Data and knowledge in A Gentle Rain and Timber Town

A Gentle Rain is a solo or “co-op”¹ game in which you place 28 tiles one at a time, matching edges each time. Making a square of 4 tiles earns you the right to place a flower that exists on that square, place all 8 of your flowers and you win, with spare tiles contributing to an increased score. That’s it, that’s the game.

It’s presented as a very Zen experience. Indeed the game describes itself as “A cozy game for one or more minds in need of a soothing reset,” and rule number 1 for setting up the game is to make yourself comfortable, consider your clothing and music choices, make tea, stretch and breathe.

Timber Town is a much more competitive experience - strictly a two-player affair it has each player strive to complete a 16-tile town from the 48 tiles available. They roughly (ignoring some special abilities) come in sets of 3, one of which you will draft, one your opponent will draft, and the other will probably fall off the end of the river and remain uncollected.

Matching is important here too, as each tile has paths that might extend in each of the 4 directions. Running a path off the edge of the board is allowed, but where two tiles meet they must agree on whether a path exists there or not².

However, beyond the matching, each tile has a certain structure on it, and those structures have different scoring rules, a la Cascadia. This makes it impossible to predict how the designers might have chosen to populate the various tile types with paths.

In both games, you are playing with a hidden, face-down stack of tiles. You can’t look at the stack of tiles during the game, but in both cases you might really want to know something about what the stack potentially contains. With no other hidden tiles in either game, either a tile is in play, discarded in the case of Timber Town, or yet to come out of the stack.

That knowledge is useless to you unless you know something about the contents of the stack in the first place, and you probably didn’t study the stack contents before your first game - especially if you were being taught the game live by someone else.

Here’s why I chose these two games to illustrate the point - they combat the problem (if you even consider this to be a problem) in two completely different ways, and in my opinion, both of them are correct in their own choices.

A Gentle Rain:

Timber Town:

What’s that? You want in on the comparison too, Which Witch? Well, I can hardly leave you out, can I?

Which Witch:

So that’s why I think the choices made in the design of these three games reflect the intent of the design. No front-loaded information enables you to explore the patterns in A Gentle Rain, a more arbitrary choice set for Timber Town begs for the playing field to be evened out on the player aid, and Which Witch chose to avoid any hidden information by being played completely out in the open.

In writing this article I first took a long hard look at the makeup of the tileset of A Gentle Rain and did the math. The conclusions may be considered spoilers so, before I dive into the details…

If you enjoyed this newsletter, we’d appreciate it if you would forward it to a friend. If you are that friend, welcome to the newsletter!

First of all, it makes sense to ask about the individual flowers themselves. There are 8 flower colours. The game has 28 tiles, and each tile has 4 half-flowers, one on each edge, meaning the game contains 112 half-flowers in total. With no indication that any flower is any rarer than any other, you would expect each flower to appear on 14 of those available slots, and that turns out to be true. Not only that but no tile contains a flower colour twice - put together that means:

Each flower exists on 14 of the 28 tiles - exactly half.

You could use this knowledge near the end of the game knowing that, for example, you still need to place a yellow flower, a quick count of how many yellow half-flowers you can see will tell you how many tiles remain carrying that colour - and that tells you how careful you need to be about completing the half-flowers you already have in play versus your ability to create a brand new flower.

However, the game really isn’t about flowers, so much as it is about corners. After all, you cannot make a square to place a flower without at some point fitting a tile into a corner - matching 2 tiles at once. Let’s have a think about how the corners might work.

Each of the tiles has 4 corners, so that’s the same 112 corners in total as the edges. How many potential corner combinations do we have? Well, each corner starts with 1 colour, and we have 8 to choose from. We already ruled out any corner existing that’s 2 instances of the same flower, as no tile ever contains the same flower twice. That leaves 7 choices for the second colour of the corner. Finally we must note that the order matters - a blue-red corner is different to a red-blue corner as we can’t reflect the tile, only rotate it.

That makes for 8 times 7, or 56 potential corner types to find in our 112 actual corners. That divides very evenly indeed - each corner should appear exactly twice in our tileset, and careful counting confirms it.

Each corner combination exists on exactly 2 tiles.

With this knowledge you can easily see if a corner you’ve created has any tiles left in the stack that will fill it. Look for tiles in play that would fit your corner - if you can already find 2 then you know there isn’t a 3rd in the stack to be found.

I did a similar counting exercise for the 3-colour combinations. However, there are far too many 3-colour combinations to expect to find - 336 combinations and only 112 places to put them means at least two-thirds of them cannot exist. What I found by analysing the edges is that almost all of the 3-colour combinations appear in the game exactly once if they appear at all - in actual fact only 9 appeared twice. Unfortunately I wasn’t able to sense a pattern amongst the repeats or in which combinations existed versus what was excluded.