Just One More Game Standalone

Category: Entertainment

Presentation Description

No description available.


Presentation Transcript

Just One More Game…:Learning as Central to Gameplay: 

Just One More Game…: Learning as Central to Gameplay Martin C. Martin www.martincmartin.com Note: When giving this talk in person, I use a different version of this PowerPoint file with almost no text. This version is designed to stand on it’s own, without my talking along side it.

Why Are Games Fun?: 

Why Are Games Fun? This presentation talks about what makes games fun. It’s a big topic, and I’ll only scratch the surface. There are, of course, many things that make a game fun. Beautiful graphics, such as this image from Fary Cry, are one thing…

Why are games fun?: 

Why are games fun? .. and an engrossing story is another. This is an image from Shenmu for the Dreamcast, one of the most expensive games to make for it’s time. It received raves for its engrossing story line and the open endedness of its world.


Interaction But movies, books, and TV can have beautiful images and engrossing stories. Yet there’s something different about games. People play games for 8 hours at a sitting. (Or 24 hours. Or 36 hours…). If a movie takes longer than 3 hours, people complain it’s too long. Yet dames typically take 40+ hours to complete, the entire time in the same world and story.


Interaction What’s more, Tetris was the most popular game of all time for 20 years, and yet it’s graphics were nothing to look at, and had no story. The difference is that games are interactive. But how, exactly, does interactivity make games fun? That is what this presentation will address. There are common answers to this question, such as shaping the story, the thrill of winning, or …


Interaction Affecting the world. This image is from Duke Nukem 3D, whose great advance over Doom was that you could shoot out windows and lights.

Only For So Long: 

Only For So Long But those only hold our interest for so long. For example, Choose Your Own Adventure books allow you to shape the story. Every few pages, they pose a question like 'To take the path going up, turn to page 24. To take the path going down, turn to page 137.' While fun at first, they’re rarely read today. Rather than becoming a new genre of book, they’ve mostly died out. And tic-tac-toe is fun at first, but becomes boring once can play perfect game. In other words, the more you win, the less the thrill of winning holds our attention.

Only For So Long: 

Only For So Long And while affecting the world by, say, breaking open crates is fun, by itself it’s not what keeps players glued to their computers for so long.

Figuring Out How To Improve: 

Figuring Out How To Improve This presentation proposes that figuring out how to do better and better at a game is what keeps us so absorbed. We’ll look at what this tells us about existing games; extract some game design principles; and suggest some pratical steps that game designers can take.

Figuring Out How To Improve: 

Figuring Out How To Improve Think of how a First Person Shooter newbie approaches the game. Aiming is tricky, so when they want to shoot, they tend to stand still Must choose: dodge OR shoot As the game gets difficult, player gets frustrated As the enemies shoot more often, the player spends most of the time dodging and rarely shooting. Then becomes clear: they could do a lot better if they could dodge AND shoot As they start to get the hang of it: They think 'Now I’ve got it! Just one more game!' They feel they have an edge over other players Confidence rises: its own reward

The Learning Cycle: 

The Learning Cycle Player always trying to achieve goals E.g. hit enemy, don’t get hit. Always has a strategy to achieve them E.g. dodging, looking for powerups, looking for cover The player looks at the world in terms of these goals Always trying to improve strategy Leads to discovery of subgoals. E.g. discovers a place with good cover but also good shooting opportunities: a sniping location Thinks 'Now I’ve got it! Just one more game!' Feeling competent is an emotional reward. It makes these goals easier.

Game Design Principles: 

Game Design Principles 1a. 'What approaches will the player try?' When confronted with a choice, player comes up with some 'approach' to solving it. I use the word 'approach' rather than 'strategy' because I want something with the connotations from strategy to tactics to hand-eye coordination. 1b. 'How can they tell what to improve?' The initial approach is often quite simple, and needs to be improved over and over.

1. What’s The Approach?: 

Asking these two questions about existing games explains a lot about the success or failure of certain genres of game. Choose your own adventure games are fun until you realize: there’s no way to figure out which choice is right. If you’re in a cave trying to get to the surface, and one path goes up while the other goes down, you’re just as likely to be eaten if you go up or down. Once you realize this, you just choose responses at random. There’ no skill. 1. What’s The Approach?

