[0001] This disclosure relates generally to cinema display technology, and more particularly to enhancing the spatial frequency response of emissive displays.

BACKGROUND

[0002] Cinema display technology may include emissive displays, such as Light Emitting Diode (LED) displays and Organic LED (OLED) displays. Such displays may be more energy efficient than a comparable projector because current is only supplied to the pixels that are meant to be illuminated. Additionally, the emitting element need not take up the entire display surface. The remainder of the display may then be covered with light absorbing material, thus decreasing the influence of ambient light.

[0003] An emissive display may be enabled for stereoscopic 3D viewing by using active shutter eyewear that synchronizes with the display. An emissive display may be enabled for stereoscopic 3D viewing using passive eyewear by applying set of polarization filters to the individual pixels. The cost of the display is dependent on the number of pixels used.

BRIEF SUMMARY

[0004] Disclosed herein are embodiments of a method for emitting light from a display. The display may have a plurality of emitters configured with a pitch. Light emitted from an emitter is collected, split into one or more light paths, and guided along the one or more light paths to one or more exit ports. The light from the emitter exits the display at the one or more exit ports. The exit ports are configured with a smaller pitch than the pitch of the emitters. A display may have a plurality of emitters and associated exit ports for left-eye viewing, and a separate plurality of emitters and associated exit ports for right-eye viewing. [0005] Also disclosed herein are embodiments of a light guide structure having a cavity configured for optical communication with an emitter on a display. The light guide structure also has one or more exit ports associated with the cavity and configured to emit light from the display. The light guide structure also has a light-guide channel associated with the cavity and configured to guide light from the emitter to the exit ports. In some embodiments, the light guide structure comprises a single piece. In some embodiments, the light guide structure comprises multiple pieces.

[0006] Embodiments are illustrated by way of example in the accompanying figures, in which like reference numbers indicate similar parts, and in which:

[0007] FIGURE 1 illustrates the screen-door effect in a view of an example emissive display;

[0008] FIGURE 2 illustrates an example tiled emissive display;

[0009] FIGURE 3 illustrates a side view of an example emissive display with a diffuser;

[0010] FIGURE 4 illustrates a side view of an example emissive display with a tiled diffuser;

[0011] FIGURE 5 illustrates a side view of an example emissive display with an active retarder;

[0012] FIGURE 6 illustrates a side view of an example emissive display with a patterned retarder;

[0013] FIGURE 7 illustrates a front view of an example emissive display with a patterned retarder,

[0014] FIGURE 8 illustrates pixel cross-talk in an example emissive display;

[0015] FIGURE 9 illustrates light-guide channels in an example emissive display,

[0016] FIGURE 10 illustrates light shaping within a light-guide channel in an example emissive display,

[0017] FIGURE 11 illustrates an example light-guide channel creating a two-dimensional pixel; [0018] FIGURE 12 illustrates an array of example light-guide channels suitable for a 2D display wall or a 3D display utilizing active shutter eyewear;

[0019] FIGURE 13 illustrates an array of nested example light-guide channels suitable for a 3D display utilizing passive eyewear,

[0020] FIGURE 14 illustrates an example process for attaching patterned retarder film (FPR) over a light guide structure;

[0021] FIGURE 15 illustrates components of a two-part molding assembly for an example light guide structure;

[0022] FIGURE 16 illustrates components of a multi-part, molding assembly for an example light guide staicture;

[0023] FIGURE 17 illustrates an example light guide structure with pixel cavities, and

[0024] FIGURE 18 illustrates example LED emitters with lens-caps.

DETAILED DESCRIPTION

[0025] As the quality of home theater and the availability and popularity of streaming and live content increase, cinema must provide an enhanced experience in order to remain competitive. High dynamic range, increased brightness, color gamut, object based sound, and improved stereo 3D can all contribute to this enhanced experience. Projector manufacturers have been improving the performance of their products, but projectors are ultimately limited in both brightness and dynamic range. Therefore there is a need for a new cinema display technology.

