Plan for Scott Shumaker
by Scott Shumaker · 05/15/2001 (4:54 am) · 2 comments
Well, I'm heading off to LA for E3 this year. This will be my second time attending; the first time was when it was held in Atlanta. That year in particular wasn't a very exciting one for gaming, but E3 was still a lot of fun. This year, with three major consoles on the verge of being released, I expect it will be awesome.
It'll be nice to catch up with old friends from my internship at EA, hang out with some of my buddies at Outrage, and meet some people in the industry who I've only spoken to online. Even though E3 is more for the marketing people than programmers and artists, it's still a good place to renew old contacts, just because so many people from the industry attend.
It's always amusing to see the huge quantities of hyped titles, many of which never make it to the shelves. I recall seeing a "booth babe" for the never-released Thrill Kill dressed up as a dominatrix. Of course, the game was riding entirely on violence and sex; Everyone who I spoke to who actually played the game agreed: it just wasn't a good game. That brings us to an important question: What metrics do we use to decide the quality of a game?
///////////////////////////////////////
I think most people would agree that fun is by far the most important factor for deciding whether or not a game is good. For all the emphasis on glitzy graphics and showy technology, the initial wow factor of a game wears off rather quickly. The only way to keep people playing is to make a game that engages, stimulates, and basically provides people with a continuous good time.
As an example, Tetris is perhaps one of the most simplistic games ever made from a technology standpoint. It was the flagship game on the Gameboy console, with 4 colors and a tiny screen. But you'll still find people playing the game today. It spawned a whole host of spin-offs, single-handedly revitalized the puzzle game genre, and made its licensees hundreds of millions of dollars. Hell, I still fire up it occasionally on my computer. The reason is simple: It's just a lot of fun to play (and addictive to boot, but that's a property of the fun factor).
The development of an enjoyable game is not something you can really plan on paper. Instead, it's an evolutionary process; even if your basic concept is sound, it will likely take a good deal of playtesting, tweaking, and modifying to get the game playable, let alone enjoyable. The amount depends greatly on the type of game, the scope, and the experience of the developers involved.
When you first start developing games, fun factor seems to arise spontaneously, almost magically, after making seemingly inconsequential modifications to the gameplay. One minute, it may be amusing but not really engaging; the next it may have suddenly turned into a helluva lot of fun to play. In my experience, this sudden emergence of fun tends to be pretty dramatic in action-based or puzzle-based games, and less so in strategy games or RPGs (but still impressive nonetheless).
It is hard to imagine why making small changes can have such a drastic effect on gameplay and enjoyment. But if you go ahead and analyze the consequences of your decision, usually you will discover you removed a barrier to fun rather than adding a feature that increased the fun of the game. This point may seem semantic, but the underlying issue is very important.
It's very difficult to delineate exactly what makes a game fun. Most of the true masterpieces of gaming (Tetris, Super Mario, etc.) are not that complicated from a gameplay standpoint. It's difficult to see what specific feature makes them fun; they just are all-around enjoyable to play, with no single aspect really standing out above the rest.
Fortunately, it is not so difficult to identify the factors that prevent games from being fun. If you can identify and remove them from your games, I can't guarantee you they will be enjoyable to play, but they'll certainly have the potential.
On that note, I've compiled a list of 4 significant yet prevalent barriers to fun. It's by no means comprehensive (it's just off the top of my head), but you can certainly consider it a list of things you don't want in your game.
1) Repetition
One of my personal pet peeves. Games are repetitive when the player is forced to perform essentially the same sequence of actions over and over again, with little or no variation. But this category can also apply to the game's art and music, when we have content that is basically all the same. Repetition will cause players to get bored and just stop playing.
For example, there are quite a few old games that were basically just mazes presented in simple first person graphics. These games typically have perfectly straight corridors with no variation whatsoever. How many of you would enjoy playing a game like that for more than a few minutes? This is not to say that mazes aren't necessarily enjoyable on paper, but usually they consist of more than straight corridors in a random arrangement; the popular paper-based ones are curvy and artistic. The ones we are used to seeing on computers have repetitive graphics and no gameplay variety.
