I'm in pre-production on a strategy game and am trying to determine if the core gameplay will be fun. A good technique to determine this is to strip the game down to its minimum viable product (MVP) and see if that is fun. If the MVP isn't fun, then no amount of extra content or features will make it fun.

I'm having difficulty determining the MVP for a strategy game as I'm a little too far into the weeds to see which of the many design features are core mechanics and which are unnecessary.

Purely as an example, lets just say StarCraft2 is the strategy game I want to make. What would be the MVP for SC2 to prove out that its core gameplay is fun?

What would be the MVP for SC2...?

If the answer were generally known, don't you think there would be a lot more competition to SC2 out there? SC2 is the product of countless hours of design decisions; every patch that was released to SC1, SC1's initial design, the lessons from WC and WC2 that went into the SC1 design, and so on.

Game design isn't an exact science. Game design is working with infinite possibilities. Sure, there are fairly standard features in RTS, but the question here isn't What is an RTS to everyone? because everyone isn't building your game, you are. So it is rather, What is an RTS to you? (and why?)

Analysing the work of others is important; but getting started on your own work is far more important. Research is important; but don't let it bog you down. Start creating fun.

MVP is a brilliant idea, but you're missing the spirit of it: MVPs are about actively prototyping your ideas, not about over-thinking yours and everyone else's work. Getting your hands dirty is more important than worrying about what the supposed minimum mechanics for an RTS are. Many games can be considered RTS which fall largely outside the usual definition of that genre. Get a demo out and have people start playing it; and they will decide whether your product is viable, as well as genre.

I'm a little too far into the weeds to see

Until you start prototyping, that will be the case, and many questions will remain difficult to answer.

Some aspects i would say are quite easy to decide what you need for an RTS in generel. Depending on your concept, you need one "unit", that can be build, ordered or the game just starts with it.

Starting with Starcraft as your example, implement maybe a worker unit, one building and one fighting unit. Your building should be able to build both of them. General i wouldnt even add a resource to harvest, but since Starcraft heavily depends on it, in that case you should do that.

The hard part is, what features do you need to implement aswell. Your fighting unit needs to be able to "fight". So can it shoot? can it attack in cc? What are the enemies? Do you need more different units (e.g. air?)

Yes, you should only start with one race, so you basicly only got mirror matches so to speak. Additionaly, you dont need a map (if that doesnt hinder any very important features), just a square to move on. What is the objective? Destruction of the enemy or victory points, controlling capture points?

I think the problem with RTS is, that you basically got so many important features and basic elements, you still need to implement to have a MVP, while it is really hard to say, what core elements of your games are.

In my opinion it comes down to compare your base game to other RTS, and there are a lot of them, and even continued in a sequel, they are not the same.

• C&C Tiberium and Red Alert Series: Starting with main building, construct buildings by menu without a building unit, construct units in menu when corresponding building exists, different types of individual units (by foot, ground vehicle or air)


• Dawn of War 1 & 3, Company of Heroes: Base building with bulding units, no active resource gathering (passive generation by controlling points), most units are group based


• Battlefleet Gothic: No Buildings, no resouces, different types of units, not replaceable, skills are extremely important

All these differences make RTSs so vastly different. Trying to strip down a game to these basics are wastly more complicated than the example Extra Credit gave with racing games: Accelerating blocks, collision, thats it.

Real-Time Strategy is a genre which actually combines multiple games in one. You have a management game (resource management and build orders) a puzzle game (building a base with a functional yet easy to defend layout) two different kinds of exploration games (exploring the map and exploring which units beat which other units) and a tactics game (controlling your units in battle). You obviously can't create all these five games at once, so you should just focus on one of them at first.

In this answer I am focusing on two of these games which I consider most essential to the genre: Unit controlling and base building.

Unit Controlling MVP

1. Create a player-controlled "tank" game object (represented as an untextured cube) on an empty plane which moves to a new position when the player clicks there.


2. Add immobile AI targets (represented by a cube in a different color) which get deleted when their HP go below 0, and the ability for the tank to fire on the closest target to reduce its HP.


3. Add the ability for the targets to shoot back when the player-tank is in range.

Now you have a playable strategy/puzzle game: Navigate your tank from target to target in a way that it doesn't fight more than one at a time and destroys them all before itself gets destroyed. This is basically how you attack a base with defensive turrets in an RTS game.

Next steps to pursue in no particular order:

• Add a simple AI to the opponents so they can move and not just shoot back


• Add multiple player units to control and the UI for selecting units


• Add buildings
  
  
  
  
  • By the player
  
  
  • By the AI
  
  
  • Buildings which can build units
  
  
  • Resource Managment
  
  
  
  


• Add more types of units with different ranges, weapon strengths and hit points


• Add blocking terrain to the map and route finding to the unit AI so they can navigate around it.


• Replace the units with properly animated graphics


• Add win and lose conditions

Base Building MVP

1. Create an empty plane and allows the player to place a building on that plane by clicking


2. Add different kinds of buildings and an UI which allows the player to choose which building to build


3. Add construction times and resource counters


4. Add buildings which create resources in regular intervals (I wouldn't implement worker-units yet because they require too much AI programming to work)and a ruleset for when you can build which building ("can only build a factory when you have at least one finished barracks").

Now you've implemented the first few minutes of a starcraft game: Figuring out the ideal build order to get the buildings you want as fast as possible.

In fact you could stay here and polish it, and you have a simple resource management game.

I am looking forward to playing your game.