% | $
Quotes you view appear here for quick access.

Intel Corporation Message Board

you are viewing a single comment's thread.

view the rest of the posts
  • singhlion2001 singhlion2001 Nov 22, 2012 12:34 PM Flag

    Intel buys U.K. graphics chip team

    Intel desperate attempts on GPU frontier to catch up with competition

    The market for portable devices (mobile devices that we can carry) continues to soar in spite of economic difficulties and general uncertainty. Although portable devices are not necessarily replacing the PC, they are outselling the PC. All of these devices have a graphics processor (GPU) integrated in the device’s system on a chip (SoC) application processor.

    The application processors are made by a dozen-plus semiconductor suppliers (Broadcom, Apple, Intel, Marvell, MediaTek, Nvidia, ST-Ericsson, Texas Instruments, Toshiba, Qualcomm, Samsung, etc.). All of those companies fall into one of two categories: vertically integrated or IP buyers. Nvidia and Qualcomm are the vertically integrated companies with their own GPU designs; all of the other companies buy GPU IP from one of four IP suppliers (ARM, DMP, Imagination Technologies, or Vivante). The exceptions to this tidy categorization are Samsung, which has an internal GPU design and buys IP from ARM and Imagination Technologies, and Broadcom, which has an internal design GPU and buys IP from ARM.

    The leading high-volume suppliers of application processors (i.e., SoCs) are Apple, Texas Instruments, Qualcomm, and Samsung. ARM supplies GPU IP for some of Samsung’s mobile phones, while Imagination Technologies’ GPU IP is used in Apple, Texas Instruments, and some of Samsung’s mobile phones and tablets.

    The upcoming SoC suppliers with impressive design wins to their credit are Intel (Imagination Technologies’ GPU) and Nvidia (proprietary GPU).

    Other SoC suppliers that buy GPU IP are Freescale, Huawei, MediaTek, Rockchip, and Wonder Media/VIA. These companies have participated in the feature phone market, and some of them have recently entered the smartphone and tablet segment.

    One area that is fueling the growth of portable SoCs is the exploding tablet market in China, which is contributing significantly to the astounding growth that ARM and Vivante are experiencing. As a result of this tablet surge in China, dramatic changes are expected during the next 12 months. Apple will introduce a 7-inch tablet, Microsoft will bring out its Surface tablet, and Texas Instruments will continue to supply Amazon (even though the company is pulling back from the smartphone market as Nokia loses market share).

    The market for SoCs with GPUs grew 91.3% from the first half of 2011, with market shifts occurring as shown in Table 1 below.

    Supplier 1H'12 1H'11
    Qualcomm 26.4% 36.4%
    Nvidia 2.5% 3.5%
    ZiiLabs 0.1% 0.2%
    ARM IP 12.9% 6.2%
    DMP IP 1.7% 2.2%
    Imagination Technologies IP 46.5% 49.1%
    Vivante IP 9.8% 2.5%
    TOTAL 100.0% 100.0%
    Table 1: Market share changes for portable devices from 1H’11 to 1H’12
    As a result of the turbulence in the market, we expect market shares to shift dramatically through 2012, and we look for a new IP company to enter the market in 2013.

    SortNewest  |  Oldest  |  Most Replied Expand all replies
    • Supplier 1H'12 1H'11
      Qualcomm 26.4% 36.4%
      Nvidia 2.5% 3.5%
      ZiiLabs 0.1% 0.2%
      ARM IP 12.9% 6.2%
      DMP IP 1.7% 2.2%
      Imagination Technologies IP 46.5% 49.1%
      Vivante IP 9.8% 2.5%
      TOTAL 100.0% 100.0%

      ARMH & Vivante taking Market share : see above numbers...

