Final Submission

README.md

@@ -1,11 +1,73 @@
 **University of Pennsylvania, CIS 565: GPU Programming and Architecture,
 Project 5 - DirectX Procedural Raytracing**
 
-* (TODO) YOUR NAME HERE
-  * (TODO) [LinkedIn](), [personal website](), [twitter](), etc.
-* Tested on: (TODO) Windows 22, i7-2222 @ 2.22GHz 22GB, GTX 222 222MB (Moore 2222 Lab)
+* Eric Micallef
+  * https://www.linkedin.com/in/eric-micallef-99291714b/
+
+* Tested on: Windows 10, i5, Nvidia GTX1660 (Personal)
 
-### (TODO: Your README)
+- [Overview](#Overview)
+- [Performance Analysis](#Performance-Analysis)
+- [Debilitating Bugs/Bloopers](#Debilitating-Bugs-AKA-Bloopers-In-CIS565-World)
+- [References](#References)
+
+# Overview
+
+Special thanks to Ziad. He is probably sick of seeing me in his OH and on Piazza. Without him I would have thrown my desktop out the window.
+
+In this project we used DirectX12 to implement a basic path tracer. In this homework we were asked to implement the CPU side of DirectX12  as well as the GPU logic of raytracing in DirectX12. Below you will see some animations of the ray tracer in motion.
+
+![](images/sceenshot.PNG)
+
+![](images/done.gif)
+
+![](images/lighting.gif)
+
+![](images/pivot.gif)
+
+# Performance Analysis
+
+Below are some data points after manipulating the recursion depth. As you can see, the further we follow the ray slower our application becomes. I also did not see any noticeable rendering difference with these increasing recursion depths.
+
+![](images/FPS.png)
+
+# Debilitating Bugs AKA Bloopers In CIS565 World
+
+![](images/looks_close_enough.PNG)
+
+After watching my program crash 10000000000000000000 times because my cpu side was wrong, finally I got something up... but then I had to sadly go back and fix alot of stuff ... 
+
+![](images/fuzzy.PNG)
+
+turns out there was alot wrong here ... alot ... But I do like the darker purple and darker yellow colors!
+
+![](images/wtf.PNG)
+
+my lerping for fading was off.. way offf.... Lerping is not to be confused with the ever popular LARPing.
+
+![](images/weird_balls.gif)
+
+If you look you can see that the two purple balls move in a very odd manner. was kind of neat.
+
+problem was my logic for keeping track of closet hit was a bit off. I was keeping track of tmax, a sensical person would have looked at the function return and saw it was evaluating thit... I became sensical after about 2 hours of debugging.
+
+![](images/floatvsfloat3.gif)
+
+This is what happens when you accidentally return a float instead of float3 ): 
+
+# References
+
+Ziad ... without him I would have thrown my computer out the window
+
+http://viclw17.github.io/2018/11/29/raymarching-algorithm/
+
+https://www.freecodecamp.org/news/understanding-linear-interpolation-in-ui-animations-74701eb9957c/	
+
+https://www.scratchapixel.com/lessons/3d-basic-rendering/minimal-ray-tracer-rendering-simple-shapes/ray-sphere-intersection
+
+https://www.scratchapixel.com/lessons/3d-basic-rendering/minimal-ray-tracer-rendering-simple-shapes/ray-box-intersection
+
+https://www.scratchapixel.com/lessons/3d-basic-rendering/phong-shader-BRDF
+
+https://www.scratchapixel.com/lessons/advanced-rendering/rendering-distance-fields/blobbies
 
-Include screenshots, analysis, etc. (Remember, this is public, so don't put
-anything here that you don't want to share with the world.)

src/D3D12RaytracingProceduralGeometry/AnalyticPrimitives.hlsli

@@ -103,7 +103,7 @@ float3 CalculateNormalForARaySphereHit(in Ray ray, in float thit, float3 center)
 }
 
