Board selection guide
Detailed comparisons of mainstream open-source boards, based on the latest CSDN community resources
Technical specification comparison
| Board model | Core processor | CPU clock speed | Memory | Storage | Network capabilities | Main interfaces | Price Range |
|---|---|---|---|---|---|---|---|
| ESP32-S3 | Dual-core Xtensa LX7 | 240MHz | 512KB SRAM + 8MB PSRAM | 16MB Flash | Wi-Fi + Bluetooth 5.0 | UART×2, I2C×2, SPI×4, GPIO×44, camera interface | ¥15-30 |
| STM32H743 | Cortex-M7 | 400MHz | 1MB SRAM | 2MB Flash | Ethernet | CAN FD, USB OTG, Ethernet, LCD controller, GPIO | ¥100-300 |
| Raspberry Pi 5 | Quad-core Cortex-A76 | 2.4GHz | 4GB/8GB LPDDR4X | microSD + PCIe NVMe | Dual-band Wi-Fi 5, Bluetooth 5.0 | HDMI×2, USB 3.0×2, Gigabit Ethernet, PCIe 2.0 x1, GPIO×40 | ¥300-600 |
| Arduino Uno | ATmega328P (8-bit AVR) | 16MHz | 2KB SRAM | 32KB Flash | No wireless (requires add-on) | UART×1, I2C×1, SPI×1, GPIO×14, analog inputs×6 | ¥15-25 |
| Jetson Xavier NX | 6-core Carmel + 384-core Volta GPU | CPU: 1.9GHz | 8GB/16GB LPDDR4x | 16GB eMMC + microSD | Wi-Fi, Bluetooth | HDMI, DisplayPort, Gigabit Ethernet, USB 3.0, MIPI CSI×6, PCIe | ¥1500-3000 |
Board details
ESP32-S3
The IoT workhorse
Built-in Wi-Fi and Bluetooth, ultra-low power, great value, and MicroPython support for fast development
Typical applications:
- Smart home control
- IoT data collection
- AI vision recognition
- Low-power sensor networks
STM32H743
Industrial-grade platform
Rich peripherals, strong real-time performance, industrial-grade reliability, -40°C to +85°C operating range
Typical applications:
- Motor Control
- Industrial automation
- Drone flight control systems
- High-end instrumentation
Raspberry Pi 5
The single-board computer flagship
The strongest ecosystem, powerful performance, full Linux support, and extensive expansion
Typical applications:
- Edge Computing
- Media server
- Robot "brain"
- Lightweight AI inference
Arduino Uno
The maker's first choice
Low learning curve, rich library ecosystem, active community, and pin compatibility with many shields
Typical applications:
- Maker beginners / education
- Simple sensor data collection
- Student electronics competitions
- Rapid prototyping
Jetson Xavier NX
Edge AI computing platform
Powerful GPU compute (21 TOPS INT8), CUDA/TensorRT support, and professional AI acceleration
Typical applications:
- Real-time video analytics
- Autonomous driving prototypes
- Drone visual navigation
- Medical imaging AI diagnosis
Technical feature comparison
| Dimension | ESP32-S3 | STM32H743 | Raspberry Pi 5 | Arduino Uno | Jetson Xavier NX |
|---|---|---|---|---|---|
| Real-time performance | Average | Excellent | Poor | Good | Medium |
| AI compute | Weak | No | CPU/GPU Inference | No | Extremely strong (21 TOPS) |
| Development difficulty | Low | High | Low | Lowest | High |
| Ecosystem maturity | High | High | Extremely high | Extremely high | Medium |
| Power consumption | Very low (<50mA) | Low | Medium (~5W) | Very low | Medium (10-15W) |
| Expandability | Limited | Good | Excellent | Medium | Excellent |
| Cost effectiveness | Extremely high | Medium | Medium | High | Low |
Selection advice
Choose by project requirement
| Project type | Recommended board | Key considerations |
|---|---|---|
| IoT sensor nodes | ESP32-S3 | Wireless connectivity, low power, low cost |
| Industrial control | STM32H743 | Real-time performance, reliability, rich peripherals |
| Smart home hub | Raspberry Pi 5 | Full OS, multimedia, ecosystem support |
| Maker beginners / education | Arduino Uno | Low learning curve, plenty of examples |
| AI vision / robotics | Jetson Xavier NX | GPU compute, AI framework support |
| Edge computing gateway | Raspberry Pi 5 / Jetson Nano | Networking, multitasking |
Choose by budget
Technology trends
The rise of RISC-V
Boards such as ESP32-C3 and HiFive1 are maturing, and their open-source nature is attracting many developers
AI moves to the edge
The Jetson family is driving edge AI adoption, and the Raspberry Pi 5 now handles lightweight AI inference
Integrated wireless
Mainstream boards now include Wi-Fi/Bluetooth, lowering the barrier to IoT development
Unified development frameworks
The Arduino framework spans platforms, and MicroPython speeds up prototyping
Related reading
Start exploring
Use the selection guide to find the board that fits your needs