      Rightware Collaborates with Freescale and Vivante to Enable Highly Advanced Automotive In-Vehicle Infotainment and Instrument Cluster Systems

      Integrated and highly optimized Rightware Kanzi UI Solution for the Freescale i.MX 6 Application Processors, powered by Vivante Graphics Cores, brings ultimate scalability and flexibility to car manufacturers and entire automotive ecosystem

      Freescale's i.MX 6 Series of application processors offer the industry's most scalable line of hardened processors targeting automotive systems. Each i.MX 6 product is optimized around CPU, Triple Play graphics, and HD video processing to bring next-gen, unbounded visual user experiences to in-vehicle infotainment and instrument cluster systems. The i.MX 6 product line scales from single to quad-core ARM® Cortex-A9s to offer a comprehensive range of products that are truly versatile, with software and hardware compatibility across the board. By combining the power-efficient processing capabilities of the ARM Cortex-A9 technology with bleeding-edge 3D and 2D graphics, as well as high-definition video, the i.MX 6 Series provides a new level of multimedia and graphics performance to enable an unbounded next-generation user experience on in-vehicle infotainment and instrument cluster systems.
      The Triple Play graphics inside the i.MX 6 consists of three Vivante GPU cores – GC2000 for graphics and OpenCL, GC320 for composition and GC355 for OpenVG. Each core works seamlessly together to ensure QoS and efficient work-load balance for best-in class reliability and performance. The differentiated cores allow the composition engine to offload all 2D/composition functions from the GPU to save power and allow the GPU to focus on rendering amazing 3D HMIs and configurable instrument clusters. The OpenVG core operates in parallel to the other engines and delivers the sharpest, fastest, and most accurate needle renderings.
      A new era of instrument and infotainment solutions begins with Kanzi accelerated by Vivante on i.MX 6. Rightware's Kanzi Engine is highly optimized and tightly integrated with the unified software stack of the i.MX 6 applications processor and Vivante's GPU, OpenVG, and composition processing cores. The turnkey solution shortens development time and significantly reduces effort needed to create the next wave of revolutionary 3D HMIs with the best performance for modern in-vehicle infotainment and instrument cluster systems.
      "Human-machine interfaces and user experience design for in-vehicle infotainment and instrument cluster systems have become one of the key differentiators in modern cars. High fidelity graphical user interfaces significantly enhance the appeal of a vehicle and eventually that of the manufacturer brand. Our collaboration with Freescale and Vivante will help car manufacturers and entire automotive ecosystem to rise to a whole new level with next-generation HMIs," said Tero Sarkkinen, CEO of Rightware.
      "High performance graphical user interfaces are playing a growing role in driver information systems. Demand for advanced user interfaces in automotive applications is growing quickly, both for infotainment systems and instrument clusters. Rightware's optimizations for Freescale's i.MX 6 Series processors on the Kanzi® UI Solution allow our customers to take advantage of our products' advanced capabilities without compromising performance or time to market," said David Patterson, Vice President and General Manager of Freescale's Multimedia Applications Division.
      "The next revolution of in-vehicle HMI systems revolves around a visual-centric user experience that converges automotive and consumer devices. With hi-res displays, larger screens, and better graphics, the GPU is becoming the major technical focal point in OEM decisions. The partnership with Rightware and Freescale will enable OEMs to realize this vision, making the transition straightforward and providing a captivating automotive experience," said Wei-Jin Dai, President and CEO of Vivante.

      • 1 Reply to singhlion2001
      • Vivante technology—at the core of innovation
        Vivante technology breaks through the limits of size, performance, and power to help customers deliver unique products quickly and cost-effectively. Our comprehensive—and growing—series of low-power, high-performance, silicon-efficient GPU cores ensure cutting-edge graphics quality previously only available on PC platforms.

        The recent trend in graphics hardware has been to replace fixed functionality with programmability in areas that have grown exceedingly complex, such as vertex processing and fragment processing. The OpenGL® Shading Language was designed to allow application programmers to express the processing that occurs at those programmable points of the OpenGL pipeline. Independently compilable units written in this language are called shaders. A program is a set of shaders that are compiled and linked together.

