Apparently the gang at NetDevil learned a lot about their experience with AA. Herman Peterscheck talked freely and at length about these lesson; lessons that are being applied across the board to all the games they are currently developing, including Jumpgate Evolution.
Probably the most important one is to polish the game early and not at the end just before launch. A time when everyone on the team is in full on "fire fighting" mode. Chaos almost always leads to poor execution and ruin. Another is that if you don't have a great looking game that runs well out of the gate it's nigh impossible to evaluate how good it actually is. According to Peterscheck, "The 'law,' if you will, is there's no such thing as a good game with a bad frame rate."
Features won't save a game. As such, development needs to be done on a few key features, and they must be worked on until they're bright and shiny and purr like kitten. Apparently this mentality takes a lot of discipline because developers are humans, and like all humans they'd rather work on (or read about) the "next cool thing."
Another major issue is to have a game that looks good yet runs well on a reasonably equipped PC. World of Warcraft runs on virtually any machine. Lower end machines allow for great accessibility, and greater accessibility leads to a lot more people playing the game.
When asked what their motivation was to return to something they've already done (Jumpgate morphing into Jumpgate Evolution), Peterscheck gave an answer that once again gets us excited about JGE. "There's no big action space MMO, something Freelancer Online or Wing Commander Online. We want to play that game and since it doesn't exist we want to make it." A second reason is as old as time itself: practice makes perfect. Based on that line of thinking, doing another space action MMO should make it exponentially better than the first one.
[Via Rock, Paper, Shotgun]