Minimizing Repetitive Play in Action Games
By:
One
of the most common problems that we’re seeing in many action games today is how
to make them fun for a long time. A majority of the action games on the market
can have a tendency to become very repetitive or boring after only a few hours
of game play. This problem usually is the fault of the game design.
There
are many different kinds of action games. Some games like Devil May Cry or Buffy: the
Vampire Slayer,
could be classified as just a pure action game, while a game like Halo or Medal of Honor would be considered a first person shooter. Games
like Tomb Raider and Legacy of Kain:
Soul Reaver 2 would be considered an
action-adventure game, while a game like Diablo
II or Baldur’s Gate: Dark Alliance
would be considered an Action-RPG. Another form of action game are the platform games like Jak and Daxter, Crash Bandicoot and Rayman 2. Another type of action game are the beat-em-ups, like Dynasty Warriors and Bruce Lee: Quest of the Dragon. Fighting
games like DOA3 and Tekken also can be considered a type of
action game, even though their rules are slightly different. You can also get
hybrid games like Oddworld: Munch’s Oddysee, which has many
different elements in it and is hard to classify. Many action games also are
predominantly focused on one kind of game play, but have elements of other
kinds of play. As you can see, there are many different kinds of action games,
but a majority of them can suffer from the same problems.
Every
game starts with an idea. Every game usually has a hook. In Soul Reaver,
you could go to another similar plane of existence, in Medal of Honor you are reliving a historical event, in Munch’s Oddysee you can possess and talk
to characters (amongst other things), in Dynasty
Warriors you can fight in huge epic battles and kill thousands of people
and in the Mark of Kri
you can fight several enemies at once in new ways. The hook is the thing which
sets the game apart from the competition. It might revolve around some new
innovative game play idea, or it might revolve around a new story, visuals or
other aspect of the game which hasn’t been done in the same way before.
The
problem with a hook is that many designers assume that all they need is a hook.
They often feel that the hook alone is enough. I’ve seen a ton of game designs
which took one simple idea and tried to make an entire 10-15 hour game out of
it. Some of this is symptomatic of our heritage. A lot of where our action game
design has evolved from is from arcade games, which have a very different
design philosophy. The other part of our heritage was early consoles, which had
small cartridges and simple games due to many different limitations. Because of
this, I think we often still think in this very limited manner when it comes to
our game ideas.
For
a game to be great and rewarding for a long period of time there is a lot that
must come together. A game with overall high production values throughout the
game will be much better accepted than a game which is strong in some areas,
but weak in others. Therefore, the most important thing you can do is to just
make a great game. It will be pointless to try and keep the game from getting
repetitive if the overall game sucks. Having a bad camera, bad controls,
horrible graphics, bad framerate or other very
noticeable problems will keep any game from reaching its full potential. People
can forgive weak stories, bad music and other flaws, as long as the rest of the
game is great.
Keeping a game rewarding for a long period of time is more
than just adding some replay incentives or ability. If the game isn’t extremely
rewarding the first time around, then most people
won’t bother to play it again. Most people also rarely finish a game, so adding
lots of replay ability also isn’t seen or used by most people. If push comes to
shove, just making a great game is much better than having a good and highly
re-playable game. Keeping your game from being repetitive and adding replayabilty however are two different things, which are
often lumped together. You need to make sure that the game is first and
foremost not repetitive and then if you toss in some replayability
it’s icing on the cake.
There
are two main types of games when it comes to their ease of use. There is simple
and there is complex. There are a lot of advantages to making a game that is
simple. Simple does not mean simplistic however. A simplistic game might be
something like an old Atari game remake or Tetris.
A simple game will help the player pick it up and figure it out quickly, it is
intuitive, it’s easy to play, doesn’t require a ton of time to play and can be
enjoyed by a wide range of people. Devil
May Cry is a good example of a fairly simple action game. Whereas a game
that requires you to read a manual, go through a long tutorial or is constantly
making you figure out new things can be considered a complex game. A lot of
games have some complexity in them. What a lot of games are missing however is
their understanding of what deep game play is.
A
deep game doesn’t have to be complex. Deep game play relies on the use of
rules. Rules in games allow people to know what they are supposed to do in a
game. These rules are often based on real world rules or laws, like Physics. In
a game like Mario Sunshine you can
fall from very high and not hurt yourself, whereas in a game like Tomb Raider you know that if the fall
looks too high, then you probably can’t survive it. The same rules can apply to
weapons. A person playing Medal of Honor
expects all of the weapons to act realistically and can predict their behavior,
whereas in a game like Halo where
there is science fiction weapons, the player must learn a new set of rules.
