A Balanced Approach
Hallford and Hallford, Rouse, and Smith strongly advocate the use of generalized simulation design techniques as opposed to a case-based approach. But is it really possible to integrate every aspect of a game within a generalized system? One eloquent thinker on this topic is game designer Marc LeBlanc: By "simulation" I mean game systems that are rich, many-faceted analog models, whereas by "emulation" I mean coarse, case-based game systems. An example of "simulation" might be the physics system in a shooter or the damage model in an RPG. The rules of the system create a space for exploration, allow for emergent complexity, and so forth. An example of "emulation" might be, in either game, a button that opens a door. There is no physical simulation of the electro-mechanical process of opening the door. It's just a single rule: "if button pushed, then door opens." The button is more of a semantic object than a physical object; in some sense it exists only on a functional level. Emulation is the smoke-and-mirror approach. [13]
LeBlanc brings to light a point we made at the start of this chapter: simulations can be embedded. Even though an entire game can be considered as a single representation, it is important to be able to identify the smaller procedural representations that make up the larger whole. Although LeBlanc shares a preference for generalized simulation design strategies, he takes a measured approach and ultimately suggests a balance of case-based and generalized techniques. No game can rely entirely on rich, open-ended, emergent simulation. Nor should it. To demonstrate this balancing act, LeBlanc takes an example from a title he helped create, the computer game Thief, in which players can use water to put out a torch:This is an emulated game system; there's a single rule that says water puts out fire 100 percent of the time. There's no simulation; no chance of using too little water and just getting steam, and no chance of drowning the torch so that it can never be relit. However, the other systems that interact with that system are simulations; you typically douse a torch by tossing a "water balloon" at it, the motion of which is physically simulated. The way the light from the torch affects your vision, and more relevantly, the vision of the opponents you want to stay hidden from, is also simulated.[14]
(Remember that LeBlanc is using his own terminology—for example, he is using the term "simulation" to mean a generalized design approach, which is only one part of what we identify as the larger issue of simulation in games.) LeBlanc's example contains intersecting systems of representation, some case-based and some generalized. In Thief, water puts out fire 100 percent of the time. Rather than a complex physics simulation with a variety of possible results, there is a predictable and consistent outcome: water extinguishes fire. But this case-based approach interacts with more complex structures. Tossing a water balloon at the torch involves the generalized simulation of motion; the diminished visibility from the extinguished torch results from a generalized simulation of light. The whole of the game representation emerges from the complex relationships of these parts. In fact, most games use an approach that combines case-based and generalized representations. The Deux Ex characters of thug, civilian, and military agent were designed with generalized heuristics that produce emergent behavior. But other characters in the game were designed with simple, case-based behavior, such as a character that activates a cinematic cut-scene when encountered. Even simple board games such as Candyland combine a generalized movement system with a number of case-based special spaces on which players can land. If Rouse, Smith, and Hallford and Hallford are correct, if a generalized approach saves time, increases emergence, and provides players with richer play, why would a digital game designer (or any game designer) use a case-based approach for structuring a procedural representation? One answer is that a truly generalized system could easily become overly complex to implement and might not save work time in the end. Another answer is that although case-based approaches can sometimes become simple and flatfooted, the opposite danger is true of generalized strategies. Taken too far, generalized simulation design can become fuzzy and ambiguous. If you are playing a game and you see a group of birds that suddenly takes to the air for no reason, their behavior will seem random and meaningless. It might be that their behavior is simulated in such detail that their internal clocks told them it was time to leave the scene and migrate south for the winter. Unfortunately, the accuracy of the simulation would be lost on the player, who has no way of knowing how the meaning of the birds' actions fit into the larger game experience. Neither case-based nor generalized design strategies guarantee a successful game experience. The goal of game design is the creation of meaningful play, which should guide the selection of representational strategies in a game. In the example of the Zelda witch, a compelling character was created out of a central case-based procedure: if the player brings a mushroom to the witch, she will turn it into magic powder. This procedural representation was well-integrated into the other more generalized representational strategies of the game (the witch's hut is isolated, difficult to find, and dangerous to reach; mushrooms are hidden in the forest; magic powder has special effects in the game). Thus the meaning of the witch gained its power from the total game context in which it was experienced.
The fact that rich meanings can emerge from a representational context not based on software complexities offers an important insight into game design and simulation. As we've mentioned, games currently suffer from a narrow range of simulated subject matter. Although there are important historical reasons for the prevalence of military and economic conflict in games, other forms of conflict, such as social, cultural, or emotional conflict, can and should be represented as well. Some presume that an increase in technological complexity will make such representations possible. For example, a widely published quote from an executive at a major console manufacturer not too long ago looked forward to the day when the faces of game characters could represent emotion. According to the executive, on that day, games would become a mature and sophisticated form of cultural production. Clearly, media such as literature, theater, and comics have been capable of sophisticated representation for centuries without relying on high-resolution animation. Furthermore, even within the history of animation, many animated tears have already rolled down numerous animated cheeks. That fact alone is no guarantee that the story was meaningful to its viewers. In the case of computer games, although animated elements do play a part, the systemic and interactive qualities of the form have to be taken into account when envisioning future directions for the medium. The procedural representation of new kinds of game content is within our grasp, but new content can only be discovered by paying attention to the fundamental principles of game design and meaningful play. Game designers need to cultivate a deeper understanding of the form in which they work. This is especially true in considering games as simulations. More than just choosing a representational design strategy, there is a complex interplay between a simulation and its simulated subject. It is to this relationship that our attention now turns. [13]RE:PLAY: Game Design + Game Culture.Online conference, 2000 <www.eyebeam.org>. [14]Ibid.