In an editorial in the January 19 issue of EDN, Michael Santarini takes EDAC to task for being too optimistic in its market projections. You can find the full article on the web at www.edn.com/article/CA6298273.html. He also accuses the consortium of inflating the EDA sector revenue by including the sale of IP modules in the total figures for the second quarter of 2005. I disagree with his position, see IP belongs in EDA on this website. I do agree with both Michael and Erach Desai (see www.aycinena.com/index2/voices.html) that the representatives of the EDA industry have been too optimistic in making market growth predictions in the last three years, Michael goes further by proposing seven points that in his opinion the EDA industry must address in order to regain its healthy growth of a few years back. He addresses his points to the executives of the four largest, publicly traded EDA companies that account for over 75% of the segment revenue. Although the points address important subjects, I disagree with his positions.

1. Michael:

Put innovation and customers first, not Wall Street. If you innovate and make a product that's worthwhile, revenue will follow, the Street will follow, and, investors may be pleased.

Gabe:

The EDA industry is one of the most innovative segments of the worldwide economy. The problem is not a lack of innovation. It is a shrinking market where the number of companies that can afford to employ the latest semiconductor fabrication technologies is diminishing. Thus the demand for leading edge, innovative products that have traditionally fueled the revenue in EDA is lower than what most analysts and the big four CEOs expected. At the 2004 DATE conference I asked the CEOs of Cadence, Mentor, and Synopsys at the CEO panel whether they thought that most of their customers would follow the pattern and quickly migrate to the 90 nm and upcoming 65 nm fabrication nodes, thus fueling the demand for new tools or be satisfied with 130 nm technology. I thought the move would prove both too difficult from an engineering point of view, and too costly from a financial point of view to be justifiable for many systems companies. They predicted business as usual. The market has proven me right. The slower growth is due to a changing market that the big four are just beginning to understand, not lack of technical innovation. Once they decide to target the correct segments of the market, innovation will follow, and so will profits.

2. Michael:

Don't sell yourself and the industry short. EDA is a difficult field and should warrant a premium price. If EDA is imploding, it is largely because companies hit the panic button and began offering customers all-you-can-eat-buffet licensing. Although you may be hitting your numbers in the short term, in the end, you are doing a disservice to yourself and the industry, dissuading start-ups from innovating and venture capitalists from funding them. In the long run, this approach also hurts customers.

Gabe:

EDA vendors have bundled products ever since they have had two products to sell. The practice is not new, it has just received a new label and the licensing time span changed from one year to three or four in order to smooth the revenue of each company and thus present a more stable financial picture to the investors. Bundling, or “buffet licensing” does present an obstacle to small companies wishing to penetrate and established corporate account, but this situation has existed for at least twenty-five years. Entrepreneurs and venture capitalists are well aware of the problem: it is not new. I do agree that the practice provides free licenses to large customers, but this behavior is the result of a free competitive economy when relatively few customers constitute the majority of revenue for each of the big four. What has complicated account penetration today is not pricing, it is the expense and knowledge required to integrate a point tool from a small vendor into the system flow required to successfully design, develop, verify, and manufacture a leading edge IC.

3. Michael:

Big companies need to embrace start-ups and the growth-by-acquisition strategy. Most big EDA companies believe that, if they stop paying 10 times more revenue for EDA start-ups, EDA entrepreneurs will work for the big EDA vendors and innovate in-house. The fact is that most big EDA innovations come from start-ups. Again, EDA is a difficult and highly specialized field. The biggest tool innovations usually start with one grad student devoting his entire academic life to solving one problem. Architects at most big EDA companies are too busy upgrading yesterday's tools to think deeply about tomorrow's problems. Thus, if you don't support the start-up model, you're killing the industry's future.

Gabe:

As a rule, companies do not buy technology; they buy market shares, either actual or potential. Technology without a market is an academic exercise that does not justify a purchase. The fact that for many years a start-up could introduce a new tool, sell it to one of the “corporate accounts” companies and thus attract the attention of both the owner of the account and its major competitors, does not mean that this market behavior is still the norm. In fact, as I said before, it is becoming more difficult for a start-up to integrate its tool into an existing leading edge design flow, thus the competitive motivator for purchasing a start-up is less common. I do not believe that the EDA segment is so different form any other in the electronic industry to make internal innovation impossible. Research labs in many semiconductor and electronic companies generate innovative products; the same could be done in EDA by investing acquisition funds into internal research.

