Implementation of float-float operators on graphics hardware
Dali, LP2A, Universite de Perpignan, 52 Avenue Paul Alduy, 66860 Perpignan Cedex, France
arXiv:cs/0603115 [cs.AR] (29 Mar 2006)
@article{da2006implementation,
title={Implementation of float-float operators on graphics hardware},
author={Da Gra{c{c}}ca, G. and Defour, D.},
journal={Arxiv preprint cs/0603115},
year={2006}
}
The Graphic Processing Unit (GPU) has evolved into a powerful and flexible processor. The latest graphic processors provide fully programmable vertex and pixel processing units that support vector operations up to single floating-point precision. This computational power is now being used for general-purpose computations. However, some applications require higher precision than single precision. This paper describes the emulation of a 44-bit floating-point number format and its corresponding operations. An implementation is presented along with performance and accuracy results.
November 13, 2010 by hgpu
