Star Power Postmortem: The story of my worst entry so far

If you haven’t played Star Power yet, you might be interested in checking it out here. It’s not spectacular, but most folks agree that it’s pretty novel, so you might still find it interesting. Anyway, here’s a quick summary of what went well, what went wrong, and, most importantly, what lessons I can share.

What Went Wrong
Unfamiliar Development Environment – Star Power is my first LD submission not made in Game Maker, my tool of choice for gamedev. The reason for this was that I chose to participate in Ludum Dare with the wonderful folks of the Tech Valley Game Space, which meant working from my laptop. Which does not run Windows. I learned the hard way that Mac OSX is not meant for game development. The tools which are of critical importance to many LD participants either aren’t available for or don’t run well on Macs. This is most apparent in Star Power’s hideous graphics, which were made in the embarrassingly sparsely featured free version of Pyxel Edit

Imbalanced Gameplay – If the player is given the opportunity to rate their opponents, they will of course want to give the opponent nothing but one-star ratings. This was something I knew of from the very beginning of Star Power’s development, and I tried to counteract it by granting future buffs to enemies who receive one-star ratings. What I failed to account for was the fact that those simple numerical buffs would quickly counteract one another if applied to multiple stats until they became fairly meaningless. As a result, giving monsters one-star ratings still carries very little penalty, and the game is ultimately broken because of that.

Theme Attachment – I had my idea for Star Power waiting in my head the day before LD32 started. It was kind of half-baked and not fully realized, but it was there and I was proud of it. When the theme was officially announced, I spent about an hour brainstorming ideas with the folks at TVGS since that was their tradition. I came up with a handful of other ideas which probably would have worked awfully well, but I cast them aside to pursue my clumsy idea of a ratings-fueled combat system simply because it was the first thing that had come to my head.

What Went Right
Simple, Understandable Mechanics – Relative to my last couple of Ludum Dare entries, I think Star Power is particularly accessible, which is crucially important for an audience of judges who are trying to blitz through as many games as possible. I’ve made games based on elaborate memory puzzles and increasingly complex boolean algebra in the past, so it was a nice change of pace to release a game whose primary mechanic revolves around basic arithmetic and number comparison.

Big Picture Concept – The implementation of Star Power is sloppy, to say the least. The graphics are scrappy, the audio and particle effects are barebones, and the mechanics are very easily solved. That said, I still think that there’s a great idea buried beneath the layers of clumsy design, and most judges seem to agree. I don’t often revisit my old games, but I feel like Star Power is something special, and it deserves a second chance to be developed into a genuinely engaging experience.

Shenanigans – After realizing that I had spent the better part of my first 24 hours building a game on fundamentally broken mechanics, I decided to just accept the problem and shift my focus away from game design for a bit. This turned out to be the highlight of my weekend. I jammed out on a plug.dj community for a bit, I ran into a bit of a fan in the chat (Here’s to you, TaoPheonix), I spoke with other developers in the venue and tried their games, and then turned in for dinner. The rest of my time was dedicated to hanging out with a great friend of mine who agreed to host me so I’d have somewhere to sleep while up in Troy, and the weekend was better for it.

Lessons Learned
Know your limits – Ludum Dare is a marvelous time to be innovative and push boundaries, but only to a certain extent. If you end up biting off more than you can chew, you’re going to have a very rough time swallowing it in a mere 48-72 hours. Don’t be a afraid to be creative, but you should also be willing to dial back your ideas a little bit if you notice them growing a bit too lofty.

Don’t plan ahead – Like many participants, I try to start making plans for different themes before a single official theme is announced. Historically speaking, I’ve had much more success with games I thought up on the spot after theme announcement, and not before. I think it’s really easy to get married to the idea of a game which you didn’t put a ton of thought into, which keeps you from sitting down and forming an idea which gets all the thought and care that it deserves.

Have fun with it – Yes, I know this sounds cliche, but it’s true. Ludum Dare is not about winning – it’s about having fun. We participate in this event because we want to enjoy ourselves, not because we’re trying to prove something or win some tremendous prize. What’s more, the LD community is overwhelmingly upbeat and positive when they provide feedback. It’s tempting to dedicate all of your time to creating the most polished game you can, but that isn’t necessary. The LD will still love you and your work even if it turns out a little poorly, so don’t be afraid to just have some fun! If fun ends up being shooting rubber bands across a CVS pharmacy counter with a friend instead of programming, then do that instead! That’s what I chose to do, and I’m glad as hell I did.