[0026] The primary display performance improvement candidates are brightness, frame rate, high dynamic range (HDR), color gamut, and resolution. Frame rate, color gamut, and resolution can, in principle, all be addressed with projection technology. Existing digital cinema projectors are mostly capable of at least 48 frames per second and newer versions should be able to display content at 60-120 frames per second. In fact, the bottleneck for increased framerate is limited acceptance by the content creators. Laser projection enables an increase in color gamut up to near that of Rec. 2020. Again, there is limited push from the creative community for increased color gamut. Similarly, newer projectors are capable of 4k resolution yet there is very little call for higher resolution and plenty of debate as to whether higher resolution would even be noticeable.

[0027] In contrast, projectors are fundamentally limited with respect to brightness and dynamic range, DLP projectors are limited to around 60000 lumens of brightness which enables 2D brightness of 50 nits on a 36m wide 2.2 gain screen. This is indeed large; however, in 3D this brightness drops to 14 nits when using a best-in-class polarization encoding system. If the goal is to produce 50 nits of brightness in 3D, then the maximum size drops to 19m. Additional brightness can be achieved by adding projectors; however, the complexity of the system more than doubles because now the two or more projectors must be aligned to each other.

[0028] The dynamic range of the brightness displayed by projectors is also limited. Lamp based projectors typically have a sequential contrast of less than 2000: 1. The first generation of laser projectors has contrast as high as 3000: 1, and the newer generation will achieve as much as 6000: 1. However, the higher contrast necessarily results in a lower total brightness projector. More importantly, the fundamental nature of projection conflicts with the requirements for truly high contrast, in other words, the projection screen will reflect any light back to the audience. Problematically, the actual projection light displayed on one part of the screen reflects off of the room and/or the audience and back onto the screen, potentially destroying the contrast in other parts of the image. Additionally, building codes require a minimum safe illumination level of egress pathways, and this light fundamentally limits the achievable contrast.

[002 1 The concept of high dynamic range is not limited to low dark levels. The term also applies to extremely high peak brightness levels. These peak values would typically only be used to better display features such as specular reflections off of shiny surfaces. Such peak values would typically only occur in a very small percentage of the image area in any scene. In order to produce locally very bright images, a projector needs to be supplying the same potential brightness across the entire surface. Even if that were possible (for instance by using multiple projectors), the vast majority of the light, and the energy to produce it, would be wasted.

[0030] Some solutions to the problems of increased brightness and dynamic range utilize an emissive display such as a Light Emitting Diode (LED) display or Organic LED (OLED) display. These displays have the potential to be significantly more energy efficient than a comparable projector because current is only supplied to the pixels that are meant to be illuminated. Consequently, power is not wasted during the majority of scenes (and even within a scene) which do not require 100% brightness. Additionally, the emitting element need not take up the entire display surface. For example, for a typical LED display with a pixel pitch of 2.0mm, the LED diodes and (reflective) connection pads and wires take up less than several hundred square microns. The rest of the display surface can then be covered with light absorbing material, typically black ink or paint. In this way, the influence of ambient light may be decreased by as much as two orders of magnitude.

[0031] A simple way to enable an emissive display for stereoscopic viewing is to use active shutter eyewear that synchronizes with the display. However, this business model has proven problematic due to the cost and maintenance of the eyewear. Inexpensive and recyclable passive eyewear may be preferable in a cinema environment. If passive eyewear is required, then emissive displays may be fitted for stereoscopic viewing by applying a set of polarization filters to the individual pixels. This has already been done for LCD televisions by patterning adjacent rows of the display with plus and minus 45 degree alternately rotated retarders. In this way inexpensive passive 3D eyewear may be used for stereoscopic viewing. [0032] The present disclosure addresses issues with the deployment of emissive displays with discrete pixels composed of isolated emitting elements that are physically well-spaced. This structure is characteristic of current LED video wall technology, but these solutions may apply to any geometrically similar display technology. Within this disclosure, the term "emitting element ^' " may be used interchangeably with the terms "emission element", "emitter", "pixel", "emitting pixel", "emission pixel," etc.

