LumonMDR
Lumon's Macrodata Refinement application inspired by Apple TV's 'Severance'.
Install / Use
/learn @andrewchilicki/LumonMDRREADME
Introduction
The goal of this project was to create a working miniature desktop display inspired by the 'Macrodata Refinement' computers featured in Apple TV's Severance. This build consists of two key aspects: the software and the hardware design, both outlined below.
Feel free to email comments, suggestions, or inquiries directly to andrewchilx@gmail.com.
<p align="center"> <img src="https://github.com/user-attachments/assets/aaed01c5-b200-4d43-9d2c-5a4399ba4960" width="98%"> </p> <p align="center"> <img src="https://github.com/user-attachments/assets/dc987a56-2a41-4780-ac42-345e78c8344e" width="49%"> <img src="https://github.com/user-attachments/assets/f46a447d-42ce-47af-b2ae-17409240cb9f" width="49%"> </p> <p align="center"> <img src="https://github.com/user-attachments/assets/15eabc19-85e8-4757-908c-e7565946b3b8" width="49%"> <img src="https://github.com/user-attachments/assets/54204491-7dae-41e7-aed4-271cc2c316bf" width="49%"> </p>I. The Software
Objectives
- Generate a randomized number grid with grouped 'bad numbers'.
- Animate and style the interface to match the aesthetics of the show.
a. The Number Grid
GENERATE
- A N×N grid is created.
- Each cell is filled with a random integer between 0-9.
MAP
- A Perlin noise map assigns smooth gradient values to the grid.
- This ensures numbers are naturally grouped rather than randomly scattered.
THRESHOLD
- Each cell's value is compared to a set threshold.
- Values above the threshold are marked as 'bad'/'scary' numbers.
GROUP
- 'Bad' numbers are grouped based on proximity to one another.
- At random intervals, a group visible in the viewport will be activated:
- If hovered over, they become 'super active', extending their active time and appearing agitated.
- If clicked, the group is 'refined', animating into a pre-determined bin and resetting as no longer 'bad'.
b. The Interface
<p align="center"> <img src="https://github.com/user-attachments/assets/14310434-81b7-4f62-8833-948708edee78" width="30%"> <img src="https://github.com/user-attachments/assets/6e2145ac-08f5-44f9-bdad-ebf272cbc0ce" width="30%"> <img src="https://github.com/user-attachments/assets/5d105682-2eb9-4c84-9806-cfafa7e4dd91" width="30%"> </p>- A moving Perlin noise map offsets each number (vertically or horizontally).
- Activated bad groups expand / contract / jitter.
- Numbers scale-up based on cursor position.
- Refined groups animate into bins, with a percent bar keeping track of your progress.
Additional Features:
- The application is cross-compatible, allowing compilation for both Linux and Raspberry Pi (ARM).
- A 'debug mode' reveals various settings (which can be saved / loaded from disk as JSON).
- An 'idle mode' can be enabled to display the Lumon logo screensaver.
- The user can navigate the full grid using arrow and zoom keys.
II. Hardware Design
Objectives
- Reconstruct the computer housing from the show as a printable 3D model.
- Design the interior to support a RaspberryPi and LCD screen
Final Design
<p align="center"> <img src="https://github.com/user-attachments/assets/874d9d46-c7f7-44cd-a9ec-21d55afb4973" width="30%"> <img src="https://github.com/user-attachments/assets/67b06a72-48b5-4889-83c1-97fb0fe5fab0" width="30%"> <img src="https://github.com/user-attachments/assets/2189913c-6f3b-453f-a94a-37ee4acf72d1" width="30%"> </p> <p align="center"> <img src="https://github.com/user-attachments/assets/2b0c8d53-02ca-4402-87ce-a6af0ef87a9b" width="60%"> <img src="https://github.com/user-attachments/assets/3c73aa3b-753d-4f1e-9f5e-b81ce0db6a6c" width="30%"> </p>- The computer’s outer shell was modeled in Blender from the ground up, closely following reference images from the show to capture the original look.
- The design was then optimized for 3D printing.
- The structure was made modular, allowing for easy assembly and disassembly.
- Each part includes a flat printing surface, enabling clean prints without the need for supports.
- Tolerances were incorporated to ensure a snug fit for all components.
- Finally, the interior was equipped with an adjustable Raspberry Pi mount.
- The mounting bracket can move laterally, and the RPi forward, allowing for fine-tuned alignment of the LCD screen with the front panel.
Final Product
<p align="center"> <img src="https://github.com/user-attachments/assets/b8fa5a15-ae32-4bd2-a9c3-7a4a64b0ad40" width="90%"> </p>Build Photos
<p align="center"> <img src="https://github.com/user-attachments/assets/661fbc69-a9db-4aee-9691-d9dad5d3d9fc" width="30%"> <img src="https://github.com/user-attachments/assets/be88fc36-a475-4b3f-a740-7af5ddd39ef7" width="30%"> <img src="https://github.com/user-attachments/assets/8f4e8245-6b6e-4912-96fd-88ff38a1b87e" width="30%"> </p> <p align="center"> <img src="https://github.com/user-attachments/assets/c5e0b223-2800-4896-ac11-5bf557b85046" width="30%"> <img src="https://github.com/user-attachments/assets/190788f9-335e-4f94-a391-0c0d07545d1f" width="30%"> <img src="https://github.com/user-attachments/assets/9e747680-1f53-4fea-b93f-71308c976430" width="30%"> </p>Related Skills
node-connect
350.1kDiagnose OpenClaw node connection and pairing failures for Android, iOS, and macOS companion apps
frontend-design
109.9kCreate distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
openai-whisper-api
350.1kTranscribe audio via OpenAI Audio Transcriptions API (Whisper).
qqbot-media
350.1kQQBot 富媒体收发能力。使用 <qqmedia> 标签,系统根据文件扩展名自动识别类型(图片/语音/视频/文件)。
