( ESNUG 453 Item 6 ) --------------------------------------------- [03/01/06]
Subject: Praesagus, Sagantec, TSMC, SoftJin
Q: As a competitor, what do you think about these specific DFM companies?
In plain English, what exactly do they do? And how do you view them?
Praesagus
Mike -- They do simulation and analysis of the CMP step. They simulate
how the geomotries will change as a result of chemical mechanical
polishing. They are very, very nichey. They model an important effect,
but they cover such a small step they're an obvious acquisition target.
Atul -- For copper, they model the CMP process to predict thickness
variations on the metal lines. These are all proximity dependant effects.
Almost every only fab I know has their own internal CMP models. All the
other players that used be in this space disappeared. This is too
esoteric to be visible to layout designers. Instead they just guardband
their interconnect tables.
Sagantec
Mike -- Basically they can do compaction and reposition of polygons to
obey design rules. You can use this to migrate to difference processes
or to repair errors in your layout (caused by DRC errors or MRC errors.)
Ironically they can solve DFM problems, but have no idea of how to detect
DFM problems. They should team up with a DRC or LRC tool house. They're
on my watch list but I don't think they have a broad enough offering to
ever become a mainstream DFM player as a standalone company.
Atul -- They need to think of seriously partnering with the real DFM
companies so they can get beyong just layout migration. They're not on
my watch list because their technology already exists in the Big 4.
TSMC
Mike -- They have two toolkits "Yield Plus DFM" and "Yield Pro DFM". It
looks like Plus provides a series of rules and advisories, while Pro is a
series of services. This should be a standard part of any fab's offerings.
I wouldn't be surprised if UMC, Chartered, IBM, and Toshiba all also had
simular toolkits. We don't see these as the fabs giving away free tools
that compete with us, instead they're a foundation from which start-ups
like Aprio can build on top of.
Atul -- I agree with Mike. I see these two toolkits as information and
guidance for the development of real DFM tools.
SoftJin
Mike -- I don't know anything about these guys. Their Nirmaan toolkit
looks like it could be used for any implementation flow. I don't see
anything on their web site that makes Nirmaan look like it's particularly
tuned to DFM applications. They seem to focus on post-GDSII flows. We
don't use Nirmaan nor do I know of any other EDA developers using it.
Atul -- They came on my radar screen because they claimed to have some
expertise in mask data set prep tools in India. They're trying to sell
development services to EDA companies. We don't use them. I heard some
companies in Japan were working with them, but I'm not sure.
Index
Next->Item
|
|
