Editor’s Note: This content is contributed by Manuel Uhm, Director of Versal Adaptive SoC Marketing, AMD Adaptive SoCs and FPGAs.
As embedded AI, video, and control applications explode, there is a growing need for heterogeneous processing to accelerate end-to-end compute within tight area and power constraints. Multi-chip processing solutions come with significant power, area, and complexity overhead; instead, designers need support for flexible, real-time preprocessing, efficient AI inference, and high-performance postprocessing on a single device.
Today, I’m excited to announce that AMD is expanding its Versal™ portfolio to meet this evolving need. AMD Versal AI Edge Series Gen 2 and AMD Versal Prime Series Gen 2 adaptive SoCs deliver single-chip intelligence for AI-driven and classic embedded systems—balancing performance, power, area, functional safety, and security capabilities.
Achieve End-to-End Acceleration for AI-Driven Embedded Systems
Versal AI Edge Series Gen 2 devices enable end-to-end acceleration with a unique integration of heterogeneous processors, supporting all three phases of embedded compute:
- Preprocessing: Sensor processing, fusion, and data conditioning
- AI Inference: Execution of deep-learning algorithms
- Postprocessing: Decision-making and the resulting actions
During preprocessing, programmable logic offers outstanding performance and flexibility, supporting custom sensor combinations and low-latency processing pipelines that help improve system response time. In the next phase, AI Engines enable efficient AI inference with the AIE-ML v2 compute tile architecture featured in the Versal AI Edge Series Gen 2. The MX6 data type in AIE-ML v2 offers up to 3X higher TOPS (trillions of operations per second)-per-watt than the AIE-ML compute tile architecture in first-generation Versal AI Edge devices.[1] During the postprocessing phase, AI Edge Series Gen 2 devices easily handle complex decision-making and control algorithms with an enhanced system of integrated CPUs. The Versal AI Edge Series Gen 2 and Versal Prime Series Gen 2 offer up to 10X more scalar compute vs. their first-generation counterparts.[2]
The second generation of the AI Edge Series includes a range of devices that allow engineers to readily achieve performance and safety targets for applications such as autonomous driving, smart cameras with sensor fusion, and medical imaging.
Unleash Next-Level Performance for Classic Embedded Systems
Versal Prime Series Gen 2 devices offer extreme performance in a minimal footprint, with flexible, real-time sensor processing—easily handling complex embedded computing workloads. Combining world-class programmable logic with up to 10X the scalar compute of first-generation Versal Prime devices[2], these adaptive SoCs raise the bar for classic embedded systems (i.e., systems not requiring AI).
With new hard IP for high-throughput video processing, including 8K multi-channel workflows, Versal Prime Series Gen 2 devices are ideal for applications such as UHD streaming and recording, robotics, and flight computers.
Bringing Standout Adaptive Solutions and Tools to Market for 40 Years
When you choose AMD Versal adaptive SoCs, you get access to a broad portfolio of devices and the confidence of working with a proven partner with 40 years of experience in embedded systems, including those serving high-security, high-reliability, long-lifecycle, and safety-critical applications.
Productivity is key when designing these embedded systems, and our adaptive SoC design tools, libraries, and embedded software ecosystem boost productivity and streamline design cycles. Combined with the latest Versal devices, these resources give designers and system architects everything they need to confidently design, develop, and scale embedded systems equipped for today and tomorrow.
Enable Advanced AI, Control, and Video Applications with Second-Generation AMD Versal Adaptive SoCs
As compute requirements for both AI-driven and classic embedded systems grow, delivering end-to-end acceleration in a single device is imperative. Adaptive solutions that balance performance, power, area, functional safety, and security are essential to meet and exceed evolving needs. At AMD, we’re proud to deliver next-generation solutions for the automotive, industrial, vision, healthcare, aerospace and defense, broadcast, and pro AV markets—and beyond.
Learn more about Versal AI Edge Series Gen 2 and Versal Prime Series Gen 2 devices.
1. Based on AMD internal performance and power projections for the AIE-ML v2 compute tile architecture in the Versal AI Edge Series Gen 2 using the MX6 data type, compared to performance specifications and AMD Power Design Manager power results for the AIE-ML compute tile architecture featured in the first-generation Versal AI Edge Series using INT8 data type. Assumptions: 2 row, 8 column sub-arrays. Operating conditions: 1 GHz Fmax, 0.7V AIE operating voltage, 100°C junction temperature, typical process, 60% vector load, % activations = 0 < 10%. Actual performance will vary when final products are released in market. Performance projections as of March 2024. (VER-023)
2. Based on pre-silicon estimates for combined total DMIPs of the Versal™ AI Edge Series Gen 2 and Versal Prime Series Gen 2 processing systems, when configured with 8 Arm® Cortex®-A78AE application cores at 2.2 GHz and 10 Arm Cortex-R52 real-time cores at 1.05 GHz, compared to the combined total DMIPs in the first-generation Versal AI Edge Series and Versal Prime Series. Versal AI Edge Series Gen 2 and Versal Prime Series Gen 2 operating conditions: Highest available speed grade, 0.88V PS operating voltage, split-mode operation, and maximum supported operating frequency. First-generation Versal AI Edge Series and Versal Prime Series operating conditions: Highest available speed grade, 0.88V PS operating voltage, and maximum supported operating frequency. Actual DMIPs performance will vary when final products are released in market. (VER-027)
© 2024 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, Versal, Vivado, Vitis, and combinations thereof are trademarks of Advanced Micro Devices, Inc. in the United States and other countries. Arm and Cortex are registered trademarks of Arm Limited (or its subsidiaries) in the US and/or elsewhere. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies.