h2020_logo

  • Introduction                                                                                      

    GPU-WEAR is an Horizon 2020 Eh2020logoU-funded project into low-power graphics devices under the topic SME Instrument Phase II. The project has started on June 1, 2016 with a grant of € 1.34 M ($1.56) targeting to advance the technology readiness level and commercialization of the NEMA-ultra-low power graphics platform.

    NEMA GPU-WEAR is a novel platform to develop and implement ultra-low power, heterogeneous, multicore Graphics Processing Unit (GPU) technology, for SoCs (System on the Chip). The aim is to extend the battery life by decreasing the power-consumption by an order of magnitude for the next generation of mobile, wearable and embedded devices while running fast, graphically rich display and 4K video applications.

    Information and Specifications

    Dealing with the levels of power consumption required by wearables and Internet of Things (IoT) devices represents a major technological challenge that demands novel multi-disciplinary approaches spanning circuit, architecture, compiler, and API level optimization techniques. The main challenge in these new gpuwear_solutionsdevices is the battery life and how to extend it to more than one day without having to search for wall plugs and charging stations. Thus, ultra‐low power devices performing energy‐conscious graphics calculations are urgently needed in the market. GPU‐WEAR is designed to further refine and validate the first heterogeneous multicore embedded graphics processing unit (GPU), the associated low‐power software library, and the run‐time system explicitly optimized for the ultra‐low power requirements posed by the wearables/IoT market.

    The envisaged heterogeneous GPU includes two different types of general purpose (GP)‐GPU cores with different microarchitectures. The unique selling point of the project’s output is that both core types are enabled by a single and disruptive low‐power Instruction Set Architecture (ISA), thus relying on a single executable and software/compiler toolchain. In addition, GPU‐WEAR will release an integrated software environment, in the form of a Software Development Kit (SDK), referred to as GPU‐WEAR‐SDK, to streamline the development process of graphics and non-graphics applications.

  •  

    Think Silicon's GPUs Nema|t (Nema tiny)  and Nema|S (Nema SMALL) are based on a disruptive Very Long Instruction Word (VLIW), highly multithreaded microarchitecture powered by compamy’s proprietary vectorized, Single Instruction Multiple Data (SIMD) and low-power ISA. Think Silicon ISA encapsulates three types of parallelism (VLIW, SIMD, & Vector processing) – the key enabler for full hardware utilization, thus ultra-low power consumption. These GPUs can exploit the imbalance in graphics workloads by building a heterogeneous multicore GPU that can reduce power consumption by more than an order of magnitude.nema_logo

    To this end, as part of the project, the company will integrate the SMALL.GPU (based on Nema|S) with the tiny.GPU (based on Nema|t). This will create the first heterogeneous, multicore GPU in the market of wearable devices. While, Nema|t and Nema|S represent today the most competitive solutions in the field of tiny and SMALL GPUs respectively, their combination will inherit the well-known benefits of heterogeneous computing. As part of the project, Think Silicon is currently developing novel policies to best-balance graphics workloads between Nema|S and Nema|t in the most energy-efficient manner and novel compression techniques in all GPU levels (geometry, pixel, framebuffer, and texture compression) to minimize memory requirements and bandwidth.

  • GPU-WEAR formulates a holistic approach (at the hardware, software and API level) in reducing the power consumption of embedded/mobile GPUs. The project has the following technical objectives:

    • Realize a family of novel heterogeneous embedded GPUs referred to as Nema|tS
    • Investigate and establish effectiveness of power reduction techniques in a holistic way
    • Develop a transparent to the programmer, display-aware, QoS-aware graphics library referred to as GPU-WEAR-LIB
    • Optimize Nema|tS GPU for GPGPU soft real-time applications
    • Release GPU-WEAR-SDK to the Open Community and support its adoption
    • Set foundations for commercial exploitation
  • New Products supported by the GPU-WEAR project

     

    NEMA|p

    NEMA|p is an ultra-low-power and ultra-low gate count 2D Graphics Processing Unit suitable for the Wearables and the broader IoT market. NEMA|p integrates Graphics hardware accelerators and a multi-threaded processing Unit based on a powerful VLIW instruction set.

    More information about NEMA|p can be found here

    Press release  adobe-pdf-document-icon 

    Product brief adobe-pdf-document-icon

     

    NEMA|GFX-API

    NEMA|GFX-API is a low level library that interfaces directly with the Nema GPUs and provides a software abstraction layer to organize and employ drawing commands with ease and efficiency. The target of NEMA|GFX-API is to be able to be used as a back-end to existing APIs (such as OpenGL, DirectFB or any proprietary one) but also to expose higher level drawing functions, so as to be used as a standalone Graphics API.

    More information about NEMA|GFX-API can be found here

    Press release  adobe-pdf-document-icon 

    User Manual  adobe-pdf-document-icon 

    NemaGFX API review by John Peddie’s TechWatch adobe-pdf-document-icon

     

    NEMA|Bits

    NEMA|Bits is an out-of-box software environment designed to showcase and validate the graphics technology developed by Think Silicon S.A. The purpose of NEMA|Bits is to act as a vertical (including both hardware and software components) demonstration platform in which the interested users and customers can download the latest developments of the company in their Zynq FPGA platforms.

    More information and the link to download the tool and the user manual can be found here

    Press release adobe-pdf-document-icon

     

    NEMA|PIX-Presso

    NEMA|PIX-Presso image converter is a utility for converting images to formats suitable for low power embedded devices. The purpose of NEMA|PIX-Presso is to act as an easy to use companion for GPUs developers in order to adapt images to applications requirements.

    More information and the link to download the tool and the user manual can be found here

    Press release adobe-pdf-document-icon

     

    NEMA|GUI-Builder

    NEMA|GUI-Builder is a friendly user interface with drag-and-drop capabilities, that offers a development environment on host PCs to reduce GUI build development time. NEMA|GUI-Builder is build on top of NEMA|GFX-API and supports power-aware code generation.

    More information can be found here

    Press release adobe-pdf-document-icon

    Product brief adobe-pdf-document-icon

     

    NEMA|SHADER-Edit

    NEMA|SHADER-Edit is a graphical user interface (GUI) for the NEMA|tS Toolchain. It is an offline editor/compiler for vertex and fragment shaders written in OpenGL Shading Language (GLSL).

    More information can be found here

    User Manual adobe-pdf-document-icon

     

    Under-Development Products supported by the GPU-WEAR project

     

    NEMA|tS

    The first heterogeneous GPU in the mobile market.

     

    NEMA|dpu

    The NEMA|dpu is the next step in the Think Silicon display controller family.

  • D6.1: Initial Press Release

    The purpose of this initial version of the press release is to increase public awareness regarding the scope, the underlying technology, and the overall research that takes place in the project.

    Download D6.1 adobe-pdf-document-icon

     

    D6.7: First Collateral Supporting Product Soft-launch

    This deliverable should be seen as a live document describing the progress and the current status of the project-wide activities for supporting the product soft-launch of the GPU-WEAR project. As such, this deliverable is not associated to a specific task or tasks, but it is dedicated to document the GPU-WEAR open platform that will allow third-party developers to develop/contribute and further extend the GPU-WEAR initiative.

    Download D6.7 adobe-pdf-document-icon

     

    D6.8: Second Collateral Supporting Product Soft-launch

    Deliverable D6.8 is the second version of D6.7 and describes the new tools released or updated during the second period of the GPU-WEAR project.

    Download D6.8 adobe-pdf-document-icon