An interactive, side-by-side visualizer for three mouse movement algorithms — from a simple linear baseline to a full biomechanical model.
Moves directly toward the target at a constant speed. No curve, no noise, no human quality. Every other algorithm is compared against this.
Simulates wind and gravity forces that push the cursor toward the target while adding natural-looking wobble. The dominant aim-assist algorithm for over a decade — visually convincing, but statistically detectable.
Original algorithm by BenLand100 (Java)
JavaScript port by arevi
Reimplemented inline for browser compatibility in this visualizer
A biomechanically-grounded model built on Plamondon's Kinematic Theory with six interacting motor-control components:
- Sigma-lognormal velocity primitives
- Fitts' Law movement-time prediction
- Ornstein-Uhlenbeck drift
- Signal-dependent noise
- Physiological tremor
- Gamma-distributed sampling intervals
Algorithm, research & C++ source by ck0i
Paper: zenodo.org/records/18872499
Ported to JavaScript for this visualizer
- Side-by-side canvas — all three algorithms run simultaneously so differences are immediately visible
- Velocity graph — overlaid speed-over-time curves for each algorithm
- Stats bar — move time, sub-movements, and path efficiency after each run
- Speed control — one slider that scales movement time across all algorithms
- Auto loop — toggle continuous animation without user input
- Click to place target — click anywhere on a canvas to set a new destination and run instantly
| Algorithm | Author | Source |
|---|---|---|
| WindMouse | BenLand100, JS port by arevi | GitHub |
| SigmaDrift | ck0i | Zenodo paper |
| Linear | — | baseline, no external author |