sdk News - Page 4 of 51 - CNX Software

April 17, 2023April 17, 2023 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

Sony IMX500-based smart camera works with AITRIOS software

Raspberry Pi recently received a strategic investment from Sony (Semiconductor Solutions Corporation) in order to provide a development platform for the company’s edge AI devices leveraging the AITRIOS platform. We don’t have many details about the upcoming Raspberry Pi / Sony device, so instead, I decided to look into the AITRIOS platform, and currently, there’s a single hardware platform, LUCID Vision Labs SENSAiZ SZP123S-001 smart camera based on Sony IMX500 intelligent vision sensor, designed to work with Sony AITRIOS software. LUCID SENSAiZ Smart camera SENSAiZ SZP123S-001 specifications: Imaging sensor – 12.33MP Sony IMX500 progressive scan CMOS sensor with rolling shutter, built-in DSP and dedicated on-chip SRAM to enable high-speed edge AI processing. Focal Length – 4.35 mm Camera Sensor Format – 1/2.3″ Pixels (H x V) – 4,056 x 3,040 Pixel Size, H x V – 1.55 x 1.55 μm Networking – 10/100M RJ45 port Power Supply – PoE+ via […]

April 7, 2023 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

AMD Alveo MA35D media accelerator transcodes up to 32 1080p60 AV1 streams in real-time

AMD ALVEO MA350 AV1 real-time encoding card

AMD Alveo MA35D media accelerator PCIe card is based on a 5nm ASIC capable of transcoding up to 32 Full HD (1080p60) AV1 streams in real-time and designed for low-latency, high-volume interactive streaming applications such as watch parties, live shopping, online auctions, and social streaming. AMD says the Alveo MA35D utilizes a purpose-built VPU to accelerate the entire video pipeline, and the ASIC can also handle up to 8x 4Kp60, or 4x 8Kp30 AV1 streams per card. H.264 and H.265 codecs are also supported, and the company claims its “next-generation AV1 transcoder engines” deliver up to a 52% reduction in bitrate at the same video quality against “an open source x264 veryfast SW model”. AMD Alveo MA350 highlights: Auxiliary CPU – 2x 64-bit quad-core RISC-V to perform control and board management tasks AI Processor – 22 TOPS per card for AI-enabled “smart streaming” for video quality optimization Memory – 16GB […]

March 31, 2023 by Jean-Luc Aufranc (CNXSoft) - 3 Comments

SOPHON BM1684/BM1684X Edge AI computer delivers up to 32 TOPS, decodes up to 32 Full HD videos simultaneously

Firefly EC-A1684JD4 FD and EC-A1684XJD4 FD are nearly identical Edge AI embedded computers based on respectively SOPHON BM1684 and BM1684X Arm AI SoC delivering up to 32 TOPS of AI inference, and capable of decoding up to 32 H.265/H.264 Full HD videos simultaneously for video analytics applications. The BM1684(X) SoCs are equipped with eight Cortex-A53 cores clocked at 2.3 GHz to run Linux, and the systems come with up to 16GB RAM, 128GB flash, two Gigabit Ethernet ports to receive the video streams, one HDMI output up to 1080p30 for monitoring, as well as RS232 and RS485 DB9 connectors, and a few USB ports. Firefly EC-A1684JD4 FD and EC-A1684XJD4 FD specifications: SoC – SOPHGO SOPHON BM1684/BM1684X CPU – Octa-core Arm Cortex-A53 processor @ up to 2.3GHz TPU BM1684 64 NPU arithmetic units with each NPU containing 16 EU arithmetic units, or 1,024 EU in total Up to 17.6 TOPS (INT8), […]

March 22, 2023March 25, 2023 by Jean-Luc Aufranc (CNXSoft) - 9 Comments

$499 NVIDIA Jetson Orin Nano Developer Kit delivers up to 80x Jetson Nano Devkit performance

NVIDIA Jetson Orin Nano Developer Kit is an upgrade to the popular Jetson Nano Developer Kit that delivers 80 times the performance, up to 50 times the performance per watt, and gives the developers the ability to design entry-level AI-powered robots, smart drones, intelligent vision systems, and more. The Jetson Orin Nano has a similar form factor as the original Jetson Nano, but is fitted with a Jetson Orin Nano 8GB module with up to 40 TOPS AI performance, and is equipped with a DisplayPort video output, USB 3.2 Gen 2 ports, two M.2 Key M sockets for SSDs, Gigabit Ethernet, a pre-installed Wi-Fi module, and connectors for cameras. NVIDIA Jetson Orin Nano Developer Kit specifications (compared to Jetson Nano Developer Kit-B01) The new developer kit is supported by the Ubuntu 20.04-based NVIDIA JetPack 5.1.1 SDK, as well as application-specific frameworks such as NVIDIA Isaac ROS and DeepStream, which are […]

