DFT Digest

I spent the morning taking advantage of proximity to vendors to solve a couple of problems, so the first presentation I saw today was the invited address given by AMD Fellow Jeff Rearick: This Is a Test: How to Tell if DFT and Test Are Adding Value to Your Company. I thought it was a very interesting address, so let’s see if I can relay some of what was said:

Rearick laid out seven questions that all DFT practitioners should be asking themselves and their management. I’ll paraphrase, and add one sentence of detail, since my shorthand sucks:

1. What value do DFT and test add to the bottom line? What test actually does: screens out defective product, and allows us to learn about our manufacturing processes…
2. What cost is imposed by DFT and test? From the obvious to the subtle: equipment, engineers, silicon area, tool costs, schedule impacts, etc.
3. How can we improve our ROI in DFT and test? Increasing the return (faster yield ramp, test features) and decreasing the investment (drive to cheaper testers, lower test times).
4. How does our investment in DFT and test rank? (Rearick presented five grades of DFT: Leadership, Proactive, Reactive, Not enough DFT and No DFT). Where is our company on this scale?
5. Are we over-investing in DFT and test? Are we running tests that don’t catch anything? Do we need to test at all? Have we designed DFT that isn’t used?
6. Do we use DFT and test to drive quality? Do we have a closed loop process?  Do we measure our results, and feed them back to design?
7. Do we exploit value throughout our value chain? Do we re-use tests in multiple stages of the product chain?  Do IC designers think about board test?

All excellent points.  Soul-searching for DFT and test engineers; a good way to determine where you’re at, and where you want to be in the long run.

After that speech, I wandered into a related session, where Scott Davidson of Sun Microsystems was presenting a quantitative approach to determining the ROI of DFT for any given product.  Louis Ungar then presented a similar paper describing some examples of DFT catastrophes, where money and lives were wasted for the want of DFT in electronics products.  There was one more speaker on this bill, but I had to make my way to the airport.

So the ITC 2008 conference finished off with a dose of good old fashioned DFT evangelism.  All you DFT and test engineers go forth into the world and add the value to your products that apparently only you know that you can!

By the way, Rearick added an important 8th question at the end of his address: “When are you going to give the DFT/test engineer a raise?”

In the next couple of days, stay tuned for my final wrap up of ITC…