The EDA Consortium's 2001 Forecast Panel
January 30, 2001, San Jose, California
By Rita Glover, EDA Today, L.C.
Moderator:
John O. Barr, Managing Director,
Robertson Stephens Company
Panelists:
Steve Carlson, Tharas Systems
Aart de Geus, Synopsys
Penny Herscher, Simplex Solutions
Y.C. “Buno” Pati, Numerical Technologies
Wally Rhines, Mentor
Graphics
Ray Bingham, Cadence Design Systems
On January 30, 2001, the
EDA Consortium assembled a panel to prognosticate on the future of the
electronic design automation industry for the coming year. Following is a transcript of the
proceedings, edited and produced by Rita Glover of EDA
Today.
Introduction by Ray Bingham, President, EDA Consortium
The third quarter of 2000 is the second in a row to be over $900 million in revenues. It’s our highest third quarter ever, with 12% growth. I think we would all like to see much more consistent record from EDA companies, and many of them are making moves that lead in that direction. So tonight we’re all here to see what the year 2001, and even the next decade might bring.
Source: EDA Consortium, Market Statistics Service
Will the year 2001 be one of increased strength for EDA? This question leads into our main event this
evening– the 2001 CEO Forecast Panel, which will consist of panelists’
presentations followed by a discussion.
We’re pleased to have John Barr, managing director and senior research
analyst at Robertson Stephens, moderating tonight's panel. John
has been a long-time EDA industry follower. Prior to joining Robertson Stephens, John was an analyst with
Needham & Company, where he was an Institutional Investor All-Star in
technical software in 1998 and was ranked by Reuters as the leader of one of
the top large-cap enterprise software teams.
He came by his knowledge of technology in the trenches for 15 years at
Interconnectix, Racal-Redac/HHB Systems and Calma/Comsat General. He earned an MBA from Harvard Business
School and a BA from Colgate University.
I’ll turn the podium over to John to introduce the panel.
John Barr, Financial Analyst, Robertson
Stephens
Thank you, Ray. I have a few slides of introduction just to give some perspective of where we think EDA is
and what the last year looked like from a stock perspective, which is where we
spend our time. Maybe we should just
step back a year and see how our forecast panel held up last year for the year
2000. A couple of things really stuck
out as I think back on that. We talked
then about EDA industry growth being somewhere in the ten to fifteen percent range. As we look back, that didn’t quite happen,
but if you factor in the license model changes that the major companies did,
you’ll find that we pretty well did hit that fifteen percent level.
We also talked about expansion
of EDA, supplying tools and services -- in terms of the service side of Cadence and
some of the other companies -- and Guido Arnout talked about pushing into the
software space with System C and some of the high-level tools. So I think that with today’s panel, we’ll be
hoping for discussion not only of the license results that we expect for the
next year, but also looking farther into the future as to where we push the boundaries of
EDA in terms of business models. How do
we really create value to grow at a twenty percent rate long term? So we have representatives from both the
large and small EDA companies today to address that.
We will first start out by
looking back over the last year in a rather dismal perspective, the
performance of the NASDAQ, which represents many of the growth elements of the
economy and our customers, as well as EDA companies. And we see turbulence, mostly toward the
downside.
However, looking behind
that, there is a very positive story of what has been happening in the past twelve
months. We have overlaid the index
of most of the public EDA companies with the NASDAQ. We can pinpoint the Dot-Com demise in that fateful March of 2000
-- I think it was March 11 or so -- when the Dot.Com's started to come down as stocks, and as
you will notice, EDA came right down with them.
But then the first-quarter results came through and looked to be
positive, and the EDA stocks continued to perform throughout the year. As you can see, on average these are up
80 percent on the year.
We also mapped
EDA against the SOXX -- the Philadelphia Semiconductor Index -- which
includes companies that supply chips. Some of the semiconductor equipment
companies like Applied Materials outperformed at the beginning of the year, but EDA was more steady throughout. Then as the semiconductor inventory correction hit the
semiconductor side, of course those stocks were hit, but Design Automation continued to perform with good results in the October through January
quarter. Of course, the debate is, “Is
this real, or is it not?” And “Will the impact of the customers’ difficulties hit design
activity, and when?” These are really the
issues at hand.
