Game Career Guide is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Get the latest Education e-news
  • A Three-Level Short Story: Level Design In Die Young: Prologue

    - Daniele Mascagna

  • CONFRONTATION. This is the actual core of the Prologue, where the player makes use of the mechanics learned during the tutorial and is free to approach the area as they see fit.

    Keeping in mind my self-imposed guidelines (use mostly ready-to-use assets, tell a story, encourage players' autonomy), I started doing research about the setting, collecting references for me and the artists (warehouses, container yards, fuel docks, etc.). We already got most of the 3D assets at the time. For the blockout, I could use our modular kits and even some finished models. While I was looking for references, I also started making a list of ideas as they popped into my mind.

    It is advisable to have a varied list of "cool moments" (gameplay or emotional beats) before proceeding with the level layout. It is kind of a wish list, an unstructured collection of ideas; there is no rule, and they could range from specific locations (a maze-like container yard, a construction site, a drug laboratory), to spaces for core mechanics (parkour among the racks of a warehouse), interactive objects (sliding gates, shutters, elevators, fuel valves), "puzzles" (moving containers with forklifts to create a path to a hard-to-reach location), AI encounters (an enclosed area where the "junkies" are locked in and threaten anyone who approaches the fence; you have to pass through them or dangerously walk above them on a plank; you can open the fence and let them fight the guards for you), scary moments (a pitch-dark tunnel filled with junkies), new items (an unknown drug, a lethal chemical), secrets (you must locate a container hinted at in a document), etc. Anything goes, in this phase.

    Once I had a stronger knowledge of the setting and listed enough ideas to fill the level with, I began to work on the layout: this is when those ideas start to be spatially and logically organized (in scale pace, and beats).

    First of all, I had to consider the extension of the level. Since the Prologue is set on the same island as the main game, I wanted it to fit inside a specific - and relatively small - area of the original map (as often happens, limitations breed creativity; this limited surface forced me to optimize spaces and make use of verticality).

    I identified seven macro-areas for the level and divided the space accordingly. For each of them I defined their look and features, enemy types, gameplay, and main beats:

    (Click to enlarge) "The Harbor" Macro Design

    To respect players' autonomy, I intended to offer multiple routes to get past the harbor. However, I wanted to make sure that each player would experience at least two intense beats during their playthrough. To this end, I placed climaxes just before the two possible final exits (Tunnels, Exit Gate), and peaks at the two main alternative midpoints (Fuel Tanks, Laboratory).

    The resulting flow is something like this:

    Now that I had a solid overview, I moved on to the proper blockout phase (I find paper design too limited for 3D games: there is a high risk of neglecting verticality, and the perception of scale is also compromised). Having modular kits is extremely useful at this stage (we designed our "buildings kit" very early in the development of the main game; it proved to be very versatile and helped us to keep our metrics consistent throughout the game). Using these kits, I began to define buildings and spaces inside the macro-areas. I wanted to squeeze as much gameplay as possible out of the limited surface I was working with, and since one of our core mechanics was parkour, I knew that I should aim for different layers of heights: virtually every floor, ledge, and roof in the level had to be walkable and used to create more possibilities for the player.

    Keeping in mind the main flow and tempo of the beats, I started working on sub-nodes inside the areas (ground-level and elevated routes, gates, rooms, ladders, hidden paths, etc.). I had to make sure to provide enough clues for those possibilities, clearly declaring the main objectives but subtly suggesting the paths to reach them.

    This is when composition, sightlines, and lights come in. Colors and lights help to keep track of the key elements of your level, since grayboxes can quickly become unreadable without. Those key elements are also the ones that should stand out more once the level is fully finished, so it is important to identify them as soon as possible and ensure they are correctly framed from the player's perspective.

    (Click to enlarge)

    While working on the blockout, in addition to our modular kit for buildings, I also reused some of the finished assets from another level of the game (e.g. the two huge fuel tanks).

    Iteration after iteration (sped up by the use of the modular kits), I got to a fully explorable space: all routes and waypoints were in place and each route could be tested to get you from start to finish. Once the navigation felt nice and smooth, it was time for enemies and covers.

    Enemy and cover placement has a huge impact in weighting the players' routes. Subconsciously guided by the principle of least resistance, players evaluate paths for their efficiency and perceived threat: an enemy guarding the fastest route forces the player to seek alternative, safer paths; similarly, an open, wide space with little or no covers (the so-called "prospect space") is perceived as potentially dangerous.

    In setting up covers, it is extremely important to ensure that there is absolutely no space for ambiguity. The player must be able to read covers at a glance, clearly distinguishing them from non-covers: no prop should lie in the "buffer zone" between previously defined heights for covers (low and high) and non-covers.

    In the harbor, I used a set of 3D assets taken from previous levels of the main game (small containers, boxes, tanks, barrels, stacks of pallets, etc.). Their heights were consistent and they could easily be categorized as low or high covers. Since we were short on time, using these finished assets already in the blockout was a production necessity. This means that by placing covers I was not only managing gameplay but also set-dressing most of the level. This is something that can easily slip your mind while working with generic gray boxes, but covers should always seem logical in the environment, as they can contribute greatly to making the gamespace feel believable.

    For the sake of readability and flow, it is better to organize covers in clusters. These clusters work as nodes, guiding the player from cover to cover, letting you - the designer - predict their movements inside your level; this allows you to work with sightlines more easily, evaluating what the player should or should not see in one location or another.

    (Click to enlarge)

    This is the view from your cover at the entrance of the harbor. Not only did I ensure that no cover cluster would block the sightlines towards the next goals, but I actually used covers to frame them, hiding any other distraction (also note that the fence in the back is purposely screening the rest of the harbor).

    After a new set of iterations, the level was ready in all its gameplay elements. Working side by side with artists, we implemented the final art. Again, color and lights were carefully chosen to achieve the desired mood and guide the player. The harbor is characterized by cool colors (mainly blue and green) to feel like an inhospitable place; warm colors (especially yellow and red) contrast more with this background, making them the colors of choice for highlighting objectives and possibilities (main gates and doors, stairs and ladders, climbable ledges, etc.).

    At first, I was not sure about using white color for the roofs, but in the end, I think that the contrast with the ground level helps to expose their affordances.

    (Click to enlarge)

    Top view of "The Harbor" level.


comments powered by Disqus