Link to game page
Our jam entry “Corponnected Ltd.” is now in the wild, and first feedback has been very positive! To hear people saying they actually finished the game / played it for half an hour is incredibly heartwarming. Here’s a quick postmortem while memories are still fresh:
Before the Jam
This week-end was my fourth LD, and the second one with teammate Manu. This time we managed to make a 3rd person, Guillaume, join the team, so with our extended workforce we felt like the sky is the limit! With our little experience with previous jams we knew we had to avoid making a too complex game, so we planned to make something that could easily start small, and then be iterated over.
We chose in advance to make the game with Phaser.io (HTML5 engine), so we could practice with it a bit before the jam (before that, our engine of choice was CraftyJS, but while we love its API, some annoying issues made us want to try something else).
Jam progress
We were quite lucky in that it took maybe 5mn between the moment I discovered the theme, and when I figured out what I’d want to make. Our concept, a factory building simulation, is actually borrowed from a game idea that has been in my head for a while. I also liked that management games are usually nothing too complicated at a technical level (or so I thought), leaving more room to focus on game design and balance. We actually discussed so much about the game mechanics that we didn’t start coding until maybe 3 hours into the jam.
When getting started, we quickly realized things were more complicated than we thought (such as the workers AIs), so the “let’s make a small game and iterate” thing kinda falled through… Actually it’s much harder to work in small steps when you’re a team of 3: the most immediate way to split the work was instead to cut the game into subsystems and make each person responsible for a part, so the whole game is actually developed concurrently. In our case here’s how it was initially split:
– Wan: Factory building screen / Workers AIs
– Guillaume: World map screen / Level design
– Manu: Game state & contents modeling / Save system
This worked pretty well, but the side effect was that the first “working” version was barely ready for sunday evening. At that point most of the game was done though – it lacked a lot of polish (like a tutorial, an endgame, music, etc.), but the engine was up and running! There was a down side though. We had huge balance issues on top of the lack of polish & contents, so we were still really worried about one thing: will the game be actually fun?!
Monday was a work day so we only had an evening left to actually turn this first version into a fleshed-out game. We were really tired at this point, but every bit of work put into the game during this final rush made us increasingly excited, as we saw everything falling into place and the game mechanics starting to make sense, turning up the game into a surprisingly cohesive experience.
By the time the game was done (about 2 hours before the deadline) we were pretty satisfied with the entry, and decided to call it a day – though Manu still did some bugfixes after that! At this point he was the only one who actually played the game from start to finish, so even I wasn’t sure how playing the game actually felt. Even at this point, we were still worried about whether people would like the game: we barely had some external feedback during the jam, so we weren’t even sure whether the tutorial and the building descriptions were enough to make the game easy to pick up.
Fortunately the feedback was eventually positive, and the main complaint instead was that the game is a bit too easy – which makes sense, especially since we completely forgot to balance the feature (monthly taxes) meant to complicate things! But overall we’re really proud of our game – I even prefer it over the one that made 2nd place during LD24. We also really felt like the experience from the previous jams played a part in this session going well… I love it when a plan comes together.
Went right/Went wrong list
[+] No major issues met, we had a nice time overall
[+] Discovery of PhaserJS was pleasant, this thing just works. Docs are a bit weird to apprehend, but otherwise we really recommend it. Even integrating GLSL shaders (animated water) was flawless.
[+] Task management and distribution went well
[+] People liked our music, despite it being the most hastily and dreadfully patched thing I’ve ever made… (Can you spot the game from which I tried to mimic the soundtrack?)
[-] We should have taken our Monday off!
[-] Initial scope for contents was way too big – a lot of planned content (factory types, etc.) didn’t make the cut
[-] The UI was made within the engine, but we eventually realized that most parts could have been made much more easily in plain HTML…
[-] Still lacked some time for polish, the engine could have supported even more variety and challenge while making factories. Some features such as having terrains of different sizes, or really complex terrains weren’t really leveraged in the levels.
We had an unusal amount of ideas for this game, it would be nice if maybe we could turn it someday in a full-featured game… But y’know, that’s what we say after every jam!
Tags: postmortem, rockets with octopuses on top
Some might know me from the Minecraft community as Jimeo Wan, the guy behind a mod called Inventory Tweaks ; I also won the Java4K competition this year (look for “The Little Scientist“). Now this is the first time I get to make a game by myself within a limited time, so I’m not very confident about all this – but I’m pretty sure this will be loads of fun, and I guess that’s what matters 😉
