test commit using svn
This commit is contained in:
erwincoumans
218
readme.txt
218
readme.txt
@@ -1,108 +1,110 @@
|
|||||||
Bullet 3.x GPU rigid body pipeline using OpenCL.
|
Bullet 3.x GPU rigid body pipeline using OpenCL.
|
||||||
|
|
||||||
Note that the Bullet 2.x svn repository from http://bullet.googlecode.com
|
Note that the Bullet 2.x svn repository from http://bullet.googlecode.com
|
||||||
is being merged into this repository.
|
is being merged into this repository.
|
||||||
|
|
||||||
1) The old Bullet2 demos are moved from ObsoleteDemos to AllBullet2Demos
|
1) The old Bullet2 demos are moved from ObsoleteDemos to AllBullet2Demos
|
||||||
2) A new Bullet 3 API is created
|
2) A new Bullet 3 API is created
|
||||||
3) All Bullet2 functionality is added to Bullet 3
|
3) All Bullet2 functionality is added to Bullet 3
|
||||||
Until this is done, you can use the Demos3/BasicGpuDemo/b3GpuDynamicsWorld
|
Until this is done, you can use the Demos3/BasicGpuDemo/b3GpuDynamicsWorld
|
||||||
or explore the Demos3/GpuDemos to check out Bullet 3.
|
or explore the Demos3/GpuDemos to check out Bullet 3.
|
||||||
|
|
||||||
The entire collision detection and rigid body dynamics is executed on the GPU.
|
You can still use svn or svn externals using the github git repository: use svn co https://github.com/erwincoumans/bullet3/trunk
|
||||||
|
|
||||||
Requirements:
|
The entire collision detection and rigid body dynamics is executed on the GPU.
|
||||||
A high-end desktop GPU, such as an AMD Radeon 7970 or NVIDIA GTX 680 or similar.
|
|
||||||
We succesfully tested the software under Windows, Linux and Mac OSX.
|
Requirements:
|
||||||
The software currently doesn't work on OpenCL CPU devices. It might run
|
A high-end desktop GPU, such as an AMD Radeon 7970 or NVIDIA GTX 680 or similar.
|
||||||
on a laptop GPU but performance is likely not very good.
|
We succesfully tested the software under Windows, Linux and Mac OSX.
|
||||||
|
The software currently doesn't work on OpenCL CPU devices. It might run
|
||||||
|
on a laptop GPU but performance is likely not very good.
|
||||||
License
|
|
||||||
All source code files are licensed under the permissive zlib license
|
|
||||||
(http://opensource.org/licenses/Zlib) unless marked differently in a particular folder/file.
|
License
|
||||||
|
All source code files are licensed under the permissive zlib license
|
||||||
|
(http://opensource.org/licenses/Zlib) unless marked differently in a particular folder/file.
|
||||||
Build instructions:
|
|
||||||
|
|
||||||
Windows:
|
Build instructions:
|
||||||
Click on build3/vs2010.bat and open build3/vs2010/0MySolution.sln
|
|
||||||
|
Windows:
|
||||||
Linux and Mac OSX gnu make
|
Click on build3/vs2010.bat and open build3/vs2010/0MySolution.sln
|
||||||
In a terminal type:
|
|
||||||
|
Linux and Mac OSX gnu make
|
||||||
cd build3
|
In a terminal type:
|
||||||
|
|
||||||
Dependend on your system (Linux 32bit, 64bit or Mac OSX) use one of the following lines
|
cd build3
|
||||||
./premake4_linux gmake
|
|
||||||
./premake4_linux64 gmake
|
Dependend on your system (Linux 32bit, 64bit or Mac OSX) use one of the following lines
|
||||||
./premake4_osx gmake
|
./premake4_linux gmake
|
||||||
|
./premake4_linux64 gmake
|
||||||
Then
|
./premake4_osx gmake
|
||||||
|
|
||||||
cd gmake
|
Then
|
||||||
make
|
|
||||||
|
cd gmake
|
||||||
Mac OSX Xcode
|
make
|
||||||
Click on build3/xcode4.command or in a terminal window execute
|
|
||||||
./premake_osx xcode4
|
Mac OSX Xcode
|
||||||
|
Click on build3/xcode4.command or in a terminal window execute
|
||||||
Usage:
|
./premake_osx xcode4
|
||||||
|
|
||||||
The main demo executable will be located in the bin folder.
|
Usage:
|
||||||
The demo starts with App_Bullet3_OpenCL_Demos_*
|
|
||||||
|
The main demo executable will be located in the bin folder.
|
||||||
You can just run it though a terminal/command prompt, or by clicking it.
|
The demo starts with App_Bullet3_OpenCL_Demos_*
|
||||||
|
|
||||||
|
You can just run it though a terminal/command prompt, or by clicking it.
