( ESNUG 509 Item 7 ) -------------------------------------------- [09/13/12]
From: [ Hello Kitty ]
Subject: User doubts NextOp/Jasper/0-in/Zocalo assertion synthesis value
Hi, John,
Anon, pls.
My management wants the pros/cons of using assertion synthesis in our ABV
suite. NextOp, Jasper, Zocalo, 0-in claim to do it.
We noticed three things about assertion (or property) synthesis:
1.) The input for assertion synthesis is STABLE simulation info.
2.) But stable simulation verification info is not available
until WAY late in the verification stage.
3.) Also, when our verification suite changes (which it does
often) we'll need to rerun the assertion synthesis again.
That means all of our older assertions are now useless.
So how useful is assertion synthesis really? It seems to be just another
add-on task involving pricey add-on tools when you can only generate your
assertions very late after most of your verification is done.
Is this assertion synthesis overhead, at a late stage, really productive?
What's the quality of the synthesized assertions? Will the time we "saved"
by synthesis be lost manually checking and rechecking the tool output?
Is it worth the price of 10's of thousands of dollars for a tool to do it?
Is it worth the price of 100 man-hours learning and maintaining the tool?
Which assertion synthesis tool is the best? Which is the worst?
Or is assertion synthesis just a new technology toy looking for a problem?
(I suspect this is the right answer but it's not PC to mention.)
- [ Hello Kitty ]
Join
Index
Next->Item
|
|
