An fpga based triangle rendering module design for computer graphics

In computer graphics, rendering is the process of coloring the animation objects whose sufaces are mathematically defined with poloygons. As the number of objects on a scene and the number of poligons used to define these objects increase, the CPU time requirement of the rendering process grows exponentially and general purpose computers become insufficient. Several approaches such as using exhanced graphics cards, specially degined computers, or parellel computers were developed. But these approaches are not cost effective and sometimes do not yield the desired performance. In this research work, as alternatives to the above mentioned approaches, two separate hadware modules were designed. The modules were designed to be used with Field Programmable Gate Array (FPGA) and can process fixed-point data. The first module uses the half-space property of the line equation. This module is able to fill a given triangle with a solid color. The second module uses the line equation itself. This module is able to fill inside a given triangle using the given intensity values at the corners of the triangle. Each pixel?s color value is calculated according to pixel?s distance to the corners. The modules were tested using randomly generated test triangles. Modules? rendering times were compared to redering softwares? CPU times. The rendering softwares were coded in C++ and were running on general purpose computers. The results showed that the first module renders the test triangles up to 31 times faster and the second module renders the test triangles up to 17 times faster.