Game Development Reference
In-Depth Information
Other times, we want to test the high-skill balance of a game. This
requires players who can play intensely for long periods of time. Usually this
means having a team of dedicated playtesters who work on their skills daily.
There are variations between these extremes. For example, in my
combat design process, I tested with coworkers who knew the game, but
who didn't know the specific combat I was working on. So their initial
knowledge approximated that of a real player encountering the combat
after playing several hours into the game. They knew the game, but not the
particular fight I was making.
There are also other ways to divide playtesters besides knowledge of
the game. You can test with children or seniors, or people of different
cultures, socioeconomic backgrounds, or interests. In general, choose a
mix of testers who resemble the people you want to play the final game.
samPle size
It 's easy to fixate on a single playtest result. Since your brain instinctively
believes that what you see is all there is (WYSIATI), it'll trick you into
thinking that that one experience is the entire game. But it often turns
out that the first test run was just one unimportant thread through a large
and diverse set of possible experiences. This is why playtesting well means
playtesting a lot.
Good design decisions can only be made when a designer has built up an
understanding of all the different experiences the game can generate.
This means doing many playtests.
Without this broad mental context, designers will tend to solve the
problems with the experiences they saw while causing problems with ex-
periences they didn't see. The game might keep changing, but it won't
improve because every solution causes more problems.
To make real progress, we have to solve problems with one experience
without causing others elsewhere. This is impossible if we've only seen
one or two of the threads players can follow through the game. We have to
know everything a game tends to do across all players. Then we can pick
out the design solutions that solve all their problems at once.
The process of gaining that context is straightforward: watch many
playtests. Each playtester shows you a new thread through the game's
possibility space. After you've absorbed enough of those, you'll develop a
