( ESNUG 430 Item 8 ) --------------------------------------------- [06/16/04]
From: [ The Emperor Has No Clothes ]
Subject: User Wonders "Why Does Synopsys Promote VCS & Vera Over Magellan?"
Dear John,
Keep this anon, please...
I recently attended a so-called "Verification Seminar" from Synopsys. As
I'm a Magellan fan, I expected to hear something great about their new
formal tool from SNPS; but I was a bit disappointed when I realized that
Magellan was presented like a tool to be used only when you do not have any
other things to do (or if you have some problems in covering portions of
your design you have no ideas how to reach and are looking for hard bugs.)
What the seminar proposed was 99% simulation-based, and the style promoted
would prevent a user from moving towards formal methods; moreover, all the
methodology was based on Vera, i.e. nothing was synthesizeable and
forget emulation/acceleration.
Now, having been in the field for a while, I can understand that many CAEs
do not have a good understanding of formal methods (and can even be scared
of them!) In any case, I expected Synopsys to present a unified vision of
the problem of the verification today, being the only vendor (as far as I
know) that has a technology that could allow the user to "easily" switch
from the dynamic to the static world, and vice-versa -- provided a correct
style is adopted from the very beginning.
Why on earth one should buy a tool like Magellan, if the vendor is not
able to present it in a unified flow? It looks like a big risk in having
no (little?) support and scarce integration with their other tools.
- [ The Emperor Has No Clothes ]
|
|
