Making a match-3 game without writing the board logic

Everyone knows how match-3 works. Three in a row clears. That is where the simplicity ends. Then you need cascade resolution, a shuffle check for dead boards, special pieces that form on four and five, a rule for what happens when two specials meet, and a scoring curve that makes a long chain feel like an event.

None of that is hard computer science. All of it is code, and it is the exact kind of code that takes a non-programmer a fortnight and a programmer an afternoon. The interesting part of your game — how it looks, how a clear feels, what the levels ask of the player — is on the other side of that fortnight.

Flockbay writes the code and leaves the interesting part with you.

You describe the rules in plain language. Board size. What counts as a match. What a five-in-a-row produces. Whether gravity pulls down or toward a corner. Whether the board is a rectangle or a shape with holes in it. Flockbay writes the GDScript that implements it, in your project, where you can open it if you want to.

Then you play it. This is the part nobody can do for you, because match-3 lives or dies on feel — the delay before a cascade resolves, the size of the pop, whether the board reads at a glance. You play, you say what is wrong, Flockbay revises. That loop is the normal way to work here.

You do not need to read the GDScript to have an opinion about the game. Not writing the code has never meant not judging the result.

More on this: Godot puzzle templates: three of the four are card games

Making a project asks 2D or 3D, shown as two pictures, with no default and nothing created until you pick. A match-3 board is flat, and picking flat here costs you nothing: the Catalogue you search inside the app is 55,073 sprites out of 60,648 objects, each with a picture, and a gem, a blocker or a panel goes into the scene at an exact position, turn and size with the mouse.

Under that it is a Godot project, open source, where sprites, tweens, particles, shaders and audio are engine features rather than things this tool has to reimplement badly. If you eventually want to hand it to someone who does write code, or take it somewhere else, it opens in Godot like any other project. That is a different position from the closed puzzle-game builders where the game only exists inside the tool that made it.

It also means the ceiling is Godot's ceiling. Match-3 with a story layer, a map screen, animated characters and a soundtrack is a normal Godot game, not an advanced feature you have to escape the tool to reach.

Flockbay installs on Windows and runs your game locally on the copy of Godot it ships with. For a match-3 game the performance argument is weak — a board of gems will run anywhere — so the honest reason to prefer a desktop app here is different.

It is that your project is on your disk, your assets are files you can point at, and the editor does not stop working when the tab reloads. Building does need an internet connection, because the chat and asset generation come over the network. Playing what you built does not.

If you specifically wanted something with no install at all, this is not it, and it is better to know that now than after a download.

A match-3 game is a hundred and fifty levels of escalating constraint, and the difference between a good one and a bad one is almost entirely in that sequence. Where the tutorial stops holding your hand. Which level first makes you fail. What the blockers are.

That is design work, and it stays yours. What Flockbay removes is the part where you cannot start designing until the board logic exists.

The fastest useful version of a match-3 game is coloured squares on a grid. No animation, no theme, no particles. It looks like nothing and it tells you everything: whether the board size is right, whether cascades happen often enough to feel good, whether the special piece rules produce interesting decisions or just noise.

Build that first. It takes far less time than a themed version and it is the version you will throw away three times while the rules settle. Committing to art before the rules are settled is the most common way to waste a month in this genre.

Once the rules are right the theme goes on top, and by then you know exactly what needs to be legible. Art comes from generation, from the web, or from your own hand, and Flockbay tells you which route it took rather than blurring the line.

More on this: AI card game maker: from an idea on paper to something you can play