Gamasutra: The Art & Business of Making Gamesspacer
Postmortem: Pangalore's Knightly Adventure
View All     RSS
December 13, 2018
arrowPress Releases
December 13, 2018
Games Press
View All     RSS






If you enjoy reading this site, you might also want to check out these UBM Tech sites:


 

Postmortem: Pangalore's Knightly Adventure


June 21, 2013 Article Start Previous Page 4 of 6 Next
 

What Went Wrong

Okay, we've talked about the good stuff so far. Now let's delve into the other side of spectrum. We've identified our five biggest missteps in the following sections.

1. Multiplatform Launch Out of Sync

We discussed the success we had with our synced multiplatform gameplay above. But one area we stumbled on was on the launch timing between different devices. This occurred because we didn't fully understand the differences between the App Store and Google's Play store and their associated hardware.

The App Store has a longer approval cycle, leading to a longer cycle not only in development but for updates. With the ability to issue quick fixes, the Play store is more suited to multiple iterations. Ideally, we might have launched first on Android where we could release hot fixes to resolve issues before launching on the larger iOS market. Because we worked hardest on the biggest platform launch (iOS), our Android version was not ready at launch and when launched, did not have full device coverage. This meant the marketing dollars we spent at launch were leveraged less effectively since we couldn't send potential players to an Android download. It did leave us with a stronger story for iOS players but we lost our multiplatform message in the process.

To compound matters, we underestimated the time required to tune the game for each supported Android device, further delaying our Android launch. And when we did launch, it was only for a few devices, which was not pleasing (understandably so) to Android players. We're still slowly filling out our roster of Android devices.

On the positive side, we did get the game out for the holidays (we launched in November 2012) in time for a positive reception before the market was inundated with holiday shopping and we had our greatest success on the larger/more unified iOS platform. Going forward, we have realigned our expectations for the two stores and have a plan for utilizing the strengths of both. In some cases, quick is good and in others, slow and careful is the right play.

2. Testing Was Too Localized

Unfortunately, our testing did not reveal some key issues in our largest market -- North America. By conducting QA where our devs were (Korea) rather than where most customers were (U.S.), we failed to spot some tech issues that were an issue at launch. That was a big oversight, and it bit us in the end like a dog going after a mailman!

South Korea has much faster and better-integrated internet infrastructure than does the US, and this made us underestimate some issues we would have in the US with downloading and syncing performance at launch. Downloads that seem fine at 20 Mbits per second are much more vexing at 1.5 Mbits/second. During spot-testing in the U.S., we thought the issue was that the development servers we were playing from were based in Korea, but the issue turned out to run deeper than that.

Going forward, we are doing QA in North America using a service company. This gives us another channel for external feedback while also insuring that testing is closer to the reality experienced by our players. Lesson learned! 


Article Start Previous Page 4 of 6 Next

Related Jobs

Wombat Studio
Wombat Studio — SANTA CLARA, California, United States
[12.13.18]

Graphics Engineer (Tech Art focused)
Cold Iron Studios
Cold Iron Studios — San Jose, California, United States
[12.13.18]

Console Gameplay Engineer
Cold Iron Studios
Cold Iron Studios — San Jose, California, United States
[12.13.18]

Infrastructure Engineer
Cold Iron Studios
Cold Iron Studios — San Jose, California, United States
[12.13.18]

Site Reliability Engineer





Loading Comments

loader image