Game Design Theory and Practice

(1)

Game Design:

Theory & Practice

Rich ard Rouse III

Illus tra tions by

Steve Ogden

Atomic Sam char ac ter designed by

Rich ard Rouse III and Steve Ogden

(2)

Game design: theory & practice / by Richard Rouse III ; illustrations by Steve Ogden. p. cm.

Includes bibliographical references and index. ISBN 1-55622-735-3 (pbk.)

1. Computer games—Programming. I. Title. QA76.76.C672 R69 2000

794.8'1526—dc21 00-053436 CIP

Plano, Texas 75074

No part of this book may be reproduced in any form or by any means without permission in writing from Wordware Publishing, Inc.

Printed in the United States of America

ISBN 1-55622-735-3 10 9 8 7 6 5 4 3 2 1 0011

Product names mentioned are used for identification purposes only and may be trademarks of their respective companies.

All inquiries for volume purchases of this book should be addressed to Wordware Publishing, Inc., at the above address. Telephone inquiries may be made by calling:

(972) 423-0090

(3)

Por tions of Chap ter 18: Inter view: Jor dan Mechner orig i nally appeared in Inside Mac Games mag a zine. Used with kind per mis sion.

Images from the 3D ver sion of Cen ti pede ® and © 2000 Atari Inter ac tive, Inc. All rights reserved. Used with kind per mis sion. Though the game is referred to as “Cen ti pede 3D” in this book in order to dif fer en ti ate it from the older game, its proper name is sim ply “Cen ti pede.” Images from Super Break out, Aster oids, Cen ti pede, Mil li pede, and Tem pest® or ™ and © 2000 Atari Inter ac tive, Inc. All rights reserved. Used with kind per mis sion.

Images from Path ways into Dark ness, Mar a thon, Mar a thon 2, Mar a thon Infin ity, and Myth: The Fallen Lords ® or ™ and © 2000 Bun gie Soft ware Prod ucts Cor po ra tion. All rights reserved. Used with kind per mis sion.

Images from Bal ance of Power, Trust and Betrayal: The Leg acy of Siboot, Bal ance of Power II: The 1990 Edi tion, Guns & But ter, Bal ance of the Planet, and the Erasmatron ® or ™ and © 2000 Chris Crawford. All rights reserved. Used with kind per mis sion.

Images from Myst ® and ©1993 Cyan, Inc. All rights reserved. Used with kind per mis sion. Images from Tomb Raider, Tomb Raider II, and Thief II ® or ™ and © 2000 Eidos Inter ac tive. All rights reserved. Used with kind per mis sion.

Images from Spellcasting 101 © 1990 Leg end Enter tain ment Com pany, Spellcasting 201 © 1991 Leg end Enter tain ment Com pany, and Superhero League of Hoboken © 1994 Leg end Enter tain -ment Com pany. All rights reserved. Used with the kind per mis sion of Infogrames, Inc.

(4)

Images from F-15 Strike Fighter, Pirates!, F-19 Stealth Fighter, Covert Action, Rail road Tycoon, Civ i li za tion, and Civ i li za tion II ® or ™ and © 2000 Microprose, Inc. All rights reserved. Used with kind per mis sion.

Images from Gaunt let®, Gaunt let II®, Xybots™, San Fran cisco Rush: The Rock Alcatraz Edi -tion™, San Fran cisco Rush: Extreme Racing®, San Fran cisco Rush 2049™, and Gaunt let Leg ends®_{© 2000 Mid way Games West, Inc. All rights reserved. Used with kind per mis sion.}

Images from Oddworld: Abe’s Oddyssee® and © 1995-2000 Oddworld Inhab it ants, Inc. All Rights Reserved. ® des ig nate trade marks of Oddworld Inhab it ants. All rights reserved. Used with kind per mis sion.

Images from The Next Tetris™ and © 1999 Elorg, sublicensed to Hasbro Inter ac tive, Inc. by The Tetris Com pany. Tetris © 1987 Elorg. Orig i nal Con cept & Design by Alexey Pajitnov. The Next Tetris™ licensed to The Tetris Com pany and sublicensed to Hasbro Inter ac tive, Inc. All rights reserved. Used with kind per mis sion.

(5)

(6)

Thanks to James Hague, Ian Parberry, and Mar ga ret Rog ers for look ing over my work and pro vid ing me with the invalu able feed back and sup port which have improved this book tre men dously.

Thanks to Chris Crawford, Ed Logg, Jor dan Mechner, Sid Meier, Steve Meretzky, and Will Wright for gra ciously sub ject ing them selves to my end less ques tion ing. To quote Mr. Wright, I’m “pretty thor ough.”

Thanks to Jim Hill, Wes Beckwith, Beth Kohler, Kellie Henderson, Mar tha McCuller, Alan McCuller, and every one at Wordware for mak ing this book become a real ity.

For their help with this book, thanks to Benson Rus sell, John Scott Lewinski, Ari Feldman, Laura J. Mixon-Gould, Jeff Buccelatto, Jayson Hill, Laura Pokrifka, Josh Moore, Lisa Sokulski, Dan Harnett, Steffan Levine, Susan Wooley, Chris Brandkamp, Kelley Gilmore, Lindsay Riehl, Pat rick Buechner, Scott Miller, Greg Rizzer, Lori Mezoff, Jenna Mitch ell, Ericka Shawcross, Maryanne Lataif, Bryce Baer, Bob Bates, James Conner, Lisa Tensfeldt, Paula Cook, Don ald Knapp, and Diana Fuentes.

Spe cial thanks to Mar ga ret Rog ers, June Oshiro and Matt Bockol, Ben Young, Alain and Annalisa Roy, Gail Jabbour, Amy Schiller, Katie Young & Eric

Pidkameny, Rafael Brown, Eloise Pasachoff, Mark Bull ock and Jane Miller, Dave Rouse, Linda, Bob and Grayson Starner, Jamie Rouse, Alan Patmore and every one at Sur real, the Leaping Liz ard crew, Brian Rice, Lee Waggoner, Pat Alphonso, Clay Heaton, Alex Dunne, Gordon Cameron, Tuncer Deniz, Bart Farkas, Peter Tamte, Nate Birkholtz, Al Schil ling, Cindy Swanson and every one at MacSoft, Doug Zartman, Alex Seropian, Jason Jones, Jim McNally, Jeff O’Connor, Ira Harmon, Gordon Marsh, Chuck Schuldiner, Glenn Fabry, and Derek Riggs.

(7)

over seven years and has played a lead design role in the devel op ment of games for the PC, Macintosh, Sega Dreamcast, Sony PlayStation, and PlayStation 2. His credits include Cen ti pede 3D, Odys sey: The Leg end of Nem e sis, and Dam age Incor

-po rated. At Sur real he cur rently spends all his wak ing hours work ing on a secret

PlayStation 2 action/adven ture pro ject, while also con trib ut ing where he can to

Drakan for PlayStation 2. Rouse has writ ten about game design for pub li ca tions

includ ing Game Devel oper, SIGGRAPH Com puter Graph ics, Gamasutra, and

Inside Mac Games.

Your feed back to this book, includ ing cor rec tions, com ments, or merely friendly ramblings, is encour aged. Please mail them to the author at

[email protected]. You will also find the web page for this book, which will be used to track cor rec tions, updates, and other items of inter est, at www.paranoidproductions.com. See you there.