[0033] Adapting emissive technology for a cinema environment introduces a new combination of technical challenges. Specifically, the cost of the display (excluding the cost of the physical structure) has a nearly linear dependence upon the number of pixels used, it is dominated by the serial physical processes of placing the emitting structures. Therefore, the ideal display will use the absolute minimum number of physical pixels to achieve the target resolution. For current cinema content, this would be either 2k or 4k (2.21M pixels or 8.85M pixels) for 2D. Unlike with projection, for which a lens can "zoom" the image size to fill the available area (in other words, the pixel pitch is variable), an emitting display must be manufactured with a fixed pitch. In order to build a larger display, an LED supplier would need to either design a new system with a larger pitch, or simply build the larger wall with additional pixels at the same pitch. The first solution is problematic because most installations would be custom builds of the LED units. The second solution is problematic because only the minimum size display makes use of the optimum number of pixels; larger displays would have higher resolution and so the content resolution would generally need to be up-scaled.

[0034] Additionally, the matrix of space between emitting elements may be visible to viewers that come relatively close to the display. This effect is referred to as the "screen-door" effect because it looks as though the display is being viewed through a screen door. If the display is used in a passive 3D mode, then the screen-door effect is exacerbated because each individual eye is only viewing half of the pixels and so the relative dark area is increased. Projectors also have a screen-door effect, but the much larger fill factor of projector pixels combined with the blur induced by the non-perfect projection optics combine to minimize it in most cases.

[0035] Some embodiments in this disclosure produce a light guide that collects the light emitted by each pixel and then splits it into multiple spatially separated emission locations. Within this disclosure, an emission location may also be referred to as an emission point, an emission port, an exit location, an exit point, and an exit port. These emission locations may be placed on a regular grid such that the new grid has higher spatial frequency than the underlying emission pixels. Depending upon the cost of manufacturing individual emission pixels relative to the cost of manufacturing the light guide structure, the splitting could be into as few as two emission locations. Consequently, in some embodiments, the light guide structure may only increase the spatial frequency in one dimension.

[0036] Figure 1 illustrates the screen-door effect in a view of an emissive display 100. For simplicity, the individual emitters (e.g., emitter 1 10) have been schematically illustrated as three adjacent and touching rectangles. This illustration is useful to indicate pixels composed of individual red, green, and blue emitters; however, it should be understood that any arrangement of subpixels may be grouped together including but not limited to RGB + yellow, monochromatic, etc. The illustration of Figure 1 ignores the detailed arrangement of emitters on the surface which may be arranged in any configuration including rows, columns, triangles, etc. Some embodiments may use vertically stacked emitters.

[0037] In Figure 1, the horizontal pixel repeat pitch 120 and the vertical pixel repeat pitch 130 are labeled on a section of a display 100. These dimensions define the unit cell of the image pixel. In the illustration, the emitting area 140 is placed in the top left corner of the unit cell 150, but this location is arbitrar'. The emitting area has height 60 and width 170. Any arrangement of the emitting area 140 within the unit cell 150 is valid as long as it is regularly repeated across the surface of the display 100.

[0038] In order to construct a cinema-sized display wail using emissive technology, it may be necessary to build the display up from a collection of tiles. For this purpose, the empty space between emitters may actually be an advantage. If the pixel were instead 100% filled by the emitting structure, then the emitting area would need to extend exactly to the edge of the tile. Otherwise a dark line would be visible at the tile boundaries. This can be technically very challenging as the emitting structure may be delicate and therefore may be damaged by any cutting, trimming, or handling of the tiles. Even if it were possible to extend the emitting area to the tile edge, the empty area surrounding each emitter may allow tolerance in aligning adjacent tiles 210 as shown in Figure 2. Typically, the empty area in the pixel is colored black in order to minimize the impact of ambient light.

[0039] When the viewer is sufficiently far from the display, the point spread function of the eye blurs the emitters such that they overlap and so the display appears smooth. As the viewer approaches the display, they will begin to resolve the independent pixels at a distance where the angle subtended by the two pixels is on the order of 1-2 arcminutes. At this point, the black region between emitters may begin to be visible and may contribute to a screen-door effect- Therefore, for highest performance it would be preferable to ensure that viewers do not approach closer than this distance.

[0040] As an example, consider a "4k" cinema display composed of 4mm pixels. Such a display would be 16.4m wide and 8.6m tall. In most auditoriums, the distance between the screen and the first row of seats is approximately half of the screen height (4.3m in this case). A more conseivative practice would be to place viewers no closer than one screen height, as this ensures better viewer comfort due to gaze direction, neck strain, and image distortion. The 4mm pixels subtend 1 arcminute for a viewer at a distance of 13.8m. Therefore, even if the viewers are seated at a conservative distance, the screen-door effect may be observed for viewers seated in the first few rows (at distances of 8.6m to 13.8m).

