Abstract
Soft-body simulation is widely used in animation, prostheses, organs, and so on. The most common way is to use 3D software. However, Their simulation models and the data processing speed are limited. Therefore, one model based on the mass-spring mechanism is proposed. To realize real-time rendering, a parallel computing architecture based on the CUDA architecture is introduced. Besides, to increase the accuracy of the simulation, the Verlet integration is employed. The work is to check whether the massively parallel computing method based on the CUDA architecture improves the rendering performance. To meet the minimum requirement to make the human eye comfortable, all the tests had at least a 60 Hz refreshing rate. Also, the soft body of mass-particles and springs has a uniform width and depth, but the height is much smaller. It was modeled to fall under the influence of gravity, and then, to impact on a rigid object. The serial and the parallel methods were not significantly different when the rendering nodes were less than 2,000, but it became apparent when the number of nodes reached 10,000. Therefore, the simulation efficiency of a soft body is improved by the proposed method.
Structure of readiness of pedagogical university students for prevention of addictive behavior
