The Making of Tragedia: Dungeon Design

This post is part of “the Making of Tragedia” series. Tragedia is a classic-style role-playing game I am currently developing in RPG Maker VX. You can learn more about the game and download the demo at the Tragedia website.



As I began to create new dungeons for Tragedia, I realized that there is a surprising lack of information online about 2D dungeon design. For any 2D game that features dungeons, these levels have extremely important puzzle and combat elements, so why are we not talking about them? In this post, I’m going to take you through my design process for the dungeons in Tragedia, as well as provide tips for creating your own dungeons.



First off, let me define what exactly a dungeon is. I consider a dungeon to be any area in the game that utilizes puzzle and combat elements, and helps break up the narrative (although it should never break the player’s immersion). Most dungeons feature bosses and some kind of reward or power-up at the end, but not all do. Designing dungeons is certainly no easy task, and there are a lot of factors to consider before even sketching a map.


So how exactly do you go about creating an interesting dungeon for your game? The first thing I start off with is a list; I simply write down a list of the dungeon’s location, theme, story (if any), and the types of enemies, puzzles, obstacles, power-ups/treasure, etc. it will contain. The goal of this is to define everything you need to know to map out the dungeon, so it’s best to keep this in a bulleted format for quick viewing. Keep in mind that any of this information could be adapted or changed to suit the needs of your game. For example, you might not have a definitive location for your dungeon yet, but you know the types of obstacles you want the player to face. When you do decide on a location, you can adjust the visual representations of those obstacles to match the locale and theme. A basic lock and key puzzle could be changed to where the player has to find a boulder and push it to a certain location so that they may pass over it and advance; it’s the same basic premise, but has been adapted for an outdoor location where having a random door to unlock wouldn’t make much sense. When making your list, be sure to take the location and theme into consideration when deciding on obstacles and puzzles, but remember that you can always adapt them later if need be.


Whether you’re personally designing all of the dungeons in your game, or if you’re working with a team, you need to compare lists for each dungeon. These levels should build off of one another, so that a technique or skill or item the player learned to use in the first level is carried over through the other levels. This can be a tricky thing to coordinate, especially if you’re working with a team, so I recommend having the lists for all dungeons in a single, easily accessed location, such as a Google Doc. This way, anyone on the team can look at what is going on in a specific dungeon, and ensure that their own dungeon’s design and obstacles don’t break the continuity. Even if you’re working alone, this will be extremely helpful, as you can keep track of all of your dungeons in one place and make sure they work together, rather than against each other. In case you’re confused on the continuity issue, allow me to give an example: in level three of the Legend of Zelda (NES), the player acquires the raft, which is then used to progress through all subsequent dungeons. If the player only used the raft in a single dungeon and never again, it would seem pointless and likely make the player feel cheated. That’s not to say that every following dungeon has to feature the use of the raft as a primary mechanic, but the subsequent dungeons have at least one room where the raft is required to progress. You can promote dungeon continuity in your game through the use of repetitive enemies, obstacles, or puzzles, but remember that you will want to upgrade those obstacles in later dungeons to make things more challenging.


Let’s take a look at a few lists from Tragedia (spoilers have been removed!):

Location: underground sewer
Enemy Types: seedy magicians, rats
Obstacles: water channels, gates, switches

Location: forest

Enemy Types: spiders, insects, plants
Obstacles: trees, elevations, water

Location: cave

Enemy Types: snakes, imps, gargoyles, bats
Obstacles: rocks, water, pitfalls

Location: prison

Enemy Types: guards, bandits, other prisoners
Obstacles: locked doors, gates, keys, switches, pitfalls


As you can see, several obstacles are repeated, but are represented and used in the actual levels in different ways. For example, level A and D both use gates and switches, but because they are different areas, with the prison being later in the game, the proximity of the switches to gates will vary. Also keep in mind that these dungeons are not finalized, so I might be introducing additional obstacles or shifting things around at a later point.



