som

Posted on by Jean-Luc Aufranc (CNXSoft) - 13 Comments on MediaTek Genio 1200 system-on-module powers Cortex-A78/A55 AIoT & robotics development kit

MediaTek Genio 1200 system-on-module powers Cortex-A78/A55 AIoT & robotics development kit

I-Pi SMARC 1200

ADLINK Technology’s SMARC 2.1 compliant LEC-MTK-I12000 system-on-module (SoM) features the MediaTek Genio 1200 octa-core Cortex-A78/A55 AIoT processor coupled with up to 8GB RAM and 256GB UFS storage, and powers the I-Pi SMARC 1200 development kit designed for robotics and AIoT applications. LEC-MTK-I1200 MediaTek Genio 1200 system-on-module LEC-MTK-I1200 SoM specifications: SoC – MediaTek Genio 1200 (MT8395) octa-core processor with 4x Cortex-A78 cores @ 2.2GHz, 4x Cortex-A55 cores @ 2.0GHz with penta-core Arm Mali-G57 GPU @ 880MHz with support for OpenGL ES 3.2/2.0/1.1, Vulkan 1.1/1.0, OpenCL 2.2, 5.0 TOPS NPU, HiFi 4 audio DSP, etc… System Memory – 4 or 8 GB LPDDR4X Storage – 32, 64, 128, or 256 GB UFS storage, compatible with UFS gear 2.1 Wireless – 802.11b/g/n/ac WiFi 5 2×2 MU-MIMO, Bluetooth 5.0 314-pin MXM connector with Storage – 1x SDIO (4-bit) compatible with SD/SDIO standard, up to version 3.0 Display HDMI 2.0b up to 4Kp60 Dual-channel LVDS […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 38 Comments on Lichee Pi 4A RISC-V SBC takes on Raspberry Pi 4 with TH1520 processor

Lichee Pi 4A RISC-V SBC takes on Raspberry Pi 4 with TH1520 processor

Lichee Pi 4A RISC V SBC

Lichee Pi 4A is a single board computer (SBC) powered by Alibaba T-Head TH1520 quad-core RISC-V Xuantie C910  processor @ 2.0 GHz with an Imagination GPU and a 4 TOPS NPU for AI that can compete against the Raspberry Pi 4 in terms of performance and features. We previously mentioned the Lichee Pi 4A (LPi4A) in our article about the Sipeed LM4A RISC-V system-on-module, but at the time we only had some benchmarks for the board and no photos and specifications about the SBC. Sipeed has now released photos, published detailed specifications, and is taking orders for the board. So let’s have a closer look. Lichee Pi 4A specifications against the ones for Raspberry Pi 4.   The TH1520 used to be advertised at up to 2.5 GHz, but it seems to have now come down to 2.0 GHz. Software support will include Debian, OpenWrt, and Android which should then […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 7 Comments on Banana Pi BPI-W3 SBC features Rockchip RK3588 SoM, M.2 NVMe socket, 2.5GbE, HDMI output and input

Banana Pi BPI-W3 SBC features Rockchip RK3588 SoM, M.2 NVMe socket, 2.5GbE, HDMI output and input

Banana Pi BPI-W3

Banana Pi BPI-W3 is another Rockchip RK3588 SBC (single board computer) with 8GB RAM, 32GB eMMC flash, an M.2 NVMe socket, 2.5GbE networking, two HDMI 2.1 output ports, one HDMI 2.0 input port, and other features. We first recovered the board last summer, but at the time the RK3588 SBC came with a PCIe x4 slot and dual Gigabit Ethernet, and the company has now redesigned the board without either using instead a Rockchip RK3588 system-on-module as part of the design and implemented 2.5GbE and M.2 NVMe SSD support with the available PCIe interfaces. Banana Pi BPI-W3 specifications: SoM – BPI-RK3588 core board SoC- Rockchip RK3588 octa-core processor with 4x Cortex-A76 cores @ up to 2.4 GHz, 4x Cortex-A55 cores @ 1.8 GHz, an Arm Mali G610MC4 GPU, a 6 TOPS NPU, 8Kp60 H.265/VP9/AVS2 10-bit decoder, 8Kp30 H.265/H.264 encoder System Memory – 8GB LPDDR4 (options for 4GB to 32GB) Storage […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 5 Comments on Mixtile Core 3588E Rockchip RK3588 system-on-module works with NVIDIA Jetson TX2 NX carrier boards

Mixtile Core 3588E Rockchip RK3588 system-on-module works with NVIDIA Jetson TX2 NX carrier boards

Rockchip RK3588 SoM NVIDIA Jetson TX2 NX alternative

Mixtile Core 3588E is a system-on-module powered by a Rockchip RK3588 SoC and equipped with the 260-pin SO-DIMM connector found in the NVIDIA Jetson TX2 NX (and Jetson Nano, Xavier NX, Orin Nano modules). We’ve seen several companies make Raspberry Pi CM4 alternatives such as Radxa CM3 or Pine64 SoQuartz, and Mixtile even has its own Core 3568M SoM that can be used as a drop-in replacement for the Raspberry Pi CM4. But I had not seen companies try to make system-on-modules based on NVIDIA Jetson’s 260-pin SO-DIMM edge connector, and that’s exactly what the Mixtile Core 3588E has to offer. Mixtile Core 3588E specifications: SoC – Rockchip RK3588 CPU – Octa-core processor with 4x Arm Cortex-A76 cores @ up to 2.4 GHz, 4x Arm Cortex-A55 cores GPU – Arm Mali-G610 MP4 GPU with support for OpenGL ES3.2, OpenCL 2.2, Vulkan1.1 AI accelerator – 6 TOPS NPU with support for […]

