2 patterns-animation-Overview
Balazs Horvath edited this page 2026-04-18 11:13:07 +02:00
This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

Animation Patterns Overview

Animation patterns demonstrate fundamental mathematical and algorithmic techniques for creating moving images. These patterns serve as both educational examples and practical templates for generative animations.

Pattern Philosophy

These patterns emphasize:

  • Mathematical foundations: Visualizing mathematical concepts
  • Algorithmic clarity: Clean, understandable implementations
  • Educational value: Learning through visual experimentation
  • Practical utility: Reusable techniques for real projects

Pattern List

Plasma Effect

Torch-based plasma effect with color cycling, demonstrating sine wave superposition.

Perlin Noise Fractal

Torch-based Perlin-like noise with varying Gaussian filter for smooth transitions.

Sine Wave Interference

Torch-based wave interference pattern demonstrating constructive and destructive interference.

Mandelbrot Zoom

Mathematical visualization of the Mandelbrot set with zoom animation.

Recursive Pattern

Mathematical recursion creating self-similar patterns through depth-controlled iteration.

flowchart TD
    A[Animation Patterns] --> B[Wave-based]
    A --> C[Noise-based]
    A --> D[Fractal-based]
    A --> E[Recursive]
    B --> B1[Plasma Effect]
    B --> B2[Sine Wave Interference]
    C --> C1[Perlin Noise Fractal]
    D --> D1[Mandelbrot Zoom]
    E --> E1[Recursive Pattern]
    B1 --> B11[Sine superposition]
    C1 --> C11[Gaussian filter]
    D1 --> D11[Complex iteration]
    E1 --> E11[Depth recursion]

Technical Concepts

Plasma Effect

Plasma effects demonstrate the superposition principle: combining multiple simple waves creates complex patterns.

Mathematical Foundation:

P(x, y, t) = sin(x + t) + sin(y + t) + sin((x + y)/2 + t)

Key Concepts:

  • Wave superposition
  • Phase shifts
  • Color mapping
  • Time-based animation

Perlin Noise Fractal

Perlin noise demonstrates multi-scale signal processing. By smoothing noise at different levels and combining them, you create organic-looking textures.

Key Concepts:

  • Gaussian filtering
  • Multi-scale analysis
  • Signal synthesis
  • Dynamic parameters

Sine Wave Interference

Wave interference demonstrates how waves interact constructively and destructively.

Key Concepts:

  • Constructive interference (peaks align)
  • Destructive interference (peak meets trough)
  • Moiré patterns
  • Phase relationships

Mandelbrot Zoom

The Mandelbrot set demonstrates complex number iteration and fractal geometry.

Formula:

zₙ₊₁ = zₙ² + c

Key Concepts:

  • Complex numbers
  • Iteration
  • Divergence detection
  • Color mapping

Recursive Pattern

Recursive patterns demonstrate mathematical recursion and self-similarity.

Key Concepts:

  • Base case
  • Recursive step
  • Depth control
  • Pattern combination

Animation Architecture

Dual-Mode Output

ScriptableImage nodes support both still images and animations through the batch dimension:

  • Still image: B=1 (single frame tensor [1,H,W,C])
  • Animation: B>1 (batch of frames [B,H,W,C])

This follows ComfyUI's native format and is compatible with SaveWEBM and SaveVideo nodes.

Frame Generation

frames = []
for t in range(num_frames):
    frame = generate_frame(t)
    frames.append(frame)
output_image = torch.stack(frames, dim=0)  # [B, H, W, C]

Performance Considerations

Resolution Trade-offs

  • 512×512: Good balance
  • 1024×1024: 4× slower, higher quality
  • 256×256: 4× faster, good for previews

Frame Count

  • 30 frames: Short animations
  • 60 frames: 1-second loops
  • 600 frames: 10-second loops

Optimization Techniques

  1. Precompute coordinate grids
  2. Use vectorized operations
  3. Limit recursion depth
  4. Cache intermediate results

Dependencies

Required:

  • torch: Tensor operations
  • numpy: Array manipulation
  • PIL: Image saving

Optional:

  • scipy: Gaussian filter
  • imageio: Video encoding
  • opencv-python: Alternative encoding

Learning Path

Beginner

  1. Plasma Effect - Learn sine waves and superposition
  2. Sine Wave Interference - Understand wave interaction
  3. Color Cycling - Practice color manipulation

Intermediate

  1. Perlin Noise Fractal - Learn signal processing
  2. Scanline Shift - Understand coordinate manipulation
  3. Gradient Patterns - Practice spatial functions

Advanced

  1. Mandelbrot Zoom - Complex number iteration
  2. Recursive Pattern - Mathematical recursion
  3. Fractal Brownian Motion - Multi-scale synthesis
  4. Looping Patterns - Mathematical loop design

References