4. Michael:

The EDAC (EDA Consortium) should not confuse IP (intellectual property) with EDA. Please, get rid of IP in MSS (Market Statistics Service) reports and compare apples-to-apples numbers-good, old EDA. If a governing body were looking into your statistics, you'd have some explaining to do. Maybe MSS could create a separate report for the IP industry if it is that desperate for new sources of member funding.

Gabe:

I have answered this point separately, as I said before. But let me add that comparing apples to apples flies in the face of innovation. Good old EDA is a thing of the past every time the semiconductor companies deploy a new fabrication technology. EDA companies serve one of the most dynamic and demanding markets in existence. There is not room for any “good, old” anything in such a competitive environment. On ly two years ago very few companies were considering purchasing tools for yield improvements: now such tools are required if you want to use 65 nm technology.

5. Michael:

A lot of people think lawsuits hurt the EDA industry when, in fact, stealing does more damage. Stop stealing and stop complaining when EDA companies sue other companies for taking their property. Avanti Chief Executive Officer Gerry Hsu set a precedent that you can build a company on stolen property, use that stolen property to get a head start on the competition, and, even if the courts find you guilty, walk away a multimillionaire. Vendors shouldn't encourage this practice, and, after the courts have handed down a decision, consumers shouldn't, either. (If you think IP theft is no big deal, wait a few years; chances are that it will likely hurt you, too).

Gabe:

No one condemning the proliferation of lawsuits in our industry justifies or condones criminal behavior. But we are seeing EDA companies using lawsuits as a deterrent to the penetration of key accounts and new markets by competitors. Litigations based on patents infringement is also increasing, aided by a US Patent Office that is under funded, under staffed, and guided by obsolete laws and regulations. Although we all want to protect the rights to intellectual property, the methods for doing so must be based on the enlightened understanding of modern technology, in particular of computer science.

6. Michael:

Embrace the esoteric aspect of the EDA industry. The business press doesn't understand EDA, because EDA is too esoteric. If the business press one day begins to understand EDA, it likely means you aren't doing your jobs well enough. Or, perhaps one day, it will mean that EDA has grown large enough to accommodate another analyst and perhaps an opposing opinion to Gary Smith's group at Gartner-Dataquest, which is currently the only analyst group covering EDA.

Gabe:

I do not think Michael meant that all financial analysts, editors, and journalists are too uneducated to understand the EDA industry. I agree that some competition to Dataquest would be good for both the industry and Dataquest itself. The reasons that the business press does not devote more time and space to our industry are financial. Our entire turnover is around four billion per year, less than what many publicly traded companies earn in the same period, and the returns on investment are lower than many other industrial sectors. In addition, the stock prices of the few EDA companies that are publicly traded move in a fairly narrow range and thus do not attract the attention of speculators: and the speculators fuel the financial markets by reading the publications, paying commission fees and interest on borrowed funds used for edge investments. Financially the EDA industry is not very interesting and it is likely to remain that way for a few years.

7. Michael:

EDA needs to work on a new slogan. When I moved from EE Times to EDN, I picked up the FPGA beat and was surprised to learn that the FPGA industry posts lower revenue than the EDA industry and has more than five industry analysts covering it. Maybe EDAC can leverage that data point and take pride in the fact that EDA posts higher revenue while getting only half the business-press coverage the FPGA industry gets. In fact, if EDAC wants to shelve the slogan “EDA: where electronics begins,” it should replace it with “Did you know we're bigger than the FPGA industry?” Better yet, instead of always coming across as the forgotten child of the electronics family, EDA should adopt a haughty attitude and slogan to match-maybe something like “We're way above your head!” That's sure to grab some attention.

Gabe:

I agree that the EDAC adopted a poor slogan for its marketing effort. Although the slogan rings true to us and to some of our customers, it is too much of an inside message. We continue to fail to realize that the vast majority of the world population is not familiar with the engineering world. “Where electronics begins” requires and understanding of IC design methods to be appreciated. What Michael, in a humoristic mood, proposes is of course not acceptable either.

To comment on this article send email to:gmoretti@gabeoneda.com