I worked on a team for Ludum Dare for the first time. This is my postmortem for our game, Ultra Hat Dimension.
This is cross posted from the Ludum Dare web site.
Our game, Ultra Hat Dimension, is a puzzle game in which you are a hat designer. You won a prestigious hat design contest on another world, but at the ball in your honor everyone suddenly went crazy, aligned themselves into factions based on which of your hats they're wearing, and started attacking each other and you. So your goal is to get out of the palace. To do so, you have to get past the crazed guests, using hats to avoid getting punched backwards and to move people so they'll get out of your way and/or attack each other.
This was my first team effort, done with Woof (@woofycakes), Yuzuki (@yuzukimasu), and Eniko (@enichan). Woof was on art, Yuzuki on music, Eniko on programming, and I did the writing, level designs, and mouth sounds. As someone who always found group work incredibly frustrating in school, I was pleasantly surprised by how awesome it was to focus on one area and let the others take care of their areas so we could end up with a game which has been very well received.
We created a private IRC channel for ourselves and when the theme was announced set to brainstorming. Eniko and I are both very against doing the obvious thing, which we felt was regular game with [insert wacky weapon here]. Woof had an idea (Edit: He doesn't remember suggesting it, so I dunno who did.) which was too close to that for Eniko's and my liking, having a rocket launcher that shot animals which did the actual attacking, but we liked the idea of the player attacking indirectly. We ended up settling on "flags" as a weapon, using some kind of flag or other object to make enemies attack each other.
By this time, it was like 4 AM for Eniko (Netherlands) and she didn't need to do anything right away, so we sent her to bed while Woof started trying to figure out an interesting and preferably non-standard setting that he could reasonably make assets for within the time limit. I went to the library for a distraction-free environment and started working out exact mechanics while Yuzuki waited anxiously for more information because he needed to see our work to have inspiration for his work with the music.
This was midday for me, here in Japan, and evening/night time for Woof (U.S. East Coast) and Yuzuki (U.S. somewhere in the middle), so on that first day Woof gave us a good idea of where he was going to go with the art. I took on the job of figuring out how pac-man-esque-ghosty-people who attack by throwing things from under their skirts (that didn't quite make it in, unfortunately) and love hats end up attacking a hat designer.
I came up with the idea of the hat designer being from a different world to explain the difference in look between the Spluffs, as they came to be called, and Bea, our hat designer. The idea for the hat design contest flowed naturally from their love of hats, and a rival hat designer ruining the celebration of her winning the contest with a magic spell that turned everyone against each other flowed naturally from the contest. The magic spell also helped explain the hats being consumed on use. Why do the hats magically disappear? Well, duh... it's magic.
Yuzuki osmosed all the art information from Woof and mechanics and story information that I was talking about and started coming up with tunes. At first, he was hampered a bit, I think, by the fact that we had been discussing a banquet setting with things like eating utensils for our "flags" and although his first intro song idea was excellent, it didn't quite fit the mood. He ended up repurposing the melody later to be used as the end-of-game music, though.
Yuzuki and Woof stayed up too late working on stuff and I designed levels until I hit a point where I needed to talk to Eniko about how the mechanics would work exactly before continuing, then goofed off for a few hours until she got up. I got several levels done before I hit that wall, though, because I was trying to take a leaf from Portal's book.
One of the things that made me fall so hard in love with Portal when I first played it was that the game gives you at most one new thing to work with at any given time and makes sure you have to use it and understand it before you can continue on to more complex puzzles. My goal was to emulate that kind of teaching and avoid the wall-of-text instructions problem that I had with Worst Gnome at the Factory a few LDs back. So since I had to start with things like movement, key unlocks door, Spluffs punch you when you walk by, and Spluffs won't punch you if you wear the wrong hat, I was able to get a decent amount done before getting to other mechanics that I either a) wanted Eniko's input/opinion on or b) needed to be sure would be something she could implement in time.
Once Eniko woke up, we had that conversation about mechanics and implementation. Then she got to work using her custom framework that she's been making on to start on the game's engine. I pulled a random tile set off of opengameart.org so that I could make a test map in Tiled. Eniko needed to see what Tiled produced for JSON exports, and together we learned a lot about Tiled and how to use it.
