1. Goals
2. Task/Idea List
3. Dev Notes
4. Changelog
5. Credits

• Meta Goals
  • Basic UI widgets that can be reused.
  • Separation of concerns.
  • Multi-processing: simulation separate from UI.
• Game Goals
  • 2 teams on the field.
    • 1 player
    • 1 AI
  • Each team has one base and N number of ants.
  • "Energy" is the main resource.
  • Every N leaves brought in gives you E energy.
  • Every 1 ant carcass gives you aE energy.
  • Energy can be used to create new ants.
  • Energy can be used to create DNA.
  • User has no direct control over the ants. User only directly controls the queen.
  • Queen ants, represented by the base, produce ants on a button click.
  • The ant produced is determined by the current DNA makeup which is also determined by the player.
  • Producing an Ant costs N base amount + kD amount based on additional DNA changes.
  • Win condition: other teams have no energy and no ants.
• Game Physics/Rules
  • Ants can't climb over each other.
    • Ants perform a short-sighted A* pathing because that's how ants behave in real life.
  • Simple terrain generation: passable and impassible.
  • Simple visual background procedurally generated.

Key

• In process

• Not Started or on hold.

• User Interface
  • Extract the minimap view from it's container view. To form the buffer, and simplify the minimap.
  • Allow preservation of aspect ratio and auto scaling via a mixin, something like: PreserveAspectRatio
    • MiniMap has the code that could be used as a basis for this.
  • Make a DebugPanel for showing:
    • fps
    • mouse coordinates
    • log statements in game.
  • Assets have the same default size and then scale according to zoom level (like 32x32 pixels, or something).
  • Views should have dirty rect capability (see pygame.display.update function for interface for passing in changed list of rects that need redrawing).
  • There should be a grid view-like capabilities. This is accomplished by allowing child views to be positioned relative to other child views or by allowing the containing parent to position the children.

• Entities
  • Add sight (square first, than radial if we want to) and visual fog.
  • Add collision detection so that entities can not run over each other.
• Ant-god controls:
  • Hostility level
  • Friendly index
  • Helpful level
  • Health minded ants or self sacrifice?
  • Expansion oriented or homebodies?
  • Conversion or kill?
• Evolutionary advancements
  • vision increase (see food and enemies better)
  • conversion capability
  • Different Ant types
  • The Swarm
• Multi-Process, Sim and Render Thread
• Saving settings...
• Game World
  • Procedural world generation.
• Spatial Index:
  • Keep the seeding of the expected index, but allow dynamic creation of cells outside of the world area if/when an entity travels outside the boundaries.
    • Note: The spatial index also needs to keep an extent of max boundaries so that if a large search value is placed in, only possible cells are searched through.

Dependencies are managed by pip and are stored in the requirements.txt file. To install dependencies on a new system, do:

pip install -r requirements.txt

To run the tests with nose, use the following command from the main directory.

nosetests -c nose.ini

It's difficult to map git to a traditional changelog. Logging entries should be stored with the commit:

# short entries
git commit -m "My log entry."

# long entries...
git commit
# ...and use vi or whatnot to enter the log entry.

A quick way to generate a changelog like format to standard out:

git log --oneline --decorate

anthill-1.jpg and anthill-2.jpg by Accretion Disc (found on a Creative Commons Search)

link of original image:

http://www.flickr.com/photos/befuddledsenses/6126386168/sizes/l/