Some issues related to double roundings - ENS de Lyon - École normale supérieure de Lyon
Reports Year : 2011

Some issues related to double roundings

Abstract

Double rounding is a phenomenon that may occur when different floating-point precisions are available on a same system, or when performing scaled operations whose final result is subnormal. Although double rounding is, in general, innocuous, it may change the behavior of some useful small floating-point algorithms. We analyze the potential influence of double roundings on the Fast2Sum and 2Sum algorithms, on some summation algorithms, and Veltkamp's splitting. We also show how to handle possible double roundings when performing scaled Newton-Raphson division iterations (to avoid possible underflow problems).
Fichier principal
Vignette du fichier
SumsDoubleRoundings.pdf (338.41 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

ensl-00644408 , version 1 (24-11-2011)
ensl-00644408 , version 2 (02-12-2011)
ensl-00644408 , version 3 (08-07-2013)

Identifiers

  • HAL Id : ensl-00644408 , version 2

Cite

Érik Martin-Dorel, Guillaume Melquiond, Jean-Michel Muller. Some issues related to double roundings. 2011, pp.42. ⟨ensl-00644408v2⟩
839 View
455 Download

Share

More