Game Development Reference
In-Depth Information
However, some hypotheses are expensive or impossible to test in reality, includ-
ing those about very large or very slow systems (such as the behavior of galaxies) or
events in the past (such as geological processes). In these cases, the scientist forms
a hypothesis from observations as before, but instead of running experiments, she
builds a simulation that models her hypothesis about how nature operates. She then
runs the simulation and compares its results with more data from the real world. If
the simulation produces results that differ from the real world, she revises both her
hypothesis and the simulation.
Once a hypothesis seems to be solid—it has been supported by many experiments
or observations and never disproven—it becomes a theory and can be used to predict
future events and plan construction or other activities. Scientists can use simulations
to predict such things as the time and place of the next solar eclipse, for example,
and engineers can use them to design buildings and aircraft.
Scientific simulations focus on accuracy: They model the important aspects of a
system as closely as they can within the limits of the available time and computing
hardware. It is very important that the simulation's model resembles the real mech-
anisms of the system the simulation represents, and in order to refine the model,
scientists and engineers check it against available real-world data. In semiotic terms,
we might say that the simulations are
iconic
: The signal (the simulation rules) resem-
bles its meaning (the real mechanisms).
simULaTiOns in Games
In ordinary games the designer's object is not accuracy but enjoyment. The designer
starts with a game idea and refines it into a game design. Although it may change
over time, the design is a static rather than an interactive thing, a collection of
documents and diagrams and notes taken at design meetings. Programmers then
write software that implements the systems specified by the design. In many genres,
the software simulates something: a vehicle, a battle, a city. Both the designer and
the programmers may borrow ideas from observations of reality (such as the law of
gravity or the performance characteristics of aircraft), but they often ignore or alter
real-world systems for entertainment purposes. This produces the peculiar con-
ventions found in video games, such as cartoon physics: Characters can fall much
longer distances without hurting themselves.
Instead of testing a game's simulation by comparing with reality, game developers
play test it for enjoyment. When we refine our simulation, we refine it to improve
the entertainment value it delivers, not the accuracy with which it reflects the real
world. We care about accuracy only if the players care about accuracy. With vehicle
simulations or sports games, the players frequently do care about some aspects of
accuracy; but in other genres, they care much less. It's important to know just which
aspects matter to your audience.
In semiotic terms, games use indexical and symbolic signs much more frequently
than they do iconic ones. Instead of trying to actually show a fighter's state of
health through the appearance of the fighter, which would require a very large
