Relative error in floating-point multiplication

DOI10.1007/BF02260500MaRDI QIDQ1058264zbMATH OpenFDO

Authors R. H. Goodman, Alan Feldstein, J. Bustoz

Publication date 1985

Published in Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/BF02260500

zbMATH Keywords

computer arithmeticfloating point multiplicationfloating point numbersfloating point precision and significancefraction errorguard digitslogarithmically distributed numbersmean and standard deviation of errorsnormalization optionsrelative errorround-off error

Mathematics Subject Classification ID

Roundoff error (65G50)

Review text

A model of the relative error in floating point multiplication is developed and is analyzed stochastically for various choices of computer design parameters. These parameters include the base, the type of rounding rule, the number of guard digits, and whether the post- arithmetic normalization shift (if needed) is done before or after rounding. Under the assumption of logarithmic distribution for the fraction (mantissa), the major stochastic conclusions are: 1. The average relative error in multiplication increases as the base increases. 2. This error is minimized by selecting the machine based to be binary (better yet, binary with a hidden bit) and is rather large for machines with base 16. 3. The classical relative error bounds are pessimistic. The average overestimation by those bounds increases as the base increases.

Recommendations

• Some models of relative error in products
• On relative errors of floating-point operations: optimal bounds and applications
• On roundoff error distributions in floating point and logarithmic arithmetic
• scientific article; zbMATH DE number 872251
• Improved estimates for error in floating point representation analysis

Cites work

• Convergence Estimates for the Distribution of Trailing Digits
• Effect of guard digits and normalization options on floating point multiplication
• scientific article; zbMATH DE number 3182507 (Why is no real title available?)
• Improved Trailing Digits Estimates Applied to Optimal Computer Arithmetic
• On Local Roundoff Errors in Floating-Point Arithmetic
• On the distributions of significant digits and roundoff errors
• Some models of error in floating point multiplication
• The Distribution of Leading Significant Digits

Cited in

(12)

• The relative accuracy of \(\mathtt{(x+y)*(x-y)}\)
• On the definition of unit roundoff
• Reducing the influence of tiny normwise relative errors on performance profiles
• scientific article; zbMATH DE number 3864502 (Why is no real title available?)
• Overflow, Underflow, and Severe Loss of Significance in Floating-Point Addition and Subtraction
• scientific article; zbMATH DE number 1086539 (Why is no real title available?)
• On relative errors of floating-point operations: optimal bounds and applications
• Algebras over floating point numbers
• Improved estimates for error in floating point representation analysis
• Number shortening algorithms
• On the maximum relative error when computing integer powers by iterated multiplications in floating-point arithmetic
• Some models of relative error in products

This page was built for publication: Relative error in floating-point multiplication

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1058264)