Do you think RV1126 may be a good choice for the reCamera V2.0?

[jennazhuxj]

To evaluate the feasibility of the RV1126 development board as a core processing unit for the second generation of reCamera, our team did a series of tests, mainly focusing on:

• Real-time video analysis capability (YOLOv5/SSD model performance)
• Long-term operational stability (temperature/power consumption)
• Development Environment Maturity (Toolchain Integrity)

Our preliminary conclusion is that the RV1126 may not be the ideal choice for our second-generation reCamera. Testing results show that its performance in multi-channel or high-precision scenarios, thermal design, real-time detection capabilities, and compatibility between OpenCV-Python and quantized models has not met our expectations.

The test details are recorded here on our Hackaday logs: https://hackaday.io/project/202943-peek-under-the-hood-how-to-build-an-ai-camera

[IoTEgr]

I think RV1126B is more suitable. As an upgraded version of RV1126, Ruixin Micro RV1126B has achieved comprehensive breakthroughs in five core areas: CPU architecture (four-core A53 replaces A7, performance improved by 2 times), NPU computing power (3 TOPS increased by 50%), visual processing (independent AI-ISP realizes day and night dual-mode adaptive + 0.01Lux low-light imaging), interface expansion (4-way MIPI-CSI+USB 3.0 + thermal imaging interface) and energy efficiency control (AOV 3.0 audio wake-up, standby power consumption as low as 1mW). At the same time, dynamic bit rate optimization (saving 50% storage bandwidth) and national secret-level security encryption (SM2/SM3/SM4+TrustZone) have been added, which significantly improves the efficiency and reliability of AI visual tasks in complex environments, but I think we should work harder on the software.