Loading...
 Start Page
 

2.5D Smartphone-like GUIs and "gaming" graphics speeds on MCU Class SoCs

 
DESCRIPTION

NEMA® | pico XL Series is the only no-compromise graphics solution offering SoC-optimized GPU IP bundled with a feature-complete design and targeting the most competitive PPA for mid to high-end wearables and embedded IoT display products. This holistic approach aims to reduce the overall system power by minimizing memory and display access without sacrificing graphics quality and performance.

MARKETS & APPLICATIONS

BIZ Class - Small to Midsized Displays (1,5" - 10")
LE Entertainment | Home Control | Appliances | Wearables | IoT | Embedded | Data Centers | Video Overlay | Instrument Cluster

XL 1000
XL 2000
XL 4000
PICO XL
PERFORMANCE
XL 1000
XL 2000
XL 4000
 
 
GPU
2.5D
2.5D
2.5D
 
 
Cores
1
2
4
 
 
Area (mm) (@ 28nm, 100MHz, TSMC 28HPC+)
0.21
0.37
0.71
 
 
Core Clock Frequency @28nm TSMC HPC+
500+ MHz
500+ MHz
500+ MHz
 
 
Core Clock Frequency @7nm TSMC HPC+
Up to 700 MHz
Up to 700 MHz
Up to 700 MHz
 
 
Threads
32 -256
32 -256
32 -256
 
 
Characterization Configuration
64 Threads AXI4:64
64 Threads AXI4:64
64 Threads AXI4:64
 
 
Colors
16.7 Million (24-bit)1
- 1+ Trillion (30-bit)2
16.7 Million (24-bit)1
- 1+ Trillion (30-bit)2
16.7 Million (24-bit)1
- 1+ Trillion (30-bit)2
 
 
Texture Read Performance @100MHz
100 Mpixel/sec
200 Mpixel/sec
400 Mpixel/sec
 
 
Frame Buffer Write @100MHz
100 Mpixel/sec
200 Mpixel/sec
400 Mpixel/sec
 
 
Frames per Second (Mickey Mouse® watch face)
70
120
130
 
 
Power Consumption
2 mW
4 mW
8 mW
 
 
CPU Utilization Typical
<5%
<5%
<5%
 
 
Resolution
Up to 1024 x 756, 4K Video Overlay

 

 

1Default

2Video Overlay Extension

MORE FEATURES – MORE FLEXIBILITY – NO COMPROMISE

The NEMA® | pico XL Series is a scalable and flexible multi-core GPU IP platform, which meets the rigid requirements of memory and power-limited SoCs with the need for the latest and greatest GPU hardware in MCU class SoCs (ARM® M class, ARC® EM5D, RISC-V). With clock frequencies as low as 100MHz (not limited), NEMA® | pico XL renders natively fluent and stunning 2.5D graphics.

The modular IP building block system allows to add additional extensions to perform Vector Graphics3 functions and HD Video Overlay2 making it a perfect candidate for higher-end embedded display and video applications. With the ability to run on bare metal / RTOS requiring only small on-chip memory and system resources, NEMA® | pico XL GPU-Series saves cost and is a GPU power house in its own class. The software driver stack, SDK and toolkit “NEMA® | GFX”, provides an easy to use programming environment with aggressive power savings, intelligent composition, rendering and bandwidth modifications to build a rich and ultra-low power graphical user experience.

2Video Overlay Extension

3VG Extension

ARCHITECTURE

NEMA® | pico XL Series has been designed for graphics efficiency in ultra-compact silicon area. Its fixed-point data path and instruction set architecture (ISA) are tailored to GUIs acceleration and small display applications leading to substantial improvements in power consumption and silicon area. NEMA® | pico XL Series microarchitecture is based on a lean version of NEMA ISA and it combines hardware-level support for multi-threading, VLIW and low-level vector processing in the most power efficient way.

INTEGRATION / VERIFICATION

A fast 2D GUI interface and the NEMA® | pico XL 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® | p GUI Configurator, a Graphical User Interface Configurator Tool. The NEMA® | pico XL GPU IP Platform is available in Verilog and is easy to integrate and verify.

NEMA® | pico XL ASIC reference designs have been verified in various process technologies. NEMA® | pico XL 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 ensures correct migration to target technology nodes.

SOFTWARE SUPPORT

The NEMA® | GFX library, which is available in pure ANSI C with no dependencies, allows easy portability to systems running with or without an RTOS. The software package includes OS drivers for Linux and Software Libraries for supporting 2D Graphics APIs such as DirectFB and Qt.

The software package includes Linux drivers, Software Libraries for 2D Graphics APIs and comes together with:

  • NEMA® | GUI-Builder (non-commercial version), a graphical cross-platform software framework enabling rapid high-end Graphics User Interface (GUI) development on low resource hardware.
  • NEMA® | PIX-Presso (non-commercial version), a utility software for converting images to/from formats suitable for low power embedded devices.
  • NEMA® | Bits, an EVK for technology evaluation and pre-silicon application development.
FEATURES

HARDWARE COMPONENTS

  • Programmable Shader engine with a VLIW instruction set
  • Command list based DMAs to minimize CPU overhead
  • Primitive Rasterizer
  • Texture Mapping Unit
  • Blending Unit
  • Display Controller (optional)

IMAGE TRANSFORMATION

  • Texture mapping
  • Point sampling
  • Bilinear filtering
  • Blit support
  • Rotation any angle
  • Mirroring
  • Stretch (independently on x and y axis)
  • Source and/or destination color keying
  • Format conversions on the fly
  • 2.5D Perspective Correct Projections

BLENDING SUPPORT

  • Fully Programmable Alpha blending modes (source and destination)
  • Source/Destination color keying

TEXT RENDERING SUPPORTS

  • Bitmap antialiased (A1 | A2 | A4 | A8 )
  • Font Kerning
  • Unicode (UTF8)

DRAWING PRIMITIVES

  • Pixel / Line drawing
  • Filled rectangles
  • Triangles (Gouraud Shaded)
  • Quadrilateral

COLOR FORMATS

  • 32-bit RGBA8888 | BGRA8888 | ABGR8888
  • 24-bit RGB
  • 16-bit RGBA5551 | RGB565
  • 8-bit Α8 | L8 | RGB332
  • 4-bit A4 | L4
  • 2-bit A2 | L2
  • 1-bit A1 | L1
  • YUV Read Only
  • TSCTM*4 (4 bits per pixel) (optional)
  • TSCTM6 | TSCTM6a (6 bits per pixel w/o alpha) (optional)

* Think Silicon Compression (proprietary)

ANTIALISING

  • 8x MSAA
  • Quadrilaterals per edge
  • Triangles per edge
  • Antialiased Thick lines
  • Antialiased Circles

 

Copyright © 2021 - 2024 Think Silicon | ULTRA-LOW POWER GPUs