NetDevil Interview
-
Category: News ArchiveHits: 700
RPS: How did the team feel in the aftermath of Auto Assault? What was learned from that project?
A lot! Honestly it could fill a book! Obviously after the game didn't do as well as expected and was eventually turned off, that's a very difficult thing to get through. As a team you end up spending years of your life working very long hours and when that doesn't pay off the way you expect, then that's certainly a painful thing. We ended up spending a lot of time thinking and evaluating why things didn't go better and learned a lot of lessons from that which we are applying across the board. There's probably a million things that contribute the success of a product but on a high level there's a few big lessons which I think are important for MMO development. The first is to polish early, and not at the end. Numerous high profile game failures result directly from breaking this rule. The idea is that at the end of development you will always be in chaos, fire fighting mode - and if performance optimizations are thrown in with that as well, it's not a very good scenario. The other problem is that if you don't have a great looking, well running game early, then it's really hard to evaluate how good the game really is. The (law,) if you will, is there's no such thing as a good game with a bad frame rate.
The other big lesson we learned is that features don't save you. This means that development needs to be a progression where you work on a few core features and stay on them until they are done and working. This takes a lot of discipline because developers always want to work on the (next cool thing.) In addition there is always enormous pressure to have all the features that every other game has. Once again you can see a number of high profile failures where games had a ton of features, but that didn't save them. Conversely, there are example of games with very limited features that did quite well - Portal is the recent example, but there are lots of others.
The third major thing is that for MMOs I feel it's important to look good and run well on reasonably spec'd hardware. If you look at the larger, more successful MMOs they tend to run on lots of machines. I think the reason for this is two-fold. The first is that you have to get lots of people playing for the game to be fun, which means the higher the system spec, the smaller the initial population. The other reason is that you need to run the game for a long time. This means that if you have super high end ultra realistic graphics, the next evolution in hardware will date your game. World of Warcraft, notably, has kept itself looking really good by being highly stylized, which means that you don't think of the game as getting graphically behind as quickly as more realistic looking games.
