Real-Time Physics Simulation Forum

Also curious is: the distance traveled along the x-axis does not agree with your speed and timestep: time = 0.05 sec rate = 250 m/sec distance = rate * time = 250 m/sec * 0.05 sec = 12.5m I wonder if you're bumping up against a maximum speed limit, especially since your total speed is so steady. May...

Alas, physics simulations require approximation and abstraction in order to run in real-time on real-life hardware. The discrepancy between how you think it should work, and how it actually works, is a manifestation of that truth. The fact of the matter is: it is possible to get the effect you want,...

Sorry about that. I updated the link to what I hope is a publicly shared google photo. Yes, I guess I misunderstood your problem. I still might not understand it but I'll take another swing (let me know if this is in the right direction). If the vehicle is being steered directly... and you're trying...

Yes, I believe I made an error in my derivation. Let me try again from scratch... I tried again on paper and took a photo . Fig1 shows the leaning motorcycle. Fig2 shows the balance of forces which apply torque on the center of mass with focal point where the tire touches the road. Fig3 shows the ve...

I have fixed a few bugs in your code, and added a bunch of comments: btRigidBody *car = common.brickCars[a]->body; btTransform worldTransform = car->getWorldTransform(); worldTransform.setOrigin(btVector3(0,0,0)); btVector3 up = btVector3(0,1,0); // When we set 'normal' to be 'up' we are assuming th...

Yes, the center of mass (COM) defines the local-frame origin which means: in order to shift the COM relative to the geometry you must shift the geometry instead. For this reason: btRigidBody::setCenterOfMassTransform() can only be understood to mean "set the COM transform in the world-frame&quo...

Yeah you don't want to use btRigidBody::setCenterOfMassTransform() . That sets the world_transform of the object's center of mass and doesn't change the center of mass in the object's local-frame. The center of mass in the object's local-frame is, by definition, the origin of the shape geometry. Con...

For a rolling motorcycle the tilt of the vehicle is a balance between the torque from gravity and that of the centripetal force. In the case where the road is flat and the motorcycle has a forward speed and some tilt angle (from the vertical) then the torque balance relation looks like: (1) mass * s...

What you want to do is to also set the velocities of the kinematic object correctly. Setting the velocities doesn't help move it to new transoforms, but it does help the collision system impart correct velocity on the dynamic object, allowing it to be knocked out of the way instead of getting into a...

Every system has its limits and you are adventuring outside those of the softbody collision system. Meanwhile, if you constantly invalidate the softbody cache (which was probably introduced to speed things up) then it would not be too surprising to suffer consequences of poor performance. Since you'...

Slamming the transform of the child object doesn't just work?

I've never personally used scaling on spheres however upon inspection of the code it looks like btSphereShape derives from btConvexInternalShape which has a m_localScaling data member. Try it and maybe it will work. Alternatively, you could use a btConvexHullShape . It is possible to set the collisi...

I should probably mention: one problem with slamming the linear velocity is it can override collision rebound effects and if the velocity blend is too strong it might lead to pushing the character through obstacle geometry (e.g. tunnel through a btBvhTriangleMeshShape terrain). To avoid this problem...

After thinking about it some more I realized: if I just wanted a very simple rolling controller that pretty much always moved in the direction I wanted then I would blend toward a target linear velocity and then slam the angular velocity to always agree. In C++ it might look something like the code ...

The broadphase keeps track of the AABBs and also overlapping AABB pairs between distinct objects that might need narrowphase collision check to compute the contact points (if any). Ignoring narrowphase costs... there is overhead in: (1) knowing all the AABBs and overlaps and (2) adding new AABBs to ...