RK3588

Performance Evaluation of Rockchip RK3588 to RK3562 NPU (Part 2)

by
Performance Evaluation of Rockchip RK3588 to RK3562 NPU (Part 2)

The following is a more detailed evaluation of the NPU performance of four Rockchip chips, covering aspects such as computing power details, hardware architecture, actual measurement data comparison, and development support: 1.RK3588: FlagshipNPU with ultimate performance Computing power and precision support Peak computing power:6 TOPS (INT8), supports mixed precision operations (INT4/FP16/BF16, etc.), and in low … Read more

openKylin Fully Compatible with OrangePi 5 Plus, Expanding the High-Performance Edge Computing Ecosystem

by
openKylin Fully Compatible with OrangePi 5 Plus, Expanding the High-Performance Edge Computing Ecosystem

Click the blue text above to follow us and discover more exciting content. Recently, the OpenAtom openKylin (referred to as “openKylin”) community has officially completed the full adaptation of openKylin version 2.0 with the OrangePi 5 Plus, successfully achieving deep compatibility between the openKylin operating system and the OrangePi 5 Plus development board. Currently, the … Read more

Beiqi Technology’s RK3588 Core Board Passes Rigorous High and Low Temperature Testing

by
Beiqi Technology's RK3588 Core Board Passes Rigorous High and Low Temperature Testing

The RK3588 development board underwent temperature and humidity testing at an environmental temperature of 65 degrees Celsius and 90% humidity, successfully passing the test. After completing the temperature and humidity testing, a low-temperature test was conducted, with an environmental temperature below -40 degrees Celsius, and the motherboard successfully passed the test. The RK3588 core board … Read more

Performance Evaluation of Rockchip RK3588 to RK3562 NPU (Part 1)

by
Performance Evaluation of Rockchip RK3588 to RK3562 NPU (Part 1)

This is a deep performance evaluation of the NPU of four chips from Rockchip: RK3588, RK3576, RK3568, and RK3562, integrating multiple practical data and technical analysis to compare their computing power, architecture, and application scenarios: 1.RK3588: Flagship-level AI computing power with full scene coverage NPU Performance: Integrating the third-generation self-developed NPU, with a computing power … Read more

Rockchip’s Profit Surges 341%, Leading the ‘Chinese Chip’ Revolution in Inference Computing

by
Rockchip's Profit Surges 341%, Leading the 'Chinese Chip' Revolution in Inference Computing

Author | Yan Xi As NVIDIA’s GB200 chip creates a billion-dollar frenzy in data centers, another revolution in computing power concerning trillion-dollar IoT terminals is quietly unfolding. Under the LCD dashboard of BYD’s smart cockpit, within the laser radar of Xiaomi’s robotic vacuum, and in the image processors of industrial quality inspection cameras, domestic ASIC … Read more

Setting Up the RK3588 NPU Development Environment

by
Setting Up the RK3588 NPU Development Environment

How to set up the RK3588 NPU development environment on an Ubuntu system (PC)? This involves running the Ubuntu system on the computer while the RK3588 board runs the Buildroot Linux system, establishing the RK3588 NPU development environment.The following are the corresponding deployment steps:0. Dependency Files RKNPU2 project download link: https://github.com/airockchip/rknn-toolkit2/tree/master/rknpu2 RKNN-Toolkit2 project download link: … Read more

Peite Technology RK3588 Development Board: Octa-Core Architecture + 6TOPS Computing Power for Accelerated Product Iteration

by
Peite Technology RK3588 Development Board: Octa-Core Architecture + 6TOPS Computing Power for Accelerated Product Iteration

Click the blueFollow us In today’s global wave of AIoT (Artificial Intelligence of Things) and edge computing, companies have higher demands for the performance, flexibility, and development efficiency of hardware platforms. How can we shorten the product cycle from prototype to mass production? How can we achieve complex functions at a lower cost? How can … Read more

From Bricking to Rebirth: My RK3588 Development Board Embarks on the Journey of Customizing Ubuntu

by
From Bricking to Rebirth: My RK3588 Development Board Embarks on the Journey of Customizing Ubuntu

When I first introduced cloud phones (Unveiling Cloud Phone Technology! Achieving Low-Cost Transformation of One Computer into Hundreds of “Virtual Phones” in Practice), I mentioned that the mobile box from Magic Cloud was a bit expensive. However, running it on a server lacked graphical acceleration, which slightly affected the experience, so I tried running it … Read more

Running Waydroid on RK3588 via Panthor

by
Running Waydroid on RK3588 via Panthor

Previously, we could run Waydroid using the official image through software rendering, as referenced in the Waydroid research, allowing Android applications to run on Linux. This time, by porting Panthor, we can utilize the official Mesa for rendering. Based on related articles from the external community, we will verify the performance of Waydroid running on … Read more

Deploying YOLOv8 on Wildfire RK3588

by
Deploying YOLOv8 on Wildfire RK3588

Description Deploying YOLOv8 on Wildfire RK3588, we use yolov8s.pt (downloaded from the YOLOv8 official website) as an example. 1. pt->onnx Do not use the official YOLOv8 code; instead, use the Rockchip YOLOv8 code, available at https://github.com/airockchip/ultralytics_yolov8 After downloading the code, execute the model conversion with the following code: from ultralytics import YOLO model = YOLO('yolov8s.pt') … Read more