This change is a preparation for moving the CPU profiler/visualization system from Atom into its own Gem by removing the dependency on local time tracking object AZ::RHI::CpuTimingStatistics
Full changes include:
- Removed all usage of AZ::RHI::CpuTimingStatistics
-- Replaced with pushing to AZ::Statistics::StatisticalProfilerProxy global instance
- Promoted VariableTimer from AZ::RHI to AZ::Debug
- Removed now unused CpuTimingStatistics.h
Signed-off-by: AMZN-ScottR 24445312+AMZN-ScottR@users.noreply.github.com
* Define fallback slots when shader is disabled
* Disable SSAO parent pass depending on the settings
* Move SSAO modulation pass into SsaoParent pass so it can be disabled with the rest of SSAO
* Enable SSAO by default
Co-authored-by: Tobias Alexander Franke <tobias.alexander.franke@huawei.com>
* Create helper function for getting threads per
group from a compute shader
Added GetComputeShaderNumThreads() functions to RPIUtils.
By default the function returns 1, 1, 1 in case of errors.
Updated existing code that was looking for 'numthreads' attribute data
with the new GetComputeShaderNumThreads() API.
Signed-off-by: garrieta <garrieta@amazon.com>
* Added a ModelReloader class that handles reloading the hierarchy of assets in the correct order. A ModelReloaderSystem that is used to make sure there is only one ModelReloader at a time for a given asset. Modified the Model, ModelLodAsset, and Buffer assets to not automatically reload, and instead handle reloads manually via AssetCatalog events. Modified the MeshLoader to kick off a reload via the ModelReloaderSystem whenever a model asset is added or changed.
Signed-off-by: amzn-tommy <waltont@amazon.com>
* Comment updates
Signed-off-by: amzn-tommy <waltont@amazon.com>
* Minor naming and comment updates based on PR feedback
Signed-off-by: amzn-tommy <waltont@amazon.com>
* Correcting previous commit. I flipped == to != when switching from count to find, when it should have stayed ==
Signed-off-by: amzn-tommy <waltont@amazon.com>
* Updating RenderCommon header
Signed-off-by: amzn-tommy <waltont@amazon.com>
* Removing unneeded headers
Signed-off-by: amzn-tommy <waltont@amazon.com>
* Reverting RayTracingFeatureProcessor change following Doug's guidance. This logic was tricky to get right initially, and leads to TDR when incorrect, so leaving as is.
Signed-off-by: amzn-tommy <waltont@amazon.com>
* Removing a tab
Signed-off-by: amzn-tommy <waltont@amazon.com>
* Added missing #include for AssetId
Signed-off-by: amzn-tommy <waltont@amazon.com>
* Adding missing RTTI header
Signed-off-by: amzn-tommy <waltont@amazon.com>
* Include ModelAsset definition intead of forward declaring it to avoid a static assert on Linux
Signed-off-by: amzn-tommy <waltont@amazon.com>
Merge pull request #4006 from aws-lumberyard-dev/Atom/santora/UseLibPng
OpenImageIO is a huge library and overkill for Atom's needs, which is to just save and load png files at this time. Here we remove all dependence on OpenImageIO and use libpng instead. I introduced a PngFile wrapper class to assist with loading and saving png files.
Note that there is a bit more work to do before merging these changes. I am still working on adding libpng to the 3rdParty package system. Currently these changes only reflect the addition of the Windows package. We'll make the other platform packages available before merging.
* Terrain feature processor improvements regarding material, mesh, and lod
- Now using a material with pbr lighting for terrain. Removed some of the old shader code that wasn't needed anymore
- Move heightmap image declaration to the material so it's not needed in the object SRG anymore
- Added prototype code (commented out) for smoothing the terrain with a b-spline weighting function
- Mesh data is all made with a proper mesh asset now.
- Added basic LOD support (no continuous LOD yet, but it does pop between lod levels)
- Moved RenderCommon to be accessible publicly. It contains stencil refs that should be public.
Signed-off-by: Ken Pruiksma <pruiksma@amazon.com>
* Fixing terrain's per object srg because of changes in the default per object srg.
Signed-off-by: Ken Pruiksma <pruiksma@amazon.com>
* PR Fixes
Signed-off-by: Ken Pruiksma <pruiksma@amazon.com>
* Removing unused static.
Signed-off-by: Ken Pruiksma <pruiksma@amazon.com>
* Remove the "TEST_SUPPORTED" traits.
