Tektronix TIF 2026
This page was machine-translated and may differ from the original. View original

Congatec Launches NXP i.MX 95-Based SMARC Module

Google 우선 소스Published2024.07.17 10:25

Up to 3 times faster GFLOPS · AI machine vision inference performance doubled

Congatec, a leader in embedded and edge computing technology, is providing a high-performance edge computing development environment with the launch of a new module for edge AI applications that delivers up to 3 times faster GFLOPS computing performance and doubles AI machine vision inference performance compared to its predecessor.

Congatec announced on the 16th that it is expanding its extensive module portfolio, including low-power NXP i.MX Arm processors, by launching a new high-performance computer-on-module (COM) equipped with NXP's i.MX 95 processor.

With a strong partnership with NXP, customers can receive direct scalability and a reliable upgrade path for existing and new energy-efficient edge AI applications with high security requirements.

The newly released module provides up to 3 times faster GFLOPS (gigaflops) computing performance compared to the previous generation based on the i.MX8 M Plus processor, and NXP's new neural network processing unit, eIQ Neutron, doubles the inference performance of AI-accelerated machine vision.

In addition, the hardware-integrated EdgeLock Secure Enclave simplifies the implementation of cybersecurity measures within an organization.

The conga-SMX95 SMARC module is ruggedly designed for industrial temperature ranges from -40 to +85 degrees.It is optimized for cost and energy-efficient applications.

With the integrated high-performance eIQ Neutron NPU, AI-accelerated workloads can be processed more effectively, closer to the local device level.

The new SMARC module is ideal for AI-accelerated low-power applications in fields such as industrial production, machine vision and visual inspection, rugged HMI, 3D printers, robot controllers for autonomous mobile robots (AMRs) and automated guided vehicles (AGVs), and medical imaging and patient monitoring systems.

It is also suitable for entertainment systems in bus and aircraft seat backs, vehicle management in the transportation sector, and construction and agricultural applications.

The conga-SMX95 SMARC 2.1 module is based on the next-generation NXP i.MX 95 application processor, which features four to six Arm Cortex-A55 cores.

NXP introduces the Arm Mali 3D graphics unit for the first time, providing up to 3 times faster GPU performance compared to the previous generation based on i.MX8 M Plus.

Along with the Image Signal Processor (ISP) for hardware-accelerated image processing, the NXP eIQ Neutron for hardware-accelerated AI inference and edge machine learning (ML) has been newly added in the new SMARC module. This eIQ software development environment from NXP provides OEMs with a high-performance development environment that simplifies the implementation of ML applications within their organizations.

It also integrates a real-time domain for the real-time controller. The conga-SMX95 SMARC module provides two Gbit Ethernets with Time Sensitive Networking (TSN) for synchronous and deterministic network data transmission and LPDDR5 (with inline ECC) for data security.

For display connectivity, the new module provides DisplayPort and the widely used LVDS display interface as standard interfaces, and for direct camera connectivity, the module supports two MIPI-CSI ports.

Congatec provides an extensive hardware and software ecosystem along with a comprehensive service design to streamline and accelerate application development. This includes evaluation and production-ready application carrier boards and custom cooling solutions, while in terms of services, Congatec offers integrated documentation, training, and signal integrity simulation services for application development.

Detailed information about the conga-SMX95 SMARC module and SMARC module standards can be found on the Congatec website.
본 기사에 대한 정정·반론·추후보도 청구는 보도 청구 안내를, 그간 게재된 보도문은 정정·반론보도 모아보기를 참고해 주세요.
배종인 기자
배종인 기자