Steve Ogden has been an art ist, illus tra tor, and car toon ist for almost 20 years, and mirac u lously, his right hand shows no sign of drop ping off. Among his pro jects in the dig i tal domain, he has worked on Bally’s Game Magic casino game as well as

Cen ti pede 3D, and has just fin ished a stint as Art Direc tor and Pro duc tion Lead on

Cyan’s realMYST (while fin ish ing the illus tra tions to this book dur ing the few hours he was sup posed to be sleep ing). He is now gear ing up for work on Cyan’s next game, if they can catch him and chain him to his desk again. To see more of his work, both of the 2D and 3D vari ety, stop by his web site: www.lunaenter-tainment.com. You can reach him at ogden@ lunaenterwww.lunaenter-tainment.com. He is now going to crawl to a beach very far away and sleep for a while.

(8)

(9)

In tro duc tion . . . xviii

Chap ter 1 What Players Want . . . 1

Why Do Players Play?. . . 2

Players Want a Chal lenge . . . 2

Players Want to So cial ize . . . 3

Players Want a Dy namic Sol i taire Ex pe ri ence. . . 5

Players Want Bragging Rights . . . 5

Players Want an Emo tional Ex pe ri ence . . . 6

Players Want to Fan ta size . . . 7

What Do Players Ex pect? . . . 8

Players Ex pect a Con sis tent World. . . 8

Players Ex pect to Un der stand the Game-World’s Bounds . . . 9

Players Ex pect Rea son able So lu tions to Work . . . 10

Players Ex pect Di rec tion . . . 10

Players Ex pect to Ac com plish a Task In cre mentally. . . 12

Players Ex pect to Be Im mersed. . . 12

Players Ex pect to Fail . . . 14

Players Ex pect a Fair Chance . . . 14

Players Ex pect to Not Need to Re peat Them selves . . . 15

Players Ex pect to Not Get Hope lessly Stuck . . . 16

Players Ex pect to Do, Not to Watch . . . 17

Players Do Not Know What They Want, But They Know It When They See It . 18 A Never-Ending List . . . 19

Chap ter 2 In ter view: Sid Meier . . . 20

Chap ter 3 Brain storm ing a Game Idea: Gameplay, Technology, and Story . . . 42

Starting Points . . . 43

Starting with Gameplay . . . 44

Starting with Tech nol ogy . . . 45

Starting with Story . . . 47

Working with Lim i ta tions . . . 50

Od ys sey: The Leg end of Nem e sis . . . 50

(10)

Dam age In cor po rated . . . 51

Cen ti pede 3D . . . 53

Em brace Your Lim i ta tions . . . 54

Es tab lished Tech nol ogy . . . 55

The Case of the Many Mush rooms . . . 55

The Time Al lotted . . . 57

If You Choose Not to De cide, You Still Have Made a Choice . . . 58

Chap ter 4 Game Anal y sis: Cen ti pede . . . 59

Clas sic Ar cade Game Traits . . . 62

In put . . . 65

In ter con nec ted ness . . . 66

Es ca lating Ten sion . . . 68

One Per son, One Game . . . 71

Chap ter 5 Fo cus. . . 73

Es tab lishing Fo cus . . . 74

An Ex am ple: Snow Car nage Derby . . . 77

The Func tion of the Fo cus . . . 79

Main taining Fo cus . . . 82

Fleshing Out the Fo cus . . . 83

Changing Fo cus . . . 84

Sub-Focuses . . . 88

Using Fo cus . . . 91

Chap ter 6 In ter view: Ed Logg . . . 93

Chap ter 7 The El e ments of Gameplay . . . 121

Unique So lu tions . . . 122

An tic i pa tory ver sus Com plex Sys tems . . . 122

Emer gence . . . 123

Non-Linearity . . . 125

Types of Non-Linearity . . . 125

Im ple men ta tion . . . 127

The Pur pose of Non-Linearity . . . 129

Modeling Re al ity . . . 130

Teaching the Player . . . 132

Re wards . . . 134

In put/Out put. . . 136

Con trols and In put . . . 136

Out put and Game-World Feed back . . . 141

Ba sic El e ments . . . 145

Chap ter 8 Game Anal y sis: Tetris . . . 146

Puz zle Game or Ac tion Game? . . . 147

Tetris as a Clas sic Ar cade Game . . . 149

(11)

The Tech nol ogy . . . 151

Ar ti fi cial In tel li gence . . . 153

Es ca lating Ten sion . . . 154

Sim plic ity and Sym me try . . . 155

Ten Years On, Who Would Pub lish Tetris? . . . 157

Chap ter 9 Ar ti fi cial In tel li gence. . . 158

Goals of Game AI. . . 160

Chal lenge the Player . . . 161

Not Do Dumb Things . . . 163

Be Un pre dict able . . . 164

As sist Sto ry tell ing . . . 167

Cre ate a Liv ing World . . . 169

The Sloped Playing Field . . . 170

How Real is Too Real? . . . 171

AI Agents and Their En vi ron ment . . . 172

How Good is Good Enough? . . . 175

Scripting. . . 177

Ar ti fi cial Stu pid ity . . . 178

Chap ter 10 In ter view: Steve Meretzky . . . 179

Chap ter 11 Sto ry tell ing . . . 214

De signer’s Story Ver sus Player’s Story . . . 216

Places for Sto ry tell ing . . . 218

Out-of-Game . . . 219

In-Game . . . 224

Ex ter nal Ma te rials . . . 227

Frus trated Lin ear Writers . . . 228

Game Stories . . . 230

Non-Linearity . . . 232

Working with the Gameplay . . . 233

The Dream . . . 234

Chap ter 12 Game Anal y sis: Loom . . . 236

Fo cused Game Me chanics . . . 238

User In ter face . . . 239

The Drafts Sys tem . . . 241

Dif fi culty . . . 243

Story . . . 244

Loom as an Ad ven ture Game . . . 245

Chap ter 13 Getting the Gameplay Working . . . 248

The Or ganic Pro cess . . . 251

Too Much Too Soon . . . 251

Keep It Sim ple . . . 253 xi

(12)

Build ing the Game . . . 254

Core Tech nol ogy . . . 254

In cre men tal Steps . . . 255

A Fully Func tional Area . . . 256

Going Through Changes . . . 257

Pro gramming . . . 259

When is It Fun? . . . 261

Chap ter 14 In ter view: Chris Crawford . . . 263

Chap ter 15 Game De vel op ment Doc u men ta tion . . . 291

Doc u ment Your Game . . . 293

Con cept Doc u ment or Pitch Doc u ment or Pro posal . . . 293

De sign Doc u ment . . . 294

Flowcharts . . . 295

Story Bi ble . . . 295

Script . . . 297

Art Bi ble . . . 300

Storyboards . . . 301

Tech ni cal De sign Doc u ment. . . 301

Sched ules and Busi ness/Mar keting Doc u ments . . . 302

No Stan dard Doc u men ta tion . . . 302

The Ben e fits of Doc u men ta tion . . . 303

Chap ter 16 Game Anal y sis: Myth: The Fallen Lords . . . 304

Use of Tech nol ogy . . . 305

Game Fo cus . . . 308

Sto ry tell ing . . . 310

Hard-Core Gaming . . . 311

Multi-Player . . . 313

Over all . . . 314

Chap ter 17 The De sign Doc u ment . . . 316

The Writ ing Style . . . 318

The Sec tions. . . 321

Ta ble of Con tents . . . 321

In tro duc tion/Over view or Ex ec u tive Sum mary . . . 322

Game Me chanics. . . 323

Ar ti fi cial In tel li gence . . . 329

Game El e ments: Char ac ters, Items, and Ob jects/Mech a nisms . . . 331

Story Over view . . . 334

Game Pro gres sion . . . 335

Sys tem Menus . . . 337

One Man’s Opin ion . . . 337

In aus pi cious De sign Doc u ments . . . 338 xii

(13)