1. What’s The Approach?: 

Why does tic-tac-toe survive, but only as a kid’s game? It’s fun while you’re learning. Imagine someone playing for the first time, and how they might approach it. They’ve been told that the first person to get three in a row wins, so they may approach it as a race: get to three in a row as quickly as you can, ignoring the other player. After being blocked a few times, and losing when they’re one step away from winning, they realize they need to block an opponent who has two in a row. They think 'Ah, now I can play better! Just one more game!' The insight spurs them on. Once they become perfect, the game becomes boring: it’s just turning the crank on an algorithm 1. What’s The Approach?

2. Combination Puzzles Aren’t Much Fun: 

2. Combination Puzzles Aren’t Much Fun Suppose a player finds a safe, and somehow knows they can’t find the combination anywhere. The only thing they can do is try numbers and find out whether that set of numbers is right or not. What’s the approach? Just try all combinations one after the other. Not much fun. It’s amazing how many forms of gameplay amount to the same thing.

2. Combination Puzzles Aren’t Much Fun: 

Games that are too easy aren’t any fun. So sometimes, to make a game more challenging, they make it so you almost need to be clairvoyant to defend against attacks. Imagine a level of an RTS where the goal is to defend a base. Suppose the AI rushes the player, sending a large force from the south only a few mintues into the game. Suppose the only way to survive is to build defenses like mad from the start, all at the southern edge of the base. 2. Combination Puzzles Aren’t Much Fun

2. Combination Puzzles Aren’t Much Fun: 

Then, a few minutes later, the enemy attacks from the west. Again, the only way to succeed is to spend that time moving defenses to the west, and building up more there, leaving the rest of your base undefended. And then, just to top it off, the same thing happens a few minutes later from the north, with the same requirement from the player. If a player does the reasonable thing and build up defenses evenly on all sides, she’ll get creamed. What approach does she learn? 'Play the game until attacked, then once I know where it will come from, revert to a saved game and defend against that.' 2. Combination Puzzles Aren’t Much Fun

2. Combination Puzzles Aren’t Much Fun: 

That’s just turning the crank on a simple algorithm. It’s not much fun. This example comes from a published game (although not Warcraft III pictured here), where the designers wanted it to make the first level very hard to appeal to a hardcore gamer audience. But build/get attacked/revert doesn’t keep your interest long, even for hardcore gamers. Simply asking 'What’s the approach they’ll learn' can cure this. Another important point: once players hit on the strategy of build/get attacked/revert, they’ll stick with this as long as it works, even if other, more enjoyable or successful strategies become possible. There are enough examples of that to turn it into it’s own principle… 2. Combination Puzzles Aren’t Much Fun

Game Design Principles: 

Game Design Principles 1a. 'What approaches will the player try?' 1b. 'How can they tell what to improve?' 'Guess The Combination' isn’t much fun Players stick with existing approach until it fails

Introducing New Units: 

Introducing New Units This is important when introducing new units or abilities. Imagine an RTS where, when you start some level, an artillery unit becomes available. If it shows up in the menus without much fanfare, people won’t think to look for it. They’ll stick with the infantry units that have worked up till then. If they can take out the enemy base using only infantry with a Hurculean effort – grabbing all the resources on the level as quickly as possible, building up the biggest force, honing all the tactics they learned on earlier levels – they’re likely to do just that, and think it’s an unusally hard level.

Introducing New Units: 

Introducing New Units But suppose there’s something that clearly can’t be solved the old way. For example, they need to traverse a bridge or other narrow path where they’ll be pelted by enemy artillery from a far – out of reach of their infantry. If the infantry gets wiped out in seconds, no matter how many units are thrown at that pass, they’ll stop an reconsider. What can take out that enemy artillery from a distance? There are a lot of units in that tech tree. Is one of them useful? When you want to player to change their approach, it helps to make the old approach clearly infeasible. If it just becomes very difficult, people may think the level is just hard.

Changing Difficulty Levels: 

Changing Difficulty Levels How do we create easier difficulties? A common method is to just weaken parts of the game: make the enemies slower, do less damage, etc. Obvious, right? Well, what does this do to the player’s approach? If the enemies are slow enough, then in a game like capture the flag, the player may learn to just run past them. Later, when the game gets harder, they’ll be forced to learn the basic of combat when the AI has more accurate aim and is better at dodging. That can be very discouraging. Unreal Tournament 2004 does this well: bots at lower skill levels simply turn off various advanced tactics, making them behave more like newbie players.

