Contact Info / Websites
A long time ago I tried to make a physics engine. It sucked. I had no idea where to start and the entire thing fell apart before it was made (this was before know any OOP).
A few years later I tried again. It was a marginal success and I managed to make a fun rag-doll simulator that could pass for actual physics at a glance (minus collisions with bodies aside from the ground which was a picture, not mathematically defined). The game was actually pretty fun and I wish that I had made something of it.
Its going to be different this time arround. I am almost done physics 12, so I have a good understanding of dynamics and such. I figured I would try again and, low and behold, after only a few hours of work I have already created a pretty sophisticated "spring" engine.
I am going to add in collisions and friction over the next few days, and if I do it properly (which I think I will seeing as I have the entire engine already planned out), rolling objects and movement on slopes should emerge naturally from the equations. As it is now, the engine handles rotating bodies almost perfectly except for the fact that it uses cords to calculate velocity instead of arcs (meaning that the objects very, very gradually slow down).
Its working out very nicely! I now have semi-rigid bodies, torque, point objects and ball objects. I think I only have to add in angular momentum for the ball object and collisions between objects.
mouse controls torque objects, arrows supply force:
http://media.putfile.com/point-mass-br idge (the centre point is more massive than the others)