The Wa fer-Thin or El lip sis Spe cial Doc u ment . . . 338

The Back-Story Tome . . . 339

The Over kill Doc u ment . . . 340

The Pie-in-the-Sky Doc u ment . . . 341

The Fos sil ized Doc u ment . . . 342

A Mat ter of Weight . . . 343

Getting It Read . . . 343

Doc u men ta tion is Only the Be gin ning . . . 344

Chap ter 18 In ter view: Jor dan Mechner. . . 346

Chap ter 19 De signing De sign Tools. . . 378

De sired Func tion al ity . . . 380

Vi su al izing the Level . . . 380

The Big Pic ture . . . 382

Jumping into the Game . . . 384

Editing the World . . . 386

Scripting Lan guages and Ob ject Be hav iors . . . 388

Us Ver sus Them . . . 390

The Best of In ten tions . . . 392

A Game Ed i tor for All Sea sons. . . 394

Chap ter 20 Game Anal y sis: The Sims . . . 395

Ab di cating Au thor ship . . . 396

Fa mil iar Sub ject Mat ter. . . 398

Safe Ex per i men ta tion . . . 399

Depth and Fo cus . . . 400

In ter face . . . 401

Con trolled Ver sus Au ton o mous Be hav ior . . . 403

A Les son to Be Learned . . . 404

Chap ter 21 Level De sign . . . 406

Levels in Dif fer ent Games . . . 408

Level Sep a ra tion . . . 409

Level Or der . . . 410

The Com po nents of a Level . . . 412

Ac tion . . . 413

Ex plo ra tion. . . 413

Puz zle Solving . . . 415

Aes thet ics . . . 416

Bal ancing It All . . . 418

Level Flow . . . 418

El e ments of Good Levels . . . 421

Player Can not Get Stuck. . . 421 xiii

(14)

Sub-Goals . . . 422

Land marks . . . 423

Crit i cal Path . . . 423

Limited Back tracking . . . 423

Suc cess the First Time . . . 424

Nav i ga ble Areas Clearly Marked . . . 424

Choices . . . 424

A Per sonal List. . . 425

The Pro cess . . . 425

step 1. Pre lim i nary . . . 425

step 2. Con cep tual and Sketched Out line . . . 427

step 3. Base Ar chi tec ture . . . 427

step 4. Re fine Ar chi tec ture Un til It is Fun . . . 428

step 5. Base Gameplay. . . 429

step 6. Re fine Gameplay Un til It is Fun. . . 430

step 7. Re fine Aes thet ics . . . 430

step 8. Playtesting . . . 431

Pro cess Vari a tions . . . 431

Who Does Level De sign?. . . 432

Col lab o ra tion . . . 433

Chap ter 22 In ter view: Will Wright . . . 434

Chap ter 23 Playtesting . . . 472

Find ing the Right Testers . . . 473

Who Should Test. . . 474

Who Should Not Test . . . 477

When to Test . . . 479

How to Test . . . 481

Guided and Un guided Testing . . . 482

Bal ancing . . . 483

Your Game is Too Hard . . . 485

The Ar tis tic Vi sion . . . 487

Con clu sion . . . 489

Art . . . 489

The Me dium. . . 490

The Mo tive . . . 491

Ap pen dix Sam ple De sign Doc u ment: Atomic Sam . . . 493

Atomic Sam: Fo cus . . . 495

Atomic Sam . . . 496

De sign Doc u ment . . . 496

Ta ble of Con tents . . . 496

I. Over view . . . 499 xiv

(15)

II. Game Me chanics. . . 500

Over view . . . 500

Cam era . . . 501

In-Game GUI . . . 502

Re playing and Saving . . . 502

Con trol Sum mary . . . 503

Gen eral Move ment. . . 503

Flying Move ment . . . 504

Sur faces . . . 507

Picking Up Ob jects . . . 507

Throwing Pro jec tiles. . . 508

Elec tric Pi ra nha . . . 510

Ac tions . . . 510

In ter ac tive Com bat En vi ron ments. . . 512

Looking . . . 513

Friends . . . 513

Speaking . . . 514

Cut-Scenes . . . 515

Levels . . . 516

III. Ar ti fi cial In tel li gence . . . 518

En emy AI . . . 519

Player De tec tion . . . 519

Mo tion . . . 519

Flying . . . 520

Pathfinding . . . 520

Taking Dam age . . . 520

Com bat At tacks . . . 520

Evading. . . 521

Spe cial Ac tions . . . 521

Trash Talking. . . 522

Falling into Traps . . . 522

Non-Combatant Agents . . . 523

Friends . . . 523

IV. Game El e ments . . . 525

Items . . . 525

Char ac ters . . . 527

V. Story Over view . . . 536

VI. Game Pro gres sion . . . 538

Set ting . . . 538

In tro duc tion . . . 540

Gargantuopolis . . . 540

The Elec tric Priest ess’ Bub ble Home . . . 540

Benthos. . . 541 xv

(16)

Har mony . . . 542

New Boston . . . 543

The Elec tric Priest ess’ Bub ble Home . . . 544

The Ikairus . . . 545

VII. Bib li og ra phy . . . 545

Glos sary . . . 546

Se lected Bib li og ra phy . . . 562

In dex . . . 565

(17)

My ear li est rec ol lec tion of play ing a com puter game was when I stum bled upon a halfheight Space Invaders at a tiny Mex i can res tau rant in my home town. I was per -haps six, and Space Invaders was cer tainly the most mar vel ous thing I had ever seen, at least next to LegoLand. I had heard of arcade games, but this was the first one I could actu ally play. Space Invaders, I knew, was better than tele vi sion, because I could con trol the lit tle ship at the bot tom of the screen using the joy stick and shoot the aliens myself instead of watch ing some one else do it. I was in love. The irony of this story is that, at the time, I failed to com pre hend that I had to stick quar ters into the game to make it work. The game was run ning in “attract” mode as arcade games do, and my young mind thought I was con trol ling the game with the joy stick when I was actu ally not con trol ling any thing. But the idea was still mind-blowing.

This book is about devel op ing orig i nal com puter games that will hope fully have the same mind-blowing effect on play ers that Space Invaders had on my young brain. This book deals with that devel op ment pro cess from the point of view of the game designer. Many books have been writ ten about the pro gram ming of com puter games, but I can remem ber my frus tra tion in being unable to find a book such as this one when I was an aspir ing game designer. In some ways, I have writ -ten this book for myself, for the per son I was a decade ago. I hope that other peo ple inter ested in design ing games will find this book infor ma tive. In my hum ble opin -ion, it is the game designer who has the most inter est ing role in the cre ation of a com puter game. It is the game’s design that dic tates the form and shape of the game’s gameplay, and this is the fac tor which dif fer en ti ates our artis tic medium from all oth ers.

(18)

What is Gameplay?

I hear you ask ing, “But what is gameplay?” Many peo ple think they know what gameplay is, and indeed there are many dif fer ent rea son able def i ni tions for it. But I have one def i ni tion that cov ers every use of the term you will find in this book. The gameplay is the com po nent of com puter games which is found in no other art form: interactivity. A game’s gameplay is the degree and nature of the interactivity that the game includes, i.e., how the player is able to inter act with the game-world and how that game-world reacts to the choices the player makes. In an action game such as Cen ti pede, the gameplay is mov ing the shooter ship around the lower quad rant of the screen and shoot ing the ene mies that attack relent lessly. In SimCity, the

gameplay is lay ing out a city and observ ing the cit i zens that start to inhabit it. In

Doom, the gameplay is run ning around a 3D world at high speed and shoot ing its

extremely hos tile inhab it ants, gath er ing some keys along the way. In San Fran cisco

Rush, the gameplay is steer ing a car down implau si ble tracks while jock ey ing for

posi tion with other rac ers. In StarCraft, the gameplay is maneu ver ing units around a map, find ing resources and exploit ing them, build ing up forces, and finally going head to head in com bat with a sim i larly equipped foe. And in Civ i li za tion, the gameplay is explor ing the world, build ing a soci ety from the ground up, dis cov er ing new tech nol o gies, and inter act ing with the other inhab it ants of the world.

