NVIDIA Jetson Xavier NX SoM Delivers up to 21 TOPS for AI Workloads at the Edge

NVIDIA Jetson Xavier NX

NVIDIA has just announced Jetson Xavier NX system-on-module, with the company claiming it is the “world’s smallest, most powerful AI supercomputer for robotic and embedded computing devices at the edge” with a 70x45mm “Jetson Nano” form factor, and delivering either up to 14 TOPS at 10 Watts or 21 TOPS at 15 Watts. The company expects the module to be used in small commercial robots, drones, intelligent high-resolution sensors for factory logistics and production lines, optical inspection, network video recorders, portable medical devices, and other industrial IoT systems. Jetson Xavier NX specifications: SoC – NVIDIA Xavier with 6-core NVIDIA Carmel ARM v8.2 64-bit CPU, 6MB L2 + 4MB L3 caches, and a 384-core NVIDIA Volta GPU with 48 Tensor Cores, 2x NVDLA deep learning accelerators delivering up to 21 TOPS at 15 Watts System Memory – 8 GB 128-bit LPDDR4x @ 51.2GB/s Storage – 16 GB eMMC 5.1 flash Video Encode 2x464MP/sec 2x 4K @ 30 (HEVC) 6x 1080p @ …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

RAKWireless RAK4260 is a Tiny LoRAWAN Module based on Microchip SAMR34 LoRa SiP

RAK4260 EVB

Rakwireless has just announced a new module part of their LPWAN family:  RAK4260 LoRaWAN module based on Microchip ATSAMR34J18B LoRa SiP and at just 15x15x1.2 mm, one of the smallest LoRaWAN modules in the market. The new module is cheaper than the company’s earlier RAK811 module and consumes less power at just 790 nA in sleep mode. The company also provides a RAK4260 evaluation board with easy access to GPIOs and serial interfaces. RAK4260 LoRaWAN Module Specifications: SiP – Microchip ATSAMR34J18 SiP with SAM L21 Arm Cortex M0+ MCU @ 48 MHz, up to 40 KB RAM, up to 256 KB Flash LoRa Connectivity Frequency Range – 862 to 1020 MHz High level of accuracy and stability (32MHz TXCO) Max Tx Power: 20dBm; Max Sensitivity: -148dBm; Rx Current: 17mA (typical) Compliant with LoRaWan 1.0.2 Expansion – Castellated holes with I2C, SPI, ADC, UART, GPIOs Power Consumption Low RX current of 13.6mA 790 nA in sleep mode Dimensions – 15×15 mm …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

Forlinx NXP LS1043A & LS1046A Networking SBC’s Support 10Gbps Ethernet

NXP LS1046A Networking SBC

NXP LS1043A quad-core Cortex-A53 communication processor was introduced in 2014, while NXP LS1046A quad-core Cortex-A72 SoC was launched about 18 months later. Both are designed for networking equipment such as CPE (Customer Premise Equipment), routers, NAS, gateways, as well as single board computers and include one or two 10 GbE interfaces. Forlinx Embedded has decided to leverage those two processors in their OK1043A-C and OK1046A industrial-grade single board computers designed for networking applications. Both boards are comprised of the same baseboard and only differ by their COM-Express Mini Type 10 module which comes with the processor, memory, and flash storage. Specifications: COM Express Mini System-on-Module (one or the other) FET1046A-C NXP LS1046A SoM CPU – NXP LS1046A quad-core  Cortex-A72 processor @ up to  1.8GHz System Memory –  2GB DDR4 RAM Storage –  8GB eMMC flash + 16MB QSPI NOR Flash Voltage Input – 12V Temperature Range – -40℃ to +75℃      Dimensions – 84 x 55mm FET1043A-C LS1043A SoM …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

Intrinsyc Unveils Open-Q 845 µSOM and Snapdragon 845 Mini-ITX Development Kit

Open-Q 845 µSOM Development Kit

Intrinsyc introduced the first Qualcomm Snapdragon 845 hardware development platform last year with its Open-Q 845 HDK designed for OEMs and device makers. But the company has now just announced a solution for embedded systems and Internet of Things (IoT) products with Open-Q 845 micro system-on-module (µSOM) powered by the Snapdragon 845 octa-core processor, as well as a complete development kit featuring the module and a Mini-ITX baseboard. Open-Q845 µSOM Specifications: SoC – Qualcomm Snapdragon SDA845 octa-core processor with 4x Kryo 385 Gold cores @ 2.649GHz + 4x Kryo 385 Silver low-power cores @ 1.766GHz cores, Hexagon  685 DSP, Adreno 630 GPU with OpenGL ES 3.2 + AEP (Android Extension Pack),  DX next, Vulkan 2, OpenCL 2.0 full profile System Memory – 4GB or 6GB dual-channel high-speed LPDDR4X SDRAM at 1866MHz Storage – 32GB or 64GB UFS Flash Storage Connectivity Wi-Fi 5 802.11a/b/g/n/ac 2.4/5Ghz 2×2 MU-MIMO (WCN3990) with 5 GHz external PA & U.FL antenna connector Bluetooth 5.x Audio & …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

