Abstract
The problem of obtaining optimal starting values for the calculation of square root using Newton-Raphson's Method is considered. This paper presents the best starting values theory in order to optimize the maximum absolute error after a given number of iterations. Two different methods are shown, and a third, which can be considered as a mixture of the previous two, is briefly discussed. The approach combines analytical and numerical methodologies, which gives more interesting results on the main characteristics of the behavior of the absolute error for different initializations. A comparison table between the traditional optimal relative error results and the absolute error ones is provided.
Zusammenfassung
Wir betrachten das Problem, optimale Startwerte für die Berechnung von Quadratwurzeln bei Verwendung der Newton-Raphson Methode zu erhalten.
In dieser Arbeit wird die Theorie zur Wahl der besten Startwerte in Hinblick auf die Optimirung des maximalen absoluten Fehlers nach einer vorgegebenen Anzahl von Iterationen dargestellt. Es werden zwei verschiedene Methoden gezeigt und eine dritte, die als Mischung dieser beiden Methoden angesehen werden kann, betrachtet. Eine Vergleichstabelle zwischen den herkömmlichen Ergebnissen bezüglich des optimalen relativen Fehlers und des absoluten Fehlers wird angegeben.
Similar content being viewed by others
References
E. H. Kaufman, G. D. Taylor: Uniform rational approximation of functions of several variables, In'l J. for Numerical Methods in Eng., Vol. 9, pp. 297–323 (1975).
IEEE Standard for Binary Floating-Point Arithmetic, IEEE Standard 754, IEEE Computer Society (1985).
P. Montuschi, M. Mezzalama: Survey of square rooting algorithms, IEEE Proceedings Part E,137, 31–40 (1990).
S. Majerski: Square-rooting algorithms for high-speed digital circuits, IEEE Trans. Comput.,C 34, 724–733 (1985).
P. Montuschi, L. Ciminiera: On the efficient implementation of higher radix square root algorithms, IEEE 9th Symposium on Computer Arithmetic, Santa Monica, CA (1989), pp. 154–161.
V. G. Oklobdzija, M. D. Ercegovac: An on-line square root algorithm, IEEE Trans. Comput.,C 31, 70–75 (1982).
H. Peng: Algorithms for extracting square roots and cube roots, Proc. 5th IEEE Symposium on Computer Arithmetic, Ann Arbor, Michigan, pp. 121–126 (1981).
G. S. Taylor: Compatible hardware for division and square root, Proc. 5th IEEE Symposium on Computer Arithmetic, Ann Arbor, Michigan, pp. 127–134 (1981).
R. W. Southworth, S. L. Deleeuw: Digital computation and numerical methods, McGraw-Hill, New York (1965).
J. E. Volder: The CORDIC Trigonometric Computing Technique, IRE Trans. Electron. Comput.,EC 8, 330–334 (1959).
B. G. DeLugish: A class of algorithms for automatic evaluation of certain elementary functions in a binary computer, PhD Dissertation, Dept. of Computer Science, University of Illinois, Urbana (1970).
T. C. Chen: The automatic computation of exponentials, logarithms, ratios and square roots, IBM Research Report RJ 1970, p. 32 (1972).
M. Andrews: Mathematical microprocessor software: a\(\sqrt x \) comparison, IEEE Micro,2, 63–75 (1982).
D. G. Morsund: Optimal starting values for newton-raphson calculation of\(\sqrt x \), Comm. of the ACM,10, 430–432 (1967).
I. Ninomiya: Best rational starting approximations and improved newton iteration for the square root, Math. Computation,24 391–404 (1970).
D. L. Phillips: Generalized logarithmic error and newton's method for them-th root, Math. Computation,24, 383–389 (1970).
P. H. Sterbenz, C. T. Fike: Optimal starting approximations for newton's method, Math. Computation,23, 313–318 (1969).
M. W. Wilson: Optimal starting approximations for generating square root for slow or no divide, Comm. of the ACM,13, 559–560, (1970).
C. T. Fike: Starting approximations for square root calculation on IBM System 360, Comm. of the ACM,9, 297–299 (1966).
J. E. Dennis, Jr., R. B. Schnabel: Numerical methods for unconstrained optimization and nonlinear equations, Prentice-Hall, Inc., Englewood Cliffs, N.J. (1981).
W. Fraser, J. F. Hart: On the computation of the rational approximation to continuous functions, Comm. of the ACM,5, 401–404 (1962).
P. Montuschi: Architetture di Elaborazioner per Applicazioni Speciali, PhD Dissertation (1989), I.R. DAI/ARC 1-89, Politecnico di Torino.
Digital Equipment Co., VAX/VMS V.3.0 Mathematical Library (1982).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Montuschi, P., Mezzalama, M. Optimal absolute error starting values for Newton-Raphson calculation of square root. Computing 46, 67–86 (1991). https://doi.org/10.1007/BF02239012
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02239012