Axiomtek AIE110-XNX is an affordable Edge AI Developer Kit based on NVIDIA Jetson Xavier NX

Axiomtek AIE110-XNX devkit

Axiomtek AIE110-XNX Edge AI developer kit provides up to 21 TOPS of accelerated AI computing performance thanks to an NVIDIA Jetson Xavier NX system-on-module with a 6-core NVIDIA Carmel ARM v8.2 64-bit CPU and a 384-core NVIDIA Volta GPU with 48 Tensor Cores. The new devkit serves as a relatively low-cost cost development kit with no minimum order, instead of getting full-feature embedded computers such as the company’s AIE900-XNX Edge AI system also based on the same Jetson Xavier NX module. Axiomtek AIE110-XNX specifications: NVIDIA Jetson Xavier NX system-on-module with CPU – 6-core NVIDIA Carmel Armv8.2 64-bit CPU with 6 MB L2 + 4 MB L3 cache GPU – 384-core NVIDIA Volta GPU with 48 Tensor Cores AI Accelerator – 2x NVDLA System Memory – 8GB 128-bit LPDDR4x onboard Storage – 16GB eMMC flash Storage – M.2 Key M 2280 with PCIe x4 NVMe SSD slot, microSD slot Video Output […]

Epi 32U4 – An Arduino-compatible USB-C development board with ATmega32U4 MCU (Crowdfunding)

Epi 32U4 Arduino USB Type-C development board

Epi 32U4 is a tiny Arduino-compatible development board based on a Microchip ATmega32U4 8-bit AVR microcontroller with just a USB-Type-C port and 23 I/O pins accessible through castellated holes that can be soldered to headers or another PCB. The board also includes some protection such as a fuse, ESD protection, a ferrite bead, and a low-pass filter for the analog reference voltage. Ping Hobbyelektronik, the developer, says it’s especially useful for compact USB keyboards, but it also has more I/O ports than a standard Pro Micro board, so it’s appropriate for a range of other projects as well. Epi 32U4 board specifications: MCU – Microchip ATmega32U4 8-bit AVR microcontroller @ 8MHz or 16 MHz with 32 KB flash, 2.5 KB SRAM, 1 KB EEPROM USB 1x USB Type-C port for power and programming USB D+ and D- lines are broken out ESD and surge protection on USB data lines and […]

Silicon Labs FG28 Sub-GHz wireless and 2.4 GHz BLE SoC supports Amazon Sidewalk, Wi-SUN, etc…

Silicon Labs FG28 block diagram

Silicon Labs dual-band FG28 Cortex-M33 SoC comes with sub-Gigahertz and 2.4 GHz Bluetooth LE radios that support long-range networks and protocols like Amazon Sidewalk, Wi-SUN, and other proprietary protocols, as well as a built-in AI/ML accelerator for machine learning inference and SiLabs Secure Vault technology. The chip looks to be an update to the sub-GHz FG25 Cortex-M33 microcontroller with an additional 2.4 GHz BLE radio, but somehow less program memory (up to 256KB) and storage (up to 1024 KB). The peripherals implemented in the chip are almost the same, as the microcontroller gains support for predictive maintenance with a small matrix vector processor for AI/ML acceleration. Silicon Labs FG28 (EFR32FG28) specifications: MCU core – Arm Cortex-M33 @ 78 MHz with DSP instructions and FPU On-chip memory – Up to 256 kB RAM data memory On-chip storage – Up to 1024 kB flash program memory AI accelerator – Matrix Vector Processor […]

Pico-Ice board combines Raspberry Pi RP2040 MCU with Lattice ICE40 UltraPlus 5K FPGA

Pico Ice Raspberry Pi RP2040 Lattice FPGA board

tinyVision.ai Pico-Ice is a development board with a Raspberry Pi RP2040 MCU and a Lattice ICE40 UltraPlus 5K FPGA connected through an 8-bit bus. The Arm Cortex-M0+ microcontroller provides the clock for the FPGA and can program the FPGA directly or the dedicated FPGA flash using a drag-drop of a UF2 file. Just a few days ago we wrote about the LILYGO T-FPGA board that combines an ESP32-S3 wireless MCU with a Gowin FPGA connected through a 6-bit bus, and the Pico-Ice board provides a similar option with different chips and without wireless connectivity. Pico-Ice specifications: MCU – Raspberry Pi RP2040 dual-core Cortex-M0+ microcontroller @ 133 MHz with 264KB SRAM with all pins exposed FPGA – Lattice UltraPlus iCE40UP5K FPGA with 5.3K LUTs, 1Mbit SPRAM, 120Kbit DPRAM, 8x multipliers with all pins brought out Memory & storage chips MCU – 4MB QSPI flash FPGA – 4MB QSPI Flash, 8MB low […]

