"How do you come up with story and/or puzzle ideas for an adventure game?"
As the author of a series of successful adventure games, I'm often asked questions such as these. In this text, I've tried to answer both these questions, which are closely related and intertwined, to the best oif my ability. These are my views and observations regarding both the creative process (story development) and gaming success. Most of this stuff I had to learn the hard way, through trial and error, so I hope I can help someone avoid some of the difficulties I went through!
Coming up with a good story, puzzles, etc, is a process that is different for everyone. What works for me may not work for you. I can only tell you how I go about this creative process, and the things I take into consideration when designing an adventure.
A Few Basic Rules
I found a few basic rules to creating an entertaining, successful adventure game:
1. TESTING -- Test your game thoroughly before release! Have your friends play the game as often as possible, while you watch. This way you can find programming errors, and also find out what kinds of things the player will try to do. I had people test each world, or major section of world, as soon as it was ready, instead of waiting until the entire game was finished. Then when the game was finished, I had it tested again. The more complex a game is, the more important testing becomes. Nothing is more frustrating than playing a game that is full of bugs, especially if a bug makes the game impossible to finish.
2. LOGIC -- A good adventure must be logical, with puzzles that can be figured out by using a little common sense and deductive reasoning. And if there is only one solution to a puzzle, make sure that there is a good, logical reason why other solutions won't work.
3. RESPONSIVE -- Program as many possible responses to player commands as possible. Too many games suffer because there is only one thing the player can do in a given situation, and anything else the player tries to do fails to bring even a simple "You can't do that here" response from the game. This is where testing is essential. The programmer can try to imagine all the ways a player might try to solve a given puzzle, but until he sees someone else actually play it, he's still likely to miss something obvious.
4. PUZZLES, NOT BATTLES -- This has been a major problem in most World Builder games. Instead of having to solve puzzles, the player is confronted with a series of creatures that must be killed, usually by shooting over and over. This may be fine for some players, but for most it is boring and repetitive! Sure, you can have monsters and villians for the player to defeat, but make the player use his head. If you want to create a DOOM-style shoot 'em up, fine; but in an adventure game the player should be sharpening his wits, not exercising his trigger finger.
One more thing - avoid mazes. Most people consider them annoying and frustrating.
5. SUBPLOTS -- To make your game more interesting, give the player plenty to do. Besides the main plot, add a couple of minor subplots. Maybe put in some extra little puzzles for finding treasure or something too. Nonlinear plots are also good, allowing the player to solve puzzles according to his or her own pace and preference.
6. SPELLING -- Believe it or not, spelling counts! an occasional typo or misspelled word is one thing, but I've seen many shareware games that were ruined by horrid spelling and grammar.
7. EXPERIENCE -- Learn from your mistakes. If your first game doesn't do well, don't give up. Find out what people didn't like, what they did like, and make your next game better.
8. CONSISTENCY -- The commands used to play the game should be consistent. And the shape and position of the windows should also be consistent, changing only when absolutely necessary.
Designing The Worlds
First off, since my particular games involve travel to strange, even bizarre, worlds, I try to come up with the basic theme or characteristic of each world first. Sometimes I want to do a certain type of world as a result of my own experiences, such as the ghost town in A Mess O'Trouble, and the desert world in Ray's Maze. Sometimes I might have a sketch or idea for a building, creature or landscape that is unusual, and create a world based on that.
Mostly, I start by making a list of the kinds of worlds that I know I want to do, plus any others that might be interesting. When I was starting to design the concept for A Mess O'Trouble, I knew right away that I wanted an authentic western ghost town, like the ones I explore on my trips to Nevada. I also wanted to do something with world that is one huge living creature. And I wanted to do some more things with Void, a really weird place I'd used in both of my previous games. (The Void has almost unlimited possiblities.)
I didn't know what other worlds I'd do besides those. But I did have an idea for a puzzle that involved getting a large bird to carry you up to the top of a cliff or something. So I played around with that a little, and came up with the idea for a place like Monument Valley, in Arizona, with huge stone pillars. And some of the pillars had been hollowed out and carved into elegant, ancient cities.
I also wanted some places where the player would seem to be trapped, but could find a way out of with a little work. A small island seemed like a natural for this, so that was added to my list of worlds.
I ended up with a list of a few definite worlds, plus several possibles to work from. Of course, any one of these worlds could have easily been done as a complete game in itself. It isn't actually necessary to have more than one world (or time period) in a game. But for this particular game, that was what I wanted.
Basic Storylines
Then I started working on the plots and puzzles for the individual worlds. Since each world had a particular theme or peculiar feature, the story and plot for each world was often derived from that world's theme.
I tried to imagine what kinds of people (or aliens) would inhabit a particular world, and what kinds of problems they might have. I also tried to imagine what kind of puzzles might fit best in each world's theme.
