AMD Versal AI Engines enable you to scale your digital signal processing (DSP) compute and future-ready designs for current and the next-generation of compute-intensive DSP applications. With Versal AI Engines, customers can expect high-performance DSP with reduced power1 and fewer programmable logic resources.2
“Versal AI Engines can increase DSP compute density at a reduced power budget,” said Udayan Sinha , the senior product marketing manager. “This efficiency allows Versal AI Engines to tackle the most demanding DSP applications operating within a strict power budget.”
Markets that can benefit from these computing capabilities include aerospace and defense, test and measurement, medical imaging/healthcare, and other markets that require high-throughput, real-time DSP operating in the GSPS throughput range.
Utilizing the AMD Vitis™ Unified Software Platform, Versal AI Engine designs can be programmed via a C-based design flow. For engineers that prefer a model-based flow, Vitis Model Composer (plugin to MathWorks Simulink®) can be used. Additionally, a collection of DSP library functions, APIs, and enhanced IP cores enable a rapid development of high-performance DSP designs.
Visit the AMD GitHub to see our Versal AI Engine GitHub Design Tutorials and DSP Libraries.
Get started with our webinar series, experts from AMD introduce the Versal AI Engine technology and demonstrate how they can be used to accelerate a variety of valuable DSP algorithms. Topics in this webinar series include:
- An introduction to Versal Adaptive SoC and AI Engine technology
- Acceleration of a polyphase channelizer design using Versal AI Engines
- Acceleration of FFT algorithms using Versal AI Engines
Register here for watching the webinar series.
To also see how Versal AI Engines performed in head-to-head DSP benchmarks against a traditional HDL implementation, Contact your sales representative for benchmark designs.
In summary, Versal AI Engines can accelerate high-performance DSP applications by reducing dynamic power and programmable logic resources. If your current DSP applications are running into potential scaling issues due to resource or power constraints, Versal AI Engines are ideal solutions to unlock your current and next-generation DSP compute capabilities.
To learn how to get started programming Versal AI Engines, visit the design flow page.
- Based on 3rd party benchmark testing commissioned by AMD in February 2024, on the AMD Versal adaptive SoC with Vitis for AI design tools versus traditional programmable logic with Vivado software and Vitis Model Composer Tool, version 2023.1 in a signal processing application FIR implementation. Results averaged over 3 designs. Results will vary depending on design specifications. (VER-034)
- Based on 3rd party benchmark testing commissioned by AMD in February 2024, on the AMD Versal adaptive SoC with AMD Vitis for AI design tool versus traditional programmable software implementation with Vivado software and Vitis Model Composer tool, version 2023.1 in a signal processing application FIR implementation. Results averaged over 4 designs. Results will vary depending on design specifications. (VER-035)