Scribd Logo
Upload

Welcome to Scribd!

  • 23 views

    Vertex Processing Selection Capability: December 2007 Revision 001

    Uploaded by

    copslock

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Vertex Processing Selection Capability: December 2007 Revision 001

    Uploaded by

    copslock
    23 views10 pages

    Original Title

    27-45508 (1)

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    23 views10 pages

    Vertex Processing Selection Capability: December 2007 Revision 001

    Uploaded by

    copslock

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 10

    Vertex Processing Selection

    Capability
    White Paper for Intel® G965, G35, and GM965 Express Chipsets

    December 2007

    Revision 001

    Document Number: 318888-001


    INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL® PRODUCTS. NO LICENSE, EXPRESS OR
    IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT
    AS PROVIDED IN INTEL'S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY
    WHATSOEVER, AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL
    PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY,
    OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT.
    UNLESS OTHERWISE AGREED IN WRITING BY INTEL, THE INTEL PRODUCTS ARE NOT DESIGNED NOR INTENDED FOR ANY
    APPLICATION IN WHICH THE FAILURE OF THE INTEL PRODUCT COULD CREATE A SITUATION WHERE PERSONAL INJURY OR
    DEATH MAY OCCUR.
    Intel may make changes to specifications and product descriptions at any time, without notice. Designers must not rely on the
    absence or characteristics of any features or instructions marked "reserved" or "undefined." Intel reserves these for future
    definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them. The
    information here is subject to change without notice. Do not finalize a design with this information.
    The products described in this document may contain design defects or errors known as errata which may cause the product to
    deviate from published specifications. Current characterized errata are available on request.
    Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing your product order.
    This document contains information on products in the design phase of development.
    All products, platforms, dates, and figures specified are preliminary based on current expectations, and are subject to change
    without notice. All dates specified are target dates, are provided for planning purposes only and are subject to change.
    This document contains information on products in the design phase of development. Do not finalize a design with this
    information. Revised information will be published when the product is available. Verify with your local sales office that you have
    the latest datasheet before finalizing a design.
    Intel and the Intel logo are trademarks of Intel Corporation in the U.S. and other countries.
    *Other names and brands may be claimed as the property of others.
    Copyright © 2007, Intel Corporation. All rights reserved.

    2 White Paper
    Contents
    1 Introduction .....................................................................................................5
    1.1 Terminology ..........................................................................................5

    2 Section 2 .........................................................................................................6
    2.1 Introduction ..........................................................................................6
    2.2 Integrated Graphics Architecture ..............................................................6
    2.3 Why Vertex Processing Selection Capability? ..............................................7
    2.4 Driver Architecture for Vertex Processing Selection Capability .......................8
    2.5 Summary ............................................................................................10

    Figures
    Figure 1: 3D and Media Kernels .........................................................................6
    Figure 2: Performance with HWVP ......................................................................7
    Figure 3: Benefits of Vertex Processing Selection Capability (SWVP) ........................8
    Figure 4: Block Diagram Showing the Graphics Engine Path - HWVP ........................9
    Figure 5: Block Diagram Showing the Processor Path - SWVP .................................9
    Figure 6: 3D Applications Supported by SWVP....................................................10

    White Paper 3
    Revision History

    Revision Description Revision Date


    Number

    -001 Initial release. December 2007

    4 White Paper
    Introduction

    1 Introduction
    The purpose of this document is to explain what the vertex processing selection
    capability is and what the benefits are. The document is targeted for OEM’s, system
    integrators, and end-users that will be building or using motherboards with
    Intel® G965, GM965, and G35 Express Chipsets.

    1.1 Terminology

    Term Description

    TnL Transform and light

    PSGP Processor Specific Graphics Pipeline: Code path that enables Vertex
    processing and TnL (transform and light) calculation to run on a
    processor.

    SWVP Software Vertex Processing: Vertex Processing that takes place on


    the CPU utilizing the PSGP

    HWVP Hardware Vertex Processing: Vertex Processing that takes place on a


    graphics chip

    White Paper 5
    Section 2

    2 Section 2

    2.1 Introduction
    Intel recently introduced the 15.6 and 14.31 Windows Vista* and Windows* XP
    graphics drivers that enables Shader Model 3.0 including support for hardware vertex
    shader and HW TnL on the Intel® G965, GM965, and G35 Express Chipsets. This
    capability has shown enhancements in game compatibility as well as game play. Along
    with this capability, Intel also introduced a vertex processing selection capability which
    allows the drivers to switch between using the integrated graphics engine and the
    Intel processors for vertex processing depending on the application. The end result is
    that Intel is able to deliver the highest possible frame rates by leveraging Intel’s world
    class processors.