Posted on by Leo Zachary - No Comments on 100ASK-V853-Pro – A feature-rich Allwinner V853 board designed for AI vision applications

100ASK-V853-Pro – A feature-rich Allwinner V853 board designed for AI vision applications

100ASK V853 Pro

The 100ASK-V853-Pro board is a development kit consisting of an Allwinner V853 system-on-module board (SoM) and a feature-rich carrier board with a large number of interfaces. Allwinner V853 supports up to 1TOPS of NPU computing power and is mainly for AI vision application development. The core board contains a DDR and eMMC as well as a PMU chip (AXP2101) and is connected to the carrier board through a board-to-board connector. All the functional resources of the V853 are drawn out through the carrier board. The carrier board comes with 2-channels CSI camera interfaces as well as RGB and MIPI DSI display interfaces. Although 1 TOPS of AI computing power is not outstanding, the NPU can still be used to accelerate AI vision applications at the edge. The board also comes with four USB 2.0 ports (two Type-A, two Type-C), an 100Mbps Ethernet port, a 22-pin header for expansion, and five […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 8 Comments on AAEON BOXER-8224AI – An NVIDIA Jetson Nano AI Edge embedded system for drones

AAEON BOXER-8224AI – An NVIDIA Jetson Nano AI Edge embedded system for drones

BOXER 8224AI board AI Edge NVIDIA Jetson Nano embedded system

AAEON BOXER-8224AI is a thin and lightweight AI edge embedded system solution based on NVIDIA Jetson Nano system-on-module and designed for drones, or other space-constrained applications such as robotics. AAEON BOXER products are usually Embedded Box PCs with an enclosure, but the BOXER-8224AI is quite different as it’s a compact and 22mm thin board with MIPI CSI interfaces designed to add computer vision capability to unmanned areal vehicles (UAV), as well as several wafers for dual GbE, USB, and other I/Os. BOXER-8224AI specifications: AI Accelerator – NVIDIA Jetson Nano CPU – Arm Cortex-A57 quad-core processor System Memory – 4GB LPDDR4 Storage Device – 16GB eMMC 5.1 flash Dimensions – 70 x 45 mm Storage – microSD slot Display Interface – 1x Mini HDMI 2.0 port Camera interface – 2x MIPI CSI connectors Networking 2x Gigabit Ethernet via wafer connector (1x NVIDIA, 1x Intel i210) Optional WiFi, Bluetooth, and/or cellular connectivity […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 9 Comments on Cool Pi CM5 evaluation board features Rockchip RK3588/RK3588J system-on-module

Cool Pi CM5 evaluation board features Rockchip RK3588/RK3588J system-on-module

Cool Pi CM5 SBC

Cool Pi CM5 is a system-on-module based on Rockchip RK3588 or RK3588J (industrial temperature range) octa-core Arm Cortex-A76/A55 processor with up to 32GB RAM, 256GB eMMC flash, offered with a development board with dual GbE, two 8K HDMI 2.1 ports, a PCIe 3.0 x2 slot, M.2 NVMe and SATA storage, and more… At the end of last year, we wrote about the Cool Pi 4 as a much faster alternative to Raspberry Pi 4 SBC, and the company has now launched a Raspberry Pi Compute Module 4 and Radxa CM5 alternative with the Cool Pi CM5 MXM 3.0 system-on-module that they offer along with a feature-rich evaluation board (EVB). Cool Pi CM5 EVB specifications: System-on-Module SoC – Rockchip RK3588(J) octa-core processor with 4x Cortex‑A76  cores @ up to 2.4GHz, 4x Cortex‑A55 core @ 1.8GHz Arm Mali-G610 MP4 “Odin” GPU Video decoder – 8Kp60 H.265, VP9, AVS2, 8Kp30 H.264 AVC/MVC, 4Kp60 […]

Posted on by Jean-Luc Aufranc (CNXSoft) - 1 Comment on IP67-rated CM4 AI camera uses Raspberry Pi Compute Module 4 for computer vision applications

IP67-rated CM4 AI camera uses Raspberry Pi Compute Module 4 for computer vision applications

EDATEC ED-AIC2020

EDATEC ED-AIC2020 is an IP67-rated, Raspberry Pi CM4-based industrial AI camera equipped with a fixed or liquid lens and LED illumination that leverages the Raspberry Pi Compute Module 4 to run computer vision applications using OpenCV, Python, And Qt. We’ve previously written about Raspberry Pi Compute Module-based smart cameras such as the Q-Wave Systems EagleEye camera (CM3+) working with OpenCV and LabVIEW NI Vision and the StereoPi v2 (CM4) with stereo vision. But the EDATEC ED-AIC2000 is the first ready-to-deploy Raspberry Pi CM4 AI camera we’ve covered so far. EDATEC “CM4 AI camera” (ED-AIC2020) specifications: SoM – Raspberry Pi Compute Module 4  up to 8GB RAM, up to 32GB eMMC flash Camera 2.0MP global shutter or 5.0MP rolling shutter Acquisition rate – Up to 70 FPS Aiming point – Red cross laser Built-in LED illumination (optional) Scanning field Electronic liquid lens Fixed focal length lens Networking Gigabit Ethernet M12 port Communication protocols – Ethernet/IP, PROFINET, Modbus […]

EmbeddedTS embedded systems design