Now that you have a good idea about what you want to be in your dungeon, it’s time to start mapping. For Tragedia, I first take a look at the assets I have available, as this can influence the level’s design, especially if working with limited resources. Then I bust out the graph paper. Since my game is 2D, it’s easiest for me to use graph paper to mark the screens, which also helps to determine the size of the dungeon. This shouldn’t be to scale, but instead, should be a general representation of where you want everything and how it should fit into the game world. I usually start off by marking where I want the entrance and the boss room to be, but keep in mind you can always move these around later. Next, I’ll take my primary mechanic and think of ways of incorporating it into the environment. For example, in a forest level in Tragedia, the primary mechanic is elevation, where the player has to locate vines they can use to climb up and down, allowing them to traverse the maze-like area. It doesn’t matter what the primary mechanic is, but whatever it is, it should be utilized the most out of all the player’s skills. In the forest level, elevation is not only used to navigate to the boss room, but it is also used to locate treasure that might seem unreachable at first glance.


When designing your dungeon’s map, there are a few other things to keep in mind: 1) critical path and progression, 2) enemy placement, and 3) playtesting and revision . Critical path is the quickest route from point A to point B. For dungeons, this means the route the player must take to go from the entrance to the boss room, without going out of their way for unnecessary treasure (I say unnecessary here because the dungeon can be completed without it). While your critical path could just be a straight line, that isn’t exactly very interesting. Instead, you can use obstacles to temporarily heed the player’s progression, thus forcing them to either find another way around, or to find a solution that will allow them to open up a path, such as a key. If you created a list as I discussed earlier, you should already have a solid idea of what types of obstacles you can implement in your dungeon. Looking back at the list for the forest dungeon, you can see that I listed elevations, trees, and water as obstacles. In the beginning of the dungeon, the player can climb up a vine to reach an elevation, where they would discover a river with a bridge on the ground level. They would then realize that they need to find a way to get to the ground level near the bridge. By placing obstacles along the critical path, it forces the player to explore more of the dungeon in an effort to find the correct route, and along the way, you can conveniently place treasure to deviate them from that path for only a moment. This is far more interesting than just letting the player walk right up to the boss room. Depending on the size of your dungeon, you will want to be careful where the critical path takes you, though. If you have a huge dungeon, making the player backtrack to the beginning is likely to get them lost, so be considerate of size when determining a critical path. If you’re having trouble understanding this concept and how to implement it, take a look at other games. The Legend of Zelda series has always been a big inspiration for me, but you’ll want to look at examples that are along the same line as your game. Study how the player progresses through a specific level and take notes if you need to.


Enemy placement can come in two forms: random encounters, where the player cannot see the enemy, but is taken to a separate battle screen every so often (think Final Fantasy); or visible encounters, where the player can see the enemy and can determine when, where, and how to strike (think Chrono Trigger or Legend of Zelda). No matter which form you use in your game, it’s imperative that you are consistent with enemy encounters. If you use random encounters, you should not suddenly introduce visible encounters in a dungeon, and vice versa, as this will be jarring to the player and probably break immersion. For dungeons with random encounters, you want to balance the frequency of encounters, which will probably mean adjusting your step counter if you’re making an RPG. While dungeons are assumed to be crawling with monsters, forcing the player to switch back and forth between the dungeon and the battle screen too often can be frustrating and cause them to lose track of what they were doing or where they were going. It will likely take a lot of tweaking to get right, but it’s an important practice, especially with larger dungeons. You can also introduce safe zones, where the player understands that they will not be attacked while in these areas. In many Final Fantasy games, safe zones are the areas with save points. If you do this, just be sure that it is clearly communicated to the player that this is a safe zone, and reinforce this with similar imagery throughout all of your dungeons. Visible encounters require a lot more planning, since the player can choose how and when to react to the enemies they see. This can easily tie into puzzle design, where the player might have to manipulate an enemy into stepping on a switch, or they have to defeat all the enemies in order to unlock the next room. If you’re utilizing visible encounters, it is essential that you take enemy placement into consideration when defining the critical path. For more help with visible encounters, take a look at Chrono Trigger for help with placement and use as obstacles, and the Legend of Zelda series (2D games) for help with use of enemies as puzzle elements.


