Guidelines

Last updated: 12 may 1999

Difficulty level

The difficulty level should be adaptable to the player's skill. It is very frustrating if you have to play a game at a level that's too easy. However, especially for games with new concepts, there should be a couple of very easy levels. It's best to let the player select at any difficulty s/he chooses.

Scores and highscores

Keeping a score is important. A score is not just something to brag about, but is also a metric that can be used to monitor cognitive performance and long-term progress. Having at least one highscore is useful as a reminder during the current playing session. Having a table of highscores or latest scores that is saved on disk is useful for monitoring long-term progress.

Levels vs no levels

Levels will be defined here as relatively static configurations of the playing area that influence the game concept, such as platforms, rooms, mazes, and initial monster locations. Nowadays, games have many complex levels, which take a lot of time to design. In fact, one might even say that level design has largely replaced game design. There have been several games that use the same engine, but only have different levels. Also, new levelsets for specific games are often released to prolong their replayability. In contrast, early games have little in the way of levels.

Levels have a strong sightseeing element: seeing the next level is a motivation factor for playing a game. However, there is also a dark side to this aspect: familiar levels have little sightseeing value, and, when sightseeing was the primary motivation, they might be experienced as extra boring. For example, in some older games, I found it rather off-putting having to play all levels again in a fixed sequence. Playing them in random order would have been much better. In modern games, this has in part been remedied by the popular `save game' feature. If you have many large levels, having a level skip or save game is a necessity.

A related problem of levels is that they dictate the flow of the game: the more constraining the level, the more linear the gameplay will become. For example, in order to go to a specific room, you might have to kill specific monsters and get a specific key. In some cases, the most difficult spots of games (such as bosses) require very specific sequences of actions to navigate. The necessity to repeat such spots over and over might turn gameplay into a drag. I noticed that many people are talking about `non-linear' gameplay as a hot item: apparently, gameplay in current games is too linear. The solutions proposed are often very complex, and few succesful implementations of such solutions exist. It seems to me that the current inclination to design levels to such a high level of detail is simply shooting yourself in the foot as far as nonlinearity is concerned.

Perhaps a better option is to specify levels only up to a certain point, and fill in everything else by means of random generation, or even simply not to have levels.

Special effects

Special effects are representation features that do not influence the game concept. They are meant to enhance sightseeing experience, and are present in almost any game, including older games. Nowadays, a lot of effort goes into special effects. In fact, perhaps it is one of the main factors that made the design of many current games such a long and laborious process. In particular, detailed 3D object and scene rendering and graphics in general take a lot of effort. In older games, special effects were achieved in a much more efficient way. The most-often used effect is simply changing the background colour in each level. Extremely easy, and, if you choose some nice colours, very effective. Other well-used effects are: changing the shape of goodies or baddies (Squish'em, Fast Eddie, Pac-man), and various motion effects, such as 2-D or 3-D starfields, and moving 2-D or 3-D surfaces (Grid Runner, Juno 1st, P.O.D.). A personal favourite is colour cycling.

Since veteran players or players that are involved in hectic action don't spend much time doing detailed sightseeing, the old techniques may be as effective as the modern ones, and a whole lot easier, so you can spend more time on more relevant aspects of game design. If you wish the player to experience 3D motion, perhaps a starfield will do the trick. If you wish a variety of scenery, perhaps all you need to do is change a palette colour.

Also remember that, if you wish to suggest a high degree of realism, you have to go all the way, otherwise the experience is unconvincing. For example, if you use some photo-realistic bitmapped graphics, all graphics must be made photo-realistic, otherwise the other graphics look out of place. Worse, the graphics must be good, which requires the skills of a graphical artist, otherwise they are little improvement over simplistic graphics. If you choose to use only low-resolution graphics instead, you can somehow get away with it, and you don't need to be a graphical artist either.

2D vs 3D

It seems everyone is hyping first-person 3D. It appears that most games that are released today are 3D games. In reality however, even the games that sell best are mostly 2D games. Further reading:

Still, many people are jumping on the 3D bandwagon, while apparently they do not always realise what repercussions the use of a first-person 3D representation has on gameplay. In first-person perspective, you cannot see behind you, and you cannot obtain a clear view of the overall structure of your surroundings. This means that maintaining a view of your surroundings has become part of the game's tasks. For example, would one be able to survive level 32 of Robotron if one is only given a first-person 3D perspective?

Alan Shiu Ho Kwan about the disadvantages of some features of modern action games