Sign In  |  Register  |  About San Rafael  |  Contact Us

San Rafael, CA
September 01, 2020 1:37pm
7-Day Forecast | Traffic
  • Search Hotels in San Rafael

  • CHECK-IN:
  • CHECK-OUT:
  • ROOMS:

Think Silicon to Showcase its Latest Ultra-Low-Power 3D Graphics and AI in One IP Architecture at Embedded World 2024

ATHENS, Greece and SAN JOSE, Calif., April 02, 2024 (GLOBE NEWSWIRE) -- Think Silicon, the leading provider of ultra-low-power GPU and AI IP for embedded systems, will showcase its latest graphics and AI solutions for edge computing devices in Hall 4, Booth 476 at Embedded World 2024 taking place in Nuremberg, Germany from April 9-11.

Think Silicon’s NEOX® | GA100 is a GPGPU solution for the MCU market based on RISC-V, offering the unique combination of performing both 3D graphics and ΑΙ in one IP architecture. The processor can be programmed using graphics and AI libraries independently or together. This flexibility provides SoC fabless customers more opportunities to package and segment chips according to use cases and markets, thereby maximizing silicon area while saving cost.

AI SDK optimized for low-power edge applications

NEOX® | GA100 is delivered with an AI SDK, a collection of open source and proprietary tools for converting, optimizing, analyzing and deploying pre-trained and post-trained neural network models targeting sports, wellness, audio, gesture and vision-based AI applications on future wearable and AIoT devices. It allows developers to perform various iterative steps in model compression and model analysis, until the desired balance between “accuracy-performance-memory” is achieved. Customers can perform offline graph optimization and AI inference based on TensorFlow Lite for MCU. 

3D graphics on a lightweight RTOS system

NEOX® | GA100 provides a new hardware and software extension that can accelerate 3D graphics (gltf), while enabling efficient transmission and loading of 3D models and applications. Effects like rotation, lighting and animation can be applied before rendering to a screen. The entire 3D/2.5D and VG API requires less than 150 KB of memory and is ideal for graphics rendering on a variety of target systems – from low-power bare-metal to RTOS and high-performance, Linux-based systems.

Supports LVGL graphics library and feature upgrades

Think Silicon will also demonstrate its vector graphics GPU flagship – NEMA® | pico VG series – which includes software feature upgrades and support for the latest version of the popular open-source graphics software library LVGL (Light Versatile Graphics Library) 9.0.

In addition, Think Silicon will showcase an array of customer demos featuring the NEMA® | pico GPU, including the Ambiq® Apollo5 Graphic Display Kit, ST Micro U5 & H7 series, Qualcomm Snapdragon W5+ and Racyics.

Think Silicon’s experts will be available for meetings and demonstrations of the NEMA® and NEOX® product series at Embedded World. Please contact us here to schedule a meeting.

About Think Silicon, an Applied Materials Company

Think Silicon Single Member P.C., an Applied Materials company, is a leading provider of ultra-low power 2D/3D GPU and display processors for MCU-based microprocessor cores in embedded systems. The flagship NEMA® GPU IP can be found in millions of chips that drive displays in resource-constrained devices and serve a variety of applications, from wearables and infotainment to micro mobility and AIoT. Launched as the industry’s first RISC-V GPGPU, NEOX® is revolutionizing graphics and AI for wearables and IoT devices in a single IP. Think Silicon’s headquarters and development center are based in Greece, with sales and technical support offices in North America, Europe, and Taiwan.

Contact:
Anastasia Boussia, anastasia_boussia(at)amat.com


Primary Logo

Data & News supplied by www.cloudquote.io
Stock quotes supplied by Barchart
Quotes delayed at least 20 minutes.
By accessing this page, you agree to the following
Privacy Policy and Terms and Conditions.
 
 
Copyright © 2010-2020 SanRafael.com & California Media Partners, LLC. All rights reserved.