Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 22, 2014
arrowPress Releases
October 22, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

The Dungeon Crawler Recipe
by Bruno Patatas on 01/19/13 05:08:00 pm   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


During 2012, the dungeon crawler genre got a great boost with “Legend of Grimrock”. LOG was one of my favourite games of the year, and based on sales number, the appetite for high-quality dungeon crawlers is alive.

As a game designer, dungeons crawlers are for me one of the most challenging game genres. Of course you can portray the genre as simply being slaying creatures, levelling up your character and collecting loot. But, design of dungeon crawlers, whether they are pen-and-paper, tabletop or in video game form, always have their own secret formula, the pillars of what makes a dungeon crawler great.


This is my dissection of the game genre and what in my point of view are the pillars of what makes a dungeon crawler a great dungeon crawler.

1. Well Defined Goal

Since the beginning that the player needs to know the reason of his quest. Of course there will be numerous loot and monsters to slay, but there needs to be a main purpose that will drive the player during the entire game. Generally, the reason for the quest is one of the three “R’s”: Revenge, Retrieval and Rescue. Look back to all the dungeon crawlers you played in the past, and you will see that in some sort of shape and form they adhere to the three “R’s” rule.

2. Theme

The environment needs to be consistent across all levels, in order to not break player immersion. If you are creating a medieval/fantasy game, do not throw to the mix cultural references from other ages. Do not mix medieval with Renaissance or Victorian and vice-versa. Keep a unifying theme that is believable for the player, because if you break that consistency you may well be ruining the players play experience.

3. Over-Arching Goal

In order to establish the setting for the game, it’s extremely important that there is a common link across all the enemies that the player will find in the game. They may all come from the same region in the world, or they may all be minions of the same master villain. What is important is that there is a common, unifying theme for the enemies, which blends with the world lore.

4. Traps & Puzzles Galore

A dungeon crawler is not a dungeon crawler if you don’t find traps and puzzles everywhere. Besides adding tension and challenge for players, they are also great reward mechanisms. When the player spends time trying to figure out how to avoid an encounter with a creature, and then he is able to successfully circumvent that encounter… That is a great moment. The player will feel he is incredibly smart, and the adrenalin boost will provide him with more excitement than if he had confronted the creature directly.

5. Pacing

When designing a dungeon crawler, you can never leave space for the players to breath. They must be constantly faced with new challenges (walking through empty labyrinths without nothing to do is not that exciting, don’t you agree?). Whether they are enemy encounters, traps or puzzles, keep the flow of content quick. Design it so that when the player finishes one problem, he is almost instantly introduced to a new one. Keep the players on the edge of their seat.

Related Jobs

DeNA Studios Canada
DeNA Studios Canada — Vancouver, British Columbia, Canada

Analytical Game Designer
University of Texas at Dallas
University of Texas at Dallas — Richardson, Texas, United States

Assistant/Associate Prof of Game Studies
Avalanche Studios
Avalanche Studios — New York, New York, United States

UI Artist/Designer
Bohemia Interactive Simulations
Bohemia Interactive Simulations — ORLANDO, Florida, United States

Game Designer


Richard Wood
profile image
Good read Bruno!

Bruno Patatas
profile image
Cheers, stranger! :)

Chris Dixon
profile image
I think two critical gameplay aspects you leave out are the character leveling and class system, and the gear/loot system. You can have all of the above things, but if your in-game progression system sucks, people won't even stick around long enough to understand your well-defined over-arching goal, explore your intricate puzzles, or immerse themselves into your awesome theme. This is definitely part of the "pacing" aspect you mention, but something that needs to be done well to make a dungeon crawler entertaining.

Good read!

Jeff Hildenbrand
profile image
I couldn’t disagree with you more. Classes and loot are completely not a necessity in these games and a lot of time are crutches that detract the from actually interesting gameplay. Would you rather delve around a dungeon where interesting things happen to the character or one where you kill things, level up and get loot, rinse, repeat? Now this is personal opinion, but I would rather my character be put through a string of interesting events and puzzles. I think a lot of designers get lazy and focus on these systems way too often. I don’t blame them. it’s easier to do because the creative muscle doesn’t need to flex as hard. coming up with complex and interesting gameplay is more difficult than coming up with the next +2 sword of magic snowballs or the next level stats. I also find that systems like loot and leveling tend to break the 4th wall and really detract from emersion. As soon as that level up marker hits you in the face, you are confronted by a system, not an experience. If designers focused more on gameplay rather than gameplay 'systems' I think we would be far more entertained and satisfied when the experience was over. Wow, sorry. I totally didn’t intend to rant. those are just my thoughts.

Michael O'Hair
profile image
Has this list of elements been applied to classic dungeon-crawler games (Wizardry, Might & Magic, Dungeon Master, ) which Legend of Grimrock is akin to? This is a great (and brief) advertisement for a game, but there isn't really much there to back up the bullet points. I'll lay one down for you...

