21 lines
2.4 KiB
Plaintext
21 lines
2.4 KiB
Plaintext
---
|
|
description: General rules and concepts for the z80 cellular automata simulation project, focusing on the environmental region grid system. This rule introduces the key concepts and overall purpose.
|
|
globs: /**/*_z80_cellular_automata*.*
|
|
---
|
|
- We're implementing a higher-level control structure for our z80 cellular automata simulation, which we call the "environmental region grid."
|
|
- Key Concepts:
|
|
- Soup Cells: The individual units of our cellular automata, which follow basic rules and interact with their neighbors.
|
|
- Regions: Larger areas that encompass multiple soup cells. Each region can have unique properties that influence the behavior of the soup cells within it.
|
|
- Environmental Region Grid: A grid overlaid on top of the soup cell grid, dividing the simulation space into discrete regions. This grid can be 4x4, 8x8, or 16x16, allowing for different levels of granularity.
|
|
- Region Parameters: Each region has a set of adjustable parameters that affect the soup cells within it.
|
|
- Obstacle (A region that blocks the movement of soup cells)
|
|
- Directional influence (biasing cell interactions in specific directions)
|
|
- Randomness factor (introducing more or less chaos in cell behavior)
|
|
- Temperature (affecting overall activity levels)
|
|
- Energy levels (influencing the likelihood of certain cell states or interactions)
|
|
- Other custom parameters as needed
|
|
- Dynamic Influence: The region parameters dynamically modify the behavior of soup cells, creating areas of distinct characteristics within the larger simulation.
|
|
- User Interaction: Users can interact with the simulation by adjusting region parameters in real-time, allowing for on-the-fly modification of the simulation's behavior.
|
|
- Visualization: The region grid and its effects are visually represented, allowing users to see the influence of their changes on the simulation.
|
|
- Purpose: This system adds a new layer of complexity and control to the cellular automata simulation. It allows for the creation of diverse environments within a single simulation, enabling users to explore how different regional properties affect the emergent behavior of the cellular automata.
|
|
- By implementing this region grid system, we're providing a powerful tool for users to experiment with large-scale influences on cellular automata behavior, potentially leading to new insights and interesting emergent phenomena. |