CL 10373564 by danny.couture
Optimize Material Baking (Phase 1)
- Introduce a mecanism to override the vertex/index buffer allocator used for dynamic meshes
- Avoid GDynamicMesh non-ticked pools build-up by using our own vertex/index buffer pool during baking
- Reduce reallocation and incurred soft page faults by reusing a single set of vertex/index buffers big enough for the biggest mesh
- Preemptively detect if smearing would result in monochrome texture to avoid useless work
- Shrink smeared monochrome textures during the baking process for huge memory savings
- Move UV smearing in worker threads to avoid blocking the game thread
- Required shaders are now built asynchronously
- Add progress bar for material baking
- 28m23 [at] 150 GB RAM -> 2m14s [at] 45 GB RAM for 6 channels [at] 512x512 when baking materials on ProxyLOD for DATASET-0008a with DDC empty
#rb Jurre.deBaare, Sebastien.Lussier
CL 10516258 by danny.couture
Optimize Material Baking (Phase 2)
- Implement pipelining with staging buffers to avoid GPU stalls when reading from render targets
- Reuse the same prepared FMeshBatch instead of rebuilding it for each draw pass
- Prepare the RenderItem in advance on other threads to reduce work on the game thread
- Move the staging surface copy out of the render thread
- Small vertex and index buffers are not reused to avoid dependency locks when mapping them
- Fix bug in Canvas Flush_RenderThread found while running HLOD rebuild commandlet on Fortnite
- Delete old and unused MaterialBakingModule.h from public files
- 4m44s -> 59s for baking 6 channel [at] 1024x1024 when baking materials on ProxyLOD for DATASET-0008a with shaders already compiled
- Time spent in Material Baking when rebuilding all HLOD on Apollo_POI_Large_HLOD (Phase 1 + 2 combined)
- 10m18s -> 2m36s for a first rebuild all in editor with no shaders in DDC (cold)
- 1m23s -> 20s for a second rebuild all in editor (warm)
#rb Jeremy.Moore, Sebastien.Lussier
CL 11135986 by sebastien.lussier
Optimized mesh merging
* Added DeletePolygons() & DeleteTriangles methods to FMeshDescription which rely on TSets<> instead of performing costly TArray::AddUnique() calls()
* Parallelized UV generation and avoided duplicate processing of the same mesh+lod pairs
* Optimized FMeshDescriptionOperations::GenerateUniqueUVsForStaticMesh()
* Goes from 100s to 10s in my test case
#rb danny.couture, jeanfrancois.dube, richard.talbotwatkin
#ROBOMERGE-OWNER: sebastien.lussier
#ROBOMERGE-AUTHOR: sebastien.lussier
#ROBOMERGE-SOURCE: CL 11206337 via CL 11206341 via CL 11206346
#ROBOMERGE-BOT: (v643-11205221)
[CL 11206493 by sebastien lussier in Main branch]
#rnx
#rb none
#ROBOMERGE-SOURCE: CL 10869240 via CL 10869516 via CL 10869902
#ROBOMERGE-BOT: (v613-10869866)
[CL 10870584 by ryan durand in Main branch]
[FYI] bryce.lumpkin
#ROBOMERGE-SOURCE: CL 10453209 via CL 10453213 via CL 10453225
#ROBOMERGE-BOT: (v595-10452236)
[CL 10453236 by sebastien lussier in Main branch]
HLOD - Baked Emissive textures do not have sRGB checked
HLOD - Baked gray-scale/mask textures (Rough/Metal/Spec/Opacity) use sRGB=True
#ROBOMERGE-SOURCE: CL 10082362 via CL 10082363 via CL 10082364 via CL 10082365
#ROBOMERGE-BOT: (v574-10069753)
[CL 10082594 by sebastien lussier in Main branch]
Fixed mesh pivot point offset not being applied in this specific case
#rb patrick.enfedaque
#ROBOMERGE-SOURCE: CL 10056720 via CL 10057694 via CL 10057717 via CL 10057792
#ROBOMERGE-BOT: (v566-10053404)
[CL 10057904 by sebastien lussier in Main branch]
#rnx
#rb none
#ROBOMERGE-OWNER: rex.hill
#ROBOMERGE-AUTHOR: rex.hill
#ROBOMERGE-SOURCE: CL 9968254 via CL 9968365
#ROBOMERGE-BOT: (v560-9963197)
[CL 9969240 by rex hill in Main branch]
HLOD useLODAImposter seems to not copy over all UV channels
Mesh merging code was assuming the UV2 & UV3 channels were free to use to store position & scale of imposter instances.
This behavior seems to have been hacked in the engine for an internal project & is not needed anymore since it's now possible to use instanced static meshes with LODActors
Still kept old behavior around (CVar) in case someone would really need it
#rb patrick.enfedaque
[CL 9868123 by Sebastien Lussier in Main branch]
Attempted to resolve merge conflict for CL 9844036 Robomerge, it caused compilation errors
#rb Jake.Romigh
#fyi sebastien.lussier
[CL 9845666 by Laz Matech in Main branch]
#jira UE-82185
HLOD useLODAImposter seems to not copy over all UV channels
Mesh merging code was assuming the UV2 & UV3 channels were free to use to store position & scale of imposter instances.
This behavior seems to have been hacked in the engine for an internal project & is not needed anymore since it's now possible to use instanced static meshes with LODActors
Still kept old behavior around (CVar) in case someone would really need it
#patrick.enfedaque
[FYI] bryce.lumpkin
#ROBOMERGE-OWNER: laz.matech
#ROBOMERGE-AUTHOR: sebastien.lussier
#ROBOMERGE-SOURCE: CL 9843087 via CL 9843965 via CL 9844036
#ROBOMERGE-BOT: (v548-9842178)
[CL 9845167 by laz matech in Main branch]
[at]Ryan.Gerleve, [at]Brian.Bekich
#rb None
#ROBOMERGE-OWNER: jon.nabozny
#ROBOMERGE-AUTHOR: jon.nabozny
#ROBOMERGE-SOURCE: CL 9290108 via CL 9290249
#ROBOMERGE-BOT: (v452-9288972)
[CL 9290520 by jon nabozny in Main branch]
#jira UE-80604
#ROBOMERGE-SOURCE: CL 8934573 via CL 8934580 via CL 8934591
#ROBOMERGE-BOT: (v435-8929571)
[CL 8934596 by sebastien lussier in Main branch]
Cherrypicked CL8214461 from Dev-Enterprise
#jira UE-79203
#ROBOMERGE-SOURCE: CL 8224679 via CL 8230150
#ROBOMERGE-BOT: (v401-8057353)
[CL 8230887 by sebastien lussier in Main branch]
