By Topic

An open question to developers of numerical software

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
Kahan, W. ; Math. & Electr. Eng. & Comput. Sci., California Univ., Berkeley, CA, USA ; Zuras, D.

IEEE 754 a standard for binary floating-point arithmetic has revolutionized the portability and reliability of programs that use binary floating-point arithmetic. Floating point is almost universally implemented with special-purpose hardware that tucks into a small corner of the CPU chip and runs in the hundreds of Mflops to Gflops range. Single-stepping through today's floating-point software to debug it often turns out to be futile. The concept of a NaN, standing for "not a number", evolved from an "indefinite" in Seymour Cray's CDC 6600. IEEE 754, by default, requires an untrapped "invalid operation", to signal itself by raising a flag and to deliver a NaN just when any other result, be it finite or infinite, would cause worse confusion. The NaN lets a program retain control unless the program or programmer directs its cancellation upon an invalid operation. Thus, a program conducting a search can return to the realm being searched after an accidental foray beyond a boundary whose existence and location were previously unknown. A sNaN differs from the other quiet NaNs by traooing any attempt to perform arithmetic upon it; then a trap-handler must interpret this sNaN.

Published in:

Computer  (Volume:38 ,  Issue: 5 )