diff options
author | Stefan Werner <stefan.werner@tangent-animation.com> | 2020-03-05 14:05:42 +0300 |
---|---|---|
committer | Stefan Werner <stefan.werner@tangent-animation.com> | 2020-03-05 14:21:38 +0300 |
commit | 51e898324de30c0985a80e5bc067358b5ccedbfc (patch) | |
tree | 5efddead1b7ca5655f1d6d2422b59e7da51fe271 /intern/cycles/kernel/kernel_passes.h | |
parent | 4ccbbd308060f0330472828b317c59e054c9ee7b (diff) |
Adaptive Sampling for Cycles.
This feature takes some inspiration from
"RenderMan: An Advanced Path Tracing Architecture for Movie Rendering" and
"A Hierarchical Automatic Stopping Condition for Monte Carlo Global Illumination"
The basic principle is as follows:
While samples are being added to a pixel, the adaptive sampler writes half
of the samples to a separate buffer. This gives it two separate estimates
of the same pixel, and by comparing their difference it estimates convergence.
Once convergence drops below a given threshold, the pixel is considered done.
When a pixel has not converged yet and needs more samples than the minimum,
its immediate neighbors are also set to take more samples. This is done in order
to more reliably detect sharp features such as caustics. A 3x3 box filter that
is run periodically over the tile buffer is used for that purpose.
After a tile has finished rendering, the values of all passes are scaled as if
they were rendered with the full number of samples. This way, any code operating
on these buffers, for example the denoiser, does not need to be changed for
per-pixel sample counts.
Reviewed By: brecht, #cycles
Differential Revision: https://developer.blender.org/D4686
Diffstat (limited to 'intern/cycles/kernel/kernel_passes.h')
-rw-r--r-- | intern/cycles/kernel/kernel_passes.h | 39 |
1 files changed, 38 insertions, 1 deletions
diff --git a/intern/cycles/kernel/kernel_passes.h b/intern/cycles/kernel/kernel_passes.h index 187e8340c82..33ec05c6048 100644 --- a/intern/cycles/kernel/kernel_passes.h +++ b/intern/cycles/kernel/kernel_passes.h @@ -29,7 +29,9 @@ ccl_device_inline void kernel_write_denoising_shadow(KernelGlobals *kg, if (kernel_data.film.pass_denoising_data == 0) return; - buffer += (sample & 1) ? DENOISING_PASS_SHADOW_B : DENOISING_PASS_SHADOW_A; + buffer += sample_is_even(kernel_data.integrator.sampling_pattern, sample) ? + DENOISING_PASS_SHADOW_B : + DENOISING_PASS_SHADOW_A; path_total = ensure_finite(path_total); path_total_shaded = ensure_finite(path_total_shaded); @@ -386,6 +388,41 @@ ccl_device_inline void kernel_write_result(KernelGlobals *kg, #ifdef __KERNEL_DEBUG__ kernel_write_debug_passes(kg, buffer, L); #endif + + /* Adaptive Sampling. Fill the additional buffer with the odd samples and calculate our stopping + criteria. This is the heuristic from "A hierarchical automatic stopping condition for Monte + Carlo global illumination" except that here it is applied per pixel and not in hierarchical + tiles. */ + if (kernel_data.film.pass_adaptive_aux_buffer && + kernel_data.integrator.adaptive_threshold > 0.0f) { + if (sample_is_even(kernel_data.integrator.sampling_pattern, sample)) { + kernel_write_pass_float4(buffer + kernel_data.film.pass_adaptive_aux_buffer, + make_float4(L_sum.x * 2.0f, L_sum.y * 2.0f, L_sum.z * 2.0f, 0.0f)); + } +#ifdef __KERNEL_CPU__ + if (sample > kernel_data.integrator.adaptive_min_samples && + (sample & (ADAPTIVE_SAMPLE_STEP - 1)) == (ADAPTIVE_SAMPLE_STEP - 1)) { + kernel_do_adaptive_stopping(kg, buffer, sample); + } +#endif + } + + /* Write the sample count as negative numbers initially to mark the samples as in progress. + * Once the tile has finished rendering, the sign gets flipped and all the pixel values + * are scaled as if they were taken at a uniform sample count. */ + if (kernel_data.film.pass_sample_count) { + /* Make sure it's a negative number. In progressive refine mode, this bit gets flipped between + * passes. */ +#ifdef __ATOMIC_PASS_WRITE__ + atomic_fetch_and_or_uint32((ccl_global uint *)(buffer + kernel_data.film.pass_sample_count), + 0x80000000); +#else + if (buffer[kernel_data.film.pass_sample_count] > 0) { + buffer[kernel_data.film.pass_sample_count] *= -1.0f; + } +#endif + kernel_write_pass_float(buffer + kernel_data.film.pass_sample_count, -1.0f); + } } CCL_NAMESPACE_END |