Apple · Filed Feb 17, 2026 · Published Oct 29, 2026

Apple Files Patent for a Tighter Dot Layout That Captures More Depth Detail

Apple's latest optics filing swaps the square grid of microscopic pillars that splits laser light for a honeycomb grid. The result is about 15% more pillars in the same space, and more of the light lands where it's supposed to.

A light sensor projects an array of beams onto an object so a device can map its 3D shape. Drawing from patent filing US 2026/0339005 A1.
A light sensor projects an array of beams onto an object so a device can map its 3D shape.
See all 16 drawings from this filing ↓
Publication number US 2026/0339005 A1 US20260339005A1
Applicant APPLE INC.
Filing date Feb 17, 2026
Publication date Oct 29, 2026
Inventors Maoz Ovadia, Moshe Kriman, Aharon R. Barnea
US classification 359/569
Status when we published Waiting for an examiner (Mar 31, 2026)
Parent application Claims priority from a provisional application 63795449 (filed 2025-04-27)
Document 20 claims

What Apple's honeycomb light-splitter does

Every time a phone or headset maps the room in front of it, it fires out a grid of tiny invisible dots and watches how they land. Those dots start as a single laser beam that has to be split into many, and any light that strays off course is wasted.

The splitting is done by a flat, thin layer covered in microscopic pillars, each smaller than a wavelength of the light itself. Apple's filing arranges those pillars in a honeycomb pattern instead of the usual square grid. A honeycomb fits about 15% more pillars into the same space, and more pillars means more of the light ends up in the dots you actually want.

The pillars still vary in width in a repeating pattern, so the dots still come out as a neat rectangular grid. Only the layout underneath changes.

From the filing · CLAIM 1
… an array of pillars, which are arranged in a hexagonal lattice having a pitch no greater than half the target wavelength, and have varying respective diameters that are chosen to define a periodic pattern over a plane of the substrate having a unit cell larger than the target wavelength.

Translation: Tiny pillars are packed tightly together in a specific honeycomb pattern to bend light in precise ways.

How hexagon-packed pillars split one laser beam

The device is a piece of glass (the substrate) topped with a metasurface, a thin layer of pillars that bend light. How wide each pillar is sets how much it delays the light passing through, and the right pattern of delays can split one beam into many.

The claims cover pillars sitting in a hexagonal lattice (a honeycomb arrangement) with a spacing no greater than half the light's wavelength. The pillar widths repeat in a tile, called a unit cell, that is larger than the wavelength. The tile can be a hexagon, a parallelogram or a rectangle. In one example, an 8 by 8 tile about 3 micrometers across splits a beam into a 5 by 5 fan, and 20 by 20 of those tiles make a 60 micrometer square element.

The filing also describes layering a second, non-repeating pattern on top, such as a quadratic (bowl-shaped) phase, so the same surface can focus or straighten the split beams. Other claims cover a version with laser emitters attached and a method for making the layer.

For the numbers, the description says a honeycomb is 15% denser than a square grid at the same spacing. In computed tests, efficiency rose by more than 14% at very steep bending angles, and by about half that at 45 degrees.

What denser pillars could mean for Apple sensors

If you own a phone, headset or smart glasses that scans the space around you in 3D, the light source inside is small and has a limited power budget. A layer that wastes less light means the same laser can do more, which could help with battery life or sensor range. The filing doesn't promise either, but it's the direction the efficiency numbers point.

The design also stays within standard chip-style manufacturing, with spacing the filing calls readily achievable (380 to 470 nanometers) for 940 nm infrared light. That keeps the idea practical, since it only changes where the pillars go.

This is the 60th Apple filing we've tracked since May in the AR glasses race, joining earlier applications on brightening lenses with overlapping patterns and frame bending detection.

Editorial take

You will never see this change, and that's the point. The benefit shows up as a 3D sensor that wastes less of its own light, which could translate into a slightly longer-lasting battery or a dot pattern that stays usable a bit farther out.

