Nvidia RTX

Nvidia RTX is a development platform for rendering graphics that was created by Nvidia, primarily for real-time ray tracing. Ray tracing is typically utilized in instances where image creation is not display time sensitive (like films), meaning that applications such as video games have had to rely on rasterization for their rendering. Nvidia created RTX to enable a significant advance in computer graphics by generating lifelike images in real time.^[1] RTX runs on Nvidia Volta- and Turing-based GPUs, specifically utilizing the Tensor cores (and new RT cores on Turing) on the architectures for ray tracing acceleration.^[2]^[3]^[4]

Nvidia worked with Microsoft to integrate RTX support with Microsoft's DirectX Raytracing API (DXR). RTX is currently available through Nvidia OptiX and for Microsoft DirectX, and is in development for Vulkan.^[5]

Components

In addition to ray tracing, RTX includes artificial intelligence integration, common asset formats, rasterization (CUDA) support, and simulation APIs. The components of RTX are:^[6]

AI-accelerated features (NGX)
Asset formats (USD and MDL)
Rasterization including advanced shaders
Raytracing via OptiX, Microsoft DXR and Vulkan
Simulation tools:
- CUDA 10
- Flex
- PhysX

Products which use Ray Tracing are as follows:

RTX 2070
RTX 2080
RTX 2080Ti

Ray tracing

In computer graphics, ray tracing generates an image by tracing the path of light as pixels in an image plane and simulating the effects of its encounters with virtual objects. Previously, conventional 3D rendering has typically used a process called rasterization. Rasterization uses objects created from a mesh of triangles or polygons to represent a 3D model of an object. The rendering pipeline then converts each triangle of the 3D models into pixels on a 2D image plane. These pixels may then be further processed or “shaded” before final display on the screen.

While rasterization is typically more efficient for producing images in real-time, adding more realistic elements such as lighting effects to a rasterization pipeline requires more parameters based on a given scene.

RTX works by using acceleration structures and algorithms to build and update spatial search data structures, which compute intersections of rays with scene geometry, in real time. Applications can optimize acceleration structures for different types of content, such as static vs. animated.^[2]

While ray tracing has long been "the future" or holy grail of computer rendering, we are now seeing the advent of consumer GPUs which have enough compute capability to do interesting ray tracing workloads in real-time. It is expected that many use cases will employ hybrid renderers which combine rasterization and ray tracing, so tight integration with an existing rendering API is very important.

Acceleration structures in RTX function on two levels, which enables rigid-body animation and instancing. Bottom-level acceleration structures are built from geometric primitives, such as triangles, where the input for primitives builds are specified through one or more geometry descriptors. Top-level acceleration structures are built from references to bottom-level structures.^[2]

Development

APIs using RTX

Nvidia OptiX

Nvidia OptiX (OptiX Application Acceleration Engine) is a general purpose ray-tracing API for rendering, baking, collision detection, A.I. queries, etc. It implements a modern shader-centric, stateless and bindless design. The computations are offloaded to the GPUs through either the low-level or the high-level API introduced with CUDA.

Nvidia OptiX is part of Nvidia DesignWorks. OptiX is a high-level, or "to-the-algorithm" API, meaning that it is designed to encapsulate the entire algorithm of which ray tracing is a part, not just the ray tracing itself. This is meant to allow the OptiX engine to execute the larger algorithm with great flexibility without application-side changes.

Aside from computer graphics rendering, OptiX also helps in optical and acoustical design, radiation and electromagnetic research, artificial intelligence queries and collision analysis.

