Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
To begin with, I question the wisdom or usefulness of such
a degree of specialization. Call me "old school" but I respect
more, someone who knows and owns the design from bottom
to top. This sort of "somebody else's problem, because I don't
feel like dealing" leads to dysfunctional design teams and bad
products.
But specifically to your question, testability begins with the
detailed design and a "backend" engineer whose only role and
authority is to faithfully implement a "frontend" design that
lacks the test "backbone" is simply doomed to fail at meeting
testability metrics and actual testability, and be simply one
link in a chain of fail that ends in blown-out test development
schedules and in-production test escapes.
Tools are secondary to responsibility, effort, skill and diligence.
And they will not substitute for decent judgment about how
to get test coverage without (unduly) compromising the other
aspects of design (performance, compactness, reliability, etc.).
Prima donna behaviors don't make for career success in
engineering. Get on the team and get your hands dirty.
For the same reasons that they deal with DFS (Design for Synthesys) and DFV (Design for Verification). They have to deliver a design that is functional, verified,synthesizable, testable and meets timing. If they don't and the back end has to clean it up then it becomes a real mess.
Back in the early days DFT engineers were a little cocky. R+D was told not to worry about test. They should only worry about functionality and the test engineers would take care of figuring out how to test it.
That lasted until the 70's when larger designs in smaller packages made testing harder and harder. Test engineers now had to ask R+D to follow DFT and DFS guidelines in order to be meet test requirements. This pressure has continued so that if you are building any chip today with high performance analog then it is costing you as much to test your chip as it did to fab it.
When one engineer is responsible for as much of the chips cost as the rest of the team combined then you do not leave work for them to do, they tell you what to deliver.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
