Lattice introduces low power FPGA based on Nexus FD-SOI process
Low power programmable device company, Lattice Semiconductor introduces a new FPGA platform, Lattice Nexus. It is based on Samsung’s FD-SOU process technology and has been architected to deliver power-efficient performance for artificial intelligence (AI) for the IoT, video, hardware security, embedded vision, 5G infrastructure and industrial/automotive automation.
“The Lattice Nexus platform augments the parallel processing and re-programmability of FPGAs with the power-efficient performance demanded by today’s technology trends, like AI inferencing at the edge and sensor management,” explained Steve Douglass, corporate vice president, R&D for Lattice Semiconductor.
“Additionally, the Lattice Nexus platform offers easy-to-use solution stacks targeting high-growth applications that help customers more quickly develop their systems, even if they are not expert in FPGA design,” he added.
The Lattice Nexus platform provides system-level solutions that combine design software and pre-engineered soft IP blocks with evaluation boards, kits and reference designs to build embedded vision, sensor bridging, sensor aggregation and image processing systems quickly, said the company.
Innovative architectural features include optimised DSP blocks and higher on-chip memory capacity for power-efficient computing, such as AI inferencing algorithms. It runs twice as fast at half the power of Lattice’s previous FPGAs.
Lattice Nexus also uses innovative circuit design to deliver programmable power-performance optimisation and very fast configuration for instant-on type applications.
Lattice Nexus is developed on high-volume 28nm, fully-depleted silicon-on-insulator (FD-SOI) process technology from Samsung. This process technology is characterised by 50 per cent lower transistor leakage compared to bulk CMOS, reported Lattice.
The first FPGA developed on the Lattice Nexus FPGA platform is CrossLink-NX. This new FPGA provides the low power, small form factor, reliability, and performance that developers need to create innovative embedded vision and AI solutions for communications, compute, industrial, automotive, and consumer systems.
According to Lattice, FPGAs are a compelling hardware platform for embedded vision and AI applications, as they perform functions in parallel. The Lattice Nexus parallel architecture significantly accelerates certain processing workloads, including data inferencing.
Bill Pu, co-founder and president of Leopard Imaging, has identified uses for the CrossLink-NX FPGAs already. “With their very low power consumption, small size, high-performance interfaces, and robust software and IP library, Lattice CrossLink-NX FPGAs let us quickly and easily develop different video signal bridging, aggregation and splitting applications for our industrial and automotive customers using just one device, saving us significant development time and resources.”
CrossLink-NX has a soft error rate (SER) up to 100 times lower than similar FPGAs in its class, for mission-critical applications that must operate safely and reliably. The initial CrossLink-NX device is designed to support ruggedised environments found in outdoor, industrial, and automotive applications.
It three key elements, says Lattice are: fast I/O support with support for multiple fast I/Os, including MIPI, PCIe and DDR3 memory; instant on performance, with I/O configuration in three milliseconds and total device configuration in less than 15 milliseconds and a high memory to logic ratio for AI inferencing in edge devices. It has 170 bits of memory for every logic cell, believed to be the highest memory to logic ratio in its class, providing twice the performance compared to prior generations.
It is available in a 6.0 x 6.0mm form factor, which is up to 10 times smaller than similar competing FPGAs in its class, says Lattice.
In addition to the new Lattice Radiant 2.0 design software, there is also a library of popular IP cores including interfaces like MIPI D-PHY, PCIe, SGMII and OpenLDI and demos for common embedded visions applications such as 4:1 image sensor aggregation.
Lattice is releasing CrossLink-NX ahead of schedule and is sampling devices with select customers.