The Rough Realizing Idea Of A Meshless Simulation Scheme Based On Continuum Mechanics To The GPU Platform

Depending on current performance of today' s commercial PC hardware and the matching physically-based dynamic simulation scheme which has potential of optimization and wide-ranged usability,the main purpose is correspondingly adjusting the inner algorithm to adapt it to the hardware' s architecture , in order to achieve higher rendering speed and meanwhile maintain certain rendering quality. From here on , such arrangements are taken as the simplified basic data structure for further innovation of the chosen dynamic simulation scheme. At the same time , some conceptual operations from the simulation scheme must be efficiently realized and the detailed theoretical explanations must be given out.Understanding and comprehensively grasping physically-based dynamic simulation scheme of objects with meshless volumetrical sampling newly presented in CG. Because of the theoretical physics foundation of this scheme, which is continuum mechanics, all the contents of continuum mechanics that' s concerned in this scheme need to be systematically recognized and appropriately operated. Hence a solid theoretic assumption is established for the derivation of such scheme. It' s also very beneficial for the thorough comprehension of its algorithmic essence. For the object with meshless volumetrical sampling as the result of simulation stage , the scheme gets a final sampling points set which is varied from original sampling points set. But what we really want to get is the rendering of object surface. From the final sampling points set of the scheme above mentioned, we can get the final sampling points set of object surface and the corresponding attribute set using relevant algorithm. Due to the iterative property of such algorithm, there arises the issues of optimizations, including algorithmic optimization and cost optimization. The main aim is to optimize the cost.Traversally reviewing the current existing dynamic simulation schemes of various types. The main principle considered is the synthesized evaluation target of performance and effect.. The potential realizing means of some conceptual operations from the simulation scheme are distinguished according to its feasibility and efficiency after the fundamental scheme being determined. Apart from the feasibility owned by the scheme , efficiency becomes the key focus.For the purpose mentioned above, a rough idea of how to transplant some general intensive physical calculation to the GPU platform is given in which the possibility of performance elevation has been enlarged to some extent. At the same time the efficient realization of some conceptual operations from the simulation scheme is carried out. In addition , a non-traditional rendering on traditional GPU is schematized.The transplantation of some specific intensive physical calculation to the GPU platform is absolutely feasible, and must result in satisfying performance. Certain peripheral computations of graphics can be carried out on the GPU platform, as well as some non-traditional rendering plot.The transplantation of similar or totally different intensive physical calculations to the GPU platform must be feasible as well , as long as its procedure runs smoothly under the constraints of GPU' s architecture. Hence the above result doubtlessly provides a theoretical demonstration and basis of approval for the feasibility to the transplantation of similar or totally different intensive physical calculations on the GPU platform.Directed by...