Though some might dis agree with me, the gameplay does not include how the game-world is rep re sented graph i cally or what game engine is used to ren der that world. Nor does it include the set ting or story line of that gameworld. These aes -thetic and con tent con sid er ations are ele ments com puter games may share with other media; they are cer tainly not what dif fer en ti ates games from those other media. Gameplay, remem ber, is what makes our art form unique.

What is Game Design?

What, then, is game design? Hav ing defined what exactly I mean when I refer to gameplay, the notion of game design is quite eas ily explained: the game design is what deter mines the form of the gameplay. The game design deter mines what choices the player will be able to make in the game-world and what ram i fi ca tions those choices will have on the rest of the game. The game design deter mines what win or loss cri te ria the game may include, how the user will be able to con trol the game, and what infor ma tion the game will com mu ni cate to him, and it estab lishes how hard the game will be. In short, the game design deter mines every detail of how the gameplay will func tion.

(19)

Who is a Game Designer?

By this point it should be obvi ous what a game designer does: she deter mines what the nature of the gameplay is by cre at ing the game’s design. The terms “game designer” and “game design” have been used in such a wide vari ety of con texts for so long that their mean ing has become dilute and hard to pin down. Some seem to refer to game design as being syn on y mous with game devel op ment. These peo ple refer to any one work ing on a com puter game, be they art ist, pro gram mer, or pro -ducer, as a game designer. I pre fer a more spe cific def i ni tion, as I have out lined above: the game designer is the per son who designs the game, who thereby estab -lishes the shape and nature of the gameplay.

It is impor tant to note some tasks in which the game designer may be involved. The game designer may do some con cept sketches or cre ate some of the art assets that are used in the game, but he does not have to do so. A game designer may write the script con tain ing all of the dia log spo ken by the char ac ters in the game, but he does not have to do so. A game designer may con trib ute to the pro gram ming of the game or even be the lead pro gram mer, but he does not have to do so. The game designer may design some or all of the game-world itself, build ing the lev els of the game (if the pro ject in ques tion has lev els to be built), but he does not have to do so. The game designer might be tak ing care of the pro ject from a man age ment and pro duc tion stand point, keep ing a care ful watch on the mem bers of the team to see that they are all per form ing their tasks effec tively and effi ciently, but he does not have to do so. All some one needs to do in order to jus ti fi ably be called the game’s designer is to estab lish the form of the game’s gameplay. Indeed, many game design ers per form a wide vari ety of tasks on a pro ject, but their cen tral con -cern should always be the game design and the gameplay.

What is in This Book?

This book con tains a breadth of infor ma tion about game design, cov er ing as many aspects as pos si ble. Of course, no sin gle book can be the defin i tive work on a par tic -u lar art form. What this book cer tainly is not is a book abo-ut pro gram ming

com puter games. There are a wealth of books avail able to teach the reader how to pro gram, and as I dis cuss later in this book, know ing how to pro gram can be a great asset to game design. How ever, it is not a nec es sary com po nent of design ing a game; many fine design ers do not know how to pro gram at all.

The chap ters in this book are divided into three cat e go ries. First are the twelve core chap ters which dis cuss var i ous aspects of the devel op ment of a com puter game, from estab lish ing the game’s focus, to doc u ment ing the game’s design, to estab lish ing the game’s mode of sto ry tell ing, to playtesting the near-final prod uct.

(20)

These chap ters dis cuss the the ory behind game design, and what a designer should strive for in order to cre ate the best game pos si ble. The chap ters also include dis -cus sions of the real ity of game devel op ment, using exam ples from my own expe ri ence, to delve into the actual prac tice of game design.

There are five anal y sis chap ters included in this book, cov er ing five excel lent games in five dif fer ent gen res. One of the most impor tant skills a game designer must have is the abil ity to ana lyze games that she enjoys in order to under stand what those games do well. By under stand ing these other games, the designer may then attempt to rep li cate those same qual i ties in her own pro jects. That is not to sug gest that good game design ers merely copy the work of other game design ers. Under stand ing the rea sons why other games suc ceed will bring the designer a more com plete under stand ing of game design as a whole. Every game designer should take the games that she finds most com pel ling and try to exam ine what makes them tick. The exam ples I include in this book, Cen ti pede, Tetris, Loom, Myth: The

Fallen Lords, and The Sims, are all very unique games. And though a given pro ject

you are work ing on may not be sim i lar to any of these games, a lot can be learned from ana lyz ing games of any sort. First-person shooter design ers have had great suc cess in revi tal iz ing their genre by look ing at adven ture games. Cer tainly, roleplaying game design ers have recently learned a lot from arcade game design -ers. Melding in tech niques from other gen res is the best way to advance the genre you are work ing on and to cre ate some thing truly orig i nal.

This book also includes a group of inter views with six of the most

respected game design ers of the indus try’s short his tory who have designed some of the best games ever released. These are lengthy inter views that go deeper than the short press kit style inter views one finds on the Internet or in most mag a zines. In each inter view the sub ject dis cusses the best titles of his career and why he believes they turned out as well as they did. The design ers also talk at length about their own tech niques for devel op ing games. Through out my own career in game devel op -ment, I have found inter views with other com puter game design ers to be

exceed ingly help ful in learn ing how to per fect my craft. There is much infor ma tion to be gleaned from these chap ters, ideas that can help any game designer, regard less of how expe ri enced he may be.

At the end of the book you will find a glos sary. Though it is far from a com -plete list ing of game design ter mi nol ogy, it does cover many of the more eso teric terms I use in the book, such as a per sonal favor ite of mine, “sur ro gate.” Every game designer has a set of jar gon she uses to refer to var i ous aspects of her craft, and this jar gon is sel dom the same from one designer to the next. If noth ing else, the glos sary should help you to under stand my own jar gon. For instance, it will tell you the dif fer ence between gameplay and game mechan ics. Fur ther more, read ers who may find the con tent of this book to assume too much knowl edge may find the

(21)

glos sary help ful in sort ing out what an RTS game is and what the two dif fer ent mean ings for FPS are. Often, dis cus sions of game design can degrade into ques -tions of seman tics, with no two sides ever mean ing exactly the same thing when they refer to a game’s “engine.” I hope that the glos sary will help read ers to avoid that prob lem with this book.

Who This Book is For

This book is for any one who wants to under stand the com puter game devel op ment pro cess better from a strictly game design stand point. As I stated ear lier, there are plenty of books avail able to teach you how to pro gram, or how to use Photoshop and 3D Stu dio Max. This book will do nei ther of these things. Instead it focuses on the more elu sive topic of game design and how you can ensure that your title has the best gameplay pos si ble. Though solid pro gram ming and art are both cen tral to a game’s suc cess, no amount of flashy graph ics or cut ting-edge cod ing will make up for lack lus ter game design. In the end, it is the gameplay that will make or break a pro ject.

