Elements of Good Levels

As you design a level, there are a seemingly infinite number of details you must keep in mind. You must be concerned that you balance the elements of action, exploration, puzzle solving, storytelling, and aesthetic appeal . You must work with the artists and programmers to achieve the effects you want. For 3D levels, you must make sure the whole level is optimized so that it can run on the target system. And in the worst of situations you have to deal with unruly level design tools that seem to thwart your every attempt to make something cool.

Often a level designer will come up with rules of thumb to follow while making a level, even if she does not write them down. Every designer will have her own list of dos and don ts that she keeps in the back of her mind, and this list can change significantly from project to project. Some games will have their own design rules established ahead of time and which the designers can then follow, but there are also rules that can apply to any project. Here I present a partial list of my own rules, which I use to attempt to make a level that is stimulating to play.

Players Cannot Get Stuck

This should be obvious but is a frequent newbie designer mistake. Players should never become hopelessly stuck when playing your level. There should be no pits that can be fallen into but not climbed out of, no objects which, when moved incorrectly, permanently block the players progress, and no doors that fail to open if players approach them a certain way. Though this goal may seem perfectly obvious, it will actually consume a large amount of your time as a level designer. Consider a puzzle where players have a certain amount of dynamite, and that dynamite needs to be used to blow a hole in a wall so players can progress in the level. What if players use up all their dynamite blowing up the wrong things? Without any more dynamite, the players are now completely stuck. Similarly, suppose players need to talk to a particular NPC to get a particular object. What if, instead of talking to that character, players kill him? Either the game must end nearly instantly, or there must be some alternate way to progress through the game. Designing your level in such a way that, whatever players do, they can still finish the level, takes a lot of thinking and planning. As a level designer, you must always be asking yourself, But what if players try it this way?

Sub-Goals

As the players play a level, they should have understandable sub-goals. Instead of playing through the whole level just trying to get to the exit or accomplish some large goal, players should be able to recognize that there are various tasks they can accomplish that contribute to the final goal. A very simple example of this would be the different keys in Doom . Players know that once they get the blue key they are that much closer to finishing the level. In an arcade racing game like San Francisco Rush , instead of having just one finish line per track, most games have multiple checkpoints along the track at which players are given a time bonus and informed of how well they are doing. In an RPG, players may be working to defeat an evil force that is tormenting the land, but along the way they are able to go on various sub-quests for villagers who need their help. These various sub-quests lead players toward the larger goal, and provide players with positive feedback that they are, in fact, playing the game well. Platformers like Ratchet & Clank are particularly good at leading players through the levels with their many pickups, with the acquisition of each basically being a tiny sub-goal. A sub-goal is useless if players do not understand what they have accomplished. Therefore, it is also important to provide players with some sort of reward for achieving the goal, whether it is audiovisual bells and whistles, a new weapon, bonus points, or more time on the racing clock. If the designer does not provide enough sub-goals on a particular level or if those sub-goals are so transparent that the players do not realize they have achieved them, players may become confused as to what they are supposed to be doing and whether they are getting any closer to succeeding.

In racing games such as the San Francisco Rush series, players are given sub-goals through checkpoints, which award more time. Pictured here: San Francisco Rush: The Rock ” Alcatraz Edition .

Landmarks

The more complex your level, the more players are likely to get confused navigating it. Unless confusion is your goal, which it usually should not be, it is a good idea to set up memorable landmarks in your level to ease the players exploration. A landmark is any unique object in your level that players will recognize the second time they see it, whether it is a particularly ornately decorated room, a large statue, or a steaming pool of lava. In terms of exploration, then, when players return to this landmark, they will know that they are returning to a location they have previously visited, and will thereby begin to understand the layout of the level. Landmarks do not necessarily need to be big red signs labeled Checkpoint A, but can instead be worked into the story and setting of the level itself. For example, Grand Theft Auto III made the very large Liberty City much easier to learn to navigate by including lots of large and unique buildings as part of the map. Each of these landmarks, in addition to how they helped the player learn the level, supported the game fiction perfectly.

Critical Path

Even though I am a big proponent of non-linear gameplay, I am also a big fan of a nice critical path in a level. A critical path gives players a sense of a direction they can use in order to complete the level. This direction may be a physical direction, such as head north or head for the rainbow, or it can be a more ambiguous goal, such as finding a creature and defeating it or retrieving an important object. Always giving players a primary goal to accomplish is crucial to making your level playable . Players should have a goal and, as I discussed, sub-goals that work toward achieving that primary goal. Players should always be aware of the goal and the related sub-goals, and should always have a sense of what they can do to progress in the level. Separate optional side-goals may be less obvious or hidden, but nothing frustrates players more than having no idea what they are supposed to do. Having a clearly established critical path is a good way to help prevent players from becoming confused.