One would hope that recent games wouldn't use the maze concept from a first-person perspective, but I can't tell you how many modern first person games are guilty here. It's acceptable if used sparingly or if the mazes have large architectural and gameplay variety, but this is not usually the case. Let's be honest; how many of you enjoy those endless sewer and ventilation duct mazes where every texture and wall looks the same?
Of course, just making the game constantly undergo radical changes in gameplay, art, or music is not the solution either. You need to have some continuity in your game. One of the best ways to accomplish while still reducing repetition is to encourage player choice (see #4).
2) Frustration
Player frustration is one of the worst possible emotions you want to elicit from those playing your game. A single frustrating sequence may make a player abandon an otherwise excellent game. Another posted mentioned the tentacle sequence in half-life. I managed to sweat through it, but it certainly wasn't fun. Perhaps the most frustrating sequence I've ever played was the root maze in Space Quest II (anyone else remember that one)? Thankfully, you could save it whenever you wanted, but it took 3 hours nonetheless.
Frustration is usually directly related to difficulty or repetition. If even a single segment is too hard and the player is forced to start from the beginning every time, you can bet there will be a lot of players that just quit playing the game in disgust after a few failed attempts. This issue is compounded if there is a lot of time and effort invested that must be entirely redone if the player fails the sequence.
A second type of frustration is caused when you don't meet the player's expectations. For example, in a shooter, most players expect their ship to only actually collide with enemies that they appear to touch on the screen. So in a shooter with overaggressive collision detection, they will be understandably frustrated when their ship takes damage if it doesn't appear like it should, regardless of what the underlying collision detection algorithm says. Another example of this is breaking expectations built up within the game. If you're in a platformer and you can run on regular blocks but lava kills your character instantly, players will quickly learn to avoid the lava. But if suddenly midway through the game you make it so the player's character dies from touching the regular blocks without any advance warning that the rules have been changed, players will view this as a violation of trust and get angry.
In general, if you hear players exclaim, "This is cheap!" whenever they die or the opponent scores, you are probably not meeting the player's expectations. When in doubt, always err on the side of the player. So in the shooter example, bias collision detection slightly in favor of the player (maybe make the player's ship's bounding rectangle smaller than it appears on screen). And in the platformer, you could get away with making a certain patch of lava not damage the player. Players don't mind games that are unfair, as long as they are unfair in their favor. :)
Another surefire way to frustrate players is to make them perform meaningless tasks not related to the main goal. If the control scheme or UI is unnecessarily difficult, players will get frustrated because they're fighting the interface, not the game itself, which is obviously not the reason they're playing the game.
This point is just as relevant for in-game tasks. If you play any of the Zelda games, you'll notice that as soon as you defeat the boss of the dungeon, you are immediately transported outside of the dungeon. Can you imagine what it would be like if you had to fight your way back out of the dungeon? And what if you died on the way back out and had to fight the boss all over again? Now you have a meaningless task combined with repetition, and you even bring in my next principle to boot:
3) Inappropriate rewards
There are many different types of rewards, such as advancement, bonus stages, secrets and Easter eggs, cinematic sequences, and high score tables. Rewards are important because they keep players playing, add replay value, and just generally make the game a lot more fun. Virtually every great game has some sort of rewards sequence. Even Tetris for the Gameboy had animated sequences on Level 9 mode B, as well as a high score table. The original Super Mario Brothers rewarded clever and lucky players with warp zones, extra lives, and bonus points if you could land on the top of flag pole or finish with specific level times.
Some rewards are given out for basic competency (beating the levels), and others are given out to advanced players who can master the more difficult game maneuvers. Make sure that you have both types in your game; the basic rewards keep players going, while the more advanced ones give them the motivation to spend the time to master your game.
But ensuring that you have rewards in place is not enough. Many games also make the mistake of either handing out rewards that are not commensurate with the achievements that earned them, or handing out rewards that are not consistent with previous awards. As mentioned under frustration, if you defy the players' expectations you can expect them to get very angry. Here's a link to a very good article on rewards and expectation levels at Gamasutra.
