Dev Blog #22 – Puzzle Design Principles For Lake Ridden

Hey guys what’s up? With summer vacations approaching we’re trying to squeeze in some work on the game before it’s time to get some well-deserved time with friends and family. In this blog post, we’ll talk a bit about puzzles in games, and how we approach building puzzles into Lake Ridden. To avoid spoiling any puzzles from the game we’ll talk in broad terms, not giving away the solutions to any puzzles from the game. I’ll use a generic location of an abandoned old school as the example here (don’t worry, we don’t have one of those in Lake Ridden).

So what do we mean with “puzzles”? Well, we’re not talking about the most traditional form of puzzles, which is the classic jigsaw puzzle which comes in many shapes and pieces. When we talk about puzzles we’re referring to “problems in the game that can be solved”. Sometimes the puzzles need to be solved in order to move forward, like finding the key to a locked door, sometimes a puzzle can be part of bonus content and really easy to solve.

Lake Ridden is a first-person thriller, set in a creepy atmosphere, filled with puzzles and story. We’re aiming for a nice balance between experiencing the atmosphere, the story and challenging the player with puzzles as well. It’s not a walking simulator, it’s definitely more of a traditional gaming experience. I think there are are some really awesome walking simulators out there, but we decided to go for the more traditional gameplay focused interaction while presenting a rich story for you guys to explore and enjoy! When working with our puzzles we divide them into four main categories:

• Big main puzzles,  such as finding a key to a door at the end of a level to be able to proceed in the game.
• Medium sized puzzles, such as finding books or letter to unlock lore or piece together an item.
• Small puzzles, the simple act of open a drawer, lighting a lantern.
• Dialogue puzzles, like talking to somebody in the right way to get information

I think many gamers and developers perhaps don’t even think about it, but even shooters are a form of puzzle games. If we remove all fancy graphics and effects, most shooters are really a situation where you need to clear a level from obstacles to be able to proceed. Sometimes you can plan your moves in detail, such as in Hitman, and sometimes it’s so much action and effects going on that you might not event feel like you’re actually solving a problem (getting from the beginning to the end in games like Doom). Games that are placed in the explicit “puzzles category” are usually quite clear about their puzzles. You are asked to do classic puzzle solving like finding combinations to open safes, assembling cogs to start a machine, place items in certain orders to turn the power back on etc. But I would say “puzzles” are more than that. They are really about solving problems, as mentioned before.

Our 10-Step Process For Making Puzzles

At Midnight Hub the game design team is the same as the story and programming team; Malin and Johan. For this post I sat down and picked their brains on how they work with puzzles in Lake Ridden. This is their general process:

1. Brainstorm. As mentioned before we have divided the whole game experience into eight chapters for us to work on. We look at the chapter and have a pretty good idea about the terrain, the story and what feelings we want to communicate in each chapter/location. This could be a creepy, abandoned school. We then brainstorm what big, main puzzle(s) we could place there. We determine what items/info the player should have then they successfully have solved the big main puzzle for that chapter.
2. Whiteboxing of the bare minimum of the puzzle to get something testable up and running. This is as little as just white boxes and placeholder graphics so that we are able to test the very mechanics of the puzzle, like placing cogs in a particular way to enable the power to come back on to the old school.
3. We test the puzzle ourselves. Does the concept hold up? Does it feel nice? Sometimes a puzzle you ‘ve designed on paper just doesn’t feel right when putting into the game.
4. Basic art. The next step is to ask the art team for some rudimental art. We record voice lines ourselves and insert voice and dialogue into the scene if the puzzle needs this.
5. External testing. When we have the basic core art in the scene, together with some light and voice we bring in the first external testers to give the puzzle a go. Before we ask the testers to step into our office and try the puzzle, we form a theory about what the “perfect playthrough” would look like. What is the average playtime? How should the tester solve the puzzle? What parts could be hard to understand/find or get right for the player? We then watch the tester play the puzzle, ask questions and talk about it afterward. It’s in this stage you usually become aware of any confusion, or if the puzzle is way too hard or easy, if it’s boring or if it’s not clear to the player what they are even supposed to be doing.
6. Tweaking. We work with the feedback from the playtests, together with adding more feedback into the puzzle. This would be the stage where we add a switch turning green when the players has restored all the power to that old school.
7. More art is added, to make the puzzle and its environments look awesome.
8. Sounds and music are added to the experience.
9. More feedback and testing, to make sure the music or the art did not impact the puzzles in any way we did not intend.
10. Final art is applied, together with final sound, music and voice dialogue.

To us, it’s important that each puzzle, big or small, is able to stand on its own. Each puzzle should feel interesting, challenging and worthwhile for the player to solve. We want our puzzles to have an impact on the game, to leave some kind of trace in the game world. It’s important that the puzzles make the player feel smart, not talked down to or outright impossible. Since people play games to be entertained we want our players to have a good time when solving the puzzles in Lake Ridden. We test our puzzles a lot, but I do feel like our testing process deserves a post of its own.

Six General Tips When Designing Game Puzzles

1. Don’t make the puzzle feel “gamified”. Integrate the puzzle into the game world. The very opposite of this could be if you’re looking to open a locked door and you just happen to find a key very next to the door. It just feels…very much like you’re playing a video game, you know. Try to make the solutions and the puzzles feel native to the game world like they are an integrated part of it.
2. Research. Adjust the difficulty based on your target audience. Do your research. Play games that your target audience probably play and enjoy. How hard are the puzzles in those games? Never steal a puzzle straight from another game, tho. Some games have insanely hard puzzles and some have almost zero difficulties. As a game designer, it’s very easy to design your own puzzles way too hard. Be careful about this. Have people test your puzzles and see how they do.
3. Reward the player. We aim to make the player feel great about solving the puzzles. If possible we want to give them an aha-moment when they figure it out, because that is such a nice feeling! Ideally, we want to reward the player both with the internal feeling of accomplishment and some kind of item/new info/story snippet or perhaps access to a new area. If you just keep solving hard stuff without getting any kind of nice feelings from it, you are very likely to get bored and feel cheated. Avoid doing that to your players.
4. Mind the learning curve. Be mindfull about the learning curve of your game. For Lake Ridden we teach the player about the game using fairly easy puzzles. In the beginning, we use simple puzzles to tech the player about the inventory. We teach them that you can pick things up and how to interact with the game world. Just remember that easy puzzles like these, that work like a tutorial, only really work in the beginning of games. After this, they will feel like too much handholding, like you don’t trust the player to use her brain to work things out.
5. Hints? What happens if you play a game and get stuck in a puzzle? If you just cant seem to find that one missing screw or key, or it seems impossible to figure out how to get out of a maze? Mots people today go straight to Youtube to search for a video with a solution. You don’t want to make people feel frustrated, you want them to feel the challenged. So to remedy this we have implemented a hint-system into the GUI. If the player gets truly stuck at some point she can choose to reveal one hint a a time, starting with the hint that gives least amount of information. This way we hope to keep the players inside the game, and to give them an option to tailor how mich help they might need, IF any at all.

I hope you enjoyed this introduction to puzzle design for Lake Ridden. There are many awesome puzzle games out there, and it’s easy to forget that almost all games are puzzles in some way or another. With our upcoming game, we hope to draw a nice balance between challenge and reward, to make the puzzles feel as natural parts of the game and to entertain the player. We use many kinds of rewards when a puzzle is solved, may it be special items, new story or access to a new, exciting area. If you want to read more game developer focused posts make sure to tell us here in the comments. Big thanks for all your support, without you this game would not be possible!

Cheers,
Sara & The Team (especially the design duo Malin & Johan!).