NVSG NVIDIA Scene Graph

Transcription

1 NVSG NVIDIA Scene Graph Leveraging the World's Fastest Scene Graph

2 Agenda Overview NVSG Shader integration Interactive ray tracing Multi-GPU support

3 NVIDIA Scene Graph (NVSG) The first cross-platform scene graph with full hardware shading support through latest visual computing technology to achieve a new level of realism.

4 Why use NVSG? Performance Reduces amount of data Structures data to allow culling and sorting Support from NVIDIA The global leader in computer graphics Productivity Manages graphics hardware, reducing requirement for OpenGL coding No-cost tool License free of charge Scalability Works seamlessly with complex hardware configurations NVIDIA SLI ready Shading Tk Takes full flladvantage of Cg/CgFX/MetaSL Abstracts low-level shader work

5 Markets Automotive Training & Visualization Vis-Sim Virtual Reality Broadcasting Digital Studio Gaming Oil&Gas Misc...

6 NVSG - NVIDIA Scene Graph p C++ scene-graph C g h API OS Independent M l i h d safe Multi-thread f Fast Shader support Cluster aware Latest GPU support Easyy to use Easy to extend Future proof Free to NVIDIA developers

7 NVSG Software Stack Loader & Saver Plug-ins 3D Application NVSG Cg-Runtime 3D API CUDA GPU

8 Device-Independent Rendering p g NVSG OpenGL DirectX Vendor-Neutral V d N l Hardware Ray Tracing CUDACUDA Compatible Hardware

9 Plug-in Architecture NVSG Plug-In Framework I18N Translation Loader API Saver API DLL / SO D B I U I U M S D M S A E B A E G R G R E E

10 Interchange Formats DCC Creator Catia Maya 3D Studio Max Shader Authoring FXComposer mental mill COLLADA OpenFlight OBJ VRML PLY nbf/nvsg CgFX MetaSL others... Image Generator NVSG

11 Differing Render Targets Framebuffer FBO 2D Overlay Ray Tracing Broadcast Graphics Hardware Transform Feedback Gelato

12 Shadows, Effects, Skinning

13 GUI, Menus, overlays,...

14 GVO / SDI Broadcast Graphics Uncompressed 8-, 10-, or 12-bit SDI formats Enabling a direct connection to broadcast monitors, switchers, tape decks, or SDI projectors. Source code example GVOWinRenderArea NVAPI

15 Multipass Rendering Depth of Field Order Independent Transparency FSAA Stereo (implementation dependent)

16 Order Independent Transparency Without OIT With OIT

17 Improved FSAA Quality No FSAA 16x MPAA MPAA: Combine HW FSAA with Software controlled FSAA to get benefits from both worlds: maintain speed and quality

18 Stereo Quadbuffered Stereo Passive stereo with independent outputs

19 Active Stereo Sequential Projection of both Views left/right Eye flipping -> Active Compute left/right Eye Views 3D Modell L R Screen 3D Effect through Shutter Glasses -> Active

20 Passive Stereo L Horizontal Polarization Filter L Permant Projection of both Views no left/right Eye flipping -> Passive Compute left/right Eye Views 3D Modell Screen R R Vertical Polarization Filter 3D Effect through Polarized Glasses no Shutter -> Passive

21 Improved Skinning Usage of skinning processors CPU skinning processor GPU skinning minimal requirement: Shader Model 4 Future: CUDA skinning Scene Old Skinning CPU Skinning GPU Skinning Chameleon.nbf Hatealien.nbf Seymour.dae na

22 Skinning Implementation App Traverser Scene Preparation Cull Traverser Cull Traverser Skinning Processor Render Traverser CPU GPU CUDA

23 Skinning Processors - CPU Vertex Attribute Set Skin Xform INPUT VBO Influence Data (Sys Mem) Weights / Indices Matrices CPU Processor v = w j j M j v OUTPUT VBO R A S T E R I Z A T I O N CgFX Effect

24 Skinning Processors - GPU Vertex Attribute Set Skin Xform CgFX Effect INPUT VBO Influence Data (Textures) Weights / Indices Matrices Cg Geometry Program Transform Feedback v = w j j M j v OUTPUT VBO R A S T E R I Z A T I O N

25 Shaders - Cg GPU Shading language inspired by C API-independent d OpenGL or Direct3D Platform-independent NVIDIA, AMD/ATI, PS3 Windows, Linux, MacOS X, Solaris Hardware and API variation managed with profiles Profile = execution environment + compiled program format Profiles can determine How types are represented Available standard library routines Semantics of execution Part of 3D content creation tool chains Write your shaders in Cg; deploy them to any API or platform

26 Cg software stack 3D Application or Game Cg Compiler Cg Run-time API Cg 3D API interface: CgGL or CgD3D 3D API OpenGL or Direct3D Graphics Processing Unit (GPU)

27 CgFX support Metafile format used for shader description. The CgFX runtime takes care where to put textures and other information for the shader No shader specific C/C++ code needed!

28 Shader Authoring Tool Integration Shader authoring tools FXComposer mental mill Visual programming Drag & Drop Debugging

29 mental mill Shader creation for all levels of technical expertise Shader creation without programming for non-technical users and Artists Familiar graph editing paradigm Simple parameter editing

30 MetaSL TM Easy to use meta language for shader writers Shaders can be exported for use in: mental ray FX Composer 2.5 NVSG

31 Workflow Shader Integration MetaSL FXComposer MetaSL & Phenomenon Compiler CgFX NVSG API CgFx StateAttribute Cg-runtime & Compiler

32 Interactive Ray Tracing CUDA Everywhere Low Level: SDK High Level: NVSG Integration

33 Ray Tracing SDK C++ SDK Shader-based (CUDA*) Dynamic Scenes Hybrid rasterization/ray tracing Use the right tools to get maximum performance and quality

34 RT Scene Traversal in NVSG NVSG Scene App Traverser App Traverser Ray Tracing Traverser Cull Traverser Cull Traverser RT-API GL Traverser CUDA OpenGL OpenGL

35 Ray y Tracing g Shaders in NVSG NVSG Scene MetaSL CgFX CgFX CgFX Rasterized MetaSL Compiler Cg Runtime Ray Traced OpenGL / DX irt SDK Ray Tracing Engine

36 MGPU SDK - NVScale Image compositor for sort first and sort last based applications Screen tiling, alpha and depth based compositing Platforms: win32/64, linux 64 Compositor implementation based on latest technologies, no migration effort for applications (next gen hardware will provide faster transport) Configurable: 1-1, n-1, hierarchical Screen Tiling + 12 Alpha compositing Depth compositing Hierarchical compositing

37 Distributed rendering with NVSG GLDistributedRenderArea Clients must derive from this and integrate with windowing system Optionally select GPUs to be used Uses MGPUSDK for image composition RenderTraverser DistributedRendGLTraverser GLDistributedTraverser RenderArea DistributedRenderArea GLDistributedRenderArea ClientDistributedRenderArea DistributionTraverser ib ti Assigns GPUs to handle scene graph objects based on distribution scheme Distribution scheme assigns objects to optimally balance load on GPUs GLDistributedTraverser Multiple instances (one per GPU) Instances run in parallel Each instance only renders objects assigned to its GPU Ti Triggers image composition ii when ready

38 Dinosaurs in the Museum Model characteristics: Model designed in Maya, AutoDesk 217 Million Triangles Rendering R d i Hardware: H d HP xw8600 with 32 GB System Memory 2xD2 s with 16GB total video memory Rendering Performance: Total max. triangle throughput = 1.3 GTri/s

39 Thank You! Feedback & Questions: