Search found 12 matches
- Thu Apr 07, 2011 7:06 am
- Forum: General Bullet Physics Support and Feedback
- Topic: Bullet on XNA
- Replies: 61
- Views: 567424
Re: Bullet on XNA
Just a FYI, it seems to work with monoxna if you hack out the XNA references in the csproj files. http://code.google.com/p/monoxna/ Mind you, I compiled a 'lite' version monoxna with just the classes needed for bullet-xna. I've put in a few bug reports on the google code page. Issue 1:Collision with...
- Thu Oct 15, 2009 5:36 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: Weird Code, maybe buffer overrun in btBoxBoxDetector::inte..
- Replies: 3
- Views: 5014
Re: Weird Code, maybe buffer overrun in btBoxBoxDetector::inte..
After debugging and stepping through it again, it turns out that it moves the pointer forward to the 8th element and then ends the for loop without accessing what's in the element so it should be fine.
- Thu Oct 15, 2009 8:35 am
- Forum: General Bullet Physics Support and Feedback
- Topic: Choosing the best CollisionShape for objects
- Replies: 2
- Views: 3419
Re: Choosing the best CollisionShape for objects
I can tell you that, yes, you do need to register the GImpact collision Algorithm to get GImpactShapes to work. I can also tell you that yes, you can use dynamic bodies with GImpactMeshShape. There is a performance penalty using GImpactMeshShape.. and a slight stability impact (simulation stability,...
- Thu Oct 15, 2009 3:32 am
- Forum: General Bullet Physics Support and Feedback
- Topic: Weird Code, maybe buffer overrun in btBoxBoxDetector::inte..
- Replies: 3
- Views: 5014
Weird Code, maybe buffer overrun in btBoxBoxDetector::inte..
The code in btBoxBoxDetector::intersectRectQuad seems to either be really strange.. or a buffer overrun. the method signature is static int intersectRectQuad2 (btScalar h[2], btScalar p[8], btScalar ret[16]) But, I paused there and ran through all of the elements.. and it looked okay.. until I tried...
- Wed Oct 14, 2009 6:59 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: Bullet Collision Detection
- Replies: 5
- Views: 6316
Re: Bullet Collision Detection
To use a directX mesh in bullet, you would need to convert the mesh to a triangle mesh with vertices, and indices that make triangles.
You can then use a btTriangleMesh, or a btTriangleIndexVertexArray to create a btGImpactShape, which can be used in bullet for collision detection.
You can then use a btTriangleMesh, or a btTriangleIndexVertexArray to create a btGImpactShape, which can be used in bullet for collision detection.
- Wed Oct 14, 2009 6:53 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: btManifoldPoint and its members
- Replies: 1
- Views: 3181
Re: btManifoldPoint and its members
It would be nice to know a 'life cycle' of a btPersistentManifold and btManifoldPoint... like.. which part of the simulation creates it, changes it, and releases it. A couple things that I do know... the btConstraintSolver (this is your btSequentialImpulseConstraintSolver) modifies it and uses it to...
- Sun Oct 11, 2009 2:41 am
- Forum: General Bullet Physics Support and Feedback
- Topic: btAxisSweep3 question
- Replies: 1
- Views: 2842
Re: btAxisSweep3 question
nevermind, I think I figured it out.
- Sat Oct 10, 2009 11:38 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: btAxisSweep3 question
- Replies: 1
- Views: 2842
btAxisSweep3 question
in btAxisSweep3Internal::sortMinDown, pEdge and pPrev are defined like; Edge* pEdge = m_pEdges[axis] + edge; Edge* pPrev = pEdge - 1; where m_pEdges[axis] + edge actually means m_pEdges[axis][edge], and Edge* pPrev = pEdge - 1; is a short way of saying, pPrev = m_pEdges[axis][edge - 1] This gets two...
- Thu Oct 08, 2009 10:25 am
- Forum: General Bullet Physics Support and Feedback
- Topic: odd syntax question.
- Replies: 1
- Views: 2651
odd syntax question.
I was just noting that some of the code in the constraints have 'odd' syntax. For example; getRigidBodyA().getCenterOfMassTransform().getBasis() * m_rbAFrame.getBasis().getColumn(2); If I were to look at this linearly, I would 'think' it's setting the origin of rigidbodyA's center of mass transform ...
- Mon Sep 21, 2009 9:23 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: What?int pairIndex = int(pair - &m_overlappingPairArray[0]);
- Replies: 4
- Views: 4563
Re: What?int pairIndex = int(pair - &m_overlappingPairArray[0]);
Ok, that makes sense.
Thanks.
Thanks.
- Mon Sep 21, 2009 8:48 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: What?int pairIndex = int(pair - &m_overlappingPairArray[0]);
- Replies: 4
- Views: 4563
Re: What?int pairIndex = int(pair - &m_overlappingPairArray[0]);
Much appreciated, sir. Helps me understand what's going on there. An aligned object array.. so the memory address of the pair - the memory address of the first element is the total number of elements between pair and the first element. :) So a compatible but slower algorithm would be ; int pairIndex...
- Mon Sep 21, 2009 6:54 pm
- Forum: General Bullet Physics Support and Feedback
- Topic: What?int pairIndex = int(pair - &m_overlappingPairArray[0]);
- Replies: 4
- Views: 4563
What?int pairIndex = int(pair - &m_overlappingPairArray[0]);
In btOverlappingPairCache.cpp in btHashedOverlappingPairCache::removeOverlappingPair(btBroadphaseProxy* proxy0, btBroadphaseProxy* proxy1,btDispatcher* dispatcher) Can someone explain, int pairIndex = int(pair - &m_overlappingPairArray[0]); It seems to do a subtraction of the btBroadPhasePair, b...