 // Test if a ray with RayFlags and segment <RayTMin(), RayTCurrent()> intersects a hollow sphere.
-bool RaySphereIntersectionTest(in Ray ray, out float thit, out float tmax, in ProceduralPrimitiveAttributes attr, in float3 center = float3(0, 0, 0), in float radius = 1)
+bool RaySphereIntersectionTest(in Ray ray, out float thit, out float tmax, out ProceduralPrimitiveAttributes attr, in float3 center = float3(0, 0, 0), in float radius = 1)
 {
     float t0, t1; // solutions for t if the ray intersects 
 
@@ -166,18 +166,39 @@ bool RaySolidSphereIntersectionTest(in Ray ray, out float thit, out float tmax,
 bool RayMultipleSpheresIntersectionTest(in Ray ray, out float thit, out ProceduralPrimitiveAttributes attr)
 {
 	// Define the spheres in local space (within the aabb)
-	float3 center = float3(-0.2, 0, -0.2);
-	float radius = 0.7f;
+	float3 center[3];
+	float radius[3];
+	bool hit = false;
+
+	center[0] = float3(-0.2, 0, -0.2);
+	radius[0] = 0.5f;
+	center[1] = float3(-.4, -.5, -0.9);
+	radius[1] = .3f;
+	center[2] = float3(0.7, 0.8, 0.2);
+	radius[2] = .2f;
 
-	thit = RayTCurrent();
 
-	float tmax;
-	if (RaySphereIntersectionTest(ray, thit, tmax, attr, center, radius))
+	thit = RayTCurrent();
+	float temp_max =0;
+	float tmax = 0;
+	float temp_hit = thit;
+	ProceduralPrimitiveAttributes temp_attr;
+	for (int i = 0; i < 3; i++) 
 	{
-		return true;
+		if (RaySphereIntersectionTest(ray, temp_hit, temp_max, temp_attr, center[i], radius[i]))
+		{
+			//not sure why thit and not max but oh well tmax causes for a mind bending bug
+			if (temp_hit < thit)
+			{
+				tmax = temp_max;
+				attr = temp_attr;
+				thit = temp_hit;
+				hit = true;
+			}
+		}
 	}
 
-	return false;
+	return hit;
 }
 
 #endif // ANALYTICPRIMITIVES_H

src/D3D12RaytracingProceduralGeometry/DXR-AccelerationStructure.cpp

@@ -31,7 +31,28 @@ void DXProceduralProject::BuildGeometryDescsForBottomLevelAS(array<vector<D3D12_
 		// The number of elements of a D3D12 resource can be accessed from GetDesc().Width (e.g m_indexBuffer.resource->GetDesc().Width)
 		auto& geometryDesc = geometryDescs[BottomLevelASType::Triangle][0];
 		geometryDesc = {};
+
+		//
+		geometryDesc.Flags = geometryFlags;
+		geometryDesc.Type = D3D12_RAYTRACING_GEOMETRY_TYPE_TRIANGLES;
+
+		//
+		geometryDesc.Triangles.IndexBuffer = m_indexBuffer.resource->GetGPUVirtualAddress();
 
+		// why is width count? because it says so above i  guess?
+		geometryDesc.Triangles.IndexCount = m_indexBuffer.resource->GetDesc().Width / sizeof(Index);
+		geometryDesc.Triangles.IndexFormat = DXGI_FORMAT_R16_UINT;
+
+		// think stride is ok based on below
+		//https://docs.microsoft.com/en-us/windows/win32/api/d3d12/ns-d3d12-d3d12_raytracing_geometry_triangles_desc
+		//  doc says zero again?
+		UINT64 stride = sizeof(Vertex);
+		geometryDesc.Triangles.VertexBuffer.StartAddress = m_vertexBuffer.resource->GetGPUVirtualAddress();
+		geometryDesc.Triangles.VertexBuffer.StrideInBytes = stride;
+
+		//how do I get count?
+		geometryDesc.Triangles.VertexCount = m_vertexBuffer.resource->GetDesc().Width/ sizeof(Vertex);
+		geometryDesc.Triangles.VertexFormat = DXGI_FORMAT_R32G32B32_FLOAT;
 	}
 