4) Lack of choices
This topic has a lot in common with repetition. Certainly, severely limiting player choice will cause very repetitive gameplay. But it's enough of a different topic to warrant its own section.
When I speak of choice, I'm really talking about meaningful choices. The decision of a player to press the up button isn't really a choice if it causes them to die every time. From another viewpoint, if pressing the up button causes the exact same result as pressing the down button, this isn't really a choice either. Both these examples may seem contrived, but who hasn't played a game where:
1) You are faced with two paths, one leading to inescapable death, the other leading to the exit
2) You have the option to equip your character with one of two items, both with identical effects (or, optionally, two items, one superior to the other in every way).
I argue that neither of these are really a choice, or at least not an interesting one.
Now, to describe the results of player choices, you really have to partition up player choices into temporal categories.
At the lowest level, you have instantaneous choices, like "should I press this button or this button?" Almost all games allow you multiple instantaneous choices. One notable exception is Dragon's Lair. If you press any buttons other than the required ones and you don't press them at the correct moment, your character dies. Not much of a choice here.
Since most other games do allow this type of choice, this isn't really the category I'm interested in discussing.
Now, at the other extreme, we have choices that affect the final outcome of the game. Some games allow these types of choices. It is appropriate in particular for story-driven games (RPGs, etc.), where decisions made by the player should logically affect the game ending. Since these types of decisions are only appropriate for certain types of games, they are not the interesting category either.
In between these two extremes we have what I like to call short-term and medium-term decisions. These are really the important ones for most games, and they are also unfortunately the ones that many mediocre games neglect. To use an analogy from literature, medium-term choices are what you would call the decisions responsible for the scenes in a novel. Does the detective investigate the old warehouse or have dinner with the wealthy widow? Short-term decisions are what happen within those scenes. How does the detective get past the security at the warehouse docks? Does he survive his brush with the mysterious gunman lying in wait inside? The distinction can be fuzzy at times, but usually you can draw the line. In Super Mario Brothers, the 'scenes' are the levels (1-4, 3-1, etc.), so medium-term decisions are which levels to visit, while short-term decisions are how the player advances through the level. Super Mario Brothers 1 doesn't offer an extraordinary amount of medium-term freedom, but does offer some in the form of warp zones. The later Super Mario Brothers 3 and Super Mario World allow a great deal of choice in the medium term, letting the player navigate between levels with through an overland map.
Short-term decisions are even more important, and it's here that many games fail. An excellent game absolutely must allow multiple player choices in the short-term. In other words, there should be either multiple solutions to each problem or multiple problems to choose from. For example, if you play a Super Mario Brothers 3 level, you are confronted with many ways to solve the level. Enter the sewer area? Fly up to the clouds above and skip the level below? Equip the raccoon suit or the fire flower? Collect coins to earn an extra life, or race through the level for a time bonus? In other games, instead of multiple paths through each level, you can choose different approaches to each task. In Fallout, there was one scene where force fields blocked your path, and you could get around them by hacking into the computer and disabling them, dropping dynamite near them to take out the generators, or even bluffing your way past them by talking to guards. What was especially cool was that your options were dependent to a good extent on actions you had taken earlier in the game. Some games even combine both; in Deus Ex, you'll often find several different entrances to enemy installations, each replete with their own set of challenges; puzzles in each route that can be solved in several different ways; one can even take entirely different approaches to the whole mission (go in guns blazing or the sneak around and get out).
Nintendo's games in particular excel in this area. Reflect for a moment on their major franchises: Mario, Zelda, and Metroid. In each case you'll find that the main character has abilities that change throughout the course of the game, adding new moves to master and more player choices. Mario gets the cape, fire flower, and Yoshi; Link gets arrows, bombs, and the ability to transform; Samus gets the ice shot and screw attack. All of these additions make the games far more interesting by giving players more options.
Developers of fighting games realized this a long time ago, which is why every character you will see in modern fighting games now have dozens of different moves, whose usefulness depends on the circumstance.
