GoaLitiuM/FlaxEngine
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Optimize thread group size for DDGI trace rays compute shader

Browse Source
This commit is contained in:
Wojciech Figat
2022-06-21 10:32:25 +02:00
parent 10adf1eba1
commit 4637017707
3 changed files with 19 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
Source/Shaders/GI/DDGI.shader
View File

@@ -18,7 +18,6 @@
// This must match C++
#define DDGI_TRACE_RAYS_LIMIT 256 // Limit of rays per-probe (runtime value can be smaller)
#define DDGI_TRACE_RAYS_GROUP_SIZE_X 32
#define DDGI_PROBE_UPDATE_BORDERS_GROUP_SIZE 8
#define DDGI_PROBE_CLASSIFY_GROUP_SIZE 32
@@ -151,7 +150,11 @@ TextureCube Skybox : register(t7);
// Compute shader for tracing rays for probes using Global SDF and Global Surface Atlas.
META_CS(true, FEATURE_LEVEL_SM5)
[numthreads(DDGI_TRACE_RAYS_GROUP_SIZE_X, 1, 1)]
META_PERMUTATION_1(DDGI_TRACE_RAYS_COUNT=96)
META_PERMUTATION_1(DDGI_TRACE_RAYS_COUNT=128)
META_PERMUTATION_1(DDGI_TRACE_RAYS_COUNT=192)
META_PERMUTATION_1(DDGI_TRACE_RAYS_COUNT=256)
[numthreads(DDGI_TRACE_RAYS_COUNT, 1, 1)]
void CS_TraceRays(uint3 DispatchThreadId : SV_DispatchThreadID)
{
uint rayIndex = DispatchThreadId.x;