I think I know why my tank was skidding sideways. I was using central force, and it was applying it along the world's X axis. So when I tried to turn it seemed to skid sideways. I've got the physics based bullets working now, and it's spectacular. Now I just have to audit their mass and velocities to get them working exactly like their real life counterparts.

I was thinking about the invisible wheel concept at the beginning, but I may be able to do this without it. I'm not sure yet. The driving physics are going to be the hardest part, so I'm saving it for last, really. I've got a lot to do. The reason I'm doing this with physics is because they're faster than c_move and other instructions (says the manual), and more realistic. The target audience is people who enjoy realistic WWII simulations like IL-2 Sturmovik, or WWIIOL. I'm trying to eliminate the arcade-style elements of it. The main way this game will be played is online, however there will be a single player mode.

I think this engine was a major win though. I doubt I'll ever regret spending the money on it. It's been able to do everything I've needed it to so far, and at a cost only the fraction of most professional engines. I'm still trying to get used to it and Lite-C though. I used to program small apps in C++, and this is quite different, but more convenient for games.

Another question though:

Do you think Lite-C has the potential to make a truly convincing and realistic flight model for aircraft? Or would I need to use a "higher" language?