If you don't give players plenty of variety and frequent opportunities for decision-making, you will likely fall into the trap of making a very linear and uninteresting game.
It'll be nice to catch up with old friends from my internship at EA, hang out with some of my buddies at Outrage, and meet some people in the industry who I've only spoken to online. Even though E3 is more for the marketing people than programmers and artists, it's still a good place to renew old contacts, just because so many people from the industry attend.
It's always amusing to see the huge quantities of hyped titles, many of which never make it to the shelves. I recall seeing a "booth babe" for the never-released Thrill Kill dressed up as a dominatrix. Of course, the game was riding entirely on violence and sex; Everyone who I spoke to who actually played the game agreed: it just wasn't a good game. That brings us to an important question: What metrics do we use to decide the quality of a game?
///////////////////////////////////////
I think most people would agree that fun is by far the most important factor for deciding whether or not a game is good. For all the emphasis on glitzy graphics and showy technology, the initial wow factor of a game wears off rather quickly. The only way to keep people playing is to make a game that engages, stimulates, and basically provides people with a continuous good time.
As an example, Tetris is perhaps one of the most simplistic games ever made from a technology standpoint. It was the flagship game on the Gameboy console, with 4 colors and a tiny screen. But you'll still find people playing the game today. It spawned a whole host of spin-offs, single-handedly revitalized the puzzle game genre, and made its licensees hundreds of millions of dollars. Hell, I still fire up it occasionally on my computer. The reason is simple: It's just a lot of fun to play (and addictive to boot, but that's a property of the fun factor).
The development of an enjoyable game is not something you can really plan on paper. Instead, it's an evolutionary process; even if your basic concept is sound, it will likely take a good deal of playtesting, tweaking, and modifying to get the game playable, let alone enjoyable. The amount depends greatly on the type of game, the scope, and the experience of the developers involved.
When you first start developing games, fun factor seems to arise spontaneously, almost magically, after making seemingly inconsequential modifications to the gameplay. One minute, it may be amusing but not really engaging; the next it may have suddenly turned into a helluva lot of fun to play. In my experience, this sudden emergence of fun tends to be pretty dramatic in action-based or puzzle-based games, and less so in strategy games or RPGs (but still impressive nonetheless).
It is hard to imagine why making small changes can have such a drastic effect on gameplay and enjoyment. But if you go ahead and analyze the consequences of your decision, usually you will discover you removed a barrier to fun rather than adding a feature that increased the fun of the game. This point may seem semantic, but the underlying issue is very important.
It's very difficult to delineate exactly what makes a game fun. Most of the true masterpieces of gaming (Tetris, Super Mario, etc.) are not that complicated from a gameplay standpoint. It's difficult to see what specific feature makes them fun; they just are all-around enjoyable to play, with no single aspect really standing out above the rest.
Fortunately, it is not so difficult to identify the factors that prevent games from being fun. If you can identify and remove them from your games, I can't guarantee you they will be enjoyable to play, but they'll certainly have the potential.
On that note, I've compiled a list of 4 significant yet prevalent barriers to fun. It's by no means comprehensive (it's just off the top of my head), but you can certainly consider it a list of things you don't want in your game.
1) Repetition
One of my personal pet peeves. Games are repetitive when the player is forced to perform essentially the same sequence of actions over and over again, with little or no variation. But this category can also apply to the game's art and music, when we have content that is basically all the same. Repetition will cause players to get bored and just stop playing.
For example, there are quite a few old games that were basically just mazes presented in simple first person graphics. These games typically have perfectly straight corridors with no variation whatsoever. How many of you would enjoy playing a game like that for more than a few minutes? This is not to say that mazes aren't necessarily enjoyable on paper, but usually they consist of more than straight corridors in a random arrangement; the popular paper-based ones are curvy and artistic. The ones we are used to seeing on computers have repetitive graphics and no gameplay variety.