        Vivante’s GPU technology is architected from the ground up to deliver a high-precision, true unified shader implementation that maximizes graphics performance and quality per milliwatt of system power consumption.

        Robust application support
        Access to high-performance shaders without programming limitations is the key to lowering application development costs and developer support costs. Vivante’s high-performance true unified shader architecture delivers best-in-class programmability:

        ➢Khronos conformant support for OpenGL ES 2.0/1.1, OpenGL 3.0/2.1 and OpenVG 1.1
        ➢Microsoft certified DirectX 11 feature level 9_3
        ➢General purpose (GPGPU) computing
        ➢Up to 180 billion floating point operations per second (GFLOPS) in the multicore versions of GC series GPUs
        ➢ Full support for the OpenGL Shading Language without Appendix A exceptions
        ➢ High precision pipeline with support for long instructions
        ➢ Dynamic branching in the pixel shader and non-constant varying indexing

        Smaller, cooler, faster
        Vivante is the die size leader, offering the smallest OpenGL ES 2.0-compliant silicon footprint on the market. Fitting next-generation technology in the previous generation footprint, we enable full OpenGL ES 2.0 and 1.1 support in the same area as previous generation OpenGL ES 1.1-only GPU cores.

        Vivante also offers the lowest power consumption (GFLOPS per milliwatt) GPU available today:

        ➢ Ultra low-power microarchitecture
        ➢ Frequency scaling
        ➢ Software-controllable power states

        Because of our patented algorithms for reducing memory bandwidth and optimizing full-scene quality, Vivante GPUs also use less memory and processing power. Features like full-scene anti-aliasing, unified vertex and pixel shading, and rich texture mapping generate the highest quality graphics:

        ➢ Unified shader: 128 threads per shader unit, programmable aspect architecture
        ➢ Low-power techniques: extensive clock gating, flexible allocation of resources in the shader unit
        ➢ Patented anti-aliasing algorithm
        ➢ High-dynamic range textures

        Vivante’s shader uses a highly programmable multi-threaded architecture for scalability. Our modular software architecture supports multiple GPU cores with a single driver infrastructure.

        Easy to integrate
        With support for free-running AXI and AHB interfaces, customers can easily integrate our cores into their systems. Vivante’s licensees quickly and consistently reach target frequency and optimal final silicon area using provided reference implementation packages for Cadence and Synopsys design tool environments.

    • Asked the same question, Mike Bryant replies: "Unlike in small systems where it is key, in big systems, which smartphones most definitely are nowadays, the instruction set used is far less relevant with only 10% to 15% of power dealing with it. So even if ARM waved magic dust on their instruction set, which isn't perfect anyway, they can only optimise 10% to 15% of the power used. More power goes in moving data around the chip, and most goes to moving data on and off the chip to memory and peripherals."

      "ARM programs can actually use a few percent more codespace than Intel but let's say they are the same," adds Bryant, "next of course the data either processor has to be moved around - be it pictures, voice data, HTML or just the call set-up protocol, is identical for each processor. Finally the video data has to be transferred onto the display and for all the hype ARM's Mali is about the same as GPUs from Intel or Nvidia, though there are slightly more efficient solutions from Imagination Technologies or Qualcomm's Adreno."

      "So for a pure ARM system against a pure Intel system, 85% of the power usage is independent of the processor instruction set and dependent purely on the process technology," concludes Bryant.

      And that's why TSMC/Apple are pushing to 20nm and it won't be cheap

      • 1 Reply to semi_equip_junkie
      • ""So for a pure ARM system against a pure Intel system, 85% of the power usage is independent of the processor instruction set and dependent purely on the process technology,"


        most of the power is used by lighting the screen
        NVIdia has DIDIM which reduces backlight power by 40%
        which is independent of process technology...

        and if it was just about process technology
        INtel would have NO need to acquire 3d patents or a 3d-chip company...

35.55+0.29(+0.82%)Aug 29 4:00 PMEDT