Information Technology Reference
In-Depth Information
0 signals the victory of the opposing team. To win, teams have to actively defend and
pursue flags.
In this paper, we present a novel method that can generate dynamic maps almost
as soon as the player press the 'Play' button, while ensuring the quality. The method
is simple, practically feasible and we have implemented and evaluated it with a test
game. In the rest of the paper, we first discuss the motivations for this work in Section 2
and then related works in Section 3. In Section 4 we describe the design goals of map
generation method. Section 5 explains the design of the map generator. Implementation
and evaluations are described in Section 6. Finally we conclude the paper with the
summary at Section 7.
2
Motivation
In this work, we are attempting to automatically create playable, balanced (fairness) and
interesting maps for multiplayer shooters, with a novel approach built using Search-
based PCG [17]. While PCG is used by some multiplayer shooter game designers, who
would procedurally generate maps and then manually tweak them to ship with the final
product, our goal is to remove the human intervention for manual tweaking completely.
This means that the generation should be completed within a span of seconds, or else
the patience of the player could wear thin. If we are successful, the development time
and cost needed to create maps for similar games could be drastically reduced. The
result would be increased longevity that stems from the near limitless amount of maps
for players to play in.
3
Related Work
G uttler et al.
[7] identified some basic spatial properties of multiplayer FPS games and
proposes several heuristics for better level design. In addition, the insights provided by
several industry leaders of leading game companies on design of a good multiplayer
game ([14], [9] and [8]) are incorporated in formulating our design goals. Search-based
PCG (SBPCG), an approach to PCG, was introduced by
Togelius et al.
[17]. We will be
employing a similar approach in our solution.
Togelius et al.
also managed to procedu-
rally generate tracks for a racing game [15] and maps for strategy game Starcraft [16].
In both cases, SBPCG was used with a simulation based fitness function.
Kerssemakers
et al.
[11] introduced a procedural PCG generator to generalise the creation of PCG to
games again with the use of a simulation-based fitness function. However, the use of
simulation-based fitness function is not suitable for our goals, due to the time it takes
to create a map is long and it is not suitable for practical implementations due to the
strict real time requirements imposed by the games and game players. Work done by
Cardamone et al.
[3] to evolve interesting maps for a FPS leveraging on SBPCG is a
great starting point for our research. However, a great amount of work have to be done
to ensure that the map can be generated in a span of seconds. Moreover, the maps that
are generated are seemingly low on navigability and aesthetics, which are basic features
of any good multiplayer game. In contrast, navigability and aesthetics are part of our
design goals.
