Journal of the Institute of Electronics Engineers of Korea SD (대한전자공학회논문지SD)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지

Implementation of a 3D Graphics Hardwired T&L Accelerator based on a SoC Platform for a Mobile System

SoC 플랫폼 기반 모바일용 3차원 그래픽 Hardwired T&L Accelerator 구현

  • Published : 2007.09.25
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we proposed an effective T&L(Transform & Lighting) Processor architecture for a real time 3D graphics acceleration SoC(System on a Chip) in a mobile system. We designed Floating point arithmetic IPs for a T&L processor. And we verified IPs using a SoC Platform. Designed T&L Processor consists of 24 bit floating point data format and 16 bit fixed point data format, and supports the pipeline keeping the balance between Transform process and Lighting process using a parallel computation of 3D graphics. The delay of pipeline processing only Transform operation is almost same as the delay processing both Transform operation and Lighting operation. Designed T&L Processor is implemented and verified using a SoC Platform. The T&L Processor operates at 80MHz frequency in Xilinx-Virtex4 FPGA. The processing speed is measured at the rate of 20M Vertexes/sec.

본 논문에서는 휴대 정보기기 시스템에서 더욱 향상된 실시간 3D 그래픽 가속 능력을 갖는 SoC(System on a Chip) 구현을 위해 효과적인 T&L(Transform & Lighting) Processor 구조를 연구하였다. T&L 과정에 필요한 IP들을 설계하였으며, 이를 바탕으로 SoC Platform 기반으로 검증하였다. 설계된 T&L Processor는 24 bits 부동소수점 형식과 16 bits 고정소수점 형식을 적절하게 혼용하고 계산식의 병렬성을 최대한 활용하여 Transform 과정 연산과 Lighting 과정 연산의 지연시간을 균일하게 배분하여 Transform 과정만 처리할 때와 Lighting과 혼용으로 처리할 때 연산 속도의 차이가 없이 동작이 가능하다. 설계된 T&L Processor는 SoC 플랫폼을 이용하여 성능 측정 실험 및 검증을 하였고, Xilinx-Virtex4 FPGA에서 80 MHz의 동작 주파수를 확인하였고 초당 20M개의 정점(Vertex) 처리 성능을 확인하였다.

Keywords

References

  1. David H. Eberly, '3D Game Engine Design,' Morgan Kaufmann, May, 2001
  2. L. Garber, 'The wild world of 3D graphics chips,' IEEE Computer, vol. 33, no. 9, pp. 12-16, Sep. 2000
  3. Udo Flohr, '3-D for Everyone,' Byte, pp.76-88, Oct. 1996
  4. C. B. Harell, F. Fouladi, 'Graphics Rendering Architecture for a High Performance Desktop Workstation,' Proceeding of SIGGRAPH '93, pp. 93-99, 1993
  5. N. Trevett, 'GLINT Gamma: A 3D Geometry and Lighting Processor for the PC,' Proceeding Notebook for HOT Chips IX, pp. 235-246, 1997
  6. Jun-Hee Lee 'Exploiting Parallelism of 3D Graphics Geometry using VLIW Geometry Processor,' KAIST, Master Thesis, 1999
  7. Cheol-Ho Jeong, 'Design of an Effective Control and Execution Method for Geometry Engines and Rasterizers within Embedded 3D Graphics Accelerators,' Yonsei Univ., PhD Thesis, 2003
  8. Asger Munk Nielsen, David W.Matula, C.N.Lyu, and Guy Even, 'An IEEE compliant floating-point adder that conforms with the pipelined Packet-Forwarding Paradigm,' IEEE Transactions on Computers, vol.49, no.1, January 2000
  9. J. C. Jeong, W. C. Park, W. Jeong, T. D. Han, M. K Lee 'A Cost-Effective Pipelined Divider with a Small Lookup Table', IEEE Transactions on Computers, 2004 , pp489-495
  10. P. Hung, H. Fahmy, O. Mencer and M.]. Flynn, 'Fast division algorithm with a small lookup table,' Conference Record of the Thirty-Third Asilomar Conference on Signals, Systems and Computers, Vol. 2, pp. 1465 -1468, May 1999
  11. Won-Suk Kim, 'The Implementation of Geometry Accelerator Simulator for 3D Graphic Accelerator Hardware Design,' Yonsei Univ., Master Thesis, 2003
  12. Behrooz Parhami, 'Computer Arithmetic Algorithms and Hardware Design,' Oxford University Press, pp. 128-211, 2000
  13. Israel Koren, 'Computer Arithmetic Algorithms,' John Wiley & Sons. Inc, pp. 35-40, 1978
  14. IEEE Std 754-1985, 'IEEE standard for binary floating-point arithmetic,' IEEE, 1985
  15. nVidia, 'Technical briefs: An in-depth look at Geforce3 features,' nVidia Corporation, http://www.nvidia.com/Products/GeForce3.nsf/tec hnical.html
  16. nVidia, 'Transform, lighting and rasterization system embodied on a single semiconductor platform,' nVidia Patent, Dec. 1999