Fata Morgana | Roald van der Scheur

World of Tanks (Wargaming)
Frontline, Fata Morgana

Fata Morgana is the first professional map I worked on that has been released. The map is built for the Frontline game mode for World of Tanks. In this game mode, you play a grand battle between 60 tanks, where one team is defending the Capture Bases/HQs and the other team has to attack and try to capture/destroy them within a time limit.

The Game mode in short:

The map consists of 3 lanes and 3 zones resulting in a map nine times the size of a regular map (9 x 9 km).
When the attackers get control over a zone, they will push their spawns forward and the defenders spawn backward.
Capturing a Capture Base (A - F) will give the attacking team additional time.
After capturing a Base in the first zone (A - B - C), the attacking team can move to the second zone (D - E - F) in that lane (example B to E).
Attackers can only switch between lanes that are actively being fought for (defenders can move wherever).
In the last zone, the Attacking team has to destroy 3 out of 5 HQs to win.
The defenders need to hold back the attackers and play the clock to win.
Each lane has a limit of vehicle classes, so there is (initially) always a variety of tanks per lane.

The creation:

A few Level Designers from Wargaming started a 2D/3D sketch of the map before they reached out to Bongfish to take over and continue. The requirements were to have it be a desert map that allows for more lane interaction and to give the attacking team stronger positions to fight the HQs at the end, unlike the second Frontline map (Kraftwerk).

Thomas, Paul, and I were responsible for the Level Design part. I took the task of creating a 2D sketch of the whole map which I received feedback on and iterated on until we were happy and all together moved to 3D.

2D Sketch

During the Sketching phase, we assembled references and information about desert-themed topics and we were mostly drawn to Egypt & Libya. The goal was to bring in a different theme per area on the map while having them make sense and be connected. We decided to differentiate the three different lanes into three topics: man-made structures, mountains/cliffs, and oases/dunes. The separation gave different gameplay opportunities to each lane and a different feeling for the player. In addition to that, we wanted to have a different theme for each zone to help the players orientate in the world and give them something fresh the further they progressed into the map. This resulted in a total of at least 9 different themes.

From the gameplay side we wanted to provide opportunities for all the tank classes in each zone, but also using the advantage of the theme to have certain classes shine. For example, the city to the West gave a boost to Heavy Tanks while the dunes/oasis to the East provided opportunities for the Light Tanks, where the Central lane had naturally more elevated positions for the Tank Destroyers to shine.

3D Sketch

When going into 3D we projected the 2D sketch on the terrain to roughly block out the heightmap and main structures such as rivers, mountains, and roads. The initial goal was to see how high we could go with the terrain and where we needed to add vis blockers for gameplay and optimization reasons. The next step was to have each of us focus on the first zone of the three lanes and bring in the basic structures. We used to playtest them internally and iterate on the feedback we received before moving to the in-between spaces and the second zone.

Art was involved quite early, which was a positive thing. The early cooperation allowed us to align the plans that worked for both departments, improve existing areas, and already plan future areas that would fit the vision from both sides. This made it easier to get to a common result and to break barriers. For example, in the last zone, where the HQs are, some of the themes for the HQs were not thought out yet and something we were struggling with from the Level Design side, where Art could jump in and come up with the idea of the Oil Refinery which became the main theme/eye-catcher of the entire map.

Heatmaps

Near the end of the Level Design production, we used to have a few bigger playtest sessions, where actual full battles were fought and where we had access to Heatmap data. We used to summarize the feedback formalized the points to possible problems and used the Heatmap data to see if these problems were true and what a possible cause could be.

We could for example see:

Positions that over- or underperformed in terms of how much damage they dealt and received in a particular spot.
Sight and shotlines, to see if there were any unintentional angles we had to solve.
Player and Tank class distribution and pathing to see if the right tanks found the right spots.
Death and kill positions to generally see where the most people died and got kills from.
Win and capture rate to see which bases were captured in what order and which side had the overall upper hand.

This data helped us to focus on objective problems rather than assumptions and helped us to make more objective changes to the design rather than assumptions and second-guessing to help balance the map and guide players to where we intended them to play.

My learnings

Designing a large open world based on real-life logic and theme distribution and connections.
Working closely together on a single map with other Level Designers.
Planning and organization of such a large project and working closely together with other departments (mostly Level Art).
Using Heatmaps to analyze problems and come up with possible solutions.