Playtesting is crucial to the development of any game, regardless of genre or platform, so it only makes sense that you’re going to have to playtest the crap out of your dungeons. When playtesting, you should check for these things, among others:

  • Bugs and Glitches
  • Collision Issues (Can the player pass through things they shouldn’t? Can they go anywhere they shouldn’t be able to?)
  • Progression Issues (Is there anything that prevents the player from completing the dungeon? Do all of the puzzles function correctly? Do any of the obstacles or puzzles feel too challenging for most players?)
  • Level Differences (Do the enemies’ levels adjust to the player’s? Can the player enter the dungeon if they are a lower or higher level than intended? Do the enemies feel challenging to the player, but not overly difficult?)
  • Treasure Importance (Are there adequate rewards for the player’s progression through the dungeon? How valuable are the rewards, and are they useful to the player?)
  • Leveling (How many levels does the player gain throughout the dungeon? Are enemies providing enough experience?)


Of course, this isn’t all that you can check for, but these are some important factors when playtesting dungeons. Still, no matter how many times you playtest a dungeon, you still know everything about it, which is why you should look for outside help for playtesting. Ask a few friends or release a brief demo with a feedback form so you can find out what other people think of the dungeon’s design. After you collect all the feedback, the next step is to apply it! Revisions are a part of the design process, and you should expect to revise your work numerous times. If you look for areas with poor design in your map sketches and revise them before implementing the level, it will save you loads of time and frustration, so be sure to take a good look over everything before beginning the digital construction. Even a map that seems perfect on paper is bound to have a few problems once you fully implement it, so don’t get frustrated if you have to revise the digital level. For the first dungeon I did in Tragedia, I redesigned the entire dungeon several times, both on paper and digital, but in the end, I’m far happier with the result than if I would have just given up and stuck to the original design. Playtesting and revision go hand in hand, so be sure to utilize them when developing your entire game, not just the dungeons.


The last thing I’d like to discuss before ending this post is environmental influence in dungeons. There are many games that I’ve seen, especially within the RPG Maker community, where all of the dungeons are walled in areas with different textures. In Tragedia, I knew this wasn’t the route I wanted to go. In my 3D level design, the environment has a huge influence, so why shouldn’t it in 2D games? While some dungeons are walled in, it is because it is story and environmentally coherent. For example, there is a prison dungeon that the player will have to escape from, so it makes sense that this dungeon is in an enclosed space. There are other dungeons, however, that utilize aspects of the environment to act as walls, such as trees in the forest dungeon. While I am using limited resources, I try to mix it up in terms of what encases the dungeon area in an attempt to keep the player from being bored with their surroundings and to make things feel less “gamey.” This also applies to boss rooms; while many games might have a traditional room where the player encounters the boss, you can use the environment to create room-like areas in order to maintain continuity with the rest of the dungeon. For example, in the forest dungeon, the boss room is actually just a clearing in the forest that can only be reached by progressing through the maze that is the dungeon. Whenever possible, you should take the environment into consideration when designing your dungeon, whether it’s for the walls or surroundings, the obstacles, the puzzles, or the enemies.


Thanks for reading! As always, feel free to leave a comment, critique, or question below, and stay tuned for the next post in “the Making of Tragedia” series!

  1. You are so cool! I don’t suppose I have read anything like this before. So wonderful to find someone with some original thoughts on this issue. Seriously.. thank you for starting this up. This website is something that is required on the internet, someone with a little originality!

  2. Hello, I want to subscribe for this blog to obtain most up-to-date updates, so where can i do it please assist.|

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: