|Given the recent focus on next-generation game consoles, with their graphics, processing, and networking improvements, it’s clear that gaming hardware is progressing at an impressive rate. But is the software—the games themselves—advancing as rapidly? At first glance, it appears not, but that may be because the improvements are less obvious and the problems are more complex.
To examine this notion, consider one of the grand challenges of game programming: creating sophisticated interactive narratives, that is, stories that adapt and unfold in real time according to the player’s direction. What is often regarded as the ultimate form of this concept, as Star Trek fans know well, are the personalized, interactive dramas that occur on the starship Enterprise’s holodeck. In this empty black cube crisscrossed with white grid lines, the ship’s computer uses force fields and energy-to-matter converters to create realistic holographic environments and characters. It then continuously adapts the behavior of the world according to the actions of the player immersed in the story to create an intense, interactive play that remains true to a powerful narrative theme.
Of course, we’re light-years from developing anything remotely like the hardware technology of the holodeck. But making even small strides in developing software that combines poignant stories with effective game-play could revolutionize entertainment as we know it. Unfortunately, integrating these two elements has not proven easy.
One of the clearest descriptions of the obstacles to meeting this challenge was presented at the recent Game Developers Conference by game expert Ernest Adams. A primary reason we’ve had so much trouble merging interactivity with narrative, he contends, has to do with a failed analogy that says there is a direct correlation between the dramatic tension found in movies and the game-play tension found in games.
On one hand, the two are alike in that they both involve conflict-between people, between a person and the environment, between a person and himself or herself, and so on. But games and stories are worlds apart with respect to the kind of tension that is required for each to be successful on its own terms, Adams explains. Here are some features of games that sustain game-play tension but would extinguish dramatic tension:
Repetition: In games, you often have to repeat the same steps over and over again, and the game-play tension is maintained, and even enhanced, as you try to achieve your goal. But in narratives, if the same event happens over and over again, the dramatic tension quickly dissipates, and you will quickly lose interest.
Randomness: In most games, such as board games, random chance adds to game-play tension. But in stories, you would find it unacceptable if a completely random event determined the outcome, especially if no groundwork had been laid for it in advance.
Backtracking: Many games, particularly adventure titles, require that you make trial-and-error explorations, which frequently result in dead ends and require you to back up and start all over. But in stories, you are not asked to spend a lot of time pursuing a course that turns out to be totally irrelevant.
Excessive Detail: In games, you are involved in every event in chronological order. And in sports titles, for instance, game-play tension is preserved because all of your actions contribute in some way to the final outcome. However, if a narrative were to recount every single detail of every event, the dramatic tension would soon be lost.
Emergent Themes: In MMORPGs (massively multiplayer online role playing games), stories tend to emerge as you interact with other players. And, depending on the environment you’re in, you can experience a full range of emotions, around which entertaining stories can develop. But emergent stories have no fixed structure. Imagine trying to play a character in tightly woven story like Casablanca without knowing the basic plot. Your actions or those of the other players could lead to any of the above problems.
All of this is not to suggest that the differences between game-play tension and dramatic tension are completely irreconcilable. In fact, over the past decade, several interactive narrative techniques have shown great promise in merging the two-including the Oz game engine, based on AI techniques developed at Carnegie Mellon; and the Erasmatron storytelling engine, built on probability-based algorithms created by Chris Crawford.
More recently, a group headed by Michael Mateas and Andrew Stern has been working on an interactive drama called Façade, which casts you as a guest at a get-together in the apartment of Grace and Trip, a couple whose marriage, you soon find out, is in serious trouble. During the real-time one-act play, you spend an intense evening with them, and your participation determines the flow and outcome of the story. The soon-to-be-released 3D-animated production uses AI programming to move beyond standard story-branching and hyper-linking techniques. The AI controls Grace and Trip, including their dialogue, expressions, and body language. It also maintains dramatic tension by determining the story direction based on your recent actions and the events that have transpired so far.
While these projects represent just a few of the efforts that have been undertaken recently, they illustrate both the great promise and challenge of creating interactive narratives. Indeed, while it seems likely that the obstacles to creating a new generation of game software will be overcome long before game hardware designers will be able to devise holodeck-like energy-to-matter converters and the like, the task is turning out to be more difficult than once imagined. Yet it’s just another of the many challenges worthy of the talent and creativity of the people working in this industry.
Final Note: This will be the last issue I will be serving as editor of Computer Graphics World. Next month, I will move over to become editor of CGW’s sister publication, Solid State Technology, which covers the semiconductor manufacturing industry. It has been an honor to have worked with the CGW staff and with so many of you for more than 10 years. If you wish to contact me, my new e-mail address is firstname.lastname@example.org. Best wishes,Phil LoPiccolo