Release Day!

Back when I was a teenager SimCity was a wondrous thing.  Our computer was so slow that even on the fastest game setting the game moved at a glacial pace but I spent hours uncountable designing the perfect city.  I sat down with graph paper to draw out the perfect city block that I could just repeat over and over again to make the ideal city, rewriting my masterpiece over and over as I gained mastery over the game.  Recently when I saw a review of the latest SimCity I was was very much tempted to buy it right away and play - it looked so pretty compared to the blocky black and white graphics of the past!

And then I looked at the player reviews and discovered that even if the game is good it doesn't matter because nobody can play it.  It has a mandatory internet connection and right now everything is a total mess because the servers can't handle the load at all.  If you can play, which most people can't, you can't do anything because trying to actually start a new city doesn't work.  Some people can play the tutorial for all the good that does them!

Lots of games have horrible launch day issues and solving the problem is tricky.  You simply can't have a realistic server model where you have 100 servers for launch day and then scale back to 50 servers three weeks later without running your costs into the stratosphere.  The solution isn't exactly hidden:  You just throttle the logins you accept so that everyone who is actually playing has a good time and give everybody else a reasonable estimate of how long it will be until they can get in.

Part of me thinks the reason that this isn't done is incompetence.  Part of me thinks it is optimism on the part of management that isn't shared by the tech team.  (Maybe that is the same thing?)  The cynic in me thinks that it might just be a legal dodge.  If you sell 1M copies and can only have 50K people online at a time you are openly admitting that people that buy the product can't use it.  If you instead sell 1M copies and let everybody on to have a craptastic experience instead you can at least claim that you are *trying* accommodate all the buyers with some plausible deniability.

Honestly guys, we figured the servers could handle it!  Sorry about this mess, but as soon as people stop wanting to use our product it will be all sorted out!  We didn't think so many people would actually want to play the game....?

I think that companies really shoot themselves in the foot when they do this, especially when the game is sold digitally and they could very easily stop selling it.  Wizardry Online did this recently and pulled sales from Steam so they could get their house in order.  Other games need to do this too, but much more aggressively. If you know you can deal with 50k logins then only sell 100k copies initially and let more go out the door based on how the servers are holding up.  It will frustrate some people who want to buy but can't but at least when they do get that first taste of your game it will be a sweet one instead of being full of bitter and hate.

I know that if I see a player review that says "I had to camp the online store to get a copy but once I did it was a blast!" I would be eager to buy and if I saw "I bought the game and camped the login but still can't play after two days." I would ignore the game completely.  I don't think I am alone in this.

Pictures from:  http://www.simcity.com/en_US/media/screenshot/downtown