Ray tracing and AI to Siggraph- Nvidia Introduction

Nvidiadisplayed a series of RTX Studio laptops and mobile workstations at Siggraph this year. The Nvidia company is said to have purpose-built for creators from major OEM. The 10 new designs from Dell, HP, Lenovo, and Boxx have raised the number of RTX Studio models to 27. The ISV demos were on display from Adobe, Autodesk and Blackmagic Design.
RTX Studio laptops

Nvidia declared the availability of a brand new Nvidia Studio Driver which optimizes the known creative apps such as Octane Render, Blender, Autodesk Arnold, Maxon Cinema 4D, Substance Painter by Adobe and Magix Vegas Pro. The newly introduced Studio Driver adds most requested feature support for 30-bit color in OpenGL apps. The OpenGL apps include Adobe Photoshop and Premiere for all GPUs. This feature assists content creators. They can work with HDR photos and videos and avoid 24-bit color banding.

The latest configurations which range from laptops with GeForce RTX 2060 GPUs to professional-grade mobile workstations with Quadro RTX 5000 graphics are inclusive of

  • BoxxGoBoxx SLM mobile workstations are available to consumers with a Quadro RTX 3000 GPU in the 15-inch system and either Quadro RTX 4000 or 3000 in the 17-inch system.
  • Dell Precision 7540 and Dell Precision 7740 mobile workstations are available today and configurable up to Quadro RTX 5000 GPUs.
  • HP ZBook 15 and 17 mobile workstations feature Quadro RTX GPUs, with the 17-inch model configurable up to a Quadro RTX 5000.
  • Lenovo Legion Y740 Laptop Studio Edition features up to GeForce RTX 2080 GPUs within 17- and 15-inch laptops will be available in the market later this fall
  • Lenovo ThinkPad P53 and P73 mobile workstations support up to Quadro RTX 5000 GPUs within 17- and 15-inch systems. The ThinkPad P53 is available now; the ThinkPad P73 will be available from August first week

RTX Studio laptops and workstations
The RTX Studio laptops and workstations are very much part of the Nvidia Studio platform. The platform consists of GPUs with the Nvidia Studio Stack of and dedicated Studio Drivers. The RTX Studio laptops and workstations suit the hardware and software specifications required to get Nvidia’s RTX Studio badge which makes it easily identifiable by the users.

The Nvidia systems undergo thorough hardware and software testing for top creative applications and workflows. The RTX Studio mobile workstations based on Quadro RTX GPUs have many benefits. The enterprise-level benefits like ISV certifications, IT management tools, and advanced video and display features.

RTX Studio systems have GeForce RTX 2080, 2070, and 2060 GPUs and Quadro RTX 5000, 4000, and 3000 GPUs. The RTX GPUs speed up content creation from video editing to 3D rendering. The performance is up to 7× faster than that of the MacBook Pro. Quadro RTX 5000 GPU-based models are inclusive of 16 GB of graphics memory. The largest available in a mobile form factor enables advanced multi-app creative workflows. RTX Studio systems aid the use of large 3D models.

RTX Studio models available in the market now are: Acer ConceptD 7, Gigabyte Aero 15 OLED and Aero 17 HDR, BoxxGoBoxx SLM 15 and 17, Dell Precision 7540 and 7740, HP ZBook 15 and 17, Lenovo ThinkPad P53, MSI P65 Creator, P75 Creator, WS65, WS75 and WE75, and Razer Blade 15 Studio Edition. Other RTX Studio laptops are expected to ship later this year from Acer, ASUS, Dell, HP, Lenovo, and Razer.

Nvidia extending RTX reach

Nvidia extending RTX reach

Nvidia after disclosing its RTX technology at Siggraph 2018, it has been able to demonstrate significant progress in gathering support for its hardware-assisted approach for a wide range of applications. The applications include product design, architecture, gaming, effects, and scientific visualization. According to Nvidia, the companies Pixar, Renault, New Balance, Woods Bagot, and Siemens Healthineers are using ray tracing in their workflows.

At Siggraph, top seven ISVs launched new applications with support for Nvidia RTX, including:

  • Adobe Substance Painter is dependent on RTX ray tracing to accelerate baking performance up to 192× faster than CPUs according to Nvidia.
  • Autodesk Flame: RTX Tensor Cores improves visual effects and compositing workflows by speeding up Autodesk’s new ML technology in Flame v2020.1 helps artists isolate, extract, and change common objects in moving footage.
  • Blender Cycles: NvidiaOptiX 7 along with CUDA speeds up Blender’s open-source renderer.
  • Dimension 5 D5 Fusion: RTX ray tracing via UE4’s implementation of DXR aids architects and designers to simulate ground truth lighting and shadows accurately.
  • Daz 3D Daz Studio: NvidiaIray aids creators to accommodate scenes with interactive RTX accelerated ray tracing to quickly build their artistic composition and render out in full fidelity.
  • Foundry MODO: RTX performance through OptiX in the completely redesigned MODO path-traced renderer offers a noticeable performance boost over CPU rendering.
  • LuxionKeyShot: RTX accelerated ray tracing and AI aids in photorealistic visualization of 3D data for product design reviews. The other areas of benefit are marketing, animations, illustrations and more via OptiX support in KeyShot 9.