To
take a longer-term view
of the cycle, we mapped Applied Materials, Cadence, and Intel against each
other. There are a lot of interesting
data points, but the first thing I’d point out is the fact that they all end up
in the same place. It’s been five
years, and all of them show about 19 percent compound annual growth, but boy,
has it been a bumpy ride for Applied and Intel! There are periods of outperformance here for Cadence, and I took
Cadence just as a proxy for the group, having the largest capitalization. Cadence
outperformed as a stock in 1996 and into 1997 after the semiconductor downturn, and then again in 1998. Then the upturn that began in 1999 just didn’t happen for
Design Automation, and I think that we are just seeing the beginnings of that in 2000, lagging by about a year. So it’s
an interesting background as we look forward to 2001.
And with that,
I’d like
to introduce our panelists. We have a
distinguished panel representing both the three largest market-cap companies, three new emerging companies, and one -- Numerical Technologies -- that is
really pushing the envelope in terms of defining a new business model and
addressing a large business base beyond EDA.
So we’ll start with Steve
Carlson of Tharas Systems. Steve joined
Tharas in August of 1999 as its president and CEO. Prior to Tharas, he had entrepreneurial experience at Escalade,
and was at Synopsys in its early days.
When they say that someone "wrote the book," well, Steve really did write
the book -- his was the book, Introduction
to HDL-Based Design, which introduced hardware description languages to the
EDA industry. So Steve has an
interesting perspective.
Steve Carlson, President and CEO, Tharas
Systems, Inc.
Thanks. I think probably a lot of you aren’t aware
of who Tharas is. Hopefully a year from
now that won’t be the case. The easy
way to think about Tharas is that we build a hardware box that makes VCS or
NC-Verilog run a hundred times faster than it would on a Sun machine.
John asked us to look at
three specific questions:
1. How is EDA related to the business or
semiconductor cycle?
2. What are some big-picture ideas for growing
the EDA industry at a 20% compound annual growth rate (CAGR) for the next
decade?
3. What specifically can we do to grow our
business at this high rate?
This chart from IC Insights
shows the cycle that the integrated circuit (IC) industry goes through. It’s in
a spiral, growing ever larger. Last
year there was record spending on capital equipment by the semiconductor
industry. This period of aggressive
capital spending is ending, witness the latest reports from Applied Materials,
and we will be seeing a period of excess inventory, and moves for inventory
reduction. We are presently moving
around to the bottom of the spiral (into the red), which are the slower parts
of the cycle. The likely duration here
is about a year and a half. What we’ve
seen historically is that without a hard crash or recession in the economy, EDA
is largely unaffected by these cycles.
Source: IC Insights and Tharas Systems, Inc.
But there are some
questions as to how rough the road ahead will be. There have been some red flags recently -- for example, the GDP
(Gross Domestic Product) numbers. Also,
a lot of the earnings reports that are disappointing the analysts, and more
troubling are the layoffs. I think we’ve
lost more than fifty thousand jobs in high tech this month, and we’re seeing a
lot of inventory reduction kind of behavior in our customers -- Cisco, for
example.
Tharas itself has been
unaffected by these things as of now.
We aren’t seeing any suspensions of projects or cancellations. People are still buying. The one red flag for us is that we see a
shift in purchasing using capital equipment lease lines, which means that
people are starting to take a closer look at the financial treatment of the tool
expenditures.
The next question is, “How
can we grow EDA twenty percent?”
History shows that it just doesn’t grow that fast. You can’t will it to grow that fast, or
extort more from the customers to make it grow that fast. So the question is, “What can you do?” There’s been a lot of focus on license
models and revenue recognition as of late.
There is an improvement to be had there, but it’s not going to get you
that twenty percent level. Do business
models get you there? Buno Pati’s
company represents that possibility in a very positive light, but ultimately
you have to be talking about new markets in order to grow. You can’t keep selling more things to the
same people, over and over.
