Hi Everyone,
While I'm working on the next update, I thought it would be interesting to share some of the ideas behind Mech Armada and why some design choices are the way they are.
To really understand the game, I always refer back to a small set of guiding principles. This is what defines Mech Armada's identity and they help me make decisions during development. In game development circles they are sometimes called pillars. This is not the only way to analyze a game, there are many other frameworks that combine to create a clearer picture.
These are some of Mech Armada's pillars:
- Each run is different
- Reward mastery
- Balanced and fair
- Feed the power fantasy
In a nutshell, I want to provide replayability through variety and improving your skill. At the same time, I want the game to feel good, meaning fair and exciting.
Variety
Fundamentally, for any game to be replayable there needs to be enough gameplay variety so that the game doesn't become repetitive and boring.
Variety through content
Because Mech Armada is a single player game, variety must be accomplished through content, by which I mean player's tools (Mech Parts), enemies, environments, encounters, modes, etc.
I'm able to create some of this content procedurally (battles/encounters) whereas some of it needs to be handcrafted (parts, monsters, environments). Even the pieces that are procedural have a limited range of what they can generate, based on the code.
Adding more content is an obvious benefit to the game that follows the pillars well. So you can expect more parts and monsters, but also more variety in encounters as well as modes.
Variety through randomness
There's another side to variety that is a bit more complicated.
One of the most common pieces of feedback I receive asks for more control over what Mech Parts you have available. This can be either selecting the parts you have when you start a run, being able to cancel out of the random part drop or having some other mechanism (like a tech tree or a store) where you can work towards specific parts.
I get it. You want to succeed and you feel like some parts are really strong, while others are not. Naturally, you want to be able to get the stronger parts. This is actually a well known problem in basically every game where you have a large collection of tools available to the player (like champions in League of Legends, for instance).
Even if a game is well balanced, players develop certain preferences and are reluctant to change. The downside is that this hurts variety. In the extreme, if you always have the same exact collection of parts, it makes sense for you to develop an optimal strategy with them, and then execute that over and over. The only opportunity for the game to offer variety is through the challenge (enemies and encounters), but that's not enough and it starts feeling repetitive, as you're basically doing the same thing again and again.
Currently in Mech Armada you don't always get the parts that you want. Maybe you really want a Gatling Gun so you can buff it to do extra damage because that worked really well in a previous run, but it doesn't drop. So you have to adapt. Can you build a team around the Energy Shield you just got? Or what if you can't really do enough damage, but you have a lot of mobility? This creates different problems every time, with different solutions, which helps the game feel different.
Mastery
I knew from the beginning that I wanted Mech Armada to reward skillful play. Not all players appreciate games that require mastery. There are plenty of great games that focus on story, or creativity, or immersion instead. Even games that have a clear win/lose condition sometimes offer players a way to succeed through means other that skill (e.g. grinding or paying).
Types of skill
When we talk about mastery or skill, it's important to recognize that there are many different kinds of skills.
Mech Armada being a turn-based game doesn't require quick reflexes, mechanical precision to move a crosshair for a headshot or insane actions-per-minute numbers. This was actually one of the guiding principles behind making the game turn-based in the first place.
Instead, Mech Armada challenges you to think strategically. But what does that exactly mean?
Sometimes it's easier to define what something is by explaining what it is not. Some games function like giant virtual spreadsheets: the game is full of stats and modifiers, there are complicated calculations going on with every interaction and a 5% increase in one of the numbers has a huge impact down the line. Other games have enormous possibility spaces and you have to predict the long-term consequences of every choice, which means you have to create a very complicated mental model for how everything is connected.
Both of these style of games reward mastery, but I'm not trying to create either of them.
I want numbers to be small and calculations as simple as possible but, most importantly, I want qualitative differences, not just quantitative. I want the Rocket Launcher with area of effect to differentiate from the Cannon with armor piercing because they do different things, not because they have slightly different numbers.
And I don't want players to feel "analysis paralysis" because they are overwhelmed by the number and/or complexity of their choices. Instead, I want to have many smaller decision points that combine together to form an overall strategy, but that in isolation have a limited impact. Which Mech Part type do you need? Which one will you pick out of the two options? How do you combine them? Should you sacrifice a unit to accomplish some goal? Rush the enemy or wait?
So if strategy is making a series of decisions, I want Mech Armada's decisions to be clearly differentiated and "weakly" related to each other (so they add up slowly over time).
I should talk specifically about the part drops. The reason there are two options is to reduce the complexity of the decision; it's actually pretty easy for most people to compare two things. The reason you can't see what your options are in advance is to introduce an element of risk to buying a part. The strategy here is not finding the optimal path with perfect information, but rather mitigating that risk as best you can and finding creative solutions with the parts you got. As a thought experiment, would Slay the Spire be a better game if you knew from the beginning which enemies were in each encounter, and which cards were rewarded? I'm sure for some people it would be, but that wouldn't be the game I'm trying to make.
Mastery vs difficulty
In order to encourage and reward skillful play, the game has to provide some kind of challenge to overcome. Some kind of "test of skill", which is effectively the game's difficulty.
