基于GPU的实时水面场景并行渲染算法

Gpu-Based Real-Time Parallel Algorithm for Rendering Water-Surface Scene

水面的模拟历来都是一个难题。近几年,随着图形处理器(GPU)的迅速发展,它的功能不再局限于图形处理,而是进入到了通用计算领域,所以可以利用GPU的并行计算优势来加速水面的模拟过程。本文研究的主要内容是海浪的实时模拟绘制。海浪渲染算法使用了投影网格技术,使用CUDA并行计算方法对投影网格顶点坐标的计算进行加速,比较使用CPU进行大数据量计算和使用GPU进行大数据量计算的速度差异。最终,改进后的程序渲染速度大幅提高,使得程序可以在生成更逼真画面的同时保持令人满意的帧率。

The simulation of water surface has always been a problem. In recent years, with the rapiddevelopment of graphic processing unit （GPU）, its function is no longer confined to graphic processing,but enters into the field of general-purpose computing, so we can take advantage of GPU parallel computing to accelerate the process of water surface simulation. The main content of this paper is the real-time simulation of sea surface. The sea surface rendering algorithm uses the projected grid technology.The calculation of vertex coordinates of projected grid using the CUDA parallel computing technologywere accelerated, compared the speed of calculating a large amount of data using CPU to calculating alarge amount of data using GPU. Finally, the program improved rendering speed significantly, so thatthe program can generate a more realistic picture, while maintaining a satisfactory frame rate.