Wow, this site’s been up for nearly a month now, and what’s more, SpaceCat Studios, as an entity has been extant for more than a year! That’s right, it was fourteen months ago when I, clueless as I was, popped up on the Kongregate forums with a few perfunctory art samples and an idea for the most awesome flash game ever. That was when Chris and I teamed up for the first time and began work on our first game: an abomination of code and art, one which would claim the lives of both my parents, Chris’ entire extended family, and several adorable and helplessly endearing puppies…
…Nah, just kidding, Our first game was Master of Fortresses, and none of that bad stuff happened.
We did, however, learn quite a few lessons, and as Master of Fortresses inches ever closer to the long-awaited and long hoped-for millionth play, I’ve decided to do a post-mortem, to show to you, the literate public, exactly what went right, what went horribly horribly wrong and what lessons we’re taking from this, first and imperfect effort (on my part anyhow) and how we’re applying those lessons to our next great undertaking.
…but only if you swear not to lynch us.
What Went Right:
-Compatible Skills: You’d think this one would be obvious, but it isn’t. All too often, you see a not-quite professional team with ludicrously massive redundancies like six writers or nine million “artists” with crappy pencil sketches on DeviantArt. (and exaggeration, but sadly not a massive one.) This leads to problems, especially when you have a team with no clear leadership (like ours), since that means you have half a dozen artists and writers who funnel their questionable product to the one poor harried soul who bothered to learn C++/VB/AS3/Whatever language you kids are using these days. The artists fight with each other, all trying to submit the “final” version while the programmer is overwhelmed. Everything spins out of control and flies apart. That tends to be rather unconducive to a productive work environment. Meanwhile, on our side, Chris is the genius with AS whose art skills leave much to be desired, while I’m the artist, writer and actor who doesn’t know a lick of code. No redundancies, no overbalanced team structure and surprisingly few gaps in competence (music comes to mind). This was what kept us working, this was what kept us united as a team at ALL.
-Modest Goals: Chris will disagree with me on this one, but I honestly went forward on Master of Fortresses with what I believed to be pretty achievable goals. Granted, much of what we did was cut, and still much more was added, but in the end, we did the most important thing; we shipped a game that wasn’t half bad. My game ideas tend to be grandiose things, flights of fancy more suited for a big budget movie or AAA project than the humble resources at my disposal. In my opinion, the two things that allowed us to get stuck into this project at all were Chris’ candour in telling me when something was just too hard or impossible to implement, and my ability to just swallow my pride and concede that the fellow mucking with the code probably had a better grasp on the reality of the project than I did.
-Passion: The main reason why I’ve never been able to finish a project before this one is pretty simple: I gave up. When I worked alone, I would eventually lose interest and the half-done project file would moulder amongst the grandiose ruins of other attempts on my hard drive. When I worked with a team, it tended to fall apart, whether through power struggles, lack of commitment or the simple realization that our task was impossible. What was different with this project was that both of us constantly wanted to make this happen. Granted, there were times when I would slack off or lose focus, and then Chris would show me the new build and I would be galvanised into making more art. I’m sure it worked the other way around. It was almost like the secular version of a religious experience: we were tested, we were tried, and sometimes, we despaired, but in the end, we kept the faith, and we made it HAPPEN.
What Went Wrong:
-No Unified Design: This was the big fuck-up. This was why Master of Fortresses took a year to complete instead of six months or even four, as we had originally planned. Believe it or not, we didn’t even start with a design document, just a disparate collection of unit stats, concept art and publicity blurbs. Hell, we didn’t even have a real NAME (Master of Fortresses was formerly known as Vauban, a perfectly good reference for a military history buff like me, but not for people with actual lives.) How we were able to make it into a game at all was a wonder in itself. In addition, we had no actual art design, and frankly, playing through the game now, it SHOWS. We changed GUI designs and Map art styles halfway through development, we changed a few basic assumptions about the art, and even now the whole thing looks a little bit too disjointed.
-Form Before Function: This was somewhat less important, but only by a little. The truth was, we only solidified basic gameplay in February. Most of the art was already done, the GUI was finished, and once we realized that what the GUI needed to do, it needed to be redone, twice. Tooltips and other bits were added to make the whole thing more intuitive than a pit of giant scorpions (but as you’d probably know, only by a little bit). This took us three months. By the end of it, we were so anxious to get the damn thing finished that we quite brusquely booted it out the door, leading to the problem which you lot probably recognized the most;
-Too Little Beta Testing: When we shipped Master of Fortresses, it was full of bugs, and this was after we spent two months fixing bugs during bidding, and a two months before that in beta testing. The fact is, we were so bloody preoccupied with getting this damn thing RELEASED that we didn’t spend enough effort debugging it, or balancing the factions, or even testing out all the gameplay elements throughly. Granted, when it was only a handful of sponsors and friends playing, it wasn’t so bad, but when we released to the general public, it kicked us in our faces in the worst possible way. I feel the lack of polish and insufficient bugtesting was what made Master of Fortresses merely a good game, rather than a great one.
-Prototype First: Most of the problems with Master of Fortresses stemmed from the fact that we didn’t have a master plan. We didn’t know what the gameplay was going to be like, there was no unified art style, and we weren’t working on a timetable, just a vague checklist. That hurt us in the end. This time, we’re prototyping gameplay, developing an art style and THEN we’re doing fluff like menus.
-Fear the Feature Creep: The problem with having no master plans means that we allow new ideas to creep in, While they might be good on their own, they detract from the general plan and slow down development on important things as we indulge our own private whims. This time, I’m thinking of coming up with a final design, and then locking down the idea factory, we won’t add new stuff unless it becomes absolutely necessary.
-POLISH: I’m going to say this now, so you can hold me up to it later; we are going to polish the SHIT out of this game. Everything’s going to be tested until it breaks, repaired, and tested again. We’ll use a wider beta tester pool (Master of Fortresses had something like a total of four beta testers) and we’ll give them more time. We’ll also start them earlier so they can catch stuff before it gets buried under mountains of other code and art and therefore, too difficult to change. SpaceCat Studios’ next game won’t be merely good; it will be GREAT.
So, now you know the dark secrets which have rested in the recesses of our minds. (well, not really) We’ve both taken away a lot from this project and hope that the lessons we’ve learned will make us a better team and more skilled individually. I’d personally like to thank the fans and friends who brought the flaws in our freshman effort to our attention. You guys know who you are, by doing what you have, you’ve hopefully made us better developers from the lessons we’ve learned.