First of all, each world would have one primary storyline, in the form of a multi-part puzzle to solve, with a few subplots woven in. Each main story or puzzle had to have some kind of motivator, something that provides a reason for the puzzles and compells the player to solve them.
In the ghost town, with its history of gold mining and "Old West" bandits, the motivator is simply the lure of lost treasure. Around this premise I created the story of a paranoid old miner who spent years hiding and booby-trapping his gold, with a subplot about a payroll robbery and unrecovered loot.
The motivator for the Voidbeast "world" was obvious: What could be more important to the inhabitants of a giant creature, than the health of that creature? So I came up with a story about a leech-like parasite that has attached itself to the Voidbeast and must be removed before it lays its eggs in the creature.
For one of the other worlds, I wanted to do a story that would eventually force the player to choose between two imprisoned creatures, one good, the other evil, based solely on their appearance. What drove this story was that the planet's sun was becoming unstable and would soon explode. Only the imprisoned "good" creature could save the day.
In the Enchanted Forest, I wanted to do some kind of story involving the Queen of the Faerie Folk, since this character was modeled after my wife. So in this case, the player finds her turned to stone, and must find a way to first communicate with her, and then eventually release her.
Beginning the Quest
Avoid simply telling the player "find this" or "do that", instead try to make the player's involvement in the story as natural as possible.
In each world, I tried to do something different. Not only a different story, but a different way for the player to become involved in each story. If the player is simply sent on a quest in each world, it becomes boring and repetitive. Having a task spelled out for you at the beginning of each world is not very interesting. It's usually more interesting if the player either is given only part of the story at the beginning, or finds it for himself. And in a game where the player encounters a different story in each world, you must find different ways to draw the player into it.
For example, on the island, the player meets no one, and is given no task, but soon discovers that he is trapped. As he explores his surroundings, he finds little clues. In this way the player is automatically drawn into the puzzle.
In the Enchanted Forest, the player encounters many interesting and unusual things, but will eventually find the Faerie Queen, turned to stone. It is then up to the player to find a way, using the available clues, to communicate with her. No one comes right out and says, "Find a way to communicate with the Faerie Queen." That would be too easy. It's much more exciting when the player can figure that out on his or her own.
Even in the Monolith World, where a character does send the player on a quest, the player must first gather clues and information from numerous sources before even meeting that character. And like real life, the "quest-giver" only knows part of the story -- the player finds more info later, which leads to a different conclusion than he was told to expect. In this way the gameplay feels more natural and less scripted.
Puzzles and Clues
Some worlds had special limitations. The ghost town is a good example of this:
There could be few people encountered there, else it wouldn't really be a ghost town. And since I wanted it to be authentic, the puzzles couldn't involve things like advanced technology, or magic, or monsters. And since this world was closest to the "real world", the player would have to find peaceful ways to get what he wanted too -- he couldn't just shoot someone who got in his way, or steal things. In the real world, that will end your adventure in a hurry!
Because the possible encounters with people were so limited, so were the possible ways of learning about the area (and the plot!)
But I knew from experience that old cabins often contained a log or journal, which held messages left by campers. Also, in the old days miners papered the walls of their cabins with old newspapers and magazines to keep out the cold in the winter. That would be another possible source of information.
And I decided to have one resident, an old miner living in a battered trailer. This was also true to form for an authentic ghost town.
Then I considered ways that local wildlife could be involved. I knew that packrats collected shiny and colorful objects, so it wouldn't be unreasonable to find gold nuggets in a packrat's nest. So I put a small nature poster in one cabin, as if it had been left behind by a camper, that told of the packrat's tendencies. The player could then offer a shiny bauble to a packrat, follow it to it's nest, and find some gold.
Rattlesnakes blocked access to a mine, and had to be dealt with in a non-lethal manner. Again, information about this animal's behavior was found in one of the cabins.
Other puzzles and storylines that came to mind included: following Indian pictographs to a hidden goal; finding treasure hidden by a paranoid miner; dealing with vandals; rescuing someone who'd fallen into a shaft; getting clues from the tombstones in the cemetary; and finding money stolen during a a payroll robbery, which had never been recovered.
When you run into that old creative block, list your resources. By resources, I mean the things available to you in your story's setting that can be used to provide plot twists, clues or tools for the player.
What kinds of people might the player meet?
What kinds of animals?
What kinds of objects might realistically be found in your world or scenario?
What kinds of businesses might be encountered?
What kinds of social activities might the locals engage in?
Are there any unusual characteristics of your world or its inhabitants?
Then think about how these things can be used to build a story or solve a puzzle. If necessary, brainstorm with someone else. Even if you don't use any of their ideas, they may set your mind of on a different direction, towards something you wouldn't have thought of otherwise.
