High-Performance, Low-Power DRAM Combines with SoC in a SiP

By Rich Nass

Contributing Editor

Embedded Computing Design

February 08, 2021

News

High-Performance, Low-Power DRAM Combines with SoC in a SiP
(Image Courtesy of Winbond)

What do you get when you cross high-bandwidth memory with artificial intelligence (AI)?

You get Winbond Electronics’ 1-Gbit LPDDR3 DRAM product. The feat is achieved when the memory device is combined with Tsing Micro’s TX510 AI system-on-chip (SoC). The two devices are housed in a single 14- by 14-mm TFBGA system-in-package (SiP).

The TX510 is an advanced AI Edge computing engine optimized that’s intended for functions like 3D sensing, facial recognition, object recognition, and gesture recognition. Those functions are attainable when the part is paired with Winbond’s LPDDR3 DRAM, which offers a maximum bandwidth of 1866 Mbits/s and operates from a dual 1.2/1.8-V supply. Also included are power-saving features such as deep power-down mode and a clock stop capability.

The TX510 SoC includes a 32-bit RISC processor, a reconfigurable neural network engine, a reconfigurable general computing engine, an image signal processor, and a 3D sensing engine. Achieving a computing throughput of up to 1.2 TOPS, this SiP can perform accurate face recognition (false acceptance ratio of 1 in 10 million) in less than 100 ms, and compare features with a library of 100,000 faces in less than 50 ms. Peak operating power consumption is just 450 mW, and the chip uses just 0.01 mW in quiescent mode.

The SiP is intended for use in applications which require high-speed image detection and recognition, including biometric sensing, video surveillance, smart retail operations, smart home automation and advanced industrial automation.

Rich Nass is a regular contributor to Embedded Computing Design. He has appeared on more than 500 episodes of the popular Embedded Executive podcast series, and is a regular contributor to the Embedded Insiders podcast.

Rich has been in the engineering OEM industry for more than 35 years, and is a recognized expert in the areas of embedded computing, Edge AI, industrial computing, the IoT, and cyber-resiliency and safety and security issues. He writes and speaks regularly on these topics and more.

Rich is currently the Liaison to Industry for the Embedded World North America Exhibition and Conference, and has held similar positions with the global Embedded World Conference and Exhibition.

Previously, Rich was the Brand Director for UBM’s award-winning Design News property. Prior to that, he led the content team for UBM Canon’s Medical Devices Group, as well all custom properties and events.  In prior stints, he led the Content Team at EE Times, handling the Embedded and Custom groups and the TechOnline DesignLine network of design engineering web sites.

Nass holds a BSEE degree from the New Jersey Institute of Technology.

Podcast/Interview Coverage

Sonatus The Garage Podcast

onalytica Interview

Dev Talk with Rich and Vin

Embedded Executive Podcast

Semiconscious Webcast

IEEE Awards Frede Blaabjerg Talks EVS

Atmosic: Embedded Executive: Energy Harvesting Podcast

 

Article Coverage

Embedded AI Isn’t Enterprise AI, and That’s a Good Thing

Tear Down: Google Pixel Watch 4

Protect Your Home from Thieves and Floods

Advantech Teams With AMD To Maximize Performance at the Edge

Tear Down: Noise Luna Ring

 

View additional information

Muck Rack

More from Rich