​​(An alliance-partner of Bender RBT Inc., NY, USA)


Contact: info@rbttechnologies.in

Software Testing: Defect prevention, defect detection and debugging

FAQ


How do we know that test set generated by RBT tools are correct and complete?

The algorithms we use are based on the hardware logic testing rules where the defect rates of less than 1 defect / millions of gates or nearly zero defect.  There is a lot of mathematics and, even more important, experience that shows their correctness and completeness.  Also, RBT is probably the best tested  product (as you can see from huge number of clients list)  you will ever see since we apply our full process to it.


What is the Learning curve for the process and tool?

You can begin to have added value productivity in just days.  Portions of the process, such as the ambiguity reviews, have no learning curve.  Becoming reasonably skilled at graphing takes a few weeks.  Mastering it takes a few months.  This is similar to the learning curve on any programming language.


Who are the other competitors in the market and why RBT tool is best in the business?

There are a lot of vendors that have test design tools.  However, all of the others only focus on just reducing the large number of potential tests down to a reasonable size set.  RBT is unique in that it is the only tool which addresses the issue of the observability of defects by designing tests that guarantee defects are propagated to an observable point – which is the whole goal of testing – i.e. ensuring that any defects will be observed.  Also, RBT is unique in factoring in the critical issue of Constraints.  This is used to ensure the logical consistency of the requirements and to ensure the tests can actually be built.  We are also very strong at negative testing (e.g. exception handling).  Even in reducing the number of tests we provide at least twice the coverage for half the tests.  These numbers come from studies done by companies like Boeing, Cisco, McAfee, etc.  In addition, RBT has multiple test design engines.  So we cover pair wise (both optimized and orthogonal pairs) plus Cause-Effect Graphing.  No one else has this variety.


What is return of investment (ROI)?

The RBT process and tool result in about a 20% to 30% reduction in the cost to deliver and the time to deliver.  This is the result of minimizing scrap and rework.  At delivery you should be at or near zero defect.  The RBT tool finds defect early. As the cost of fixing error increases exponentially with development time. When you calculate the cost of defects in production, this is a huge savings and the figure is much more than 30%.  


How does the RBT tool handle the situation where the requirements changes frequently?

RBT has a special feature called ‘Evaluate and Design Both’. This process allow you to test only affected functional variations, instead of testing whole test set, to get 100% functional coverage - i.e., RBT protects the investment in the existing tests while still ensuring complete coverage. RBT is applied to many systems that are very volatile.  For example, one has 60,000 business rules; 25% change every month; and they do new releases two to three times a week.  All of these rules are modeled in RBT.  Just 12 people maintain 20,000 fully automated scripts using this process. Thus BenderRBT is very apt for Agile software development process. 


What is Price for the tool and RBT Process training?

Please see the quotation


How does RBT save time and effort?

It facilitates ‘Fast Tracking’ (Doing activities in parallel that would normally be done in sequence) eg. starting to design test case before coding starts. There are many other testing activities that fall under critical path of the project can be expedited (crashing), eg. One can export the test descriptions to most of the major playback tools in the form of comments in shell scripts. This helps create self- documenting tests saving scripting time.