FUNDAMENTALS OF GAME DESIGN, SECOND EDITION
Planning Phase
Armed with the list and sketch created in the first stage, you now start to plan the level in detail. Use pencil and paper to work out the sequence of events: both what you expect the player(s) to do and how the game will respond. Begin to document your decisions in the following key areas: gameplay, art, performance, and code requirements.
As you plan the gameplay for your level, you will need to consider all the following issues:
■ Layout (discussed extensively in the "Layouts" section earlier). Where can the player-controlled characters (avatar, party, or units) go and where can they not go? What paths can they use to get there? Many parts of your level may be cosmetic: The player can see them but cannot reach them.
■ Areas devoted to major challenges. Which areas carry strategic importance? Which will offer the biggest challenges? If the game involves combat, where would you like it to occur?
■ Pacing. How will the intensity of action vary throughout the level? Where will the key events and the rest periods occur?
■ Termination conditions. How does the player win or lose the level?
■ Resource placements. Are depots of weapons, health points, powerups, or any other resources hidden in the environment? Where? What resources, and how much?
■ Player start and end points. Do the player-controlled characters begin the level at one or more specific locations? Where? Do the characters end at one or more locations? Where?
■ NPC positions and spawn points. If NPCs—whether enemies, friends, or neutrals—appear in the level, where are they initially positioned? Can they suddenly appear in the level at a specific location or spawn point during play? Where?
■ Elevations. How much vertical movement does the level permit, and how does that affect play? Higher elevations naturally allow the player to see farther in first - and third-person perspectives; will this cause problems or constitute a positive feature of your level?
■ Secret areas. Do you plan to incorporate hidden areas or secret shortcuts? Where will they be, and what clues will be available to suggest they might be present?
■ Special event issues. What special events, unique to this level, can occur? Where will they occur? What will set them off? How do the special events reflect the setting and tone of the level?
■ Landmarks. How does the player find her way around? How can she tell where she is? Establishing major landmarks will help her out.
■ Destruction. Can any part of the level be destroyed or its landscape radically altered? Where does this happen and what causes it? How does it affect the game - play? Does it have the potential to introduce anomalies, such as enemies who wander off the edge of the world and never return?
■ Storytelling. How does the sequence of events the player experiences integrate with the game's story? Which events are dramatically meaningful and which are not? Where and when do you want cut-scenes or other narrative events to occur?
■ Save points and checkpoints. Does the level include save points or checkpoints? Where? In games in which the player fails frequently and has to reload, positioning the save points is a critically important part of balancing the game.
In the art planning phase, you determine the scope of your level and decide how much artwork it will need. Scope refers to the magnitude and complexity of the level, both in terms of the number of objects and characters that it contains and the special events that it includes. You can make a serious error by choosing too large a scope, because if you overload your art staff, you may never get the level finished at all. See "Get the Scope Right" near the end of this chapter.
You already have your sketch and a general idea of what the environment will be like, whether on the sea floor, in outer space, or inside an anthill. First decide on the scale of the level: How big will this level be in the game world's units of measure? This will help you to determine just how many other features the level needs. In almost every genre, if you've balanced the challenges correctly, the size of the level is directly proportional to the length of time that it takes the player to play through that level, so the scale you choose will, in a rough way, determine how much gameplay you can offer.
Next, start thinking about the kinds of objects that should be present in the level. Do research at the library or on the Internet for visual reference material to give you inspiration. Count the number of unique types of props that the level will require and plan in a general way where to put them. Certain generic items such as streetlights (or the infamous crates in first-person shooters) can simply be duplicated, but natural objects such as trees and boulders should come in several types, and the art team will need to know this. Try to avoid including too many identical objects in a level; it destroys realism.
Create a list of textures that the level will probably need. In an office, you may need tiles for the floor coverings, wood or metal for the desks, fabric for the chairs, and so on. Some offices may be streamlined, with severe geometric shapes, whereas others may be ornate, featuring a Louis XIV desk and antique chairs.
Decide on the visual appearance of any special effects that the artists will have to implement. It may take a while for the artists to come up with the visuals for a never-before-seen eruption of semisentient magma at zero gravity, so you need to plan ahead.
You normally think of performance as the programmers' problem, but it's up to the level designer not to build a world that bogs down the machine. You will need to sit down with the programmers and set some boundaries. How complex can the geometry be? How far into the distance will the graphics engine be able to render objects? How many autonomously moving units or creatures can the game support at one time? Know your machine's limitations as you plan your level.
Finally, as part of the planning process, identify specific requests that you intend to make of the programmers for features unique to this level. These may take the form of special events (sometimes called gags) that require coding, unique NPCs who appear only in this level but need their own behavior model and artificial intelligence, or special development tools you may require in order to build and test the level effectively. The more of these special coding problems you identify during planning and can discuss with the programmers in advance, the more likely that implementation will go smoothly.
Working through these steps results in an initial plan for the level. Don't expect the numbers and details in this plan to exactly match what you end up with in the finished level, but working out in advance as much as you can will ensure a smoother design process. Charging in without a plan and making it up as you go along creates more problems in the long run.
