Search found 49 matches
- Thu Nov 29, 2012 9:12 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: Optimized line intersection with height field
- Replies: 0
- Views: 3140
Optimized line intersection with height field
This is not a complete description of the modification since I have modified more things in the height field class. Just see processTrianglesAlongLine as pseudo code for how to improve the speed for line intersections. The code assume that btScalar is defined as float since that is what I use. Some ...
- Tue May 29, 2012 1:58 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: Collision with interior of a collision shape.
- Replies: 2
- Views: 3626
Re: Collision with interior of a collision shape.
The inside would be concave and can't use convex to convex collision algorithms. It would be easier to just have a btBvhTriangleMeshShape and keep it static.
- Sat May 05, 2012 9:56 am
- Forum: General Bullet Physics Support and Feedback
- Topic: Creating Custom Collision Object
- Replies: 1
- Views: 2675
Re: Creating Custom Collision Object
You do not have to make your own collision algorithm for static triangle shapes. If you inherit btConcaveShape like in the heightfield class does, the engine will call processAllTriangles with an AABB for the section that it is interested in and use it for collisions.
- Fri May 04, 2012 7:02 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: I made a new convex hull shape optimizer
- Replies: 0
- Views: 2800
I made a new convex hull shape optimizer
Since the convex hull optimizer that I tried caused my low detailed convex hulls to become more complex than before, I made a new convex hull optimizer that will remove every point inside the convex hull without adding any new points given that the input does not have any close pair of points. To re...
- Wed May 02, 2012 11:56 am
- Forum: General Bullet Physics Support and Feedback
- Topic: Please help me bust some myths about really large worlds
- Replies: 3
- Views: 4564
Re: Please help me bust some myths about really large worlds
Are you defining btScalar as a 64 bit float?
- Tue May 01, 2012 6:17 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: Optimizing world with double link list
- Replies: 0
- Views: 2811
Optimizing world with double link list
I am trying to optimize the Bullet engine by keeping a non allocating double link list of each active rigid body so that the world don't have to ask each sleeping body if it is active. The only thing missing is a way to know when a rigid body has changed it's activation state so that isActive() give...
- Tue May 01, 2012 12:08 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: Wheel Control / Spinning question
- Replies: 1
- Views: 2404
Re: Wheel Control / Spinning question
You might be able to make your own friction function. Use a low viscous friction and a high stribeck friction to make the wheels loose grip when spinning.
- Tue May 01, 2012 11:22 am
- Forum: General Bullet Physics Support and Feedback
- Topic: Reusing shapes - simple question
- Replies: 1
- Views: 2945
Re: Reusing shapes - simple question
Each size of the cube need a different shape.
- Sat Apr 28, 2012 3:17 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: Cross-platform determinism
- Replies: 3
- Views: 4548
Re: Cross-platform determinism
Even on the same version of Windows, different CPUs do floatingpoint calculations in different ways so that the result is noticeably different after a divergent simulation.
- Sat Apr 28, 2012 3:04 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: Smooth Rotation of a Rigid Body
- Replies: 1
- Views: 2806
Re: Smooth Rotation of a Rigid Body
If your character is a round shape then you only need to rotate the visual representation. By running backwards and sidestepping, your character can change direction using a fast acceleration while the direction is slowly turning to the walking direction. The head can turn to the walking direction f...
- Sat Apr 28, 2012 2:53 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: Rolling back solved constraints
- Replies: 2
- Views: 2999
Re: Rolling back solved constraints
I would use a compound shape for each heavy body so that the small details will always be in the right place.
You can also use class inheritance if you want to modify the engine while still being able to upgrade to a new version of Bullet.
You can also use class inheritance if you want to modify the engine while still being able to upgrade to a new version of Bullet.
- Thu Apr 26, 2012 6:25 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: Grid Collision
- Replies: 5
- Views: 5168
Re: Grid Collision
The default broadphase is using a dynamic tree for each group of bodies and therefor the lookup cost for small dynamic bodies is only O(log n) where n is the number of static bodies. My game use thousands of static bodies without any problem.
- Thu Apr 26, 2012 6:15 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: btGhostObject assertion failed on delete
- Replies: 2
- Views: 3011
Re: btGhostObject assertion failed on delete
You must remove the object from the world before deleting it.
Code: Select all
btGhostObject::~btGhostObject()
{
///btGhostObject should have been removed from the world, so no overlapping objects
btAssert(!m_overlappingObjects.size());
}
- Thu Apr 26, 2012 5:57 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: General and building questions
- Replies: 3
- Views: 5858
Re: General and building questions
A wrapper might give some built in functionality for debug drawing and synchronization with graphical objects.
I recomend that you make a collision shape that generate triangles from a dynamic octree.
I recomend that you make a collision shape that generate triangles from a dynamic octree.
- Sun Apr 22, 2012 3:53 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: Two ways of getting world transform.
- Replies: 2
- Views: 3060
Re: Two ways of getting world transform.
I don't see how one can be removed without forcing a lot of people to rewrite their code.