How can we test multiple inputs with millions of combinations?
we have to QA a display Input system.(Testing a display firmware) There is a special type of display which have 10 multiple input slots.
Each input get separate signal(a number between 1-100. example- this is bit similar to contrast value or brightness value of a monitor which have 1-100 range values )
So at a time all 10 inputs get 10 different signals(1-100 number) and output of the signals display on the monitor.
At a moment
At another instance-
All Input=(34,54,43,65,87,09,43,22,11,99) - totally different input
There are millions of different combinations.This is a manual testing process.No way to automate this.To test a one input combinations it will take at least 2-3 minutes.
So how to decide what combinations we should test.Do you think Boundary-value analysis is enough ?
So what is the best method to check that system
I saw another thread with a similar question http://www.sqaforums.com/forums/gene...tor-tools.html
You may find something useful there.
Thanks for making easy to find answers.
Apart from the technical information, testing abilities, analyzer ought to have some individual level aptitudes which will offer them to manufacture a decent affinity in the testing some assistance with teaming.
Boundry-value or partition analysis are the 2 main ways of organizing the testing space.
Originally Posted by chamara
If there are discrete boundaries, use boundary analysis. If the boundaries are too numerous to effectively try them all, you might want use partitioning. In partitioning you divide up using use cases and user profiles, then pick a case that represents each category.
Partitioning is a useful techique for things like, testing browser compatibility. Since there are so many browser settings, you might simplify to top 3 browsers/config for 1) a office user, 2) a consumer, 3) a technical user.