So
how does knowing the rules of a game add to the minimization of repetitive game
play? Very simple rules, or what we can also call game mechanics, can be
combines together in order to form more complex or deeper game play rules or
mechanics. Say for instance you have a gun, and you know that the gun shoots.
You then figure out that barrels explode when shot, so it’s now logical that
you can shoot at barrels next to enemies to kill them. However, later on, maybe
there are some barrels next to a door which you can’t open, so you find that
shooting the barrels will blow open the door. Later on in the game you might
then get a force field, which protects you from explosions, and you find that
if you shoot a barrel it will explode and send you flying, which if you have
your force field on could actually propel you somewhere you were unable to
reach before. While this is also a kind of puzzle, what you have done is take
several rules which the player knows: guns shoot and barrels explode when shot
and combined then together in some different ways to add some new twists to the
game play. The player still has only had to learn two rules.
The best way to understand these rules is to look at some
games and see how they used their rules to either create a fun long term
experience for the player or how they accidentally created some long term
problems in their game which made the game unfulfilling for players. Every rule
or mechanic in the game therefore has relationships with other rules, and it is
these relationships which can add a tremendous amount of depth to your game,
without adding a lot of additional complexity.
There is another reason for establishing rules and
relationships in games. Another kind of game play rules is called Emergent
Behavior. Emergence is defined as something new: appearing, arising, occurring,
or developing, especially for the first time. Therefore Emergent game play is
basically when you have a global set of rules, but these rules can interact
with each other in a wide variety of different ways. Grand Theft Auto 3 is the best example of a game which is primarily
based around emergent game play. Some people also talk about emergent game play
as building a sandbox. It follows that if you give the person (player) a bunch
of tools (mechanics) and a big area to play in (the game) then they will figure
out what they want to build (how they are going to play the game). Emergent
game play has a lot of benefits, many challenges and some disadvantages. If you
are trying to build a game using this kind of a system, you really have to take
a close look at what you’re building and what kinds of rules are appropriate.
Not every game can benefit from this type of rules structure.
Emergent play can be continually rewarding for the
player and help minimize repetition because you’re able to do so many different
things. In GTA3, you can follow the story and do exactly what you are told, or
if you get bored or stuck you can go off on your own and just play in the world
for awhile. GTA3 is successful because it lets you live in the world, steal
cars, beat up people, break the law and get into all kinds of trouble yet it
still allows you to follow a linear adventure. A good example of emergent game
play in GTA3 can be found in a situation I encountered while playing. Some
thugs jumped out of an alley and started mugging some old lady. I had the
choice of helping them, helping the old lady, beating them off and then mugging
her myself, or just ignoring the situation and walking on. While none of these
solutions directly affected my game, I was still allowed to do anything I
wanted, and the game would react accordingly.
Emergent game play has the disadvantage of not being
able to easily tell a cohesive story. You can tell parts of stories, but it’s
hard to tell a very well timed and precise story, since you don’t know where
the player is going to go or what he is going to do. There may be some
solution, but it’s not going to be an easy problem to solve.
You will probably start to find a lot of people
talking about emergent game play and how to do it correctly. No matter what you
do, trying to create a very open world, with lots of different possibilities is
extremely challenging, and needs to be properly evaluated before you jump too
far into it.
There are many different areas in games which can make them
repetitive, tedious, frustrating, annoying and boring. Most of the problems
listed in this section would seem to be obvious, but it’s surprising how often
we still see them in games. A good usability tester or playtest
should help you find and solve many of these problems before you ship your
game. Just remember that you need to perform deep gameplay
tests, to make sure that the game stays fun for everyone after the first few
hours of play. I have seen many games pass the first hour or two of testing
with flying colors, only to get a sharp dropoff after
the first few hours of play.
Every
character has a certain number of moves they can do. Sometimes all of the moves
are given to the player and taught to them at the beginning of the game, while
other games spread out the moves. Most first person shooters have little real
moves, mostly forward, backwards, strafe, jump, duck type moves. Platform games
typically have the most moves, with the player being able to often run, jump,
attack in many different ways and more. Fighting games are almost nothing but a
series of moves.
The
problem with moves is complicated by the need to keep the game simple and easy
to play and with the difficulty of actually controlling a character in 3D
through a wide variety of moves. You can see this problem most easily in
fighting games. When fighting games transitioned from 2D to 3D (and I’m talking
about fighting games which allow for a full range of 3D movement, not just a 3D
engine), there were a lot of problems. It was difficult enough to have to have
players memorize 50-100 different moves and then try and fit them all onto a
console controller, but when you added the need for 3D movement and often
camera control, things got a lot trickier. Fighting games had an easier time
transitioning though, since they take place in smaller arenas and had less
complexity than action adventure games.
