Box2D Forums

Hi, I’ve been trying to port the most recent code of Box2D over to my Physics2D.Net Engine and I am having some problems.

I have basically ported the b2ContactSolver class and kept the rest of my code the same. (I’m starting small)

The first problem I have come across is how I represent immovable platforms. I currently set the Body’s Mass to Positive Infinity and has worked until now. The problem is with the b2ContactSolver::SolvePositionConstraints method. It does some math that results with NAN being added to the Body’s position. Naturally this is not a desired behavior. I have tried to do some changes but they either result with bodies slowly sinking into the immovable platforms or them jumping around like popcorn.

Basically I’m wondering if someone could give a few pointers on how to resolve this problem. Also how does Box2D represent immovable platforms?

For static bodies the mass and inverse mass should be zero. Same for the rotational inertia.

Let us know when you finish the C# port. I don't think one exists yet. Thanks!

Maybe this is a stupid question, but to me it looks like there is no "joint" with a "torque" property. Any comment, please?

I believe this c# port of box2d is dead. Box2DX is the one you want... I noticed you've already seen the other thread.

Physics2D.Net is 2 dimensional unbending corpse physics engine printed in C#. And well Box2DX is a C# port of Box2D - Erin Catto's 2D Physics Engine.
To effort with XNA, Silver light and .Net. It is an absolute redraft of my unique physics 2D found on source fake. It is CLS acquiescent. At present it has only one crash solver. The narrow phase smash detector is measured to be part of the solver. The Solver is chronological impulses with a coldness grid used for thin phase detection. There are 4 Broad phase detectors to choose from.
---------------------
hp coupon codes

Also, the one I'm working on, Box2CS, which is a frontend to Box2C, and runs at native C speed.

-P

It is an complete abstract of my different physics 2D begin on antecedent fake. It is CLS acquiescent. At present it has alone one blast solver. The attenuated appearance accident detector is abstinent to be allotment of the solver.