Terrain unit tests should be usable on all platforms, so they shouldn't need a platform-specific trait to enable/disable.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Fix a few misc terrain bugs.
* Change Activate/Deactivate to happen immediately instead of deferring. There were too many order-of-operation bugs caused by trying to defer this.
* Added implementation for calculating normals.
* Fixed bug where GetHeightSynchronous wasn't stopping at the highest-priority layer.
* Added locks for SurfaceData bus to help ensure we lock our mutexes in the correct order and avoid deadlocks.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Add trivial TerrainSystem tests.
Tests construction, Activate(), Deactivate(), and destruction.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Unified Terrain system calls on single bus.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Added mock for TerrainDataNotificationBus listener.
Also added unit tests to verify the listener, and added in missing notification events.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Removed extra Sampler class.
Fixed up APIs to correctly pass Sampler and terrainExistsPtr around.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Switched MockTerrainSystem to be proper gmock.
This makes it for flexible to use and easier to reuse from other test environments.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Fix settings bug caused by bad order of operations that occurred when the methods moved to a different bus.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Eliminate extra EBus by simplifying area initialization.
Previously, there was a back-and-forth ebus signal used for the terrain system to find any terrain spawners that were created prior to the terrain system activation. Now it uses the more simple technique of just grabbing all the spawners that are currently hooked up to the spawner ebus.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Switch to NiceMock so that "uninteresting" mock calls get ignored.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Addressed PR feedback.
Filled in terrainExistsPtr at the end, and added it to GetNormal as well.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Fixed shader height calculation.
It was off by half a pixel, and it was interpolating, both of which were wrong.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Create initial LmbrCentral mocks that other Gems can use.
To help improve mock maintenance over time, this creates mocks in the same Gem as the systems being mocked, instead of the other Gems that need to use mocked systems during testing. This way, the mocks should more easily stay in sync with the interface that they mock out.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Null-guard any uses of Atom to make the class easier to unit test.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Add more mocked terrain services
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Moved Terrain mocks to publicly-available Mocks directory. Also added more unit tests.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Change debug code to use EXACT instead of BILINEAR height sampling, because it's specifically using the terrain grid.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Added support for the sampler filters.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Fix bad merge.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Add unit test to verify terrain layers define terrain regions.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Moved the AABB component mock into a private terrain header, since it's a specialized mock just for the terrain tests.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Moved sampler and normal calculations into TerrainSystem so that they can work across multiple adjacent areas.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Missed a couple of unit test changes.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Removed normal query as well, that needs a seprate unit test.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Removed unused variable that was only caught in Linux/Android builds.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Addressed PR feedback
Made the parameter names consistently listed in MockShapes.h.
Added comments to TerrainHeightGradientListComponent explaining why terrainExists is always true when a gradient exists.
Also fixed a bug where terrainExists should technically be *false* if no gradient exists.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Added more descriptive comments and names for ClampPosition.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Unit tests to verify height samplers work correctly.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Fix bug where wireframe didn't refresh when the setting changed.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Created TerrainWorldRenderer Component and moved all terrain rendering management code into it.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Fixed the icon references
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Removed expensive profile marker.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Addressed PR feedback
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Fixed AR error - constexpr doesn't need to be in a lambda capture.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
No longer need to assign material overrides to set/edit properties in UI or script
Material component will load and use the default material for a slot if no override is assigned
Signed-off-by: Guthrie Adams <guthadam@amazon.com>
- Removed all unnecessary includes to Atom CpuProfiler.h
- Added includes to AzCore Profiler.h where necessary
Signed-off-by: AMZN-ScottR 24445312+AMZN-ScottR@users.noreply.github.com
* Add HDR color grading pass and shaders.
* Remove color grading pass from light adaptation parent pass due to failing ASV tests. This will be added back later on.
* Editor HDR Color Grading Component implemented under the PostFx stack (#219)
* Add softmin/max for unbounded color grading properties.
* Revert channel mixer properties to vector3. Refactor mix to weight. Fix color grading pass to convert any color variable into vec4.
Signed-off-by: rbarrand <rbarrand@amazon.com>
Co-authored-by: Robin <rbarrand@amazon.com>
- Added new interface type AZ::Debug::Profiler to externally register profiler systems
- Modified the Atom CPU profiler to register as an AzCore profiler
-- This allows full engine markers to be visualized in the associated ImGui tool
- Converted all AZ_ATOM_PROFILE_* macros to use AZ_PROFILE_* macros instead
Signed-off-by: AMZN-ScottR 24445312+AMZN-ScottR@users.noreply.github.com
* [ATOM][RHI][Vulkan] Make sure to set super-variant for non-MSAA pipeline. Separate out reflection probe draw packet checks.
