J4 ›› 2015, Vol. 42 ›› Issue (6): 124-130.doi: 10.3969/j.issn.1001-2400.2015.06.022

• Original Articles • Previous Articles     Next Articles

Design and implementation of the graphics accelerator oriented to OpenGL

DENG Junyong1;LI Tao1;JIANG Lin1;HAN Jungang1;SHEN Xubang2   

  1. (1. School of Electronic Engineering, Xi'an Univ. of Posts ＆ Telecommunications, Xi'an  710121, China;
    2. School of Microelectronic, Xidian Univ., Xi'an  710071, China)
  • Received:2014-09-18 Online:2015-12-20 Published:2016-01-25
  • Contact: DENG Junyong E-mail:djy@xupt.edu.cn

PDF (PC)

576

Like

Cited

3

Abstract:

In order to explore the self-development of the graphics processing unit, this paper presents the design of a graphics accelerator which utilizes the classical pipelined structure and supports OpenGL primary commands. The matrix computational unit, which is the critical operation of geometric transformation, projection transformation and viewport transformation, primitive assembly unit and rasterization unit have been optimized to improve the efficiency. The accelerator realizes the rendering effects of Gouraud shading, lighting, full-screen anti-aliasing and texture mapping. In order to verify the circuit, the software and hardware simulation workbench based on C/C++ and SystemVerilog respectively have been established. The prototype system is implemented on EP2C70F896C6 of Altera which takes up to 83％ of the FPGA resource, with the speed being up to 100MHz. The test results of plenty of rendering applications show that the accelerator possesses the capability of graphics rendering.

Key words: graphics accelerator, matrix computation, primitive assembly, full-screen anti-aliasing, prototype system

CLC Number: 

  • TP302

Copyright © 2018 Journal of Xidian University, All Rights Reserved.
Powered by Beijing Magtech Co. Ltd