Since the invention of floating point arithmetic for computers, real number calculations have produced unreported errors, sometimes catastrophically.
Bounded Floating Point is a patent pending method of performing operations on floating numbers that maintains boundaries for the real numbers they represent. This is accomplished by adding a bound field to the floating point representation of real numbers. Implemented in hardware, this circuit provides real-time, fail-safe representation of real numbers.
For information about the availability of this patent for licensing or acquisition, please contact
Alan at BoundedFloatingPoint.com
(Please convert to a proper email address.)
History of floating point error and the need for bounded floating point.
Functional description of bounded floating point processor with diagrams.
Abstract from the patent application.