Floating-point computation with just enough accuracy

Hank Dietz, Bill Dieter, Randy Fisher, Kungyen Chang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

8 Scopus citations


Most mathematical formulae are defined in terms of operations on real numbers, but computers can only operate on numeric values with finite precision and range. Using floating-point values as real numbers does not clearly identify the precision with which each value must be represented. Too little precision yields inaccurate results; too much wastes computational resources. The popularity of multimedia applications has made fast hardware support for low-precision floating-point arithmetic common in Digital Signal Processors (DSPs), SIMD Within A Register (SWAR) instruction set extensions for general purpose processors, and in Graphics Processing Units (GPUs). In this paper, we describe a simple approach by which the speed of these low-precision operations can be speculatively employed to meet user-specified accuracy constraints. Where the native precision(s) yield insufficient accuracy, a simple technique is used to efficiently synthesize enhanced precision using pairs of native values.

Original languageEnglish
Title of host publicationComputational Science - ICCS 2006
Subtitle of host publication6th International Conference, Proceedings
Number of pages8
StatePublished - 2006
EventICCS 2006: 6th International Conference on Computational Science - Reading, United Kingdom
Duration: May 28 2006May 31 2006

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3991 LNCS - I
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


ConferenceICCS 2006: 6th International Conference on Computational Science
Country/TerritoryUnited Kingdom

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Floating-point computation with just enough accuracy'. Together they form a unique fingerprint.

Cite this