First, a quick aside because I'm really excited to share...

I'm making a real game!

It's called Moonsigil Atlas, and it's a deckbuilding roguelike with a geometry-based resource system. There's no conventional mana or action limit - your only restriction is piecing together the shapes of your cards in the space you have.

If this looks a little familiar, it might be because it's based on the main mechanic of my LD52 game, Fallowtide, but more in the direction of a traditional roguelike deckbuilding game like Slay the Spire. I've been developing the full version with @daniel-alhadeff for the last several months and we're really excited to start sharing it with people!

There's no Steam page up yet, but if you want to stay in the loop for the development, release, and potential demos or playtesting opportunities, please join the Discord server by clicking the banner!

Onto the postmortem!

The postmortem

This year, my LD entry was My Keyboard is Full of Ants!, a game about your keyboard being full of ants.

Choosing an idea

In a couple past events, I've considered making games where the location of keys on a keyboard has gameplay relevance. This is partly because I think it's a fun design space, and perhaps partly because my first ever Ludum Dare game tried this but was not fun at all, so it's an opportunity to take a second crack at it.

"Tiny Creatures" turned out to be a great theme for a keyboard-based game, because you can contextualize it to the player's real-life keyboard, which could have tiny creatures living in it! The concept of ants living in your keyboard and typing in order to smoosh them struck me as more funny and appealing than it ought to be, so I ran with it and used it to inform the specific mechanics and other design decisions.

The good

Several things went well this game:

• The graphics: I went a fairly safe direction with the visuals, but everything came together into a clean whole, and I'm happy with the consistency and clarity of the final product.
• The bugs: There are tons of bugs in the final game, but they are (almost) all intentional. There was no huge technical hurdle or major bug that took up a lot of time, unlike last event where I spent a lot of time figuring out hosting.
• The web build: I continue to get much more traffic to my LD games on Itch with a web build. Even though it's a little bit sketchy to do with Python/PyGame, it's proven to be worth the effort every event I've done it. For a little while, I was in the top ~30 games on Itch with the Ludum Dare 56 tag, which would be way harder with a download-only game!
• That one super obscure joke absolutely nobody saw but I'm way too proud of: The soundtrack has the word "ANTS" spoken a couple times, but I also included a "karaoke version" in the build files that is identical without the word "ANTS." Worth it? Absolutely.

The bad

• The secondary game loop: I think I spent a bit too long fleshing out the core mechanics and adding enemies and power-ups that the game's high-level structure suffered. The game is level-based, with a fixed ending, and it's not possible to lose --- you just get a better or worse score based on how efficiently you squash ants and apply multipliers. I think this is a less compelling way to play than, for instance, a wave-based game where your goal is to survive as long as possible, even for people who care about high-scores. However, in the time I had, it seemed like the best way to get the average player to see most of the game's content without extensive balancing.
• The humor: It wasn't bad per se, but I think I could have leaned further into the game's absurd premise and found more areas to inject jokes and comedy (I think I did this better in LD53). As it stands, I think the game sits in the middle of the road in a lot of respects, including humor, without excelling at any one thing, and humor would have been a good opportunity to take to the next level.
• The play-and-rate: I've had an extremely busy few weeks, especially with the development of Moonsigil Atlas, and have not been able to play and rate nearly as many games as I'd like. That unfortunately means there's no fan art compilation this time around, although I did draw a couple quick sketches for some games you may find floating around!

Takeaways

• Be deliberate about your game's identity. I think I found an appealing and consistent visual direction, but failed to make strong decisions on more fundamental parts of the game. Is it trying to be an arcade-style high-score game? Is it trying to appeal primarily through humor? Are they supposed to be pharaoh ants or grease ants? On game-jam timelines, you usually want to do one thing really well and use the other parts of the game to support it; but I never quite figured out what I wanted this game to do best.
• Take the time to make systems extensible, when it's worth it. I very early on made the system for text effects when you smoosh an ant, then used that constantly for bonus points, multipliers, and special ant effects. This saved me a bunch of time and also made the game more usable and legible for the player when a lot of things are happening. You do need to be careful about this on game jam timelines though --- sometimes it makes more sense to do the hacky solution in 20 minutes instead of the clean one in 2 hours, especially on the last day!
• Brass instruments are hard to play. Every time you beat a level, you get to hear my very underwhelming bugle playing. I'm thinking I should stick to piano.

The Results

This game did wildly better in results than I expected!

Thanks so much to everyone who played the game during the rating period. Ludum Dare has been an amazing experience over the 7 years I've participated, and while results are far from the most important part, it's extremely heartwarming and validating to get positive feedback because it means people actually enjoy the games you make.

Interestingly, I expected Graphics to be a strong point and Humor to be more of a point for improvement --- I think the results just go to show that a) sometimes you're not always able to view your game from an outside perspective, which is why playtesting is so important and b) there's enough randomness in LD results, especially in the top ranks, that the specific numbers don't matter as much as high-level trends over multiple events.

Congrats to everybody else who participated, and especially those who made it into the top 15 --- every game ranked that high is fantastic and just as easily could have taken the top spot, and I wish I had the time to play all of them. (If you haven't played Prince Charmony yet and like puzzle platformers, play it now!)

Thanks for making it all the way to the end, and here's to an even better LD57 next year!