Postmortem: Capybara Games' Pillowfight

Some of the most intense and stunning battles are fought not with guns, swords or fists, but with pillows!

A unique casual fighting game, Pillowfight lets players serve up powerful pillow beatings with simple controls and fast paced gameplay. With sexy characters, intense special attacks and lots of player customization, Pillowfight is quite unlike anything on the market.

Pillowfight's story began when some of Capybara's founders were taking a break from GDC 2006 preparations. We had wanted to bring a couple small concepts to GDC 2006 to flash in front of our publishing partners, but we hadn't really had time to formulate anything. During that break, creative director Kris Piotrowski sarcastically said "Here's an amazing idea: Let's make a pillow fighting game starring total babes!" We all had a good laugh, but the laughter quickly died off as we realized: "No seriously, let's make a pillow fighting game starring total babes." We created a one-page pitch for a casual-minded fighting game (starring total babes), slapped on a couple gorgeous mock-screenshots and headed off to GDC not really expecting much to come out of it...

Pillowfight

Fast forward a year, and Pillowfight has become one of Capybara's highlight projects. Our publisher I-play proved to be a perfect fit for the game, and I-play producer Chris Johnson helped us shape a simple yet engrossing one-thumb fighter that's more than just the sum of its.. um... parts.

Pillowfight was recently reviewed on both IGN and Mobilegamefaqs, receiving the IGN Editors Choice Award and the Mobilegamefaqs Gold Award respectively, proving that titles with sexy subject matter don't have to be devoid of great gaming experiences.

What went Right

1) Simple tools build better games
We made heavy use of tools during development, many of which were new, untested or developed specifically for this game. One tool we found to be a particular success was Capybara's custom "Fighter-Maker". Initially we began with an overcomplicated and un-flexible GUI, but when it became clear that the tool wasn't working, lead programmer Adam Rivard stepped in and quickly developed a simple yet robust spreadsheet-based tool that provided designers a greater level of control while taking very little time to learn. This let us easily test new fighter patterns, tweak timing and continuously refine fighter difficulty resulting in the fun and challenging opponents you see in the game today. Determining where to put Stef Summers' panty-flashing split attack took a lot of fine tuning... but it was worth it. We proved that effective and robust tools can be developed easily and inexpensively via tailoring commercial or freeware applications to our needs. It isn't always necessary to reinvent the wheel when it comes to tool creation.

2) Seeing eye-to-eye with I-play
From the get-go, Capybara and I-play were on the same page. We understood that, due to our subject matter (total babes with pillows) we had to make a great game to back it up, or the game could be cast aside as drivel like so many other sexy-themed games. I-play provided focus testing information early and often, which helped us figure out where we were succeeding, and where we were failing. Constant constructive feedback flowing from I-play producer Chris Johnson to Capybara producer Tim Maly meant we could always clear up issues as they arose, keeping us from wasting precious development time. From the first menus, to the character customization, to slamming your gorgeous opponent in the face with goose-down, Pillowfight development played out as a team effort between developer and publisher, with the end result being a game we are both proud of (and can actually show our mothers...maybe).

3) Hey Pretty Ladies (wink)
 We didn't kid ourselves about Pillowfight - we knew from the first conversation this game would live or die on the quality of visuals, and in the end the art in Pillowfight definitely went right....sexily right. We knew we wanted larger characters in order to give each fighter some personality, but there were huge issues with both heap and JAR limitations. Tasked with creating these beautiful pillow assassins, Lead Artists (and twins) Mike and Vic Nguyen focused immediately on making each character special. Everything from idle animations to special attacks were drawn, animated, broken into body-parts so that they could be compressed on the heap and jar, and then put back together through our code/art tools combination. Supporting the Nguyen's amazing sprite art was programmer Kenneth Yeung's amazing procedural effect touches on everything from the menus to spewing feathers on impact. The end result was immensely satisfying visuals that we're all very proud of...and often stare at...

4) One-Thumb Fighter
We researched as many mobile fighting games as we could get our hands on and we rarely enjoyed what we played. Because of the extreme control limitations every mobile game faces, it was clear to us that trying to emulate arcade-style fighting would just lead us to an over-bloated and clunky experience. We really wanted to avoid that, so designer Matt Repetski and the team decided try to create something that was more akin to Mike Tyson's Punch Out than it was to Street Fighter. By focusing on a very simple move set, our game became much more about timing and reading the opponent's attack pattern, instead of cumbersome movement and complex button inputs. Focusing on these key areas helped us deliver a fighter that is appealing to the hardcore crowd, but was easy enough for a casual player to pick up and play. Furthermore, this approach let players spend less time fumbling with buttons and memorizing special moves and more time looking at pretty ladies hitting each other with pillows, which was also one of our objectives. Mission accomplished!

