2.44-GFLOPS 300-MHz floating-point vector-processing unit for high-performance 3-D graphics computing

被引:32
作者
Ide, N [1 ]
Hirano, M
Endo, Y
Yoshioka, S
Murakami, H
Kunimatsu, A
Sato, T
Kamei, T
Okada, T
Suzuoki, M
机构
[1] Toshiba Corp, Syst LSI Res & Dev Ctr, Kawasaki, Kanagawa 2108520, Japan
[2] Kyushu Inst Technol, Iizuka, Fukuoka 820, Japan
[3] Sony Comp Entertainment Inc, Tokyo 1070052, Japan
关键词
D O I
10.1109/4.848212
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
A vector unit for high-performance three-dimensional graphics computing has been developed, We implement four floating-point multiply-accumulate units, which execute multiply-add operations with one throughput; one floating-point divide/square root unit, which executes division and square-root operations with six cycles at 300 MHz; and one vector general-purpose register file, which has 128 bits x 32 words, The parallel execution of all units delivers a peak performance of 2.44 GFLOPS at 300 MHz.
引用
收藏
页码:1025 / 1033
页数:9
相关论文
共 9 条
  • [1] HEIKES C, 1996, P IEEE ISSCC FEB
  • [2] IDE N, 1999, P EUR SOL STAT CIRC, P106
  • [3] *INT CORP, 1999, INT ARCH OPT REF MAN, pD1
  • [4] Kamei T., 2000, Proceedings ASP-DAC 2000. Asia and South Pacific Design Automation Conference 2000 with EDA TechnoFair 2000. (Cat. No.00EX389), P635, DOI 10.1109/ASPDAC.2000.835177
  • [5] KUNIMATSU A, 1999, HOT CHIPS 11 C REC A, P71
  • [6] KUTARAGI K, 1999, P IEEE ISSCC FEB, P256
  • [7] DESIGN OF THE IBM RISC SYSTEM-6000 FLOATING-POINT EXECUTION UNIT
    MONTOYE, RK
    HOKENEK, E
    RUNYON, SL
    [J]. IBM JOURNAL OF RESEARCH AND DEVELOPMENT, 1990, 34 (01) : 59 - 70
  • [8] RAAM FM, 1999, ISSCC FEB, P258
  • [9] Leading-zero anticipatory logic for high-speed floating point addition
    Suzuki, H
    Morinaka, H
    Makino, H
    Nakase, Y
    Mashiko, K
    Sumi, T
    [J]. IEEE JOURNAL OF SOLID-STATE CIRCUITS, 1996, 31 (08) : 1157 - 1164