Signed-off-by: Peng <tonypeng@amazon.com>
* Missed from previous commit to make sure non-MSAA super-variant is used.
Signed-off-by: Peng <tonypeng@amazon.com>
* Minor comment edit.
Signed-off-by: Peng <tonypeng@amazon.com>
* Replacing the old gaussian blur with a much faster, better quality kawase blur
Signed-off-by: mrieggeramzn <mriegger@amazon.com>
* Removing atomtesting outdated msg
Signed-off-by: mrieggeramzn <mriegger@amazon.com>
* Some recommendations from Tommy
Signed-off-by: mrieggeramzn <mriegger@amazon.com>
* Adding early termination from previous gaussian filtering algorithm that kawase replaces
Signed-off-by: mrieggeramzn <mriegger@amazon.com>
* Removing pcf method
Signed-off-by: mrieggeramzn <mriegger@amazon.com>
* removing the old blur and adding in the new kawase blur into .cmake file
Signed-off-by: mrieggeramzn <mriegger@amazon.com>
* Removing the boundary search method. Bicubic is now the default and only PCF filtering method
* Removing padding (based upon feedback)
* Removing PCF method from py auto testing
Signed-off-by: mrieggeramzn <mriegger@amazon.com>
ATOM-16273 Compiling SceneSRG before updating it can cause a gpu crash
Changes include:
1. Removed Scene::SetShaderResourceGroupCallback() function and clean up code which use this function.
2. Moved SceneTimeSrg.azsli to RPI's DefaultSceneSrg folder and setup the constants in RPI::Scene
3. Add AZ::Event for Scene's update srg event which features and update scene srg at proper place
4. UpdateTransformServcie FP to use PrepareSceneSrg event handler.
5. Clean up shaders and srgs used in project templates.
Signed-off-by: Qing Tao <qingtao@amazon.com>
Many of these are just extra noise in the profile, but the one in Archive.cpp could also cause PIX to crash.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Add material property names to material assets, disable FBX dependency on materialtype files.
Signed-off-by: Robin <rbarrand@amazon.com>
* Add reflection for MaterialAssets. Update member variable comment.
Signed-off-by: Robin <rbarrand@amazon.com>
* Switch cvar to using bus value. Refine comments.
Signed-off-by: Robin <rbarrand@amazon.com>
* Refactor functions and refine comments.
Signed-off-by: Robin <rbarrand@amazon.com>
* Realign property values when material property names are populated.
Signed-off-by: Robin <rbarrand@amazon.com>
* Switch PostLoadInit check to on asset status ready. Add realign property values code to PostLoadInit as well.
Signed-off-by: Robin <rbarrand@amazon.com>
* Stash@{1} code.
Signed-off-by: Robin <rbarrand@amazon.com>
* Refactor realignment code into the right places.
Signed-off-by: Robin <rbarrand@amazon.com>
* Remove pragma optmize off.
Signed-off-by: Robin <rbarrand@amazon.com>
* More refactoring.
Signed-off-by: Robin <rbarrand@amazon.com>
* Refactor comments and remove code no longer needed.
Signed-off-by: Robin <rbarrand@amazon.com>
* Refactor comments and remove unused include.
Signed-off-by: Robin <rbarrand@amazon.com>
* Comment refactor, corrected some code.
Signed-off-by: Robin <rbarrand@amazon.com>
Co-authored-by: Robin <rbarrand@amazon.com>
* Release build fix for Windows
* Release build fix for Android
* Release build fix for Windows
* Release build fix for Android
* Release build fix for Linux
* Release build fix for Mac
* Release build fix for iOS
Signed-off-by: Steve Pham <spham@amazon.com>
* ATOM-16237 Using setting registry to RPI system settings
Changes include:
- Remove RHISystemDesriptor since the pre-registered draw list tag is not needed.
- Remove EitorContext which was for system component settings.
- Add atom_rpi.setreg file
- Add getting RPISystemDescriptor from setting registry.
Signed-off-by: qingtao <qingtao@amazon.com>
NOTE: The memory driller is still intact for now to avoid needing to
modify allocators, but the frame/cpu portions of driller and the
standalone executable are now gone.
