A Hardware Acceleration Engine for Ray Tracing

摘要

Ray tracing, as a global illumination based rendering method, is famous for generating high-fidelity images of 3D scene and is widely used in computer graphics. Implementations of ray tracing based on Central Processing Unit (CPU), Graphic Processing Unit (GPU) and Field-programmable Gate Array (FPGA) have been proposed in the past few decades. In this paper, we present a hardware acceleration engine for ray tracing using Application Specific Integrated Circuit (ASIC) technology. Working with a high clock rate of 500MHz, it achieves high performance of up to 43.5 million rays per second. In addition, it offers all the usual rendering effects, including exact shadows, correct reflections and refractions, which makes ray tracing so popular. We evaluate the acceleration engine and its performance using a set of scenes with cycle-accurate simulations.