Game Development Reference
In-Depth Information
15
Changing a Decision
In the last chapter, we illustrated many aspects of what goes into a making a de-
cision. Indeed, throughout much of this topic, we have thought about a decision
as a static, isolated event—a puzzle to be solved. We have even examined some
of the examples from various starting states.
In the example in Chapter 13, we looked at how the weighted decision model
for “what to do for dinner� yielded a different output
if
we offered it different inputs.
We didn't address the situation in terms of
when
the inputs change. The scenario
for a given evening (i.e., the time and price restrictions and our desire for types of
food) was presented as inputs to the algorithm. The algorithm crunched the num-
bers and gave us an output. We were done. There was no further consideration
given to what to have for dinner. One situation led to a single choice.
Similarly, in Chapter 14, we laid out numerous starting conditions for our
Dude assault. What types of Dudes are we facing? What weapons are they carrying?
Where are they? Where is the detonator? We can represent all of those conditions
in a single slice of time—the world as it is
now.
That slice is as thin as the simulation
rate—a single frame of information… literally, a fraction of a second. We proceed
with the decision process based only on that snapshot and yield a decision just as
static. This is what we are doing
now.
But then what?
The problem with a static, thin-sliced mentality is that time and circumstances
are usually fluid. Situations change. Information varies. Inputs fluctuate. Accordingly,
our decisions must adapt as well. On the other hand, changing our decision can be
a weakness as well as a strength, as we shall examine in a little bit. Before we get too
involved with the caveats, however, we need to address whether we need to change
our minds in the first place.
395
