FLinearColor ctor from FColor is mainly three table lookups, make it inline and move to the header.
Quantize, QuantizeFloor, QuantizeRound are similarly trivial, now inline.
Move Clamp01 to FColor as static method so it's namespaced; rename to Clamp01NansTo0 to explain why it's not just FMath::Clamp to [0,1]
SSE2-ified FLinearColor::ToFColorSRGB for x86 targets, bitwise exact same results as before for all input floats (tested exhaustively, test included)
Don't want to make ToFColorSRGB inline, it's a bit large for that, but make an array version that does multiple at once. Mainly try to avoid doing a cross-DLL call per pixel for large batches.
Switch ImageCore Linear->sRGB conv to use array version.
FLinearColor R,G,B,A members union'd with "deprecated" RGBA[] array, Component() updated to use it (matching recent fix for FVector)
ImageCore vector loads from FLinearColors updated to use &Component(0) which gives &RGBA[0], pointer to a 4-elem array
Add tests to UnrealMathTest.cpp to verify that sRGB<->Linear and UNORM<->Linear conversions behave as expected, plus the separate ColorConversionHeavy test that does the exhaustive equivalence check between the SSE2 ToFColorSRGB() and reference (warning: takes a while)
#rb danny.couture, martins.mozeiko
#preflight 6238c86f04769ab493469a3f
[CL 19456245 by fabian giesen in ue5-main branch]
For large enough floats x (e.g. 1e+7f), x*255 is too large to convert to int. On x86 this turns into INT_MIN which then gets clamped to 0, on ARM it gets saturated and turns into INT_MAX which gets clamped to 255.
Fix it so that over-large values result in 255 (which is the more natural behavior) everywhere and note the conversion rules used for both the sRGB and the linear channels while I'm here. Also update ImageCore optimized kernels to include the QuantizeRound fix.
Confirmed to produce bitwise identical results for textures on AncientGame (i.e. nop on typical content).
#jira UE-146294
#preflight 6234ffcc48746817f14025dc
#rb martins.mozeiko
[CL 19442181 by fabian giesen in ue5-main branch]
FImage is now the standard preferred type for a bag of pixels
FImageView can point at pixels without owning an allocation
ERawImageFormat (FImage) converts to ETextureSourceFormat
FImageUtils provides generic load/save and get/set from FImage
major cleanup in the ImageWrappers
new preferred API is through ImageWrapperModule Compress/Decompress
SetRaw/GetRaw functions cleaned up to not have undefined behavior on unexpected formats
ImageWrapper output added for HDR,BMP,TGA
RGBA32F format added and supported throughout import/export
EditorFactories import/export made more generic, most image types handled the same way using FImage now
Deprecate old TSF RGBA order pixel formats
Fix many crashes or bad handling of unusual pixel formats
Pixel access functions should be used instead of switches on pixel type
#preflight 6230ade7e65a7e65d68a187c
#rb julien.stjean,martins.mozeiko,dan.thompson,fabian.giesen
[CL 19397199 by charles bloom in ue5-main branch]
verified nop no change in behavior
prefix scopes with Texture.
#rb dan.thompson,fabian.giesen
#preflight none
[CL 19042123 by charles bloom in ue5-main branch]
add Insights markup
factor out job count comptutation and ensure it doesn't make tiny jobs
make VT tiles non-parallel because they are already parallel at the tile level
#rb fabian.giesen,arne.schober
#preflight none
[CL 18918813 by charles bloom in ue5-main branch]
This represents UE4/Main @17911760, Release-5.0 @17915875 and Dev-PerfTest @17914035
[CL 17918595 by aurel cordonnier in ue5-release-engine-test branch]
- This dramatically improves editor framerate under high load
Cleanup all the hacks that are not needed anymore
#rb Andriy.Tylychko
[CL 14676999 by danny couture in ue5-main branch]
- Feature can be activated in the Experimental section of the Editor Settings
- Replace Texture2D/TextureCube resources by placeholders until their PlatformData is ready
- Add a utility class allowing to encapsulate raw field pointers without breaking compatibility
- Protect PlatformData from unsafe access through encapsulation.