The gain is modest and the filing says so: a 15% denser layout, and efficiency up by roughly 14% at the most extreme bending angles. At the more moderate angles it's about half of that. Those are computed results, not measurements from a built device.

Still, it's a tidy trick. Changing only the layout, with no new materials or manufacturing steps, is the easiest kind of upgrade for a device maker to adopt, and it could make face and room scanning a little more efficient.

Get our take in your Top Stories

Liked this breakdown? Add Patentlyze as a preferred source on Google, and our plain-English take shows up more often in your Top Stories the next time Apple patent news breaks.

There are more where this came from

We read every patent application Big Tech publishes and send you the ones worth knowing. Plain English, free, every week.

The drawings

16 drawing sheets from US 2026/0339005 A1 · click any drawing to enlarge

How tiny surface pillars split light to project dot patterns, drawing sheet 1 from US 2026/0339005 A1
Sheet 1 How tiny surface pillars split light to project dot patterns
Two honeycomb arrangements for packing light-steering pillars, drawing sheet 2 from US 2026/0339005 A1
Sheet 2 Two honeycomb arrangements for packing light-steering pillars
A rectangular repeating group on the first honeycomb layout, drawing sheet 3 from US 2026/0339005 A1
Sheet 3 A rectangular repeating group on the first honeycomb layout
Slanted boundaries grouping pillars across the first layout, drawing sheet 4 from US 2026/0339005 A1
Sheet 4 Slanted boundaries grouping pillars across the first layout
Angled repeating groups arranged along tilted honeycomb axes, drawing sheet 5 from US 2026/0339005 A1
Sheet 5 Angled repeating groups arranged along tilted honeycomb axes
Hexagonal groupings that tile together to direct beams, drawing sheet 6 from US 2026/0339005 A1
Sheet 6 Hexagonal groupings that tile together to direct beams
Rectangular pillar blocks applied to the second layout, drawing sheet 7 from US 2026/0339005 A1
Sheet 7 Rectangular pillar blocks applied to the second layout
Flattened diamond groupings on the second honeycomb pattern, drawing sheet 8 from US 2026/0339005 A1
Sheet 8 Flattened diamond groupings on the second honeycomb pattern
Show the other 8 sheets
A diamond shaped grouping of hexagonal cells, drawing sheet 9 from US 2026/0339005 A1
Sheet 9 A diamond shaped grouping of hexagonal cells
An alternative large hexagonal unit cell layout, drawing sheet 10 from US 2026/0339005 A1
Sheet 10 An alternative large hexagonal unit cell layout
Phase shifts mapped across a tiny beam splitting block, drawing sheet 11 from US 2026/0339005 A1
Sheet 11 Phase shifts mapped across a tiny beam splitting block
Repeated unit cells forming a complete light splitter, drawing sheet 12 from US 2026/0339005 A1
Sheet 12 Repeated unit cells forming a complete light splitter
Adding curved phase shifts to bend projected beams, drawing sheet 13 from US 2026/0339005 A1
Sheet 13 Adding curved phase shifts to bend projected beams
Higher order phase shaping across the optical surface, drawing sheet 14 from US 2026/0339005 A1
Sheet 14 Higher order phase shaping across the optical surface
Comparing light efficiency between hexagonal and square grids, drawing sheet 15 from US 2026/0339005 A1
Sheet 15 Comparing light efficiency between hexagonal and square grids
Hexagonal efficiency gains across wider projection angles, drawing sheet 16 from US 2026/0339005 A1
Sheet 16 Hexagonal efficiency gains across wider projection angles
Browse all the drawings 16 drawing sheets
Patent filing page

Source. Full patent text and figures from the official USPTO publication PDF.
Reader comments

Be the first to weigh in

Start the discussion

Real name or a handle, either is fine. Comments are read by a person before they appear, so allow a little time. Keep it about the filing.