In
an action adventure game, you have the possibility of needing a stupid number
of different moves and abilities in the game. If you offer too few moves, like
if you take away jumping, players can get very frustrated. If you offer too
many moves, players can’t control their character very well, especially if
they’re fighting a bad camera and other quirks in the controls.
Mario Sunshine
is a good example of a game that has a ton of different moves on one hand, but
they can only be used in a few ways. The problem however is that many of the
moves are so difficult to pull off, that you end up having to repeat the same
thing over and over, because you just can’t physically do the move, even though
you know what you’re suppose to do. This is complicated by a very manual camera
that hinders you as much as it helps you, so that pulling off the move can be a
challenge since you can’t see what you’re doing. Devil May Cry also suffers from a similar problem occasionally
because of it’s fixed cameras, which make it so that you occasionally can’t see
what you’re doing, or you’re trying to attack enemies off camera or something,
making it impossible to know which moves you need to pull off.
The
best solution overall I’ve found however is to allow the player to either learn
new moves, discover new moves, buy new moves or somehow learn new moves for at
least the first half of the game. If you combine this with the introduction of
new weapons, abilities and items throughout the game, you can provide the
player with a lot of depth in the game.
In the majority of action games these days, you’re bound to
have some kind of a weapon. Whether it is a hand to hand weapon, shoots
projectiles, launches missiles, fires a laser, drops a bomb or does any one of
a million different possible attacks, the problem is often the same. Many games
have weapons which are either too few, too limited in their use, too similar or
too realistic to add a lot of depth to the game. If you’re making a realistic
game like Medal of Honor, then you
may not have too many choices in your selection of weapons and your job is
going to be a lot harder. Weapons can also take the form of magic and other
kinds of abilities.
The extra abilities that a player has also have the same
problem as the weapons and moves in the game. Whether you’re giving your
character night vision goggles, a jet pack or some other gadget, they all have
the same long term problem. No matter what the player uses, it will get old
after awhile. If he has to continually keep solving the same puzzle, or doing
the same thing in the same way with the same item, it will get old. The idea
here is to limit repetition and add depth to the game.
Just
adding more weapons, items and abilities isn’t always the right choice however.
Some games like Diablo have a
tremendous number of weapons and items that you can collect. Granted, this is
part of the addiction of the game, but it also reaches a point where you stop
caring. You find your favorite weapon and just stick with it, only occasionally
getting enough money or lucky enough to get something really cool. So while
variety is important, it is more important to introduce new things which add game
play value and variety to the game, not just some different art and a few stat
changes.
Every
weapon and ability in a game should be carefully chosen, and add something new
to it. What does it achieve to have a ton of different weapons, if they all function
basically the same? Would the average player know the difference between an
M-16 and an AK-47, or between a long sword and a bastard sword? If you have the
choice on what kinds of weapons you are giving the player, it is therefore most
useful if you think about adding classes of weapons. Each class of weapon
should serve a different purpose. There are many possible types of weapons
classes: Handguns might have plentiful ammo, but only be good at close range
and are fairly weak, rifles shoot long range, bazookas and rocket launchers
blow things up, grenades are thrown by hand, etc. So instead of giving the
player 8 different weapons (or whatever the number is), you should think about
giving the player classes of weapons throughout the game which will effect the
way he plays the game. A sniper rifle is a good example of a different class of
weapon, which can drastically alter the way someone could play a game.
Something
else to consider when creating more interesting weapons, which don’t add a ton
of complexity to the game is to add new weapon powerups
or modifications. In Diablo II you
can add gems to some weapons to make them more powerful, in other games you
might be able to add a grenade launcher to the machine gun you’re using. The
player early on might learn that there is two ways to attack with each weapon,
and then when they get a modification and they choose to equip it, they don’t
have to learn anything new.
You
want to be careful though and make sure that the new weapon, item, ability or powerup you give the player is rewarding however. If the
player doesn’t see an obvious improvement in the way they can fight, or get
some other kind of reward, it can lead to a lack of satisfaction on the part of
the player. In Devil May Cry each
weapon and move you get is visually cooler and better than the last, and each
new weapon you get can do different and flashier stuff. Getting new things in
the game feel very rewarding and satisfying to the player, so that you’re
driven to want to do better and buy the next thing.
A small annoyance, which is sometimes easy to solve, is the
repetitive audio in games. This can sometimes be the music, but is often the
voice acting in the game. If your character says something after every kill
they make, and you’re going to kill a ton of people, and what they say is
always the same, people will often stop playing the game just because the audio
has gotten too annoying. You need to make sure that you offer a variety of
sounds in the game and that they don’t totally frustrate the player because
they play too often.
