blender-archive/intern/cycles/kernel/kernel_random.h

/*
 * Copyright 2011-2013 Blender Foundation
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#include "kernel/kernel_jitter.h"
#include "util/util_hash.h"

CCL_NAMESPACE_BEGIN

/* Pseudo random numbers, uncomment this for debugging correlations. Only run
 * this single threaded on a CPU for repeatable results. */
//#define __DEBUG_CORRELATION__

/* High Dimensional Sobol.
 *
 * Multidimensional sobol with generator matrices. Dimension 0 and 1 are equal
 * to classic Van der Corput and Sobol sequences. */

#ifdef __SOBOL__

/* Skip initial numbers that for some dimensions have clear patterns that
 * don't cover the entire sample space. Ideally we would have a better
 * progressive pattern that doesn't suffer from this problem, because even
 * with this offset some dimensions are quite poor.
 */
#  define SOBOL_SKIP 64

ccl_device uint sobol_dimension(KernelGlobals *kg, int index, int dimension)
{
  uint result = 0;
  uint i = index + SOBOL_SKIP;
  for (int j = 0, x; (x = find_first_set(i)); i >>= x) {
    j += x;
    result ^= kernel_tex_fetch(__sample_pattern_lut, 32 * dimension + j - 1);
  }
  return result;
}

#endif /* __SOBOL__ */

ccl_device_forceinline float path_rng_1D(
    KernelGlobals *kg, uint rng_hash, int sample, int num_samples, int dimension)
{
#ifdef __DEBUG_CORRELATION__
  return (float)drand48();
#endif
  if (kernel_data.integrator.sampling_pattern == SAMPLING_PATTERN_PMJ) {
    return pmj_sample_1D(kg, sample, rng_hash, dimension);
  }
#ifdef __CMJ__
#  ifdef __SOBOL__
  if (kernel_data.integrator.sampling_pattern == SAMPLING_PATTERN_CMJ)
#  endif
  {
    /* Correlated multi-jitter. */
    int p = rng_hash + dimension;
    return cmj_sample_1D(sample, num_samples, p);
  }
#endif

#ifdef __SOBOL__
  /* Sobol sequence value using direction vectors. */
  uint result = sobol_dimension(kg, sample, dimension);
  float r = (float)result * (1.0f / (float)0xFFFFFFFF);

  /* Cranly-Patterson rotation using rng seed */
  float shift;

  /* Hash rng with dimension to solve correlation issues.
   * See T38710, T50116.
   */
  uint tmp_rng = cmj_hash_simple(dimension, rng_hash);
  shift = tmp_rng * (1.0f / (float)0xFFFFFFFF);

  return r + shift - floorf(r + shift);
#endif
}

ccl_device_forceinline void path_rng_2D(KernelGlobals *kg,
                                        uint rng_hash,
                                        int sample,
                                        int num_samples,
                                        int dimension,
                                        float *fx,
                                        float *fy)
{
#ifdef __DEBUG_CORRELATION__
  *fx = (float)drand48();
  *fy = (float)drand48();
  return;
#endif
  if (kernel_data.integrator.sampling_pattern == SAMPLING_PATTERN_PMJ) {
    const float2 f = pmj_sample_2D(kg, sample, rng_hash, dimension);
    *fx = f.x;
    *fy = f.y;
    return;
  }
#ifdef __CMJ__
#  ifdef __SOBOL__
  if (kernel_data.integrator.sampling_pattern == SAMPLING_PATTERN_CMJ)
#  endif
  {
    /* Correlated multi-jitter. */
    int p = rng_hash + dimension;
    cmj_sample_2D(sample, num_samples, p, fx, fy);
    return;
  }
#endif

#ifdef __SOBOL__
  /* Sobol. */
  *fx = path_rng_1D(kg, rng_hash, sample, num_samples, dimension);
  *fy = path_rng_1D(kg, rng_hash, sample, num_samples, dimension + 1);
#endif
}

ccl_device_inline void path_rng_init(KernelGlobals *kg,
                                     int sample,
                                     int num_samples,
                                     uint *rng_hash,
                                     int x,
                                     int y,
                                     float *fx,
                                     float *fy)
{
  /* load state */
  *rng_hash = hash_uint2(x, y);
  *rng_hash ^= kernel_data.integrator.seed;

#ifdef __DEBUG_CORRELATION__
  srand48(*rng_hash + sample);
#endif

  if (sample == 0) {
    *fx = 0.5f;
    *fy = 0.5f;
  }
  else {
    path_rng_2D(kg, *rng_hash, sample, num_samples, PRNG_FILTER_U, fx, fy);
  }
}

/* Linear Congruential Generator */

ccl_device uint lcg_step_uint(uint *rng)
{
  /* implicit mod 2^32 */
  *rng = (1103515245 * (*rng) + 12345);
  return *rng;
}

ccl_device float lcg_step_float(uint *rng)
{
  /* implicit mod 2^32 */
  *rng = (1103515245 * (*rng) + 12345);
  return (float)*rng * (1.0f / (float)0xFFFFFFFF);
}

ccl_device uint lcg_init(uint seed)
{
  uint rng = seed;
  lcg_step_uint(&rng);
  return rng;
}

