Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/index.php?threads/efinix%C2%AE-announces-trion%C2%AE-titanium-fpga-family.12779/
        )

    [addOns] => Array
        (
            [DL6/MLTP] => 13
            [Hampel/TimeZoneDebug] => 1000070
            [SV/ChangePostDate] => 2010200
            [SemiWiki/Newsletter] => 1000010
            [SemiWiki/WPMenu] => 1000010
            [SemiWiki/XPressExtend] => 1000010
            [ThemeHouse/XLink] => 1000970
            [ThemeHouse/XPress] => 1010570
            [XF] => 2021370
            [XFI] => 1050270
        )

    [wordpress] => /var/www/html
)

Efinix® Announces Trion® Titanium FPGA Family

Daniel Nenni

Admin
Staff member
July 15, 2020 03:01 ET | Source: Efinix, Inc.

SANTA CLARA, Calif., July 15, 2020 (GLOBE NEWSWIRE) -- Efinix®, an innovator in programmable product platforms and technology, today announced its Trion® Titanium FPGA family.

Trion Titanium FPGAs are fabricated on a 16nm process node and feature Efinix’s Quantum™ compute fabric. Inspired by the Quantum fabric underlying Efinix’s first-generation Trion FPGAs, the Quantum compute fabric adds additional compute and routing capability into its enhanced eXchangeable logic and routing (XLR) cells. Enhanced compute, united with the 3X clock frequency boost afforded by the 16nm process, make Trion Titanium FPGAs ideal for computational acceleration applications while the increased routing flexibility delivers unprecedented utilization ratios.

With the 16nm process node and the 2X efficiency improvement of the Quantum compute fabric, Titanium FPGAs pack loads of processing power into an extremely small die size, taking just a quarter of the area of the previous Trion generation. Additionally, the low power consumption of the 16nm node means that Titanium devices consume a third of the power of Trion devices and overcome all the thermal issues associated with highly integrated applications. This combination makes them ideal for multi-chip, system-in-package (SIP) designs such as those found in mobile, edge compute, AI and IoT.

“FPGAs are increasingly being used as accelerators for computation rather than fixed function logic,” said Sammy Cheung, Efinix founder, president and CEO. “With a re-engineered Quantum compute fabric and a low-power 16nm process, Titanium FPGAs naturally offload edge processors in space constrained applications. When combined with the Efinix RISC-V SoCs announced last month, Titanium FPGAs form the compute core and adaptive hardware acceleration for complete SIP SoCs.”
The Titanium family comprises FPGAs ranging from 25K to 500K logic elements that are available in familiar and easy to mount BGA packages. The FPGAs have a range of hardened IP such as PCIe Gen4, DDR4, 10 Gbps Ethernet, and 2.5 Gbps MIPI controllers for optimum performance and seamless system connectivity in applications ranging from vision systems and industrial automation to edge computing.

“When designing the Titanium family, we started from the ground up,” said Tony Ngai, Efinix co-founder, CTO and SVP of Engineering. “From the re-engineered XLR and the highly efficient DSP capability, to the optimized embedded RAM blocks and the custom designed I/Os, we analyzed system requirements and set about designing the ultimate reconfigurable acceleration FPGA. The result is a truly unique collection of innovations based on Efinix’s patented Quantum architecture.”

About Efinix
Efinix®, an innovator in programmable products, drives the future of edge AI computing with its Trion® and Trion Titanium FPGA silicon platforms. At the Trion family’s core is Efinix’s disruptive Quantum™ FPGA technology which delivers power, performance and area advantage over traditional FPGA technologies. Trion FPGAs, offering 4K to 500K logic elements, have a small form-factor, low-power, and are priced for high-volume production. Our Efinity® Integrated Development Environment provides a complete FPGA design suite from RTL to bitstream. With their Power-Performance-Area advantage, Trion FPGAs address applications such as custom logic, compute acceleration, machine learning and deep learning.
For more information, visit http://www.efinixinc.com.

Contact:
Steve Stratz
Relevanz Public Relations for Efinix
206.300.9134
steve@relevanzpr.com
 
Back
Top