Developer kit creates entry-level AI-powered systems says Nvidia

Claimed to set a new standard for creating entry-level AI-powered robots, smart drones and intelligent vision systems, the Jetson Orin Nano developer kit has been introduced to support developers using the Jetson Orin Nano series modules.

The kit is compact, with numerous connectors and up to 40 TOPS of AI performance. It consists of a Jetson Orin Nano 8Gbyte module with an SD-card slot, a reference carrier board, preassembled heatsink/fan, a 19 V DC power supply, and an M.2-Key E based wireless networking module. In addition to the bootable microSD card slot, two M.2 Key-M NVMe sockets are provided on the underside of the carrier for high-speed storage.

It also includes a reference carrier board that can accommodate all Nvidia Jetson Orin Nano and Nvidia Jetson Orin NX modules.

The Jetson Orin Nano module features an Nvidia Ampere architecture GPU with 1024 CUDA cores, 32 third-generation Tensor cores and a six-core Arm CPU. These enable multiple concurrent AI application pipelines and high performance inference, said Nvidia. The developer kit carrier board boasts an array of connectors, including two MIPI CSI connectors supporting camera modules with up to four lanes.

While the earlier generation Jetson Nano developer kit made AI accessible to everyone, this Jetson Orin Nano developer kit raises the bar for entry-level AI development, claimed Nvidia, with 80 times the performance, enabling developers to run any kind of modern AI models, including transformer and advanced robotics models. Jetson Orin provides a boost in AI performance over the Jetson Nano with 5.4 times the CUDA compute, 6.6 times the CPU performance and 50 times the performance per Watt.

The Jetson Orin Nano developer kit runs the Nvidia AI software stack, with available use-case-specific application frameworks, including Isaac for robotics, DeepStream for vision AI and Riva for conversational AI. Omniverse Replicator is includes for synthetic data generation (SDG), and there is also the TAO Toolkit for fine-tuning pretrained AI models from NGC, Nvidia’s portal of enterprise services, software, management tools, and support for end-to-end AI and digital twin workflows.

http://www.nvidia.com

> Read More

Reference design offers early access to i.MX 93 applications processor  

A reference development platform based on NXP Semiconductors; i.MX 93 applications processors offers developers access to prototype ideas for a variety of applications including smart home and smart city projects.  

The EIC-i.MX93-210 reference development platform kit is based on NXP’s i.MX 93 applications processor which includes the scalable Arm Cortex-A55 core. This enables efficient machine learning acceleration and advanced security with an integrated EdgeLock secure enclave to support edge computing, said eInfochips, an Arrow company. 

The NXP i.MX 93 processor provides new use cases and opens a new gateway to a faster, more secure, powerful, and advanced edge-based system for industrial automation, smart home and intelligent appliances, smart buildings and automotive applications, said eInfochips.

The development platform enables developers to kick-start the development of creating more capable high-performance, cost-effective, and energy-efficient machine learning applications, said eInfochips. The company can help engineers progress product development with hardware development, firmware, edge-AI enablement, end-to-end design and manufacturing expertise.

“NXP’s i.MX 93 applications processor combines both a level of performance and low power that will enable a wide range of innovative edge devices,” said Jeff Steinheider, vice president and general manager, Industrial Edge, NXP.

eInfochips is an Arrow Electronics company. It provides digital transformation and product engineering services and accelerates time to market for customers with expertise in IoT, AI / ML, security, sensors, wireless, cloud, and power.

https://www.einfochips.com

> Read More

Video processing architecture by AMD is basis for 5nm ASIC-MAC

The Alveo MA35D media accelerator card (MAC) by AMD, has two 5nm, ASIC-based video processing units to support the AV1 compression standard. According to the company, it is purpose-built to power a new era of live interactive streaming services at scale. 

AMD reported that more than 70 per cent of the global video market is dominated by live content and that a new class of low latency, high volume interactive streaming applications are emerging. These include watch parties, live shopping, online auctions and social streaming.

The Alveo MA35D media accelerator delivers high channel density, with up to 32x 1080p60 streams per card. It is also power efficient with low latency performance which are critical to reduce rising infrastructure costs required for scaling such compute intensive content delivery. Compared to the previous generation Alveo U30 media accelerator, the Alveo MA35D delivers up to four times higher channel density and lower latency by a factor of four and 1.8 times greater compression efficiency, to achieve the same VMAF score as a video quality metric.