I updated the wiki on the github repo with ALL the things and then went to bed. After that, it was mostly everyone doing their own thing, making sure we were all on the same page (or if not, figuring out what to do about it), and Eniko working her code-magic to weave it all together. We were never all awake at the same time for more than 10-15 minutes after that initial brainstorming session because of time differences, but there was always overlap and Eniko and I were able to ferry messages/details across the river of time.
All that said, here's a bullet point rundown of specific things!
People like it! \o/ Criticisms have been minor and we're getting lots of happy comments. I've seen a few people say they stopped in the middle for some reason but plan to come back to it later, and I feel like people wanting to come back is one of the biggest compliments a game jam entry can receive.
The game feels like a complete game, for the most part, astoundingly so for something out of a game jam. There are not as many levels as I wanted to put in, but there are 21 and we have a beginning, middle, and end. From what I've seen, the game takes people about 45-60 minutes to complete.
Everyone on the team did their part well. The graphics are lovely (I frelling love the hats!), the music is lovely, people have been enjoying the mechanics/puzzles, and although players haven't been praising Eniko's mad coding skills, they wouldn't be enjoying the rest of the things if she hadn't done her job well.
Although our team members are spread across the globe, the time difference helped our teamwork rather than hindering it. The programmer, who needs assets and mechanics to work with, slept first; the artist and musician had time to get started before going to bed; and I (game mechanics/level designer) had plenty of time to come up with puzzle and mechanic ideas before the programmer woke up. But the programmer and I also had plenty of overlap to discuss things whenever we needed to, with time between discussions for one or the other to work. No, the time zones were our friends, in this case, and I believe it helped us be more efficient than we could have been otherwise.
Windows doesn't like Eniko's custom framework for a reason she doesn't yet understand and the Windows build (which is the only one we have until she ports it to browser here soon) crashes on load for many people. There's a workaround that fixes it for most, but many people don't read the game description on the LD page.
Difficulty spike! Level 2-6 is the hardest in the game and there aren't enough other puzzles leading up to it. I probably should have made it last, but in terms of mechanical theme, it doesn't fit on the third floor. The third floor is about reactionary punch chains, which 2-6 doesn't use. This could be fixed with more levels, and that's the plan for the version we're polishing up to sell afterwards.
No browser version (yet). Eniko was exhausted by submission time and has been busy since then with things regarding promoting MidBoss (which was prototyped during LD25, has a free beta, and was Greenlit just a few days ago!) and setting up her business as a legal entity. The browser version should be up in the next couple of days, though.
I tried to design the puzzles to teach the players the mechanics and we included a chart in the upper left to show the strength/weakness pecking order, but the mechanics still aren't as clear as they should be. People do still get through the game, so it was a moderate success, but we have plans to improve this for the version we will sell later.
As always, being in Japan shortened the amount of time I could devote to Ludum Dare. In the past I've done the 48-hour Compo and had more like 36 hours to work. This time, I lost an entire workday and a half, about 12 hours of the 72, and wasn't able to find much time to design more puzzles at work because I'd been sick with a nasty sinus infection the week before and had catching up to do. I still got 21 puzzles done, but I'd been shooting for 30 and lack of more puzzles is part of why level 2-6 is such a difficulty spike.
Our artist got locked into the initial plan for mechanics and pursued asset creation for it in a one-track sort of fashion even though we wrote modified needs on the github wiki fairly early on. We were planning for multiple enemy types the whole time but changed what kinds of extra enemy types we were going for... and then got graphics for the initial type. I should have been more active in making sure the artist understood, he probably could have paid more attention to the wiki, etc., but we didn't end up implementing new enemy types anyway because the base enemy type and the punching and rock-paper-scissors mechanics provide a great deal of versatility by default.
Overall, we're really happy with our game. A lot was done well, especially given the time limit, but there are, of course, things that could be improved on both the teamwork and game design fronts. It's been awesome to see the reactions from other people. For me, personally, it's been great to have people I respect gushing about the game. We're hopeful about getting high scores overall. I went crazy designing more levels yesterday; the version we're going to sell later will probably have about 60 puzzles on four floors. The first floor has already had new puzzles added to help fill out the process of teaching the mechanics and give players more practice, and I look forward to making more. There's also going to be a boss puzzle in which Bea gets to confront Fab, the rival hat designer who cast the evil spell on the palace. \:D/ If you wanna try the game, you can find it... HEEYAH!
Read more about:
BlogsYou May Also Like
.png?width=300&auto=webp&quality=80&disable=upscale)
