How Good Planning Can Cement Your Game in Player's Minds


Raise your hand if you enjoy exploring dungeons in video games. Be they the often-understated dungeons of Pokemon or the dingy halls of Legend of Grimrock, these can be the most memorable parts of video games. Even the term itself invokes a certain air and feeling that calls to mind different games for each of us. However, think of a game you played last year that had dungeons. Can you name all of the dungeons? Do you even remember what some of them looked like or how they played? Making a dungeon memorable can be pretty challenging, especially if your game has a lot of them. Fear not, for there are a few things you can do when designing your dungeons that can help them stand out not only within the context of your own game, but among others in the same genre as well. However, this is an enormous topic. I want to make sure that I cover everything I can, if only on a surface level, but it's just too much for one article. So, this week we will be looking at how to build a solid groundwork from which to launch your designs, and how you can make your dungeons memorable from a structural perspective.

Before we continue, I do need to clarify that we won't be focusing on how to make puzzles good or specifics on how to make them flow well. This is mostly about the planning stage, because execution is something that can be honed and refined through playtesting and iteration. Also, we're going to be talking a lot about Zelda, since those games tend to be the standard to which other dungeon-based games are held to. If you're a developer and haven't played any of those games, well... You should probably play a game or two from one of gaming's most iconic franchises. So grab your torches and compasses everyone, because we're in for a long discussion.

Angle of Approach

Strictly speaking, a dungeon is nothing but a series of problems; keys and locks to put them in. The first thing you need to do is figure out in which direction you are going to be designing said dungeon. There are three ways to do this: the top-down approach, the bottom-up approach, or the hybrid approach. Unless you are your own boss, chances are this is actually going to be assigned to you from the start. For example, if your lead comes to you and says, "You're in charge of the design for our game's second dungeon. It takes place in an old monastery atop a mountain." They then go on to describe how the dungeon should feel and all sorts of other artistic aspects. In this case, you would be designing the dungeon from the top-down; you are designing from visual elements, ideas, and emotion. The opposite of this would be if your lead came to you and said, "You're in charge of designing our game's second dungeon. We'll be introducing a new projectile weapon here and the dungeon should take about 45 minutes to complete. Oh, also we want to show off the game's lighting engine here, so make sure there are several areas where we can utilize that." See the difference? This is the bottom-up approach, one where the elements that were prescribed to you in the top-down approach are instead the ones that you create yourself. Now this second dungeon could take place in a haunted mansion or dormant volcano.

Simply put: do you start with flavor or mechanics?

Truthfully, you probably won't be taking either of these approaches at their most extreme. More than likely you'll be doing the third approach: the hybrid approach. In this scenario, your lead approaches you and says some combination of both of the directives above. That may seem like a lot of limitations, but don't be afraid or discouraged by that. Limitations and restrictions breed creativity and may cause you to build dungeons you otherwise would not have. But, let's say you are your own boss and you have the privilege (or burden) of deciding how the entirety of the dungeon is made. The hybrid approach may still be the best one for you. Sit down and identify where the dungeon will be and what it needs to accomplish. Whether or not you get to decide these elements is less important than the fact that you have some sort of guiding direction from the outset. If your direction is unfocused or misguided, you will never create a memorable and fun experience. If you take nothing else away from this article, let it be this.

Structurally Sound

Now that you have your core ideas in place, let's move on to some of the mechanical and structural elements that have a bit more fluidity to them. We'll start with an easy one: is your dungeon puzzle based or combat based?

Both Oracle of Ages and Oracle of Seasons were designed from this question. Oracle of Ages focuses more on puzzles, while Oracle of Seasons is more combat oriented.

Just like your type of approach, this isn't a binary scale. In fact, a dungeon probably shouldn't be one or the other, but knowing which side you're going to be leaning toward can help define how areas are built. In the Legend of Zelda series, this is an important distinction because it sets the pace of the dungeon. A dungeon that leans more toward combat is going to be faster and more thrilling. That kind of dungeon will appeal more toward someone who fancies themselves to be a strong hero saving the princess through his grit and willpower. However, a puzzle-based dungeon is going to be slower and appeal toward someone who sees themselves as an underdog outwitting the big bad.

The Stone Tower from Majora's Mask is an example of a dungeon that favors puzzles over combat, with its mind-bending mechanic of spinning the dungeon upside-down.

This balance isn't only achieved by volume or number of enemies versus puzzles, but also in their difficulty. A puzzle based dungeon should have easier enemies, or enemies that rely more on wit and timing rather than ones with huge health bars. Likewise, a combat based dungeon's puzzles should reflect the idea of strength in some way.

The next thing to determine is whether or not your dungeon is going to be linear or non-linear. The knee-jerk reaction from most gamers seems to be that linearity itself is bad, but that's simply not the case. Linearity has its own uses and certain types of dungeons may benefit from being mostly if not completely linear. An important thing to remember when designing a linear level is to make it not seem like it's linear. It shouldn't feel like the game is holding your hand through the whole experience. On the flip side, a non-linear dungeon shouldn't be a confusing or obtuse one. The dungeon still needs to have a sense of direction. In his YouTube series entitled "Boss Keys", Mark Brown describes linearity as, "Whether the way forward is directly in front of the player or whether they are winding back and forth through the dungeon, continuously exploring the same space and unlocking more of its secrets." In his series, Brown created a number of flow charts that break various Zelda dungeons down into a more digestible form.

I highly encourage you to go watch his videos, as he has over an hour's worth of analysis that I can't possibly begin to get into in this article.

In these charts we can see how truly linear or non-linear any given dungeon is on a mechanical level. Something to think about is what I call the "speed-runner's route", this being the absolute fastest way of finishing the dungeon and how long it takes. Is this number flexible? Could potential speed-runners figure out this route quickly, or would they need to attempt multiple approaches to determine what this route is? It’s worth thinking about even if you're not designing with speed-running in mind (which many developers don't), because how quickly your dungeon is unraveled leaves an impression on the player about its complexity.

The final thing to consider from a structural perspective is the dungeon's primary mechanic. This may be something you've already figured out or something that was part of your assignment, but if it isn't, now is a good time to get that hashed out. This mechanic might be an item found within, or a recurring puzzle that's used throughout the dungeon.

The bulk of the Water Temple's puzzles are done through the manipulating of the water level, but its navigation is made much easier once you acquire the Hookshot.

We can't really go too deep into specifics here, as this is going to be highly dependent on whatever game it is you're developing. This is pretty important, as the primary mechanic feeds into what people are going to remember about this dungeon the most: what it is they had to do the most. This combined with the setting of the dungeon are what your players will remember, in addition to whether or not it was good.


Alright, that was a lot to take in all at once. I encourage you to look at your favorite dungeons from the perspectives we've outlined here. How linear are they, and in what direction do they lean? Can you identify the main mechanic, or is the whole dungeon a hodge-podge of disconnected puzzles? Think about that between now and next week, where we'll talk about how to make your dungeon aesthetically memorable and learn how setting can influence the layout and flow of a dungeon.

See you next time!