[0041] One solution to this problem is to use smaller pixels. If the same display were constructed using 2mm pixels, then the critical distance for observing the screen-door effect would be only 6.9m. If 1 ,33mm pixels were used, then the critical distance would be 4.6m, which is approximately the same as half the screen height. Unfortunately, the costs of the display that are dominated by the serial aspects of manufacturing (pick and place of pixels) increase as the square of the number of pixels. Therefore, halving the pixel pitch from 4mm to 2mm increases the pixel costs by 4x. In order to achieve a viewing distance of approximately half the screen height, the cost must increase by 9x.

[0042] Another solution to this problem is to place a diffuser 3 10 over the emissive display surface 320. As shown in Figure 3, a diffuser 310 can increase the apparent size of the emitting area for viewers at a distance and reduce or eliminate the screen-door effect. There may be many possible choices for diffuser material including bulk diffusers, surface diffusers, and a combination of both bulk and surface diffusers. Furthermore, there are design principles that should be optimized in order to maximize the image uniformity without sacrificing resolution. However, at the tile junctions a diffuser may cause significant problems. A tile boundary 410 is illustrated in Figure 4. Rays that cross the tile boundary are likely to be disrupted with the effect being visible to some viewers. A solution to this problem has been deployed and consists of a single diffuser film that is stretched over the entire front of the display. While this can produce excellent performance in some cases, it becomes increasingly difficult for large walls. It is also largely incompatible with polarization based 3D solutions as discussed below.

[0043] New challenges appear when the display is adapted for stereoscopic 3D. One method to accomplish this technically is to use active shutter-glass eyewear synchronized to the display. In this method, the only additional requirement for the display is that it must be capable of displaying content at a sufficiently high frame-rate: preferably at least 120 Hz. The drawback is the large relative cost of the eyewear and the need for the exhibitor to manage eyewear cleaning, charging, and collection.

[0044] A passive solution may be preferred due to the much lower cost of the passive eyewear and passive eyewear management. A first passive eyewear solution is to employ a polarizer 510 and an active liquid crystal retarder as shown in Figure 5. An example of such a system is a switchable half-wave retarder 520 in combination with a quarter wave retarder 530 as exemplified by the RealD ¹ Monitor Z Screen ¹ . In this way it is possible to actively switch the polarized output of the display between right and left circular polarization. The active component must necessarily be tiled over the surface of the display and may suffer from the same boundary effects illustrated in Figure 4 for diffusers. Furthermore, many readily available diffusing components may be incompatible with preserving polarization of light and may need to be located prior to the polarizer 510 and active retarder,

[0045] A second passive eyewear solution is to apply a linear polarizer 510 combined with a patterned quarter-wave retarder 610 to the front of the display, as i llustrated in Figure 6. There are also technologies available to pattern the linear polarizer orientation rather than the quarter wave orientation, and the pixel geometry requirements for this arrangement are the same. In this configuration, alternating pixels are permanently dedicated to display each eye when the display is used in stereo 3D mode. Consequently, twice as many physical pixels are required in order to display a full resolution image to each eye. Two possible arrangements of pixels are shown in Figure 7. The pixels may be arranged in a column striped configuration 710, in a row striped configuration (not shown), or in a checkerboard configuration 720. When FPR Stereo 3D films are deployed on LCD panels, the row arrangement may be chosen primarily to fix cross-talk issues caused by parallax. The horizontal field of view may be more important for a television situation and so it may be better to confine cross-talk to the vertical viewing direction which can be easier to control. In this case horizontal rows may be a better solution.

[0046] Unfortunately, this solution may be even more difficult to implement in the presence of a diffuser. If the diffuser is placed after the polarizer then it must be carefully chosen to preserve polarization. If the diffuser is placed prior to the polarizer, then only limited diffusion is possible because the light from neighboring pixels must remain spatially separate prior to traversing the patterned retarder. A patterned retarder solution can also cause problems with the perception of the screen-door effect. When the display is viewed in 3D mode, then the spatial frequency of the pixelation, as viewed by either eye, is decreased by either a factor of two in one dimension (for striped retarder 710) or uniformly by a factor of the square root of two (for checkerboard retarder 720).

