DocumentCode
1996823
Title
16-bit floating point instructions for embedded multimedia applications
Author
Lacassagne, L. ; Etiemble, D. ; Kablia, S. A Ould
Author_Institution
Inst. d´´Electronique Fondamentale, Univ. Paris Sud, Orsay, France
fYear
2005
fDate
4-6 July 2005
Firstpage
198
Lastpage
203
Abstract
We have simulated the implementation of 16-bit floating point instructions on a Pentium4 and PowerPC G4 and G5 to evaluate the performance impact of these instructions in embedded processors for graphics and multimedia applications. Both accuracy of the computations and the execution time have been considered. For low-end embedded processors, the 16-bit FP instructions deliver a larger dynamic range than 16-bit integer with the same memory footprint. For high-end embedded processors, we add the speed up coming from wider SIMD instructions.
Keywords
embedded systems; floating point arithmetic; instruction sets; microprocessor chips; multimedia systems; 16 bit; 16-bit floating point instructions; Pentium4; PowerPC G4; computer graphics; embedded multimedia system; embedded processors; Computational modeling; Digital signal processing; Dynamic range; Graphics; Image processing; Image storage; Instruction sets; Microprocessors; Pixel; Visual effects;
fLanguage
English
Publisher
ieee
Conference_Titel
Computer Architecture for Machine Perception, 2005. CAMP 2005. Proceedings. Seventh International Workshop on
Print_ISBN
0-7695-2255-6
Type
conf
DOI
10.1109/CAMP.2005.1
Filename
1508186
Link To Document