Inventing The Future Of GamingThe University of California's Santa Cruz campus held a Research Review Day last week to let the public take a look at some of the research projects underway in Engineering. The field of gaming was well represented, with a number of the computer science professors talking about research projects, and students showing some of the games and tools they had created. It was a glimpse of what we may see in games of the future, and the future may not be far away.
The essence of UCSC's efforts was neatly summarized by Associate Professor Michael Mateas in his talk, The Quest For Procedural Everything. “What is games research, and what is the role of the university?” Mateas asked rhetorically. “Educate students, of course, but also to perforrm cutting edge research and engage in the production of new knowledge.” He went on to cite some examples of important things that university research has produced, including Arpanet (which became the Internet), TCP/IP (the basic communications protocol for the Internet), and PageRank (the algorithm at the foundation of Google). Mateas noted that “Game research is often considered an oxymoron”, but research has an important impact on games.
Mateas used a simple analogy to explain the importance of game research, and the tools that research can create. “If you're given crayons, it limits what you can draw. You can't produce the Mona Lisa with just a box of crayons.” The game industry has been working hard on graphics, as evidenced by the progression from Pong to photorealism. Mateas feels that the game industry's focus has really been about Renaissance artwork's prime interest, that of mapping 3D objects onto a 2D surface. “Games are mostly graphics with movement and collision detection,” said Mateas. But shouldn't they be more than that?
“What about representing human nature? What are the equivalent of graphics for relationships and interaction?” Mateas went on to say that “Systems are the heart of what makes games unique.” He feels that the quest for their research is to make fundamental technical representations to enable game designs. “We want to create procedural everything,” Mateas said. He envisions systems to generate content, the equivalent of physics engines for characters, interaction, and stories.
"Systems are the heart of what makes games unique." -- Michael Mateas
Jim Whitehead, Professor of Computer Science and founder of the game design degree at UCSC, took up that thread when he talked about Infinite Games: Procedural Content in Game Design. He noted that computers can generate amazing shapes and patterns, and can be creative by themselves or with people. Procedural content generation can use the computer to augment a designer's insights. In fact, the Augmented Design Laboratory at UCSC has several projects underway with procedural content generation. Tanagra is a 2D platformer design tool that works with a human to generate levels. Further on, what if a player is providing input to a procedural content generator, rather than a designer? Levels could be generated automatically based on the player's actions. In other words, infinite content.
Assistant Professor Arnav Jhala talked about efforts to create an AI for camera placement. He noted that technologies for film and games are converging, as some of the techniques used to interpret facial expressions for Avatar were used in L.A. Noire. Jhala sees the emergence of computational cinematography in the future. One recent experiment they ran used a 3D version of Pac-Man, but varied the camera angles and tracking, while measuring the heart rate and skin conductance of the players. As you might expect, players with more limited information (where the camera showed them very little of the playfield and what was coming up next) exhibited more frustration. Such information will be fed back into the design of AI for cameras as they work to create a more intelligent camera for games.
The day ended with a number of graduate students showing examples of their research efforts. Many of the ideas and concepts in these games could be utilized in games on nearly any platform. Game developers and publishers would be well served to look into the research being performed at UCSC and other universities; the ideas and tools being developed in research labs could create some richer game experiences. While much of the research is still in the early phases, many things could be used in games being built today.
