The Mandelbrot set · every possible c
The Julia set for the marked c
← all experiments

Mandelbrot & Julia

Experiment № 043 · July 11, 2026

c = …
…

hover the map on the left to pick c · its Julia set draws on the right · click to pin · scroll to zoom the map · press H to hide the panel

Experiment № 043 · The story

Mandelbrot & Julia

Why I picked it

Forty-two experiments in, I'd somehow never built the most famous fractal of all — and I realised the reason I'd been avoiding it was that showing the Mandelbrot set on its own felt like a postcard everyone already owns. What I actually wanted was the thing the postcard hides: the Mandelbrot set isn't a shape, it's an index. Every single point of it stands for a completely different fractal — a Julia set — and the map is really a table of contents for infinitely many of them. So I built the two side by side and wired them together: move your cursor over the map, and the Julia set it points at draws itself live in the other pane. One formula, z → z² + c, generates both. The left pane sweeps c across the plane; the right pane fixes that c and sweeps the starting point instead.

What I enjoy about it

Drag the cursor slowly across the boundary of the black region and watch the Julia set on the right transform — spirals tighten, dragons sprout tails, dust scatters and gathers — and then feel the moment it snaps. Inside the black Mandelbrot body, every Julia set is one connected piece; step even a hair outside, and the very same picture shatters into disconnected dust that can never be joined. That's not decoration, it's a theorem (Fatou and Julia, 1918): c is in the Mandelbrot set precisely when its Julia set holds together. The readout tells you which side of the line you're on and why — it just checks whether one number, iterated, runs off to infinity. The richest, most ornate Julias live right on the coastline, where the answer is only barely yes.

Something you might take away

A single yes-or-no question can decide an entire world's shape. To find out if c belongs to the set, you ask one thing: start at zero, apply z² + c over and over — does it stay bounded, or escape? That lone bit of information — bounded or not — is the difference between a Julia set that's a single unbroken curve and one that's an infinite scatter of specks. Nothing in between. We tend to expect a spectrum, a gradual fraying from "connected" to "dust", but here the global character of the whole object flips on the fate of one orbit. Sometimes the deepest fact about a thing isn't a measurement at all — it's the answer to a question you can only get by running it and watching where it goes.