Posted on by Jean-Luc Aufranc (CNXSoft) - 9 Comments on NVIDIA Jetson Nano 2GB Developer Kit Launched for $54 and up

NVIDIA Jetson Nano 2GB Developer Kit Launched for $54 and up

NVIDIA Jetson Nano Developer kit was introduced in March 2019 for $99. With a quad-core Cortex-A57 processor, a 128-core Maxwell GPU, and 4GB LPPDR4 RAM, it’s a great low-cost AI platform as we wrote in our Jetson Nano getting started guide where we show how to perform inferences on still images and an RTSP video stream. The company has now gone further in providing an affordable AI computer for developers with the launch of NVIDIA Jetson Nano 2GB Developer Kit with similar features except for the 2GB RAM, and pricing starting at $54 without a wireless adapter or $59 with 802.11b/g/n/ac WiFi 5 USB dongle. Pre-orders are open on sites like Amazon or Seeed Studio and shipping is scheduled to start at the end of the month. If we look at the photo above, there are very few differences against the $99 version, and indeed most of Jetson Nano 2GB Developer Kit specifications are the same as the original devkit: […]

Support CNX Software – Donate via PayPal or cryptocurrencies, become a Patron on Patreon, or buy review samples

Posted on by Jean-Luc Aufranc (CNXSoft) - 6 Comments on Mango-DVK OpenWrt Devkit Features 2.5GbE, WiFi 6 SoM powered by Qualcomm IPQ6000/IP6010 SoC

Mango-DVK OpenWrt Devkit Features 2.5GbE, WiFi 6 SoM powered by Qualcomm IPQ6000/IP6010 SoC

We’ve been covering some 802.11ax WiFI 6 embedded networking boards from Wallys Communications and Compex in the last few months, but 8devices has now entered the fray with Qualcomm IPQ6000/IPQ6010 powered Mango WiFi 6 system-in-module available in commercial or industrial temperature range, as well as as a development kit based on the module. Mango SoM supports dual-band 2×2 MiMo 802.11ax and multi-Gigabit Ethernet connectivity, features up to 2GB RAM, 256MB NAND flash, and runs OpenWrt Linux with target applications including routers, gateways, and access points. Mango SoM Specifications: SoC (one or the other) Qualcomm IPQ6000 quad-core Cortex-A53 processor @ up to 1.2 GHz with hardware NAT engine, HW crypto Qualcomm IPQ6010 quad-core Cortex-A53 processor @ up to 1.8 GHz with hardware NAT engine, HW crypto System Memory – 512MB DDR3L 512MB (up to 2GB) Storage – 32MB NOR FLASH, up to 256MB NAND flash Networking Wireless 2.4GHz 802.11 b/g/n/ac/ax WiFi 6 2×2 MU-MIMO, explicit beamforming; 22dBm; up to 573.5 Mbps […]

Support CNX Software – Donate via PayPal or cryptocurrencies, become a Patron on Patreon, or buy review samples

Posted on by Jean-Luc Aufranc (CNXSoft) - 4 Comments on New STM32H7 Cortex-M7 MCUs Clock at 550 MHz, Feature Octal SPI Flash and Ethernet Interfaces

New STM32H7 Cortex-M7 MCUs Clock at 550 MHz, Feature Octal SPI Flash and Ethernet Interfaces

STMicro launched STM32H7 single-core Cortex-M7 microcontroller family a while ago, followed by some dual-core Cortex-M7/M4 models, with most clocked up to 480 MHz. The company has now announced five faster parts clocked at up to 550 MHz with STM32H723, STM32H733, STM32H725, STM32H735, and STM32H730 which STMicro claims is “the fastest core speed in the market among MCUs that integrate Flash storage on-chip to run deeply embedded applications”. The embedded flash storage is important, as you may now NXP i.MX RT1170 Cortex-M7/M4 crossover processor can reach up to 1 GHz but does not include flash storage. Key differences in STM32H7 550 Mhz MCUs We’ve highlighted the five new microcontrollers in the table above, and beside the higher 550 MHz frequency delivering 2778 CoreMark and 1177 DMIPS, we can see those are the only parts that support both OctoSPI flash and Ethernet, and all five parts are designed for HMI applications with a TFT-LCD interface. The blog post following the announcement also […]

