( ESNUG 491 Item 5 ) -------------------------------------------- [04/14/11]
Subject: (ESNUG 485 #5) Reader doubts Tessent TestKompress compression data
> Our goal was to share with DFT engineers around the world a solution which
> integrates Talus Design with Tessent TestKompress to get a compression
> ratio of 100x. We described how to insert scan logic and an EDT logic
> block into a scan compression DFT flow and how to perform ATPG. Now,
> let's compare the non-compression ATPG results to the scan compression
> ATPG results to find if a compression ratio of 100X has been achieved:
>
> Non-Compression Scan Compression
>
> Test Coverage: 92.03% 92.76%
> Fault Coverage: 84.89% 90.55%
> ATPG Effectiveness: 99.82% 99.82%
>
> Total Scan Test Patterns: 2,475,535,200 23,065,800
>
> Total Scan Test Pattern Volume: 2.48 Gb 0.023 Gb
>
> Notice the scan test pattern volume (0.023 Gb) for the scan compression
> ATPG is 100 times smaller than that of the non-compression ATPG (2.48 Gb)!
>
> - Debo Sekoni
> AltaSens, Inc. Westlake Village, CA
From: [ Doubting Thomas ]
Hi, John,
Please keep me anon.
First, I want to thank the authors for a well-written cheat sheet on how
to insert & run scan.
However, I'm not sure I trust the results. Typically, fault coverage is
roughly the same or slightly less *with* scan compression, compared to
the non-compressed run. The results here are opposite; the compressed
fault coverage is 90.55% while the non-compressed run is 84.89%, which
is significantly less (~5%).
Maybe the faults are getting counted differently? The authors included a
rep stat output for the compressed run, but not for the non-compressed run.
If the authors can share the latter, we can compare the various fault types
side-by-side to see what is going on to explain the discrepancy in the
fault coverage.
Of course, an explanation from the authors would also help since they are
familiar with the design.
- [ Doubting Thomas ]
Join
Index
Next->Item
|
|