 	{
@@ -49,6 +70,15 @@ void DXProceduralProject::BuildGeometryDescsForBottomLevelAS(array<vector<D3D12_
 		// Remember to use m_aabbBuffer to get the AABB geometry data you previously filled in.
 		// Note: Having separate geometries allows of separate shader record binding per geometry.
 		//		 In this project, this lets us specify custom hit groups per AABB geometry.
+		//D3D12_RAYTRACING_GEOMETRY_AABBS_DESC aabb_desc = {}; // just needs start and stride?
+		// loop through the primitive and analytic types getting the start and stride
+		for (UINT count = 0; count < IntersectionShaderType::TotalPrimitiveCount; count++)
+		{
+			auto& geometryDesc = geometryDescs[BottomLevelASType::AABB][count];
+			// https://docs.microsoft.com/en-us/windows/win32/api/d3d12/ns-d3d12-d3d12_raytracing_geometry_aabbs_desc
+			UINT stride = sizeof(D3D12_RAYTRACING_AABB)*count;
+			geometryDesc.AABBs.AABBs.StartAddress = m_aabbBuffer.resource->GetGPUVirtualAddress() + stride;
+		}
 
 	}
 }
@@ -61,14 +91,17 @@ AccelerationStructureBuffers DXProceduralProject::BuildBottomLevelAS(const vecto
 	ComPtr<ID3D12Resource> scratch; // temporary AS data
 	ComPtr<ID3D12Resource> bottomLevelAS; // actual bottom-level AS resource
 
-	// This is what we need to fill in.
 	D3D12_BUILD_RAYTRACING_ACCELERATION_STRUCTURE_DESC bottomLevelBuildDesc = {};
 
 	// The bottom-level inputs are the geometries you build in BuildGeometryDescsForBottomLevelAS()
 	// Again, these tell the AS where the actual geometry data is and how it is laid out.
 	// TODO-2.6: fill the bottom-level inputs. Consider using D3D12_ELEMENTS_LAYOUT_ARRAY as the DescsLayout.
 	D3D12_BUILD_RAYTRACING_ACCELERATION_STRUCTURE_INPUTS &bottomLevelInputs = bottomLevelBuildDesc.Inputs;
-
+	bottomLevelInputs.Type = D3D12_RAYTRACING_ACCELERATION_STRUCTURE_TYPE_BOTTOM_LEVEL;
+	bottomLevelInputs.Flags = buildFlags;
+	bottomLevelInputs.DescsLayout = D3D12_ELEMENTS_LAYOUT_ARRAY;
+	bottomLevelInputs.NumDescs = static_cast<UINT>( geometryDescs.size() );
+	bottomLevelInputs.pGeometryDescs = geometryDescs.data();
 
 	// Query the driver for resource requirements to build an acceleration structure. We've done this for you.
 	D3D12_RAYTRACING_ACCELERATION_STRUCTURE_PREBUILD_INFO bottomLevelPrebuildInfo = {};
@@ -108,7 +141,9 @@ AccelerationStructureBuffers DXProceduralProject::BuildBottomLevelAS(const vecto
 	// TODO-2.6: Now that you have the scratch and actual bottom-level AS desc, pass their GPU addresses to the bottomLevelBuildDesc.
 	// Consider reading about D3D12_BUILD_RAYTRACING_ACCELERATION_STRUCTURE_DESC.
 	// This should be as easy as passing the GPU addresses to the struct using GetGPUVirtualAddress() calls.
-
+	// there are two virtual address fields in the structure so we set them here
+	bottomLevelBuildDesc.ScratchAccelerationStructureData = scratch->GetGPUVirtualAddress();
+	bottomLevelBuildDesc.DestAccelerationStructureData = bottomLevelAS->GetGPUVirtualAddress();
 