Limited Backtracking

If your game relies on exploration for a large part of its gameplay value, it is probably a bad idea to make players backtrack through large sections of the level that they have already explored in order to continue in the game. That is not to say that your level cannot have branching paths for players to explore. It merely means that each branch should loop back to the main path without players needing to backtrack along the same path. If your game is more of a role-playing or adventure game where creating the illusion of reality is important, the necessity of backtracking may be more acceptable. Grand Theft Auto III is certainly an example of a realistic setting resulting in a need for backtracking, though the game ameliorates the situation by making driving around the city fun no matter how many times you do it. Certainly in an RTS, sports, or death-match game, players will be covering the same ground over and over again, but the appeal of a basketball game or WarCraft is not so tied to exploration as Super Mario 64 , which does a very good job of eliminating the need for backtracking entirely.

Success the First Time

If most players are able to beat your level the first time they play it, you have probably made a level that is too easy. Nonetheless, the possibility should exist that players could make it all the way through your level on the first try. I do not mean, however, that players could make all of the right choices just by happenstance. Instead, you should provide enough data to the players that they have a reasonable chance of avoiding all the obstacles put in their path if they are observant and quick-witted enough.

Whenever players fail in your level, they should feel that they had a fair chance of avoiding that failure if they had only been more observant or had thought more before they acted. Nothing frustrates players more than realizing that the only way to make it through the level is by trial and error combined with blind luck. Of course, your level can still be hard. Your clues as to what to do can be quite subtle, the monsters to be defeated can be really strong, or the choices to be made can be truly challenging, but if players do everything perfectly, they should be able to get through your level the first time they play it.

Navigable Areas Clearly Marked

Players should have a clear idea of where they will be able to go in the level simply by looking at the environment. Slopes that players will slide on should appear to be significantly steeper than the slopes that can be walked on. Textures may be used to differentiate between areas players can navigate and those they cannot. It can be very frustrating to players when an area that appeared to be unnavigable turns out to be the only way out of a particular area. Another example might be a room with ten doors in it. Players try three of these doors, and they are all locked. At this point, players will probably conclude that the doors are there only for show and will stop trying any of the other doors. No information, whether visual or through a verbal clue, is given to players to indicate that the other doors might be openable when the first three they tried were not. If it turns out that the only way out of this room is through the sole unlocked door, I would suggest that this area has been poorly designed. The only way out of such a room is through tedious trial and error. The fun in a game may involve trying to get to certain areas or the thrill of running around in those areas, but there is little fun to be found in determining which areas the designer arbitrarily decided could be navigated and which could not.

Choices

This may seem obvious, but choices are something level designers can often forget to keep in mind as they are building their levels. Good levels give players choices of how to accomplish goals, just as good gameplay gives players lots of choices for how they will play the game. Choices do not necessarily mean multiple paths through a level, though that may be a good idea as well. In a first-person shooter, choices could mean giving players different options for how to take out all of the enemies in a room ” plenty of different places to hide, different locations that the enemies can be shot from, and so forth. Such a setup creates a variety of different strategies that will successfully defeat the horde of advancing demons. Choices could also mean bonus objects that are challenging for players to get, such as a rocket launcher in the middle of a pool of lava ” players have the choice to risk going for it or not. In a strategy game, interesting choices mean different places where battles may play out or different places players can choose to rally their troops or gather resources. In adventure games, the genre most notorious for not giving players enough options, choices mean multiple solutions to the game s puzzles, different characters to talk to, and plenty of different ways to move through the game. Indeed if a designer is going to add choices to her levels, it is important that she make sure she is adding interesting choices. The decision to go left or right around a pillar is a choice, but if both lead to the same place and produce basically the same experience the choice is not terribly interesting. If one side of the pillar is on fire while the other side has an intimidating thug guarding it, the choice is more interesting. Players become frustrated when they feel that they are locked into just one way of playing the game, especially if that one way is not the way they would like to play it.

A Personal List

Certainly the list I have provided above is far from complete. As you work as a level designer, it makes sense to establish your own list of design goals to keep in mind while creating your level. As you work on levels that are received well by your peers or players, try to analyze the levels to see what you did well. Then try to abstract these accomplishments into a list of goals to keep in mind as you work on subsequent levels. This list does not necessarily need to be formally written down; just keeping a mental checklist may be sufficient. The options I listed here may be a start for your own list, or you may find yourself coming up with a completely different set of goals. Every designer approaches level design in her own way.