Скачать или смотреть The Algorithm That Makes Ray Tracing 10x Faster

Basic path tracing is incredibly slow and inefficient at finding light sources. Today, we're fixing the biggest flaw in our ray tracer by implementing Multiple Importance Sampling (MIS) to get clean and optimized renders!

#RayTracing #ComputerGraphics #CreativeCoding #PathTracing

Building on a standard path tracing foundation, we dive into the math behind Monte Carlo Integration, Probability Density Functions (PDFs), and Eric Veach's legendary Balance Heuristic.

Resources & Links:
Code: You can find the code here: github.com/ComputationallyBased/RayTracingMIS (based on Sebastian Lague's code from his 2nd part)
Sebastian Lague's incredible series:
Part 1:    • Coding Adventure: Ray Tracing  
Part 2:    • Coding Adventure: More Ray Tracing!  
Part 3:    • Coding Adventure: Ray-Tracing Glass and Ca...  
Eric Veach's Thesis: graphics.stanford.edu/papers/veach_thesis/thesis-bw.pdf
Nice paper about MIS: link.springer.com/chapter/10.1007/978-1-4842-7185-8_20
Music:
Background music: https://dos88.itch.io/dos-88-music-li... (released under the following license: https://creativecommons.org/licenses/...)
Also another track from: https://dos88.itch.io/dos-88musicpack
Piano Loop Em 160 BPM.wav by holizna -- https://freesound.org/s/629175/ -- License: Creative Commons 0

Timestamps:
0:00 - Introduction
0:47 - Basic Path Tracing
1:20 - Concept of Multiple Importance Sampling
2:09 - Direct Light Sampling
3:54 - G Factor
5:20 - PDF & Monte Carlo Integration
8:18 - MIS and the Balance Heuristic