Looking at the same animation over and over again can get
very boring as well. While this is a minor complaint by many people, you should
think about ways to add lots of random animations in the game. In a fighting
game, if you continually pressed high punch, and your character did the same
move over and over, it would be bad. However, you as a designer know what the
players intention is, so a high punch could translate into a jab, a hook, a
spinning backfist, a chop, an elbow or any number of
different moves and yet still accomplish the same results and not require any
additional learning or thought from the player. In The Mark of Kri they did a wonderful job
at giving your character over 20 different ways to stealth kill enemies, and
not only is it cool to watch, but you never get bored
watching it, since there is a lot of variety. Think about ways to add some
variety into your animations to keep players interested.
A lot of action games have stories. Very few action games
have good stories. Only a couple of games have anything close to a great story.
If players are constantly being forced to watch long movies and are unable to
skip through them, it can get very annoying. If this pattern continues
throughout the game, players can get annoyed.
However,
some games like Final Fantasy, Metal Gear Solid and The Bouncer are almost just large
interactive movies. The difference is that the expectation for these games has
been set so that players expect this and actually often look forward to it. So
it’s not impossible to include a very long and deep story, but it has to be
done well if the players are going to stay interested throughout the game.
Games like Soul Reaver
2 have a tremendous amount of story, and while well done, it gets very
tedious and is so complex and long that you tend to loose track of what is
going on in the game and stop caring.
If
you’re going to put a lot of story throughout your game, try to keep it’s length to a minimum and spread it out. It’s also
critical that the player can bypass the story scenes and that nothing critical
to the level or game is told in the story sequence that would keep the player
from progressing in the game if they miss the information or skip the cutscenes.
Levels are boring for a lot of different reasons. It’s very
hard to make levels which are well balanced, fun, challenging and interesting.
Levels which seem to be the same time and time again,
either in look or layout will get very repetitive. It’s good to make sure that
the objectives, goals, look, feel and game play of each level is as distinct as
possible, especially if you’re reusing art or levels. You need to make sure
that players want to keep going from one level to the next.
It’s
also possible for levels to become very repetitive. Mario Sunshine drives me crazy. You have to play through each level
in many different ways, but often do the same kinds of tasks from one level to
the next. While the levels have some interest in them, they also make you do
the same thing over and over and over and over and …. Well you get the point.
This is also compounded by the fact that you might be climbing up the side of a
hill for a long time, and miss the final jump and then have to repeat it again
and again. Some people like repetitive play, but it drives some of us mad.
Levels
can also be repetitive if you’re fighting the same thing, in the same way for
too long. You need to remember that it’s important to pace your levels and this
will help ensure that they don’t get too repetitive. We fell into this trap in Munch’s Oddysee, where at times you
needed to move something with a crane, which was fine, but you actually had to
move 25 of them according to the story. It was very boring after five, but
there was nothing we could do about it in the end. This kind of repetition has
to be carefully controlled and monitored.
Another
problem that many games have is large amounts of empty or dead space. This is
really common in games where you have to travel through the same area many
times. Some games simply leave the areas open, while other games respawn the same enemies at the same locations. This can be
seen in games like Soul Reaver 2. While this is often unavoidable, you need to
keep in mind that simply running through a level for 5 minutes isn’t fun,
especially for the third or forth time.
Try
to minimize dead space in your games. While a game can’t be one solid battle,
you need to be aware of large areas which have nothing to do in them, or force
you to do the same thing over and over again. Try to avoid having the same
enemies just reappear over and over in the game, in the same places, in order
to make it less obvious that the player is going through the same area. If the
player has to go through the same area twice, think about adding a script that
triggers a whole new set of people or enemies once the player accomplishes a
goal, but before he has to go back through the area. This will keep the players
at least thinking that the world responded to them. If you killed a bunch of
people in one area (and if you have enough memory to keep their dead bodies
around), maybe you could do something like have another group of soldiers
there, along with some doctors, running through some kind of a script which
makes it look like they are trying to figure out what happened. This might even
become a gameplay device, since you may want to then
hide bodies or make sure everyone is dead before you move on. Otherwise you may
have to deal with the consequences. You could even do something like, if they
find someone still
alive, that person tells the new group that you attacked them and
they’re looking for you, but if you kill everyone, they may be looking for
someone and questioning people, but they don’t know it’s you and you can
actually get past them. If you hid the bodies of the dead, the group may just
be moving through the area and not stopped, and you might be able to bypass
them, sneak around them, hide from them or just kill them. You can see how a
simple mechanic like being able to hide the bodies in an area could add a ton
of different game play possibilities. The same action of say dragging a body, could also be used to drag other things or move things
in the game. Or, maybe you could pickup branches and cover the bodies, but
picking up those same branches and placing them over a hole could create a
pitfall trap.
