用旧手机替代树莓派，作者质疑高价Pi的必要性

I’m working on a project that needs a high resolution camera, GPS, storage, and a microcontroller to control some lights. (More on the project soon!) I’ve been designing PCB adapters around a Raspberry Pi, an Arducam 64MP camera, and a GPS module. Every time I add another part, the bill grows.

I’m also trying an ESP32-P4 board, which will get its own post soon. But the alternative I keep coming back to is an old smartphone.

I have a Redmi 13C 5G and a Huawei Mate 30 5G. Phones like these turn up second hand on Taobao for roughly ¥300–500 (£33–56 / €40–67 / $45–75). The Redmi gives me a camera that saves 50MP JPEGs; the Mate 30 has a 40MP main camera and a telephoto lens. Both already contain GPS, storage, a clock, a battery, mobile data, and a touchscreen. Their USB-C ports give me a route to the PCB that will handle the lights and other sensors.

Here is the rough parts comparison I made while deciding what to build next:

These are illustrative November 2026 listings. The more fully equipped Pi total uses the cellular HAT’s GPS instead of buying the separate GPS HAT; its modem is 4G, while the phones are 5G models. Neither side includes my custom light controller PCB, a SIM or data plan, or an enclosure. The conversions use 10 November 2026 mid market rates of approximately ¥1 = £0.111531 / €0.133643 / $0.14991, rounded to whole units in the table.

The phone is already becoming more than a price exercise. I’ve unlocked and rooted the Redmi, tested its camera, and got it sleeping and waking for scheduled work. I can plug it into my computer and have Codex configure and test the pieces I need. The Huawei is harder to root, but I can still run a kiosk app on it. I’ve made a stand to photograph the same target with the Redmi and Arducam side by side; I want to see what detail each camera actually captures. A 50MP JPEG file does not, by itself, settle that question.

The open question is supply. Finding two good used phones is easy; finding a few hundred matching ones, with healthy batteries and predictable camera behaviour, would take more work. I feel increasingly confident that it’s possible.

This could be useful well beyond the current project. For digital artworks and installations, we’ve often used a Pi or a Mac mini. Imagine instead posting a client a configured phone: plug it into the display and it runs. With mobile data and remote access, I can fix something without asking a gallery for Wi-Fi privileges or an open port.

There’s something fun about taking a device made in enormous numbers, giving it a second life, and finding that most of the hardware I was about to assemble is already inside it.