I wanted to make a comment
on the license models -- just a different perspective. The old model was the big license fee up
front and then a small trickle of maintenance dollars after that, versus the
new annual subscription model, with equal monthly revenue recognition. I think everybody is in general agreement that
the new subscription model is a very positive thing for the industry, but it
may not be without a cost. It increases
the predictability of the revenue recognition, and there is a strong
possibility that it is also going to decrease the discounting pressure that we
see at the ends of the quarters, because it is not an all or nothing
proposition for the EDA vendor. They
would be walking away from less immediate revenue, should the customer insist
on unreasonable discount levels.
Source: Tharas Systems, Inc.
But again, if we look what
the EDA industry does, we talk about “retooling” a lot. “Retooling” means we’re replacing old tools
and methodologies with new ones, rather than selling to new markets. We are just replacing one tool revenue stream
for another. Recognizing this, if you
look at the old license model, you can actually end up with a situation where
there is a greater cumulative revenue than compared with the new subscription
model. In fact, you’re trading this
higher growth for predictability. So
there is some kind of a trade-off that we’re making.
Source: Tharas Systems, Inc.
In EDA we always talk about
the cost of the sales and the distribution channel as being major obstacles to
diversification and to entering new marketplaces. The perception is that it’s easier to fight over the existing
marketplace and steal market share from our competitors rather than developing
new markets. But it’s a real question
in my mind as to how much channel leverage there actually is. When you look at formal verification tools
and selling them via the same sales and AE force as layout extraction tools,
are you really gaining leverage? In
fact, if you look at the numbers, you would expect the larger companies to
spend a smaller percentage of their revenues on sales and marketing. But that is not the case, obviously.
We have seen movement out
into non-traditional kinds of markets -- design reuse or semiconductor
intellectual property (SIP) is one area that is beginning to solidify a little
bit, but maybe not living up to all of its promise. There has been movement up into the somewhat mythical systems
designer, a realm that hasn’t developed as fast as we had hoped, and down to
semiconductor equipment -- maybe a smaller audience, but willing to entertain
new business models, because they haven’t been doing business with us for
twenty years. The services side is an
interesting one, because I think there is a lot to be exploited there. So much of our customers’ budget does go
into personnel costs -- over 90%. There
is a lot of money here to tap into.
We really want to look at
the major impediments to diversification in the channels. Let’s think about starting by just
leveraging the channels we have. If you
look at potential leverage of our channels, by selling more products to the
same people we reach today, what are other things could we actually sell? What are they buying? Maybe we could sell workstations, ASICs,
services, components, or other things as well.
With actual leverage in the channel, rather than the supposed leverage,
we could afford to contemplate selling products that are commonly thought of as
lower margin offerings.
Source: Tharas Systems, Inc.
There is one well-known
example of a diversification opportunity that perhaps represents the road not
taken. One of the major companies in
EDA bought a small company comprised, primarily, of a world-class wireless
design group. They had the opportunity
to go out and develop products in an emerging wireless marketplace, or to use
those resources to provide services to wireless companies. They actually chose not to create their own
products, because they did not want to compete with potential or existing
customers. They thought that would be
dangerous, not being able to sell them tools.
That was the road not taken in this example. One company that did emerge in wireless -- Qualcomm -- now does
$3.2 billion dollars in revenue, with a $55 billion dollar market cap. Maybe we shouldn’t have cared about selling
those wireless tools so much.
Source: Tharas Systems, Inc.
So what can we do to grow
EDA at twenty percent? Basically, we
have to redefine the marketplace. We
can’t just sell these replacement tools to the same old people. We need to essentially get out of EDA in order to see that level of growth.
For our part at Tharas,
we’re working in a segment of the market where we serve big design
projects. This is a very vital
segment. Last year we had about a
thousand designs of over a million gates underway, and in a couple of years
we’ll have over four thousand. So it’s
a very rapidly growing area. TAM (Total
Available Market) is not going to be an issue for us -- it’s really market reach
that could act to throttle our growth.
