Slashdot Mirror
An Inside Look At Tabula Rasa's Failure
Massively notes a couple of posts from people who worked at NCSoft while Tabula Rasa was in development. Adam Martin says the lengthy, wandering development cycle led management to push it through before it was ready. "Very late, they eventually hit upon a good formula, a good core game," but, "Before they could actually make that game, a difficult decision was taken to push the team to the wall and force an early beta test." Scott Jennings suggests that early warning signs, like the tepid reaction to the beta, were largely ignored. "One of the mantras that went around production discussions after Auto Assault's launch square into the pavement was that if you can't get people to play the beta for free, you have serious, serious issues. Tabula Rasa had those issues. Not as bad as Auto Assault — there were people doggedly playing every night and presumably enjoying themselves, and metrics were duly assembled to measure every movement those testers took. But it was pretty clear, at least from my completely disassociated and busy with my own thing viewpoint, that there wasn't a lot of excitement."
Only so long you can take. While it is nice to say "We'll just keep working on it till it's done," that isn't realistic. There's two major problems that can result:
1) It costs too much. Publishers have to make money on the games they publish. You can't ask them to just toss money down a black hole. Well if a development team spends a decade working on something with nothing to show for it, all the while collecting their salaries, needing equipment and so on, this puts the publisher in a bad position. They are so far in the hole that even if the game is successful, they may not make back all they spent. So while I can certainly understand why publishers are going to push a release at some point.
2) Duke Nukem Forever syndrome. Here's an example where they are self financing so they CAN do development for many years. The problem is it isn't working. At a certain point, you can't improve you game. Part of the reason is just that with really long cycles focus can get lost and such. However the bigger reason is technical. You are developing for 1995 hardware. It is now 2000, your game is outdated. So you have to redo your art assets, you have to rewrite or buy a new engine and so on. You get stuck on a treadmill of doing the same shit over and over. DNF has gone from Quake to Unreal Engine 1 to Unreal Engine 2 and will have to buy Unreal Engine 3 if they want to release a game today. That's a lot of respent effort (and money).
So I certainly don't condone publishers hurrying releases out the door, but I can understand how after a time they'll say "Enough is enough, we move forward with the launch." You can't ask them to wait forever. Often the end prodcut won't be good, or even get done (see DNF) and even if it is good, it may still not make enough money dur to the costs.