From Point 2: "If you are creating a medieval/fantasy game, do not throw to the mix cultural references from other ages."

According to this point, naming a sword after a piece of modern kitchen equipment (Cuisinart, for example) would break this point and therefore break "immersion".

Here's the thing about immersion; your players will play games with differing levels of emotional investment and commitment, even the same players on different play sessions. One player may think up background elements and side-stories while navigating the dark dungeon to break up the monotony of movement forward and turning and resting to camp... while another player isn't so much concerned about immersion and environment, and their attention may be pointed more towards detailed spreadsheets and graph paper maps and inventory management and set encounter locations.

Nothing pigeon-holes things like rules such as "Thou shalt not mix and match sci-fi with fantasy, or other combinations, in games and stick to the source material". The result of following such rules is providing cookie-cutter experiences, of which there are already multitudes.

Trebor sux

Mark Ludlow
profile image
An interesting list, but not one that I'd say is a hard and fast list and is possibly a little subjective.

A lot of dungeon crawlers have strata to keep things different and encourage a sense of discovery in the player. Think of something like Diablo which starts off in cellars and dungeons, then progresses to underground caves and finally ends up in the nightmarish and twisted corridors of Hell itself. Having elements that don't belong add interest and contributes to the unwritten story. Why is that there? Who built it? Was there an invasion at some point? Even archaeology itself shows us that different cultures merge and mix over time creating an interesting tapestry of mixed technologies and themes.

However, it is important to be consistent within each of those sub areas and ensure you don't just randomly put incongruous things in without thinking about why you are putting it there. Number 3 is the same. Enemies should be consistent in an area but not necessarily linked across the whole game. It is expected there will be mixes of native and foreign enemies.

Having too many traps and puzzles can sometimes slow the player down and even turn them off if they keep getting back into the action only to be stopped by another puzzle, or die because of yet another unexpected trap. Same with pacing, the player needs some down time to explore and manage inventory or browse any literature scattered around. Sure, you don't want endless mazes with nothing happening, but you don't want to overwhelm the player either with constant action.

Robert Crouch
profile image
When I think of a dungeon crawler, the things that spawn to mind are games like Dungeon Master, Bloodwych, Legend of Grimrock.

The thing I think that defines these games for myself is primarily the mental (or physical) mapping.

I don't really remember them for their character or item progression. It's the feeling of being placed into a completely unknown situation.

- You enter a floor by a staircase, and you see the few walls around you. The rest of the level is a blank.
- You explore the safe area around the staircase, knowing you can return and getting your bearings.
- You venture a little further out, remembering how you got there and how you got back, trying to find the next goal, whether it be a staircase, or a way to open that unknown door, or get to that room visible on the other side of a grate with treasure in it.
- You become comfortable with the level, moving around it with ease. You've discovered where the monsters are and how to avoid them, or you've killed them. You know how the switches affect the environment, you've solved the puzzles, got the secret treasures, and found the staircase to the next level.
- You enter that floor by a staircase, and you see barren walls around you. The rest of the level is a blank.

That, for me, describes my own archetypal idea of a dungeon crawler.

The sweet spot is to create each level such that it is:
1) Initially unsolved.
2) Ultimately understandable and abstractable. (This is the level where I get the gold key. I need to open that door at the beginning of the level to get it. There's a series of secret doors whose positions I now know, and have marked by dropping rocks by them.)

Then the game should be comprised of these sort of abstractable levels so it's like:
- I choose my party on the first level
- I go down and get the food and items I need to survive
- This is the floor where there's a second staircase behind a door locked with the gold key
- Going down the unlocked staircase is the puzzle room with the staircase on the other side.
- The next level down is a room with pits and a staircase down. Down this staircase I can see the gold key through a grate. I need to go down a pit instead where I can reach the key and open the grate to climb back up.
- etc.

When the levels "goals" get too complex, I think it's less satisifying. If you can't really say "this is the level where I 'x'" then I think it's gotten too complex. I mean the game is still fun and playable, but our brains really enjoy it when we can wrap stuff up in a package like that. If the level's too complex, players minds will break up the level into sections, and you might lose that sort of relationship between the externally enforced structure, and the players own organizational structure. When they're in sync, I think it's more satisfying.

I think a good dungeon crawler is much more quickly completed the second time around. It's a game about organization. It's taking a large amount of unknown information and relationships and synthesizing it, whether by making physical maps or mental maps, and understanding the relationships between different elements. (This switches each open one door but close another door. To finish this puzzle I pull the first and third switches, get the crystal, then go back and pull the third and second switches to raise the pedestal where I place the crystal, then all the doors open.) The first time through you end up spending time clicking the switches, looking at how the environment changes, mapping a pattern. The second time, you know the pattern, you just flip the switches.

Replaying this sort of game is a little less satisfying in some ways (the puzzles are solved) but a little more satisfying in others (damn, I'm smart, I know all the puzzles). It's fun to replay them after you've had a chance to forget most of it, but remember bits and pieces.