During development, I mentioned many games that inspired me in one way or another. But if I shorten the list, the closest analogs are Into The Breach, Polytopia, and Frostpunk. The objective in the game seems simple: discover resources, extract them, complete tasks, and leave the sector before a strong earthquake hits.

But it's not that simple!

Each level represents a new biome with its unique set of monsters, "flora," and points of interest. In each sector, the rules change slightly, and new mechanics are added.

While in novels or quests the player experience remains relatively consistent, in a strategy game, it's quite different. I offer tools, rules, objectives, and methods of achieving them, but the player has to decide every second what to do next and exactly how to achieve the result.
I aimed to make the gameplay as random as possible, so initially, the map of each level was generated completely randomly. I like it when players are encouraged to explore when there's no complete understanding of what awaits them. Even plot objects may be hidden in one playthrough but revealed in another.

Incidentally, I also don't have a visual map editor. Maps are created through code. In my case, it works, but I wouldn't recommend this approach to others.

Naturally, randomness led to imbalance: playthroughs could be either too easy or excessively difficult. Although it sounds obvious now, the idea initially seemed good to me.

As a result, I had to return to the map generation code many times. Today, in the story campaign, the map is created taking into account predefined rules: the base, resources, and plot objects are distributed in "fair" regions, avoiding extremes.

Another rule I followed: to make sure something crazy happens every turn. In a good way. The thing is, if you don't invest, don't use perks and a special locator, you're essentially drilling blindly...

The following resources are present in the game:

• Petroleum
  
• Fuel
  
• Energy
  
• Water
  
• Oxygen
  
• Gas
  
• Technology Points

People are also a resource. They are set at the beginning of the first chapter. You lose the game if you lose your entire team. In addition, their mental state needs to be constantly monitored. Gameplay is influenced by various factors such as temperature, radiation, and other biome features.
Also, to not make life too easy, I implemented some abilities as randomly obtained perks for special Innovation Points, which can only be obtained by completing story quests and killing monsters.

Anoxia is led by a high command of heroes – officers with various specializations and unique abilities. At the start of the game, you choose your hero-avatar. Their death means game over.

Anoxia Station offers two game modes:

• Story Campaign: The main mode with a narrative.
  
• Quick Game: This mode has a significantly higher element of randomness, creating unique challenges for players. The difficulty setting also influences this randomness.

I think the game turned out challenging. And possibly, not everyone will enjoy the plot. But my theory is that interest in a game is born in the learning process. When you first encounter the rules, begin to understand them, make mistakes, find new paths – that's where the magic lies.

Unfortunately, in Game Maker, at least currently, there is no visual UI editor. This means that I have to manually place each button at specific coordinates. Then I need to compile the game, see how it looks, and if something is wrong, repeat the process. And so for each available resolution.
At some point, I started using a special extension that allowed me not to recompile the game every time. This slightly sped up the process, but still didn't completely solve the problem and didn't save much time.

The save system in a strategy game with hundreds of variables is a nontrivial task.
I'm proud that I managed to implement exactly what I wanted. The game only has one save slot, but technology and characters are carried over between chapters. Of course, players can replay chapters as they wish.

Generally, a strategy game is essentially a collection of arrays and loops; lists. Therefore, I didn't reinvent the wheel, I simply save the objects at the current moment. However, then, when the level is recreated on reload, I simply delete everything and load the objects and their variables that I saved. It's crude. But it works.