Some Thoughts on Computer Game Design
=========================================

I've been thinking about what it is about my favorite games that
I like...I've thought about Game Design on and off over the years,
and have begun to pull some things together for myself. 

Here's a start of a breakdown...feel free to add upon it and respond to
individual ideas...I'd like to quote different reactions to these ideas in a
non-profit hypermedia project I'm working on at school, and thought the life
of these groups would have much to add. 

Dave Seah
(dis6218@ultb.isc.rit.edu, pccdaves@aol.com)

--- cut here ---

The Player

Why do we play games? We like challenges, and we derive 
pleasure when we overcome them. Games provide a safe context 
in which we can express our fantasies and dreams. We achieve 
a sense of stature and accomplishment by engaging in 
stimulating gameplay. Computer games are no different in the 
way they provide challenge and stimulation. As game 
designers, though, we are responsible for designing every 
part of every interaction within the self-contained computer 
world. The player has to be able to find and overcome the 
challenges we design into a game without feeling unfairly 
constrained or manipulated. 

For games in which the player plays an active role in a 
simulated social environment, emphasize the hero qualities 
of the character. By controlling the character throughout 
the game, the player establishes an alter ego. That 
joystick-controlled blip on the screen is the player, for 
all intents and purposes. The game environment becomes the 
world, and the player is the protagonist, with the potential 
to become a hero. Heros are always above average in their 
abilities, for it enables them to go forth and kick-ass in a 
definitive manner.

The protagonist must undertake some form of the hero's 
journey. This can take the form of a developmental cycle, as 
it is in role playing games (RPGs.) A new character in the 
Ultima series, for example,  starts out as a naked 
stripling, equipped with the wimpiest of weapons and armor. 
The character grows in experience by exploring the world, 
slaying monsters and engaging in quests. Eventually, the 
protagonist achieves a level of ability that is truly epic. 
The player experiences a vicarious sense of accomplishment 
through the actions and ultimate success of his/her 
character. 

In games that emphasize manual dexterity, the hero's journey 
is externalized. In an arcade game, the player usually 
controls some kind of ship. The ship itself may already be 
imbued with awesome weaponry and abilities; by surviving and 
wreaking destruction on the computer world, the player is 
shown to be a hero (or at least skilled) to his or her 
peers. 

A good example is the original version of Wing Commander. In 
this game, a chalkboard kept track of how many enemy kills 
were attributed to the player's character. It lists the 
game's non-player characters (NPCs) ranks as well, and the 
player could see how well he is doing as the story 
progresses. As the player grows more experienced, he moves 
up in the ratings. The chalkboard also makes it possible for 
players to compare their skills with their friends. "I 
fragged 105 furballs before I finished the game" was a 
common boast on the computer networks. In Wing Commander II, 
the chalkboard was moved to a less prominent area, and the 
emphasis was placed on the wooden story line, drawing the 
player into a simulated social environment.

Game Environment

As the player establishes and sets goals in the context of 
the game, the game must respond in kind by providing some 
kind of feedback. Without feedback, the player will feel 
lost and impotent, trapped inside an artificial world 
instead of participating in it. The interactive gaming 
experience must support the actions of the player, not 
thwart them.

To maintain the illusion of a functioning, living world, the 
game designer has to ensure that there is some minimum 
interaction between the player's character and the simulated 
environment. As an example, the Super Nintendo version of 
Zelda provides lots of audio-visual feedback. As the 
player's character walks through tall grass, a rustling 
sound is heard, and the character's body below the waist is 
swallowed up. When walking in a shady forest, shadows fall 
across both the ground and the character. NPCs on the screen 
react to the presence of the player in a variety of ways or 
not at all. The player is drawn into the game because the 
game environment interacts with his/her character. The high 
level of interaction between player character and game 
environment is enjoyable in itself.

The player should also receive some kind of feedback on 
his/her progress. In arcade games, this is accomplished with 
some kind of indicator (score, level, etc). The player 
ideally feels that he or she is in control at least most of 
the time. All hell may be breaking loose, but the player 
should know how much that is affecting his/her character. 
The rule of thumb is that for every action, there should be 
some kind of feedback or reaction.

The game universe should have well-established conventions 
of how things work -- its own "law of physics." If the 
laws change, there must be a rational explanation. Arbitrary 
changes, especially those that impact the player's 
character, can have a negative impact on the gaming 
experience. 

Premise and Player's Expectations

The player must know what to expect from the game world, 
journey or no journey. Usually, some rudimentary explanation 
is given to explain the game world. Ignoring purely abstract 
games like Go or Checkers, the game designer must establish 
both premise (story) and setting. The premise helps 
establish an initial motivation for why the game is the way 
it is, and the setting helps establish the tone. This can 
take rather trivial forms, but if the scope of the story 
matches the scope of the game, then I have no argument 
against it. By "scope of the game", I mean the range of 
possible associations and interactions. A complex political 
thriller as the story and setting for a straight shoot-em-up 
arcade game doesn't carry much credibility. Likewise, a game 
that is based on a popular movie is asking for trouble. 
People will carry their memories of the movie over to their 
expectations of the game, and are setting themselves up for 
disapointment. 