I have writ ten this book in such a way as to encom pass pro jects of dif fer ent scopes and sizes. It does not mat ter if the game you are work ing on is des tined for com mer cial release, if you hope to some day release it as share ware, or if you are only mak ing a game for you and your friends to play; this book should be help ful to a game designer work ing in any of those cir cum stances. Fur ther more, it does not mat ter if you are work ing on the game with a large team, with only a few accom pli -ces, or going com pletely solo. In the book I often make ref er ence to the “staff” of your pro ject. When I refer to “your pro gram ming staff” I may be refer ring to a team of ten sea soned cod ers com mand ing mas sive sal a ries and push ing the bound aries of real-time 3D tech nol ogy, or I may be refer ring to just you, cod ing up every last aspect of the game your self. When I refer to “your playtesting staff” I may be refer -ring to an expe ri enced and thor oughly pro fes sional test ing staff of fif teen who will pride them selves on giv ing your game a thor ough going-over, or I may be refer ring to your cous ins Bob and Judith who, like you, enjoy games and would love to play your game. Good games cer tainly do not always come from the big gest teams. Even today, when multi-million dol lar bud gets are the norm, the best games still often result from the vision and deter mi na tion of a lone indi vid ual, and he need not always sur round him self with a mas sive team to see that vision through to

com ple tion.

Many places in this book make ref er ence to you lead ing the design on the pro -ject on which you are work ing. Of course, not every designer can be in the lead posi tion on every pro ject, and even if you are the lead, you will often find your self with out the abso lute final say on what takes place in the game. In this regard, this

(22)

book is writ ten from a some what ide al is tic point of view. But regard less of how much author ity you actu ally have over the direc tion of the pro ject, the impor tant point is to always know what you would do with the pro ject if you could do what -ever you wanted. Then you should cam paign for this direc tion with the other peo ple on the team. If you are per sua sive enough and if you are, in fact, cor rect in your instincts, you have a good chance of con vinc ing them to do it your way. Pro jects are often led not by the peo ple with the most senior ity or who have the right title on their busi ness card; pro jects are lead by the peo ple who “show up” to the task, who care about their pro jects and are com mit ted to them, and who are will ing to put in the time and effort to make the game the best it can be.

Theory and Practice

Every medium has a unique voice with which it can speak, and it is the respon si bil -ity of the user of a medium to find that voice. Com puter games have a voice that I firmly believe to be as strong as that avail able in any other media. Com puter games are a rel a tively young form when com pared with the likes of the printed word, music, the visual arts, or the the ater, and I think this cur rently works against the like li hood of com puter games truly find ing their most pow er ful voice. This book is an attempt to help read ers find that voice in their own pro jects. This can come in both the more the o ret i cal form of ques tion ing why it is that play ers play games, but also in the entirely more prac ti cal form of how to most effec tively work with playtesters. To have any chance of pro duc ing a great game, the game designer must under stand both the the o ret i cal aspects and the prac ti cal neces si ties of game design.

(23)

Chap ter 1

What Players Want

“But when I come to think more on it, the big gest rea son it has be

-come that pop u lar is Mr. Tajiri, the main de vel oper and cre ator of

Pokemon, did n’t start this pro ject with a busi ness sense. In other words,

he was not in tend ing to make some thing that would be come very pop

-u lar. He j-ust wanted to make some thing he wanted to play. There was

no busi ness sense in cluded, only his love in volved in the cre ation.

Some how, what he wanted to cre ate for him self was ap pre ci ated by

oth ers in this coun try and is shared by peo ple in other coun tries.

. . . And that’s the point: not to make some thing sell, some thing very

pop u lar, but to love some thing, and make some thing that we cre ators

can love. It’s the very core feel ing we should have in mak ing games.”

— Shigeru Miyamoto, talk ing about the cre ation of

Pokemon

(24)

G

ame design ers spend a lot of time con cern ing them selves with what game play ers are look ing for in a com puter game. What can they put in their com puter games that has not been done before and will excite play ers? Often game design ers are so bereft of an idea of what gamers want that they instead only include gameplay ideas that have been tried before, rehash ing what was pop u -lar with game play ers last year. Surely if play ers liked it last year, they will like it this year. But therein lies the rub. Gamers gen er ally do not want to buy a game that is only a clone of another game, a “new” game that only offers old ideas and brings noth ing orig i nal to the table. None the less, suc cess ful games can be use ful, not for clon ing, but for anal y sis. As game design ers, we can look at the games that have come out pre vi ously, that we have enjoyed in years past, and try to deter mine a set of direc tives that explain what com pelled us to try those games in the first place, and why they held our inter est once we started play ing them.

Why Do Players Play?

The first ques tion we should con sider is: why do play ers play games in the first place? Why do they choose to turn on their com puter and run Doom instead of vis it -ing the art museum or go-ing to see a movie? What is unique about com puter games ver sus other human enter tain ment pur suits? What do games offer that other activ i -ties do not? It is by under stand ing what is attrac tive about games that other media do not offer that we can try to empha size the dif fer ences, to dif fer en ti ate our art form from oth ers. To be suc cess ful, our games need to take these dif fer ences and play them up, exploit them to make the best gameplay expe ri ence pos si ble.

Players Want a Challenge

Many play ers enjoy play ing games since they pro vide them with a chal lenge. This pro vides one of the pri mary moti vat ing fac tors for sin gle-player home games, where social or brag ging rights moti va tions are less of an issue. Games can enter tain play -ers over time, dif fer ently each time they play, while engag ing their minds in an entirely dif fer ent way than a book, movie, or other form of art. In some what the same way some one might fid dle with a Rubik’s Cube or a steel “remove the ring” puz zle, games force play ers to think actively, to try out dif fer ent solu tions to prob -lems, to under stand a given game mech a nism.

When a per son faces a chal lenge and then over comes it, that per son has learned some thing. It does not mat ter if that chal lenge is in a math text book or in a com -puter game. So, chal leng ing games can be learn ing expe ri ences. Players will learn from games, even if that learn ing is lim ited to the con text of the game, such as how to get by level eight, and so forth. In the best games, play ers will learn les sons through gameplay that can be applied to other aspects of their life, even if they do

(25)

not real ize it. This may mean that they can apply prob lem solv ing meth ods to their work, use their improved spa tial skills to better arrange their fur ni ture, or per haps even learn greater empa thy through game role-playing. Many play ers thrive on and long for the chal lenges games pro vide, and are enriched by the learn ing that fol lows.

Players Want to Socialize

I have a friend who main tains that games are anti so cial. This is, of course, absurd, as nearly all non-computer games require a social group in order to func tion. Games arose as a com mu nal activ ity many mil len nia ago out of a desire to have a chal leng -ing activ ity in which a group of friends and fam ily could engage in. Com puter game design ers need to remem ber that the roots of gam ing, and an impor tant part of its appeal, are in its social nature.

For most peo ple, the pri mary rea son they play games is to have a social expe ri -ence with their friends or fam ily. I am not talk ing about com puter games here, but rather board and card games like chess, Monop oly, bridge, Scrab ble, Diplo macy, or

The Set tlers of Catan. Peo ple like to play these games because they like being with

their friends and want to engage in a shared activ ity that is more social than going to a movie or watch ing TV. It is true that lots of peo ple enjoy play ing sol i taire card games as well, but there are many more multi-player games than there are sin player. This is because peo ple enjoy a social gameplaying expe ri ence.

But how does this apply to com puter games? If one con sid ers all the com puter games ever cre ated, the major ity of them are sin gle-player only expe ri ences. But of course there are plenty of multi-player games, rang ing from the “death-matches” found in Doom and its imi ta tors, to the clas sic M.U.L.E. game of wheeling and deal ing, to the per sis tent worlds founds in MUDs (Multi User Dun geons) or their com mer cial equiv a lent, Ultima Online.

