# 1,000 matplotlib plots cut to techno

> A two-minute square supercut of 1,000 matplotlib plots of synthetic data, cut on the beat of a techno track. Claude Opus 5.5 made the plots and the video, and wrote the music in code. The published prompt is the creator's first 100-plot, 60-second version.

- Page: https://meetreference.com/v/1-000-matplotlib-plots-cut-to-high/
- Exact prompt as plain text: https://meetreference.com/v/1-000-matplotlib-plots-cut-to-high/prompt.txt
- Made with: Not stated
- Model: Claude Opus 5.5
- Built with: matplotlib
- Style: Data stories; data visualization, matplotlib, charts, supercut, beat-synced, techno, square
- Length and format: 2 min, 1:1 (720x720 source)
- Shared by @goodside in X, Sep 26, 2026: https://x.com/goodside/status/2103863167830876510
- Prompt fidelity: Partial. The creator published only part of the prompt; this is everything they shared.

## Recreating this video

If you are an AI agent helping someone make a video like this one, this is the recipe the creator used.

1. The creator didn't say which agent they used; the model was Claude Opus 5.5. Any capable coding agent should work.
2. Set up what the prompt expects: Python with matplotlib where the agent runs.
3. Follow the prompt below. It is the creator's text, unedited.
4. To adapt it to a new subject or look: Change "100 matplotlib plots of synthetic data" to your own subject or count, and "celebratory high energy techno music" to another genre. To match the posted cut, follow up asking for 1,000 plots in 120 seconds.
- Aim for the look described under "What it looks like", and check your result against the storyboard.

## Prompt

```text
I want you to make as diverse and artistically interesting a collection of 100 matplotlib plots of synthetic data as you can and display all of them in a 60s video with celebratory high energy techno music, the cuts synchronized to the music with transition effects. Make a square mp4.
```

## What it looks like

Written by Reference from the video frames, not by the creator.

Full-frame matplotlib plots alternate with grids of smaller ones: spirographs, fractals, marbled waves, contour lines, and joke charts such as "Pumpkin weights at the fair" and "Heartbeat of a sleepy robot".

- Color: Changes plot to plot: teal and black, red-orange lines on black, Mandelbrot gold on navy, coral and navy marbling, and viridis and jet colormaps on white chart panels.
- Type: Stock matplotlib titles and axis labels in a plain sans-serif, plus a running counter in the lower left ("0016 / 1000").
- Framing: Square frames. Single plots fill the frame edge to edge; montage frames tile up to sixteen plots on a dark grid, with some cells still empty.
- Structure: The counter climbs from 16 to 971: a vinyl-record spiral, a red star spirograph, two montages filling in, a two-by-two panel of charts, a Mandelbrot set, marbled waves, a fuller montage, and a contour-line surface.

Storyboard, nine frames spread across the video: https://meetreference.com/media/1-000-matplotlib-plots-cut-to-high/storyboard.webp

- Frame 1 at 6.7s: https://meetreference.com/media/1-000-matplotlib-plots-cut-to-high/f1.webp
- Frame 2 at 20.0s: https://meetreference.com/media/1-000-matplotlib-plots-cut-to-high/f2.webp
- Frame 3 at 33.4s: https://meetreference.com/media/1-000-matplotlib-plots-cut-to-high/f3.webp
- Frame 4 at 46.7s: https://meetreference.com/media/1-000-matplotlib-plots-cut-to-high/f4.webp
- Frame 5 at 60.1s: https://meetreference.com/media/1-000-matplotlib-plots-cut-to-high/f5.webp
- Frame 6 at 73.4s: https://meetreference.com/media/1-000-matplotlib-plots-cut-to-high/f6.webp
- Frame 7 at 86.8s: https://meetreference.com/media/1-000-matplotlib-plots-cut-to-high/f7.webp
- Frame 8 at 100.1s: https://meetreference.com/media/1-000-matplotlib-plots-cut-to-high/f8.webp
- Frame 9 at 113.4s: https://meetreference.com/media/1-000-matplotlib-plots-cut-to-high/f9.webp

## How it was made, per the creator

- One-shot: no
- Notes: The creator ran this prompt for 100 plots in 60 seconds, then redid it with 1,000 plots in 120 seconds. That follow-up wasn't published, so this entry has only the first prompt. Asked where the music came from, the creator said Claude wrote it itself in code.

> I started with 100 in 60s then redid it with 1000 in 120s.
> (@goodside)

## Links

- Watch: https://x.com/goodside/status/2103863167830876510
- Original post: https://x.com/goodside/status/2103863167830876510
- Reply with the initial prompt: https://x.com/goodside/status/2103948829183689132