我此前已经得出过一个结论，每个独立游戏开发商此生都会制作出至少一款平台游戏。Twisted Pixel有《Splosion Man》，Jonathan Blow有《Braid》，Odd Gentelmen制作过《The Misadventures of PB Winterbottom》。还有无数的创意性flash游戏属于这种题材。去年，我第9次阅读HG Wells的《First Men in the Moon》时，发现了《Limbo》这款游戏。于是，我决定也要制作出一款属于自己的平台游戏。《Cavorite》就这样诞生了。
当然，游戏与我最初的设想并不相同。《Cavorite》不像《Limbo》那般阴暗恐怖。坦白说，Cascadia Games并不具有制作艺术性游戏的能力。因而在游戏原型制作完成后，整个游戏搁置了8个月的时间。我不知道如何利用现有的人员和时间来把游戏制作出来。幸运的是，在如此长期的等待之后，我又有了灵感。瑞典独立开发商AcnePlay发布了《Pizza Boy》，这款16位平台游戏似乎把我带回了1992年。看到这款游戏，我忽然明白了要如何将我的想法变成现实。
《Cavorite》是款更侧重益智元素的平台游戏。我通常把它当成是《超级马里奥2》和《Professor Fizzwizzle》的交集。这也是款16位游戏，最初的灵感来自于HG Wells。
游戏开发始于11月，到3月份时就可以对《Cavorite》进行测试了。我让朋友、家人和其他开发者同伴帮我测试游戏。需要特别指出的是，我得到了Appy Entertainment的Paul O’Connor和开发者同伴Matt Mitman和Ronny Bangsund的帮助。他们都很能清楚地看到游戏的长处和不足之处，他们的反馈能够极大地改善游戏。
Postmortem – Cascadia Games’ Cavorite
I’ve come to the conclusion that every independent game developer must create at least one puzzle platformer in their lifetime. Twisted Pixel has Splosion Man. Jonathan Blow has Braid. The Odd Gentelmen has The Misadventures of PB Winterbottom. And there are countless creative flash games in the sub-genre. Last year, I discovered Limbo while re-reading HG Wells’ First Men in the Moon for the Nth time. It was then that I decided I had to make my own platformer game. Cavorite was born.
Of course, nothing even close to my original vision happened. Cavorite is not grisly and dark like Limbo. To be frank, that type of artsy game is not within Cascadia Games’ abilities. So after an initial prototype, the game sat idle for 8 months. I simply couldn’t see how we could get the game done with the talent and time onboard. Fortunately, after the long wait, I was inspired again. Swedish indie developer AcnePlay had released Pizza Boy, a 16-bit platformer that felt like it had leapt straight from 1992 onto my iPhone. Suddenly, I knew how to bring my vision to life.
Cavorite is a puzzle platformer hybrid with extra emphasis on the puzzle. I often describe it as a cross between Super Mario 2 and Professor Fizzwizzle. It takes place in a 16-bit, steampunk universe inspired primarily by HG Wells.
What Went Right
Outsourcing the Art
‘Retro’ art is in the midst of a big comeback. Early thirty-somethings like me love the style and are willing to spend for the nostalgia. Because of this, there’s an excellent (dare I say, thriving) market for pixel artists. Finding a capable and affordable artist proved surprisingly easy. In fact, I almost don’t want to reveal who I hired, because I don’t want to share him. Tom Filhol was that fantastic to work with.
Even though I outsourced the pixel art, I still had the notion that I should handle the UI myself – as I had done for all our previous games. It’s a good thing that I felt dissatisfied enough with my efforts this time around because fellow indie Matt Mitman was available. He handled the game’s logo, UI, and story comics. And he really nailed it. I don’t think the game would have received as much high praise without his excellent contribution.
Finally, even though I consider it more in-house than outsourcing, frequent Cascadia musician Erhan Ergenekan once again took the visuals plus virtually zero guidance from me and made an awesome soundtrack as well as the majority of the sound FX. Erhan consistently blows me away. The music is beautiful and really sets the tone of the game.
My goal is always the same: design something that can be finished. I knew that designing the 60+ puzzle levels would be a major challenge and decided early on that the best simplification would be to cap them all at single-screen size. I was initially concerned that this would hamper our ability to create a distinct experience per level. As it turned out, it was more liberating in how it forced creativity.
One of the best habits we formed was to design each level with pencil and graph paper. That allowed the designers (myself and my wife, Katie) to quickly try ideas, discuss them, and tweak them without involving the computer. Each evening I would have a stack of finished levels to implement. The only time we struggled with level design was when we tried to circumvent this path.