Almost all death-match style multi-player games are basi cally adap ta tions of sin gle-player games into multi-player incar na tions. Though there are excep tions, such as Quake III or Unreal Tour na ment, these games usu ally pro vide a sin player (SP) game in addi tion to the multi-player (MP) game. The SP and MP games are played with nearly the same set of rules and game mechan ics. But even in these sin gle-player-turned-multi-player games, play ers like to social ize while play ing. Any one who has ever played one of these games over a LAN in a room with a bunch of their friends can tes tify to this. These LANfests are usu ally rich with con -ver sa tion as play ers shout back and forth to each other, brag ging o-ver their most recent “frag” or pro claim ing how close they came to being killed. Games such as

Quake can also be played over the Internet, where the expe ri ence is quite a bit less

social, since play ers may be miles apart and are thus only able to com mu ni cate through the com puter. And the high-intensity and fast-action nature of these games

(26)

does n’t leave play ers much time to type mes sages to their oppo nents, if they hope to sur vive for long. But these games do still pro vide chat func tion al ity, and play ers, when they are in a safe cor ner, after they have died, or between games, can send con ver sa tional mes sages to each other. At more hec tic points in the gameplay the mes sages are short and typed on the fly, con sist ing of only a cou ple of let ters. The fact that play ers still try to chat with each other in these highvelocity games is tes -ta ment to the play ers’ desire to social ize.

A sep a rate cat e gory of multiplayer games is what has come to be called “per -sis tent uni verse” or “mas sively multi-player” games. These games tend to be more in the style of role-playing games, where play ers wan der around “vir tual worlds” and meet and inter act with the other char ac ters in these worlds, char ac ters who are con trolled by other play ers. These games tend to be played over large net works such as the Internet, instead of over LANs, and as a result play ers only social ize with each other through what they type into the com puter. Since these games are con sid er ably slower paced than deathmatch games, there is a much greater oppor -tu nity for the play ers to chat with each other while play ing. MUDs were the first pop u lar incar na tion of this style of game, which were played pri mar ily by col lege stu dents from the late 1980s on. At the time, col lege stu dents were the main group of peo ple with free time who were hooked to the Internet. These games are

textonly, and pro vide their play ers with quests to accom plish in mostly fan tasy set -tings. The quests, how ever, take a backseat to the social iza tion and role-playing, with play ers spend ing the vast major ity of their time chat ting with other play ers. A lot of peo ple are drawn into play ing these games as a way to inter act with their friends, despite the fact that these friends are peo ple they met online and who they

Unreal

Tour na ment is an

exam ple of a game which focuses pri mar ily on pro vid ing a multi-player expe ri ence.

(27)

have never seen in per son. Indeed, the per sis tent worlds, MUDs in par tic u lar, draw in a legion of play ers who are not inter ested in play ing any sin gle-player com puter games. These peo ple play games in order to meet and talk to other peo ple. The games are an activ ity these peo ple can engage in together while social iz ing.

As multi-player games have become more and more com mon, many game devel op ers have been quick to point out their advan tages in terms of com pet i tive AI. Human oppo nents are much more unpre dict able and chal leng ing than any AI that could be rea son ably cre ated for most games. This, they sug gested, is why peo -ple are drawn to multi-player games. But the big gest advan tage of these

multi player games is that they trans form com puter games into truly social expe ri -ences, which is one of the larg est moti vat ing fac tors for peo ple to play games.

Players Want a Dynamic Solitaire Experience

Per haps I have con fused the reader by say ing first that play ers want to social ize and then sug gest ing that play ers want a sol i taire expe ri ence. Of course the two do not hap pen at the same time; some game play ers are look ing for a social expe ri ence, and a dif fer ent set are look ing for some thing dynamic that they can engage in by them selves. Some times friends are not avail able, or a player is tired of his friends, or sim ply tired of hav ing to talk to other peo ple all the time. Sim i lar to the dif fer -ence between going to a movie the ater with an audi -ence ver sus rent ing a video alone at home, the anti so cial nature of sin gle-player games attracts a lot of peo ple who have had enough of the other mem bers of the human race.

But games are dis tinct from other sol i taire expe ri ences such as read ing a book or watch ing a video since they pro vide the play ers with some thing to inter act with, an expe ri ence that reacts to them as a human would, or at least in a man ner resem -bling a human’s reac tions. But the play ers are always in con trol, and can start and stop play ing at any time. Thus the com puter game “fakes” the inter est ing part of human inter ac tion with out all of the poten tial annoy ances. In this way, peo ple are able to turn to com puter games for a dynamic and inter ac tive yet anti so cial expe ri ence.

Players Want Bragging Rights

Par tic u larly in multi-player gam ing, play ers play games to win respect. Being able to frag all of your friends in Doom will force them to have a grudg ing respect for you: “Bob isn’t very good in alge bra class, but he can sure anni hi late me in a deathmatch.” Even in sin gle-player games, play ers will talk with their friends about how they fin ished one game or about how good they are at another. Players will brag about how they played the whole game through on the hard est dif fi culty in only a few hours. If one looks at arcade games both old and new, the high-score table and the abil ity to enter one’s name into the game, even if only three let ters, pro vides a

(28)

tre men dous incen tive for peo ple to play a game repeat edly. Players who may not have much to brag about in their ordi nary lives, who may not be ter ri bly phys i cally coor di nated at sports or book ish enough to do well in school, can go down to the arcade and point out to all their friends their ini tials in the Cen ti pede game. Even with out tell ing any one, play ers can feel a tre men dous sense of self-satisfaction when they beat a par tic u lar game. When play ers are vic to ri ous at a chal leng ing game, they real ize they can do some thing well, prob a bly better than most peo ple, which makes them feel better about them selves.

Players Want an Emotional Experience

As with other forms of enter tain ment, play ers may be seek ing some form of emo -tional pay off when they play a com puter game. This can be as sim ple as the adren a line rush and ten sion of a fastaction game like Doom. Or it can be con sid er -ably more com plex, such as the player’s feel ing of loss when her friendly robot com pan ion sac ri fices him self for the player in Steve Meretzky’s Planetfall. Sadly, many games’ emo tional ranges are lim ited to excite ment/ten sion dur ing a con flict, despair at repeated fail ure at a given task, and then ela tion and a sense of accom -plish ment when the player finally suc ceeds. It may seem strange that play ers would play a game in order to feel despair. But many peo ple enjoy watch ing plays that are trag e dies or mov ies that have sad end ings, or lis ten ing to music that is out-and-out depress ing. Peo ple want to feel some thing when they inter act with art, and it does not nec es sar ily need to be a pos i tive, happy feel ing. Per haps the sense of cathar sis peo ple obtain from these works makes them worth expe ri enc ing. Many clas sic arcade games, such as Cen ti pede or Space Invaders, are unwinnable. No mat ter what the player does, even tu ally the game will beat him. These games are, in a sense, les sons in defeat—trag e dies every time the player plays them. Yet the player keeps pump ing in his quar ters. This is why a player’s feel ing of hope less ness as a game repeat edly bests him is not to be ignored. The player is feel ing some thing, and some would say that is the goal of art.

Emo tional range is not some thing com puter games have explored as much as they could. The exam ple from Planetfall I cited above is one of the very few exam -ples in com puter games of a player becom ing attached to a char ac ter in a game, only to have him killed later on. Many devel op ers are wary of mak ing a game too sad. But in the case of Planetfall, the tragic story twist of that game was exploited for all the pathos it was worth by designer Steve Meretzky. It is a moment of trag -edy that has stuck in many gamers’ mem o ries. Game design ers would be wise to con cen trate on expand ing the emo tional expe ri ence in games beyond excite ment and accom plish ment, into more unex plored and uncharted emo tional ter ri tory.

