The GTC, short for GPU Technology Conference, is nVidia’s annual gathering taking place in San Jose.
nVidia’s CEO Jen-Hsun Huang opened the event withg his traditional keynote, this year the theme being « A New Computing Model ». The message being that nVidias innovation helps to make breakthroughs to change the world. And in order to be able to achieve these breakthrougs, Jen-Hsun Huang insists that researchers and R&D departments need supercharged computing : GPU Accelerated Computing.

New nVidia SDKs to fuel and help development efforts
Huang calls the nVidia SDK an essential resource for GPU developers. nVidia’s new SDK comes in several flavors :

• nVidia Gameworks allows to include physics into games, hair look realistic, waves to shimmer and fire to flame. « Recreating the world is essential to create great games » says Huang.
• The second SDK is nVidia DesignWorks, created to make photo-real graphics through the company’s iRay 3D creation and rendering platform. An additional library called MDL of real surface textures for glass, gold, carbon allow photorealistic renderings.
• The third SDK is nVidia VRWorks, specifically targeting VR developments for graphics close-ups and reduce latency in realtime VR scenes.
• nVidia DriveWorks is a suite of algorithmes created for car developers and startups to help them create self-driving cars.
• nVidia JetPack is to help developers of nVidia’s Jetson platform for robots and drones to implement deep learning algorithmes into their development projects.

Lastly, Jen-Hsun Huang announced that CUDA 8 was close to completion, along with the cuDNN library which will allow neural net developers to create neural frameworks and run them with the best possible performance. nvGraph will complement the CUDA toolkit allowing to browse through large volumes of relational databases. And indeX, touted as the largest visualization platform, will help rendering volumetric images of data simulations out of terabytes of data.