SAN FRANCISCO--Sluggish growth in the EDA industry, which
has frustrated vendors for a decade, can be reversed by tapping into new
budgets and application areas.
That was the message from Wally Rhines, Mentor Graphics CEO,
during a presentation here at the 51st Design Automation Conference.
Rhines (pictured right), recalling a DAC keynote he delivered a decade ago,
noted that in most areas of EDA in the intervening years, the revenue from
tools has been essentially flat.
"100 percent of the growth in tools has been new
methodologies" such as design for manufacturing, ESL, formal verification,
power analysis, emulation, signal integrity analysis, and so on. IP as well has grown to a $2 billion market in the
same period of time.
In the future, thornier design issues still have to be
solved from an R&D model that represents two percent of the worldwide
semiconductor market. As the cost per transistor has decreased 30 percent per
year over the past decade, so too has EDA revenue per transistor "as it
must be because otherwise we'd become a disproportionate part of semiconductor
costs," Rhines said. "That creates a difficulty."
"What about tapping into new budgets?" he asked.
Resolution enhancement technologies, for example, has jumped from zero to $400
million a year, by tapping into semiconductor wafer manufacturers. The "sleepy" emulation market has
doubled recently to $330 million annually, largely because it moved out of the
laboratory as systems designers began pulling software design farther left in
the design process, Rhines said. That tapped into (for EDA) a new software
development budget stream.
Three major opportunities present themselves for the
industry to exploit, he argued:
Embedded software ("Increasingly, it'll be
not so deeply embedded and even application software that will be pulled forward, requiring
more capacity for simulation/emulation and that can tap into new money.")
System design: Automotive and aerospace is a $2
trillion electronics market "automated to the same level as chip design was in 1970 and overflowing with electronics. It will have
to be automated."
Hardware cyber security. This has "almost
unlimited potential. Our capability in EDA today is largely focused on being
able to verify that a chip does what it's supposed to do. The problem of
verifying that it doesn't do anything it's NOT supposed to do is a much more
difficult one, a bigger one, but one for which governments and corporations
would pay billions of dollars for to even partially solve."
-- DAC 2014 Keynote: Imagination CEO Charts New Opportunities for Semiconductors