Be observant of the real world too, that's where some of the best ideas come from. For instance, while changing the ribbon on my typewriter, I noticed how the words I'd typed appeared on the used ribbon. It then occurred to me that a player could find some clue by reading the words on an old typewriter ribbon in an adventure, just like a detective! To make it more interesting, suppose that only the correcting tape was available. In this case, the words would be backwards, and the player would have to "decode" it by re-writing the letters in the proper order.
This brings me to my next point:
Puzzle Design
Puzzles should usually require several steps to solve. The player should get a feeling of real accomplishment when a puzzle is solved. Each step of the puzzle can be a little puzzle in itself, providing a small thrill that builds to real excitement as the player completes the task.
When creating puzzles for my games, I usually start with the most basic form and work it from there. Say for instance the player has to get over a wall. Make a list of all the possible ways that a person could possibly get over, under or around a wall:
Climb over
Jump over
Fly over
Tunnel under
Break through
Then list some ways that a player could accomplish each of these methods. For instance, The player could climb over with a ladder, or a rope, or a pile of rocks, boxes or other objects. He could jump over with springs, a trampoline, a bouncy mattress, or maybe by mixing up a secret formula that provides super jumping abilities.
Now, you could just have him find a ladder, but that wouldn't be much fun. So, maybe he has to make a ladder. That adds a step to the puzzle. Suppose he has to buy the materials, or maybe find an abandoned lumber mill and cut the boards. That adds yet another step.
But...What if the player finds or builds a ladder, but it breaks? Now he has to find some other route. There are a number of possiblities. He could trick a large bird into carrying him, which could easily add a few more steps to the process. He could dig a tunnel, which might require the use of timbers for support (more steps!).
False leads and misdirection are useful too. Like the ladder that breaks. Or add some rocks; the player could pile them up, but they don't come up high enough to get over the wall. Putting in some fairly obvious solution that doesn't pan out adds to the fun, and makes the player think harder for the real solution.
In A Mess O'Trouble, the player begins the game by going through a "Jump Door", an interdimensional passage. The first place he lands is at the bottom of an old mine. A rope hangs down from the shaft, but the end is just out of reach. A tunnel runs to the north and to the south. At the north end, the player finds the tunnel has caved in, killing some earlier explorer. To the south, the tunnel ends, and ceiling of the tunnel is dangerously fragmented, supported by a single timber.
In this scenario, there are several false leads. The player can stand on a rock to try to reach the rope, but the rock isn't tall enough. When the player finally finds a way to reach the rope, it isn't securely tied, and falls, so the player can't use it to climb up. The player can dig out the cave-in, but it collapses again before he can pass through. And any attempt to climb up the shaft is futile. The only real escape is to pull the timber out from under the fractured ceiling at the south end, but that must be done with the rope or else the player would be crushed by the falling rock.
Solving the puzzle requires several steps: Finding a loose timber from the cave-in, using it to get the rope, tying the rope to the beam, and pulling out the beam to collapse the ceiling and create a new passage. And all the false leads add to the puzzle. If there were no false leads, the solution would be too easy to find, simply because it was the only possible thing to do.
Here's another example of how a simple puzzle can be broken down into a series of steps that enhance the puzzle: After passing through another Jump Door, the player lands on a small island of rock floating in the nothingness of the Void world. Nearby is an old building, slightly damaged. There is a door in the building -- this is the only exit. One corner of the building has been damaged, creating a gap the player can pass through into a small room.
In this case, the puzzle is to open the door. Now, the player could just find a key and unlock the door. But... what if the door is electronically controlled? That gives us more ways to add steps to the puzzle. The power is off, so the player has to find and activate the generator. And the numbered keypad is missing from the electronic lock, so the player has to find that too.
Inside the small room, there is a cabinet in one corner, and a panel on the wall. The panel opens easily, revealing a wheel, but the cabinet is locked. So now the player has to unlock the cabinet too. A search of the room turns up a small sliver of scrap metal which can be used to pick the lock. Inside the cabinet is the keypad, and a page torn from an encyclopedia. This page serves two purposes. It provides important imformation the player will need later in the game, and the page number is also the code number needed to open the door.
There is still the question of finding the generator, and also that wheel in the wall. Attempts to turn it are fruitless, so it too must be locked somehow. And what does it actuate? The only other object is the cabinet, the rest of the room is featureless. So...maybe there is something under or behind the cabinet? Yep, a panel in the floor. After throwing the latch on the panel, the wheel can now be used to open the panel.
Below the player finds a strange device, with a pair of tanks. This is the generator. The player must then experiment to find the right settings of hydrogen and oxygen to start up the generator.
Now the player has everything he needs to open the door. He has the keypad, the code number, and the power is turned on. So how can we add another step to the puzzle? How about a broken power cable in the generator room!
See how I've taken a simple task -- opening a door -- and added more and more steps to it, until it becomes a puzzle that really makes the player work for the solution? Everytime you have the player perform some task, ask yourself if there is some way to complicate it.
