Introduction

The rules that govern single player level design are becoming more and more well known. They help make sure that the gamer’s experience is controlled in terms of difficulty, rhythm, renewal etc. But multiplayer level design does not follow the same constraints as those of single player level design. I will start by describing the specific constraints of multiplayer level design.

Technical Constraints

The first technical constraint is the infamous bandwidth bottleneck. A game machine may be high-powered and capable of processing a huge amount of information, but if the “pipe” that links it to other machines is too narrow, little information can be exchanged and the game is therefore slowed down or impoverished.

What are the main points of a multiplayer game that eats up bandwidth? First there is character movement and animations. In most multiplayer FPS games, character animation is very limited. What characters do most often is run, jump or crouch. But in games such as the multiplayer version of Splinter Cell, the wealth of animations is at the heart of the game.

Characters can grab each other, perform acrobatics, hit each other etc. This wealth of animations is very demanding in terms of bandwidth. I am convinced that such a high quality of animations will become more and more present in future multiplayer games, not only because they provide more realism, but also because they enlarge gameplay possibilities, as it may be noticed when playing Splinter Cell in versus mode.

One of the numerous complex animations available in the multiplayer version of Splinter Cell – Pandora Tomorrow and Chaos Theory

Other large bandwidth consumers are special effects, such as the explosions with their particle display, and the dynamic events in the levels. The latter are map animations, such as the movement of a crane or of an elevator or the destruction of a wall. If it is possible to interact with such events (by throwing a grenade into an elevator or by positioning a character on a moving object), the exact position of these moving parts of the map must be followed image by image. Once again, I am convinced that maps of future multiplayer games will contain more and more special effects and map animations.

Special effects in a game are not only used for cosmetic purposes. They may be used to disturb vision, darken or light up the environment, leave footprints etc. Their use in terms of gameplay is obvious.

The same is true for map events. We have widely used such events in the multiplayer maps of Splinter Cell – Chaos Theory. For example, in the Aquarius map, the level designer positioned a destructible ventilation grille near a locked entry, with one of the map’s mission objectives behind (see the illustration below). If a defender uses a grenade close to this grille to kill an opponent, he destroys the grille and thus provides the latter with a quick route to his objective.

In this way, the constructive use of map events helps bring a new dimension to the gameplay of the map by changing both its layout and the defenders’ tactics. The maps of Splinter Cell – Chaos Theory and certain maps of Splinter Cell – Pandora Tomorrow are full of such interactions and use of special effects for gameplay purposes.

The amount of information exchanged among machines during a game session is proportional to the number of players. One of the direct consequences of the bandwidth bottleneck is the decision that has to be made regarding the number of players and the richness of the game (animations, special effects, map events etc.).

A destructible ventilation grille in Aquarius, one of the multiplayer maps in
Splinter Cell – Chaos Theory

The second technical constraint is the need to synchronize the events of the game. In the case of action games, the synchronization of the key events, such as shooting, is essential for the quality of the game. Synchronization problems account for the main cause of dissatisfaction among gamers. When a good player places the cursor on a moving target and pulls the trigger, he expects the bullet to reach the target immediately and exactly where he aimed at. A half-second delay is unacceptable. Good players are very precise in their game and feel “cheated” if the game does not react as quickly as they do. Some games, such as Soldier of Fortune on Xbox, handle this problem very efficiently

The third technical constraint is the lack of control on what the players will do. If too many players are found in the same place on the map and start generating explosions and causing many map events, all concentrated in the same area the amount of information exchanged between machines becomes too large and the refresh rate of the images plummets.

This problem does not occur in single player games, because the level designer would have made sure that the events which are likely to slow down the machine are evenly distributed. But in the multiplayer level, everything is possible. It is difficult to prevent the players from willingly “stuttering” their game session by overloading the CPU and the bandwidth, but good level design should minimize the risk of overload by encouraging players to spread out in the map.

Intensive Use of the Maps

I will deal now with one of the major differences between the maps designed for single player games and those for multiplayer games. In a single player game, the player goes through a level with a single objective in mind, finishes it and passes to the next. He only spends little time in each level. But in multiplayer games, the players will spend hundreds of hours on each map. All map weaknesses will then be found.

Thus, design errors or bugs that allow cheating are revealed and exchanged among players. A second consequence of this hyper-use of the maps is the risk of player boredom if the map is not tactically rich enough. Multiplayer maps must support thousands of hours of play without letting the player feel bored. One year after the marketing of Splinter Cell – Pandora Tomorrow, thousands of multiplayer sessions were still being played every day, this is the same for other tactically rich maps such as some Halo 2 maps.

The Search for Efficient Gaming

The third typical constraint of the multiplayer level design is the consequence of the highly competitive game style that is specific to this type of game (except for cooperative modes). Since the essence of the multiplayer game is to crush the opponents, the players search for the most efficient tactics, whereas in a single player game, the players tend to play at their own pace and explore all the possibilities provided by the game.

What are the consequences? First, players completely ignore many game features (weapons, animations, specific map functions etc.), even if they show a real potential. They will only use the most efficient features.

The second consequence is the strong incentive to cheat or to take advantage of the map’s weaknesses or bugs. This problem is so important that it renders the classification in many multiplayer games null and void.

The Difficulty in Getting Casual Gamers to Play Multiplayer Games

The fourth constraint is the difficulty in getting average or casual gamers to engage in multiplayer games. The reason for this is simple: nobody likes being humiliated by losing repeatedly to gamers that give you no chance. Playing against a human opponent generates a lot of tension and makes the game more exciting, but also increases the stress level of an inexperienced gamer.

It will then be really difficult for him to put up with the three challenges he must handle simultaneously: control of the interface, knowledge of the maps and tactical vision of the game. There are many classification systems that regroup the gamers by level, but the majority only provides an incomplete solution to the problem of integrating the beginners.

At the moment, multiplayer games are reserved for the hardcore gamers. If we want multiplayer games to get out of their niche, it is vital that we design them with this in mind and not simply adapt them.

The Weight of the Gamers’ Community

Finally, the last major constraint is the weight of the gamer community.
A multiplayer game exists thanks to its players, who are hungry for new content (new levels), improvements, competitions and possibilities to adapt the game to their own style of play. The creation of a community of gamers around a game may be a blessing for a developer and its publisher, but the development of the game must be prepared in view of this.

In subsequent installments of this series of articles devoted to the multiplayer design, I will tackle my suggestions regarding: