Pixel size requirements for AR/MR

For conventional projection displays, the sizes of the microdisplay and imaging lens aperture stop typically limit the optical throughput or etendue of the system. However, for AR/MR projection systems using diffractive pupil-replicating waveguide combiners, we examine how re-interaction with the in-coupler grating at one FOV extreme and pupil replication sparsity at the other FOV extreme severely restrict the optical throughput. Diffractive waveguides are the most commonly used AR/MR combiner technology due to their low cost, attractive form factor, and large achievable FOV. Nevertheless, we show that current and anticipated diffractive combiners only support etendues up to approximately 6.2 mm²sr, equivalent to no larger than a 0.34” display operating at f/2. At this display diagonal, maintaining 60 PPD for retinal resolution over a 50°-diagonal FOV requires a 3-μm pixel pitch. Future combiners with larger FOVs and higher resolutions will require even smaller pixels. Additionally, we find that optical engine volume varies quadratically with pixel pitch. Finally, we illustrate how the push for small pixels will intensify as the dominant array-based microdisplay technology used in AR/MR devices transitions from LCoS to inorganic microLED (micro-iLED). The emissive nature of micro-iLED displays offers exceptional potential power savings for sparse AR/MR content. Nonetheless, the broad angular emission and spatially multiplexed color subpixels of many proposed micro-iLED architectures further strain the bandwidth of the limited waveguide throughput. We demonstrate how CP Display has anticipated the requirement of 3-μm and smaller pixels in architecting its IntelliPix™ display platform.