Bookends at Cadence: In a way, the torch has been passed ...
******************************** This article is about Cadence Design Systems, its future and its past. As Executive Chairman Ray Bingham announced his retirement earlier this week – after 12 years with the company, including 5 as CEO – it seems appropriate to revisit two recent conversations, one with the first CEO of Cadence and one with the man who as of this week now fully embraces job. The first interview is a discussion with the original CEO of Cadence, Joe Costello, who was named last year as the 2004 winner of the EDA Consortium's prestigious Phil Kaufmann Award. When people speak about Costello’s era at Cadence, it’s often in wistful terms. His tenure at the company from 1987 to 1997 coincided with the glory days in EDA – those years when the industry was expanding rapidly and optimism abounded with regards to the business and technology potential in electronic design automation software. The second interview is with Cadence's current President & CEO, Mike Fister, who in this interview is looking forward to the challenges and vigorous sport of building the company into an even bigger player – both in the EDA industry and in the larger semiconductor supply chain. In talking to these two men, what strikes you is their remarkable similarity. They are both agile in their thinking, seemingly unfettered by conventional (read "formulaic") wisdom, and both bring a great deal of intense energy and confidence to the table – confidence in technology, confidence in a future based on that technology, and most importantly, confidence in their own abilities to shape and influence the future of the industries around them. These interviews took place in mid-September 2004. Costello had just been announced as the Kaufmann Award winner. Fister had been on the job for less than 6 months. ******************************** ******************************** * Chapter 1 – Joe Costello Joe Costello and I last chatted back in October 2001, as I was writing up an article profiling his career. This time around, when we spoke by phone in late September 2004, the conversation was prompted by Costello having been selected to receive EDAC's 2004 Phil Kaufmann Award. Here are the highlights of our visit. Q: So, Joe – why the Kaufmann Award and why now? Costello: [Laughing] Well, I got this message from my assistant that [Synopsys CEO] Aart de Geus had called. So I called him back, not having any idea of why he wanted to talk to me. Of course, all kinds of things go through your mind, so when I reached him and he said, ‘I’ve been chartered to tell you that you’ve been chosen to receive the Phil Kaufmann Award,’ I was definitely surprised! I haven’t been involved in EDA, except tangentially, since 1997. Although obviously, I still think about the industry a lot because I’ve been on several Boards of Directors since then, and am still sitting on Barcelona’s Board. Also, I continue to be on the Engineering Advisory Board at U.C. Berkeley assisting [Engineering Chair] Richard Newton. Anyway, my receiving the Kaufmann Award is one of those really great things for me. I was just beaming when Aart told me and thinking what a tremendous honor it is. It’s fantastic! I think that in the group of people who have received the honor before me, it’s kind of like Where’s Waldo. And I’m Waldo! Q: Are you as disaffected today with the EDA industry as you appear to have been in the past? Costello: I’m really not thinking about EDA much these days, but I think it’s true today even more than before. I don’t think electronic design automation gets the value it deserves, because it still creates more value than it gets. Between the tools, the technology, and the services in EDA – people are still trying to work out the value equation. It’s not horrible, though – it’s not ridiculously undervalued. I mean, we don’t pay our schoolteachers anywhere near enough. Those services are horribly undervalued! EDA is not as undervalued as that, but still… Q: Can the EDA industry evolve and improve? Costello: Yes, it can evolve, although right now the market’s a bit stagnant. It feels like it’s sliding laterally, but things can happen during times of big change – which happens when markets emerge with a new view of things. How do we grow? I still believe the way to change growth and value in an industry, is to deliver solutions, not just tools and technology. The difference is really understanding the customer’s problems inside and out. What are the core objectives of your customer’s business? What are their hopes and dreams on the one side, and their anxieties on the other side? EDA is still coming at things with, ‘We’ve got pliers, and here’s a saw, and here’s a plumber’s helper.’ But none of the customers are saying, ‘I’ve got a hammer question.’ The customers are saying, ‘We have a problem that requires a comprehensive solution.’ Q: When will the EDA industry link their financial success to that of their customers? Costello: I agree. If you wanted to crack these things, tie your revenues and your profits to your customers’ successes. That would change the equation and create opportunity for growth. Essentially what the EDA industry does today is, they limit their upside by protecting their downside. How many EDA vendors have their payments tied to the actual success of their customers? When I pushed into services at Cadence, it was when we were trying to make that happen, although there's a limit there. In electronics and other industries, you may actually get push back in the other direction, because sometimes customers act the same way. You see them saying, ‘I'll have my smart guys take your technology, and we'll do it on the cheap.’ So, you see it in both directions. The 90’s were definitely the go-go years for software. Boards of Directors loved their companies to sell software. It was like selling drugs, because there were such high margins. You could sell very expensive packages with big labels like ERP, CRM, EDA, and you could make a lot of money that way through expensive sales channels and high prices. Selling services is something different, however, and people are looking for something different today. Customers are saying, ‘I already have a lot of software, and it didn’t really work for me. I paid a lot, but didn’t get real solutions.’ These days, people have changed the phrase ‘Total Solution’ into a political sound bite, but if you tie your revenue to your actual customer’s success, people won’t keep buying your software unless you really provide a solution. I’m talking about the kind of pro-active success that creates a real bond between a vendor and a customer. Something that people feel proud of, and that you can measure. Q: Does it strike you as interesting that people often say that things in EDA were so much better when ‘Joe’ was around? Costello: I did my piece in EDA and I’m not there anymore. So I don’t really think about it. Q: Okay – so, what are you going to talk about in your acceptance address at the Kaufmann Award dinner in October? Costello: I’ll probably talk about all of this stuff we’ve been talking about here. Of course, they’ll say it’s just the same old Joe, but I’ve seen this stuff succeed in other business segments. And, of course, now that I’ve worked in the MCAD industry, I’ve seen how much easier folks have it in EDA than in MCAD, where things change so much more slowly. In EDA, Moore’s Law forces the customer to continuously upgrade the technology, so it’s a nice business-forcing function for the EDA industry. In MCAD, there isn’t a Moore’s Law, which forces the [MCAD vendors] to look at what the true value of their software is. In MCAD, you have to prove that you’re solving a really big pain for your customers. If you’re not, you’re not going to get diddly from them. It’s completely different in EDA. What EDA needs today are leaders who have a higher motivation to move things forward. It’s not just money that motivates people – it’s the pride of making a change and making the industry better. Q: Why do you think there aren’t more women in EDA? Costello: Well, generally people come into EDA from EE, and it’s scary, but there just aren’t many women there. I heard this story from Richard Newton that people claimed there was no gender bias in a college application for an electrical engineering program. But it turned out that there was this one question on the application that asked high school seniors if they had ever done any serious work on a computer science or technology project in high school. It turns out, that by eliminating that one question on the application, the school was able to change their screening and increase their acceptance rate for women students to 30 to 40 percent. Because, although the women applicants did well in science and math, the gender bias showed up in that one question. The hopeful things, however, is that there are more and more women going into business, even though EDA is biased towards engineers who tend still not to be women. But really, if you want to be successful in running one of these companies, it just doesn’t make any sense for somebody to try it who’s not completely immersed in the technology. After all, that’s the only way you’re going to understand your customers and your own guys. If you don’t have that technical background, it’s really difficult to lead the company. The flip side is, you can surround yourself with those kinds of people, the technical people. But even if you surround yourself with fabulous people, you’re the one who’s got to make the final decisions, and you shouldn’t find yourself having to flip a coin to make important decisions. Q: What do you think about the concern these days over the outsourcing of high-tech jobs? Costello: I understand the underlying issues – most people are hysterically highlighting the thing right now. But instead of attacking it, we should embrace it. People outsource all sorts of things, so the important thing is how does our country, our economy, and our industry deal with it?
