Exactly what I was thinking as I was kicking the idea around with a coffee buddy of mine, the amount of active artists can make or break a project, but a flaky/buggy engine can too.

It's a pyramid of engine, graphics and game design, but notice the "content" part makes up 2/3rds of the game. You could never hope to get a 1st class level of graphics to start, but it can improve over time. But expectations too high and too early can only frustrate people.

If I had to do art for it I would probably go with something like Blender as is has some nice actor model animation exporting features that are nice (and uses Python which I really like). Then all you have to do is find free models and re-skin and modify them for what you need like creatures, people, terrain, an buildings. But finding a group of guys to get pumped about it is the toughie.

Hmm have to check it out, thanks :)