Sketching out our levels independently also proved a great way to let knew ideas organically sneak into the design. Typically, I would introduce a new element and Katie would come up with about 5 uses that I’d never envisioned. Or she would introduce an idea and I’d grumble about how hard it was to implement, then add it and be amazed at how much it would open up the game. I’ll just note that Katie is directly responsible for the design of about 1/3 of the levels (typically the more brutal ones) and pioneered many of the fundamental game mechanics, like using platforms as makeshift timers. We had a lot of fun designing boards and sprinkling in twists to keep the ideas fresh.
Heavy development started in November and by March I had declared Cavorite in ‘beta’. From there I showed it to the usual suspects – friends, family, and fellow developers. In particular, I want to call out the help of Paul O’Connor from Appy Entertainment and my fellow Torque developers Matt Mitman and Ronny Bangsund. They all had great insights into the game’s strengths and weaknesses. I was able to take their feedback and improve the game greatly.
After this initial round of testing, I announced the game around May in a few key places, including the TouchArcade forums. I was taken aback by the number of volunteers for beta testing. I hadn’t planned on a big public beta test but decided on a whim to go for it. Originally, I assumed the testing would take about a week. It ended up taking over a month with easily over 100 bugs / issues reported. Especially helpful were ‘Red1’ and ‘Doomfan’ with their very thorough feedback. Small issues that I had forgotten or ignored were brought to the forefront. New ideas (such as collecting stars for level times) were offered up and worked well once added. It was a fantastic experience that made the game significantly better and more broadly appealing as a result.
What Went Wrong
I admit it. I’m a terrible marketer. We’ve also never had a “big” game before. Prior to Cavorite, our previous apps have primarily been proof of concepts or experiments. My expectations were always low. With Cavorite, however, I knew we had our first fun, polished, and unique game. Still, I had no clue how to market it.
The day after launch, Apple prominently displayed the game’s logo just beneath the iPhone Game of the Week promo slot in the iPhone App Store’s Games section. We also got some nice promotion from our pals at Appy and mentions from the beta testers, early adopters, and other indies. Several game news sites picked up on the game and offered praise. I then started contacting reviewers.
Of course, this is the classic mistake with the App Store. Visibility is everything. And rank equals visibility. While we continue to get high scores from prominent review sites such as AppAdvice, Appolicious, and 148apps, they have all appeared at different times. We never got the additive push of all the reviews together. Cavorite hovered in the high 40s to low 80s for awhile before starting to slip. Our All Games rank peaked at 294. I wish I could have seen all of the review rank bumps added into one.
Finally, the game’s social aspects are far too weak. Yes, you can tweet your level times in-game. But that’s basically it. The ability to TwitPic was cut, as was the idea for solutions sharing. This combined with sporadic review visibility has left Cavorite largely unfound. More than one happy buyer has complained about how hard it was to find the game.
Cavorite took approximately 8 months to develop. About 1 month could have been trimmed had I not insisted on writing the core game logic in TorqueScript initially. TorqueScript is a handy, flexible language. But it wasn’t designed to do an app’s heavy lifting. As a result, I had to spend weeks porting functionality to C++, which reintroduced bugs, timing issues, and a variety of other challenges that could have been avoided had I planned things better.
The time crunch meant certain features were dropped: social sharing, retina support, an iPad UI option. I’m positive these features were desired by players at launch. But toward the end I started to run out of steam and didn’t have the energy to tack them on. When you write apps full time, then come home and write apps at a part-time pace for eight months, you expend a lot of energy. Better planning could have kept me going longer and made the development more efficient.
I was warned that the winds of the market were blowing toward freemium. And I didn’t listen. Cavorite simply wasn’t designed to be a free game with in-app purchases. Reviewers and buyers alike complained that it cost $1.99. But I felt it was justified by the quality and quantity of content.
In retrospect, Cascadia Games isn’t a known brand that consumers are willing to risk their highly sought dollars on. I should have looked at free-to-play more seriously. Right now, the game is on track to profit, but it’s not what we had hoped for.
We launched on June 29th – exactly one day before every major iOS publisher, from Gameloft to EA, started putting their apps on 4th of July Summer sales. Our game was buried by these $0.99 and free app deals. If I could launch the game again, I would have executed a more organized launch in July after the sales wave died down.
Everyone involved with Cavorite is proud of what we accomplished. Cascadia Games went from just another app-churner to the creators of something special. While we didn’t strike it big with this app, we now have the clout and experience to do things better and get wider exposure in the future. Whether that’s additions to the current game or a new project entirely, we haven’t explored yet. But we’re higher than we’ve ever been and are happy to be here. (Source: Gamasutra)