You are here

Home » Content

A SINGLE/DOUBLE PRECISION FLOATING-POINT MULTIPLIER DESIGN FOR MULTIMEDIA APPLICATIONS

Journal Name:

  • Istanbul University Journal of Electrical & Electronics Engineering

Publication Year:

  • 2009

Key Words:

Author NameUniversity of AuthorFaculty of Author
Abstract (2. Language): 
Modern graphic processors, multimedia processors, and general-purpose processors with multimedia extensions provide SIMD floating-point instructions. SIMD floating-point multiplication is commonly used in 2D and 3D applications, which mostly use single precision floating-point operands. Consequently, efficient single precision multiplier units are crucial for high performance systems. This paper introduces a novel floating point multiplier that can perform either a double precision or two parallel single-precision floating-point multiplications. The proposed design uses approximately 10% more hardware and has 33% more delay compared to a conventional double precision floatingpoint multiplier.
Bookmark/Search this post with
FULL TEXT (PDF): 
PDF icon double-precision-floating-point-multiplier-design-multimedia-applications.pdf
  • 1
827-831
  • English

REFERENCES

References: 

[1] V. Lappalainen, T.D. Hämäläinen, P. Liuha,
“Overview of Research Efforts on Media ISA
Extensions and Their Usage in Video Coding”,
IEEE Transactions on Circuits And Systems For
Video Technology, Vol: 12, No: 8, 2002.
[2] G. Even, S.M. Mueller, P.M. Seidel, “A dual
mode IEEE multiplier Proceedings”, Second
Annual IEEE International Conference, pp. 282-
289, 1997.
[3] R.B. Lee, “Multimedia extensions for general
purpose processors, Signal Processing Systems,
1997”, SIPS 97 - Design and Implementation.,
IEEE Workshop, pp. 9-23, 1997.
[4] ANSI-IEEE Standard 754-1985: “IEEE
Standard for binary floating-point arithmetic”,
1985.
[5] Intel, “IA-32 Intel™ Architecture Software
Developer’s Manual”, http://download.intel.com/
design/Pentium4/manuals/25366520.pdf, 2006.
[6] Advanced Micro Devices, “AMD64
Architecture Programmers Manual”, White
Paper, http://www.amd.com/usen/assets/content
type/white_papers_and_tech_docs/26569.pdf,
2006.
[7] K. Diefendorff, P.K. Dubey, R. Hochsprung,
H. Scale, “AltiVec extension to PowerPC
accelerates media processing”, IEEE Micro, Vol:
20 Iss: 2, pp. 85-95, 2000.
[8] M.D. Jennings, T.M. Conte, “Subword
Extensions for Video Processing on Mobile
Systems”, IEEE Concurrency, Vol: 6, No: 3, pp.
13-16, 1998.
[9] M. Gök, S. Krithivasan, M.J. Schulte,
“Designs for Subword-Parallel Multiplications
and Dot Product Operations”, Workshop on
Application Specific Processors, pp. 27-31,
2004.
Metin Mete ÖZBĐLEN received his B.Sc. degree in
Electrics and Electronics from Gaziantep University in
Turkey in 1996, and his M.Sc in Electrics and
Electronics from Çukurova University in Adana,
Turkey in 2001. He is pursuing Ph.D. degree in
Electrics and Electronics at Çukurova University in
Adana. He has been a research assistant in Computer
Engineering Department in Mersin University since
2001. His current areas of research include computer
arithmetic and the design of IEEE compliant floatingpoint
units, multimedia acceleration algorithms.
Mustafa Gök received his B.Sc. degree in Electronics
Engineering from the Istanbul University, Istanbul in
1995, the M.Sc. degree in Electrical Engineering from
the Lehigh University, Bethlehem in 2000, and the
Ph.D. degree in Electrical Engineering from the
Lehigh University, Bethlehem in 2003. He is an
Assistant Professor in Cukurova University. His
research interests include computer arithmetic,
computer architecture, and design of hardware
accelerators.

Thank you for copying data from http://www.arastirmax.com