Fix #114515: Cycles: Numerical precision issues in triangle light sampling #119224

Open
Lukas Stockner wants to merge 1 commits from LukasStockner/blender:triangle-sampling-precision into main

When changing the target branch, be careful to rebase the branch in your fork to match. See documentation.
1 changed files with 12 additions and 12 deletions

View File

@ -177,27 +177,27 @@ ccl_device_forceinline bool triangle_light_sample(KernelGlobals kg,
const float cos_a = dot(B, C);
const float cos_b = dot(A, C);
const float cos_c = dot(A, B);
const float sin_b_sin_c_2 = (1.0f - sqr(cos_b)) * (1.0f - sqr(cos_c));
const float mixed_product = fabsf(dot(A, cross(B, C)));
/* The area of the spherical triangle is equal to the subtended solid angle. */
const float solid_angle = 2.0f * fast_atan2f(mixed_product, (1.0f + cos_a + cos_b + cos_c));
/* Compute the angle at A. */
const float cos_alpha = dot(safe_normalize(cross(A, B)), safe_normalize(cross(A, C)));
const float sin_alpha = sin_from_cos(cos_alpha);
const float alpha = safe_acosf(cos_alpha);
/* Select a random sub-area of the spherical triangle and calculate the third vertex C_ of that
* new triangle. */
const float A_hat = rand.x * solid_angle;
float sin_A_hat, cos_A_hat;
fast_sincosf(A_hat, &sin_A_hat, &cos_A_hat);
/* These values lack a `sin_b * sin_c` factor, will divide when computing `temp`. */
const float cos_alpha = cos_a - cos_b * cos_c;
const float sin_alpha = mixed_product;
const float t = cos_A_hat * cos_alpha + sin_A_hat * sin_alpha;
const float temp = (cos_c - 1.0f) * t * cos_alpha / sin_b_sin_c_2;
const float q = (cos_A_hat - cos_c + temp) / (1.0f - cos_A_hat * cos_c + temp);
float sin_phi, cos_phi;
fast_sincosf(A_hat - alpha, &sin_phi, &cos_phi);
const float u = cos_phi - cos_alpha;
const float v = sin_phi + sin_alpha * cos_c;
const float num = (v * cos_phi - u * sin_phi) * cos_alpha - v;
const float den = (v * sin_phi + u * cos_phi) * sin_alpha;
const float q = (den == 0.0f) ? 1.0f : num / den;
const float3 U = safe_normalize(C - cos_b * A);
const float3 C_ = safe_normalize(q * A + sin_from_cos(q) * U);