Sipeed Maix-III devkit features AXERA AX620A SoC for AI vision applications

Maix III Board

Sipeed Maix-III development kit is composed of a core board equipped with AXERA AX620A AI processor and a carrier board exposing several interfaces such as Ethernet, USB, and camera connectors. resources. As noted in our earlier article about the AXERA AX620A chip, it is a  quad-core Arm Cortex-A7 chip with an NPU offering a maximum of 14.4 TOPS @ INT4 or 3.6TOPSs @ INT8, and as such, this development board mainly targets AI vision applications. The board is an upgrade to the earlier Speed Maix II with an Allwinner V831 processor. Thanks to the low power consumption of the AXERA AX620A SoC, the Sipeed Maix-III board can be powered directly through its USB port as the system consumers 5W ([email protected]) with the carrier board, the system-on-module, WiFi, Ethernet, a camera, and a 5-inch display. Using the display does not require am additional heatsink or fan, nor does it require an […]

Allwinner T3 automotive-grade processor powers industrial-grade SoM

Allwinner T3 SOM TLT3

Allwinner T3 is a quad-core Cortex-A7 automotive-grade processor that supports a wide industrial temperature range of -40°C to +85°C. After comparing the specifications of Allwinner T3, I think it is the same as Allwinner A40i, as Allwinner has different business units and the T-series is for the automotive-grade market, while the A-series has historically been for the tablet market, but is now also used in the industrial-grade market. Tronlong SOM-TLT3 and SOM-TLT3-B are Allwinner T3 system-on-modules (SoM) of Allwinner T3. Both modules have basically the same specifications but the SOM-TLT3 comes with castellated holes for soldering to the carrier board, while the SOM-TLT3-B features board-to-board connectors. Tronlong SOM-TLT3 / SOM-TLT3-B specifications: CPU – Allwinner T3 quad-core Arm Cortex-A7 @ 1.2 GHz with 32KB L1 I-cache + 32KB L1 D-cache, 512KB L2 cache. GPU – Arm Mali-400 MP2 with support for OpenGL ES 1.1/2.0, Open VG 1.1 Memory – 1/2GByte DDR3 […]

FNIRSI 1013D teardown and mini review – A portable oscilloscope based on Allwinner CPU & Anlogic FGPA

FNIRSI 1013D

The FNIRSI 1013D is a dual-channel flat-panel oscilloscope with a rich set of features. It is cost-effective and useful to people in the maintenance and R&D industries. Although it has been on the market for a few years, I purchased one, and I decided to introduce it and disassemble it to check out the hardware design. FNIRSI describes its oscilloscope as “small and portable”, so I assume it should include a lithium battery, a TFT LCD screen, a processor to handle the display, and another chip to process the digital signals. Let’s take it apart first to find out. FNIRSI 1013D oscilloscope unboxing FNIRSI 1013D package content: 1x FNIRSI 1013D oscilloscope 2x 100MHz matching probes (1X and 10X) 1x USB cable 1x Charging adapter 1x Instruction manual The FNIRSI 1013D adopts a 7-inch 800 x 480 resolution color TFT LCD with a capacitive touch screen. There are two input channels […]

AXERA AX620A 4K AI SoC delivers up to 14.4 TOPS for computer vision applications

AX620A

AXERA AX620A is a high-performance, low-power AI SoC with a quad-core Arm Cortex-A7 processor and a 14.4TOPs @ INT4 or 3.6TOPs @ INT8’s NPU that is slightly inferior to the Amlogic A311D, and mainly used for AI vision applications. With high computing power and built-in image processing capabilities, the AX620A can support a wide range of AI workloads. It also offers low power consumption with low standby power and fast wake-up, so the chip can be integrated into battery-powered products. AXERA AX620A specifications: CPU – Quad-core Arm Cortex-A7 @ 1.0 GHz with 32KB L1 I-cache + 32KB L1 D-cache per core, 256KB L2 cache, FPU and NEON NPU – 4 TOPS @ INT4, 3.6 TOPS @ INT8 with support for Imagenet, AlexNet, VGG, ResNet, GoogLeNet, Faster R-CNN, SSD, FPN, Yolo V3, and other neural networks. ISP Proton AI-ISP up to 4Kp30 4 channels of camera support up to 4x 1080p30 Support […]

DongshanPI-D1s – An Allwinner D1s RISC-V development board designed to teach programming

DongShanPI D1s

The DongshanPI-D1s development board is comprised of a soldered-on Allwinner D1s RISC-V system-on-module board (SoM) and a carrier board with two 40-pin headers and a 2.0mm dedicated header. This development board is specifically designed to teach programming with a focus on the RISC-V architecture. The development board was designed by 100ask. They previously designed the Dongshan NeZha STU a development board based on the Allwinner D1. The main difference between the two is that 100ask did not include the Ethernet and HDMI interfaces on the DongshanPI-D1s board. The pinout of the headers is also slightly different because they opted to make the headers compatible with the widely used 40-pin GPIO from Raspberry Pi single board computers. DongshanPI-D1s preliminary specifications: D1s Core Lite SoC – Allwinner D1s single-core XuanTie C906 64-bit RISC-V processor @ 1.0 GHz with with 32 KB I-cache + 32 KB D-cache Memory – 64 MB DDR2 (SIP) […]

BIGTREETECH CB1 – A Raspberry Pi CM4 compatible Allwinner H616 system-on-module

BIGTREETECH-CB1

The BIGTREETECH CB1 core board is an Allwinner H616 quad-core Arm Cortex-A53 system-on-module (SoM) that follows the Raspberry Pi CM4 form factor and was designed by BIGTREETECH, a company whose main business is related to 3D printer motherboards and their peripherals. The BIGTREETECH CB1 comes with 1GB RAM, an HDMI output interface, 2.4GHz WiFi, and 100Mbps Ethernet. The CB1 has better multimedia capability than the CM4 with support for 4Kp60 H.265/H.264 video decoding and 1080p60 H.264 video encoding, while the Broadcom BCM2711 processor on the CM4 can only handle 4Kp60 H.265 and 1080p60 H.264 video decoding, and 1080p30 H.264 video encoding. Some disadvantages include the lack of CSI and DSI interfaces on the Allwinner H616 system-on-module and the presence of only one video output interface against two for the Raspberry Pi CM4 module. BIGTREETECH CB1 specifications: SoC – Allwinner H616 quad-core Cortex-A53 @ 1.5GHz with Mali-G31 MP2 GPU with OpenGL 3.2, […]