Having
dead space in a game, and having new areas to explore are different things.
First of all, you must know your audience and what they’re going to want.
Someone who’s expecting a fast paced action game may not want to do lots of
exploring. Exploration can be fun however as long as it is kept interesting.
This can be done by providing new things to look at, enemies to fight and other
stuff to do while you explore. Most importantly, exploration should be rewarded
with some kind of rewards, whether it is items or something else. In strategy
games, if you do lots of exploration, you can often find random stuff worth money,
special items, you discover where the enemy is hiding, where piles of resources
are at, and gain other benefits. This keeps exploration interesting on each
level. Exploration early in strategy games also fills dead time when you’re
waiting to build your base and army, so it fills a secondary need.
How
you move about in a level is related to the dead space issue above.
Games
like Dynasty Warriors, have a lot of problems with
running around a level for a long time looking for the person you need to
fight. You also have the problem when you run low on health, since you’re given
health randomly by enemies after you defeat them, or you can find it in
sparsely placed boxes. You can’t afford to fight people when your health is
low, so you have to go look for it, which can take forever, especially when
you’re avoiding a fight. Other games can suffer from a similar problem when you
have to obtain health or healing from a far away place. Diablo 2 and Neverwinter Nights solved this by giving you a
teleport spell, which allows you to return and heal whenever. While this may
seem too easy, there is a cost, but it keeps the game moving forward with only
a short pause to get healed before returning to battle or exploration.
Something else which gets very old in games is moving
through a level and just having a bunch of static enemies that come to life and
charge at you when you get close. Not only does a bunch of static or simply
animated enemies visually look bad, but fighting them over and over gets boring
very quickly. Likewise a bunch of enemies that just keep teleporting in or
appearing randomly also get old. Fighter Force did a great job at having enemies enter the level in
a wide variety of ways. Why can’t you have the player enter an
are and then have enemies come at you from different places. If you’re
walking down a street, and you want the player to fight a dozen enemies, they
all don’t need to be standing there looking stupid. Maybe you could have a
lookout that blows an alarm if you get close, or a few guys standing around a
fire warming their hands who jump you first. Other enemies could then stream
out of a door, a van could pull up and guys could jump out of it, guys could
rappel down from a helicopter, jump in off the roofs, come up through a sewer
cover, break through a window and jump out of it, or just open a door and come
out it at you. Constantly introducing new enemies in new ways will keep the
player guessing and on his toes.
Also,
think about ways to use these entry or exit points as
an enemy spawner. This point may continue to generate
X# of enemies unless it is stopped or blocked. Maybe you could blowup a car and
have it land against the wall, blocking the door, or use your flamethrower on a
manhole cover to melt it shut. Players can then even more be able to use the
skills they have and the few rules they know to do different things which
actually affect the game, how it plays and how difficult it is. However you
choose to do it, a little attention to detail in this area will keep the
players much more interested in the game as it goes on.
Having
a wide enough variety of enemies is a problem that most games face. It is still
very difficult to create good enemy characters. There are a lot of gating
factors in design, art and programming which stop you from creating a lot of
enemies in the game. Some games create variations on enemies, which are
alright, but the game needs to have more than the same enemy with a different
weapon and a different outfit on. A lack of enemy variety may also occur in
games which are trying to be historically accurate and only want the correct
types of enemies in the game.
So,
if you only have a limited number of enemies, there are some important things
you should do. First of all, you need to make sure that a variety of enemies
are delivered throughout the game, with new enemies still appearing at least
half way to three quarters of the way through the game.
So
what makes for good enemies that keep the game from getting repetitive? First
of all you should try and make the enemies fit the game. They need to look, act
and be as authentic to the game as possible – this goes for non-historical
games as well. If the enemies are interesting, the player will enjoy fighting
against them. This means that we have to learn how to script out custom AI
behaviors and events better. Look at the kind of response games like Halflife got. Halflife didn’t
have the best enemies ever, but it sure had some of the most interesting.
Making
smart and somewhat realistic characters can really help propel your game
forward. Players are always moaning about how dumb the enemies are in the game.
We need to make better, smarter enemies that seem more realistic. This doesn’t
mean that the enemies need to be more challenging or tougher to kill, be able
to aim better or any of that. They need to be more interesting.
