using a pair of such active 3D ready projectors but operated in a 2D mode as part of a dual projector passive 3D system equipped with the Omega 3D filters, the results may be a somewhat brighter image with the passive system. This will vary depending on the specific projectors being used and their available 2D and 3D operating modes. 3D Crosstalk/Ghosting – 3D crosstalk occurs when the image intended only for one eye is not adequately blocked from being visible to the other eye. As I noted in an earlier blog in this series, many people will not find the visible 3D Crosstalk/Ghosting obvious when the cosstalk level is kept to 0.5% or less. I measured both the light level for a ‘right image’ as would be see by the left eye and for a ‘left image’ as would be seen by the right eye. In both cases the measured 3D crosstalk was just under 0.4% which means 3D crosstalk/ghosting will only rarely be visible. Impact on Color Balance – Most consumer projectors, including my JVC projector, use UHP lamps which have a rather non-uniform color spectrum (see my previous blog – Part 8) and tend to have low output at the red end of the optical spectrum. The color balance achieved will be the result of the color spectrum of the light coming from the projector (i.e., the spectrum from the projector’s lamp as modified by the projector’s internal filters) as then altered by the Omega filter mounted external to that projector. Since the right and left Omega filters each pass different sets of wavelengths (see spectrum diagram above), the overall color balance of the right and left projected images (i.e., through the different Omega filters) were not the same and both showed a significant change in color balance from the 2D image (without the filters) from that same projector. The projector I used to evaluate the Omega 3D system was first calibrated (i.e., without the Omega filters) to have a 6500K color temperature with the red, blue and green levels closely match. The following two plots show the red, blue and green levels first measured through the Omega right projector filter then through the Omega left projector filter. The results shown in the following plots were for just the use of the projector filters, while results from measurement that also included the 3D glasses did not change the results very much. The projected images through the right filter had a moderate blue tint while the projected images through the left filter had a moderate yellow tint. Right Color Balance w/Omega Filter Left Color Balance w/Omega Filter It may be possible to create a custom calibration for each of the two projectors to at least partially compensate for the impact on color balance caused by the filters. This would involve adjusting the red, green and blue gains to restore a more correct color balance. However, this could result in a lower overall light output from the projector. Color Uniformity – The color bandpass characteristics of the Omega filters change as the angle of the light passing through the filter moves away from the ideal 90 degrees (i.e., perpendicular to the filter surface). The idea location for these filters would be inside the projector before the light enters the projector’s lens. However, since external mounting just in front of the projector’s lens is the only practical way for the hobbyist to use these filters, the characteristics of the projector’s lens is important. A technical representative from Omega has advised that a throw ratio of 1.8 or higher is best to reduce color uniformity issues. Throw Ratio = Throw Distance / Width of Projected Image Note: Throw distance is the distance from the projector to the screen The normal setup for my JVC projector, as installed in my home theater, has the zoom set such that the throw ratio is approx. 1.5 which is less than the recommended minimum value for use with the Omega 3D filter. With this setting I did note a considerable color shift in going from the center of the projected image to the corners. This was most obvious when the left filter was in place and the center of the projected image had a moderate yellow tint while the corners had a moderate blue tint. I used my projector’s zoom to reduce the image size thus increasing the throw ratio to over 2. While the overall color uniformity did improve with the larger throw ratio, some color shift was still visible between the image center and the image corners. 3D Glasses – The Omega 3D glasses are relatively light and I found them to have acceptable comfort for the short term (I did not wear them for more than a few minutes at one time). However, as with all glasses the comfort level will surely be judged differently by different wearers. Since the optical coatings on the glasses’ lenses are very reflective it is essential to prevent light sources from the sides or rear of the viewer from being reflected off the inside surface of the lens back into the viewer’s eyes. The oversized design of the frame appears intended to help block such reflections and this should be sufficient in a very dark home theater environment but I have no first hand experience of how well they perform if there are light sources located to the sides or rear of the viewer. As with the Omega projector filters, the color balance of the viewed image can be degraded if not viewing at nearly right angles through the glasses’ lens. This should not be a significant issue in most cases. Conclusions and Observations – The Positives: – The good news is the Omega 3D system offers very good performance at minimizing 3D crosstalk/ghosting. – Users do not need to use a silver screen material as is required for passive 3D projection systems that use polarization. Thus a screen that is well suited for 2D projection can also be used for passive 3D projection when using the Omega 3D filter system. – The Omega 3D filter system is compatible with projectors that output polarized light as well as those that don’t. – As