Engines

This article deals with engines: games centred around resources and their conversion.

The resource web

If one resource just does one thing, that's boring and obvious. There's no choice there.

Instead, your resources should form an interesting web, where each resource has multiple different and unrelated uses, so the player must intuit how to use it. The resources of your game form a complex and deeply linked web. That's fun. It means that resources are always useful, and require thought.

At the other end, it's also good if resources have multiple different sources, so that a player who wants to take a certain action can get there via multiple different paths.

For maximum webbiness, try to connect distant parts of your resource web. Try to connect each thing to a few different game systems.

In my farm game, I wanted to add another resource to the mining part of the game. This system already produced stone, which was linked to construction-themed systems. I really wanted to connect mining to the distant cooking and animal systems, to increase webbiness. I achieved this by adding salt. You mine it, and you can use it in cooking, or use it to build a salt lick, that helps your animals.

This is where your resource (and wider theme) choices really pay off. If you've chosen simple, obvious, thematic resources, you can use them in all kinds of ways that make strong thematic sense.

Stone is for building, but the bandits can be fought in a variety of ways — throwing stone at them being one.

Try to make new connections between your existing resources and actions. Don't overdo this. There should be two or three ways to do each thing, not ten.

You will find obvious sequences in your resource tree, and want to make them more webby. As an example, imagine your players always turn clay into bricks. Yes, you could add another use for clay, or another way to make bricks, to add even more webbiness. Often, you don't want to do this. In this case, you can deal with this problem at the action level, by making the clay action give clay and something else. In general, I don't like actions that do multiple things, as it reduces focus. However, a few multi-actions that give two unrelated resources can really tie together the least webby and most distant parts of your web together.

In my farm game, I had a space that made spears. I was originally using them for both fighting and for fishing, but the fishing part withered away, and there were only two uses for spears. I also had shovels, which were used in mining. They were very linear. Each of these was an important and thematic resource. I made a "tools" space, where you got both spears and shovels. This gave the player a big crossroads: use the spears, or use the shovels. Ideally, you'd want to find a way to use both.

Preventing repetition

You want players moving through your web, but you don't want them just repeating an action, or two actions. There's no choice to be made, and the game becomes uninteresting.

Usually, the answer is just to make loops longer. That's part of the reason for linking distant parts of the resource web.

My farm game had irrigation, which would make your trees and crops grow more. If I made the irrigation cost wood, the loop would be a bit too small, as you'd irrigate, get more wood, and then do more irrigation. Instead, I made irrigation cost stone, which was a much more distant resource.

Usually, you don't need to stop resource explosion. Just make it not useful. However, this doesn't work when the resource is points. The player will just repeat the loop.

In Race for the Galaxy, there are some interesting paths through the web. However, it's sometimes best to just produce goods, then sell them for points, then produce again etc. This loop is too small, and nothing really causes the player to want to leave the loop.

You can also give "by-products". Add on another, ideally unrelated resource, that will incentivise the player to do something else.

Make points less repeatable. Towards the end of a game, don't give actions that just give points. Instead, give goals, cards or buildings, that give points based on something. This way, at least the player has to do at least two things to get the points. These goals/cards/buildings are also non-repeatable, while a board space or ability that gets points is inherently repeatable.

Combined actions

If an action gives the players something, it should give them one of that thing. It's perfectly fine if it gives more, as long as there's a reason for it. There's no need for your game to deal with unnecessarily large quantities of things.

Also, actions that convert things, should generally be unlimited, or at least repeatable. These are big moves.

As above, I avoid actions that do two things, but feel free to bundle actions together, if they're unexciting, or if they're typically small.

If actions have the same kind of cost, you can combine them. Then, the player is given a new and interesting choice, of how much to spend on each action.

In Agricola, you can sow your grain, and you can bake it into bread. These are not particularly interesting, and the player will usually have a few grain at most. So, they're combined into one action: Sow and/or Bake Bread. Now, for each grain, you must choose whether to sow it, or bake it. This combined action is much more interesting and fulfilling.

Paths to victory

All paths through the game should be viable ways to win. Nothing should peter out, or become a dead end. Set up your points system so that every possible kind of gameplay results in comparable points. Do this by making all terminal resources be worth points, and making a path for all other resources to be converted into terminal resources.

If a player wants to go crazy making a million hamburgers, let them do that. Just make sure that these stupid strategies are rare, and the deep strategies are the common ones.

In Radlands, most of your cards are people cards. They're fun and interesting. You don't necessarily have to play them, however. Occasionally, you'll have a game where you really don't play many people. However, it's rare enough that it's an interesting change.