On the flip side, we need to be cautious in the U.S. We’ve gotten fat, dumb, and happy here. We need to step up the pace in this knowledge-based industry. The only thing that sets us apart is our education, the educational level of our population. We need the government to encourage education, and for people to take advantage of the educational opportunity they have here. An engineering and science education is an extremely powerful foundation for people. I was not a practicing engineer for a huge number of years, but the foundation I got through that experience helped me to know how to approach problems and analyze things. I strongly encourage that everyone be exposed more to that kind of education than they are today. A science and engineering foundation is an incredible platform that’s extremely useful later in life, whether it’s in management, sales, or even the arts. What people get paid highly for in our society is Sales and Marketing – and even Finance. All of those people tend to get paid in our society more than our engineers, although many times engineers have a higher IQ. But the first group brings the higher value add to the thing, and just as we treat entertainers and athletes as celebrities, there’s a personality thing that’s required to be a success in Sales and Marketing. The thing that engineers can get paid for is innovation, which is always highly valued. Q: What do you predict going forward to the technology sector? Costello: My forecast says move to China, for the reason that things are happening there at a really incredible rate. Maybe some of the environmental things there aren’t as good as they are in the U.S., but there are really smart people there and the market is really growing there. I mean, where do you think the next great innovations in cell phones are going to come from? China has had some successes and some failures, but it’s an exciting and a formidable place. We’ve got to take advantage of that, and I mean all players – whether you’re in manufacturing or in electronics. I say, don’t play sour grapes and put up walls because of fears of outsourcing to China. We used to have it easy here in the U.S., but look at this summer’s Olympic games. America didn’t win at basketball and China was Number 2 in the overall number of medals won. That’s where the world is going. Q: So, are you looking forward to the EDAC Awards dinner? Costello: Absolutely! I’m going to bask in the award, although I hope it isn’t too obvious. [Laughing] My number one message at the dinner will be, ‘Get closer to your customers. It’s true that sometimes customers can be mean spirited, arrogant, or cheap – picking away about small costs. But in general, you’ve got to embrace those customers. There’s a whole new generation of semiconductor companies, and a whole new generation of things happening. So, there’s a new generation of EDA that needs to be managed. There’s an opportunity here for both sides to make a change. Break out of your defensive arrogance, and make things happen!’ [Laughing again] They probably all know I’ll say these things, so you’ve got to see that it took a lot of guts for EDAC to give me this award! ******************************** Chapter 2 – Mike Fister You can skip this bit if you already know that: Mike Fister is President and Chief Executive Officer of Cadence Design Systems, Inc. Prior to joining Cadence, he spent 17 years at Intel Corporation, where he was most recently Senior Vice President and General Manager of the company's Enterprise Platforms Group. Fister is a graduate of the University of Cincinnati where he received a BS and MS in electrical engineering in 1977 and 1978 respectively. He spent his early years in a variety of executive and engineering management positions at Wyse, Machine Vision International, and Cincinnati Milacron. He currently sits on the Board of Directors of Autodesk Corporation. Mike Fister was named President & CEO at Cadence in the spring of 2004, just in time for DAC. Fister’s appointment caused quite a stir at Cadence, and in the industry, because change always does. We spoke by phone on September 15, 2004 – the same week as the Cadence Users Group Conference in San Jose. Q: How is the Cadence Users Group Conference going? Fister: [Very well], although I’ve only been able to be there for a little bit this week. Q: Presuming you’ve moved in from out of the area to take over at Cadence, how has the relocation to San Jose gone for you? Fister: Our home is in Portland, Oregon and [we still have children in the area], so I’ll be commuting down to San Jose. [But it doesn’t matter where I am because] I’m very much a work-all-the-time person, which means I work about 100 hours a week. But, I’ve always loved what I do and I’m very committed to our Cadence team, [so it’s not going to be a problem]. I’m also a big bicycle rider. I ride a lot in Oregon, and it’s great that there’s lots of biking in this area as