Some dynamic memory allocations have been replace by pool allocation or stack allocations.
quantized aabb versus quantized aabb overlap check is made branch-free (helps a lot on consoles PS3/XBox 360)
Collision algorithms are now created through a new btDefaultCollisionConfiguration, to decouple dependency (this is the API change):
Example:
btDefaultCollisionConfiguration* collisionConfiguration = new btDefaultCollisionConfiguration();
m_dispatcher = new btCollisionDispatcher(collisionConfiguration);
Added btStackAlloc to Bullet (right now only used by btGjkEpa)
removed default constructors of btCollisionWorld/btDiscreteDynamicsWorld, to reduce link-time dependencies
Improved vehicle interpolation of wheels, and added Z-up axis option for the Demo
made 'getWorldTransform' const method in btMotionState
added future 'deactivationCallback'(not used yet)
Fixed memoryleak in btOptimizedBvh (delete []m_contiguousNodes;)
Changed DemoApplication::localCreateRigidBody, so it adds the rigidbody to the btDynamicsWorld.
Added check for duplicate objects in world when adding.
Added assert to prevent setLinearVelocity on static rigidbodies
Added btCollisionFilterGroups to btBroadphaseProxy
removed duplicate 'btBroadphaseProxy* m_broadphaseProxy;' in btRigidBody
removed two cached optimizations, type in btTransform and cached inverse transform (todo: test performance impact)
committed fixes that make the code adhere to 'who creates it, also destroys it'
* Unused variables.
* Missing newlines at ends of #included files.
* signed int loop variables where the termination condition is an unsigned 'get number of' function.
* 'NULL' used inappropriately for an integer or character constant (NULL is a pointer)
* abstract base classes with no virtual destructor.
* Floating point constants used to initialise integer variables.
Moved optional code to Extras: AlgebraicCCD,EPA,quickstep
Moved SimpleBroadphase data to OverlappingPairCache, and derive both SimpleBroadphase and AxisSweep3 from OverlappingPairCache.
Added ParallelPhysicsEnvironment (prepair more parallel mainloop)
Upgraded hardcoded limit from 1024/8192 to 32766/65535 (max objects / max overlapping pairs)
