» Current | 2021 | 2020 | 2019 | Subscribe

Linley Newsletter

Flex Logix Diversifies eFPGA

July 24, 2018

Author: Bob Wheeler

Embedded FPGAs are starting to take off, driven by customer demand in diverse applications. In a little more than four years from its founding, Flex Logix has responded to that demand by proving its embedded-FPGA (eFPGA) core in several IC processes while improving its core technology with a second-generation design. By using standard-cell designs, the startup has delivered an impressive roadmap despite having a small team. Its primary target markets are networking, wireless infrastructure, and aerospace/defense. Investors have rewarded its progress with two funding rounds, most recently $5 million raised in May 2017.

Flex Logix proved its first-generation cores in a TSMC 28nm process in 2015. To address a secondary market in microcontrollers (MCUs), it added in 2016 a 40nm version of its low-end core. Learning from early customer engagements, the startup developed a second-generation core, the EFLX4K, targeting 16nm FFC technology at TSMC. The most important change to the second-generation EFLX core is a move from four- to six-input LUTs, enabling greater logic density and higher performance. Both generations scale to 7x7 arrays, but the company made undisclosed improvements to its interconnect to boost large-array performance. In the maximum 7x7 configuration, the EFLX4K delivers up to 123,500 LUT6s, equivalent to 185,000 LUT4s.

Meant for integration in ASICs, ASSPs, and MCUs, eFPGAs seldom replace external FPGAs. Instead, they add some programmability to a fixed-function device or one that performs signal processing. Wireless-infrastructure OEMs have adopted them for use in ASICs, which may appear in baseband processors or digital front ends. An Ethernet switch-ASSP vendor can use an eFPGA to add programmability to its packet-processing pipeline. MCU vendors employ eFPGAs to create product variants (SKUs) from a single mask set. The technology has the potential to accelerate other workloads, including machine-learning inference.

Subscribers can view the full article in the Microprocessor Report.

Subscribe to the Microprocessor Report and always get the full story!

Purchase the full article

Free Newsletter

Linley Newsletter
Analysis of new developments in microprocessors and other semiconductor products
Subscribe to our Newsletter »


Linley Fall Processor Conference 2021
Coming October 20-21, 2021
Hyatt Regency Hotel, Santa Clara, CA
Register Now!
Linley Spring Processor Conference 2021
April 19 - 23, 2021
Proceedings Available
More Events »