The SimCity Debacle: A Case for EA

I was one of the nerd horde desperately following news of the new SimCity’s release. Eating up every glassbox demo video on youtube. Visiting the website daily to check the release date. When March 5th rolled around, I convinced my wife to let me install Windows on her MacBook Pro and sat there eagerly awaiting my download.

That was 10 days ago.

Today, to put it lightly, the game is not living up to expectations.

A short list of the reported issues with the game:

  • Server capacity issues leaving game unplayable.
  • Always-On “DRM” annoying gamers.
  • Broken AI, causing serious gameplay issues.
  • Broken simulation, in the form of incorrect population reports.
  • Map size too small.

And on and on and on. Take a look at the unofficial subreddit for more details if you’re so inclined.

Many redditors and gamers in general have called foul, accusing EA and Maxis of lying and deceiving customers in an attempt to boast sales. Frankly those claims are not unfounded. EA has a history of poor customer service, etc.

But what if EA simply under estimated the number of people interested in the game. What if EA has no clue who buys their games? This would explain the utter brokeness of the servers at launch.

It might explain why they decided to release such a buggy game. If they completely failed to understand how many people would buy the game, they may have thought it would take much longer for the bugs to surface. They might have assumed they had more time to address the bugs.

The always-online gameplay model allows EA to iterate. If AI is broken, they can fix it, release a patch and instantaneously fix everyone’s experience. Iteration is good.


Leave a Reply

Only people in my network can comment.