search

xurxodiz / MarioLevels

My entry for the Mario AI Championship 2012, implementing adaptive procedural level generation. Holds 1st place at the all-time points and percentage tally.
MIT License
2 stars 1 forks source link

Implement the core automaton #17

Closed xurxodiz closed 12 years ago

xurxodiz commented 12 years ago

Issue #16 defined an automaton. The new idea shaping the system is to generate the chunks by traversing it, each state producing a new part of the stage lineally.

We should then create classes for both the Automaton, the State interface, the several individual States defined in #16 and methods for traversing them that call the appropriate level building functions.

Build upon the methods already created in #11 and #12, reusing the salvageable code.

xurxodiz commented 12 years ago

Work on it has progressed, we have the automaton, several states, and the system produces levels flawlessly (albeit simple, and the probabilities have to be worked a bit). We now take a break to reflect on #16 before implementing new states.

xurxodiz commented 12 years ago

Design (#16) and implementation (current issue) seem to go hand in hand, although design obviously goes one tiny step ahead.

We think of a state flow, implement it, test it, go back and refine it, etc.

The architecture part has been sort of set in stone, to be honest, and the only thing remaining is creating some new states and connecting them with transitions (a bit overlapped with #26).

xurxodiz commented 12 years ago

Closed, work continues in #48.