FGJX Postmortem

This year I went back to Next Games to join the Finnish Game Jam 18. My friends participated in the last Ludum Dare and were pretty hyped about doing it again together. The plan from the very beginning was to use Unity and try to get something as decent as the game from last year, which we were pretty happy about.

Seeing our current group had expanded a bit, with another coder, Nico (who worked with us at Remedy) and Kim, who was going to do some design, this time I decided to switch gears and help Áureo with art. I usually go solo for Ludum Dare, and I’ve been trying to improve my pixel art chops, so I thought this was a good enough challenge.

Both me and Aureo were a bit hyped about doing a platformer because Pedro Medeiros just posted a couple of tutorials and we thought it would be pretty fun to do some cool animation. From my side, I already had some experience with platformer animation (from my first LD entry) and with the new Unity 2D tools looked like a suitable genre.

As always in this kind of team project, this postmortem is written from my own experience and with the learnings I got from it.

What went right

Aseprite: I had used in my last jams and I though it was good enough for the kind of art I can do, but spending the full jam working with it forced me to explore it a bit more than usual (and even had some time to do some learning). I was doing the environment tile set and I totally fell in love with how easy can you make tileable stuff without having to play around with offsets or how good the palette manager worked. Definitely it’s a worthy investment if you are doing pixel graphics of any kind.

Change of role: It was a totally new challenge to do art, it was a good opportunity of working side by side with Áureo, who even if not a professional pixel artist, is a proper artist and did a lot of work thinking on a good visual style for the game (and a super kick-ass key art). Also, working for 15 hours on art helped me get a bit better. I was pretty happy with the tileable rock/brick texture, but maybe invested too much time in it.

Technical art: Besides doing some environment art, I also did most of the Unity art integration and setting up the camera, etc. Unity has gone much better with 2D but it’s still a bit far from drag and drop. The animations specially were pretty tiresome to set up (we had a lot of animation frames in some animations). I think that if we had experimented with Aseprite Importer we would have decreased the time I spent in the Mecanim Animator window.

Placeholders: We realised pretty quick that we needed a placeholder art for everything. We did it for the environment, which allowed the level design to go forward pretty fast. Maybe we should have had them for the character animations so we would have had a shippable (or playstestable) version earlier.

Improvements from 2017

More energy: I feel had more energy throughout the jam (even though we had some questionable food choices). I had a party on Saturday, so I had to leave a bit early, but until the last minute we were pushing changes. Also, because it was at Next Games office one thing to remind for future is that not having access 24/7 you will have a bit less time than in an average LD.

Deployment: This time I I had a simple script to deploy the build. I didn’t end up pushing too many builds, but it was pretty easy to update the build, just exporting and running my Python script. At least this has been solved and will re-use the script in future jams.

Video: I am happy that I did a bit of video editing this year. I had wanted to learn to use HitFilm (a free video production tool, kudos to Baxa for the tip) but never had the excuse/material too. I may not had the best footage, as I didn’t make a special capture build with the right resolution to not avoid scaling the image, because I did it on Sunday afternoon and I was a bit tired after the jam, and wanted to just get done with this project and rest.

What went wrong

Theme: From the beginning, we didn’t really like it too much, specially since it sounded a bit too close to last year’s Waves. All the Transmission ideas that were thrown over dinner on Friday seemed a bit lukewarm. Luckily we stumbled in a diversifier that we really liked (Power Down) and spin out the game idea from there. The transmission concept got a bit “lost in translation” because…

More than we could tackle: (on a weekend) The art production speed that we could achieve was pretty slow. I think we need to find a more effective way of pushing out content for future jams. I think Áureo did a good job finding and assembling a palette/mood for the game and I don’t think the time we invested there (that wasn’t too much anyways) wasn’t well spent. My big error, I think, is that when looking for references I think in Super NES or Sword & Sworcery and those are super detailed and high quality pixel art. I need to do some research and find art styles where I can do more and faster.

Starting flow: We had the tutorial levels in the planning but fell out from the tasklist. Our original idea was to teach you everything with simple levels and, then, start the “real game” where you have to start giving your skills away. After the first lunch, we started to worry if we could design so much content, and we decided to create the levels backwards, with the idea that we’ll have something shipable all the time. I think it was the correct idea, but we didn’t had any Plan B for the Tutorial in mind and we didn’t event tell people the keys in game.

Game difficulty: Kim did the levels and I think he managed to do something decent pretty fast, but maybe the levels are a bit too hard for a gamejam game. The project was so big that we were pushing features all the time and we didn’t do enough playthroughs to get the game to the right level of challenge. Also the slow (but awesome) death animation doesn’t give you a fast enough retry for such a challenging game. With such a difficult game, it felt as a feature that the game doesn’t fully reset so it’s a bit easier than designed.

Needs to improve

Áureo didn’t had a build: He liked in the previous LD, where they worked with PICO-8, because he could iterate the stuff in the game. This time at least I sat besides him and I put his stuff in much faster, but I would have liked to help him set up the environment beforehand.

Not the best food choices: Jasmin took care of getting us some pretty healthy snacks but with such a big group settling in a place to lunch was a bit hard. So we ended with kebab and burrito. And on Saturday night I was on a party so I think my diet on that weekend had some room for improvement. Maybe I’ll prep some lunch for next one.

Conclusion

I think, even though that we didn’t love the theme and we didn’t got an idea that we loved, we were able to finish something we could deliver.

I’m not as happy with the outcome as last year, but I had lots of fun in this jam and. People from Next Games are super cool to hang out with. I intentionally set up this jam as a challenge to learn more about art, and in that, I succeeded.

Also, working with so big groups in a jam was pretty challenging and for sure we will have to learn how to scope the projects better and how we can work more efficiently now that we are so many in our group of merry jammers.