For the designer's expectations of the game, it is important 
to set the challenge clearly. You know what kind of 
experience you want to deliver, so make that clear right up 
front when establishing the story and setting. The premise 
(or premises) of the game will vary wildly. If a game is 
intended to be a visually confusing, surreal exploration of 
the inner mind, then let it be said up front. If the game is 
intended to be an accurate, uncompromising World War II 
simulation of the Battle of the Bulge, then different 
expectations are set. The designer must set these guidelines 
explicitly, and adhere to them. The game experience should 
fully support the premise and the player's expectations.

Consider two space gaming genres: the side-scrolling 
"shoot-em-up" (or shooter) and the space combat simulator. 
In a classic shooter like Defender, the premise is pretty 
simple: Destroy all the waves of attacking aliens and 
protect your people. The gameplay is simple to understand, 
though the game itself requires superb eye-hand coordination 
of a level that is seldom seen today. For the gamer, the 
premise is secondary to the nerve-racking adrenaline rush of 
playing which is the main reason for playing. As a more 
recent example, the space combat simulator X-Wing is based 
on the Star Wars universe. Since it is based on the popular 
movie, players expect a lot from the game. The premise is 
that you are a rookie joining the Rebel Alliance. Although 
you can jump right into the missions and just start blasting 
TIE fighters, you are encouraged to go through the training 
process. Animated sequences and accurate portrayal of the 
Star Wars universe build the framework in which you play the 
game, and an included novella sets the tone for your tour of 
duty: You are a new recruit, and the Alliance is being 
ruthlessly hunted down by the Imperial fleet. You are not 
just flying any old spacecraft...you are flying an X-wing, 
escorting a crucial shipment of grain through dangerous 
Imperial patrols! There is a sense of an expanded game 
universe that exists outside the confines of the game, 
carefully reinforced with information from the mission 
briefing screens, but the story isn't stated overtly. You 
are not Luke Skywalker, but you know that he's out there 
somewhere fighting the same cause. 

Balancing of Goals

From the initial motivation, the player can form (or be 
informed of) several long-term goals. The default goal is to 
win the game. Other long term goals may be to "find the 
Golden Sword" or "Return Peace to the Land." Players will 
formulate a number of short-term goals along the way. "To 
find the Golden Sword, I have to talk to everyone," or "To 
return peace to the land, I need to collect the Peace 
Crystals." Shorter term goals might be, "To get to the 
next level, I have to blow up all these ships," or "I have 
to find the Power Booster if I want to survive!" Some 
short-term goals are standing goals. These goals are always 
on the player's mind, as they continually contribute to the 
well-being of the character. For example, "I better not get 
hit by enemy blaster fire, or I'll lose a ship." The game 
designer manages the difficulty of these goals, and the 
player juggles them.

For flexibility, I like to allow flexible goal completion. 
In other words, I like to allow some leeway in the order 
that goals are brought to completion. There is no single 
order that guarantees success. Putting together a structure 
that allows this is not trivial, and it depends on the kind 
of game. To take an old example, consider the classic Apple 
II game, "Castle Wolfenstein." The object (long term goal) 
of the game is to escape from a Nazi castle during WWII, 
initially armed with nothing but a pistol. Finding the "War 
Plans" was a secondary long-term objective. Since the game 
begins with the player trapped in a guarded room, clad in 
bright purple prisoner duds, the short term goal is to get 
out of the room without getting captured or shot. As the 
player moves from room to room, the standing goals are 
finding uniforms, keeping stocked with ammunition, and 
keeping the SS from getting on your tail.

Balance short term with long term goals. You want short term 
goals, punctuated by moments of sheer terror. This is part 
of the "hero building" process. The short term goals 
should be immediate and compelling. This can be something as 
simple as "blow up more bad guys." Long term goals should 
be what drives the short term goals, giving them purpose and 
meaning. Not all games need purpose and meaning. DOOM is a 
good example: the gameplay is so compelling that you have no 
time to think about any long-term goals other than, 
"survive." It is the most absorbing game of its kind on 
any platform, in my opinion.

Another aspect of goal-setting that I like is the idea of 
open play. In the case of Castle Wolfenstein, the goal of 
the game is to escape with the war plans. However, the 
player can take as long as he/she wants. In fact, the player 
can choose to use the game as an open play environment, 
effectively creating a new game. A common game was "Nazi 
hunting", in which players would just run around shooting 
everyone. Other players would try to get out of the castle 
as quickly as possible, ignoring all other aspects of the 
game. Castle Wolfenstein did not punish the player, because 
its structure was open enough to contain these kind of 
alternate games. 

Gameplay

Part of the pleasure in a game comes from interacting with 
it. I enjoy controlling characters on the screen, making 
them solve problems or overcome challenges. I also derive 
satisfaction at overcoming those challenges with skill and 
dexterity. I set a limited goal ("solve the puzzle") and 
then I reach it ("puzzle solved.") 

