Cavalry 2D Procedural Motion Graphics Guide

1. The Procedural Philosophy: Cavalry vs. After Effects

In traditional animation software, creating 100 animating geometric dots requires creating 100 separate layers and keyframing each one manually:

• Procedural Nodes: In Cavalry, a single Duplicator Node generates hundreds of shapes distributed along curves, grids, or Archimedean spirals. Altering one master shape instantly updates all duplicates in real time.
• Real-Time 60 FPS Viewport: Built on modern GPU vector rendering, Cavalry plays back complex generative motion designs smoothly at 60 fps without tedious pre-rendering cache bars.

2. Duplicators, Falloffs & Noise Fields

Complex generative motion design in Cavalry relies on connecting three core node types:

3. Data-Driven Animation with CSV & JSON Feeds

Automating social media graphics, sports scoreboards, or financial charts:

Connect a live CSV spreadsheet or JSON API endpoint to Cavalry's text and bar-chart components. Cavalry automatically generates distinct animated video variations for every row in your dataset without manual typing.

4. Interactive Web Export: Lottie, SVG & MP4

Cavalry bridges the gap between motion graphics and UI web development:

Export pixel-perfect vector animations directly to Lottie JSON and animated SVG for lightweight web apps, or render ProRes 4444 with alpha channels for video editors in Premiere and DaVinci Resolve.

Where This Fits

Cavalry works in 2D, and Blender for Motion Graphics covers the 3D side in a free toolchain, from render engine to edit.

What goes wrong

The failure modes worth recognising, and the symptom that identifies each one.

• Building procedurally when keyframes would do. A duplicator and a falloff are the right tools for thirty repeating elements and the wrong ones for three. Procedural setups take longer to build and are harder for a collaborator to change, and that cost only pays back at scale.
• Data that changes shape. A CSV feed drives the animation through its column names. Adding, renaming or reordering a column breaks every binding that referenced it, usually silently, leaving elements at default values. Fix the schema before building against it.
• Exporting Lottie with unsupported features. Lottie carries a subset of what the application can render. Effects outside that subset are dropped on export rather than flagged, so the file plays and simply looks wrong. Preview the exported Lottie rather than the composition.
• Noise fields that look random and are not. A noise falloff is deterministic: the same seed produces the same pattern every time, which is what makes a procedural setup repeatable. It also means every element sharing a seed moves identically, and a composition that should feel organic reads as mechanically synchronised. Offset the seed per element rather than reusing one across the scene.
• Render times that scale worse than expected. Duplicators multiply geometry, and a falloff evaluated per duplicate multiplies the evaluation with it. A setup that previews smoothly at thirty copies can take minutes per frame at three hundred, and the cost appears at render rather than while building. Test at the final count early.

Working Vocabulary

Key Cavalry and procedural 2D motion design terms introduced in this guide:

• Duplicator. A procedural node that clones input shapes across geometric distributions with automated index offsets.
• Falloff. A spatial influencer (spherical, linear, or noise) that modulates animation parameters based on subject proximity.
• Lottie. A JSON-based vector animation file format rendered natively on web and mobile devices.
• SceneForge. Cavalry rendering architecture designed for automated batch video generation from live data.
• Connectors. Visual wire links that feed output values from one node into the input parameters of another.

Sources & Further Reading

• Blender Motion Graphics: The Complete Guide
• Kinetic Typography in After Effects
• Lower Thirds Design & Broadcast Graphics
• Aspect Ratio & Resolution Calculator