Game Design Principles: 

Game Design Principles 1a. 'What approaches will the player try?' 1b. 'How can they tell what to improve?' 'Guess The Combination' isn’t much fun Players stick with existing approach until it fails Player must be able to find viable approach

Only Advanced Maneuvers Succeed: 

Only Advanced Maneuvers Succeed Realistic military airplane simulators were a viable genre in the 1980s, but have died out. What initial approach will users try? Almost no one reads the manual, so they fly around until they see an enemy plane, then turn toward them. But they have a big turning radius, and the enemy is turning towards the player as well. They end up going in circles, chasing each other’s tail. What will they try next? Maybe turning at random, or climbing andamp; diving, or some other desperate gambit. But even when you somehow get mostly lined up, it’s hard to hit the enemy since it’s hard to 'fine tune' your aim with all your momentum, not to mention that you must lead the target.

Only Advanced Maneuvers Succeed: 

Only Advanced Maneuvers Succeed Instead, they must learn maneuvers like the Immelman (pictured here). Typically, only military buffs are motivated enough to study the manual and practice these maneuvers. No wonder they’ve become niche games. Lesson: it’s not enough for there to be a viable approach; the player needs enough feedback to discover it.

Players Need Feedback: 

Players Need Feedback In SimCity, why is nobody moving in to your city? The game provides lots of feedback displays for just this purpose: maps showing pollution levels, police andamp; fire coverage, traffic congestion, etc. Without these, the player is left to just guess. 'Maybe I need to increase desirability by adding a big park.' The game’s responses seem random. With those, you can say 'Oh, the crime and pollution from the industrial zones are spilling over into the residential zones. Next time I’ll separate them! Just one more game…'


Advisors? Complex simulation games (such as Civilization II shown here) can be overwhelming for newbies. One way to help is to provide advisors to suggest actions to take. But what will that do to the player’s approach? They’ll listen to the advisor. If there are multiple, conflicting advisors, how will they know which one to listen to? Just doing what an advisor says is turning a crank, and you don’t learn anything about why a given suggestion is considered best. So it’s not surprising that info screens are essential to all sim games, while advisors seem to come and go without contributing much to the gameplay.

Existing Gameplay:Dialog Trees: 

Existing Gameplay: Dialog Trees What do these principles say about existing games? Let’s look at adventure games, starting with the dialog trees. What’s the approach? Choose each item in turn No real penalty for wrong answer No strategy for recognizing correct one Even if, say, 'get them upset so they make a mistake' worked for one character, the other characters will be completely different Answer often non-obvious, to make game more 'difficult' Result is essentially the same as the 'choose your own adventure:' a combination puzzle

Existing Gameplay: Adventure Games: 

Existing Gameplay: Adventure Games Outside of dialog trees, the situation isn’t much better Sometimes the answer is straightforward (e.g. give the membership card to the bouncer) Sometimes out of the blue, i.e. use a towel to 'flick' someone in your way. Either way, there’s no way to learn from wrong solutions Last resort: apply every object to every other Ends up being the same as a combination puzzle

MUDs and MMORPGs: 

MUDs and MMORPGs MUDs and Massive Multiplayer games are the opposite First levels: learn interface, how to use weapons to kill easy creatures Once you get the hang of this you can reach level 5 where easy monsters don’t give much XP, but more powerful monsters are too difficult Forces you to form party, work on social skills As you progress, you’re always learning something new: spells, monsters, quests, parts of the world… These games are all learning curve These games are some of the most absorbing out there

Dialog With Internal Logic: 

Dialog With Internal Logic So how could you create a dialog system with rules underneath, and feedback to help the player figure them out? The otherwise unremarkable game Law and Order: Criminal Intent (2005), based on the TV show, provieds an interesting clue. In the dialog tree, instead of choosing the text you say, you choose the effect you want to have on the other person’s emotions E.g. flattering, empathetic, confrontational, etc. So what internal logic could these dialogs have?

Dialog with Internal Logic: 

Dialog with Internal Logic In this scene from the game, the bartender has just found out that one of last night’s patrons is dead. Goren asks 'Ever seen him before last night?' She responds, visibly shaken, 'No, he’s not one of the regulars. Look, this is getting a little heavy for me. OK? My girlfriend gave the cops information on a robbery lsat year and they made her testify and everything and it was awful. I want to help but, ummm… I don’t know.