TechNexion XORE is a tiny NXP i.MX 8M Mini LGA System-on-Module

TechNexion XORE LGA System-on-Module

There are several ways to design a system-on-module to carrier board interface, and the most common solutions are edge connectors (e.g. SO-DIMM), board-to-board connectors (placed under the module), and castellated holes where the board is soldered directly to the baseboard. Another less common method is to design an LGA (Land Grid Array) module, which also have to be soldered straight unto the carrier board, but enables much more compact system-on-modules, and that’s exacty what TechNexion has done with their XORE family of LGA system-on-module currently powered by the 14-nm NXP i.MX 8M Mini processor. XORE-IMX8M-Mini Module TechNexion XORE module specifications: SoC (one of the other) NXP i.MX8M Mini Solo single Arm Cortex-A53  @ 1.8 GHz + M4 processor, Vivante GC7000Lite 3D GPU NXP i.MX8M Mini Dual 2x Arm Cortex-A53 @ 1.8 GHz + M4 processor, Vivante GC7000Lite 3D GPU NXP i.MX8M Mini Quad 4x Arm Cortex-A53 @ 1.8 GHz + M4 processor, Vivante GC7000Lite 3D GPU System Memory + storage – …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

Geniatech SOM3399 RK3399 SoM Complies with 96Boards System-on-Module Specification

CBD96-3399

96boards System-on-Module specification was introduced last April together with their first SOM-CA (2x board-to-board connectors) and SOM-CB (4x board-to-board connectors), namely TB-96AI & TB-96AIoT modules powered by respectively AI capable Rockchip RK1808 and RK3399Pro processors that would fit into a compatible baseboard. Those started to ship early this summer, and there’s now a new module that complies with 96Boards SoM specification courtesy of Geniatech SOM3399 powered by a Rockchip RK3399 hexa-core processor without the extra NPU / AI accelerator found in the aforementioned modules. Geniatech SOM3399 Specifications: SoC – Rockchip RK3399 dual-core Cortex-A72 up to 1.8GHz, quad-core Cortex-A53 up to 1.4GHz,  Mali-T860 MP4 GPU with support for OpenGL ES1.1/2.0/3.0, OpenCL1.2, DirectX11.1 System Memory –  2GB LPDDR4 (4GB as option) Storage – 8GB eMMC flash (16, 32, and 64GB as option) Board-to-board connector – 4x 100-pin high-speed connectors (X1, X2, X3, and X4 as defined by the spec) with USB 3.0, PCIe, Gigabit Ethernet, MIPI  DSI/eDP/LVDS, HDMI, MIPI CSI, I2S, I2C, …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

Boardcon Idea3399 Features-Rich SBC Comes with M.2 NVMe SSD and 4G LTE PCIe Sockets

SBC M.2 NVMe SSD & mPCIe 4G LTE card

Back in 2017, Boardcon introduced EM3399 single board computer powered by a Rockchip RK3399 processor through the company’s PICO3399 SO-DIMM system-on-module. They’ve now designed another RK3399 SBC – Idea3399 – comprised of a baseboard and module, but instead of re-using the SO-DIMM module, CM3399 system-on-module with castellated holes was used instead. The new board comes with many of the same features as their first board but adds an M.2 NVMe SSD slot and mPCIe socket for 4G LTE modem on the back of the board. CM3399 SoM Key features and specifications: SoC – Rockchip RK3399 dual Cortex-A72 @ 1.8GHz + quad Cortex-A53 @ 1.4GHz System Memory – 4GB LPDDR4 Storage – 8GB eMMC flash 202 castellated pin (1.3mm pitch) with  USB2.0 host, USB3.0 host, USB OTG, UART, MIPI, Ethernet, SPI, HDMI out, I2C, I2S, PCI Express, SDIO, SD/MMC, eDP, Camera, PWM, ADC IN, etc… Power – Supply Voltage: 5V; RK808 PMU Dimensions – 55  x 50mm (8 Layers, complying with …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

Raspberry Pi CM3+ based EagleEye Smart Camera Works with OpenCV and LabVIEW NI Vision

Raspberry Pi CM3 Industrial Smart Camera

We previously covered Q-Wave Systems’ Melon S3 board combining a Xilinx Spartan 3E FPGA with ESP8266, but the Thai company is back is a completely different product: EagleEye Smart Camera. The board is powered by Raspberry Pi Compute Module 3+ (CM3+) with 16GB or 32GB flash, and equipped with a 5 MP camera for machine vision and robotics applications.  There are two version of the board Uno and Industrial with the latter adding 24V digital input and outputs, circuit protections and support for industrial temperature range. EagleEye smart camera key features & specifications: SoM – Raspberry Pi CM3+ with Broadcom BCM2837B0 quad core cortex-A53 processor, 1 GB RAM, and 16GB or 32 GB flash Camera – 5 MP OV5647 image sensor,  CS/M12 lens holder + 4mm CS lens Video Output – mini HDMI port Networking – 10/100M Ethernet USB – 1x USB 2.0 host port up to 1.2A Expansion Uno – 4x 3.3V digital input; 4x 3.3V digital output; both …

Support CNX Software – Donate via PayPal or become a Patron on Patreon