    2.2 Integrated Graphics Architecture


    The Intel GMA X3x00 architecture is based upon a unified shader architecture that is
    comprised of a collection of highly programmable and scalable 32-bit floating point
    engines. This programmable architecture is one based upon a unified shader model,
    which means both vertex shaders and pixels shaders run on the same programmable
    engines. These same programmable engines are also able to process media codecs,
    as well as various filters for text and image processing. The Intel graphics driver
    architecture delivers scalability and flexibility by supporting features within kernels
    that are compiled for the Intel graphics architecture. At the high level there are
    kernels that support 3D and media. Within 3D there are kernels for DirectX* and
    OpenGL*. DirectX* then has kernels that support pixel and vertex shading, and
    filtering. OpenGL has fragment and vertex shading, as well as filtering kernels. The
    media kernels are codec dependent, but capabilities like IDCT (Inverse Discrete
    Cosine Transform), motion compensation, and filtering techniques like advanced de-
    interlacing would all be supported by a kernel. Figure 1 is a view of the driver
    architecture with kernels.

    Figure 1: 3D and Media Kernels

    6 White Paper
    Section 2

    2.3 Why Vertex Processing Selection Capability?


    In the process of designing and enabling a driver that supports Vertex Processing and
    TnL, Intel engineers discovered that some applications, mainly 3D games, performed
    better with vertex processing done on the processor rather than on the graphics
    engine. As mentioned above, the architecture uses the same programmable engines
    to process all shaders. By off-loading some of the vertex shader processing to the
    processor, the graphics engine is able to process more pixel shader data and do
    additional work like anisotropic filtering. Therefore, the graphics engine is able to do
    more while staying within market required power, thermal, and cost envelopes.

    Separate vertex shader and pixel shader units within the graphics engine was a
    potential design option for Intel graphics engines. However, upon analysis of various
    3D applications, it was discovered that the vertex traffic is at the beginning of each
    scene, and only accounted for one-third of the compute requirements for a scene. On
    the other hand, pixel traffic is about two-thirds of the scene and required most of the
    compute resources. So for much of the time the vertex shaders would be sitting idle.
    Of course this is application specific and one we will explore further. Figure 2 shows a
    small sample of applications that perform better with Vertex Processing being done on
    our integrated graphics parts. As one can see Age of Empires 3* receives a modest
    gain of 15%, while Far Cry* and Hal-Life 2 Episode 1* see extremely large gains of
    95% and 230% respectively. Figure 3 shows a small sample of applications and the
    benefits of software vertex processing. When using software vertex processing, the
    performance benefit gained over hardware vertex processing ranges from 30-65%
    across a small sample of games. A 30% performance benefit was measured for
    FEAR*, 60% for Guild Wars* and 65% for Grand Theft Auto*. See Figure 3 below.

    Figure 2: Performance with HWVP

    White Paper 7
    Section 2

    Figure 3: Benefits of Vertex Processing Selection Capability (SWVP)

    2.4 Driver Architecture for Vertex Processing


    Selection Capability
    Within any given 3D API there are certain capabilities that are exposed to the
    application. The Intel graphics driver always exposes Vertex Shader 3.0, and below,
    as well as HW TnL capabilities bits by default. Many applications perform best when
    the vertex processing is done on Intel graphics engines. As stated above, there are
    several exceptions where the application may perform better when vertex processing
    is done on a processor. If the application performs better with processor-based vertex
    processing, and has met Intel’s selection criteria, it is placed in a list of applications
    contained within the driver INF file. The driver detects the application executables,
    turns off the integrated graphics vertex processing path and turns on the PSGP path
    that runs on an Intel processor. Figure 4 is a block diagram showing the graphics
    engine path (HWVP). Figure 5 is a block diagram showing the processor path (SWVP).

    Applications are validated across different configurations and results are reviewed to
    determine which applications qualify for inclusion into the PSGP list. As new
    applications are launched, they are reviewed for inclusion on the list as shown below
    in Figure 6 (a list of 3D applications that are supported by the PSGP (SWVP)).

    8 White Paper
    Section 2

    Figure 4: Block Diagram Showing the Graphics Engine Path - HWVP

    Figure 5: Block Diagram Showing the Processor Path - SWVP

    White Paper 9
    Section 2

    Figure 6: 3D Applications Supported by SWVP

    NOTE: Other names and brands may be claimed as the property of others.

    2.5 Summary
    Intel recently introduced graphic driver versions 15.6 and 14.31 for Windows Vista
    and Windows XP that enable Shader Model 3.0 (including support for hardware vertex
    shader and HW TnL on the Intel® G965, GM965, and G35 Express Chipsets.) Intel
    also introduced the capability to switch between HWVP and SWVP based upon the 3D
    application running on Intel integrated graphics engines. Results showed that some
    3D applications perform best with HWVP, which is the default configuration. In other
    instances, results showed that some 3D applications perform best with SWVP. Intel’s
    goal is to provide the best user experience possible for 3D applications while
    leveraging the performance of the processor where it makes sense.

    10 White Paper

    You might also like