Mai Lavelle 6238214159 Cycles: Faster split branched path tracing by sharing samples with inactive threads ... Unlike regular path tracing, branched path tracing is usually used with lower sample counts, at least for primary rays. This means that are less samples for the GPU to work on in parallel and rendering is slower. As there is less work overall there is also more inactive threads during rendering with BPT. This patch makes use of those inactive rays to render branched samples in parallel with other samples. Each thread that is preparing for a branched sample will attempt to find an inactive thread and if one is found the state for the sample is copied to that thread. Potentially, if there are enough inactive threads, 100s of branched samples could be generated from the same originating thread and ran in parallel giving large speed ups. Gives 70% faster render for pavillion midday scene. 20-60% faster on BMW with car paint replaced with SSS/volumes.		2017-06-10 04:08:49 -04:00
..
kernel_branched.h	Cycles: Faster split branched path tracing by sharing samples with inactive threads	2017-06-10 04:08:49 -04:00
kernel_buffer_update.h	Fix T51537: Light passes are summed twice for split kernel since denoise commit	2017-05-19 12:14:03 +02:00
kernel_data_init.h	Cycles: Fix building with native only option	2017-05-16 16:05:04 -04:00
kernel_direct_lighting.h	Cycles: Branched path tracing for the split kernel	2017-05-02 14:26:46 -04:00
kernel_do_volume.h	Cycles: Faster split branched path tracing by sharing samples with inactive threads	2017-06-10 04:08:49 -04:00
kernel_enqueue_inactive.h	Cycles: Add kernel to enqueue inactive rays	2017-06-10 03:51:18 -04:00
kernel_holdout_emission_blurring_pathtermination_ao.h	Cycles: Implement denoising option for reducing noise in the rendered image	2017-05-07 14:40:58 +02:00
kernel_indirect_background.h	Cycles: fix AO approximation for split kernel	2017-05-11 11:58:25 +02:00
kernel_indirect_subsurface.h	Cycles: Branched path tracing for the split kernel	2017-05-02 14:26:46 -04:00
kernel_lamp_emission.h	Cycles: Avoid some ccl_local in various kernels	2017-03-16 11:27:17 +01:00
kernel_next_iteration_setup.h	Cycles: Faster split branched path tracing by sharing samples with inactive threads	2017-06-10 04:08:49 -04:00
kernel_path_init.h	Cycles: Remove ccl_addr_space from RNG passed to functions	2017-03-27 10:46:28 +02:00
kernel_queue_enqueue.h	Cycles: Define ccl_local variables in kernel functions	2017-03-16 11:27:17 +01:00
kernel_scene_intersect.h	Cycles: Faster split branched path tracing by sharing samples with inactive threads	2017-06-10 04:08:49 -04:00
kernel_shader_eval.h	Cycles: Split kernel - sort shaders	2017-05-03 15:30:45 +02:00
kernel_shader_setup.h	Cycles: Split kernel - sort shaders	2017-05-03 15:30:45 +02:00
kernel_shader_sort.h	Cycles: Split kernel - sort shaders	2017-05-03 15:30:45 +02:00
kernel_shadow_blocked_ao.h	Cycles: Branched path tracing for the split kernel	2017-05-02 14:26:46 -04:00
kernel_shadow_blocked_dl.h	Cycles: Faster split branched path tracing by sharing samples with inactive threads	2017-06-10 04:08:49 -04:00
kernel_split_common.h	Cycles: Faster split branched path tracing by sharing samples with inactive threads	2017-06-10 04:08:49 -04:00
kernel_split_data_types.h	Cycles: Faster split branched path tracing by sharing samples with inactive threads	2017-06-10 04:08:49 -04:00
kernel_split_data.h	Cycles: Split kernel SSS & Volume data definitions cleanup	2017-05-05 13:42:26 +02:00
kernel_subsurface_scatter.h	Cycles: Faster split branched path tracing by sharing samples with inactive threads	2017-06-10 04:08:49 -04:00