The "skill floor" (the lowest skill that you need to succeed) of the game will always be too high for certain people. This is true of every game that requires skill, from Minesweeper to Street Fighter. Ideally, you want a game with a low skill floor and a high skill ceiling (the highest skill the game rewards), so there is a lot of room for people of different skills to be fairly challenged.
There is a problem I've found when the skill floor is too low. If it's too easy to win the game, the consequences of your decisions become unimportant and all the strategy basically disappears. It's not fun to play anymore. Picture someone playing a tennis match where they can't hit the ball properly so you lower the net and extend the court so they can win.
The flipside of course is accessibility. I want as many people as possible to play and enjoy Mech Armada, and I will make the game as accessible as I can. But without some degree of skill the game won't be enjoyable, so the focus shouldn't be on lowering the skill floor, but helping struggling players elevate themselves by learning how to play better.
The challenge system in Mech Armada (increasing difficulty levels from 0 to 12 that players can select when starting a run) exists to raise the skill ceiling. And trust me, at higher challenge levels, the game becomes hard.
Making the best strategy the most fun
Players want to win and if the game requires skill they will use everything at their disposal to succeed. Unfortunately this means that sometimes they'll do something that's not enjoyable simply because it's effective.
The energy economy is the lifeblood in Mech Armada, you need energy to spawn new mechs. It's natural and expected for players to want to collect as much as possible. The Fog Creature and the energy cap exist to prevent players from indefinitely exploiting a state where they are not in danger and can collect enormous amounts of energy, therefore breaking the game's challenge. Even with this mechanism in place, energy is so powerful that many players feel compelled to do as much as they can, realize that just sitting there collecting energy is not much fun, but still feel frustrated that they can't do more.
Even though the design goals of limiting energy collection are valid, the actual way this is accomplished can be improved. This is an area I'll be exploring in future updates.
Balance
Balance in a game means that the obstacles you face are fair, so that when you lose you understand the mistakes you made and feel that you can do better next time. Balance can also mean that all the tools available to the player have a time and a place where they are desirable, and none are always indisputably better than others (this is sometimes referred to as "dominant strategies").
It's basically impossible to reward skillful play if a game is not balanced well. Also, while a game where everything is the same would be perfectly balanced, it wouldn't be very interesting. Maximizing variety while pursuing balance is an important constraint.
Balancing a game is extremely hard, in big part because the balance can be quite different depending on the player. In practice, I use both spreadsheets and data regularly to balance Mech Armada, though most of the balance input comes from the community.
I will post with more details about how I use spreadsheets and what kinds of data I analyze in the future, but essentially it comes down to combining all the numbers that affect an element in the game (a part or a monster) into a single number (something like "power") that can be compared with others. Then you assign a cost to that element that is representative of their power.
Outside of the purely numeric analysis, there is also the idea of counter-play. Parts and monsters are designed so they can both combine and counter each other in interesting ways, providing challenge and strategy, but without becoming dominant or dictating a specific way to use them. Needless to say, this is a fine line to walk.
Unfortunately, these tools are generally too simple to fully solve the problem of balancing a dynamic, ever-changing game with complex strategies. They just provide a starting points. Fine tuning the balance is one of the main reasons why Mech Armada is in Early Access, and I will continue to make adjustments to the game's balance throughout, as a result of player feedback.
Power Fantasy
This one's the easiest to justify: who doesn't want to feel powerful? Mech Armada tries to accomplish this in two main ways: visuals and progression.
I always wanted the game to look and sound good and move well. A big part of how a game feels comes both from the way it looks (clear and pleasant visuals, punchy sound) and how smoothly everything moves and responds to the player. Not being an artist by training, the current implementation is the best I can do with my resources. Hopefully you find the explosions rewarding.
The other part of the power fantasy is getting stronger over time. You start with almost nothing but by the end you're so ludicrously overpowered that it almost feels like cheating. As long as the sense of challenge remains (see Mastery), this feeling of power is wonderful and an integral part of playing Mech Armada.
The Early Access roadmap
Understanding Mech Armada's pillars helps me read through player feedback and prioritize the work needed to get the game ready for a full launch, while staying true to the game's identity. This analysis is but a small portion of what goes on regularly as I sift through a long list of tasks and improvements.
My plan is to release a major update once a month, with a minor update in between. In this way you can see something new in the game every two weeks.
I will provide new parts and new monsters regularly. There will be some new features that will change the gameplay that I'm not ready to reveal right now (I keep a list of ideas to make the gameplay more varied and interesting). I want to have more variety in both encounters (enhance the procedural generation) and modes (going beyond the roguelite campaign).
At the same time I'm exploring solutions to some of the feedback I've received:
- Energy farming is not fun
- You want more control over starting parts
- Being restricted to use parts only in a single blueprint at a time feels unintuitive
- The Campaign needs some kind of world map
Lastly, there are some technical features that are needed to make the game feel complete, like Achievements, controller support, accessibility features and localization.
Be sure to Follow Mech Armada on Steam to get notified as these updates get released.
/Sergio
https://store.steampowered.com/app/1389360/Mech_Armada