archive/blender-archive
Archived
1
1
Fork 0
Code Pull Requests Activity

Cycles: more accurately count main paths for adding work tiles

Browse Source 
Easy now thanks to the main and shadow path decoupling. Doesn't help
in an benchmark scene except Spring, where it reduces render time by
maybe 2-3%.

Ref T87836
This commit is contained in:
Brecht Van Lommel
2021-10-20 14:47:04 +02:00
parent 52c5300214
commit 0c52eed863
1 changed files with 4 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
intern/cycles/integrator/path_trace_work_gpu.cpp
View File

@@ -731,7 +731,6 @@ void PathTraceWorkGPU::enqueue_work_tiles(DeviceKernel kernel,
int PathTraceWorkGPU::num_active_main_paths_paths()
{
/* TODO: this is wrong, does not account for duplicates with shadow! */
IntegratorQueueCounter *queue_counter = integrator_queue_counter_.data();
int num_paths = 0;
@@ -739,7 +738,10 @@ int PathTraceWorkGPU::num_active_main_paths_paths()
DCHECK_GE(queue_counter->num_queued[i], 0)
<< "Invalid number of queued states for kernel "
<< device_kernel_as_string(static_cast<DeviceKernel>(i));
num_paths += queue_counter->num_queued[i];
if (!kernel_is_shadow_path((DeviceKernel)i)) {
num_paths += queue_counter->num_queued[i];
}
}
return num_paths;