well. Q: Where were you born and where did you grow up? Fister: I was born in Savannah, Georgia, but I grew up in Cincinnati. My father was an EE – he was infamous in the [design of the] instrumentation in jet engines at GE. So, I grew up wanting to be like Dad. I attended the University of Cincinnati, where I got my BSEE and my MSEE. [While I was in college], I did an engineering co-op at Intel, where I worked on the 4040 and 4004 and got a lot of notoriety for some special stuff I did on those products when they were first coming out. So, after I got my MSEE, I went to work for Intel full time. [At one point], I left Intel to join a small company working in the area of robotics. We were working on machine vision and the problems were pretty intriguing. We developed non-traditional computer architecture approaches [to the problem] based on an abstract algebraic approach – not a statistical approach. It’s the computer architecture guys who [usually succeed in applying] these techniques to problems like traditional data processing, vision, and control. After I left the start-up, I went back to Intel to develop microprocessors and micro-controllers, where several thousand of my buddies and I built product lines that you’re probably typing on right now. We pursued the thoughtful evolution of products through to today’s Pentium III and IV, which allowed us to change the world. I’ll never be able to fully show the total strategy behind all of that – you’ve just got to meet guys like me and some of my colleagues to understand. It was a very considered evolution of those products – different products [to address the needs] of many different types of users. We always planned the microprocessors at Intel with architectural innovations that would also show a product continuum. The why and the sense of the product flows were always implicit in the success of those product lines, because in the design process, it’s the approach you’re taking and knowing what kind of product you want to build. And, of course, always at the root [of everything] is a design method that’s dependent on the human dynamic in the tools. All of that left me interested in going out and trying to help our industry supercharge the inevitable, which is to be increasingly more dependent on the computer. You can see how I’m a logical guy to a fault. Although I was trained as an engineer, I’ve always been more interested in how people think as opposed to how to get things done. My old man derives a formula [to solve a problem] when he needs it, whereas I develop a theory. Electrical engineering training is at the root of everything I’ve done, even in this adventure into EDA. Because in the design process, it’s the approach you’re taking which is just as much a part of the thing as what kind of product you want to build. As the world continues to evolve, the complexity and dynamics increase – we’ll have larger and larger devices, devices with crazy performance [metrics], and products that combine mixed signal with analog/digital stuff. All of us are going to increasingly benefit from the companies in our industry. I have an organizational approach for integrating that experience and [helping to push that process forward]. Q: At the Gartner/Dataquest Semiconductor Conference this week, Intel CTO Patrick Gelsinger said that everything in design revolves around Platform, Package, and Power. Would you agree? Fister: For me, it’s more about the social issues. Q: At that same conference, Gartner Analyst Jim Tully closed by summarizing the three major themes he saw at work in the industry – Consumer, China, and Consolidation. What do you think? Fister: Yes, there is going to be further consolidation in the industry. At Cadence, we’re very thoughtful about that and the roles we are going to play in doing that. We’re busy trying to integrate disparate technology and [looking to achieve] a better holistic toolset, rather a tool for each domain. And while we’re bragging about our demonstrable [technology], we’re also walking the talk about Open Access. [The ability to develop tools in an open environment promises] incredible rewards to the industry. We want to be the driving force behind Open Access – to drive our tools for interoperability, which may [actually] help drive our competitors’ tools. So, it’s a bold move for us, because we’re daring ourselves to be good. But it’s promising a new era in open industry participation – and a breath of fresh air. Also, there are issues about licensing that we’re addressing. There are a number of different business models that can go in a number [of different directions]. I’m very open to the view of trying the different things that will link us to the manufacturing ability of our customers, so we’re also very concerned about] tools for manufacturability. [Many customers] have an idea that yield is more than the raw counting of dies, but [also in considering] the goodness of the