Invertebrae: a Post-Mortem

Editor’s note: this post is about a game I created for Ludum Dare 44, and was originally lifted from there.

Ludum Dare 44 is over! It’s actually been over for a couple of days now, so let’s just get to it.




This is my third Ludum Dare where I actually completed a game, and as I continue to improve as a dev, my Overall score also continues to improve. Extremely happy to be in the top 50–next time, maybe the top 10?

Jokes aside, it makes me happy to see that a text-adventure game can do this well. Glad to see that the LD audience is diverse in their interests.


Oof. I know entries that stretch interpretations of LD themes typically don’t do too well, but I definitely was hit pretty hard. In case you’re wondering how I feel my game tackles the theme, I’m going to actually cite what my friend had to say, as he put it way better than I could:

Your life is the sum of your earned experiences. You have to manage how you spend each word because they are the currency you use the act.

In that regard, as we’ll talk about later, the game even punishes you for amassing too much currency ;)


I had gotten a lot of great feedback on Invertebrae from friends, family, and of course the Ludum Dare community. While some of the feedback I had expected the moment I hit the publish button, there was also completely valid criticism that I hadn’t seem coming at all.

How Long Is Your Game Really?

Nearing the end of the compo period, I was running through Invertebrae nearly every half hour, making sure that grammar was sound, the UI never bugged out, and that I hadn’t inadvertently created any soft-locks with the puzzle mechanism. Each of these playthroughs would take me roughly 5-10 minutes, so in my head, I just pegged my game as a 20-minute game at most.

Imagine my surprise when the comments started coming in with playthrough times of 30 minutes, an hour, two hours, heck my friend even stopped playing to go to sleep and then continued it the next day.

First off, it makes me so happy that anyone would play a game of mine for more than like 20 minutes, much less two hours. The idea of creating something that could hold someone’s attention for that long really is some of the nicest feedback I can receive as a game developer. Plus, let’s be frank–most Ludum Dare games can be fully experienced in 5-10 minutes. In the time it takes to play a one-hour text-adventure, you could have rated like ten other great games.

However, not knowing that my game would have such a playtime means that I didn’t design the game for such a playtime, and that was definitely reflected in the criticism. As I addressed in a post-compo update, the core drag-and-drop UX is pretty, but also pretty inefficient. Dragging words back and forth leads to a lot of unnecessary movement, and this is only exacerbated by the large number of words you could amass.

The Moon Is Exponentially Far Away

Due to one of the core puzzles in the game, a lot of people ended up hoarding a lot of words. This was definitely not the intention–the multi-page feature of the word bank was actually only added in the last five hours after I realized it was even possible to collect enough words such that they went off the bottom of the screen. However, with puzzles like open computer that rely on you having to re-use words you played prior, and navigating between nodes of the story being a slow affair, it was silly of me to not expect players to store every word they saw in worry that it would be necessary later.

Instead of puzzles that are designed to be worked out from a pool of four to seven words, now even the simplest puzzles are obscene, with tons of accidentally reasonable combinations of words not satisfying the one or two answers I had in mind. Plus, the moment you learn that some puzzles have multiple answers is the moment you start thinking that all puzzles have multiple answers. I tried to poke fun at this with things like the stairwell ending, where you play pet cat after walking on a stairwell that squeaks like a dying cat”. But while that ending is quite funny (well, only 190 places of funny as the results showed), the hundreds of non-existent endings that players tried to get to certainly aren’t.

A lot of people had noted that a couple of the puzzles were really difficult. While I agree that a few were a bit obtuse, I do think it was primarily due to the excess of words that people could collect. Even sort of sketchy phrases like pretend security are way easier to reason if they’re only two of five words in your word bank.

If I were to work on Intevertebrae in the future, I would definitely design the puzzles to have fewer words overall and give each word more uses. Not only would this decrease the frustration of cycling through endless combinations, but it would also increase the number of actual word puzzles, which is the real difficulty (and source of fun) that I wanted present in my game.

OK People Said Good Things Too

I’ve gone on about the most popular critical feedback for a bit now, but people also plenty of nice things to say:

Issues aside, I’m really proud of how I did this Ludum Dare, and I’m looking forward to the next one. See y’all at LD45!

2019-05-23 · gamedev · ludum-dare

Previous:Invertebrae: A Pre-Mortem
Next:Existential Types in Typescript Through Continuations