A tower defense game maker where the balance is the whole job

Things walk a path. You spend money on things that stop them. If the things that stop them are too strong, the game is over before it starts; too weak and it is a treadmill. The pleasure is in the gap between those, and the gap is narrow.

Everything else — theme, art, story — sits on top of that arithmetic. This is why so many tower defense games feel the same underneath and why the good ones feel distinctly different: the numbers are the design, not the decoration.

It is also why the genre is a good place to start. The scope is small, the loop is short, and you can tell within one playthrough whether something is wrong.

A tower defense map is a puzzle about placement. Where the path bends, where a tower can cover two lanes at once, how much buildable ground sits near the choke, whether high ground is worth walking to.

You build that by hand in Stage Engine. Terrain, water, buildings, the shape of the route, the pockets of ground that are worth fighting over. Then you play it, and you find the one placement that trivialises the level, and you go back and change the shape.

Doing it in 3D changes the genre slightly and usefully. Height, line of sight and cover become real rather than notional, and a tower that cannot see over a ridge is a design element rather than a bug.

Every tower is a small set of numbers with a rule attached: range, rate, damage, cost, what it is good against, what it cannot hit. Every wave is a composition and a timing. All of it is code, and Director 1 writes it as GDScript inside your project.

Which means the loop is: play a run, notice that the cheap tower is always the right answer, say so, play again. That is the entire craft of the genre and it depends on being able to change a number without it costing an evening.

The judgement is yours from beginning to end. Not writing the code does not mean not seeing that wave nine is where everyone will quit.

Adding tower types. It always feels like progress and it usually is not. Five towers that interact meaningfully beat fifteen that are the same tower at different prices, and the fifteen take three times as long to balance.

The test is simple: if there is a tower you would never build, it is not a choice, it is a mistake with a price tag. Either give it a reason to exist or delete it.

The same goes for enemy types. Two enemies that demand opposite answers create more interesting play than eight that all die to the same thing.

A tower that is firing and doing nothing is indistinguishable from a tower that is broken. Players read a tower defense game almost entirely through feedback: the hit, the number, the health bar, the sound when something dies.

Which means presentation work here is not polish, it is function. If a player cannot tell that their new tower is worse than the old one, they will not learn the system, and a system nobody learns is not a game.

The cheapest test is to play a wave and ask yourself which tower is contributing least. If you cannot tell, the player certainly cannot, and the fix is in what the game shows rather than in what it computes.

Stage Engine is a desktop app for Mac and Windows with Godot bundled inside it, so the game runs locally on your machine. A late wave with a few hundred units in it is your hardware's problem rather than a browser tab's.

Your project is an ordinary Godot project on your own disk, and Godot is open source, so it opens without this app.

Building needs an internet connection, because Director 1, asset generation and world data come over the network. Playing what you built does not. Free to download with a free plan, Indie $10 a month, Pro $50 for more usage.