My puzzle making process

Making puzzles is a lot like solving one, only in reverse.

The more I make puzzles, the more I learn what works and what doesn’t work and more importantly, I’ve learned why.

When I make a puzzle game I start with an idea such as “what if you could take items in and out of dreams?” and then I build from there and ask a bunch of questions. What do you take out of dreams? What effects do they have inside of the dream?

Then you go from there thinking about elements and how they interact with each other. Some of the most basic elements of a puzzle is taking one object from one location and putting it somewhere else. Crates for examples, appear in numerous puzzles to hold down buttons or clog up holes.

The reason they appear in so many puzzles is because they’re such an effective, yet simple piece. You see a crate in a puzzle and you already know what to do with it.

Your elements should be versatile and compliment each other. The more uses an element has, the better that element is, within reason of course. Take for instances the crate again. The crate can be used for three things, clogging holes, pressing buttons and helping you reach higher places.

That allows you to make a puzzle where you need a crate to accomplish all of these things, yet you only have one crate. This leads to a contradiction of sorts, where the puzzle, at first glance, looks impossible.

The opposite route to go, is making a puzzle look so simple that the player immediately starts trying to solve it, only to get stuck on the final move. There are many ways to accomplish this, such as misdirection and establishing expectations in previous puzzles.

So how do I make a level for a puzzle? Once I have all my elements, it’s time to build a level. I start by saying “I wanna use this element and I want the player to have to do this with it.” One of the puzzles I’m working on right now involves a grid and the way the game (Penelope) is deigned it’s easier to go down than go back up.

I start by just making the grid, sectioning off the room and I notice almost instantly that once you choose which path you’re going to go down, you basically can’t reach the others. Meaning the puzzle would’ve been solved in the first move and that’s not a puzzle at all.

So I had to let you travel sideways between the rooms, which of course made me run into my second problem. If you can reach all of the rooms sideways, what does it matter which one you go down first? So I have these walls that you can go through depending on what state their in and these were perfect for sectioning off the grid.

But there was one last problem. The grid was 3 rooms wide and about 5 rooms deep. You could still very easily solve the puzzle but just plunging straight down the middle, which again, is not a puzzle. So I expanded the grid to 4 rooms wide and viola! Now you can’t just plunge down the middle of the grid.

Now I fill each room with items you need to progress and items you need to collect to win the puzzle. The last part in the process is to test the level until you’re able to solve it. After I solve the puzzle I strip it down to the minimum that I needed to solve it and then I try to solve it using even less stuff.

Now, this is important, when you make a puzzle you probably already have a predetermined idea of how the puzzle “should” be solved. You need to chuck your supposed solution out the window and really find places where you could do better.

You may find that you didn’t actually have to do all the steps you thought you did, and in fact all you had to do was one relatively simple thing to just cut through the whole puzzle. And since you gave the player a bunch more resources than they really needed, they’re going to have a real easy time just stumbling through the level and win it by mistake.

So that’s all I’ve got for you today. If you want more posts like this, let me know!