A
good example of where AI hasn’t gone very far in action games, but is starting
to show up in Strategy games, is in the use of formations and tactics. If you
look at games like Dynasty Warriors 3
or the Mark of Kri,
where you fight a lot of enemies at once, you will see that there is little to
no coordination between enemies. In fact, you can often kill a guy right next
to another guy and the other enemy will just keep standing there looking
clueless. This kind of “stupid” behavior has the biggest criticism with gamers
and needs to end.
Any
kind of soldier you are fighting against would be trained to fight and know how
to work as a team. Soldiers don’t often work alone. When was the last time you
saw a group of 5 enemies really working well together, covering each other,
fighting side by side? Metal Gear Solid 2
has some good use of this, but it is especially more noticeable in games where melee
weapons, not guns are used primarily. Granted it can be tough for you as an
individual soldier to fight against a well armed and disciplined group of guys
working together, but there are ways to compensate. You don’t want to introduce
impossible situations for the player early on, but this is where the rules come
in. Instead of relying on AI which is intelligent, a lot of this behavior can
be scripted and gives the player the illusion of intelligence.
New
AI and enemy tactics and behaviors can be slowly introduced throughout the
game. The player may start off fighting only a few guys at a time and then
slowly fight more and more enemies which gradually get smarter and smarter.
This doesn’t mean harder, but just think more interesting. For instance, groups
of soldiers may get together and fight in a group, with shield guys in the
front, blocking you, a commander with a sword and an archer in the rear. The
player then has the choice of how to fight the group, which might include
trying to take out the archer first, the commander or dispatching of the shield
guys. If the player takes out the archer however, the shield guys may pull
their swords and attack, but if he can take out the commander the others will
flee. These simple choices would not only look cool, but be interesting to the
player. They can also subtly change with the types of units which are in the
mix, but not require that the player really learn any new rules, but adapt the
ones he already knows how to use.
Later
on in the game, the AI could also change to make things more interesting for
the player. Imagine that same group of 4 guys, but now you take out the
commander and two of the guys go to flee, but the third guy hollers and keeps
them together, taking command, making you realize that he was a corporal, not a
private and therefore his rules changed slightly and made the player adapt.
This would keep the player on his toes and more observant on whom he was
fighting.
Some of the changes in AI could also come from
adaptive learning or pattern recognition. While you don’t need an intelligent
AI to do this, it is very simple to track how many kills you have made and
maybe categorize them. Imagine that you’ve killed 50 Orcs
in the game, so now you’re known as the great Orc killer. If you encounter a
few Orc’s in the game, they will flee from your
might. This is fun for awhile, until you realize that now they’re coming back
and bringing help in the way of reinforcements. The player will feel that the
AI is adapting or learning, when it’s really just changing. This can also be
used by enemies when you fight them, which can consist of the ability to detect
when you’re using the same move over and over, and then provide a block to the
move and force the player to adapt and change his patterns. Set play patterns
are what makes the game boring. Rune: the
Viking Warlord was criticized for being winnable with the use of only one
button and a single move used throughout the entire game. People need reasons
to do new things, and it starts with the use of better AI.
The
topic of Psychology and game design could fill a book or two of it’s own. It is as much of a science as it is a concept or a
theory for us. It’s important to understand that it is a science. Most of these
topics are things we may understand the concept of, but never really realize
that there are actually formulas and scientific data to backup our concepts. So
when I talk about some of this, I may not go into the exact formulas and
scientific research behind it, but it’s good to know that if you’re interested
in learning more about the subject, there are a lot of great books out there to
draw from. These issues are some of the fundamental issues revolving around why
games become repetitive. You may also want to read an article or a book on
reinforcement schedules and how they work, to have a better idea of how to plan
out when new items should come into a game in order to keep players playing the
game.
One
of the best ways we remember details is through repetition. It is a great
technique which we can use, to try and reinforce things for people, but we also
have to be careful in that we don’t force so much repetition that it bothers
the player.
Repetition
of important story elements can be an important part of the player understanding
they need to do that or that it is important. Possibly telling the story from
different sides, or in different ways may reinforce
the story through the repetition of hearing it, but by telling it different
ways, the player may not tire of it as soon.
In
terms of familiarizing a player with a control system, or having them learn
certain procedures within a game, repetition is both a way to help the player
remember what to do, and provides the opportunity to practice these actions. As
long as the repetition is skillfully paced and well integrated into the overall
flow of the game, it should do its job without becoming tedious.
