NEMA®|t is the industry’s smallest Internet-of-Things (IoT) GPU with 3D functionality. NEMA®|t features OpenGL® ES capabilities, implements a fully configurable and programmable 3D graphics rendering engine, accelerates a comprehensive super-set of 2D graphics drawings, and has smart composition functions.
NEMA®|t is, designed to support midrange till higher level quality wearable and IoT devices sporting an SoC with 32-bit MCU or MPU (e.g. ARM® Cortex® M and A processors).
With NEMA®|t, you are able to create compelling 3D Graphical User Interfaces (GUIs) and software applications with ultra-long battery life or lower power consumption at a significantly lower cost for power–memory–area constrained IoT devices.
NEMA®|t is a modular architecture and is available in one, two or four-core configurations. Each flavor includes one additional fully programmable vertex processor in addition to the fragment processor cores. All cores utilize a fixed-point data path and custom VLIW Instruction Set Architecture (ISA) in a heavily multi-threaded execution pipeline which is optimized for extracting the maximum performance in 3D acceleration in a power and silicon area constrained system.
NEMA®|t connects to the rest of the system over an AHB or AXI bus and features an optional, smart IOMMU that eliminates unnecessary data traffic between host CPU and the GPU.
Fully programmable engine with a VLIW instruction set
Scalable to multiple cores
DMA based Command lists to minimize CPU overhead
TSCTM4 (4 bits per pixel)
TSCTM6 / TSCTM6a (6 bits per pixel without/with alpha)
Support for OpenGL® ES
Perspective correct image projection
Z-Buffer (Early test, Late test)
Pixel / Line drawing
Triangles (Gouraud Shaded)
One pass composition
Hidden areas not read from memory
YUV layers are automatically processed without intermediate conversion
OpenGL®|ES library provides OpenGL|ES 2.0 API for graphics applications. It uses NEMA®|t innovative architecture to accelerate rendering operations but it also comes with a built-in SW renderer to provide full OpenGL|ES functionality.
OpenGL®|ES library is implemented on top of a subset of Vulkan API, necessary to support OpenGL®|ES features. Vulkan API is utilized only as the basis layer underneath GL and cannot be used as a standalone API for graphics applications.
Beyond OpenGL®|ES, NEMA®|t supports:
NEMA®|GFX enabling high quality 2.5D graphics on systems running RTOS or even on OS-less systems.
DirectFB enabling software development with application and Graphic User Interface (GUI) creation frameworks, such as Qt and GTK.
Software package comes together
NEMA®|SHADER-Edit is a shader editing tool that enables users to rapidly create and compile GLSL shaders offline. Compilation framework incorporates both NEMA®|t and ARM back ends, automatically choosing whether the shader can be executed by NEMA®|t or SW renderer and producing the appropriate executable.
NEMA®|GUI-Builder, a graphical cross-platform software framework enabling rapid high-end Graphics User Interface (GUI) development on low resource hardware
NEMA®|PIX-Presso, a utility software for converting images to/from formats suitable for low power embedded devices
Configuration/ Integration/ Verification
NEMA®|t can be customized at design-time choosing from a number of parameters that enable or disable several features of the design. This is done using NEMA®|t GUI Configurator, a Graphical User Interface Configurator Tool.
The NEMA®|t GPU IP Platform is available in Verilog and easy to integrate and verify. NEMA®|t ASIC reference designs have been evaluated in various process technologies.
NEMA®|t is designed with AMBA interfaces.
The core has been verified through extensive simulation and rigorous code coverage measurements. It comes together with a complete verification suite that compares reference images with rendered images.
Think Silicom® S.A ist ein Haftungsbeschränktes privat geführtes Unternehmen, mit Hautpsitz in Patras/Athen Griechenland sowie Vertriebs & Support Niederlassungen in Toronto/Kanada, Köln/EMEA, San Jose/USA, Tokyo/Japan und Taipei/Taiwan. Think Silicon ist spezialisiert in die Entwicklung von extrem energiesparenden Microprocessortechnology für Grafikanwendungen und auf künstlicher Intelligenz basierenden Computer Vision Amwendungen für Mobile (Smartphone) Wearables (Smartwatch), Embedded Systems (Industrie) und IoT/Edge Endgeräte mit Bildschirm und/oder einer Kamera. Think Silicon lizenziert seine Technologie an führende sog. „Fabless Microprocessor“ Unternehmen weltweit.