Bullet 물리엔진에서 대규모 강체 충돌 처리 성능 프로파일링
Performance Profiling of Massive Rigid Body Collisions under Bullet Physics Engine
1
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
In large scale game battle scenes or the virtual reality environment, the scale of the collision handling has been getting bigger and the importance of it has been gradually increasing. In order to compare the collision handling algorithms, this paper develops six different benchmark based on the Bullet physics engine. By massive rigid body collision simulations, the number of collision pairs, the CPU utilization, the memory usage, and the computation time are profiled. For the CPU utilization, the Simple algorithm is the best and provides 19.3% less compared to the other algorithms. In terms of the computation time, the Dbvt algorithm is the best and provides 8.85 times speedup than the worst Simple algorithm. The memory usage of the 32bitAxisSweep3 algorithm is the largest and is average 2.57 times more than the other algorithms. The optimal collision handling algorithm can be selected through performance profiling by considering the criteria such as the minimum computation time, the limited computing resources, the collision shapes and the number of rigid bodies.
Abstract
In large scale game battle scenes or the virtual reality environment, the scale of the collision handling has been getting bigger and the importance of it has been gradually increasing. In order to compare the collision handling algorithms, this paper develops six different benchmark based on the Bullet physics engine. By massive rigid body collision simulations, the number of collision pairs, the CPU utilization, the memory usage, and the computation time are profiled. For the CPU utilization, the Simple algorithm is the best and provides 19.3% less compared to the other algorithms. In terms of the computation time, the Dbvt algorithm is the best and provides 8.85 times speedup than the worst Simple algorithm. The memory usage of the 32bitAxisSweep3 algorithm is the largest and is average 2.57 times more than the other algorithms. The optimal collision handling algorithm can be selected through performance profiling by considering the criteria such as the minimum computation time, the limited computing resources, the collision shapes and the number of rigid bodies.
참고문헌 (15)
- S. Kockara, "Collision detection: A survey", Systems, Man and Cybernetics, 2007. ISIC. IEEE International Conference on, pp.4046~4051 (2007)
- Gino Van Den Bergen, "Efficient collision detection of complex deformable models using aabb trees", J. Graph. Tools, 2(4), pp.1~13 (1997)
- Stefan Gottschalk, "Obbtree: A hierarchical structure for rapid interference detection", Proceedings of the ACM Conference on Computer Graphics, pp.171~180 (1996), ACM
- Jonathan D. Cohen, "I-collide : An interactive and exact collision detection system for large-scale environments", SI3D, pp.189~218 (1995)
- 김영식, "3D 게임 공간 분할 트리에서 트리 빌드 휴리스틱", 한국게임학회 논문지, 13(4), pp.25~34 (2013), 한국게임학회[DOI]
- 박태정, "게임 충돌 처리 및 실시간 레이트레이싱을 위한 OptiX 실시간 광선추적 프레임워크 상의 Binary Radix Tree 기반 병렬 kd Tree 구현 및 분석", 한국컴퓨터게임학회논문지, 27(3), pp.53~60 (2014), (사)한국컴퓨터게임학회
- 박태정, "GPU 기반 kd-tree를 이용한 병렬 충돌 처리 기법", 한국컴퓨터게임학회논문지, 25(2), pp.195~203 (2012), (사)한국컴퓨터게임학회[DOI]
- 박태정, "게임 오브젝트 Polygon Soup 모델의 근사화 및 보간을 위한 고속 병렬 GPU 적분형 MLS 연산", 한국컴퓨터게임학회논문지, 25(1), pp.179~192 (2012), (사)한국컴퓨터게임학회[DOI]
- E. G. Gilbert, "A fast procedure for computing the distance between complex objects in three-dimensional space", IEEE Journal of Robotics and Automation, 4, pp.193~203 (1988)
- "GIMPACT project"[URL]
- F. Liu, "Real-time collision culling of a million bodies on graphics processing unit", ACM Transactions on Graphics (TOG), 29(6), pp.154~161 (2010)
- Quentin AVRIL, "A broad phase collision detection algorithm adapted to multi-cores architectures", VRIC 2010 Proceedings, pp.12~ (2010)
- Min Tang, "Multi-core collision detection between deformable models", Computers & Graphics (2008)
- Everton Hermann, "Interactive physical simulation on multicore architectures", Eurographics Workshop on Parallel and Graphics and Visualization, EGPGV’09 (2009)
- "Bullet Physics Engine"[URL]