[0047] Due to the relatively high index of refraction of inorganic LED materials (die), internal reflections may limit the amount of light that eventually exits the die. It may often be necessary to encase the die in an intermediate dielectric material such as epoxy or silicone in order to help couple light out of the emitter. For LED emitters that are individually mounted to form discrete pixels (surface mounted device (SMD) technology), each emitter is individually pre-encapsulated. For LED emitters that are bonded directly to the final module board (chip on board (COB) technology), the generally-used method is to mold a protective shell over the entire module board.

[0048] In order to achieve high stereo contrast (discrimination of left and right-eye images) and thus high quality stereo viewing, the pixels must be sufficiently optically isolated from each other. If the emitting pixel structure is encased within a protective (or optically functional) dielectric material 810 such as epoxy or silicone, then light may totally internally reflect (TIR) within the dielectric 810 and thus travel from a right-eye pixel 820 to a left-eye pixel 830. It may then reflect, scatter, or othenvise couple out of the structure from the physical location of the left-eye pixel 830 and thus incorrectly code for left-eye viewing as shown in Figure 8. [0049] The present disclosure provides solutions to the screen-door effect problem by increasing the spatial frequency of the luminance without increasing the number of pixels. These solutions may be operable both for 2D-oniy displays as well as 3D displays. Some embodiments create a light guide structure that collects the light from each pixel and then splits it into multiple emission locations.

[0050] The fundamental cause of the screen-door effect is an insufficiently high spatial frequency of the emitters. Therefore, some embodiments collect the light from one emitter and distribute it through a light guide so that the light emanates from two or more locations. Consider the one-dimensional example illustrated in Figure 9. The emitters on the display, for example pixels 970, have a regular spacing 910 of P. The light from each pixel is collected in the integrating region before splitting into two separate paths, for example a first light path 940 and a second light path 950. The exiting light 960 then exits from openings, for example emission ports 920, with a spacing of P/2. This doubling of the spatial frequency (halving of the pitch) effectively halves the minimum viewing distance of the display. The structure through which the light travels is referred to as a light-guide channel 930.

[0051] In order to achieve optimum performance, the exit ports 920 of the light-guide channel 930 should have a number of properties. The intensity radiated from each of the ports should be approximately equal. Preferably, the change in intensity between ports should be imperceptible to a viewer. For example, if all pixels are driven to equal brightness, then a viewer situated at a distance farther than the minimum viewing distance should not be able to perceive any local changes in brightness; in other words, the brightness changes should be less than one just noticeable difference (JND). In many cases, this will require less than 50% change in brightness between any port and preferably less than 10%. Similarly, the angular emission profile of each of the emission locations should be substantially the same within the required angular viewing region. In other words, for any desired viewing direction, the intensity change observed from each of the individual emission locations should be substantially less than a JND.

[0052] Ideally, the light-guide channel 930 should transmit as much of the emitted light as possible out of the emission locations/ports. If possible, the light-guide channel should employ TIR in order to confine and transmit the light. However this may not be possible for many geometries due to the large number of different reflection angles present. Therefore, the surfaces of the light-guide channel may need to be coated with a reflective material, for example silver, aluminum, or any other reflective material. The light-guide channel may also be coated with some form of diffuse or semi-specular white material. In 3D implementations, the white material may be sufficiently opaque to prevent pixel cross-talk as discussed below,

[0053] The primary purpose of the light-guide channel is to split singular sources into multiple (nominally equally spaced) sources. However, it can also incorporate other optical functions. The shape of the light-guide channel may be optimized to produce a desired angular distribution of the emitted light. For example, the exit ports may have a tapered profile 1010 in order to increase the head-on brightness of the display as illustrated in Figure 10. This shaping may be performed in either the vertical or horizontal direction or in some combination of the vertical and horizontal direction.

[0054] Each light-guide channel 930 may be individually molded and then bonded to the appropriate pixel 970, While this may enable a higher quality registration between each light- guide channel and its pixel, the process may be time consuming and expensive. Furthermore, the final structure may be unnecessarily fragile. Another choice may be to assemble all of the light- guide channels into a monolithic slab and bond that entire slab to the LED PCB substrate in one step.