Support CNX Software – Donate via PayPal or cryptocurrencies, become a Patron on Patreon, or buy review samples

Posted on by Jean-Luc Aufranc (CNXSoft) - 2 Comments on FPGA Meets Breadboard with Mercury 2 Xilinx Artix-7 FPGA Development Board

FPGA Meets Breadboard with Mercury 2 Xilinx Artix-7 FPGA Development Board

While we now come across more and more FPGA development boards, very few are designed to be inserted into a breadboard. We’ve previously covered TinyFPGA BX, Fipsy, and QuickFeather FPGA breadboard-friendly boards. All those boards are however based on entry-level FPGAs like Lattice Semi ICE40 or QuickLogic EOS S3, and if you’d like a more powerful FPGA board that fits into a breadboard, MicroNova Mercury 2 development board may meet your requirements with a Xilinx Artix-7 FPGA. MicroNova Mercury 2 specifications: FPGA – Xilinx Artix-7A FPGA (XC7A35T) with 33,280 logic cells or XC7A100T with 101,440 logic cells System Memory – 4 Mbit (512K x 8-bit) asynchronous SRAM Storage – 32 Mbit SPI flash for configuration & user data Networking – On-board Microchip LAN8720A 10/100M Ethernet PHY USB – Dual-channel high-speed USB 2.0 micro USB port (ch A for config, ch B for user) 2x 32-pin headers Fast Ethernet signals 8-channel, 200 KSPS, 10-bit ADC via Microchip MCP-3008 2-channel, 225 kHz, […]

Support CNX Software – Donate via PayPal or cryptocurrencies, become a Patron on Patreon, or buy review samples

Posted on by Jean-Luc Aufranc (CNXSoft) - 2 Comments on Heart Rate Development Kit Features ESP32 WiSoC, MAX30102 Pulse Oximeter & HR sensor

Heart Rate Development Kit Features ESP32 WiSoC, MAX30102 Pulse Oximeter & HR sensor

LilyGO has launched a fair amount of ESP32 boards with some less common features like Ethernet and OLED Display, cellular connectivity, LoRa & GPS, PoE, watch form factor, and more. Their latest product is somewhat different from their earlier boards with the TTGO Heart rate programming development kit being closer to a reference design since it comes with an enclosure similar to commercial products. The heart rate devkit features ESP32 WiFi and Bluetooth SoC, a small LCD, as well as a Maxim Integrated MAX30102 pulse oximeter and heart-rate sensor.   TTGO heart rate development kit specifications: SoC – Espressif Systems ESP32 dual-core Tensilica SoC with WiFi 4 & Bluetooth connectivity Display – 0.96-inch IPS LCD via ST7735 SPI controller Sensors MAX30102 heart rate blood oxygen I2C sensor MPU6050 6-axis accelerometer and gyroscope (I2C) Debugging – Micro USB port via CP2104 USB to TTL bridge Misc -PCF8563 RTC, user button Battery – 200 mAh; charging voltage and current: 5V, 500mah via […]

Support CNX Software – Donate via PayPal or cryptocurrencies, become a Patron on Patreon, or buy review samples

Posted on by Jean-Luc Aufranc (CNXSoft) - 7 Comments on PicoCore RT1 System-on-Module Runs FreeRTOS on NXP i.MX RT1050 Crossover Processor

PicoCore RT1 System-on-Module Runs FreeRTOS on NXP i.MX RT1050 Crossover Processor

