Constrained Text Generation with AI

I was discussing how AI text generation, such as ChatGPT, might end up getting used in computer games. So far, designers are fairly reluctant to adopt the technology. One of the key problems is that you just can’t control the output enough. Language models will break character or respond in inappropriate and toxic ways. Finding a good solution to this is a huge research field, and not likely to get cracked soon.

For the foreseeable future, AI in games is much more likely to be used offline – assets and dialog generation generated up front, so it can be vetted before being integrated into the game.

But it got me thinking, can we vet the AI’s output in advance, but still get the benefits of intelligent decision making at runtime? It turns out, we can! I doubt it’ll be useful in every circumstance, but I can certainly see uses for it, like chatbots, games.

The code and demonstration for this article is available here.

Continue reading

2d Marching Cubes with Multiple Colors

2d Marching cubes (sometimes called marching squares) is a way of drawing a contour around an area. Alternatively, you can think of it as a drawing a dividing line between two different areas. The areas are determined by a boolean or signed number value on each vertex of a grid:

But what if we didn’t have a boolean value? What if we had n possible colors for each vertex, and we wanted to draw separating lines between all of them?

Continue reading

Classification of Tilesets

Many years ago I started looking at different sorts of tiles sets used by artists. A good tile set is flexible enough to allow tiles to be re-used in a lot of situation, but simple enough that the tiles can be easily created. Ideally, it would enable autotiling or otherwise be easy to design levels with.

Though I covered a few different techniques back then, I fell short of any systematic discussion of tiles. Here I plan to take a more rigorous approach, in the hopes of making a common language for referring to different tile sets, and pointing out the key variations in design. Maybe we’ll even discover something new, like Mendelev predicting new elements for the periodic table.

You can explore many of the tilesets discussed using the Tileset Creator tool I made.

Continue reading

Infinite Modifying in Blocks

I’m going to share with you a technique I’ve found for doing lazy, reliable, deterministic, constant-time infinite generation of tile based levels using Wave Function Collapse (WFC). But first, let’s cover some background, Modifying in Blocks, and lazy chunk based infinite generation.

Continue reading

Constraint-Based Tile Generators

Last article, we were comparing WaveFunctionCollapse (WFC), and Model Synthesis (MS). These are both similar procedural generation techniques that work along similar lines. Specifically, they generate a grid of tiles (or pixels), using a set of hard constraints, and some generalized solver technique to find a solution, a set of tiles that satisfies all the constraints provided. Let’s call this overall class of thing constraint-based tile generators.

The techniques have been pretty popular in recent years, and part of that is because it’s actually a very easy concept to experiment with and extend. Many of the more practical examples in the wild include their own extensions, as basic WFC tends to produce levels that look a bit repetitive and structureless.

Today I thought I’d do a review of all the different ways that you can customize this generation technique, showing how WFC and MS are subsets of a greater class. The whole family of such algorithms is much larger and I think there’s a lot of potential still to explore.

Continue reading