If you know me the slightest, I think it’s pretty clear that I love adventure game jams. The 14 day format is definitely my comfort zone, and I may be delusional but maybe there’s someone out there who’s interested in how I approach these events, so… here’s a recap of what I did during the $108 Adventure Game Challenge, where I submitted The death of Fo-gu.
I knew I had to be as fast as I could, because I had a very busy schedule those weeks. I wasn’t meant to be home on Saturday afternoon, Sunday, and we were leaving for a vacation on the last Friday. And since I usually produce the most on weekends, I was scared I couldn’t make it. Luckily I got my first Saturday back, but still I had to sprint to finish the game in 10 days and submit on Thursday.
My plan
For the $1xx Adventure Challenge I usually spend the first day away from my computer. The very specific theme means I really have to design everything carefully first before even thinking of drawing or implementing. This time I was forced to do so as I was traveling to my office, so there were no temptations. While on the train I discussed some plots with my friends and let them brew for a while. Here’s what I had on my google doc that morning:
- Office setting, big guest coming, you’re given a secondary task because the boss fears you’ll mess it up, you end up saving the day
- Religion, you’re the only one who doesn’t believe it but you’re right
- Food, everyone cooks crap and you’re the only one with taste
- Frankenstein monster in a group of monsters who bully you because you have the brain of an idiot but it turns out they don’t even have a brain
- You could be a clone or a genetic experiment
- A robot who everyone bullies because you’re “just a robot” but you’re self aware (but you don’t know you are)
During the lunch break I gave some time to each one, imagining the story. I could imagine funny events and lines for office, monsters and religion. However when it came to plot and puzzles, I had this feeling that the religion setting could give me more freedom while still sticking to the theme, so I focused on that one. At 15 I had envisioned a three part game (titled repentance, rebirth, wrath). I always try to make my games in three parts. The bottlenecks are a good way of giving the player a sense of completion, but most importantly they’re a good way of giving yourself the chance to cut the game short if you overscoped.
I wrote down a bunch of settings and puzzles I could use, although they were all disconnected. I have an iterative approach to puzzle design, sometimes I think of the mechanic and try to fit it into the setting, sometimes I think of the objects that are needed for the plot to advance and try to add obstacles. And if I find some matches, then I make a chain, and so on. When I have enough elements, I write down the chains in nomnoml (https://www.nomnoml.com/) and see what it looks like. I usually try to have some parallelism in the chains, but I also try to keep in mind that I’m definitely going to revise the puzzle chains during development. So even if I had my graph ready, I knew it wasn’t set in stone.
But since I had my setting and I knew roughly what act 1 was about, I started drawing. By end of day 1 I had my first room, the protagonist and the Fire Priest.
On day 2 I started implementing. I usually go full detail on the first room / bottleneck, because if things go south I’m able to deliver a playable game anyway. In fact this technique saved me for The Will of Arthur Flabbington, because after two days in AdvXJam2021 I knew I wouldn’t have time to work on it so I just focused on polishing act 1 as much as I could.
I try to add a different kind of puzzle in every jam game, so I spent all my efforts in trying to implement the water dripping mechanic. I’m very happy with how it turned out, and now I know I can use a “footstep trail” kind of puzzle in my next game if I want to. And since it was a very particular puzzle, I decided to leave it as basically the only puzzle in the first part. I wanted to move on quickly.
After my “playable and slightly polished first act” is ready, I usually try to bring the game to a completable state as quickly as I can. I spent day 3 creating rooms full of placeholders just to be able to navigate the world. Not really programmer art (except for the map) but empty rooms with no detail. Once the world was fully connected, I tried to tackle down everything one chain at a time. For example, I needed to add the training camp interaction… so I drew everything I needed. I drew the warrior and his idle animation right away, I drew the room, and if I needed an inventory object, I added it to the code and drew its icon, then went to the room where I wanted to put it and added the required room props. Then I coded the interactions that made the thunder puzzle solvable. Once it was solvable, I moved on to another puzzle chain.
After three days act 2 was completable. The rooms went slowly from empty to full with the needed characters and hotspots. Once the second act was solvable I focused on polishing it instead of working on act 3, for the same reason as before: should anything happen, I can still deliver a two act game. Polishing means adding extra “useless” objects, dialogue, descriptions, signposting. When I do so I often get new ideas that make me tweak the puzzles some more. For example, the water quiz was just three easy questions, it became five after I noticed that I could integrate the newly created mythology into the puzzle.
Act 3 was very short, so it took me less than one day to implement. By day 6 my game was completable. I started adding animations and cutscenes, and the morning of day 7 I sent my first alpha to my friends and went for a trip with my family. When I came back I had enough feedback so I spent the evening adding signposting for the puzzles that weren’t that clear and started adding sounds.
I spent a lot of time on the music, because I didn’t feel inspired. I wanted something that gave a “prehistoric” feeling so I tried to restrict myself to pan flutes and simple percussions. But I’ve never written anything like that, so it felt flat. After a day or two I decided to stick to my mantra of “good enough for a game jam” and moved on… and it was indeed good enough, I guess. People seemed to like it.
Since the game was done I kept tweaking a few things but I was also extremely tired at that point. After 10 days, the game was ready. I submitted and left for a much needed vacation.
My takeaway
I’m very happy with how things turned out. I was able to add a new type of puzzle to my arsenal, and I like how the quiz puzzle makes the story an integral part of the gameplay – which is also a kind of puzzle I never had in my games, now that I think of it.
This time I didn’t hide my logo anywhere. I was too tired to think of a clever way to do it. Maybe if I had the last three days I’d have done it, and I would also have added some easy animations like picking up objects or the down / up animations. But I left a small Easter Egg for my Sardinian fans (all two of them) – all deities have Sardinian names that refer to their element. It makes very easy to remember, at least for me. It was very fun to watch people play the game and struggle to remember who each was.