One would hope that recent games wouldn't use the maze concept from a first-person perspective, but I can't tell you how many modern first person games are guilty here. It's acceptable if used sparingly or if the mazes have large architectural and gameplay variety, but this is not usually the case. Let's be honest; how many of you enjoy those endless sewer and ventilation duct mazes where every texture and wall looks the same?
Of course, just making the game constantly undergo radical changes in gameplay, art, or music is not the solution either. You need to have some continuity in your game. One of the best ways to accomplish while still reducing repetition is to encourage player choice (see #4).
2) Frustration
Player frustration is one of the worst possible emotions you want to elicit from those playing your game. A single frustrating sequence may make a player abandon an otherwise excellent game. Another posted mentioned the tentacle sequence in half-life. I managed to sweat through it, but it certainly wasn't fun. Perhaps the most frustrating sequence I've ever played was the root maze in Space Quest II (anyone else remember that one)? Thankfully, you could save it whenever you wanted, but it took 3 hours nonetheless.
Frustration is usually directly related to difficulty or repetition. If even a single segment is too hard and the player is forced to start from the beginning every time, you can bet there will be a lot of players that just quit playing the game in disgust after a few failed attempts. This issue is compounded if there is a lot of time and effort invested that must be entirely redone if the player fails the sequence.
A second type of frustration is caused when you don't meet the player's expectations. For example, in a shooter, most players expect their ship to only actually collide with enemies that they appear to touch on the screen. So in a shooter with overaggressive collision detection, they will be understandably frustrated when their ship takes damage if it doesn't appear like it should, regardless of what the underlying collision detection algorithm says. Another example of this is breaking expectations built up within the game. If you're in a platformer and you can run on regular blocks but lava kills your character instantly, players will quickly learn to avoid the lava. But if suddenly midway through the game you make it so the player's character dies from touching the regular blocks without any advance warning that the rules have been changed, players will view this as a violation of trust and get angry.
In general, if you hear players exclaim, "This is cheap!" whenever they die or the opponent scores, you are probably not meeting the player's expectations. When in doubt, always err on the side of the player. So in the shooter example, bias collision detection slightly in favor of the player (maybe make the player's ship's bounding rectangle smaller than it appears on screen). And in the platformer, you could get away with making a certain patch of lava not damage the player. Players don't mind games that are unfair, as long as they are unfair in their favor. :)
Another surefire way to frustrate players is to make them perform meaningless tasks not related to the main goal. If the control scheme or UI is unnecessarily difficult, players will get frustrated because they're fighting the interface, not the game itself, which is obviously not the reason they're playing the game.
This point is just as relevant for in-game tasks. If you play any of the Zelda games, you'll notice that as soon as you defeat the boss of the dungeon, you are immediately transported outside of the dungeon. Can you imagine what it would be like if you had to fight your way back out of the dungeon? And what if you died on the way back out and had to fight the boss all over again? Now you have a meaningless task combined with repetition, and you even bring in my next principle to boot:
3) Inappropriate rewards
There are many different types of rewards, such as advancement, bonus stages, secrets and Easter eggs, cinematic sequences, and high score tables. Rewards are important because they keep players playing, add replay value, and just generally make the game a lot more fun. Virtually every great game has some sort of rewards sequence. Even Tetris for the Gameboy had animated sequences on Level 9 mode B, as well as a high score table. The original Super Mario Brothers rewarded clever and lucky players with warp zones, extra lives, and bonus points if you could land on the top of flag pole or finish with specific level times.
Some rewards are given out for basic competency (beating the levels), and others are given out to advanced players who can master the more difficult game maneuvers. Make sure that you have both types in your game; the basic rewards keep players going, while the more advanced ones give them the motivation to spend the time to master your game.
But ensuring that you have rewards in place is not enough. Many games also make the mistake of either handing out rewards that are not commensurate with the achievements that earned them, or handing out rewards that are not consistent with previous awards. As mentioned under frustration, if you defy the players' expectations you can expect them to get very angry. Here's a link to a very good article on rewards and expectation levels at Gamasutra.
4) Lack of choices
This topic has a lot in common with repetition. Certainly, severely limiting player choice will cause very repetitive gameplay. But it's enough of a different topic to warrant its own section.
