Foundations

Complex numbers as a plane

A complex number combines a horizontal real component and a vertical imaginary component. Treating it as a point turns arithmetic into motion, stretching and rotation across a plane.

Color-rendered Julia set with winding boundary bands
Julia set in MandelKit: one iteration rule produces a boundary with structure down to the display limit.Image: Björn Kindler / MandelKit · MandelKit · Eigene Darstellung · Own work

Coordinates with their own multiplication

Write z = x + yi. Addition translates points. Multiplication combines magnitudes and adds angles; squaring doubles the angle and squares the distance from the origin. That geometric behavior explains the rotational folds produced by z² before a constant c is added.

Geometrically, complex multiplication combines scaling and rotation. That is why very short polynomials can produce twisting, highly symmetric orbits. The imaginary unit is not a decorative addition; it gives an algebraic setting in which polynomial equations and their iteration can be treated naturally. A renderer stores real and imaginary parts as two coordinates, and every multiplication couples the two axes according to a fixed rule.

Orbit, magnitude and escape

Starting from a point, a renderer repeatedly applies a function and watches the resulting orbit. The magnitude |z| is its distance from the origin. For the quadratic family, once the orbit crosses an escape radius of 2 it cannot return to a bounded path.

Magnitude supplies a simple escape test because an orbit beyond a sufficient radius cannot return for the usual quadratic maps. Iteration is still required until that proof is reached. Points near the boundary can appear bounded for a long time before escaping. Bailout radius and iteration limit are therefore different decisions: the radius provides a mathematical bound, while the limit ends a finite computation without proving membership for every point that remains.

Parameter plane and dynamical plane

In the Mandelbrot set, each pixel is the parameter c and the orbit starts at zero. In a Julia set, c is fixed and each pixel supplies the starting z. The equations are closely related, but the image asks a different question.

The two planes are complementary. Parameter space asks which constant produces a particular global behavior; the dynamical plane shows how starting points divide under that chosen constant. Selecting a parameter inside the Mandelbrot set yields a connected filled Julia set, while a parameter outside yields a disconnected one. Interactive tools become especially instructive when a selection in parameter space updates its Julia view alongside it.

Sources and further reading

This article summarizes the following specialist sources in original wording. Accessed and editorially reviewed 12 August 2026.

  1. Mandelbrot SetWolfram MathWorld
  2. Julia SetWolfram MathWorld