graphic hardware opera house
GraphicsHardwarevienna 3 september
austria 2006

Keynote Speakers

  • Matt Pharr, Neoptica
    “Interactive Rendering in The Post-GPU Era”

    Two decades ago, offline renderers introduced programmable shading into the graphics pipeline. Half a decade ago, hardware programmable shading revolutionized interactive rendering. Today, heterogeneous parallel rendering architectures with hundreds of GFLOPS of both CPU and GPU computational capability (e.g. the PS3, XBox 360, and future multi-core PCs) are once again redefining interactive graphics. These architectures will be even more disruptive than programmable GPUs were as they free the graphics pipeline from the constraints of the unidirectional, data- parallel computation model, replacing it with an interactive rendering pipeline that is able to efficiently incorporate dynamic, data-dependent algorithms, complex data structures, as well as the traditional (and very efficient) rasterization and data-parallel shading model.

    These architectures are a huge step forward for developers, providing computational resources that are suited to a much broader range of graphics computation than today's PCs with GPUs. In this talk, I will discuss the opportunities that these architectures present to developers of graphics software as well as the substantial challenges that they pose. The transition from fixed-function to programmable GPUs was not painless; now developers must cope with a number of difficult problems that are hidden on today's PCs with GPUs, including exposed parallelism, asynchronous parallel computation, and distributing computation across heterogeneous processing elements with varying capabilities. I will argue that if the algorithms used on these architectures are carefully redesigned accounting for the new capabilities of these systems, then interactive graphics may soon see even more revolutionary change than has been seen with programmable GPUs over the past few years.

    Matt Pharr recently cofounded Neoptica, a company devoted to developing software for advanced graphics on next-generation architectures. Previously, he was a member of the technical staff in the Software Architecture group at NVIDIA, where he also served as the editor of the book “GPU Gems 2: Programming Techniques for High-Performance Graphics and General Purpose Computation”, a collection covering the latest ideas in GPU programming written by experts in the industry. He was one of the founders of Exluna, a company that developed rendering software and tools; Exluna was acquired by NVIDIA in 2002. He previously worked in the Rendering R&D group at Pixar, working on the RenderMan rendering system.

    Matt and Greg Humphreys are the authors of the textbook “Physically Based Rendering: From Theory To Implementation”, which has been used in graduate-level computer graphics courses at over ten universities, including Stanford, University of North Carolina, and University of Virginia. He holds a B.S. from Yale University and a M.S. and Ph.D. from Stanford University, where he researched theoretical and systems issues related to rendering.

  • Peter-Pike Sloan, Microsoft
    “Direct3D 10 and Beyond”

    Direct3D 10 represents a significant collaborative effort between Microsoft, application developers and hardware vendors. The graphics pipeline has been extended and enhanced in several new ways: A programmable “Geometry Shader” stage that receives whole primitives and can output zero or more primitives, the ability to stream out vertices/geometry to buffers, and a common shader instruction set that includes integer and bitwise operations. Many of the driver model, API and pipeline changes were motivated by experiences and inefficiencies observed when creating Direct3D 9 applications. This talk covers the new pipeline, motivation, goals and process used in creating Direct3D 10. It will also touch on current thinking regarding challenges and possible improvements that could be made in future versions of Direct3D.

    Peter-Pike Sloan has been in the DirectX group at Microsoft for the past four years. Prior to that he was a member of the graphics group in Microsoft Research, a staff member at the Scientific Computing and Imaging group at the University of Utah and worked at both Evans and Sutherland and Parametric Technologies. He is interested in most aspects of computer graphics and most of his publications are available online at

please address your comments and feedback to