Jörg Reitterer | TriLite Technologies: How can a panel-less projector double luminance and resolution purely in software?
05:53 - 07:27
Other snippets from this talk
Summary of the clip:
How can a panel-less projector double luminance and resolution purely in software?
With no panel, there is no physical pixel grid. The pixel grid is defined in software, enabling dynamic field-of-view scaling. A maximum field of view is set by the MEMS mirror scanning angle, and it can be scaled down dynamically based on content. Launching the same optical power or luminous flux into the system then yields higher luminance at the output.
Resolution at the refresh rate can also increase because the scanning pattern becomes denser. Currently this works on one axis: scaling by 50% gives two times the luminance, dynamically depending on displayed content. The next generation is doing this in the second dimension too. Planned products are to implement what is called phobiation.
Phobiation dynamically changes the scanning pattern, again purely by software, to make it denser in a certain region of interest. A region of interest can be selected via software, increasing brightness, refresh rate, and resolution dynamically in that area. A very big misconception that always comes up is that since lasers are used, some kind of speckle artifacts must be seen.
In this short video, you can learn:
* Without a panel, the pixel grid is software-defined, so field of view can be scaled dynamically.
* Scaling by 50% on one axis doubles luminance and densifies the scan pattern to raise resolution at refresh rate.
* Phobiation is planned to concentrate a denser scan pattern on a software-selected region of interest, boosting brightness, refresh rate, and resolution there.
📋 **Clip Abstract** Laser beam scanning removes the physical pixel grid, replacing it with a software-defined grid. Dynamic field-of-view scaling, one-axis 50% scaling for two times luminance, second-dimension work, and planned phobiation are ways to trade field of view for luminance and resolution at refresh rate.
About the speaker:
* Speaker: Jörg Reitterer
* Company: TriLite Technologies
* Event: Eindhoven 2026
* Location: High Tech Campus, Eindhoven
#LaserBeamScanning, #MEMS, #SoftwareDefinedPixelGrid, #DynamicFieldOfView, #ARDisplays, #LightEngines
This is a highlight of the presentation:
Beyond mLED and LCoS: Laser beam scanning for high-efficiency AR displays
More Highlights from the same talk.
02:56 - 04:59
Why do laser light engines reach 5x the wall-plug efficiency of LED systems and 10x versus microLED?
Why do laser light engines reach 5x the wall-plug efficiency of LED systems and 10x versus microLED?
LEDs in LCoS and microLED systems emit by spontaneous emission: broadband, unpolarized and with large étendue, where the product of emission surface and emission angle is very large. Lasers, specifically semimode lasers, emit by stimulated emission. That yields narrow-band, inherently polarized light with small étendue: the beam's emission size on the chip is very small and emission angles are much more confined than LED sources. This intrinsically allows much smaller light sources.
Wall-plug efficiency is critical here. Compared with an LED system, laser engines typically deliver a 5x wall-plug efficiency gain; compared with microLEDs, it is more like 10x, because microLEDs are inherently less efficient. That ignores polarization. Accounting for polarization adds another 2x factor on top. Polarized light can be exploited in many combiner architectures, and some waveguide architectures are more efficient when a certain polarization is coupled in at the input.
Color gamut also separates the sources. Typical high-end computer monitors are barely above 100% sRGB. With laser light you can get well above 200% sRGB, so colors appear much richer and more vivid compared to LED light sources. Combined with intrinsically smaller light sources and the polarization advantage, this matters for next-gen architectures.
In this short video, you can learn:
* LEDs emit by spontaneous emission, giving broadband, unpolarized light with large étendue, whereas semimode lasers emit by stimulated emission with narrow-band, polarized output and small étendue.
* Laser light engines reach about 5x wall-plug efficiency versus LED systems and roughly 10x versus microLEDs, with polarization adding a further 2x factor.
* Laser light gives well above 200% sRGB color gamut, compared with barely above 100% for typical high-end computer monitors.
📋 **Clip Abstract** The clip contrasts LED and laser light sources for displays: spontaneous emission versus stimulated emission, large versus small étendue, and unpolarized versus inherently polarized output. It quantifies laser advantages at 5x wall-plug efficiency versus LED systems, 10x versus microLEDs, an extra 2x from polarization, and over 200% sRGB color gamut versus barely above 100% for high-end monitors.