The player should be able to do any action that seems 
reasonable, given the premise of the game and the goals 
formed based on those premises. If there is a situation in 
which the natural response would be to run like hell, then 
the character should be able to do just that. There should 
be a counter or defense for every move. If you are shot at, 
you should be able to move out of the way. If attacked, you 
should be able to counter-attack, or at the very least be 
allowed to escape. In the event that the player has just 
made some horrible tactical mistake, this should be made 
very clear so the player knows that it was "his fault." 
There should also be the knowledge that, "there is another 
way."

The player will also want some way to gain an advantage, 
however temporary, by changing the rules slightly. For 
example, in the arcade game Assault, your highly-
manueverable tank can enter "jump zones" that propel it 
above enemy vehicles on the ground, giving the player a 
unique advantage. Players want the biggest guns and the best 
armor. Give them a way of earning that advantage, and then 
give them a reason for really needing it.

When placing constraints on the player's action, there must 
be a rationale. The premise may help establish why these 
constraints exist. If the player's magic hammer has been 
able to smash all walls, then it has to work all the time 
unless a "Special Wall" is encountered. The hammer must 
react with the wall in some way, perhaps signaled by a 
different noise or animation sequence. If something has 
worked in the past, it should work the same way in the 
future, unless some kind of new variable has entered into 
the picture. Make that variable clear, but you don't have to 
give away all its secrets.

Don't forget to emphasize the heroic, epic quality of the 
player's struggle. Explosions must be large and gratifying. 
Climactic duels with the villian must satisfy! The player 
wants to look good, so provide those opportunities that 
reward dexterity, cleverness, and experience. 

Japanese arcade games often reward pattern memorization. 
Instead of providing a way out, many of the Japanese arcade 
games rely heavily on memorization to advance. The reward is 
for experience, not necessarily cleverness. By varying a set 
of patterns, you can make the game more dynamic and still 
reward memorization. For example, in Prince of Persia, 
fighting the first few swordsmen is fairly easy if you 
recognize a pattern in their fighting style. Later swordsmen 
have slightly different appearances, and have different 
patterns. I remember encountering the fat swordsman for the 
first time and getting slaughtered when I fell in my old 
patterns. The best patterns, in my opinion, are the ones 
that don't rely on a fixed sequence of actions or moves. The 
player instead recognizes of a particular tactic or set of 
circumstances that allows him/her to act prudently, while 
retaining flexibility in the goal-setting process. An 
excellent example is X-Wing. A particular Imperial tactic 
that is used over and over is the launching of TIE bombers 
after a wave of TIE fighters. The TIE fighters are a mere 
annoyance compared to the bombers, but the Fighters must be 
dispatched lest they attack lightly-protected Rebel vessels. 
Depending on distance and mission, the X-Wing pilot has to 
react based on the experience that has been drilled into him 
or her.

Graphics and Sound 

The demand for high quality graphics and sound in today's 
oftware has advanced hundreds of times in capability 
since the early days of personal computing. We are no longer 
required to use our imagination to equate a blocky, low-
resolution screen image with a rich fantasy world. Today, we 
fully expect to see recognizable animated images, and we 
increasingly demand music and sound. Games have become less 
abstract and more complex, made possible by faster computers 
and near-photorealistic graphics.

The visual and audible aspects of a game should mesh 
smoothly with all previous aspects of the game. They should 
not be the primary focus of the game. Great graphics and 
sound should enhance an already great gaming experience, not 
usurp it. A mediocre game with great graphics and sound 
becomes a "technology showcase" with limited play value. A 
brilliantly-conceived game with poor audiovisual design 
limits its mass appeal. 

Having "good" graphics does not necessarily require 
millions of colors and high resolutions. The graphics must 
reflect the premise and setting of the game. They must be 
clear, well-rendered, and stylistically consistent. Great 
graphics will transcend the limitations of resolution and 
color. The games from LucasArts are among my favorites. 
Although they are seen on a 320x200, 256 color IBM-PC 
display, the characterization and visual design stands on 
its own. The chunky pixels and blocky graphics fade from 
consideration because the game successfully pulls you into 
its world.

Engagement of the Senses

To engage the senses is to invite the player into the game 
world. The vicarious experience of the player must be 
consistent in presentation and context. This is a balance 
between all the elements discussed above: Player and 
Environment, Story/Setting and Player Expectation, Goal-
Setting and Gameplay, Graphics and Sound.

The game designer can also exploit second- and third-order 
experience. These are the subtler details that are not 
immediately obvious, but nevertheless contribute to the 
depth of the gaming experience (the term "second-order" is 
a mathematically expression. In engineering, it distinguises 
between a simple model and a more accurate but complex one.) 

An excellent example of this is Street Fighter II. On the 
surface, this game is just another chop-sockey kung-fu 
fighting game. This is a simple categorization. When it was 
first introduced, it distinguished itself from other games 
by offering silky-smooth, state-of-the-art animation. This 
naturally attracted players. What distinguished SF2 from 
other "hit-the-button punch-punch-punch" games was its 
outstanding second-order design. It wasn't enough to just