(29)

Players Want to Fantasize

A major com po nent of the pop u lar ity of sto ry tell ing art forms is the ele ment of fantasy. Whether one con sid ers nov els, films, or comic books, many peo ple expe ri -ence these works to “get away” from their own “mun dane” lives and escape to an alto gether dif fer ent world, one filled with char ac ters who engage in excit ing, inter -est ing activ i ties, travel to exotic locales, and meet other fas ci nat ing peo ple.

Cer tainly not all sto ry tell ing works por tray excit ing and glam or ous pro tag o nists, but there is cer tainly a large seg ment of works that is labeled “escap ist.” Some crit ics deride such escap ist pieces of art, and indeed a lot of very good books, mov ies, and com ics deal with more real is tic set tings and top ics to great effect. The fact remains, how ever, that many peo ple want to be trans ported to a world more glam or ous than their own.

Com puter games, then, have the poten tial to be an even more immersive form of escap ism. In games, play ers get the chance to actu ally be some one more excit -ing, to con trol a pulp-fiction adven turer, dar ing swords man, or space-opera hero. While in books or films the audi ence can merely watch as the char ac ters lead excit -ing lives, in a well-designed com puter game a player will actu ally get the chance to live those lives them selves. Even better, these fan tasy lives are not weighed down with the mun dane events of life. In most games, play ers do not have to worry about eat ing, need ing to get some sleep, or going to the bath room. Thus, a game can cre -ate a fan tasy life with out the tedious details. And, most impor tantly, the level of fan tasy immer sion is height ened from that of other art forms because of the inter ac -tive nature of gam ing.

Another part of the fan tasy ful fill ment ele ment of com puter games is enabling the player to engage in socially unac cept able behav ior in a safe envi ron ment. Many pop u lar games have allowed play ers to pre tend they are crim i nals or assas sins.

Driver is a good exam ple of this. Though the back-story explains that the player is

actu ally play ing an under cover police offi cer, in Driver the player gets to pre tend she is a crim i nal who must evade the police in elab o rate car chases. There is a dev -il ish thr-ill to out run ning police cars, espe cially for any one who has ever been pulled over by one. Though most play ers would never con sider driv ing in car chases in real life, there’s some thing tempt ing and entic ing about engag ing in taboo activ i ties. Com puter games pro vide a good medium for play ers to explore sides of their per -son al ity that they keep sub merged in their daily lives.

Players may also fan ta size about events in his tory. If the player could have been Napo leon, would Waterloo have turned out dif fer ently? If the player were a rail road baron in the twen ti eth cen tury, would he be able to cre ate a pow er ful finan cial empire? A whole line of his tor i cal games, from wargames to eco nomic sim u la tions, allow play ers to explore events in his tory, and see how mak ing dif fer ent choices than the his tor i cal fig ures involved made will result in wildly dif fer ent out comes.

(30)

While many peo ple spend their time dwell ing on the past, won der ing how events could have tran spired dif fer ently if alter nate deci sions had been made, games can give play ers a chance to find out how his tory might have been dif fer ent.

Even with out the ele ments of excite ment and glam our, even if another per son’s life is not actu ally that excit ing, it can be inter est ing to spend time as that per son. Good com puter games can pro vide play ers with the oth er wise unavail able oppor tu -nity to see the world through some one else’s eyes. As mil lions of gamers can attest, it is fun to role-play and it is fun to fan ta size.

What Do Players Expect?

Once a player has decided he wants to play a given game because of one moti vat ing fac tor or another, he will have expec ta tions for the game itself. Beyond the game not crash ing and look ing rea son ably pretty, play ers have cer tain gameplay expec ta -tions, and if these are not met, the player will soon become frus trated and find another game to play. It is the game designer’s job to make sure the game meets these expec ta tions. So once they start play ing, what do play ers want?

Players Expect a Consistent World

As play ers play a game, they come to under stand what actions they are allowed to per form in the world, and what results those actions will pro duce. Few things are more frus trat ing than when the player comes to antic i pate a cer tain result from an action and then the game, for no per ceiv able rea son, pro duces a dif fer ent result. Worse still is when the con se quences of the player’s actions are so unpre dict able that a player can not estab lish any sort of expec ta tion. Hav ing no expec ta tion of what will hap pen if a cer tain maneu ver is attempted will only frus trate and con fuse play ers, who will soon find a dif fer ent, more con sis tent game to play. It is the con -sis tency of actions and their results that must be main tained, for an unpre dict able world is a frus trat ing one to live in.

Fighting games are a par tic u larly appro pri ate exam ple of the impor tance of pre -dict able out comes from actions. Players do not want a maneu ver to work

some times and fail other times, with out a readily appar ent rea son for the dif fer ent out comes. For instance, in Tekken, if the player misses a kick, it has to be because her oppo nent jumped, blocked, was too far away, or some other rea son that the player can per ceive. The player’s per cep tion of the rea son for the move’s fail ure is impor tant to empha size. It may be that the inter nal game logic, in this case the col li -sion sys tem, will know why the player’s kick missed, but it is as bad as hav ing no rea son if the player can not eas ily rec og nize why the maneu ver failed. Fur ther more, if only expert play ers can under stand why their action failed, many nov ices will become frus trated as they are defeated for no rea son they can under stand. If a kick

(31)

fails in a sit u a tion that closely resem bles another sit u a tion in which the same kick suc ceeded, play ers will throw their hands up in frus tra tion.

Pin ball games are another inter est ing exam ple. Of course, a pin ball game is a com pletely pre dict able game-world, since it is based on real-world phys ics. An expert pin ball player knows this, and will use it to his advan tage. But the prob lem comes with the nov ice. Inex pe ri enced play ers will often fail to see what they “did wrong” when the ball goes straight down between their flip pers, or rolls down one of the side gut ters. These play ers will curse the pin ball game as a “game of luck” and not want to play any more. Of course, the fact that play ers of dif fer ent skill lev -els will have rad i cally dif fer ent lev -els of suc cess at a given pin ball game shows that it is not just a game of luck. But only those play ers who stick with the game

through numer ous early fail ures will find this out. I am not sug gest ing that pin ball games should be aban doned or rad i cally sim pli fied, but one of their short com ings is that they alien ate new play ers who can not see the con nec tions between their actions and the out come of the game.

Players Expect to Understand the Game-World’s Bounds

When play ing a game, a player wants to under stand which actions are pos si ble and which are not. He does not need to imme di ately see which actions are needed for a given sit u a tion, but he should under stand which actions it is pos si ble to per form and which are out side the scope of the game’s play-space.

For instance, in Doom, a player will intu itively fig ure out that she is not going to be able to hold a dis cus sion with the demons she is fight ing. The player will not

In Doom II, the

player will not expect to be able to start a

con ver sa tion with the mon sters he is attack ing.

(32)

even want to ini ti ate a con ver sa tion with a demon dur ing which she sug gests sur -ren der as the most log i cal course of action. The player under stands that such inter per sonal dis cus sion is out of the scope of the game. Sup pose that Doom had included a mon ster late in the game, a foe that could only be defeated if the player was friendly to it, win ning it over with her witty con ver sa tion. Players would have been frus trated, since they came to under stand, through play ing the lev els that led up to that level, that in Doom all that is needed for vic tory is to blast every thing that moves, while avoid ing get ting hit. Talking is com pletely out of the scope of the game.

