GPGPU Acceleration for Skeletal Animation-comparing OpenCL with CUDA and GLSL

Shousheng Liu, Ge Chen, Chunyong Ma, Yong Han
College of Information Science and Engineering, Ocean University of China, Qingdao 266100, China
Journal of Computational Information Systems 10: 16, 7043-7051, 2014


   title={GPGPU Acceleration for Skeletal Animation-comparing OpenCL with CUDA and GLSL},

   author={LIU, Shousheng and CHEN, Ge},

   journal={Journal of Computational Information Systems},






Download Download (PDF)   View View   Source Source   



The existing matrix palette algorithms for skeletal animation are accelerated by the technique GPGPU based on GLSL or CUDA. Because GLSL is extended from graphics library OpenGL, it couples the rendering and calculations together closely and forces itself not convenient to reuse, meanwhile CUDA is designed only for NVIDIA GPUs. In this paper GPGPU based on OpenCL is proposed for accelerating skeletal animations. OpenCL brings portability both on software and hardware. The experimental results show that the parallel scheme based on OpenCL can run on GPUs from AMD and NVIDIA. And the speedup is comparable with CUDA or GLSL.
Rating: 2.0/5. From 2 votes.
Please wait...

* * *

* * *

HGPU group © 2010-2021 hgpu.org

All rights belong to the respective authors

Contact us: