- YouTuber bitluni has unveiled a DIY RISC-V graphics processing unit
- Builder has a library of crazy DIY projects on his YouTube channel
- RISC-V microcontrollers were obtained from AliExpress
YouTuber bitluni has unveiled his latest project, which could put any concerns you have about GPU pricing into context. The German DIYer, who has built a huge following with his selection of intriguing and quirky projects, has assembled a GPU using a large collection of affordable RISC-V microcontrollers (MCUs).
While the results are impressive for a DIY build, unfortunately the GPU is quite modest and would likely have been abandoned much sooner had a PCB design company not been in touch to explore a build partnership.
Demonstrating the project on his channel, bitluni (whose real name is Matthias Balwierz) expressed how difficult the build was and how it almost drove him crazy.
Output of more than 8000 RISC-V MCUs with 320×200 resolution
In his latest video, bitluni recalls a previous attempt to build a GPU in late 2025 that led to PCB designers Altium getting involved, explaining: “The clusters I made before were already challenging my sanity. I thought I was done with it, but the budget and these tools would allow for a cluster of a different magnitude, and the magnitude I had in mind was just crazy.”
Taking things to the next level is a hallmark of bitluni’s videos, and the final build has a whopping 8,192 CH570 RISC-V microcontrollers, each running at 100 MHz, with 12 KB of SRAM. These are mounted on blades, which went through several revisions as the PCB company declared them “too complicated.”
Each CH570 has a series of LEDs mounted on it, and these correspond to each MCU and the equivalent QVGA pixel. With 8,192 MCUs, the resulting 320×200 may seem modest, but bitluni already has plans for a more powerful version with 32,000 MCUs.
Graphs or hashes?
The rise in GPU prices before the rise of AI was due to industrial-scale cryptocurrency hashing, for which graphics cards are particularly well-suited. It is this relationship between hash and graphics that gives the bitluni project some hope beyond displaying images and videos.
Identifying the serial port as the bottleneck to the project’s success as a functional (albeit unwieldy) GPU, bitluni asked its community of viewers for help in finding an alternative solution, but that didn’t close the project.
Changing gears, he used microcontrollers for hashing and found that they outperformed his PC’s 8-core CPU, with a low power consumption of 4 watts, so perhaps this approach has some use after all.
Follow TechRadar on Google News and add us as a preferred source to receive news, reviews and opinions from our experts in your feeds.
