Implementation and Performance Analysis of a Spectral Monte Carlo Path Tracer with Physical Optics Simulation and Adaptive Variance Reduction

A physically-based spectral Monte Carlo path tracer implementing wavelength-dependent light transport across the 380-780nm visible spectrum with 32 discrete samples. Features include GGX microfacet BRDF, Cauchy equation-based dielectric dispersion, complex Fresnel equations for conductors, Surface Area Heuristic BVH acceleration, Multiple Importance Sampling with power heuristic, heterogeneous volumetric media using 6-octave Perlin fractal Brownian motion, and adaptive sampling with statistical variance analysis. Achieves 1.81 megasamples/second performance on 512×512 resolution with 128 samples per pixel. Implementation includes OpenMP parallelization, PCG32 random generation, and feature-guided denoising.