February 9, 2023 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

Android 14 developer preview brings enhancements to performance, privacy, security, and user customization

Google has just released the first developer preview of Android 14 with productivity improvements for developers, as well as enhancements to performance, privacy, security, and user customization. Android 14 aims to work better across devices and form factors with improved support for tablets and foldables and adds window size classes, sliding pane layout, Activity embedding, and box with constraints, etc… To help developers, the company also published “Get started with large screens” documentation and released a Cross-device SDK preview. The new version of the mobile operating systems also further streamlines background work to optimize system health and battery life and provide a better end-user experience. This is achieved through updates to JobScheduler and Foreground Services, optimized broadcasts most of which are internal to Android 14, and a new “Exact alarms” permission since it consumes more resources. Android 14 also introduced some user-facing changes with bigger fonts up to 200% with […]

January 19, 2023January 19, 2023 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

CHIPSEA CST85F01 480 MHz Cortex-M4 MCU supports dual-band WiFi 6 and Bluetooth 5.0 LE

CHIPSEA CST85F01 is an Arm Cortex-M4F microcontroller clocked at up to 480 MHz and designed for IoT applications with dual-band (2.4/5.0 GHz) WiFi 6 with TWT (Target Wake Time) support, Bluetooth 5.0 LE, and a range of I/Os. Following the recent availability of 2.4 GHz WiFi 6 IoT chips such as Espressif Systems ESP32-C6 and Bouffalo Lab BL616, CHIPSEA CST85F01 offers an alternative with dual-band WiFi 6 connectivity while we are waiting for the launch of the ESP32-C5 RISC-V microcontroller. CST85F01 specifications: CPU core – Arm Cortex-M4F CPU with MPU and FPU @ up to 480 MHz Memory – 992 KB SRAM, SDR/DDR PSRAM interfaces Storage – 752 KB ROM, 8 Mbit to 128 Mbit flash WiFi features 2.4GHz/5GHz Wi-Fi 6 Data rates up to 286.8 Mbps (Tx) and 229.4 Mbps (Rx) with 20/40 MHz bandwidth Rx sensitivity – -98dBm in 11b mode, -93.5dBm in MCS0 HT20 mode Tx power […]

January 9, 2023January 10, 2023 by Jean-Luc Aufranc (CNXSoft) - 3 Comments

Panthronics PTX30W is a 1W NFC wireless charging listener IC

Panthronics PTX30W is an NFC wireless charging listener chip that can receive up to 1W as per the NFC Wireless Charging (WLC) specification unveiled in 2020 and integrates a power management unit and a Li-Ion battery charger. Offered in a compact 3.2mm2 (1.78 x 1.78 mm) WL-CSP package, the PTX30W will enable small battery-powered products, such as smartwatches and other wearables, to be charged over NFC. The chip can operate in standalone mode or connected to a host microcontroller, and the company claims it is around four times smaller than existing designs based on multiple discrete components. PTX30W features and specifications: Integrated NFC Wireless Charging Listener device Efficient Active Rectifier RF interface according to Forum Type 2 Tag Li-Ion battery charger with charging current from 5mA to 250mA MCU LDO with 1.8V or 3.3V output, up to 50mA Embedded power negotiation logic High-efficiency NFC wireless charging listener IC with up […]

January 2, 2023January 7, 2023 by Jean-Luc Aufranc (CNXSoft) - 14 Comments

$150 Axelera M.2 AI accelerator module claims to deliver up to 214 TOPS

Axelera M.2 AI accelerator module is said to deliver up to 214 TOPS of AI inference and up to 3200 FPS with ResNet -50 in a compact M.2 2280 form factor. Few details are available at this time, but the module is based on the company’s Metis AIPU (AI Processing Unit) using in-memory computing based on arrays of SRAM memory devices used to “store a matrix and perform matrix-vector multiplications “in-place” without intermediate movement of data”. This technology is said to “radically” increase the number of operations per computer cycle with without suffering from issues such as noise or lower accuracy. The Metis AI platform delivers 50+ TOPS per core (RISC-V-controlled dataflow engine), offers FP32 equivalent accuracy, and has a 15 TOPS/W energy efficiency. The last point is impressive, but that means 214 TOPS won’t be reachable with the module shown above, since the M.2 form factor is designed to […]