[0055] The light guide structure may consist of two regions: 1) the light-guide channel 930 with functional properties as described above, and 2) the filler region with dual purposes of providing structural integrity and of providing optical isolation between left and right-eye pixels for 3D display. Therefore, an ideal optical channel would be optically clear or slightly diffuse. The channel could be completely empty, in other words, a void in the final structure. The filler region needs sufficient opacity to isolate the left-eye and right-eye pixels. Therefore, a black material may be a good choice. This has the added benefit of presenting an absorbing surface at the exit to minimize the effect of ambient light. Alternatively, a highly diffuse white material may be chosen, but the top surface would need to be colored black to the extent that ambient performance is important.

[0056] Having discussed doubling the spatial frequency of the emission in one direction, it may be straightfonvard to extend that concept to increases of three or more times the initial spatial frequency. As long as the individual physical components of the light-guide channels remain manufacturabie (both for physical yield and for cost reasons), the number of emission ports may be increased in order to enable any desired minimum viewing distance. Similarly, there may be no reason to confine the increase in spatial frequency to one dimension. Figure 1 illustrates a light-guide channel that doubles the spatial frequency in both the horizontal and vertical directions. For simplicity only the light-guide channel is shown and not the encasing structure. The spatial pitch between emission ports is P _h 12 in the horizontal direction and P _v 12 in the vertical direction where P,-, and P _v are the horizontal and vertical pitches of the underlying pixels. The physical shape in this illustration is meant to indicate the functional aspect of increasing the number of emission locations and is not meant to indicate the other optical functions of the light-guide channel.

[0057] Figure 12 illustrates an array of two-by-two light-guide channels that would be suitable for a 2D display wall or a 3D display utilizing active shutter eyewear. For clarity, only one of the light-guide channels is shown. However, the positions of the source pixels and emission ports are shown in the array structure. The unit repeat cell 1210 of the structure consists of one emitting pixel capped by one light-guide channel "tree" with four emission ports. This unit structure is tiled regularly across the display modules so that the pixels are regularly spaced with pitch P _h in the horizontal direction and P _v in the vertical direction. P _h and P _v may be equal. The emission ports are regularly spaced with pitch P _h 12 and P _v 12. This structure doubles the spatial frequency in both axes and therefore cuts the minimum acceptable viewing distance by half.

[0058] In order to enable 3D display using passive eyewear, some embodiments nest two light-guide channels (one for each eye) within the same unit repeat cell 1210 as illustrated in Figure 13, The unit repeat cell of the display consists of two emitting pixels each capped by a light-guide channel so that there are eight total emission points per tile (four for each eye). The full ensemble of left-eye and right-eye emission ports are separately located on grids with regular spacings of P _h in the horizontal direction and P _v in the vertical direction. Figure 13 illustrates the general case in which the physical offset between emission ports of the left-eye light guide 1310 and emission ports of the right-eye light guide 1320 is arbitrarily chosen. Some embodiments may offset the left-eye and right-eye emission ports by P _h /4 and P _v /4 in the horizontal and vertical directions, respectively. This places each right-eye emission port 1420 exactly in the center of the square formed by its neighboring four left-eye emission ports 1410. In 2D mode, if both sets of pixels are used, this has the benefit of either optimizing the brightness uniformity or maximizing the available additional resolution. However, in some cases the physical constmction of the light guide structure may prevent the obtainment of the ideal offset between sets of ports. 2D display functionality may also be obtained by only driving half of the pixels; in other words, driving only the left-eye pixels or only the right-eye pixels.

[0059] Figure 14 illustrates a process of attaching patterned retarder film for the left-eye 1330 and for the right eye 1340 over the top of the light guide structure. In this case a striped pattern is used to encode the light for left-eye and right-eye views, respectively. In some embodiments, a more complex retarder structure such as a checkerboard may be used.

[0060] There are numerous different methods to manufacture this structure. In some embodiments useful for rapid prototyping, the structure may be directly printed using stereo lithography techniques or 3D printing. If two optically distinct print resins are available then the structure may be fabricated in one step. However, if a reflective coating for the channels is required, then multiple steps may be required. Step one may be to print the structural filler region. Step two may be to coat the interior of the channels with reflective material (typically silver or aluminum but also possibly diffuse white ink or paint). Step three may be to then inject clear or slightly diffuse resin into the light-guide channels. Optionally the light-guide channels could be left empty.

