Search found 66 matches

by reltham
Thu Dec 10, 2009 10:13 pm
Forum: General Bullet Physics Support and Feedback
Topic: Issue with convexSweepTest() not reporting hits
Replies: 8
Views: 5595

Re: Issue with convexSweepTest() not reporting hits

So I commented out a check near the bottom and calcTimeOfImpact() in that file. if ((projectedLinearVelocity+ maxAngularProjectedVelocity)<=result.m_allowedPenetration)//SIMD_EPSILON) return false; Commenting out this made my sweeps work the way I wanted them to, and it didn't seem to have any ill e...
by reltham
Thu Dec 10, 2009 5:17 pm
Forum: General Bullet Physics Support and Feedback
Topic: Issue with convexSweepTest() not reporting hits
Replies: 8
Views: 5595

Re: Issue with convexSweepTest() not reporting hits

That would be great. In our current usage case, even if it just reported that it started in collision with an object that would be enough. Is this something you need to add or is there something I can do to make it happen? I spent a little time digging through the bullet code related to the convex s...
by reltham
Wed Dec 09, 2009 10:56 pm
Forum: General Bullet Physics Support and Feedback
Topic: Issue with convexSweepTest() not reporting hits
Replies: 8
Views: 5595

Re: Issue with convexSweepTest() not reporting hits

Yes, the starting point is in penetration.

Can I make it report these cases somehow?
by reltham
Wed Dec 09, 2009 8:18 pm
Forum: General Bullet Physics Support and Feedback
Topic: Issue with convexSweepTest() not reporting hits
Replies: 8
Views: 5595

Issue with convexSweepTest() not reporting hits

I am having an issue with convexSweepTest(). Testing a box shape against a world containing bvhTriangleMeshShapes. I have diagnosed it down to the following case being the problem: If a triangle in the mesh is large enough that the box object being swept can be in contact with the triangle but not b...
by reltham
Wed Apr 15, 2009 5:40 am
Forum: Applications, Games, Demos or Movies using Bullet
Topic: FreeRealms uses Bullet
Replies: 0
Views: 4661

FreeRealms uses Bullet

We're using Bullet collision & dynamics. We use it for our character controller (our own custom one), line of sight checks, and so forth as well as for vehicle physics in the Racing and Demolition Derby instances. Check out the beta here . Go sign up, we are letting people into the beta in waves ove...
by reltham
Tue Feb 24, 2009 6:40 pm
Forum: General Bullet Physics Support and Feedback
Topic: Massive debug performance loss with composite shape
Replies: 3
Views: 1675

Re: Massive debug performance loss with composite shape

It's a debug build... I'd say it's very expected.
by reltham
Fri Dec 05, 2008 9:41 pm
Forum: General Bullet Physics Support and Feedback
Topic: Bullet 2.73 has been updated to Bullet 2.73 SP1.
Replies: 4
Views: 2705

Re: Bullet 2.73 has been updated to Bullet 2.73 SP1.

Is there some MSVC setting that needs to be enabled for auto-vectorization? With VS 2005 (and I'm sure it's the same for 2008 (and probably 2003)) use this: /arch:SSE OR /arch:SSE2 AND /fp:fast (instead of /fp:precise) If you don't set /fp:fast then it will block some of the SSE/SSE2 optimizations....
by reltham
Thu Nov 20, 2008 6:41 pm
Forum: General Bullet Physics Support and Feedback
Topic: How to deal with "HUGE" world ? Really Need help!~
Replies: 6
Views: 2780

Re: How to deal with "HUGE" world ? Really Need help!~

In the game I am working on, the world is seamless and pretty large. We manage a set of "chunks" around where the player is at. For Bullet we use the btDbvtBroadphase, and it is working great for us. You just add/remove objects and it dynamically adjusts. Also, I would switch to using meters for you...
by reltham
Thu Nov 13, 2008 9:54 pm
Forum: General Bullet Physics Support and Feedback
Topic: bottleneck on btRaycastVehicle
Replies: 8
Views: 3583

Re: bottleneck on btRaycastVehicle

The debug build of bullet is considerably slower. Especially due to things like the vector accessors that optimize out in a release build. In a profile of a debug build, btVector3 accessors (getx, gety, etc) show up as a huge portion of the run, but in release they are gone. I would never do perform...
by reltham
Thu Nov 13, 2008 9:43 pm
Forum: General Bullet Physics Support and Feedback
Topic: My problems with Kinematic Character control
Replies: 2
Views: 1923

Re: My problems with Kinematic Character control

I think you want to apply a force or set a velocity on the rigid body. The bullet internal code will then call the SetWorldTransform() functions when you step the simulation.
by reltham
Thu Nov 13, 2008 6:45 pm
Forum: General Bullet Physics Support and Feedback
Topic: Assertion in btQuantizedBvh
Replies: 1
Views: 1912

Re: Assertion in btQuantizedBvh

I've run into this a couple times myself, and in my case it turned out to be bad index/vertex data both times. First, I had a situation where one of my verts had NANs in it. Second, I had a situation that was changing the vertex data after the bullet object was created and this made the vertices go ...
by reltham
Mon Sep 29, 2008 10:17 pm
Forum: Release Announcements
Topic: Bullet 2.71 released: Disney Maya Plugin, CUDA experiments
Replies: 10
Views: 41210

Re: Bullet 2.71 released: Disney Maya Plugin, CUDA experiments

Just wanted to report that I have been using btScaledBvhTriangleMeshShapes for a few days now and it appears to be working great! Thanks!

Our exported data size dropped almost in half thanks to the BVH sharing. We are using serialized BVH data to speed load times.

Thanks you!
by reltham
Fri Sep 12, 2008 1:20 am
Forum: General Bullet Physics Support and Feedback
Topic: Issue with btOptimizedBvh serialize/deserialize.
Replies: 6
Views: 3060

Re: Issue with btOptimizedBvh serialize/deserialize.

I got this sorted out and submitted issue 94 in the tracker on google.

Roy
by reltham
Wed Sep 10, 2008 9:38 pm
Forum: General Bullet Physics Support and Feedback
Topic: Issue with btOptimizedBvh serialize/deserialize.
Replies: 6
Views: 3060

Re: Issue with btOptimizedBvh serialize/deserialize.

Yeah zeros would work. The main thing is that serializing the same exact object each time should result in the same output even if the destination memory buffer is at a different location.

I will try and get a patch worked up soon. Unless you are going to just go fix it?