VPro™ Graphics Provide Extreme Visual Quality and Accuracy

Silicon Graphics Fuel® offers two breakthrough VPro graphics options:

• V10 with 32MB configurable graphics memory
• V12 with 128MB configurable graphics memory

• Up to 128MB graphics memory including 104MB texture memory capacity
• Hardware acceleration of OpenGL® 1.2 core features and imaging extensions
• Hardware-accelerated specular shading (or per-pixel normal lighting with interpolation) for realistic Phong effects without a performance penalty
• Advanced texture management with asynchronous texture download capability
• 48-bit (12-bit per component) RGBA
• 96-bit hardware-accelerated accumulation buffer for depth of field, full-scene anti-aliasing, motion blurs, and other effects (V12 only)
• Perspective-correct textures and colors
• High-performance hardware clipping

VPro graphics features a 12-bit per color component path (48-bit RGBA) for all operations throughout the pixel pipeline. The higher resolution, combined with perspective-correct color and texture, equates to more accurate color, more realistic blending of colors for transparent objects, higher-quality volume visualization using 3D textures, and expanded image processing capabilities. High resolution is retained through the display interface, where 10-bit per component digital-to-analog converters ensure the very highest color accuracy.

Hardware-accelerated specular shading within VPro graphics also introduces a new level of accuracy to desktop graphics. Specular shading produces accurate highlights. Traditional Gouraud lighting can distort highlights, especially when rendering coarsely tessellated solids. This is the result of calculating the color only at each triangle vertex and then interpolating that color across the triangle. By contrast, specular shading interpolates normals (not colors) and does an interpolation for each pixel. In this way, specular shading provides a more accurate representation of the curvature of surfaces, improving the viewer's ultimate understanding of the model.

VPro graphics features OpenGL on a Chip™ for full hardware acceleration of the OpenGL 1.2 pipeline and also offers hardware acceleration of the OpenGL ARB imaging extensions. The integrated image and texture engine dramatically speeds up texture loading for fast rendering of large, complex models or volumetric data. The asynchronous texture download capability streamlines texture management for even better graphics performance when processing large textures. High-speed, configurable graphics memory allows you to customize and vary the resolution and the size of the frame buffer for the requirements of particular imaging operations.

Graphics Libraries

OpenGL
OpenGL is an industry-standard graphics development environment. OpenGL is a cross-platform portable API that enables 2D, 3D, and imaging applications to be developed once for deployment to a variety of hardware platforms with different operating systems and windowing environments. OpenGL includes operations for geometric and raster primitives, viewing and modeling transformations, lighting, shading, blending, fog, hidden surface removal, and texture mapping.

Open Inventor™
Open Inventor is an object-oriented 3D graphics toolkit that presents a programming model based on a 3D scene database. It provides a rich set of objects to speed up your development and extend 3D programming beyond OpenGL, X11, and Motif™ . The objects include cubes, polygons, text, material, camera, lights, trackballs, handle boxes, and 3D viewers and editors. Like OpenGL, Open Inventor is a cross-platform portable API.

OpenGL Performer™
The IRIS Performer toolkit gives developers the means to achieve a fast, consistent frame rate for applications in the areas of visual simulation, interactive entertainment, and simulation-based design. IRIS Performer has the intelligence to evaluate underlying hardware and optimize for best performance. Key benefits include simplified management of the visual database and maximum performance for scene rendering for all SGI® workstations.

OpenGL Optimizer™
The OpenGL Optimizer application programming interface (API) provides the technology necessary to build applications that enable users to realize the benefits of digital prototyping and large database interaction. OpenGL Optimizer is a C++, 3D graphics API built on the industry standard, OpenGL. OpenGL Optimizer extends the power of OpenGL by enabling high-performance rendering of and interaction with extremely large databases. Within the extensible, object-oriented framework of the OpenGL Optimizer API, developers can simplify large models, reduce scene complexity by removing objects that are behind other objects, and transparently utilize all available processors in the machine. The OpenGL Optimizer API takes advantage of the fact that distant objects don't need the detail that close-up objects require. It provides tools that enable the developer to create as many levels of detail (LOD) as necessary for efficient rendering. Many graphics systems today provide developers with view frustum culling--the removal of objects that are not in view. Though useful, this function does little for complex scenes where the majority of the geometry is in view. The OpenGL Optimizer API provides a general framework for culling that includes view frustum culling as well as occlusion culling the removal of objects that are hidden behind other objects and back patch culling. Simplification and culling techniques help to minimize the amount of data that gets rendered. Ultimately, however, the graphics hardware renders polygons. The OpenGL Optimizer API provides many advanced tessellation routines that generate optimal polygonal surfaces from trimmed parametric surfaces including NURBs. Highly curved surfaces are densely tessellated while flatter areas are sparsely tessellated.