[0061] In some embodiments useful for mass production, 3D printing techniques may be too slow or expensive to be practical. In this case, the structure may be formed by a molding process. Figure 15 illustrates the components of an example two-part molding assembly. The top piece 1510 of the mold defines most of the exit assembly. The bottom piece 1520 of the moid defines the cavities 1530 which enclose each emitting pixel. The assembly of these two pieces can be accomplished by filling the voids with a transparent or diffusely scattering plastic resin. This resin would then make up the light-guide channel. Reflective treatment such as aluminum or silver can be added to the top and bottom mold pieces separately prior to assembly. Alternatively, the top and bottom pieces may be adhesively joined and then coated with reflective material. This adhesive step may be accomplished using an adhesive resin or by heat- joining/welding the top and bottom pieces. The light-guide channels could then be filled using an optional final step.

[0062] Another embodiment is illustrated in Figure 16. The light guide structure may be composed of multiple layers 1610a, 1610b, 1610c to simplify the structure of each individual layer. These layers may then be adhesively laminated or heat welded to form the final structure 1620. This method enables multiple different slopes within the final assembly that may not be possible with only a two-part mold. It also enables a manufacturer to utilize cutting processes such as stamping or laser cutting to cut the holes for the light guide.

[0063] In order to attach the light guide assembly to the emitting display surface, it may be necessary to incorporate recesses within the light guide structure to contain the emitting pixel structure. Figure 17 illustrates a finished light guide structure with a cavity 1710 at the bottom of the light-guide channels. The bottom surface of this cavity may be textured in order to facilitate mixing of the light before it enters the multiple emitting ports. This structure may then be adhesively bonded to the substrate of the emitting pixels.

[0064] For COB LED designs, the emitting LED dies may require encapsulation in dielectric resin in order to enhance efficiency. However, coating the entire display board 1820 in dielectric may increase the cross-talk between adjacent pixels. Therefore, some embodiments that prevent this may mold individual dielectric structures (lens-caps) onto each pixel as shown in Figure 18. This structure may be a simple lens 1810 as illustrated or may have a more complicated or rough structure to more effectively diffuse the localized light sources. In both cases, the cavity in the light guide structure must be large enough to accept the dielectric lens-caps.

[0065] As may be used herein, the terms "substantially" and "approximately" provide an industry-accepted tolerance for its corresponding term and/or relativity between items. Such an industry-accepted tolerance ranges from zero to ten percent and corresponds to, but is not limited to, component values, angles, etc. Such relativity between items ranges between approximately zero percent to ten percent.

[0066] While various embodiments in accordance with the principles disclosed herein have been described above, it should be understood that they have been presented by way of example only, and not limitation. Thus, the breadth and scope of this disclosure should not be limited by any of the above-described exemplary embodiments, but should be defined only in accordance with any claims and their equivalents issuing from this disclosure. Furthermore, the above advantages and features are provided in described embodiments, but shall not limit the application of such issued claims to processes and structures accomplishing any or all of the above advantages.

[0067] Additionally, the section headings herein are provided for consistency with the suggestions under 37 CFR 1.77 or otherwise to provide organizational cues. These headings shall not limit or characterize the embodiment(s) set out in any claims that may issue from this disclosure. Specifically and by way of example, although the headings refer to a "Technical Field," the claims should not be limited by the language chosen under this heading to describe the so-called field. Further, a description of a technology in the "Background' ^' ' is not to be construed as an admission that certain technology is prior art to any embodiment(s) in this disclosure. Neither is the "Summary" to be considered as a characterization of the embodiment(s) set forth in issued claims. Furthermore, any reference in this disclosure to "invention" in the singular should not be used to argue that there is only a single point of novelty in this disclosure. Multiple embodiments may be set forth according to the limitations of the multiple claims issuing from this disclosure, and such claims accordingly define the embodiment(s), and their equivalents, that are protected thereby. In all instances, the scope of such claims shall be considered on their own merits in light of this disclosure, but should not be constrained by the headings set forth herein.