F&S Elektronik Systeme PicoCore RT1 is a System-on-Module powered by NXP i.MX RT1050 crossover processor designed to offer the real-time capabilities of microcontrollers and the performance of application processors thanks to an Arm Cortex-M7 core clocked at up to 600 MHz. The 40×35 mm module is also equipped with 32MB SDRAM, 256MB “high-reliability QSPI NOR flash, and 16KB EEPROM, and offers various interfaces including Ethernet, RGB LCD, and USB via board-to-board connectors. PicoCore RT1 (V1) SoM specifications: SoC – NXP i.MX RT1050 Arm Cortex-M7 processor @ up to 600 MHz with 32KB L2 cache, 2D graphics accelerator System Memory – 32MB SDRAM Storage – 256MB QSPI Flash, 16KB EEPROM; optional NAND flash 2x 80-pin board-to-board connector with Storage – SD Card Display – 16-bit RG interface, analog resistive and PCAP touch via I2C Networking – 1x 10/100M Ethernet Audio – Line-In, Line-Out, microphone, headphone, I2S USB – 1x USB 2.0 host, 1x USB 2.0 OTG Serial – Up to 2x […]

Support CNX Software – Donate via PayPal or cryptocurrencies, become a Patron on Patreon, or buy review samples

Posted on by Jean-Luc Aufranc (CNXSoft) - No Comments on Nordic Semi nRF21540 RF front end module Extends the Range of nRF52/nRF53 Bluetooth SoC’s

Nordic Semi nRF21540 RF front end module Extends the Range of nRF52/nRF53 Bluetooth SoC’s

A little while ago, we wrote about RFCat N32 board integrating Meshtek-H52 module with nRF52832 2.4GHz multi-protocol SoC and a Skyworks power amplifier increasing power by up to 30 times to extend the wireless range to up to 200 meters. It turns out Nordic Semi outed its own range extender module late last year with nRF21540 RF front end module (FEM) that integrates a power amplifier (PA) and low noise amplifier (LNA) optimized to boost the link budget of the nRF52 and nRF53 multi-protocol wireless SoCs. nRF21540 key features and specifications: Supported wireless standards Bluetooth Low Energy (incl. Bluetooth mesh) Thread and Zigbee (802.15.4) Proprietary 2.4 GHz Output Power – Adjustable in small increments up to +21 dBm Receive Gain – +13 dB with 2.5 dB noise figure Antennas – 2x antenna ports for antenna diversity Host Interface – I/Os, SPI, or a combination Supply Voltage – 1.7 V to 3.6V Current consumption Tx tuned to +20 dBm: 115 mA […]

Support CNX Software – Donate via PayPal or cryptocurrencies, become a Patron on Patreon, or buy review samples

Posted on by Jean-Luc Aufranc (CNXSoft) - 8 Comments on J721E DRA829/TDA4VM/AM752x – Texas Instruments Cortex-A72 based Monster SoC’s

J721E DRA829/TDA4VM/AM752x – Texas Instruments Cortex-A72 based Monster SoC’s

Texas Instruments unveiled their first 64-bit processor in 2018 with TI AM654 “Keystone III” quad-core Arm Cortex-A53 + dual lockstep Cortex-R5F processor designed for general embedded and industrial applications. The company is now working on a more powerful processor with J721E SoC with Cortex-A72 cores belonging to the K3 Multicore SoC architecture platform appearing in TI Linux git repository. Ti J721E is a monster of an SoC, not necessarily in terms of CPU processing power, but it has an amazing amount of features and peripherals. I’m not going to list all specifications of this monster SoC, and we’ll do with J721E highlights instead: CPU Dual Cortex-A72 up to 2.0 GHz in a single cluster Up to three clusters of lockstep capable dual Cortex-R5F MCUs @ 1.0 GHz AI Accelerator / DSP Deep-learning Matrix Multiply Accelerator (MMA) @ up to 1.0 GHz (8 TOPS for 8-bit inference) C7x floating-point Vector DSP @ up to 1.0 GHz (80 GFLOPS) Up to two […]

Support CNX Software – Donate via PayPal or cryptocurrencies, become a Patron on Patreon, or buy review samples