Of course, a chatty mon ster in Doom is an extreme exam ple of a game hav ing unpre dict able bounds, but plenty of games break this design prin ci ple. These games have play ers per form ing actions and com plet ing lev els using a cer tain type of game mech a nism, and then later on insert puz zles that can only be solved using an

entirely new mech a nism. The prob lem is that the player has been taught to play the game a cer tain way, and sud denly the game requires the player to do some thing else entirely. Once play ers come to under stand all of the gameplay mech a nisms that a game uses, they don’t want new, unintuitive mech a nisms to be ran domly

intro duced.

Players Expect Reasonable Solutions to Work

Once a player has spent some time play ing a game, he comes to under stand the bounds of the game-world. He has solved numer ous puz zles, and he has seen what sort of solu tions will pay off. Later in the game, then, when faced with a new puz -zle, the player will see what he regards as a per fectly rea son able solu tion. If he then tries that solu tion and it fails to work for no good rea son, he will be frus trated, and he will feel cheated by the game.

This sort of dif fi culty in game design is par tic u larly true in games that try to model the real-world to some degree. In the real-world there are almost always mul ti ple ways to accom plish a given objec tive. There fore, so too must it be in a com puter game set in the real-world. Of course, a designer always pro vides at least one solu tion to a puz zle, and granted that solu tion may be per fectly rea son able. But there may be other equally rea son able solu tions, and unless the designer makes sure those solu tions work as well, play ers will dis cover and attempt these functioning alter nate solu tions and will be irri tated when they do not work. It is the game designer’s task to antic i pate what the player will try to do in the game-world, and then make sure that some thing rea son able hap pens when the player attempts that action.

Players Expect Direction

Good games are about let ting the play ers do what they want, to a point. Players want to cre ate their own suc cess sto ries, their own meth ods for defeat ing the game,

(33)

some thing that is uniquely theirs. But at the same time, play ers need to have some idea of what they are sup posed to accom plish in this game. Not hav ing direc tion is a bit too much like real life, and play ers already have a real life. Many gamers are prob a bly play ing the game in order to get away from their real lives, to fan ta size and escape. They usu ally do not play games in order to sim u late real life on their com puter.

Players want to have some idea of what their goal is and be given some sug ges -tion of how they might achieve that goal. With a goal but no idea of how to achieve it, play ers will inev i ta bly flail around, try ing every thing they can think of, and become frus trated when the maneu vers they attempt do not bring them any closer to their goal. Of course, with out an idea of what their goal is, play ers are left to just wan der aim lessly, per haps enjoy ing the scen ery, mar vel ing at the immersive game-world. Yet with out some thing to do in that game-world, it is point less as a game. If the play ers do not know what their goal is, the goal might as well not exist.

The clas sic exam ple of the goal-less game is SimCity. In fact, Will Wright, the game’s cre ator, calls it a “soft ware toy” instead of a game. SimCity is like a toy in that the player can do what ever she wants with it, with out ever explic itly being told that she has failed or suc ceeded. In some ways SimCity is like a set of Legos, where a player can build what ever she wants just for the thrill of cre ation. The trick, how -ever, is that SimCity is a city sim u la tor, wherein the player is allowed to set up a city how ever she wants. But since the game sim u lates real ity (con struct ing and run -ning a city), and the player knows what is con sid ered “suc cess” in real ity (a

boom ing city full of lovely sta di ums, pala tial librar ies, and happy cit i zens), she will nat u rally tend to impose her own rules for suc cess on the game. She will strive to

SimCity 3000

is the third in a series of city sim u la tion “soft ware toys,” which let users play with out giv ing them a spe cific goal.

(34)

make her idea of the per fect city, and keep its cit i zens happy and its econ omy buoy -ant. In a sub tle way, the player is directed by her own expe ri ence with real ity. If

SimCity had been a sim u la tion of a sys tem that play ers were com pletely unfa mil iar

with, it would cer tainly have been less pop u lar. Though the game does not explic -itly have a goal, the very nature of the game and its ground ing in real ity encour ages play ers to come up with their own goals. And so, what starts out as a toy becomes a game, and thus the play ers are com pelled to keep play ing.

Players Expect to Accomplish a Task Incrementally

Given that play ers under stand what their goal in the game-world is, play ers like to know that they are on the right track toward accom plish ing that goal. The best way to do this is to pro vide numer ous sub-goals along the way, which are com mu ni cated to the player just as is the main goal. Then, a player is rewarded for achiev ing these sub-goals just as he is for the main goal, but with a pro por tion ally smaller reward. Of course one can take this down to any level of detail, with the sub-goals hav ing sub-sub-goals, as much as is nec es sary to clue the player in that he is on the right track. With out pro vid ing feed back of this kind, and if the steps nec es sary to obtain a goal are par tic u larly long and involved, a player may well be on the right track and not real ize it. When there is no pos i tive rein force ment to keep him on that track, a player is likely to try some thing else. And when he can not fig ure out the solu tion to a par tic u lar obsta cle, he will become frus trated, stop play ing, and tell all his friends what a mis er a ble time he had play ing your game.

Players Expect to Be Immersed

A direc tor of a musi cal I was once in would become incensed when actors wait ing in the wings would bump into the cur tains. She sug gested that once the audi ence sees the cur tains mov ing, their con cen tra tion is taken away from the actors on the stage. Their sus pen sion of dis be lief is shat tered. They are reminded that it is only a play they are watch ing, not real at all, and that there are peo ple jos tling the cur tains sur round ing this whole cha rade. Per haps exag ger at ing a bit, this direc tor sug gested that all of Broad way would col lapse if the cur tains were seen shak ing.

But she had a point, and it is a point that can be directly applied to com puter games. Once a player is into a game, she is in a level, she has a good under stand ing of the game’s con trols, she is excited, and she is role-playing a fan tasy; she does not want to be snapped out of her expe ri ence. Cer tainly the game should not crash. That would be the most jar ring expe ri ence pos si ble. Beyond that, the player does not want to think about the game’s GUI. If the GUI is not designed to be trans par -ent and to fit in with the rest of the game-world art, it will stick out and ruin her immer sion. If a char ac ter that is sup posed to be walk ing on the ground starts walk -ing into the air for no rec og niz able rea son, the player will real ize it is a bug and her

(35)

sus pen sion of dis be lief will be shat tered. If the player comes to a puz zle, fig ures out a per fectly rea son able solu tion to it, and that solu tion does not work, the player will again be reminded that she is “only” play ing a com puter game. All of these pit falls and many oth ers detract from the player’s feel ing of immer sion, and each time the player is rudely awak ened from her game-world fan tasy, the harder it is to

reimmerse her self in the game-world. Remem ber that many play ers want to play games in order to ful fill fan ta sies. And it is very hard to ful fill a fan tasy when the game’s idio syn cra sies keep remind ing the player that it is just a game.

Another impor tant aspect of player immer sion is the char ac ter the player is con -trol ling in the game. Most all games are about role-playing to some extent. And if the char ac ter the player is con trol ling, his sur ro gate in the gameworld, is not some -one the player likes or can see him self as being, the player’s immer sion will be dis rupted. For instance, in the third-person action/adven ture game Super Mario 64, the player is pre sented with a char ac ter to con trol, Mario, who does not have a very dis tinct per son al ity. Mario has a fairly unique look in his pseudo-plumber getup, but he never really says much, and acts as some thing of a blank slate on which the player can impose his own per son al ity. On the other hand, some adven ture games have starred char ac ters who acted like spoiled brats, and the player has to watch as his char ac ter says annoy ing, idi otic things over and over again. Each time the char -ac ter says some thing that the player would never say if he had the choice, the player is reminded that he is play ing a game, that he is not really in con trol of his char ac ter as much as he would like to be. In order for the player to become truly immersed, he must come to see him self as his game-world sur ro gate.

Despite all his fame, Mario does not have a very dis tinct per son al ity. He is pic tured here in