Object Oriented Software Engineering   View all facts   Glossary   Help

subject > problem > defect > defect in a numerical algorithm

defect in a numerical algorithm comparison table

Subject	have testing strategy	occur	force
assuming a floating point value will be exactly equal to some other value	standard boundary testing
not using enough bits or digits to store maximum values	test using very large numbers to ensure the system has a wide enough margin of error
not using enough places after the decimal point	perform calculations that involve many significant figures, and large differences in magnitude		the system to round excessively, which can mean that data is stored inaccurately and can also lead to a build-up of errors
not using enough significant figures	perform calculations that involve many significant figures, and large differences in magnitude		the system to round excessively, which can mean that data is stored inaccurately and can also lead to a build-up of errors
ordering operations poorly so errors build up	if a numerical application is designed to work with floating-point numbers, then make sure it works with inputs that vary widely in magnitude, including both large positive and large negative exponents pay particular attention to the accuracy of the result when a floating point value is being repeatedly decremented or incremented by small amounts	when you do small operations on large floating point numbers, and excessive rounding or truncation errors build up

Next defect: defect in an ordinary algorithm    Up: defect    Previous defect: timing and co-ordination defect