Dan Gibbons, general manager of AECG data centre group, AMD, commented: “We developed the Alveo MA35D with an ASIC architecture tailored to meet the bespoke needs of these providers to reduce both capital and operating expenses for delivering immersive experiences to their users and content creators at scale.”

The Alveo MA35D uses a purpose-built VPU to accelerate the video pipeline. By performing all video processing functions on the VPU, data movement between the CPU and accelerator is minimised, reducing overall latency and maximising channel density with up to 32x 1080p60, 8x 4Kp60, or 4x 8Kp30 streams per card. It provides low latency support for the mainstream H.264 and H.265 codecs and features next-generation AV1 transcoder engines delivering up to a 52 per cent reduction in bit rate for bandwidth savings versus a comparable software implementation.

The accelerator features an integrated AI processor and dedicated video quality engines designed to improve the quality of experience at reduced bandwidth. The AI processor evaluates content, frame-by-frame, and dynamically adjusts encoder settings to improve perceived visual quality while minimising bit rate. Optimisation techniques include region-of-interest (ROI) encoding for text and face resolution, artifact detection to correct scenes with high levels of motion and complexity, and content-aware encoding for predictive insights for bit rate optimisation. 

Scaling high-volume streaming services requires maximizing the number of channels per server while minimizing power and bandwidth-per-stream. By delivering up to 32x 1080p60 streams per card at 1W per stream, a 1U rack server equipped with eight cards delivers up to 256 channels to maximis the number of streams per server, rack or data centre. 

The Alveo MA35D MAC is accessible with the AMD Media Acceleration software development kit, supporting the FFmpeg and Gstreamer video frameworks for ease of development.

Alveo MA35D media accelerators are sampling now with production shipments expected in Q3. To accelerate development, an Early Access Program is available to qualified customers with comprehensive documentation and software tools for architectural exploration. 

https://www.amd.com/

> Read More

Cellular-to-cloud dev kits connect to Microsoft Azure cloud services 

Development kits that are Azure-certified are available from Renesas Electronics, powered by the company’s 32-bit microcontrollers. Renesas offers two versions of the kits, the CK-RA6M5 and CK-RX65N.

The kits also include the RYZ014A Cat-M1 Pmod, six on-board sensors, and a set of peripherals, including a USB FS connector, USB-to-UART serial converter, two Pmod connectors and a battery connector. For additional layers of security, Renesas collaborated with Crypto Quantique to implement a software-based IoT cryptography platform that securely connects IoT devices to server-hosted applications on-premises or in the cloud. Hardware-based security, such as the Renesas Secure Crypto Engine 9 (SCE9) and Trusted Secure IP (TSIP) driver has been incorporated on the RA6M5 and RX65N MCUs respectively. 

Both development kits enable users to connect and manage wireless IoT devices via the cloud and IoT device developers to exploit features and services available through the cloud computing platform.

Both cloud development kits incorporate Renesas’ LTE CAT-M1 module, which wirelessly connects MCUs to cloud services without a gateway. The kits have Microsoft’s new Embedded Wireless Framework, which provides a simplified link to the Azure cloud platform. It lets any network adapter connect securely to the cloud, providing greater flexibility and better lifecycle management of cloud-based products, said Renesas. As a result the kits can be used in asset tracking, building automation, medical data management, voice command and industrial control applications.

Both the CK-RA6M5 and CK-RX65N kits have been certified by Microsoft Azure and are therefore validated to connect with Azure IoT hub and can be securely provisioned through the device provisioning service. According to Renesas, both can be integrated using pre-defined device models which eliminates the need to manually configure hardware.  

The CK-RA6M5 and CK-RX65N cloud kits allow users to collect data from onboard sensors, upload them to the cloud for AI / machine learning processing and tools, and download the model back to the end point. Simplified development software available  includes Reality AI tools and Edge Impulse Studio. Reality AI Tools provide a true endpoint AI for a range of applications, including anomaly detection and voice recognition. The Edge Impulse Studio integrated cloud environment enables end-to-end AI / ML development using the kits’ accelerometer for asset and package tracking use cases.  

Both the CK-RA6M5 and CK-RX65N are available today. 

http://www.renesas.com

> Read More

About Smart Cities

This news story is brought to you by smartcitieselectronics.com, the specialist site dedicated to delivering information about what’s new in the Smart City Electronics industry, with daily news updates, new products and industry news. To stay up-to-date, register to receive our weekly newsletters and keep yourself informed on the latest technology news and new products from around the globe. Simply click this link to register here: Smart Cities Registration