 	// Fill up the command list with a command that tells the GPU how to build the bottom-level AS.
 	if (m_raytracingAPI == RaytracingAPI::FallbackLayer)
@@ -127,7 +162,17 @@ AccelerationStructureBuffers DXProceduralProject::BuildBottomLevelAS(const vecto
 	// the AccelerationStructureBuffers struct so the top-level AS can use it! 
 	// Don't forget that this is the return value.
 	// Consider looking into the AccelerationStructureBuffers struct in DXR-Structs.h
-	return AccelerationStructureBuffers{};
+	//ComPtr<ID3D12Resource> scratch;						// temp data needed by the GPU to build the AS
+	//ComPtr<ID3D12Resource> accelerationStructure;		// actual AS data
+	//ComPtr<ID3D12Resource> instanceDesc;				// used only for top-level AS
+	//UINT64                 ResultDataMaxSizeInBytes;	// this can be acquired from the prebuild info
+	AccelerationStructureBuffers return_buff = {};
+	return_buff.scratch = scratch;
+	return_buff.accelerationStructure = bottomLevelAS;
+	return_buff.instanceDesc = nullptr; // used only for top level 
+	return_buff.ResultDataMaxSizeInBytes = bottomLevelPrebuildInfo.ResultDataMaxSizeInBytes;
+
+	return return_buff;
 }
 
 // TODO-2.6: Build the instance descriptor for each bottom-level AS you built before.
@@ -179,7 +224,19 @@ void DXProceduralProject::BuildBottomLevelASInstanceDescs(BLASPtrType *bottomLev
 	//		Where do you think procedural shader records would start then? Hint: right after.
 	// * Make each instance hover above the ground by ~ half its width
 	{
+		auto& instanceDesc = instanceDescs[BottomLevelASType::AABB];
+		instanceDesc = {};
+		instanceDesc.InstanceMask = 1; // making the mask same as triangle for now
+		// how the fuck is this intuitive? I try numbers until stuff appeared on my screen .. great use of my time and money
+		instanceDesc.InstanceContributionToHitGroupIndex = BottomLevelASType::AABB * RayType::Count;// maybe this?
+		instanceDesc.AccelerationStructure = bottomLevelASaddresses[BottomLevelASType::AABB];
+
+		// TODO do I need anything else here?
+		// also not intuitive
+		XMMATRIX mTranslation = XMMatrixTranslationFromVector(XMLoadFloat3(&XMFLOAT3(0, c_aabbWidth / 2, 0)));
 
+		// Store the transform in the instanceDesc.
+		XMStoreFloat3x4(reinterpret_cast<XMFLOAT3X4*>(instanceDesc.Transform), mTranslation);
 	}
 
 	// Upload all these instances to the GPU, and make sure the resouce is set to instanceDescsResource.
@@ -202,7 +259,10 @@ AccelerationStructureBuffers DXProceduralProject::BuildTopLevelAS(AccelerationSt
 	// TODO-2.6: fill in the topLevelInputs, read about D3D12_BUILD_RAYTRACING_ACCELERATION_STRUCTURE_INPUTS.
 	// Consider using D3D12_ELEMENTS_LAYOUT_ARRAY as a DescsLayout since we are using an array of bottom-level AS.
 	D3D12_BUILD_RAYTRACING_ACCELERATION_STRUCTURE_INPUTS &topLevelInputs = topLevelBuildDesc.Inputs;
-
+	topLevelInputs.Type = D3D12_RAYTRACING_ACCELERATION_STRUCTURE_TYPE_TOP_LEVEL;
+	topLevelInputs.Flags = buildFlags;
+	topLevelInputs.NumDescs = BottomLevelASType::Count;// trianlge and AABB or are there more or some value ERIC
+	topLevelInputs.DescsLayout = D3D12_ELEMENTS_LAYOUT_ARRAY;
 