About the speaker:
* Speaker: Jörg Reitterer
* Company: TriLite Technologies
* Event: Eindhoven 2026
* Location: High Tech Campus, Eindhoven
#StimulatedEmission, #WallPlugEfficiency, #SemimodeLasers, #sRGBGamut, #ARDisplays, #DisplayLightEngines
19:10 - 20:15
Why must waveguides and light engines be co-optimised for laser light rather than bought separately?
Why must waveguides and light engines be co-optimised for laser light rather than bought separately?
Customers are only interested in system solutions; a great display without a combiner does not make sense. That is why the light engine and waveguide must be co-developed, not bought separately. Working with waveguide companies optimizes designs for laser light, and participation in the AR Alliance laser display for AR working group aims to jointly advance laser display, including waveguides optimized for that.
If a waveguide is not designed for laser systems and laser light is coupled in, the performance is typically not optimized. This mismatch is not a minor detail: it affects the complete system the customer sees. Co-optimizing the combiner and the light engine together is therefore essential. There is confidence this can be resolved, but only through joint design rather than separate component purchases.
Laser light has specific properties that demand a waveguide designed for it. A generic combiner cannot deliver the same result. The co-optimization is not a one-way street: the light engine and combiner must always be tuned as a pair. This is why separate purchasing fails in practice and why collaborative design with waveguide companies matters.
In this short video, you can learn:
* Customers are only interested in system solutions, so a great display without a combiner does not make sense.
* If a waveguide is not designed for laser systems, coupling in laser light typically does not give optimized performance.
* Co-optimizing the combiner and the light engine together is essential, and the AR Alliance laser display for AR working group aims to jointly advance laser display with waveguides optimized for that.
📋 **Clip Abstract** Customers want system solutions, not separate components, because a great display without a combiner does not make sense. Waveguides not designed for laser systems perform poorly when laser light is coupled in, so co-optimizing the combiner and light engine is essential, with confidence this can be resolved through joint work such as the AR Alliance laser display for AR working group.
About the speaker:
* Speaker: Jörg Reitterer
* Company: TriLite Technologies
* Event: Eindhoven 2026
* Location: High Tech Campus, Eindhoven
#LaserWaveguide, #CombinerOptimization, #LightEngine, #LaserDisplay, #ARDisplays, #WaveguideOptics
11:02 - 12:30
Does a laser beam scanner really beat an X-Cube microLED engine on power at 20-30% APL?
Does a laser beam scanner really beat an X-Cube microLED engine on power at 20-30% APL?
For brightness-critical applications, the LBS engine can use multi-mode laser diodes or multiple single-mode lasers to boost output. The current generation LBS system, called the trixel free cube, measures 0.8 cc, delivers a 40-degree field of view, and reaches up to 15 lumens at the output. Asked whether a light engine can achieve 30 lumens, the answer is yes, using higher-output laser diodes.
For combiners, high-power laser diodes are typically not required, but if an architecture needs more light, using lasers is not the issue. The power comparison is against a state-of-the-art X-Cube micro LED architecture; the red dots are X-Cube power figures for different APL values. That data is not ours; it is from literature, and lab measurements reached pretty much the same conclusion.
Those X-Cube numbers are real measurements from the system, including the panel and driving electronics. The LBS system measurement includes the lasers, the mirror, and all driving electronics. Even today, LBS is much more competitive on power consumption. For typical sparse AR content at 20-30% APL, the system goes well below 300 milliwatts.
In this short video, you can learn:
* The current-generation LBS system, called the trixel free cube, is 0.8 cc, offers a 40-degree field of view, and outputs up to 15 lumens.
* A 30-lumen light-engine output is achievable with higher-output laser diodes, and architectures needing more light can use lasers without issue.
* For sparse AR content at 20-30% APL, the LBS system consumes below 300 milliwatts compared with X-Cube micro LED power figures that match literature and lab measurements.
📋 **Clip Abstract** This clip compares a laser beam scanning light engine with an X-Cube micro LED architecture on power consumption. It reports the trixel free cube at 0.8 cc, 40-degree field of view, up to 15 lumens, a 30-lumen capability with higher-output lasers, and below 300 milliwatts for 20-30% APL sparse AR content.
About the speaker:
* Speaker: Jörg Reitterer
* Company: TriLite Technologies
* Event: Eindhoven 2026
* Location: High Tech Campus, Eindhoven
#LaserBeamScanning, #MultiModeLaserDiodes, #XCubeMicroLED, #SparseContentAR, #ARDisplays, #LightEngines