As an aspiring designer, my first design task in my first job is to create encounters for PVE levels in our F2P mobile strategy game, League of War: Mercenaries. The primary goal is to make these levels interesting. PVE experience exists in almost all F2P mobile games as the first and basic part players interact with. And they are kind of similar in terms of design philosophy — matching player growth with increasing difficulty to keep players in flow. However, for low-level design practices for this task, there are a lot of uncertainties for me. And I am inexperienced on it. So the first thing I am trying to do is search on the Internet for some pioneer’s suggestions because a lot of designers must have done similar tasks before. Disappointingly, I can’t find any article or discussion in this topic. Thus, I think it is great to write down and share my personal journey on this design task. Hopefully, other designers will be able to view my experience as a reference and improve their process of creating F2P PVE levels and share them again.
To be specific, my goal is to make levels “interesting”, which implies I should focus on the experience instead of the economy. So I would expect my levels to:Introduce in-game elements and show their unique mechanics and features.
- Introduce in-game elements and show their unique mechanics and features.
- Contribute to the sense of the virtual world in the game.
- Bring joy and happiness to players.
The game I am working on is about wars between mercenaries, the gameplay is sort of similar with Clash Royale with more strategic units options and in-depth elements like weather and defensive buildings. To be specific on this game, the first two goals can be translated as:
- Introduce various combat units and buildings. Show their relationship of supporting or countering each other.
- Create the sense of mercenaries war.
Emotional arc is always one of the most fundamental concern when doing an experience-focused design. From my perspective, the arc is similar in most F2P PVE experience, described as “Confusion – Understanding – Insight – Expectation – Surprise (cycling from insight to surprise)”. Players always start with confusion when they interact with new contents. One of the design goals is to make players understand the mechanics behind new contents as quickly as possible. Moreover, establishing their cognition around new contents and giving them insights of how to play with it. With the repetition of these contents, players should gradually form expectations about what is coming next. At that moment, something fresh, a surprise, either like a new element or a new pattern will add great spices to the game.
The above is more like the ideology. For practical purpose, I also have to understand what are my resource to use and what is my constraints before starting. In this game, enemy units levels, environment, and loots for each level is already set. For level structures, there are 21 zones. Each zone has three difficulties. And each difficulty has 8 to 12 levels. Each zone has different enemy commander figure and weather environment. In sum, there are 600+ levels waiting for me to feed. I should not make any change on these during the design unless necessary. Instead, my design focus is on structuring and tuning enemy units and buildings combinations for each level.
Learn in-game elements
My first step into this design task is playing the game. I play the game with randomized enemy units and buildings to be familiar with elements I am going to deal with. During the play, I took some notes and looked into some units details. This step serves to provide me with a general picture of the game and elements in it. Here are some examples:
- “The four units combinations in each level is highly flexible.”
- “The similarity among units of the same type makes them very hard to be recognized differently when placed together in one same level.”
- “The UI design places units in the first place, but buildings and weather environment are in the second.”
- “Buildings change the pacing and experience of the play a lot.”
There are more notes about very specific units and buildings, I won’t list all of them here.
For the next, I will start my design attempts by drafting out principles for this design. Since principles are decisive, it is a very important step in the whole process. Be very careful and thoughtful. Even through multiple iterations, the general direction of the design set by principles is hard to change and it determines how your design evolves a lot. Based on my basic understanding of the game and design visions from our lead, these are my initial principles:
- Repetition in adjacent levels helps players be familiar with new elements.
- Contrast may stress a certain unit among the four units group in each level.
- Patterns through levels should be used to help player set up expectations for following levels and lower their cost of cognitive learning.
- Sometimes, apply some rarely used combinations to create surprise.
Due to the fact that I am not 100% understanding the differences between buildings and units, assuming these are principles applied to both (which is no longer true in the later phase with my understanding of the game growing).
Enumerations on potential design
After having these principles, I do not start creating levels directly according to the principles. Instead, I start to lay out all potential level design based on the above principles. For each potential design pattern, I only repeat it once or twice before moving to the next potential pattern. I call this method as “enumerations with increasing factors”. For example, at the beginning, I only used 6 different units to lay out different unit combination patterns:
- Fix one strong unit and switch the other three units for each adjacent level;
- Switch one unit in different slots for each adjacent level;
- Keep two units fixed (strongest and weakest), and cycling the rest of them.
After enumerating all potential design patterns from principle #1, I add new units into the levels and trying out different ways to introducing them as principle #2, followed by principle #3 then buildings. The fact shows it is very efficient that even enumerating all these design patterns I only use less than one-third of the levels I need to actually design finally.
In summary, by enumerations with increasing factors, I want to answer two general questions:
- What are the best design patterns to deliver a certain principle?What are the best design patterns to resonate among different principles?