 	D3D12_RAYTRACING_ACCELERATION_STRUCTURE_PREBUILD_INFO topLevelPrebuildInfo = {};
 	if (m_raytracingAPI == RaytracingAPI::FallbackLayer)
@@ -216,7 +276,8 @@ AccelerationStructureBuffers DXProceduralProject::BuildTopLevelAS(AccelerationSt
 	ThrowIfFalse(topLevelPrebuildInfo.ResultDataMaxSizeInBytes > 0);
 
 	// TODO-2.6: Allocate a UAV buffer for the scracth/temporary top-level AS data.
-
+	// give unordered access the explanation is on line 121
+	AllocateUAVBuffer(device, topLevelPrebuildInfo.ScratchDataSizeInBytes, &scratch, D3D12_RESOURCE_STATE_UNORDERED_ACCESS, L"ScratchTODO2.6"); 
 
 	// Allocate space for the top-level AS.
 	{
@@ -231,7 +292,7 @@ AccelerationStructureBuffers DXProceduralProject::BuildTopLevelAS(AccelerationSt
 		}
 
 		// TODO-2.6: Allocate a UAV buffer for the actual top-level AS.
-
+		AllocateUAVBuffer(device, topLevelPrebuildInfo.ResultDataMaxSizeInBytes, &topLevelAS, initialResourceState, L"TopLevelAS2.6");
 	}
 
 	// Note on Emulated GPU pointers (AKA Wrapped pointers) requirement in Fallback Layer:
@@ -259,7 +320,8 @@ AccelerationStructureBuffers DXProceduralProject::BuildTopLevelAS(AccelerationSt
 		};
 
 		// TODO-2.6: Call the fallback-templated version of BuildBottomLevelASInstanceDescs() you completed above.
-
+		// what do I pass in for the template. WRAPPED_GPU_POINTER seems wrong
+		BuildBottomLevelASInstanceDescs< D3D12_RAYTRACING_FALLBACK_INSTANCE_DESC, WRAPPED_GPU_POINTER>(bottomLevelASaddresses, &instanceDescsResource);
 	}
 	else // DirectX Raytracing
 	{
@@ -271,7 +333,7 @@ AccelerationStructureBuffers DXProceduralProject::BuildTopLevelAS(AccelerationSt
 		};
 
 		// TODO-2.6: Call the DXR-templated version of BuildBottomLevelASInstanceDescs() you completed above.
-
+		BuildBottomLevelASInstanceDescs< D3D12_RAYTRACING_INSTANCE_DESC, D3D12_GPU_VIRTUAL_ADDRESS>(bottomLevelASaddresses, &instanceDescsResource);
 	}
 
 	// Create a wrapped pointer to the acceleration structure.
@@ -283,7 +345,10 @@ AccelerationStructureBuffers DXProceduralProject::BuildTopLevelAS(AccelerationSt
 
 	// TODO-2.6: fill in the topLevelBuildDesc. Read about D3D12_BUILD_RAYTRACING_ACCELERATION_STRUCTURE_DESC.
 	// This should be as easy as passing the GPU addresses to the struct using GetGPUVirtualAddress() calls.
-
+	// like above except we add the instance since we are top level
+	topLevelInputs.InstanceDescs = instanceDescsResource->GetGPUVirtualAddress();
+	topLevelBuildDesc.DestAccelerationStructureData = topLevelAS->GetGPUVirtualAddress();
+	topLevelBuildDesc.ScratchAccelerationStructureData = scratch->GetGPUVirtualAddress();
 
