EGGH98: SIGGRAPH/Eurographics Workshop on Graphics Hardware 1998

Permanent URI for this collection

https://diglib.eg.org/handle/10.2312/337

Browse

Browsing EGGH98: SIGGRAPH/Eurographics Workshop on Graphics Hardware 1998 by Subject "B.3.2 [Memory Structures]"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    IMEM: An Intelligent Memory for Bump- and Reflection-Mapping
    (The Eurographics Association, 1998) Kugler, Anders; S. N. Spencer
    Data path simplification in the context of reflection- and bumpmapping hardware opens new solutions in the design of rendering and shading circuits. We are proposing a novel approach to rendering bump- and reflection-mapped surfaces, where the local geometry defining bump-maps is transformed on-the-fly prior to surface shading. Applying angular encoding to normal vectors results in narrower data paths and permits hardware integration of look-up tables of acceptable size. A special-purpose logic-embedded memory architecture is presented, where bump- and reflection-mapping of textured surfaces are executed by an intelligent memory device. High-performance surface shading is achieved by making use of precomputed shading- and reflection-map coordinate generation tables, and considering cache coherence of pixel-to-pixel normal vectors. Such a dedicated memory chip can easily be interfaced to a standard rasterizer, in place of texture memory to offer bump-, texture- and reflection-mapping hardware support.
  • Thumbnail Image
    Item
    Neon: A Single-Chip 3D Workstation Graphics Accelerator
    (The Eurographics Association, 1998) McCormack, Joel; McNamara, Robert; Gianos, Christopher; Seiler, Larry; Jouppi, Norman P.; Correll, Ken; S. N. Spencer
    High-performance 3D graphics accelerators traditionally require multiple chips on multiple boards, including geometry, rasterizing, pixel processing, and texture mapping chips. These designs are often scalable: they can increase performance by using more chips. Scalability has obvious costs: a minimal configuration needs several chips, and some configurations must replicate texture maps. A less obvious cost is the almost irresistible temptation to replicate chips to increase performance, rather than to design individual chips for higher performance in the first place. In contrast, Neon is a single chip that performs like a multichip design. Neon accelerates OpenGL [19] 3D rendering, as well as X11 [20] and Windows/NT 2D rendering. Since our pin budget limited peak memory bandwidth, we designed Neon from the memory system upward in order to reduce bandwidth requirements. Neon has no special-purpose memories; its eight independent 32-bit memory controllers can access color buffers, 1. depth buffers, stencil buffers, and texture data. To fit our gate budget, we shared logic among different operations with similar implementation requirements, and left floating point calculations to Digital s Alpha CPUs. Neon s performance is between HP s Visualize fx<sup>4</sup> and fx<sup>6</sup>, and is well above SGI s MXE for most operations. Neon-based boards cost much less than these competitors, due to a small part count and use of commodity SDRAMs.