Archive: April 11th, 2008

The Simulation Gap

[General,Research] (04.11.08, 10:59 pm)

Simulation is important because of the way that it works at a fundamental level. Simulation consists of:

  1. A mapping between things that are outside the simulation to things that are inside the simulation, a representation. The representation converts things in the world to tokens or symbols in the simulation itself.
  2. A set of rules that defines the relationship and interactions between the tokens, a model. Models are interpretations of how things work.
  3. An actual execution and application of the simulation, where the state changes from some set of initial conditions to some later conditions after a period of time. That is the run of the simulation.

Among these is the gap. The gap is a property that emerges from simulation, that is a consequence of its definition. A simulation is dependent on both rules and representation. Without rules, the run of a simulation is nothing more than an arbitrary sequence of states. Without representation, a simulation is just a mathematical entity in abstraction, disconnected from reality. However, simulations can take advantage of one more than the other and still be called simulations. This means that there must be some sort of negotiation between what is handled by the model and what is handled by the representation. The disconnect and distance between these is the simulation gap.

The magic of simulation occurs when the model and representation work synergetically to produce something that seems to work beyond both. Ultimately, the simulation here is something experienced. Simulation authors must therefore use the gap to creative and expressive advantage. Many games are successful because of their care in choosing what to simulate versus what to represent.

Closely related to this is the immersive fallacy. The nut of the immersive fallacy is the idea that a game can be made better by expanding player’s agency in the game world. The line drawn between in this idea is between player agency and simulation. I guess that would be defined as the agency gap. Generally I have been concerned with the simulation gap, but the agency gap is very significant due to its construction of believability through participation. The simulation gap is significant because its construction of believability through mechanics.