Curious TechieDev Toolbox
← All Guides/PRIVACY & DIGITAL TRACKING•8 min read

What is Browser Fingerprinting & How Does It Track You?

An investigative look at cookieless tracking: Canvas rasterization entropy, GPU profiling, AudioContext mathematics, and anti-fingerprinting defenses.

ⓘKey Takeaways
  • Browser fingerprinting is a stateless device identification technique that collects hardware, graphics, audio, and configuration entropy from web browsers.
  • Unlike cookies and local storage, browser fingerprints cannot be cleared simply by opening an Incognito / Private window.
  • Key entropy vectors include Canvas rendering variance, WebGL GPU unmasked renderers, AudioContext oscillator frequency decay, installed system fonts, and screen color geometry.
  • Combining 8-12 subtle system traits can yield over 30-40 bits of entropy, uniquely identifying 1 device out of several million internet users.
  • Defenses against fingerprinting include Tor Browser uniformity, Firefox anti-fingerprinting protections (FPP), and randomized Canvas noise injection.

1. What is Browser Fingerprinting?

Browser Fingerprinting is a sophisticated identification method that gathers technical information about a user's browser configuration, graphics card, operating system, and hardware.

While any single trait (such as having a 1920x1080 screen) is shared by millions of people, combining dozens of microscopic system variables creates a statistically unique digital fingerprint that can track a user across different websites without storing any cookies on their machine.

2. Major Fingerprinting Vectors

Canvas RasterizationSlight variations in font anti-aliasing and subpixel smoothing between GPU drivers.
WebGL GPU ProfilingExtracts unmasked GPU vendor, renderer model, and floating-point shader precision limits.
AudioContext ProcessingMeasures acoustic wave decay rates processed through system DACs and audio buffers.
Hardware ConcurrencyReveals number of logical CPU cores and approximate system memory (RAM).

3. Canvas & WebGL Rendering Variance

When an invisible HTML5 <canvas> draws complex 2D text with colored alpha gradients, the exact RGB pixel output differs across machines by tiny floating-point rounding margins caused by differences in OS graphics subsystems (DirectX, Metal, OpenGL) and physical GPU silicon.

Hashing the rendered canvas base64 image data URL creates a reliable identifier across sessions.

4. Why Incognito Mode Cannot Prevent Fingerprinting

Incognito / Private Browsing mode isolates cookies, cache, and history. However, your physical GPU, CPU architecture, screen resolution, and font rasterization engines remain identical in private windows.

5. Modern Privacy Defenses & Anti-Fingerprinting

Modern browsers fight fingerprinting using two strategies:

  • Uniformity (e.g. Tor Browser): Normalizes all users to look identical (standard window size, generic fonts, spoofed timezone).
  • Noise Injection (e.g. Brave, Safari): Injects imperceptible pseudo-random noise into Canvas and WebGL calls so every script reading produces a different ephemeral hash.