Duopoly is bad for EDA - Comment - Electronic Design AutomationJacques Benkoski
WE ALL KNOW A monopoly is bad. It stifles innovation, locks customers in and leads to degraded service and rising prices. But a duopoly? It must be better--it offers the combination of the benefits of competition and one-stop shopping. But not so fast. As with many other aspects, our industry does not behave quite like most of the major, consumer-oriented, industries of the same size.
This is not the first time the EDA industry has had a duopoly fighting it off at the top. We can learn from our own history that, in fact, anything close to this model has repeatedly shown its limits and, more importantly, has historically led to periods of dissatisfaction and stifled innovation.
In a small company, everyone is focused on survival--developing and delivering useful, innovative products to customers. This razor-sharp focus to succeed leads to innovation, a dedication to influence the customer methodology, genuine customer partnerships and rapid deployment of the technology among leading edge customers.
Some startups are acquired by the large companies which have found this to be an economical way to stay current on technology. The larger companies wait until the technology leader surfaces from the startup battle and then they acquire it.
However, when a small startup is swallowed by a larger company, the sense of urgency disappears. The product gets diluted in a larger offering, and the ability to cause methodology changes is dramatically reduced. The key people that were driving those changes unfortunately are forced to become inwardly focused and no longer have the fierce competitive drive that keeps them with an eye on what the customer needs and developing the best technology to meet those needs. But the largest barrier to innovation within a large company is not the lack of talented engineers, but rather economics.
It is hard to retain and financially motivate top talent to remain in a large company when going back to the garage has proven to yield them substantial payoffs. And the bundling of large amounts of technology--mixing leading edge products, which require an enormous investment in R&D with cash cows that are basically in maintenance mode--makes it virtually impossible to properly apportion the revenue and hence the R&D focus to the products that require it.
Some companies will eventually grow to become medium-size contenders and perhaps go on to become one of the large companies of tomorrow. The talent, which is slowly drifting from the large companies, creates the next generation of startups, and the cycle continues. This ecosystem--with all its faults--has proven to work and to provide sustained value to our customer base.
The players in the duopoly can harm smaller niche tool vendors in a number of ways. They can undercut prices and use market segments in which they have virtual locks to keep the small vendors out. They can bundle tools and price the bundles attractively, making it difficult for customers to go out and find the best tool for the job. They can degrade the interoperability of their tools with niche tools. They can configure their software to automatically invoke their own product rather than a competing product.
These two monolithic vendors supplying the platforms and offering the majority of tools in the flow will increasingly offer closed systems. The idea in theory is that with the two platforms, you can use either as a base with a number of tools included and add others in. Though they might claim interoperability and open databases, the customer will effectively have to choose to live in one or the other entire system.
With the advent of common process rules and standard libraries, silicon implementation methodologies have become the main competitive advantage between our customers. The step functions in design technologies are being brought to market as we speak by the new players. Some customers may decide to adopt old-fashioned methodologies supplied by the duopoly vendors. The customers that will choose to make methodology an integral part of their strategic focus will understand the dangers the industry is facing. It is likely that they will continue to push forward for their own sake. lathe process, they will contribute to the development of a new, thriving and diverse ecosystem. Nature is trying to tell us something--we'd better listen.
Jacques Benkoski is president and CEO of Monterey Design Systems Inc. in Sunnyvale, Calif. The duopoly of which he speaks is, of course, Cadence and the newly merged Synopsys/Avant!.
COPYRIGHT 2002 Reed Business Information
COPYRIGHT 2002 Gale Group