Learning
is an adaptive process in which the tendency to perform a particular behavior
is changed by experience. As conditions change, we learn new behaviors and
eliminate old ones. It’s important to understand how people learn. This is
important form the moment the player starts playing the game, until they reach
a point where there is nothing new in the game (which is hopefully at the very
end of the game). You never want to force a player to learn too many things at
the same time, and you don’t want to introduce new elements into the game,
without the player being able to first learn how to use it.
Some
designers have the philosophy that they can put the instructions for the game
into the manual, and that if the player can’t figure it out, they can always
read the manual. While this may be an acceptable solution for simulators and
certain hard core games, this solution is very questionable on most games these
days. Players do much better if they are slowly taught a skill in the game, and
slowly shown how to use it, instead of just given something and expected to
figure it out. This doesn’t mean that you can’t have puzzles the player needs
to figure out, but the core of the
Many
events cause us to react automatically. For example, a sudden, unexpected noise
causes an orientating response: We become alert and turn our heads or move the
camera towards the source of the sound. However, if the noise occurs
repeatedly, we gradually cease to respond to it; we eventually ignore it. So an
Orientating response is a reaction to some kind of stimulus which causes us to
direct ourselves towards the source of the stimulus. Habituation, learning not
to respond to an important event that occurs repeatedly, is the simplest form
of learning.
This
can be important for games, since people will learn to ignore things which are
repeatedly show, told or heard. It’s the boy who cried wolf syndrome. If the
player is constantly told something, which turns out to be false, and then at
one point it is true, the player won’t know whether to believe it or not. This
can also be true, when the same audio cue is given for the same or similar
events. For instance, in Age of Empires, whenever you’re under attack, it plays
a short sound. If you gather up a group of guys and attack the enemy, this
sound will play. In a large and long battle, the noise can play several times,
or very often. As a player, you begin to tune out the noise, knowing that you
are under attack, because you’re busy kicking the computer or other persons butt(you hope). The problem occurs that if you get attacked
somewhere else, which you’re not watching, and you hear the audio alert, you’re likely to not realize you’re under attack
somewhere else, and are likely to suffer a major blow because of the habit of
tuning out the alert sound while you’re attacking. I’ve had this happen to me
many times, and have returned to find my base half destroyed before I realized
it was under attack.
Habituation
is a simple but useful form of learning which permits us to remain relatively
free from distraction by petty events and allows us to focus on more important
things. As game designers, we just need to be aware of Habituation, and try and
account for it. In the case of Age of Empires, the solution might be to provide
and additional audio alert which is specific to your base being attacked.
Whatever the solution is, it’s important to look out for anywhere in the game
we use the same thing over and over again. As Albert Einstein said “Everything
should be made as simple as possible, but not simpler.” This is true for the
way we use things in the game, because sometimes by simplifying say an audio
cue too much, it may lead to the player developing a bad habit and ignoring the
alert all together.
Unlike
habituation, classical conditioning involves learning about the conditions that
predict that a significant event will occur. We acquire a lot of out learning
this way. Games are about pattern recognition, which is the basis for classical
conditioning. For instance, if you’re walking through a level in the game, and
an alarm goes off and then 10 seconds later, the ship you are on begins to
explode which kills you, the player will die and then realize when they play
the game the second time, and the alarm goes off that they better run and get
to an escape pod. This particular even is classical conditioning, but it also
only works if you have shown the player where an escape pod is and made sure
that one is close, so that the solution to the puzzle is obtainable by the
player.
Flinching is another form of
classical conditioning. Many boys like myself always
played the flinch game growing up with their friends. Most of us have the
natural reaction to flinch when someone tries to hit us in the face or on our
body. We flinch because we know it will hurt, because it might scare us, or
because we see something coming for us which we naturally wish to avoid. In
playing the flinch game, you try to get your friend to flinch, by pretending to
hit him, and in order to see what his reaction is. The reaction to your friend
trying to hit you is classical conditioning. However, if your friend pretends
over and over again to try and hit you, you can also enter a habituation
learning phase, where you stop flinching, since you realize that it’s all just
a ploy. So some forms of conditioning can be combined.
Classical
conditioning accomplishes two main functions. First, the ability to learn to
recognize stimuli that predict the occurrence of an important event slows the
learner to make the appropriate response faster and perhaps more effectively.
For example, hearing the sound of an incoming mortar round in the game will
cause you to dive or run for cover more quickly the more often it occurs. The
second function of classical conditioning is even more significant. Through
classical conditioning, stimuli that were previously unimportant acquire some
of the properties of the important stimuli with which they have been associated
and thus become able to modify behavior. In a sense, the stimulus takes on a
symbolic value. For example, we respond differently to the sight of a stack of
paper on the desk than to a stack of paper on the desk. The reason for this
special reaction to money, is that money has in the
past, been associated with good things and stuff you’d like to buy.
