US20020116431
System and method for improving the accuracy of reciprocal operations performed by a floating-point unit
System and method for improving the accuracy of reciprocal operations performed by a floating-point unit