Unga Bunga Postmortem
From GDWiki
By Jordan Magnuson, November 22, 2004
[edit] Introduction
I've always wanted to make games. When I was a child I used to think up game ideas, and then scratch out notes and drawings in pads of paper I kept under my bed. I even drew out some logo ideas for my dream game company... I think it was called "Topnotch Software." To cut a long story short, when I was fourteen I purchased a copy of Clickteam's Multimedia Fusion and started coding away. I had big dreams for big games, and of course I tried to realize those dreams right away: fortunately it didn't take me long to figure out that I had better start someplace humble if I ever wanted to see anything completed. Unsurprisingly there were a few "humble" false starts, but one day towards the end of my freshman year of high school I sat down and started coding a game that was destined to be finished, albeit over two years later. The game was Unga Bunga. Here is my take on that project four-and-a-half years after it was started.
[edit] Back Story
Unga Bunga is a multiplayer (same-keyboard only) platform deathmatch game in which players run around as club wielding cavemen and bonk each other to death. The game was inspired by 2D deathmatch classics like Tremor and Liero, and features most of the offerings you'd expect from a deathmatch game on this scale: multiple levels, 2 or 3 player support, and a variety of game types (3, to be exact).
On the innovation side, Unga Bunga was built with the idea of extreme customization in mind, which means users get a lot of control over game options, and can, in theory, have a different experience every time they play. Also, unlike other deathmatch games I am aware of, Unga Bunga features a technology tree, which allows players to gain access to better weapons by killing computer controlled bears and "meese."
[edit] Game Data
Release date: October 1, 2002
Developer: Fusion Apple Entertainment
Distributors: Xing Interactive, Magnussoft
Genre: 2D platform deathmatch
Demo Download Size: 9.8 MB
Platforms: Windows 98/2000/ME/XP
Number of developers: 1
Budget: $0
Length of development: Approx. 2 years
Software used: Multimedia Fusion
Notable technologies: None
Project size: Core code is approx. a thousand lines of MMF script.
[edit] What Went Right
[edit] 1. It got finished
Peter Molyneux begins his post mortem of Black & White with this point, saying, "This sounds stupid, but we encountered some big problems, and there were times when we doubted that the game would get released." I think every game post mortem should begin the same way, especially indie game post mortems (unless, of course, the game really didn't get finished). With the number of games that get started every day, and the number of games that burn every day, each game completed is a miracle, as far as I'm concerned. I started Unga Bunga thinking that here was a small, simple project that I could kick off nicely in a few months or so... Haha: when I finished the game over two years later I rejoiced. Because in between starting and finishing, that dark world of bugs and not-enough-time and feature-creep and "oops-I-underestimatedagain" struck hard.
Sometimes I would go weeks without working on the game at all, talking about getting back to it, but wondering, in my heart. Other times I would work away, but still be uncertain of where the finish line was. Sometimes I'd get stumped on a bug that would eat up days and days of valuable time, and I'd wonder how many of these kinds of bugs were yet to come. But in the end it did get finished.
[edit] 2. Using Multimedia Fusion
Multimedia Fusion was both a blessing and a curse, as you will see in the "What went wrong" section below, but here I will focus on the positive. Prototyping in MMF is fast. Period. It is also extremely easy. These are two very good things: point A allows you to jump right into implementing design without worrying about details, and if not for point B, Unga Bunga would simply not exist today.
I had been programming for a couple of years when I started the project, but in languages like Pascal and Visual Basic: I was not ready to jump into programming a major game project. Enter MMF, and all is taken care of: don't know how to program with graphics? No problem, just point and click. Want an exe? No problem, just point and click. In fact, that phrase basically sums up the wonder of MMF: point and click. It works. It works so well, in fact, that I will probably be using it for prototyping game ideas well into the future. The problems start to arise when you want to do more than prototype: i.e. build a solid, fast, professional-quality game.
[edit] 3. Gameplay
Unga Bunga is a fun game. I know it's a fun game because I'm still playing it-I kid you not-four years down the road... after playing it about ten thousand times. I also know it's fun because I've watched other people play it, and it shows. What's more, the game continues to get a conversion rate that astounds me, considering all of its flaws (2-3%); I attribute these figures to the simple fact that the game is fun.
This is huge. I think that the elusive "fun" factor is the most critical element to any game's long-term success, and I think it's the element most often lacking. How did Unga Bunga end up being so fun? I think it was pure luck, mostly: I plugged away at an idea, and it turned out to be a good idea. I think partly, though, my success in this area was due to efforts to incorporate mechanics that I knew were fun from other games. Also, my basic style of development was "play and tweak," which I think is a good idea, generally. Design documents have their uses, but setting everything in stone too early can be disastrous, in my opinion. Unga Bunga was a constantly morphing game: it morphed to the fun.
[edit] 4. Premise
Apparently people really dig the idea of cavemen running around bashing each other. Apparently people also dig the name "Unga Bunga." What I'm trying to say is that the basic premise of the game has turned out to be a really, really good marketing decision (albeit my only good marketing decision) -- and I wasn't thinking along those lines at all when I started the project.
I think coming up with an original idea that has mass appeal is a very smart move when designing a game. This has, in a way, very little to do with actual gameplay, but it has everything to do with how people will perceive the game, and whether they will want to download it or not. The moral of the story, as far as I'm concerned is don't make a game about spaceships and call it "Space Shooter." The gameplay may be awesome, but people have seen too many spaceships and too many games with names like "Space Shooter." Add an original twist to the exterior of the game, and give it a "cool" name that will make people interested, and voila: same game with different clothes, and a lot more successful. That's my theory anyway ;). Unga Bunga gets a lot of looks -- a lot more looks than it would get, I think, if it featured robots with guns and was called "Deathmatch Action."