Source: Tharas Systems, Inc.
We have very easy-to-absorb
technology and a ready marketplace. We
have the enviable position right now of being a company that can look at triple-digit
growth realistically for the next two, maybe four years. We’re only going to be limited by the number
of application engineers that we can hire.
The real challenge now is in shaping the company for growth in years
three and four. One of the elements of
the strategy that we’re working on is looking at ourselves as a
high-intellectual-property content, application-specific hardware
provider. Right now we’re developing a
channel for EDA. We’re looking to add
value-added channel partners with people to add services, support, technology
augmentation, maybe hosting, and wrap all kinds of facilities -- maybe VARs,
around them. It really is the value-added part of the strategy that will ensure
that there will be enough margin to make it an interesting business for
everybody. This will enable us to go
and invest in new marketplaces (like biotech, pharmaceuticals, chemicals, etc.)
where we’ll co-develop these channels and move on to new opportunities over
time. Thanks.
Source: Tharas Systems, Inc.
John Barr, Moderator: Thank you,
Steve. Next we’d like to introduce Aart
de Geus. Aart is the Chairman and CEO
of Synopsys. He was a co-founder of the
company in 1986, which was started in North Carolina and moved to Silicon
Valley, where they have grown to be one of the most imminent software companies
of any type in the industry.
Aart de Geus, Chairman & CEO,
Synopsys, Inc.
As you know, we do these
forecast panels once a year, and I remember only too well last year, because
following this we had the biggest stock dip in our history. So I’ve got to be pretty careful of what I
say in these predictions. But it is
clear that last year we were in a very different situation, because at that
point in time we were surrounded by great opportunities in the world of
Dot-Com’s.
Source: Synopsys
It seemed at that time
that any of our employees who were not key participants in Pets.Com were
morons. This was not a complete
surprise, because those of you who have grown up with our industry know that
dog food has always been close to our hearts.
Unfortunately for EDA, it was easier for our employees to work in this
area, greatly fueled by the belief that shipping dog biscuits is indeed a job
that requires a high degree of intelligence and algorithmic power. So quite a number of the employees in our
industry actually went there. What made
this all worse was the fact that some of our own decided to give up their
belief in our industry.
It’s interesting to look
at the people that have been aspiring to file for IPOs (Initial Public
Offerings) recently. We have quite a
number here. Notice that during all of
2000, EDAC was regarded as a four-letter word.
What found is that of all those companies, not a single one used the
term “EDA” in their IPO statement. So
it will be interesting to hear the next speaker explain this to us. But we certainly don’t begrudge them a place
at the table, because I just checked just ten minutes ago and was assured that
they did pay their EDAC dues for 2001.
So they’re okay from that perspective.
Source: Synopsys
What everybody seemed to forget in the melee last year was that Moore’s Law was alive and well. This graph is just another way of looking at it. It’s a rather constant Compound Annual Growth Rate, which is now clearly into the millions of transistors. Keep in mind that this is actually what’s being designed, not what’s being fabricated. So no matter what, we’re in an incredible opportunity space.
You may also recall that a year ago, we talked about the fact that the silicon opportunity, Moore’s Law, was accompanied by a designer productivity that didn’t grow quite as much as we would have liked in the ‘90s. As a matter of fact, you may recall that many of the articles essentially predicted the demise of EDA due to the design gap. You will also recall that, despite those dire predictions, all of us foolishly predicted that EDA in 2001 was going to be quite good. I think it is fair to say that we were right. At 0.18 microns or below, there are substantial investments in design that not only will be made, but also actually must be made in order to get chips out. Many of the companies here have greatly benefited from both the dangerous electronics industry issues as well as the opportunity of doing designs.
Source: Synopsys
Let’s take a look at what’s been happening from the point of view of growth rates. Focusing a bit more on the larger companies, we can see that over a long period of time, our growth rate has actually been reasonably decent. It’s a little bit difficult to calculate exactly what it is because of the licensing model changes. But I would predict that it will increase going forward. You notice also that in the process, if you look at the percentage of the market that is covered by the top four companies, there is a steady increase.