/* Path Tracing Utility Functions
 *
 * For each random number in each step of the path we must have a unique
 * dimension to avoid using the same sequence twice.
 *
 * For branches in the path we must be careful not to reuse the same number
 * in a sequence and offset accordingly.
 */

ccl_device_inline float path_state_rng_1D(KernelGlobals *kg,
                                          const ccl_addr_space PathState *state,
                                          int dimension)
{
  return path_rng_1D(
      kg, state->rng_hash, state->sample, state->num_samples, state->rng_offset + dimension);
}

ccl_device_inline void path_state_rng_2D(
    KernelGlobals *kg, const ccl_addr_space PathState *state, int dimension, float *fx, float *fy)
{
  path_rng_2D(kg,
              state->rng_hash,
              state->sample,
              state->num_samples,
              state->rng_offset + dimension,
              fx,
              fy);
}

ccl_device_inline float path_state_rng_1D_hash(KernelGlobals *kg,
                                               const ccl_addr_space PathState *state,
                                               uint hash)
{
  /* Use a hash instead of dimension, this is not great but avoids adding
   * more dimensions to each bounce which reduces quality of dimensions we
   * are already using. */
  return path_rng_1D(kg,
                     cmj_hash_simple(state->rng_hash, hash),
                     state->sample,
                     state->num_samples,
                     state->rng_offset);
}

ccl_device_inline float path_branched_rng_1D(KernelGlobals *kg,
                                             uint rng_hash,
                                             const ccl_addr_space PathState *state,
                                             int branch,
                                             int num_branches,
                                             int dimension)
{
  return path_rng_1D(kg,
                     rng_hash,
                     state->sample * num_branches + branch,
                     state->num_samples * num_branches,
                     state->rng_offset + dimension);
}

ccl_device_inline void path_branched_rng_2D(KernelGlobals *kg,
                                            uint rng_hash,
                                            const ccl_addr_space PathState *state,
                                            int branch,
                                            int num_branches,
                                            int dimension,
                                            float *fx,
                                            float *fy)
{
  path_rng_2D(kg,
              rng_hash,
              state->sample * num_branches + branch,
              state->num_samples * num_branches,
              state->rng_offset + dimension,
              fx,
              fy);
}

/* Utility functions to get light termination value,
 * since it might not be needed in many cases.
 */
ccl_device_inline float path_state_rng_light_termination(KernelGlobals *kg,
                                                         const ccl_addr_space PathState *state)
{
  if (kernel_data.integrator.light_inv_rr_threshold > 0.0f) {
    return path_state_rng_1D(kg, state, PRNG_LIGHT_TERMINATE);
  }
  return 0.0f;
}

ccl_device_inline float path_branched_rng_light_termination(KernelGlobals *kg,
                                                            uint rng_hash,
                                                            const ccl_addr_space PathState *state,
                                                            int branch,
                                                            int num_branches)
{
  if (kernel_data.integrator.light_inv_rr_threshold > 0.0f) {
    return path_branched_rng_1D(kg, rng_hash, state, branch, num_branches, PRNG_LIGHT_TERMINATE);
  }
  return 0.0f;
}

ccl_device_inline uint lcg_state_init(PathState *state, uint scramble)
{
  return lcg_init(state->rng_hash + state->rng_offset + state->sample * scramble);
}

ccl_device_inline uint lcg_state_init_addrspace(ccl_addr_space PathState *state, uint scramble)
{
  return lcg_init(state->rng_hash + state->rng_offset + state->sample * scramble);
}

ccl_device float lcg_step_float_addrspace(ccl_addr_space uint *rng)
{
  /* Implicit mod 2^32 */
  *rng = (1103515245 * (*rng) + 12345);
  return (float)*rng * (1.0f / (float)0xFFFFFFFF);
}

ccl_device_inline bool sample_is_even(int pattern, int sample)
{
  if (pattern == SAMPLING_PATTERN_PMJ) {
    /* See Section 10.2.1, "Progressive Multi-Jittered Sample Sequences", Christensen et al.
     * We can use this to get divide sample sequence into two classes for easier variance
     * estimation. */
#if defined(__GNUC__) && !defined(__KERNEL_GPU__)
    return __builtin_popcount(sample & 0xaaaaaaaa) & 1;
#elif defined(__NVCC__)
    return __popc(sample & 0xaaaaaaaa) & 1;
#elif defined(__KERNEL_OPENCL__)
    return popcount(sample & 0xaaaaaaaa) & 1;
#else
    /* TODO(Stefan): popcnt intrinsic for Windows with fallback for older CPUs. */
    int i = sample & 0xaaaaaaaa;
    i = i - ((i >> 1) & 0x55555555);
    i = (i & 0x33333333) + ((i >> 2) & 0x33333333);
    i = (((i + (i >> 4)) & 0xF0F0F0F) * 0x1010101) >> 24;
    return i & 1;
#endif
  }
  else {
    /* TODO(Stefan): Are there reliable ways of dividing CMJ and Sobol into two classes? */
    return sample & 0x1;
  }
}

CCL_NAMESPACE_END