How I animate 3Blue1Brown | A Manim demo with Ben Sparks

My name is Grant Sanderson. Videos here cover a variety of topics in math, or adjacent fields like physics and CS, all with an emphasis on visualizing the core ideas. The goal is to use animation to help elucidate and motivate otherwise tricky topics, and for difficult problems to be made simple with changes in perspective.

For more information, other projects, FAQs, and inquiries see the website: https://www.3blue1brown.com

3Blue1Brown

Introduction to Manim and the video's purpose

The most frequent question about 3Blue1Brown's animation software is answered.

The software, Manim, is a custom Python library, programmatically created and highly specific to 3Blue1Brown's needs.

The video aims to give a behind-the-scenes look at Manim, explaining its use and workflow.

The video features a conversation with Ben Sparks, known for his Numberphile videos, about Manim.

The existence of two Manim versions is highlighted: the original and the community version.

History and differences between Manim versions

The original Manim's development is intertwined with the channel's history, starting as a personal project during the creator's undergraduate studies.

The creator lacked the capacity to manage the open-source project effectively, leading to community involvement.

The Manim Community version, a fork of the original, focuses on issue resolution, pull requests, testing, and documentation.

The demo uses the creator's version, but the community version is recommended for its better documentation and testing.

Manim workflow demonstration: Hello World example

A "Hello World" example in Python is shown, using Sublime Text as the text editor.

Scenes are classes in Python, with the `construct` method containing rendering code.

A Python terminal interacts with the scene, allowing for real-time adjustments and testing.

Keyboard shortcuts streamline the workflow, enabling quick testing and immediate visual feedback.

Larger scenes are discussed, highlighting the benefit of keeping all code in one place despite its length.

A 'checkpoint paste' feature allows running specific sections of code, reverting to a previous scene state for testing.

The workflow is compared to Jupyter notebooks, highlighting the advantages of using a single text file for scene creation.

The Lorenz attractor, a shape from chaos theory, is introduced, explained as a result of 3D differential equations.

ChatGPT is used to generate the Python function for solving the Lorenz equations using SciPy's `solve_ivp` function.

Initial conditions are tested, and the resulting points are used to create a curve using Manim's functions.

The creation of the curve is animated, showing the system's evolution over time, with adjustments to smoothing and color.

Multiple curves with slightly varied initial conditions are created to illustrate the chaotic divergence of the system.

A coding bug related to global variables in the IPython environment is explained and a workaround is provided.

The process of rendering the scene into an MP4 file is demonstrated using command-line arguments.

The evolution of the workflow from command-line based to interactive shell-based is discussed.

Sublime Text scripts and terminal integration are explained, with a promise to document the process in the video description.

Adding equations to the scene using LaTeX objects and MathPix for LaTeX conversion is shown.

Example scenes are mentioned as a resource for beginners to explore Manim's functionalities.

How I animate 3Blue1Brown | A Manim demo with Ben Sparks

Scripts