- Protect texture's resource from race conditions between game and render threads through encapsulation.
- This allows to get rid of FlushRenderingCommands and long game-thread stutters when Updating a texture's resource.
- UpdateResource was never safe to call without a FlushRenderingCommands and multiple call-site are doing exactly that, this will fix those cases.
- Those were probably undetected due to their low occurence rate under normal conditions but can easily be reproed during async texture compilation on 32 cores.
- Force wait on required texture compilations for MaterialBaking, ProxyMesh, Thumbnail generation for disk usage
- Wait on all textures compilation whenever a wait for all shaders compilation is requested for safety (i.e. screenshot)
- Compile UI and heightmap textures with higher priority to reduce visual artefacts
- Increase priority of texture that have been rendered to improve time-to-usefulness of the editor under low core count
- Async compilation is disabled for -game / non-editor mode as there is currently no support for async bulk data loading from external files
- Properly cancel async tasks when UTexture is garbage collected before the compilation is finished
- Show progress when explicitly waiting on compilation
- Changing the mip settings in the texture editor (or any settings requiring the running platform data to be recomputed) will now be processed asynchronously.
DEBUGGING
- Can be forcibly enabled/disabled through command-line via -asynctexturecompilation=[off, on, paused]
- Can pause texture compilation using Editor.AsyncTextureCompilation = 2 or -asynctexturecompilation=paused
- Can manually resume a specified amount of paused compilation using Editor.AsyncTextureCompilationResume [Num]
- Can forcibly wait on all compilation using Editor.AsyncTextureCompilationFlushAll
BENCHMARKS
- 3m15s to 1m20s when loading Apollo_Terrain with no textures in DDC (AMD TR 3970X)
- 6m45s to 1m11s when loading Apollo_Terrain with no textures in DDC (-corelimit=8)
- 3m10s to 1m54s when lauching PIE on Apollo_Terrain with no textures in DDC (AMD TR 3970X)
- 7m43s to 1m36s when lauching PIE on Apollo_Terrain with no textures in DDC (-corelimit=8)
- 0m57s to 0m42s when importing Attic_NVIDIA.usd with no textures in DDC (AMD TR 3970X)
- 2m14s to 0m35s when importing Attic_NVIDIA.usd with no textures in DDC (-corelimit=4)
TESTS
- Success on all material baking tests from EngineTests with -asynctexturecompilation=paused
- Runned with -corelimit=1 all the way to unlimited
- Cooking worked
- Opening the texture editor/material editor will force the compilation to finish like expected.
- Changing a setting in the texture editor will recompile async, even allowing to close the editor and continue doing other changes.
- Unpausing the compilation will update the texture thumbnails properly.
- Started with -asynctexturecompilation=paused, and then unpaused after a map loading, and then into a PIE session to stresstest UpdateResources.
- Tested both dx11/dx12
- Vulkan fails on Fortnite even with -asynctexturecompilation=off because of Landscape weigthmap, not this CL.
- Compiled and tested FortniteGame / UE4 / ShooterGame projects
#rb Uriel.Doyon, Francis.Hurteau
[CL 13694814 by danny couture in ue5-main branch]
Support for Half (R16F) textures
[FYI] Dmitriy.Dyomin, Brian.White, Will.Damon, Luke.Thatcher
#jira UE-91681
#ROBOMERGE-SOURCE: CL 12766470 via CL 12766473 via CL 12766477
#ROBOMERGE-BOT: RELEASE (Release-Engine-Staging -> Main) (v678-12743200)
[CL 12766484 by rolando caloca in Main branch]
#rnx
#rb none
#ROBOMERGE-OWNER: ryan.durand
#ROBOMERGE-AUTHOR: ryan.durand
#ROBOMERGE-SOURCE: CL 10869210 via CL 10869511 via CL 10869900
#ROBOMERGE-BOT: (v613-10869866)
[CL 10870549 by ryan durand in Main branch]
Needed for heightmaps
#ROBOMERGE-SOURCE: CL 8966460 via CL 8966466 via CL 8966471
#ROBOMERGE-BOT: (v436-8951088)
[CL 8966475 by jeremy moore in Main branch]