Known apps like Adobe Dimension CC and Chaos Group V-Ray provide users the ability to create fluidly, with great realism powered by RTX technology’s RT Cores. Nvidia says Adobe Substance designer can deliver speed. The increase in speed is up to 200% when compared with earlier CPU-based technology. The increase in speed is due to the integration of RTX through DXR for light baking.

Along with the above, BlackMagic and Nvidia conveysDaVinci Resolve with DaVinci Neural Engine has implemented AI using the Nvidia AI libraries. The speed they are gaining is from the Tensor Cores found on Nvidia RTX GPUs to accelerate inferencing.

In just 12 months, RTX has become widely known. Nvidia’s VP of corporate marketing and developer programs Greg Estes claims every major 3D design application in the world has committed to supporting RTX by year’s end.

The gaming industry is on board through a robust ecosystem that is inclusive of industry-standard APIs. The APIs are DirectX 12 and Vulkan, as well as leading game engines like Unity and Unreal Engine.

In the data center:

Nvidia RTX Servers include fully optimized software stacks. They are available for OptiX RTX rendering, gaming, VR and AR, and professional visualization applications.

Nvidia’sOmniverse is a collaboration platform for 3D production pipelines which aids to streamline workflows for studios. The platform aids creative professionals to share assets across a range of industry-standard applications. The Omniverse Viewer gives an option to interact with those assets in a realtime. The ray-traced viewport is speeding up by Nvidia RTX RT Cores, CUDA cores, and Tensor Cores available with Nvidia RTX Server with QuadrovDWS.

Nvidia has also launched an RTX server for cloud gaming. The Nvidia RTX Server packs 40 Nvidia Tesla GPUs along with ray tracing technology into an 8U blade server form factor.

The RTX Server with Nvidia GRID Vs gaming speeds up streaming and gaming PC deployment applications. The applications are designed and optimized for cloud operators. Nvidia’s server technology is also applicable to Mobile Edge Computing (MEC). With MEC it can deliver AR and VR apps and game streaming with low latency and high concurrency.

AR R&D

At Siggraph, in the Emerging Technologies section, Nvidia displayed some R&D work. The R & D is done in AR for eye-tracking and fovea rendering on smart glasses.

The research demonstrates, “Foveated AR,” was a headset that adapts to your gaze in realtime with the help of deep learning. The headset adjusts the resolution of the images displayed and their focal depth based on the distance and area you are viewing. This feature gives both sharper images and a wider field of view than any previous AR display.

To get the above effect, the feature combines two different displays per eye. A high-resolution small field of view where visual acuity is highest, and a low-resolution display for peripheral vision. The end result is high-quality visual experiences with reduced power and computation.

Nvidia also displayed an option to change the display image of AR glasses. This option compensates for one’s optical prescription.

The display is thinner by many times and lighter. The display has a wider range of view. Virtual objects appear throughout the natural field of view instead of clustered in the center. The display has your prescription built right into corrective optics you wear. This option brings us much closer to the goal of comfortable, practical, and socially acceptable AR displays than anything available right now.

Bea great artist
NVIDIA GauGAN neural network makes masterpieces

NvidiaGauGAN is an interactive paint program. The program uses GANs (generative adversarial networks) to create works of art with simple brush strokes. So you can be an artist too.

GauGAN, named after post-Impressionist painter Paul Gauguin, creates photorealistic images from segmentation maps. The sketches that depict the layout of a scene.

People can use paintbrush and paint bucket tools to design their own landscapes. You can include river, grass, rock, and cloud. A style transfer algorithm allows users to apply filters, changing the color composition of the generated image. You can turn it from a photorealistic scene to a painting.

The paint program is developed using the PyTorch deep learning framework. The programmers trained the neural network behind GauGAN on a million images using the Nvidia DGX-1 deep learning system. The demo shown to the audience at GTC ran on an Nvidia Titan RTX GPU. The web app offers Nvidia GPUs through Amazon Web Services.

Liu has developed the deep neural network and accompanying app along with fellow researchers Taesung Park, Ting-Chun Wang, and Jun-Yan Zhu.

GauGAN was an instant hit at Nvidia’s GPU Technology Conference (GTC) in Silicon Valley. The latest release of the app at Siggraph lets users upload their own filters to layer onto their masterpieces. This lets the users adapt the lighting of a perfect sunset photo or copy the style of a favorite painter.

The GauGAN team has publicly released source code for the neural network behind GauGAN. It is available for non-commercial use to anyone who would like to experiment and build their own applications. GauGAN is available on the Nvidia AI Playground for visitors to experience it firsthand.

The GauGan users have created more than 500,000 images since the beta version was made publicly available last month.

Nvidia’s viral real-time AI art application, GauGAN has won two major Siggraph awards: “Best of Show” and “Audience Choice” awards.

The company has also revealed late-breaking research results in the ‘open problems in realtime rendering’ course at Siggraph. The Nvidia researchers displayed the first realtime rendering of scenes from Zero-Day, a short film leaving people in awe.