product. However, you don’t know how good it is, until you get to the end of the [manufacturing] loop. That’s the reason why design is inexplicably linked to manufacturing. A company must never stand still in its technology or its business model. Our leadership in the industry means we’re going to lead the industry and [at the same time] compete with ourselves. We will be innovating not only in the technical areas, but also in business areas with relationships that we’ve formed with key customers – relationships, which are very methodical, while also allowing us to sustain our [leadership in the industry]. [All in all], we are all about trying to help our customers. By the way, this may sound like a Cadence commercial, but it really isn’t. [Laughing] Q: How do you feel things are going in the EDA industry? I’m a business pragmatist. I believe product lines should be the drivers – I’m not a technology zealot like some. At Intel, we were always careful about the evolution of the technology and timing of the technology, so that the product flow and the timing were very thoughtful. From Intel’s standpoint, it’s the product guys who are the most powerful manifestation of what the industry is doing. In the EDA industry, we [frequently and unfortunately] get people to delve into unbelievably deep detail about stuff that doesn’t have a lot of relevance to what the customer is trying to do. My fiber is [all about] looking at the problems that people are trying to solve – it’s about a timing process synchronization. I want to look at the domains that will be the most relevant to users of our tools, [particularly] on a two-to-three year cycle. As an example – somebody's attacking a mixed-mode design with analog, digital, high performance specs – there may be a technology that’s preserved or reserved to do all of that, but laboring through and debating whether our technology is better than somebody else’s ability to solve the problem is largely an academic exercise. I want to train our technology into a useable tool kit that attacks the application domain. I want to see [our customers getting] more value out of the holism of the approach – an entirely differently mentality. In the quest for leadership in EDA, however, some people have decided to base their [position] totally on a technical foundation that sounds like techno-babble to customers. In fact, however, EDA is all about time to market, managing complexity, and garnering value for the integration of what our customers are doing. When the Apple guys built the best MP3 player in the world, they weren’t trying to say that they wanted to integrate this magnetic storage with that interface and [so on]. They were saying they wanted to create a package [with these specific features] that would fit into a pocket. They worked with the rest of the industry to integrate [specific technologies] to create a product that was smaller, sleeker, and cooler. And that’s where our efforts should be directed. I want us [to work with the customers, who are working] with hierarchical flows at 90 nanometers. It’s a Fister passion that we’re going to be in that thin strata of people, who actually talk to our customers’ customers, because those are the penultimate "What?" guys. They’ll give us the insight as to whether or not our tools and technology are relevant. We’ll benefit [by adding] more value to our tools, our technology, and our services. Q: When will Cadence buy Wind River? Fister: [Big laugh] I haven't thought that much about it. Q: A VC recently told me that the Cadence organization of late has been nothing short of a train wreck – a place of internal squabbles and contentious fiefdoms. How would you respond to that criticism? Fister: Cadence has grown by acquisition and that’s the [source] of some of that criticism. People wonder if we’re going to be able to fully integrate those acquisitions – so it’s fair to ask those questions. But, our post-acquisition retention characteristics are very strong and very high. Ping Chao is one of those guys and is a key member of the acquisition team. I would tell you that, coming into this thing, we could be a better-integrated team, but we’ve got some very positive things going on at Cadence. My reputation is one of being a special person who’s able to [communicate] inclusivity, camaraderie, and teamwork. So, watch the situation at Cadence play itself out. We’re going to bulk up the team in a few spots, and I know you’re going to be impressed with the [additions to the team]. The people we’re bringing in will be a great complement to the [existing structure], just like I was a complement to a great team. Oh yeah – and we’re going to have fun! ******************************** Editor's Note: An edited version of this article appeared on-line in September 2004 in EDA Nation. ********************************
Peggy Aycinena owns and operates EDA Confidential. She can be reached at peggy@aycinena.com
|