What Went Wrong

1) Starting off on the wrong foot (foot = device)
Pillowfight began development on a mid-end device, which meant that our first set of prototypes often left programmers Adam Rivard and Ziya Kadioglu concentrating on device limitation-based problems when we should have been focusing all attention on developing the core gameplay. It is extremely tough to manage device limitations and create new gameplay at the same time, especially early in development. Initial prototypes simply did not show our vision of the game - without time to optimize code and resources we could not fit all the animations, characters or speed we desired on a lower-caliber device. In the end, we achieved a better result for the high-end than we likely would have otherwise and had a solid base to bring to lower-end devices. Switching to the high-end target device helped things move much more smoothly for the remainder of schedule, but had we started with the high-end build we could have likely avoided a lot of these problems altogether.

2) A little help, please?
One of our most obvious mistakes was significantly underestimating the art resources required. Initially, we had not scheduled sufficient artist time for the work that would be necessary to achieve the level of quality we aimed for - or for getting the required art completed at all. Artist Mike Nguyen was more than a little overwhelmed, and did not have the time needed to create the sexed-up characters he envisioned. Combining this high demand with a rather complicated character animation system, involving breaking each fighter into 9 different parts, meant we found ourselves consistently coming up short on art for the first set of milestones. Some of the early builds contained very generic looking fighters instead of the sexy and dynamic characters that we all wanted. Fortunately for us, Mike had a twin! Adding Vic Nguyen to the project added the firepower necessary to create a visually impressive fighting game. On future projects we will make sure that we have all the right people in place as soon as they are needed.

3) FOCUS!
We had some extremely lofty/fluffy ideas for Pillowfight - so much so that we overwhelmed ourselves with them. We originally planned a very complex career mode, tons more fighters with a great deal of interchangeability and much more. We pride ourselves in being a creative company, but we took it to the extreme with Pillowfight. Most ideas were scrapped outright, but some lingered through development. We finally focused ourselves on a final product that excels in key areas and has a lot of personality, and the decision to do fewer things better, rather than more things poorly, was the right strategy, one that we should have chosen from day one.

4) Your arm is a bit off
Another aspect of development shortcoming was never completing a tool to allow artist to properly test their work on the fly. Because of the complex nature of Pillowfight art, coordinates and alignment were constant sources of defects. Our total babes would often show up with detached torsos, missing hair, or one arm short. We never spent the time to create a tool or technique for artists to easily diagnose and fix these issues before implementing art in the game, meaning that problems had to be caught later on during testing, and could in some cases be hard to reproduce (you try catching a slightly misaligned arm sprite on frame three of a six frame leaping pillow smash). Considering the complexity of our character animation system, we should have had a tool for letting artists view the cut up and reassembled images frame by frame prior to integrating them into a build.

Conclusion

The end result was a game that both Capybara and I-play are extremely proud of. While development had some minor speed bumps, our combined vision to create the best (and only) pillow fighting game ever has shone through. At the end of the day, we feel we managed to create a great casual fighting game for mobile game lovers everywhere. What started as a crazy idea has become a viable mobile brand, and a hot one at that.

I-play has done a tremendous job marketing Pillowfight, building buzz around this sexy title, and the game has been well received by the press and gaming community. We've got our fingers crossed for Pillow Fight 2: Extra Fluffy Edition.

[Producer Tim Maly and Designer Matt Repetski are founding members of Capybara Games (www.capybaragames.com), a leading developer of mobile games. The two have played key roles in many of Capybara's premier titles, including Warner Bros' Happy Feet, Disney/Pixar's Cars, Super Shove It! and Monkey on Your Back. Of the two, only Repetski is presently single.]

Game Data

Release Date: May 2007
Publisher: I-play
Developer: Capybara Games
Platforms: GSM, J2ME, BREW (US and EU)
Languages: English, French, German, Italian, Spanish
Number of Developers: Producer 1, Designer 1, Programmers 2, Art Lead 2
Development Time: 6 months