with dual projector systems using polarization, there are benefits as compared to the single projector active 3D alternatives. With such dual projector passive 3D systems, both the right and left images are being projected at the same time (i.e., time synchronized) and blanking frames are not present between visible frames. Some viewers may notice less flicker in the projected image and also some may find that fast motion sequences in the video appear to provide a more solid 3D effect than with active 3D solutions. Although these advantages may not be immediately obvious to many viewers during a short viewing session, they may result in less viewer eye (or brain) fatigue when viewing 3D video for an extended time (such as by the end of a 2 hour movie or a 3 hour football game. Some viewers may find this improvement significant over active 3D projection solutions. The Negatives: – The impact on color balance can be significant and the extent of the impact will vary depending on the specific projectors being used. From observations posted by hobbyists that have purchased the Omega filter system, it appears that the results I obtained are representative of using these filters with some, and perhaps most, consumer-class projectors. In my case the image as seen by the right eye appeared too blue while the image as seen by the left eye appeared too yellow (i.e., yellow results from a lack of blue). It has been suggested by a representative of the manufacturer that the viewer’s brain will combine the two images to produce one with nearly correct colors. However, my observations, and that posted by some other hobbyists, seem to indicate that some viewers are much more sensitive to the right/left color differences than are other viewers. Either the projector’s internal adjustments or those of an external video processor could be used to at least partially correct the color balance for both the right and left images, but this could be expected to reduce the overall light output of the projector. However, if high lumen output projectors are being used this may be a viable option. Also if the projectors use Xenon lamps (not common with consumer-class projectors), which have more uniform light output across the visible spectrum than the more common UHP lamps, then the negative effect of the filters on color balance can be expected to be less. – As a result of the right and left filters having a different level of light transmission (the amount will vary depending on the specific projectors being used) and also because the different color balance in the projected right vs. left images, there is a potential for “eye rivalry” (as some users have termed it). When this occurs it detracts from creating the ideal 3D image. It may be possible to overcome this by reducing the light output of the brighter projector (the one equipped with the filter with the lower light loss) and by correcting the color balance to provide a better match between the two projected images. Positive or Negative ? – The Omega 3D system may not achieve your goals if you are looking to a dual projector passive 3D setup as the means to getting a much brighter 3D image as compared to using a single projector, active 3D projector setup. This will likely be the case unless you use a pair of 2D projectors with significantly higher lumens output than would be found in consumer-class active 3D projectors operated in active 3D mode. The Omega 3D filters plus 3D glasses will result in a image that is approx. 20% as bright as a 2D image from the same projector. This may be similar efficiency to what is achieved with a typical consumer-class 3D projector using active shutter 3D glasses. However, with those 3D ready projectors where their active 3D mode has lower total light output than the same projector’s brightest acceptable 2D mode, a somewhat brighter image may be obtain by using two of these projectors, each operating in 2D mode, as part of a passive 3D system with the Omega 3D filters. Conclusions – – Let me first say that I have seen this system of filters and glasses used by Runco with their Model D-113d high-end dual projector passive 3D system. These are 3-chip DLP projectors using two high power lamps in each projector providing very high lumens output. The total cost of this Runco system as demonstrated was over $200,000 and is intended for use with very large screens (e.g., 300 to 400 inches). During the few minutes during which I viewed the Runco demo of their system virtually no 3D crosstalk/ghosting was visible, the overall color balance seemed good and the 3D image was very bright. This would indicate that with the right projectors and setup the Omega 3D/Panavision system is capable of good performance. – Using projectors with a lens that supports longer throw distances and installing the projectors at a location that provides a long throw ratio will provide more uniform colors in the projected image when the Omega 3D filters are installed. – The results when using the Omega 3D filter system with typical consumer-class projectors equipped with UHP lamps may have issues, as noted above, that some home theater owners may find objectionable. Ultimately, the ability to select compatible projectors will decide on the success of creating a dual projector passive 3D system that uses the Omega 3D system. At the current time it appears that DLP projectors with relative high lumens output are a better match, in terms of the projected color spectrum, for the Omega 3D system than are the typical LCD and LCoS based consumer projectors. Better still are projectors having Xenon lamps rather than the lower cost UHP lamps used in most consumer-class projectors. However, such projectors with Xenon lamps are out of the price range for most home theater owners. _____________________________________ For my upcoming blogs I will post some final thoughts on passive 3D projection and then move on to some new topics.