Advanced Micro Devices (AMD) has expanded its Versal adaptive system on chip (SoC) portfolio with the introduction of the new Versal AI Edge Series Gen 2 and Versal Prime Series Gen 2 adaptive SoCs. These devices integrate preprocessing, AI inference and postprocessing capabilities into a single device to provide end-to-end acceleration for AI-driven embedded systems.
The Versal Series Gen 2 devices build upon the first generation with new AI engines expected to deliver up to 3x higher TOPs-per-watt and up to 10x more scalar compute, AMD says.
The advancements aim to balance performance, power, area and functional safety and security features, with the Gen 2 solutions enabling “the design of high-performance, edge-optimized products for the automotive, aerospace and defense, industrial, vision, healthcare, broadcast and pro AV markets”, AMD says.
Subaru Corporation has chosen Versal AI Edge Series Gen 2 devices to power its next-generation advanced driver-assistance system (ADAS) vision system, known as EyeSight. This system is integrated into select Subaru car models to enable advanced safety features, and AMD adaptive SoC technology is already used in current EyeSight-equipped vehicles.
“Subaru has selected Versal AI Edge Series Gen 2 to deliver the next generation of automotive AI performance and safety for future EyeSight-equipped vehicles,” said Satoshi Katahira, GM of advanced integration system department and ADAS development department of the engineering division at Subaru.
The Versal AI Edge Series Gen 2 devices incorporate FPGA programmable logic for real-time preprocessing, next-gen AI Engines for efficient AI inference, and Arm CPU cores for postprocessing power. This single-chip intelligence is designed to eliminate the need for multi-chip processing solutions.
In comparison, the Versal Prime Series Gen 2 provides end-to-end acceleration for traditional, non-AI-based embedded systems by combining programmable logic for sensor processing with high-performance embedded Arm CPUs. AMD says the Prime Series devices have the capability to offer up to 10x more scalar compute compared to the first generation, making them suitable for applications such as UHD video streaming, industrial PCs and flight computers.
“The demand for AI-enabled embedded applications is exploding and driving the need for single-chip solutions for the most efficient end-to-end acceleration within the power and area constraints of embedded systems,” said Salil Raje, senior vice president and general manager of adaptive and embedded computing group at AMD.
“Versal AI Edge Gen 2 devices are designed to provide the AI inference performance, ultra-low latency and functional safety capabilities required to put cutting-edge AI-based safety features in the hands of drivers,” said Katahira.
Early access documentation and evaluation kits for Versal Series Gen 2 devices are available now, with silicon samples expected in the first half of 2025 and production silicon anticipated in late 2025.