[edit] 5. Beta testing
I got the game out to some forums relatively early on, and received a lot of quality feedback. Consequently the game changed quite a bit, and for the better. Everyone probably realizes this, but can it be stressed too much? Working in a vacuum is bad. Getting fresh input from people who have never seen the game before is good. Enough said?
[edit] 6. Tweaking the Demo
Shortly after I released the first public version of the game I discovered the indie developer mountain of gold known as dexterity.com. The indie developer forums, along with Steve's articles, proved (and have continued to prove) invaluable to me. The particular article I want to mention here is "How to Permanently Increase Your Sales by 50% or More in Only One Day." I followed this article's advice and Unga Bunga's conversion rate went up not by a factor of 1.5, but by a factor of more than 10! Just by tweaking the demo and the purchasing process; I didn't change the actual game at all! I think you can draw your own conclusions from this one.
[edit] What Went Wrong
[edit] 1. Unpolished graphics and sound
I pixeled all the graphics for Unga Bunga by hand, by myself (well, a friend helped a bit). Big, big, big mistake. I think the biggest hindrance to the success of Unga Bunga today is the unpolished, unprofessional look of the game. One look at a screenshot and distributors say "no way." They don't care if the game has a 3% conversion rate. A lot of potential players also look at the screenshots and say no way, and I've talked to people who can't focus on the game even after they start playing because they're turned off by the graphics. The same points apply to the game's sound, though the graphics are a bigger deal.
[edit] 2. Multiplayer same-machine gameplay only
I love multiplayer-on-one-keyboard games. Always have, always will. But how many multiplayer same-machine only games do you see out there? Not too many, and it turns out this is for good reason. Unga Bunga needs single player gameplay to draw casual gamers in. It needs online and LAN capabilities to get people hooked. Multiplayer same-machine will still do it for some people, as the game's conversion rate attests, but how much more people could I be reaching?
[edit] 3. Not easy to get into the game
This one has been difficult to judge, but I think Unga Bunga can be hard for players to "get into." This kind of builds off of the last point, as many players download the game and start it up right away, only to find that it doesn't support single player gameplay, and any attempts to "mess around with the game anyway" prove predictably futile, as the game is simply not fun (not playable) with only one player. Even if they do get further, and try playing the game with a friend, the lack of any real in-game help can make the function of the meese and bears, and the role of technology, difficult to understand. As I said, I'm really not sure how big of a problem this whole issue is, but I sense that it's a problem.
[edit] 4. Lack of a strict (read any) development schedule
Working on a game part-time as you feel like it is not a good idea if you want to finish a small game in less than two years. What more is there to say? There needs to be a schedule, and some way of sticking to it. Even if the schedule isn't that strict, there needs to be continual work being done: going two or three months without touching the game is simply unacceptable. Unga Bunga almost stalled out, and I don't want that to be an issue with any of my future projects.
[edit] 5. Using Multimedia Fusion
Multimedia Fusion is great for prototyping, as I've already said. When it comes to making "full-blown" games, however, there are some major issues that arise. A prime example is the presence of bugs within MMF. Finding and squashing one's own bugs is, I think, enough to be asked of any programmer, but there were numerous times when coding Unga Bunga that I had to figure out, first, if a bug was in my code or in MMF itself, and then, if it was in MMF, figure out how to get around it. I always found a way, in the end, but sometimes it took me weeks. Usually I could still keep working on other parts of the game in the meantime, but sometimes I couldn't.
Another problem with using MMF is that, once a certain level of complexity is reached, it becomes less efficient to code the project in MMF than it would be to program it using a standard language, thus defeating one of the main reasons for using MMF in the first place. Unga Bunga didn't quite reach this level, but it's something to keep in mind.
Another issue: code written in MMF is very slow compared to code written in standard languages like C++ or Java (forget the speed differences between C++ and Java themselves!). Optimizing can make a big difference (something I only partially succeeded at with Unga Bunga), but this is still a major issue for all but the lowest profile games. Unga Bunga requires a 600Mhz processor to run smoothly, and some of the levels don't run well on anything less than 1Ghz! Fortunately for me these specs were fairly standard by the time the game came out, but you can see that the requirements are ridiculous for a 2D platform game! (Optimizing the code more thoroughly would make a noticeable difference though, I suspect.)
Finally, coding in MMF squashes from the start any dreams you may have of porting your game out of Windows. No big deal, I thought-until I found out that the Mac shareware market is around 20% larger than the PC one. Oops.
[edit] Conclusion
I have to consider Unga Bunga a success. Heck, from the first point of "What Went Right" alone I'd have to consider it a success. The game got finished, I honed my logical thinking skills, I developed some serious MMF know-how, and I learned a lot about what to do better next time. Of course, it would have been nice if the game was finished in three months, and if it were bringing in more money, but how much can you ask for? ;) The fact that the game has sold any copies at all makes me happy, and the fact that it gets a 3% conversion rate just blows me away. But the best part, in my opinion, is being able to play the game today with my college roommates-four-and-ahalf years after I started it-and have a blast. Frankly, I can't wait to start working on Unga Bunga 2.
By Jordan Magnuson, November 22, 2004