|
||||||
There are some command-line options, you can see using the --help option. For example, this will perform a benchmark writing to some files:
|
|
||||||
|
|
||||||
./App_Bullet3_OpenCL_Demos_clew_gmake --benchmark
|
There are some command-line options, you can see using the --help option. For example, this will perform a benchmark writing to some files:
|
||||||
|
|
||||||
[--selected_demo=<int>] Start with a selected demo
|
./App_Bullet3_OpenCL_Demos_clew_gmake --benchmark
|
||||||
[--benchmark] Run benchmark and export results to file
|
|
||||||
[--maxFrameCount=<int>] Run the benchmark for <int> frames
|
[--selected_demo=<int>] Start with a selected demo
|
||||||
[--dump_timings] Print the profile timings to console
|
[--benchmark] Run benchmark and export results to file
|
||||||
[--cl_device=<int>] Choose a certain OpenCL device
|
[--maxFrameCount=<int>] Run the benchmark for <int> frames
|
||||||
[--cl_platform=<int>] Choose a certain OpenCL platform
|
[--dump_timings] Print the profile timings to console
|
||||||
[--disable_cached_cl_kernels] Disable loading cached binary OpenCL kernels
|
[--cl_device=<int>] Choose a certain OpenCL device
|
||||||
[--x_dim=<int>] Change default demo settings (x,y,z)
|
[--cl_platform=<int>] Choose a certain OpenCL platform
|
||||||
[--pair_benchmark_file=<filename>] Load AABB's from disk for the PairBench
|
[--disable_cached_cl_kernels] Disable loading cached binary OpenCL kernels
|
||||||
[--no_instanced_collision_shapes] Disable collision shape instancing (for tests)
|
[--x_dim=<int>] Change default demo settings (x,y,z)
|
||||||
[--no_shadow_map] Disable shadowmap rendering
|
[--pair_benchmark_file=<filename>] Load AABB's from disk for the PairBench
|
||||||
[--shadowmap_resolution=<int>] Change the resolution of the shadowmap
|
[--no_instanced_collision_shapes] Disable collision shape instancing (for tests)
|
||||||
[--shadowmap_size=<int>] Change the worldspace size of the shadowmap
|
[--no_shadow_map] Disable shadowmap rendering
|
||||||
[--use_uniform_grid] Use uniform grid broadphase (no all scenes work)
|
[--shadowmap_resolution=<int>] Change the resolution of the shadowmap
|
||||||
[--use_jacobi] Use GPU parallel Jacobi solver (instead of PGS)
|
[--shadowmap_size=<int>] Change the worldspace size of the shadowmap
|
||||||
[--use_large_batches] Use a different strategy for the constrains solver
|
[--use_uniform_grid] Use uniform grid broadphase (no all scenes work)
|
||||||
[--debug_kernel_launch] Show debug info at start/end of each kernel launch
|
[--use_jacobi] Use GPU parallel Jacobi solver (instead of PGS)
|
||||||
[--use_dbvt] Use the CPU dynamic BVH tree broadphase
|
[--use_large_batches] Use a different strategy for the constrains solver
|
||||||
[--allow_opencl_cpu] Allow to use an OpenCL CPU device
|
[--debug_kernel_launch] Show debug info at start/end of each kernel launch
|
||||||
|
[--use_dbvt] Use the CPU dynamic BVH tree broadphase
|
||||||
|
[--allow_opencl_cpu] Allow to use an OpenCL CPU device
|
||||||
You can use mouse picking to grab objects. When holding the ALT of CONTROL key, you have Maya style camera mouse controls.
|
|
||||||
Press F1 to create a screenshot. Hit ESCAPE to exit the demo app.
|
|
||||||
|
You can use mouse picking to grab objects. When holding the ALT of CONTROL key, you have Maya style camera mouse controls.
|
||||||
|
Press F1 to create a screenshot. Hit ESCAPE to exit the demo app.
|
||||||
Bullet 3.x only implements a small set of collision shapes and constraints:
|
|
||||||
|
|
||||||
Static plane
|
Bullet 3.x only implements a small set of collision shapes and constraints:
|
||||||
Static concave triangle mesh
|
|
||||||
Sphere
|
Static plane
|
||||||
Convex Polyhedron
|
Static concave triangle mesh
|
||||||
Compound of Convex Polyhedra
|
Sphere
|
||||||
|
Convex Polyhedron
|
||||||
Bullet 3.x uses the separating axis test (SAT) between convex polyhedra, testing all vertex - face and edge - edge combinations. For performance it is best to keep the number of edges in a convex polyhedron low, using simple shapes such as a tetrahedron or a box.
|
Compound of Convex Polyhedra
|
||||||
|
|
||||||
The constraint solver currently supports two constraints:
|
Bullet 3.x uses the separating axis test (SAT) between convex polyhedra, testing all vertex - face and edge - edge combinations. For performance it is best to keep the number of edges in a convex polyhedron low, using simple shapes such as a tetrahedron or a box.
|
||||||
|
|
||||||
point to point constraint (ball-socket
|
The constraint solver currently supports two constraints:
|
||||||
fixed constraint
|
|
||||||
|
point to point constraint (ball-socket
|
||||||
It can be extended to other constraint types. The constraint solver uses the same algorithm as Bullet 2.x.
|
fixed constraint
|
||||||
|
|
||||||
It is possibly to try out Bullet 3.x using the Bullet 2.x API, using the b3GpuDynamicsWorld bridge:
|
It can be extended to other constraint types. The constraint solver uses the same algorithm as Bullet 2.x.
|
||||||
Copy the source code of both versions in the same location, and click on build3/vs2010_bullet2gpu.bat to generate Visual Studio project files.
|
|
||||||
|
It is possibly to try out Bullet 3.x using the Bullet 2.x API, using the b3GpuDynamicsWorld bridge:
|
||||||
See docs folder for further information.
|
Copy the source code of both versions in the same location, and click on build3/vs2010_bullet2gpu.bat to generate Visual Studio project files.
|
||||||
|
|
||||||
|
See docs folder for further information.
|
||||||
|
|
||||||
|
|||||||
Reference in New Issue
Block a user