Classical
conditioning can mean many things for game designers. It means that through
proper prompting, players can be taught what types of events are going to occur
and how they can react more quickly to them. Just like the alarm when you’re
attacked in Age of Empires. If the alarm sounds and you’re not ready for a
fight, a sense of urgency, adrenaline and fright can enter the player. The
reaction here can be different than if the player just suddenly finds himself
under attack. Classic conditioning also acts to foreshadow events, which can
also subconsciously make players become frightened, excited or in some other
mood. Music and sound is another effective way to elicit a classical learning
response, for if you begin to play scary music, just before something scary
actually happens, it’s been shown to be even scarier than just having someone
jump out at you. So therefore, understanding what classical conditioning is,
can be a good first step to understanding how and why people react to events in
different ways.
Habituation
and classical conditioning teach us about stimuli in the environment. We learn
to ignore unimportant stimuli, and we learn about those that predict the
occurrence of important ones. They dealt with the way different things would
react together in order to teach you something. In contrast, operant
conditioning teaches us the relationship between environmental stimuli and our
own behavior. The term operant, refers to the fact
that a person learns something through responding, or operating on the
environment. In other words, when a particular action has good consequences the
action will tend to be repeated, and if an action had bad consequences, the
action will tend to not be repeated.
Understanding
that these learning conditions are also rules for you to follow is also
important. In other words, if something helps a player one time or especially
multiple times, if the player does it, then the same action shouldn’t hurt the
player later on unless some warning has been given. If you condition the player
to do do something and that they’ll get rewarded for
it, don’t hurt or punish the player if they later perform the same action
without first teaching them the new rule.
Operant conditioning occurs
all throughout games. Later on, you need to learn how to develop the operant
conditioning method into a reinforcement schedule, so that you’ll know when
it’s appropriate to give out good and bad rewards for things.
As
players practice skills and problems in a game they come to learn the sequence
of moves required to solve the problem or portions of the problem. Tactical
learning refers to the improvement that comes about because people learn
familiar subsequences of problem solving steps that appear in multiple
problems. Players are able to solve problems more easily, the more often they
encounter it, until they are able to overcome it. Many RTS games take this
approach. The design of many RTS games is such that the player, unless very
skilled, must play through the level a few times in order to understand the problem,
and then must play the level a few more times in order to figure out how to
solve the level. While this approach isn’t ideal, it takes into account the
tactical learning that must go on during a game, and understands that the
average player must go through the process.
As
the game goes on, players begin to learn the patterns of the game. In an RTS,
this may be that they’re beginning to understand how the AI is working, and are
able to compensate more quickly for it’s tactics. This
should allow players to begin to recognize similar problems later in the game,
like how to deal with their base being under attack, and allow them to deal
more quickly or more efficiently with the problem.
Strategic
learning refers to the improvement that comes about because people learn the
optimal way to organize their problem solving for a particular problem. Whereas
Tactical learning is more of a linear path to problem solving, Strategic
learning involves the player learning the correct approach to solving the
problem. Strategic learning requires the player to organize themselves in a
different manner and is more likely to take a non-linear or less
straightforward approach. An example of strategic learning could be in an RTS,
where a player learns to sacrifice a group of troops in order to draw the enemy
into a trap or away from their base, so that some type of sacrifice must be
made in order for progress to continue. This type of maneuver is initially
counter intuitive, since the loss of units is usually associated with doing
poorly and loosing for beginning players.
It
has been shown that novice players tend to work backwards to solve a problem.
They figure out what the problem is they need to solve, like they need to
destroy an enemy building, and then they work backwards from that, analyzing
what they need to destroy the building, then what they need to destroy the
buildings defenses, then what they need to get to the building and so on until
they figure out what to start with. In contrast, experts tend to think forward,
understanding what they have to work with and what it’s capable of doing and
figuring out a solution looking forward at the problem.
As you can see, there are a lot of issues revolving around
how to make games fun. You’ve seen that repetitive games come in many shapes
and sizes. If there is one thing that makes a game repetitive, it’s our own
lack of ability to see that a simple single idea, no matter how cool it is,
doesn’t always make for an exciting game in the long run.
I
haven’t talked about all of the issues on how to make games less repetitive,
but hopefully I’ve opened your eyes up to the seriousness of the problem we’re
facing as an industry and an art. I can’t stress to you enough how often I see
this problem occurring still in games being made today, and that if we expect
to make the next hit title, we’re going to need to stop looking at other games
as what the acceptable bar is, but look back into ourselves to first recognize
that there is a problem and then come up with some easy solutions on how to
solve them.