Hello! First of all, thank you for stopping by. We would assume that you've played our game too, so we thank you for that too! If you haven't go try it out!

It's a little bit weird to call this a post-mortem, because our team still has something in the works (we'll get to that later), but technically, as  a game jam game, it's complete! We'd like to tell our experiences of working on this project.

Preface

PhaseJam has been going on for a while now, and it has reached number #5. Our team skipped #3 and #4 because we were occupied with something else, and as the timing provides, we had the chance to free up schedule to join #5. When the theme was finally announced, we immediately decided to make a puzzle game (again) themed around Jelly Hoshiumi (again), but the big question then was "what kind?". After some brainstorming session, xeliosphere showed a sidequest game from the mobile game Wuthering Waves called Encryption Blocks, and that's where we lifted the "light up tiles with a movable light source" idea. Then, peppy added the idea of turning it into a sokoban, which made the game at least two degrees more complex, both gameplay wise and programming wise, but in hindsight it was a super genius idea. We then got straight to work, but learning from Jelly Labs, where we crunched the last few hours making and testing levels, we decided to have some extra hands in this. Peppy brought in some clever Starknights (Meow Mix and Sheepiro) to be dedicated level designers. In retrospect it could have been overkill to have five members in a game jam team, but, well, we couldn't have finished if we didn't do that.

What went right

Plenty went right, really! Building up from our experience in making Jelly Labs as well as various prototypes/other works between then and now, we've sort of found a good 'flow' in working on a game project. Those include:

Reuse and Recycle

A lot, and we mean A LOT of elements in Save the Con was recycled from old projects. The most obvious example is the menu system, as we literally just copy-pasted the components from Jelly Labs and tweaked it to work with Save the Con. Art and audio too, as we ended up reusing some assets and sfx, first as a placeholder, then permanently because we thought it's fitting enough. The unsung hero of this game is the gridmap and the undoable movement command system, as it was frankensteined from shelved prototypes we attempted to make (a tower defense and a card game, respectively). This sped up development so much that we were able to have a complete flow of the game (start up, level select, one playable level) within the first day of the game jam.

Borrowing Wheels from Others

Being a sokoban, Save the Con definitely needs some sort of a level editor to, uh, edit levels. A super-efficient, competent team would have made this themselves, but we are neither competent nor super-efficient, so we opted to use Tiled Map Editor to serve as our level editor. It pretty much has everything we needed, such as tile/object placement and layering, as well as custom classes and enums. The most important thing was it was able to export to JSON, so we were able to put the files natively into our game engine and read whatever Tiled exported to correctly generate and render a level. This also allowed our level designers to work in parallel with the game development, as they could make modifications to their levels without breaking any work in progress, and vice versa. We were able to create 20+ levels within the jam timeframe and we'd consider that as a great success.

Extra Time + Extra Members = ExtraExtra Polish

PhaseJam #5 was quite long (about 9-10 days), and was made so to accommodate smaller teams and participants with bigger life commitments. We decided to use ALL of that 10 days. The core gameplay, as mentioned, was effectively done within the first day, so we were confident that we  could expend effort to make the game play and feel as best as it can. As we're already using the exact UI system from Jelly Labs, we could reuse animations and popups we've made before, and add improvements in places that needs it (screen transition animation is one example). Both peppy and xeliosphere went above and beyond to deliver art and music to spice up the game; peppy provided all art assets within the game and made TWO comics to serve as intro/outro cutscenes, and xeliosphere made FOUR unique arrangements of Jelly's available themes. Meow + Sheepiro duo busted out their brains over multiple days to curate those 20 levels that tingle your brain just right. Magna also managed to create a save data system (to avoid a certain someone breaking it again) as well as virtual controls to support mobile gamers. All of these combined resulted in an end product that we're all super proud of. 

What went not-so right

Looking back, there wasn't much that went OVERLY wrong, but some parts we think could definitely executed better:

Confidence + Time = Overscoping = Burnout

We didn't really need a comic panel cutscene. We didn't really need four unique BGMs. We definitely didn't need to make an exportable save system either. It's a game jam project after all. But we did anyway, because those sounded cool on first pitch, and we thought we had enough time to do it. And that we did, but it was at the cost of some 11th hour crunching to make sure everything was in place. We could have used that last period of the jam to leisurely test an already working game, but we ended up doing some frantic bug hunt instead. It wasn't really that bad, but it could have been better.

Here's a direct quote from peppy when I asked her about her experience in this gamejam:

"i would like to be unemployed" - peppy

What's next?

Well, Jelly has played our game! She also has expressed interest in revisiting to finish the game. She also broke the game on stream in front of a live audience worth hundreds of people. Again. So our immediate target is to fix these bugs and add some few more optional levels for the game and deliver that as v1.1. We haven't decided when to do this, but it's no sooner than after the PhaseJam #5 voting period ends. 

Some players have expressed interest in having a public level editor and to play custom levels. While we do have the technology to have a public level editor (we can simply release our Tiled Map Editor project files to the masses), we don't have any way to read levels on runtime just yet. Right now the JSON files export needs to be put into the game engine editor and built within the game. Considering Save the Con runs in JavaScript, there should be a way to have user upload a file on runtime and read it from there. But this is sort of unexplored territory for us and we might have to put this back for v1.2. Hopefully.

Anything else?

Aside from those two planned updates, nothing! The game has potential to be scaled up to become a 'big and proper' title, and we have some extra ideas and mechanics thrown around here and there, but we realize that doing so would turn this project into a way bigger commitment than a game jam, and that is a scary step to make. We'll reassess after we deliver the updates we have in the works, but honestly, don't count on it too much.

Endface

Once again, thank you for stopping by and for playing Save the Con, Jelly! This has been our best work yet, and hopefully you enjoyed playing it. If you have any feedback or ideas, please don't hesitate to leave comments in this post or straight at the game page. Thank you for reading this log/mortem this far too, and we hope to see you again on the next chance! Do check out the rest of the PhaseJam #5 entries too!

-/skk/ Atelier-