 	// Build acceleration structure.
 	if (m_raytracingAPI == RaytracingAPI::FallbackLayer)
@@ -302,7 +367,17 @@ AccelerationStructureBuffers DXProceduralProject::BuildTopLevelAS(AccelerationSt
 	// Very similar to how you did this in BuildBottomLevelAS() except now you have to worry about topLevelASBuffers.instanceDesc.
 	// Consider looking into the AccelerationStructureBuffers struct in DXR-Structs.h.
 	// Make sure to return the topLevelASBuffers before you exit the function.
-	return AccelerationStructureBuffers{};
+	//ComPtr<ID3D12Resource> scratch;						// temp data needed by the GPU to build the AS
+	//ComPtr<ID3D12Resource> accelerationStructure;		// actual AS data
+	//ComPtr<ID3D12Resource> instanceDesc;				// used only for top-level AS
+	//UINT64                 ResultDataMaxSizeInBytes;	// this can be acquired from the prebuild info
+	AccelerationStructureBuffers return_buff = {};
+	return_buff.scratch = scratch;
+	return_buff.accelerationStructure = topLevelAS;
+	return_buff.instanceDesc = instanceDescsResource;
+	return_buff.ResultDataMaxSizeInBytes = topLevelPrebuildInfo.ResultDataMaxSizeInBytes;
+
+	return return_buff;
 }
 
 // TODO-2.6: This will wrap building the Acceleration Structure! This is what we will call when building our scene.
@@ -312,18 +387,23 @@ void DXProceduralProject::BuildAccelerationStructures()
 	auto commandList = m_deviceResources->GetCommandList();
 	auto commandQueue = m_deviceResources->GetCommandQueue();
 	auto commandAllocator = m_deviceResources->GetCommandAllocator();
+	auto flag = D3D12_RAYTRACING_ACCELERATION_STRUCTURE_BUILD_FLAG_NONE; // the other flags seem too fancy for me
 
 	// Reset the command list for the acceleration structure construction.
 	commandList->Reset(commandAllocator, nullptr);
 
 	// TODO-2.6: Build the geometry descriptors. Hint: you filled in a function that does this.
 	array<vector<D3D12_RAYTRACING_GEOMETRY_DESC>, BottomLevelASType::Count> geometryDescs;
-
+
+	BuildGeometryDescsForBottomLevelAS(geometryDescs);
 
 	// TODO-2.6: For each bottom-level object (triangle, procedural), build a bottom-level AS.
 	// Hint: you filled in a function that does this.
 	AccelerationStructureBuffers bottomLevelAS[BottomLevelASType::Count];
-
+	for (UINT count = 0; count < BottomLevelASType::Count; count++)
+	{
+		bottomLevelAS[count] = BuildBottomLevelAS(geometryDescs[count], flag);
+	}
 
 	// Batch all resource barriers for bottom-level AS builds.
 	// This will Notifies the driver that it needs to synchronize multiple accesses to resources.
@@ -336,7 +416,7 @@ void DXProceduralProject::BuildAccelerationStructures()
 
 	// TODO-2.6: Build top-level AS. Hint, you already made a function that does this.
 	AccelerationStructureBuffers topLevelAS;
-
+	topLevelAS = BuildTopLevelAS(bottomLevelAS, flag);
 
 	// Kick off acceleration structure construction.
 	m_deviceResources->ExecuteCommandList();
@@ -347,5 +427,11 @@ void DXProceduralProject::BuildAccelerationStructures()
 	// TODO-2.6: Store the AS buffers. The rest of the buffers will be released once we exit the function.
 	// Do this for both the bottom-level and the top-level AS. Consider re-reading the DXProceduralProject class
 	// to find what member variables should be set.
-
+	// set all bottoms
+	for (UINT count = 0; count < BottomLevelASType::Count; count++) 
+	{
+		m_bottomLevelAS[count] = bottomLevelAS[count].accelerationStructure;
+	}
+	// set the top level we only have one for this prj
+	m_topLevelAS = topLevelAS.accelerationStructure;
 }