Signed-off-by: Jeremy Ong <jcong@amazon.com>
Made Model Material Conversion Optional
Added a new registry setting that disables automatic conversion of materials from model files like FBX.
By default, processing of model files (like FBX) automatically convert the included materials to Atom materials, using StandardPBR. This adds a job dependency on StandardPBR.materialtype, which propagates to any related azsl files as well. Thus any change to azsl code will cause all model files in the project to rebuild.
Some game teams have no interest in using the auto-converted materials; they always use a Material Component to apply material overrides for every mesh. This new setting allows teams to disable material auto-conversion for the entire project, thus removing the job dependency on StandardPBR.materialtype. Instead, every mesh will be assigned the same default material. Any change to azsl code will cause that one default material to rebuild, but this will not trigger any models to rebuild.
Details:
- Added /O3DE/SceneAPI/MaterialConverter registry settings for configuring the scene material converter. It includes an enable flag, and a default material to use when conversion is disabled.
- Added SceneBuilderDependencyRequests::AddFingerprintInfo which allows SceneAPI components to modify the scene builder analysis fingerprint. We use this to reprocess scene files when the material converter settings change.
- Updated SceneAPI's material asset builder to skip the StandardPBR dependency when material conversion is disabled.
- Added some code to MaterialComponentController to handle an edge case that may when disabling material conversion on an existing project, and assigned materials disappear.
Testing:
- Changing the registry setting does trigger a rebuild of the fbx files.
- When material conversion is disabled, changing an azsl file does not cause fbx files to rebuild, but the shader still reloads as expected.
- Made a test level using multiple models with multiple meshes, made various adjustments to the material slots for each mesh, and tried switching the material conversion registry setting from true to false. (Details below)
- Merged this change to a customer's fork and tested on their existing content.
Details about my test level:
- Made a new test level AtomTest project
- Added two entities, both using multi-mat_mesh-groups_1m_cubes.fbx
- Added a material component to both entities
- Entity 1 material assignments
- Blue_Zaxis: left as-is
- Green_Yaxis: exported the material
- Red_Xaxis: exported the material, and changed the material instance color to pink
- StingrayPBS1: exported the material, scaled the UVs in the exported material source, and changed the material instance color to green.
- With_Texture: selected an existing brick material, changed the material instance color to red.
- Entity 2 material assignments
- Default Material: set to an existing brick material
- Blue_Zaxis: manually assigned built-in material that was converted from fbx
- Green_Yaxis: manually assigned built-in material that was converted from fbx, and changed the material instance color to orange
Fixes GPU crashes when releasing Indirect draw and query related dx12 objects
Fixes imgui profiler related to viewing buffer allocations
Signed-off-by: moudgils <moudgils@amazon.com>
* Fixed log2 shaper equations. Added bspline sampling for lut. Added options for custom log2 or linear lut with custom exposure ranges.
Signed-off-by: Ken Pruiksma <pruiksma@amazon.com>
* Added support for PQ shaper. Added shader option & cvar for lut sampling quality. Fixed issues in the blend lut shader that were causing considerable quality loss. No longer always changing to the log2 1000 nit shaper when blending luts - if the source luts all use the same shaper, keep using that shaper in the blended lut.
Signed-off-by: Ken Pruiksma <pruiksma@amazon.com>
* Fixed an integer -> float
Signed-off-by: Ken Pruiksma <pruiksma@amazon.com>
* Minor PR reveiw updates
Signed-off-by: Ken Pruiksma <pruiksma@amazon.com>
* Profiler: implement loading from saved capture
Adds functionality for finding a saved capture on disk and then
deserializing it using rapidjson's built-in buffered stream reader. This
does require use of raw file pointers since saved captures can be
hundreds of megabytes.
Actually showing the data in the visualizer is TODO.
* Profiler: use heap buffer over stack buffer
* Profiler: move deserialization logic to ImGuiCpuProfiler
Signed-off-by: Jacob Hilliard <jhlliar@amazon.com>
Scott Romero
Tobias Alexander Franke
Tommy Walton
Yuriy Toporovskyy
galibzon
Tommy Walton
Guthrie Adams
mrieggeramzn
dmcdiarmid-ly
dmcdiar
mrieggeramzn
Guthrie Adams
Artur K
santorac
Ken Pruiksma
Mike Balfour
hershey5045
AMZN-tpeng
Esteban Papp
Qing Tao
Steve Pham
Jeremy Ong
carlitosan
michabr
moudgils
Kyle Birnbaum
Jacob Hilliard
pappeste