During Snapdragon Summit 2022, Qualcomm Technologies, Inc. revealed the Snapdragon AR2 Gen 1 Platform, which delivers groundbreaking AR technology that will unlock a new generation of sleek, highly capable glasses. The Company built Snapdragon AR2 from the ground up to revolutionise the headworn glass form factor and usher in a new era of spatial computing experiences for the real-world/metaverse mix.
Purpose-built for AR
To help create the thinnest possible, high-performance AR glass, we have built a multi-chip distributed processing architecture combined with customised IP blocks. The main processor occupies a 40% smaller2 PCB area on glass and the overall platform delivers 2.5x better AI performance while consuming 50% less power to help achieve AR glasses that consume <1W power. This enables rich AR experiences on glasses that can be comfortably worn for extended periods of time and meet the demands of both consumers and enterprise use cases.
AR distributed processing architecture
To better balance the weight and decrease the arm width on either side of the glasses, Snapdragon AR2 uses a multi-chip architecture that includes an AR processor, AR co-processor and connectivity platform. Snapdragon AR2 dynamically works to distribute the processing of latency-sensitive perception data directly on the glasses and offloads more complex data processing requirements to a Snapdragon-powered smartphone, PC or to other compatible host devices.
- The AR processor is optimised for low motion-to-photon latency while supporting up to nine (9) concurrent cameras for user and environmental understanding. Its enhanced perception capabilities include a dedicated hardware acceleration engine that improves user motion tracking and localisation, an AI accelerator to reduce latency for sensitive input interactions such as hand tracking or 6DoF, and a reprojection engine for a smoother experience.
- The AR co-processor aggregates camera and sensor data and enables eye tracking and iris authentication for foveated rendering, to optimise workloads only where the user is looking. This helps reduce power consumption.
- The connectivity platform utilises Qualcomm FastConnect 7800 connectivity system to unlock the world’s fastest Wi-Fi 7 technology available and reaches <2ms latency between the AR glasses and the smartphone or host device. Embedded support for the FastConnect XR Software Suite 2.0 enables better control of XR data to improve latency, reduce jitter and avoid unwanted interference.
Beyond the transformative Snapdragon AR2 technology, it is critical to deliver an end-to-end solution consisting of hardware, a suite of perception technologies, and the software tools to build immersive experiences. To allow developers to build incredible headworn AR applications, Snapdragon AR2 and the Snapdragon 8 Gen 2 Platform are optimised to be Snapdragon Spaces Ready. The Snapdragon Spaces XR Developer Platform is designed to be the foundation that will pave the way for developers to reimagine headworn AR content and help propel the entire AR glass segment.
“We built Snapdragon AR2 to address the challenges of headworn AR and provide processing, AI and connectivity that can fit inside a stylish form factor,” says Hugo Swart, vice president of XR product management, Qualcomm Technologies, Inc. “With the technical and physical requirements for VR/MR and AR diverging, Snapdragon AR2 represents another metaverse-defining platform in our XR portfolio to help our OEM partners revolutionise AR glasses.”
“Microsoft worked closely with Qualcomm on the platform requirements for Snapdragon AR2 to help define the purpose-built, foundational technologies to unlock new possibilities in AR experiences,” says Ruben Caballero, corporate vice president of mixed reality, devices & technology, Microsoft. “Snapdragon AR2 platform innovations will revolutionise headworn AR devices that will transform immersive productivity and collaboration and we look forward to seeing the innovation that Qualcomm and its partners will bring to market.”
Follow us and Comment on Twitter @TheEE_io