Skip to main content

Progress on Procedural Generation.

From the start the idea of this project has been that everything should be done as much as possible with procedural generation. I want people to come back and play it again and again.

That's not to say that there won't be a story, and non random elements, but the key thing is that you can replay it and get a different experience.

Anyway, I've been working on the random dungeon generator for some time, more than a year it seems, though I haven't been working on it continuously.  I got some ideas from existing random dungeon generators such as this one:

There were three things I wanted it to do:
  1. Create interesting rooms which are more than just rectangles.
  2. Allow different ways of joining the rooms through corridors.
  3. Use different tile types in the same map to avoid looking repetitive.
 The first thing required a bit of a different approach from normal generators. Normally we just set a room size and make a room but I used a dictionary of interesting room shapes. At the moment the dictionary contains a small sample so rooms are a little repetitive, but later I'll be expanding the sample and adding different dictionaries for each tileset.

I used arrays to represent the rooms which look like this:

The arrays can be rotated or mirrored to create new rooms from existing ones. I may write a function to merge room types too.    

I used A* pathfinding to achieve the second goal. Corridors are drawn automatically from each room exit making every room accessible. They can join in sequence or join all to the entrance to the level, or all to the exit.

The builder uses several different tile types to make the rooms, some can transition to other types directly, some can't. I used a transition dictionary to set which can and which can't That can be changed for different tilesets.

Here's what the result looked like using simple colored objects as placeholders:

 The generator created them very quickly, meaning that to create a whole dungeon of levels would probably be the work of seconds rather than minutes (I hate long loading times).

One other thing I'm working on is to create granular rock, some hard some soft, which helps clump the rooms together in to more organic shapes and also causes the corridors to snake more pleasingly. Where corridors have to be cut through hard rock they can be set to have a different appearance.

Finally here's a look at how it all fits together in a video:

You can see the red circles show where there is hard rock. Setting granularity high makes a very noisy image with small pockets of hard rock, this snakes the corridors just how I want, lower granularity makes the corridors avoid certain areas of the map which is also interesting.

Next up I have to work on getting the tiles ready to walk on. I'll be using empties added to the mesh to show which squares are walkable. When the level is loaded the empties will be used to generate a A* graph (taking note of which nodes are linked to doors) which will be saved as a set of dictionaries which can be accessed one at a time to prevent creating a too large graph for the A* algorithm.

There already exists several variables that can be changed to create more varied levels, such as room mirroring, number of rooms and level size, as well as type of room linkages and average room sizes. These are mostly not active at the moment but have been tested and seem to work fine.

Later I'll be adding further refinements to the generator script so that it can add multiple stairs up and down to exit the level, match the location of previous levels stairs to the current level, and also match up rooms from the level above in cases where there is a pit trap. But for now it works well enough for testing purposes.


Popular posts from this blog

Make your game models POP with fake rim lighting.

I was watching one of my son's cartoons today and I noticed they models were using serious amounts of simulated rim lighting. Even though it wasn't a dark scene where you'd usually see such an effect, the result was actually quite effective.

The white edge highlighting and ambient occluded creases give a kind of high contrast that is similar to, but different from traditional comic book ink work.

I'll be honest, I don't know if there's a specific term for this effect in 3d design, since my major at university was in traditional art. I learned it as part of photography.

You can find plenty of tutorials on "what is rim lighting" for photography. It basically means putting your main sources of light behind your subject so that they are lit around the edges. It can produce very arresting photographs, either with an obvious effect when used on a dark subject...

..,or as part of a fully lit scene to add some subtle highlights. See how alive the subject look…

How to... build a strong art concept.

So you want to make some art assets for your game. The first on the list is a Steampunk Revolver for your main character to shoot up Cthulhu with. Quickly opening your internet browser you start with a Google image search. Ah, there is is!

It might be a good idea to find a few influences so you don't accidentally end up copying a famous design.

Just mash them up and you're ready to go! Off to your favorite modeling program.
But wait! isn't there more to building a strong design concept than that?

Of course there is.
One of the diseases of modern design is that of recursion. Everything is a copy of a copy of a copy. This is especially a problem with "historical" concepts. Over the course of that recursive process the concept becomes infected with modern design elements, and ends up looking very similar to everything else that anyone else has ever made.
If you want to come up with a really fresh idea, you have to get beyond secondary references and go look at real …


Ok, so it's not exactly skynet, but I have got my first AI state working, kind of.

The first state is "HOLD" in which case the agent stays in place where they are and shoots at any unit that comes in range. When I started writing this module, I found that the existing method of triggering actions wasn't good enough to allow the AI to choose the best weapon or target. It worked by simply sending a command to the unit to trigger the currently selected action.

If the action is valid, it triggered, if not it didn't.
That's fine for play controlled units, as that's all they need to do. But AI needs to know in advance if the action is valid. The player can get that info from UI feedback, but that wasn't available to the AI player.

There were three problems:

1. The UI feedback duplicated code in the action trigger function. These  two sets of code could get out of phase so that UI feedback was wrong.

2. The action trigger didn't give enough feedback for …