Finally done with content! Now let's polish and test a bit more for the final hours.

Dungeon Break is a puzzle game, where each level is a one-screen puzzle set in a magical fantasy city. Escape the dungeon in which you are imprisoned and explore 10 levels including a wine cellar, clock tower, the under city and even the throne room!

Still not convinced? How about this: In Dungeon Break you can even customize your character!

Edit: Ooops. Seem to have forgotten the link! Here you go.

Try Dungeon Break - the game where you can equip the playable cat-character with a magenta hat or an iconic yellow raincoat!

Do you chose the blue portal or the red portal?

Going in to the jam we had only one restricting: we wanted this jam to be a bit more chill than some we had participated in in the past. With two people (3D Arts + Programmer) our team was also significantly smaller than the team we usually work with, so we wanted to keep things nice and small and get enough sleep.

Part 1 - Choosing a project

The theme 'stuck in a loop' wasn't our favourite going into the reveal because it felt rather predictable. What could you do with it, other than some version of "groundhog day, but it's a game". However we quickly went through a few ideas and found our self intrigued by a few of them:

• 'A days work' would have had the players do a precise and tedious sequence of actions in a factory setting in order to produce as many products as possible in a set amount of time. In the end the player would have been paid depending on the amount of products produced. This money would then have to be spent on items such as rent, food or medicine for the player and their family (think: papers please) and there would be consequences if one of the items could not be satisfied. The game would have had different endings depending on players choice (e.g. do you join a union and strike for better working conditions or is the potential punishment in form of less immediate pay something you cannot afford right now?), but in the end we weren't sure we would be able to do the idea justice in a weekend and felt unsure about making the core game loop tedious by design.
• There were a few ideas that took "loop" simply as the name of a location, say a space station and would have probably ended up being some kind of escape game.
• We thought about having a main character design where the character would always wear a rubber ring with a flamingo head around their waist.

In the end we settled on the idea of having a puzzle game that would teleport you back to the beginning after a set amount of steps. An inspiration for this was 'minit', but we wanted it turn based rather than time based and with more focus on puzzling rather than exploration.

Part 2 - Prototyping

Before we started coding we wanted to make sure that the elements we had thought of at the time (Levers, Gates and sinking stones, that would return after a while or could be raised by a lever) would be able to make fun puzzles, so we started sketching out puzzle designs with pen and paper. The one that convinced us that this was sufficient to make fun puzzles looked something like this and a slightly modified version of it actually ended up in the game (it's level 5 - the training room).

draft of level 5

Now that we were convinced, we started coding. The base functionality was done pretty quickly and we were able to fully design levels and play through the prototyped level 6 hours into the jam.

fully functional prototype version of level 5

Part 3 - Creating Content

*Puzzles*
We had decided early on that we wanted to have 10 levels, giving us time to introduce mechanics, have a few easy levels to make people familiar with the core concepts, a few medium puzzles around the difficulty of the one we had originally prototyped and a few really hard ones to give everyone a challenge. Between the tutorials and the prototyped level we had 3 already and quickly added 2 easy levels and one rather difficult one for a total of 5.

In order to mix things up, we looked at additional mechanics we could add and settled on portals. Portals were great for a few reasons: They can be activated by levers and they can create shortcuts, allowing the player to travel further within the step limit of the particular level.

portals at work

In the end we ended up creating a couple of really hard levels in order to give players a real challenge that was worth sitting down and really thinking about.

*Art*
We knew that we wouldn't have the time to animate much, so we decided to have a boardgame style player character, that would look good without animations and just setting new player positions rather than moving the player character over time. This had the added upside of allowing the character to move as fast as the player is able to input stuff, making it feel snappy and less tedious, especially on the larger levels.

the player character

We also knew that we wanted the game to run decendly as a webbuild, so we'd have to be economical with verticies and thus decided to go with a textured style, relying on normal maps for the athmospheric dynamic lighing.

We ended up with a decend amount of assets, allowing us to use some of them selectivly and giving levels a unique look and feel. These assets were created gradually, as we didn't know what decoration items we would need or even have time to place from the get go, so we started with the functional items: Platforms, levers, gates and portals. Later we added walls, torches and railings in order to kit out the scene and in the end added smaller assets such as tools, barrels, banners in order to add detail to the scenes.

All 3D assets used in the game

Actual geometry

Part 4 - Decorating the levels

Now that the game was working as intended and we had the full 10 levels we were going for it was time to make them look really pretty. This escalated a bit. While we started out placing a torch or two and a few backwalls we soon realized we wanted to give each level it's own unique character. We went through all levels and decided on themes and specific settings and started to decorate them accordingly. This was by far the greatest time sink in the jam and took us until just before the submission to finish.

At this point we also started to look at lighting and particle effects. Creating a basic fire effect and a magical portal effects for the portals helped fill the levels with believable lighting and made them more colourful.

final decorated version of level 5

Part 5 - Polish

Since we had started giving the levels specific settings we had created a simple narrative that would frame the path through all those locations so we ended up creating two simple cutscenes. We were again able to profit from the lack of animation on the player character and the fixed camera position, as it allowed us to create these rather quickly by stepping through a simple set of states that would set player position, environmental assets, subtitles etc. without dealing with complex timings, camera angles or animations.

We created the Main menu in a similar fashion, making it loop to reemphasize the jams theme and giving a few fun QoL options such as customizing the material colours on the pc. Especially since this is a jam game we decided to also give players the option to jump straight to any of the 10 levels via level select, allowing them to skip levels they were stuck on or pick up where they left of after restarting the game.

Next we added sounds to key actions (gates opening/closing, platforms sinking/rising, stepping, respawning and teleporting) and also picked a few tracks to play during the levels to emphasize the mood each level aims to convey. We did not create either of those ourselves, but relied on a library of game music and sound effects we often use in jams or while prototyping, even thought some sound effects had to be altered and cut a little to fit the game.

Part 6 - What we didn't get to

Surprisingly and in spite of being a team of only two we didn't have terribly many things that we ended up cutting. Here are a few things we thought about adding at some point or another but never actually decided to do: * patrolling enemies that need to be avoided (moving according to simple observable rules, as in 'Rocks'n'Diamonds') * having an antagonist that would feature in the cutscenes, giving a bit more backstory to the player character and revealing why they were imprisoned * upgrading our logic grid to work in 3 dimensions rather than only 2, allowing for puzzles where you would lower a platform and then use it as bridge on a lower level * a fancier level select screen

The Road Ahead

People often try and turn successful jam games into full games and while we have definitely thought about doing this in the past we never actually ended up doing it, mainly because jam games usually rely on a central gimmicky mechanic that doesn't scale very well to a full game. This entry however could work really well on a larger scale, with a few more mechanics and a ton of extra levels, so once we have caught up on sleep and have processed the feedback that will come in we might have another look at the game and where it could go.

If you want to give Dungeon Break a try, you can do so here.

Hey everyone,

I'll play some entries on stream in ~45 minutes (8:30 CET).

I don't usually stream, so don't expect many people to watch it, but if you want to watch me fail at your game just comment here or come by and post the link to your game in chat.