- What are the best design patterns to resonate among different principles?
Modify principles and design patterns
With these precious enumerations, the next step is testing and learning from them. I find this step very stressful but fruitful meanwhile. Even the amount of levels is relatively small, to make fully use of them, I have to divide my attention into two parts and keep switching back and forth. First, I need to recall what are my expected goals for the level. Meanwhile, I also need to touch the actual experience as objectively as possible to feel whether the design truly delivers the expected goals or not. Moreover, it is better to figure out reasons why these designs work or not. The followings are selected notes:
- “Be familiar and understand one new unit need about five times repetitions.”
- “Introducing two fresh units at same time overwhelms players.”
- “Zone transition is a great moment for pace changing. Design should collaborate with it. Create something different between zones and some patterns within each zone.”
- “Unified colors or defensive buildings (counter a certain type of unit) is very painful to players depend on the certain unit a lot.”
- “It is very neglectful for buildings and weather environment. So they require higher repetition than combat units to be noticed by players.”
After the intensive test, I gradually develop a deeper sense of all in-game elements and finally come up with the final design principles:
- For each zone, there should be an ambiguous story in mind, such as the commander of this zone loves tank.
- Consider each zone as a block for design. Make levels in adjacent zones different and levels within a zone have patterns.
- Introducing one to three units for each zone according to its length and the situation.
- Besides of the new units, choose other units from a preset library, “Common Units”, and use them repetitively to make sure new units are stressed by contrast.
- It is a reference pattern in a zone:
- New, Common A, Common A, Common B
- Same New, Common A, Common A+, Common B
- Same New, Common A, Common A+, Common C
- Same New, Common D, Common A+, Common C
- Same New, Same New (Different Color), Common A+/D, Common C
- Insert some irregular zones occasionally.
- Unify weather and buildings based on zones to reduce distraction from them.
- There are mainly three different types of buildings: Early-phase attack, purely defensive, and unit enhancement.
Now let’s go back to see whether these principles deliver the intended emotional arc. The repetition among the same zone stress out the new primary unit to push players to notice its unique features and gain insights on it. Zone as a basic repetitive block help players to form patterns through playing, which greatly reduces their cost of conversion from confusion to understanding and also builds up expectations. The transition between zones will bring player the surprise of new units.
From problem-solving perspective, the above design principles make players project their attention on the fresh units. The levels enjoy diversity and regularity based on zones, which is make sense for our real world. Moreover, a great pacing of “expectation – surprise” cycling are conducted by encounters in zones. And the unexpected irregular zones add an extra layer of surprise to the game.
Arrange resource & finalize design
Once I am satisfied with the principles, the rest of works are relatively simple and repetitive. First, list all units and choose one-third of the units as “Common Library”. The rest of them go as primary units. Then arrange these units to all levels accordingly with one to three primary units and 3 to 5 common units for each zone. Finally, add building combinations into these zones. That is the end of the design.
Since the design have not been put into release yet nor serious player test, it is hard for me to conclude. However, besides of the process, following points are meaningful to share as well:
Be brave on using binary tree search (It is not the algorithm)! When tuning a value, the intuitive way is to start with you intuition and gradually change the value into the right one by iterations. When I am trying to figure out what is a good block size (units are repetitive in the same block) for this design, I start with three level for each block. It feels messy. So I increase the size to five, then eight, then finally realize the zone (about ten levels) can be a block. If using the binary search, I should start with my intuition, three, then adjust with a big step like 20. Each time, adjust the step with half of the previous step either forward or backward. It is more efficient on finding the right number. Moreover, it is easy to be dumb if you adjust the value with minor changes, binary search avoids it as well.
Bottom-up during exploration. Usually, I start with digging into the game to learn these elements through playing instead of proposing high-level framework. A lot of design principles come from a deep understanding of the elements. In my case, the understanding of UI confirms units are the primary design focus then generates the principle about different repetition frequency for the elements.
Top-down during design production. However, it is very important to have a big picture in mind and follow it strictly when doing the final design. It ensures the design will not run off the rail and lose original intentions. One of my total failure iteration happens when I try to arrange all units into these levels without assigning a set of units to each zone. It turns out to be a complete mess that I use up too many units at the beginning to have enough units in the later levels. And I even lose track which units have been used.
Efficient tools matter. (And they do not have to come from engineers.) The first thing to avoid when doing design is distractions. A lot of subtle things need 100% attention to capturing. Generally, I see three main distractions during design: personal mood, the cost of acquiring information, the cost of iteration. The first one depends on designer him/herself. Tools help to lower the last two distractions a lot. Before I start to tune the units combinations, I use Spreadsheet formula and conditional formatting to extract the most important information of the unit and present them in a straightforward way. It greatly lowers my cost of acquiring this information and speed up the whole process.