Made available in DSpace on 2014-12-17T15:48:00Z (GMT). No. of bitstreams: 1 PhilipMD_DISSERT.pdf: 3999230 bytes, checksum: fd3b6c186bc10448c63b511e10e6017e (MD5) Previous issue date: 2012-02-26 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico / The game industry has been experiencing a consistent increase in production costs of games lately. Part of this increase refers to the current trend of having bigger, more interactive and replayable environments. This trend translates to an increase in both team size and development time, which makes game development a even more risky investment and may reduce innovation in the area. As a possible solution to this problem, the scientific community is focusing on the generation of procedural content and, more specifically, on procedurally generated levels. Given the great diversity and complexity of games, most works choose to deal with a specific genre, platform games being one of the most studied. This work aims at proposing a procedural level generation method for platform/adventure games, a fairly more complex genre than most classic platformers which so far has not been the subject of study from other works. The level generation process was divided in two steps, planning and viusal generation, respectively responsible for generating a compact representation of the level and determining its view. The planning stage was divided in game design and level design, and uses a goaloriented process to output a set of rooms. The visual generation step receives a set of rooms and fills its interior with the appropriate parts of previously authored geometry / Recentemente a ind?stria de jogos vem experimentando um aumento consistente nos custos de produ??o de jogos. Parte deste aumento ? referente ? tend?ncia atual de se ter ambientes cada vez maiores, mais interativos e rejog?veis. Esta tend?ncia se reflete num aumento das equipes e do tempo de desenvolvimento, o que torna o desenvolvimento de jogos um investimento de risco e pode reduzir a inova??o na ?rea. Como uma poss?vel solu??o para este problema, a comunidade cient?fica vem apostando na gera??o procedural de conte?do e, mais especificamente, na gera??o procedural de cen?rios. Dada a grande diversidade e complexidade dos jogos, a maioria dos trabalhos opta por trabalhar em g?neros espec?ficos, sendo os jogos de plataforma um dos g?neros mais estudados. Este trabalho prop?e um m?todo de gera??o de cen?rios para jogos de plataforma/ aventura, um g?nero mais complexo que jogos de plataforma cl?ssicos e que at? o momento n?o foi alvo de estudo de outros trabalhos. Dividimos a gera??o de cen?rios em etapas de planejamento e gera??o visual, respons?veis respectivamente por gerar proceduralmente uma representa??o compacta de cen?rio e determinar sua visualiza??o. A etapa de planejamento se divide em game design e level design, e se utiliza de um processo orientado a objetivos para produzir como sa?da um conjunto de salas. A etapa de gera??o visual recebe um conjunto de salas e preenche seus interiores com partes adequadas de geometria previamente constru?das

Context. Procedural content generation (PCG) refers to algorithmical creation of game content (e.g. levels, maps, characters). Since PCG generators are able to produce huge amounts of game content, it becomes impractical for humans to evaluate them manually. Thus it is desirable to automate the process of evaluation. Objectives. This work presents an automatic method for evaluation of procedural level generators for 2D platform games. The method was used for comparative evaluation of four procedural level generators developed within the research community. Methods. The evaluation method relies on simulation of the human player's behaviour in a 2D platform game environment. It is made up of three components: (1) a 2D platform game Infinite Mario Bros with levels generated by the compared generators, (2) a human-like bot and (3) quantitative models of player experience. The bot plays the levels and collects the data which are input to the models. The generators are evaluated based on the values output by the models. A method based on the simple moving average (SMA) is suggested for testing if the number of performed simulations is sufficient. Results. The bot played all 6000 evaluated levels in less than ten minutes. The method based on the SMA showed that the number of simulations was sufficiently large. Conclusions. It has been shown that the automatic method is much more efficient than the traditional evaluation made by humans while being consistent with human assessments.