Path tracing

Path tracing is a rendering algorithm in computer graphics that simulates how light interacts with objects, voxels, and participating media to generate realistic (physically plausible) images.

This ray tracing technique uses the Monte Carlo method to accurately model global illumination, simulate different surface characteristics, and capture a wide range of effects observable in a camera system, such as optical properties of lenses (e.g., depth of field and bokeh) or the impact of shutter speed (e.g., motion blur and exposure). By incorporating physically accurate materials and light transport models, it can produce photorealistic results but requires significant computational power. Performance is often constrained by VRAM/RAM capacity and memory bandwidth, especially in complex scenes, necessitating denoising techniques for practical use. Additionally, the Garbage In, Garbage Out (GIGO) principle applies - inaccurate scene data, poor geometry, low-quality materials, or incorrect rendering settings can negatively impact the final output, regardless of rendering precision.

Due to its accuracy, unbiased nature, and algorithmic simplicity, path tracing is commonly used to generate reference images when testing the quality of other rendering algorithms. Fundamentally, the algorithm works by integrating the light arriving at a point on an object’s surface, where this illuminance is then modified by a surface reflectance function (BRDF) to determine how much light contributes to the final image, as seen by the camera. This integration procedure is repeated for every pixel in the output image, ensuring detailed evaluation of each one. The number of samples per pixel (spp) determines the level of detail and quality of the final render, with more samples generally improving image clarity. Rendering performance is often measured in mega samples per second (Ms/sec), which reflects how many millions of samples can be processed per second, directly impacting rendering speed. Several variants of path tracing, such as bidirectional path tracing and Metropolis light transport, have been developed to improve efficiency in various types of scenes, reducing noise and speeding up convergence.