Nvidia Announces Fermi-Optimized Application Acceleration Engines
August 9, 2010

Nvidia Announces Fermi-Optimized Application Acceleration Engines

Los Angeles, Calif. – Nvidia announced new releases of its Application Acceleration Engines (AXE), optimized for the newest Quadro graphics processing units (GPUs) based on the company’s new Fermi architecture. These latest versions, designed for the professional software development community, include: Nvidia SceniX 6, Nvidia Cg Tooklit 3, and Nvidia OptiX 2.
SceniX 6, Cg Toolkit 3, OptiX 2 technology, and iray rendering solutions from mental images, offer application developers a host of new features and benefits that they can offer to end-users across a range of markets, including medical imaging, automotive styling, architectural design, and energy exploration.

New features and benefits include:

Nvidia SceniX 6
New Bezier Patch geometry class, using Cg tessellation programs for the smoothest of surfaces on Nvidia Fermi architecture class GPUs;

OptiX 2 support for faster, interactive ray tracing;

Continued improvements in overall performance and fidelity, and;

Future support for iray by mental images.

Nvidia Cg Toolkit 3
New tessellation programs, allowing displacement and procedural surfaces to dynamically adapt their tessellation in real-time on the latest NVIDIA Fermi architecture class GPUs, and;

OpenGL 4 and DirectX 11 level of programmability for the latest in portable, cross platform effects.

Nvidia OptiX 2
Optimizations for new Nvidia Fermi architecture class GPUs, delivering up to 4X performance over previous generation (GT200) GPUs and >10X over G92;

Support for all Nvidia CUDA architecture-capable NVIDIA GPUs (G92 or later) on Windows, Linux, and Mac OSX, and;

Direct3D and fast interoperability in Direct3D and OpenGL; for flexible compositing and hybrid rendering opportunities.

iray from mental images
Physically correct and fully deterministic global illumination rendering;

Superb scaling between multiple NVIDIA Fermi architecture class GPUs on the same machine, and;

Distributed rendering support for efficiently leveraging multiple Quadro and Tesla™ GPUs in a network.

Nvidia Application Acceleration Engines are available immediately to developers for download.