Microsoft DirectX Raytracing

Microsoft DXR is an extension of its previous DirectX 12. The API works by giving the application significant low-level control, reflected in several design decisions that integrate ray tracing with other work such as rasterization. Ray tracing shaders are also dispatched as grids of work items, letting implementation utilize parallel processing and perform low-level scheduling work of items. The application retains the responsibility of synchronizing GPU work and resources, meaning that developers can optimize the amount of overlap between ray tracing, rasterization, compute work, and memory transfers.^[2]

DXR introduces three new concepts to DirectX that the application must manage:

Ray Tracing Pipeline State Objects contain the compiled shader code that gets executed during a ray tracing dispatch.
Acceleration Structures contain the data structures used to accelerate ray tracing itself, i.e. the search for intersections between rays and scene geometry.
Shader Tables define the relationship between ray tracing shaders, their resources (textures, constants, etc.), and scene geometry.^[2]

Vulkan

Currently, Nvidia is in the process of developing a ray-tracing extension to the Vulkan cross-platform graphics and compute API, which is designed by the Khronos Group. Vulkan is a cross-platform graphics and compute API, and the extension intends to enable Vulkan developers to access RTX.^[7]

Requirements

Nvidia OptiX

Developers who wish to use NVIDIA OptiX need:

Nvidia Turing or Volta GPU (suggested), Nvidia Kepler or newer GPU (required)
Nvidia driver version 396 or higher
Windows 10 64-bit; Linux 64-bit; Mac OS 10.9 or higher
Nvidia CUDA Toolkit
OptiX SDK 5.0 or higher

RTX for DirectX

Developers who wish to use DirectX Raytracing applications accelerated by RTX need:^[2]

Nvidia Turing or Volta GPU
Nvidia driver version 396 or higher
Windows 10 RS4
Microsoft’s DXR developer package, consisting of DXR-enabled D3D runtimes, HLSL compiler, and headers
A complete set of denoisers will be available in the upcoming release of GameWorks for Ray Tracing

Nvidia also recommends Nvidia NSight Graphics, and notes that DXR is an experimental feature of DirectX, meaning that developer mode must be enabled in windows in order to run DXR applications.^[2]

References

↑ Alwani, Rishi. "Microsoft and Nvidia Tech to Bring Photorealistic Games With Ray Tracing". Gadgets 360. Retrieved March 21, 2018.
1 2 3 4 5 6 7 Atlavilla, Dave. "Nvidia And Microsoft Lay Foundation For Photorealistic Gaming With Real-Time Ray Tracing". Forbes. Retrieved March 19, 2018.
↑ "Nvidia announces RTX 2000 GPU series with '6 times more performance' and ray-tracing". The Verge. Retrieved 2018-08-20.
↑ "Nvidia reveals $800 GeForce RTX 2080 at Gamescom 2018". CNet.
↑ Walton, Jarred. "Nvidia talks ray tracing and Volta hardware". PC Gamer. Retrieved March 19, 2018.
↑ "NVIDIA RTX™ platform". Nvidia.
↑ Trevett, Neil. "More on Vulkan and SPIR-V: The future of high-performance graphics" (PDF). Khronos Group. Khronos Group. Retrieved March 9, 2015.

External links

Nvidia developer page on RTX

This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.

[1] Alwani, Rishi. "Microsoft and Nvidia Tech to Bring Photorealistic Games With Ray Tracing". Gadgets 360. Retrieved March 21, 2018.

[Forbes2018-2] 1 2 3 4 5 6 7 Atlavilla, Dave. "Nvidia And Microsoft Lay Foundation For Photorealistic Gaming With Real-Time Ray Tracing". Forbes. Retrieved March 19, 2018.

[3] "Nvidia announces RTX 2000 GPU series with '6 times more performance' and ray-tracing". The Verge. Retrieved 2018-08-20.

[4] "Nvidia reveals $800 GeForce RTX 2080 at Gamescom 2018". CNet.

[5] Walton, Jarred. "Nvidia talks ray tracing and Volta hardware". PC Gamer. Retrieved March 19, 2018.

[6] "NVIDIA RTX™ platform". Nvidia.

[7] Trevett, Neil. "More on Vulkan and SPIR-V: The future of high-performance graphics" (PDF). Khronos Group. Khronos Group. Retrieved March 9, 2015.