In this article, we take a look at the 2019 GDC talk, during which Aubrey Serr shared an overview of non-linear design techniques in single-player sandbox levels.
In 2019, during a Game Developers Conference, Aubrey Serr shared an overview of non-linear design techniques in single-player sandbox levels, the distinct value and challenges of non-linear levels, and how level designers can adapt these techniques to otherwise linear levels.
So, in this article, we decided to take a look at this great talk and show you the great level design techniques, which can be used to create so-called Radically Non-Linear Levels, that Serr showed during it.
First of all, what is a Radically Non-Linear Level? According to Serr, it is a sandbox level that is designed to encourage unpredictable player movement and exploration of the space. This type of level is very distinct and interesting, as it never allows players to become bored. The oldest and cleanest example of such a level that Serr found was Miss Pac-Man.
The map is offering free-form tactical choices, thanks to its openness, and large enough for strategic choices, available to the player. Serr commented, that the designer of Miss Pac-Man confirmed the various behaviour of each ghost, which makes each ghost a unique risk factor that the player needs to consider.
Serr mentions an additional list of factors that make the map in Miss Pac-Man a great example of a Radically Non-Linear Level:
An important thing Serr mentions is that Radically Non-Linear Levels ≠ open-world maps. If it were an open-world map, the pacing would be destroyed by unlimited power pellets and the ability to outrun the ghosts no matter the situation. In other words, the gameplay would have been completely different.
Following the explanation of the term "Radically Non-Linear Levels", Serr explains, what makes such levels cool and appealing to the players:
First of all, before starting a map you have to come up with the layout of it. By creating a strong layout you'll be able to differentiate each map you make, rather than ending up with a messy brew of obstacles, enemies, and rewards. There are a lot of different possibilities for a layout, but Serr proposes 4 solid examples.
The first one is the symmetrical layout. We have a low tension centerline with a left/right choice. This is an interesting layout because the player gets to choose where to go. This helps to break up the tempo and creates pacing. This is a great layout when the major feature of your map is some kind of path, like a highway.
With the concentric layout, the outer rim of the map is safer and rewards scouting before diving headfirst into the enemy territory. These maps work well for maps about assaulting a base or a castle
Node layout provides a series of clearings that intertwine in some way or the other. This kind of map keeps the high tempo, because the player is constantly in danger, with only small pockets of air to recuperate. This layout fits perfectly with ruins or slums.
The field layout is basically a large area with increasing risk. This layout promotes attacking and retreating to get the best strategical position and gives the player a lot of control in terms of tempo. This is a typical layout for something like a beach assault.
It is important to remember that you can create any layout you wish, using gameplay mechanics as the driving force behind it.
Usually, it is a good thing to pick one gameplay concept and utilize it to its fullest. Serr explains, using the game Brigador as an example, how to do it correctly. In this game, every part of the terrain is destructible, so it was possible to make the layout revolve around this aspect. The player has the opportunity to make their own path to achieve victory.
Another layout Serr shows is the exposed layout. The main gimmick of it is that the objectives are located on the different sides of the map, thus making the player vulnerable. This shows the importance of picking a concept and a layout that work well together.
At the end of the talk, Serr talks about problems that Level Designers may encounter during working on a Non-Linear Level.