When I speak of choice, I'm really talking about meaningful choices. The decision of a player to press the up button isn't really a choice if it causes them to die every time. From another viewpoint, if pressing the up button causes the exact same result as pressing the down button, this isn't really a choice either. Both these examples may seem contrived, but who hasn't played a game where:
1) You are faced with two paths, one leading to inescapable death, the other leading to the exit
2) You have the option to equip your character with one of two items, both with identical effects (or, optionally, two items, one superior to the other in every way).
I argue that neither of these are really a choice, or at least not an interesting one.
Now, to describe the results of player choices, you really have to partition up player choices into temporal categories.
At the lowest level, you have instantaneous choices, like "should I press this button or this button?" Almost all games allow you multiple instantaneous choices. One notable exception is Dragon's Lair. If you press any buttons other than the required ones and you don't press them at the correct moment, your character dies. Not much of a choice here.
Since most other games do allow this type of choice, this isn't really the category I'm interested in discussing.
Now, at the other extreme, we have choices that affect the final outcome of the game. Some games allow these types of choices. It is appropriate in particular for story-driven games (RPGs, etc.), where decisions made by the player should logically affect the game ending. Since these types of decisions are only appropriate for certain types of games, they are not the interesting category either.
In between these two extremes we have what I like to call short-term and medium-term decisions. These are really the important ones for most games, and they are also unfortunately the ones that many mediocre games neglect. To use an analogy from literature, medium-term choices are what you would call the decisions responsible for the scenes in a novel. Does the detective investigate the old warehouse or have dinner with the wealthy widow? Short-term decisions are what happen within those scenes. How does the detective get past the security at the warehouse docks? Does he survive his brush with the mysterious gunman lying in wait inside? The distinction can be fuzzy at times, but usually you can draw the line. In Super Mario Brothers, the 'scenes' are the levels (1-4, 3-1, etc.), so medium-term decisions are which levels to visit, while short-term decisions are how the player advances through the level. Super Mario Brothers 1 doesn't offer an extraordinary amount of medium-term freedom, but does offer some in the form of warp zones. The later Super Mario Brothers 3 and Super Mario World allow a great deal of choice in the medium term, letting the player navigate between levels with through an overland map.
Short-term decisions are even more important, and it's here that many games fail. An excellent game absolutely must allow multiple player choices in the short-term. In other words, there should be either multiple solutions to each problem or multiple problems to choose from. For example, if you play a Super Mario Brothers 3 level, you are confronted with many ways to solve the level. Enter the sewer area? Fly up to the clouds above and skip the level below? Equip the raccoon suit or the fire flower? Collect coins to earn an extra life, or race through the level for a time bonus? In other games, instead of multiple paths through each level, you can choose different approaches to each task. In Fallout, there was one scene where force fields blocked your path, and you could get around them by hacking into the computer and disabling them, dropping dynamite near them to take out the generators, or even bluffing your way past them by talking to guards. What was especially cool was that your options were dependent to a good extent on actions you had taken earlier in the game. Some games even combine both; in Deus Ex, you'll often find several different entrances to enemy installations, each replete with their own set of challenges; puzzles in each route that can be solved in several different ways; one can even take entirely different approaches to the whole mission (go in guns blazing or the sneak around and get out).
Nintendo's games in particular excel in this area. Reflect for a moment on their major franchises: Mario, Zelda, and Metroid. In each case you'll find that the main character has abilities that change throughout the course of the game, adding new moves to master and more player choices. Mario gets the cape, fire flower, and Yoshi; Link gets arrows, bombs, and the ability to transform; Samus gets the ice shot and screw attack. All of these additions make the games far more interesting by giving players more options.
Developers of fighting games realized this a long time ago, which is why every character you will see in modern fighting games now have dozens of different moves, whose usefulness depends on the circumstance.
If you don't give players plenty of variety and frequent opportunities for decision-making, you will likely fall into the trap of making a very linear and uninteresting game.
About the author
Associate Matt Fairfax
PopCap