Compared with mobile processors, desktop processors and server processors, everyone may not pay so much attention to AMD's embedded products. In fact, AMD's embedded processors are important for edge devices, including industrial, automotive, medical, digital game consoles, and small client systems. Today, AMD unveiled its latest embedded architecture, Embedded+, which combines Zen+-based Ryzen™ embedded processors with Versal adaptive SoC onto a single PCB.
The EmbedDED+ architecture integrates the capabilities of Ryzen™ embedded processors with its Versal AI Edge adaptive SoC onto a single package board. AMD targets key areas that require good computing power and energy efficiency, and this synergy enables Embedded+ to process AI inference and manage complex sensor data in real time, which is critical for applications in dynamic and demanding environments.
Packaging Ryzen™ embedded processors with Versal SoCs is especially beneficial for industries that require low-latency response times between hardware and software, including autonomous vehicles, diagnostic devices in healthcare, and precision mechanics in industrial automation. The AMD Embedded+ architecture also supports a wide range of workloads across different processor types, including x86 and ARM, as well as AI engines and FPGA fabrics, providing flexibility and scalability for embedded computing solutions in the industry.
The AMD Embedded+ platform offers a great deal of compatibility with a wide range of sensor types and their corresponding interfaces. It facilitates direct connection to standard peripherals and industrial sensors via Ethernet, USB, and HDMI DP interfaces. The AMD Ryzen™ embedded processors in this architecture can process inputs from traditional image sensors such as RGB, monochrome, and even advanced neuromorphic types, while supporting industry-standard image sensor interfaces such as MIPI and LVDS.
The AMD Versal AI Edge Adaptive SoC on Embedded+ motherboards further enhances its capabilities, providing adaptive IO options for real-time sensor input and industrial networks. This includes interfaces to lidar, radar, and other sophisticated and complex sensors required for modern embedded systems in the industrial, medical, and automotive sectors. The platform supports a wide range of production-level sensor interfaces, such as GMSL and Ethernet-based vision protocols, which means it is designed to be integrated into complex sensor-driven systems.
AMD is also announcing a new pre-integrated solution that will be available to ODM vendors starting today. The Sapphire VPR-4616-MB platform is based on the ITX form factor, using the AMD Versal AI Edge 2302 SoC and AMD Ryzen embedded R2314 processor, which is based on the Zen+ architecture, with four cores and four threads, and is equipped with six VEGA CU units. The motherboard is equipped with two DDR4 SO-DIMM memory slots, with a maximum memory capacity of 64GB, and offers a PCIe 30 x4 m.2 slots and a SATA port, and 2 for the network5 GB network card and an M. for expanding the wifi wireless network card2 Key E port, it also supports Linux-based Ubuntu 2204 Operating system.
Also released at the same time is a series of expansion boards that significantly expand support for the Embedded+ architecture. The OCTO GMSL Camera IO board is particularly noteworthy for its ability to connect with multiple cameras at the same time, making it suitable for high-bandwidth vision-based systems and can be used in areas such as advanced driver assistance systems and automated monitoring systems. These systems often require the integration of a large number of image inputs for real-time processing and analysis, and OCTO GMSL boards are designed to meet this need.
In addition, there are dual Gigabit Ethernet ports I.O. boards to meet the needs of environments that require high-speed network communication. Dual 10 GB SFP+ boards with 16 GPIOs meet higher bandwidth requirements, providing ample data transfer rates for tasks such as real-time streaming and large-scale sensor data aggregation. These expansion options expand the capabilities of the Embedded+ architecture in both edge and industrial scenarios.