Xilinx set to demo 16nm All Programmable MPSoCs

Xilinx Inc. is set to unveil what it boasts as the industry's first 16nm All Programmable MPSoC, the Zynq UltraScale+ MPSoC at ARM TechCon 2015, through a series of presentations and demonstrations. In addition, the company will also feature solutions for important industry megatrends.

The UltraScale MPSoC Architecture is built on TSMC's 16FinFET+ process technology and enables next-generation Zynq UltraScale+ MPSoCs. Building on the Zynq-7000 All Programmable SoC family, the latest UltraScale MPSoC architecture extends Xilinx SoCs to enable true heterogeneous multi-processing with 'the right engines for the right tasks' for smarter systems including smarter control, vision and network.

The architecture aims to enable: connected control/M2M though flexible/adaptable manufacturing, factory throughput, quality and safety; advanced 2D/3D through evolving video-processing algorithms, object detection and analytics; and wireless and wired communications and data centres by supporting multiple spectral bands and smart antennas, multiple wired communications standards and context-aware network services, and SDN, data pre-processing and analytics, respectively.

Zync-7000 diagram

The EyeTech demonstration features the Zynq-7000 All Programmable SoC and allows users to experience first-hand, this touch free control interface driven by the human eye. The PFP Cybersecurity demonstrates monitoring, detecting, reporting, and remediating malware intrusions in machine time, using the Zynq-7000 All Programmable SoC.

The 2D/3D graphics with Display Port for HMI demonstration features the Zynq UltraScale+ MPSoC with integrated ARM Mali GPUs/graphics processors to optimise performance and power through multiple processing domains. In addition, the video processing system acceleration with C/C++ demonstration showcases maximising performance and power through a heterogeneous multiprocessing system using quad core ARM Cortex-A53 based SoCs for application processing, dual ARM Cortex-R5 based SoCs for real-time processing.