Note:
- Currently, booting is only supported from SD Card or NVMe.
- UFS is available only as external storage, but support for UFS boot will be added soon.
- To enable booting from NVMe, the bootloader must be installed on SPI NOR using either
armbian-installorrkdeveloptool.
- Installation to NVME or USB is done via standard armbian install utility (
sudo armbian-install). Make sure to generate partitions and format them before start. - For UFS ready images use an image with the Extension UFS and flash it directly either via an USB-Adapter or via `rkdeveloptool`. See this Docs page for more information on how to flash Rockchip boards via Maskrom mode.
Flashing instructions:
-
- Hold MASK button and power the device up to enter MASKROM mode
- You will need mini loader binary for this device https://dl.armbian.com/nanopi-r76s/bootloader/MiniLoaderAll.bin
- Follow instruction for flashing with RKdevtools: https://docs.armbian.com/User-Guide_Getting-Started/#rockchip
How to install
Radxa-Nio-12L uses EFI Boot via Grub with DeviceTree. To make use of it follow this Guide from Collabora to flash Uboot onto the UFS storage. Then you can flash one of the provided images onto a SD-Card and power on the SBC which will automatically boot into Armbian. Keep in mind the Boot order has UFS at first place so you need to erase any OS you have on on UFS to boot from a SD Card.
Q&A for Genio-Family:
- Can I boot from UFS?
- Yes just make sure to use the Armbian image with the UFS extension as these are partioned with 4096 sector size instead of the 512 used commonly for SD-Card / NVMe / etc.
- Which kernels are available?
- We’re making use of Collabora’s mediatek-next branch which is based on mainline Linux.
- XYZ is not working.
- Check this support matrix to see if that feature has been implemented already.
BeagleY®-AI is a low-cost, open-source, community-supported development platform for developers and hobbyists in a familiar form-factor compatible with accessories available for other popular single board computers. Users benefit from BeagleBoard.org-provided Debian Linux software images with a built-in development environment and the ability to run artificial intelligence applications on a dedicated 4 TOPS co-processor along with real-time I/O tasks on a dedicated 800MHz microcontroller. BeagleY®-AI is designed to meet the needs of professional developers and classroom-environments alike being affordable and easy-to-use, while being open source hardware such that developers barriers are eliminated to how deep the lessons can go or how far you can take the design in practical applications.
Checkout https://pinout.beagley.ai for Pinmuxing and accessory compatibility info.
The SK-AM69 Starter Kit/Evaluation Module (EVM) is based on the AM69x AI vision processor which includes an image signal processor (ISP) supporting up to 1440MP/s, 32 tera-operations-per-second (TOPS) AI accelerator, eight 64-bit Arm®-Cortex® A72 microprocessor, and H.264/H.265 video encode/decode. The SK-AM69x is an ideal choice for machine vision, traffic monitoring, retail automation, and factory automation.
The starter kit also includes two RPI 22-Pin CSI connector and two MIPI CSI-2 40-Pin expansion connectors for up to 12 cameras, making it suitable for a variety of embedded vision and AI applications on the edge. Our ISP can handle different lighting conditions in video streams by using HDR encoding for camera support. The TI deep-learning accelerator enables classification and object detection at low latency and high frame rate.
Features
- Performance: AM69 processor enables 32 TOPS deep-learning performance and hardware-accelerated edge AI at low power
- Camera interfaces: three CSI-2 ports compatible with Raspberry Pi or a high-speed 40-pin Samtec camera connector allowing up to twelve cameras (requires multi camera hub)
- Connectivity: three SuperSpeed USB Type A ports, one SuperSpeed USB Type C port, one ethernet port, one M.2 key E connector and one M.2 key M connector, four CAN-FD interfaces, four UART terminals over one USB bridge
- Memoy: four DRAM, LPDDR4‐4266, 32 gigabytes total memory, support for inline ECC
- Display: DisplayPort with up to 4K resolution with MST support and 1080 HDMI
The most powerful AI open-source platform available
BeagleBone® AI-64, is built on a proven open source Linux approach, bringing a massive amount of computing power to the hands of developers in an easy to use single board computer. Leveraging the Texas Instruments TDA4VM SOC with dual 64-bit Arm® Cortex®-A72, C7x DSP along with deep learning, vision and multimedia accelerators, developers have access to faster analytics, more data storage options, more high speed interfaces including all the connectors you’ll need on board to build applications such as Autonomous Robotics, Vision and Video Analytics, Hi-End Drones, Media Servers, and Smart Buildings.
BeagleBone® AI-64 brings a complete AI and Machine Learning System to developers with the convenience and expandability of the BeagleBone® platform and the peripherals on board to get started right away learning and rapidly prototyping applications.
Leveraging over a decade of success in open hardware single board Linux computers, BeagleBone® AI-64 puts a massive amount of computing power in an easy to use platform.