Khadas VIM4 Amlogic A311D2 SBC gets 3.2 TOPS NPU

Khadas VIM4 Amlogic A311D2-N0D SBC

First introduced in 2022, the Khadas VIM4 SBC will now sell with an Amlogic A311D2-N0D Cortex-A73/A53 processor equipped with a 3.2 TOPS NPU, and the version without AI accelerator will be phased out. I reviewed the Khadas VIM4 with both Android 11 and Ubuntu 22.04 last year, and really liked the user experience notably thanks to the OOWOW system that makes it really easy to try different operating systems downloaded from Khadas servers, although some features like support for GPU acceleration still needed to be worked on at the time. One thing missing compared to the previous generation Khadas VIM3 SBC was an on-chip NPU to accelerate AI workloads, and the latest revision of the Amlogic A311D2 processor (with N0D suffix) fixes that on the Khadas VIM4. “Khadas VIM4 with NPU” specifications: SoC – Amlogic A311D2-N0D octa-core processor with 4x Arm Cortex-A73 cores @ up to 2.2 GHz and 4x […]

MangoPi mCore-R818 module powers CyberPad 3.1-inch handheld android “tablet”

CyberPad Components

mCore-R818 is the first time that MangoPi adopts the design combination of a core-lite module and a carrier board. As its name suggests, it is an AllWinner R818 SoM whose SoC integrates an Imagination PowerVR GE8300 GPU for UI rendering, can drive MIPI DSI, LVDS, and RGB displays, as well as cameras through a MIPI CSI interface, 8MP/5MP/2MP interfaces. The Allwinner R818 system-on-module powers a feature-rich carrier board as well as an upcoming Cyberpad Android “tablet” with a 3.1-inch display. MangoPi mCore-R818 The package design of the processor itself is small, so the MCore-R818 core board is only 3x3cm in size, but still contains four components with the Allwinner R818, the eMMC flash, LPDDR4 memory, and the AXP717 PMU. MangoPi provides two hardware configurations: 2GB DDR with 16GB eMMC flash, and 4GB DDR with 32GB eMMC flash. MCore-R818 Core Lite Specifications: SoC – Allwinner R818 CPU – Quad-core Arm Cortex-A53 […]

ARIES v3.0 development board features India-made VEGA RISC-V processor

Made-in-India VEGA RISC-V board

ARIES v3.0 is an Arduino-inspired development board with the made-in-India 100 MHz THEAJS32 ASIC with the VEGA ET1031 RISC-V microprocessor, 256KB of internal SRAM, and various I/Os. Not to be confused by the recent FIVEberry RISC-V SBC made by ARIES Embedded… We first wrote about VEGA RISC-V processors designed in India in February 2022, and at the time, the government worked on five RISC-V processor designs ranging from a single-core 32-bit RISC-V microcontroller-class processor to a Linux-capable quad-core 64-bit out-of-order processor, with the entry-level THEJAS32 and THEJAS64 SoCs based on respectively VEGA ET1031 and VEGA AS1061 cores running on an ARTY A7 FPGA board. But good progress has been made since then, and they have tapped out the THEJAS32 processor and manufactured some development boards such as the ARIES v3.0. ARIES v3.0 development board specifications: MCU – THEJAS32 VEGA ET1031 32-bit 3-stage in-order RISC-V ( RV32IM) microcontroller @ 100 MHz […]

LILYGO T-FPGA devkit combines ESP32-S3 WiSoC with Gowin GW1NSR-4C FPGA

ESP32-S3 Gowin FPGA devkit

LILYGO T-FPGA development kit is comprised of an M.2 module with an ESP32-S3 WiFi and Bluetooth wireless microcontroller and a Gowin GW1NSR-4C FPGA with 4,608 LUTs, and a carrier board that gives easy access to I/Os and comes with a 18650 battery holder. The module comes with a USB -C OTG port connected to the ESP32-S3, and a USB-C FPGA port for the FPGA. The “T-FPGA shield” carrier board includes four Pmod interfaces and a STEMMA Qt/Qwiic connector for expansion, plus several pins for 3.3 and GND, and another 20-pin I/O header. LILYGO T-FPGA specifications: Wireless MCU – Espressif Systems ESP32-S3R8 dual-core Tensilica LX7 @ up to 240 MHz with vector instructions for AI acceleration, 512KB RAM, 8MB PSRAM, wireless connectivity Storage – 16MB flash FPGA – Gowin Semiconductors GW1NSR-LV4CQN48PC6/I5 with 4,608 LUT4, 256Kb flash, 64Mb PSRAM, 64Mb Hyperflash, 32Mb NOR flash Connectivity via ESP32-S3 2.4 GHz 802.11 b/g/n Wi-Fi […]