Dialog With Internal Logic: 

Dialog With Internal Logic When someone wants to help but is afraid, should you be straightforward, confrontational, or empathetic? If you’re straightforward, they’ll still be hesitant and avoid your question If you’re confrontational, it’s even worse: they said they were afraid to help, and you’re making them more afraid! You need to be empathetic.

Dialog With Internal Logic: 

Dialog With Internal Logic The game doesn’t really communicate this logic to the player, and there’s not a lot of penalty for guessing wrong. But if game designers could find a way to communicate to the player that they’re affecting the character’s emotions, and that mixes with their temperament to produce their actions, I think dialog in games could have much more depth, and be a lot more fun than simply 'try every possibility.'

Save Anywhere vs. Save Points: 

Save Anywhere vs. Save Points Game design community is divided. Some feel 'save anywhere' is far superior to save points; others feel the opposite just as strongly. From other theories of game design, should be minor World is just as interactive, doesn’t change units, abilities, etc. Yet it affects learning, sometimes drastically One isn’t better or worse – they’re different Differ in how well skills must be learned Save anywhere: even 1 in 10 chance of clearing a screen means you can progress Save points: player must learn skills well enough to solve challenges regularly Each challenge can’t be as hard

Practical Steps: 

Practical Steps Next, some practical steps for game designers Because of too much involvement, [game designers] are unable to objectively comprehend how the actual players would feel when they play the game for the first time. - Shigeru Miyamoto, Nintendo Creator of Donkey Kong, Mario Bros. and Zelda

Practical Steps: 

Practical Steps Put range of approaches in design document At very least, forces designers to ask 'what’s the approach?' Kleenex Testing: play testing for design Use each play tester only once Team member watches but can’t say anything Done for Half-Life, settled many design arguments Sims 2 team swears by it Like user testing in HCI; verbal protocols in AI

Example: Space Invaders: 

Example: Space Invaders What might you see during such a test? On the next two slides are a hypothetical Kleenex Testing result It’s also a good example of the core learning dynamic

Example: Space Invaders: 

Example: Space Invaders Shoot blindly into crowd Few shots go between columns Slows firing rate  think how to avoid Pay attention to aiming, align with column Aligning difficult, so try for entire column at a time Invaders don’t stay still long enough, so move half way through

Example: Space Invaders: 

Example: Space Invaders Move back to column after dodging [Learning fine motor skills to position base] Soon, leading targets and trading off shooting vs. dodging Later: shoot hole through bunker Take out lower rows first Take out outside columns first Infrequent fast missiles: can’t dodge, so stay covered when not firing

Design in Space Invaders: 

Design in Space Invaders To other theories of game design, the points below are minor. But they make a big difference to how someone learns Space Invaders One shot on screen at a time  rewards aiming Worse aim  game slows down Better aim  game gets frantic Unlimited could lead to firing faster, not better Then forced to learn accuracy at higher levels, when also trying to dodge andamp; strategize Invaders shoot sparingly at first Player focuses on shooting columns before dodging, encouraging fine control

What’s the most fun AI?: 

What’s the most fun AI? A feature’s effect on gameplay: Not just how it changes the rules But how it changes approaches andamp; learning People agree AI should be fun, not smart Is a coordinated enemy more fun than mindless hordes? Mindless hordes are fun in e.g. Space Invaders, Doom Should units have morale? Without any further framework, discussion ends up on realism See, for example, the transcripts of the GDC roundtables


Realism In an RTS: Simulate physics of missiles? Could lead to defender micromanaging movement Whoever clicks fastest to move their defending units wins the game. Not much fun. AI in FPS: Learning? Shell Shock? Stress? Panic? Morale? Effect on approach essentially arbitrary Strategy games: Alliances? Must be understandable, not schitzophrenic AI logic transparent: lots of warning before a pact is broken, emissary explain why happy/mad Player must know how to exploit it


Summary When confronted with choices: Players develop theory of game world Come up with strategy/approach Apply approach Find strengths/weakness, always thinking how to improve it When it succeeds, they feel rewarded and competent Game is a journey: the path is at least as important as the outcome www.martincmartin.com

authorStream Live Help