Source: Synopsys
Tying into some of the slides that John Barr and Steve Carlson showed us earlier, if we correlate this out to what’s happening in the semiconductor market, you can see that every time the semiconductor market goes down, the large EDA companies actually gain in market share. This means that customers are more careful about where they spend their money and try to have fewer sources of supply.
Now, in my opinion, there was a subtle thing happening in the last year -- the change in the business model. And in all fairness, Avant! was already fully into this model a couple of years ago. They claimed at that time that 15 percent of their license bookings were coming in in a ratable (subscription) form. Cadence has made great progress and most recently reported about 36 percent subscriptions. As you know, Synopsys went “cold turkey” during the summer, with all the pros and cons associated with that, but I can argue that it was absolutely “pro” overall. And it’s good to have Mentor show us what life can be on the other side of that perspective.
Source: Synopsys
The bottom line is that we feel a heck of a lot better as an industry than we felt twelve months ago. As much as I think that it’s great to have our stock prices up now, because that is a reflection of how other people think about our market, it is useful to reflect for just a minute on why we felt so bad during the year, even though we knew that, fundamentally, the technology was coming our way and always has been.
Source: Synopsys
It’s in that context that
as an industry, we most definitely need to increase our pride in our industry,
and to increase our understanding that we are central to what chip design is
all about. Therefore, the valuation of
our companies and the return on our business will be there.
Penny Herscher, Chairman & CEO,
Simplex Solutions
[Editor’s Note: Penny Herscher’s comments have been omitted
from this version of the transcript to comply with the quiet period prior to
the Simplex IPO. The transcript of her
remarks will be inserted after this quiet period has passed.]
John Barr, Moderator: Our next panelist, Buno Pati, is the President and CEO of
Numerical Technologies. He gave up a
very promising academic career at Harvard and Stanford to co-found his company
in 1995, and stayed on at Harvard for another year while his company was
getting started in Silicon Valley. He
then joined the company full-time shortly thereafter. They were one of the most successful IPOs in 2000, and it’s been
a great success story in developing a business, leveraging a big market
opportunity, and presenting a very interesting thing in EDA.
Y.C. “Buno” Pati, President & CEO,
Numerical Technologies, Inc.
[Editor’s Note: Buno’s slides are unavailable.]
John Barr, Moderator: Next, I’d like to introduce Wally Rhines, the Chairman and CEO of
Mentor Graphics. Wally spent
substantial time at Texas Instruments, so he brings a great perspective as a
real customer to his current position, as he was Executive Vice President of
the Semiconductor operations of TI before joining Mentor. I really like the fact that early in his
career, he was responsible for the speech synthesizers which led to Speak &
Spell. Wally joined Mentor in 1993.
Walden C. (Wally) Rhines, CEO, Mentor
Graphics Corporation
Thank you. I might note that John is one of the few
analysts, unlike the Dot-Com analysts, who actually tells you to buy a stock
before it goes up, and to sell it before it goes down.
You will all be pleased to
know that I have brought only two slides.
Addressing the first question, how do we grow the whole industry and our
individual companies, at twenty percent plus, and move into markets beyond what
EDA provides? As most of you know, the
number of designers only increases about three or four percent per year. So if you make your living selling more of
the same things to the people who are already buying your stuff, you are going
to have very, very weak growth. Well, obviously, since our industry averages
somewhere in the 15-16 percent range of growth per year, that’s not what we’re
doing. What we’re doing is selling new
capabilities. So the issue is how can
we find more new capabilities so we can grow the market faster.
Yet another factor, which
most of you are probably familiar with, is that although your customers will
tell you, “Oh, if you’ll just make tools better, faster, cheaper, etc., I will
buy them,” they’re lying to you. The
fact is nobody changes design methodology, or buys significant amounts of
tools, until the ones they are using start to break. So you can benchmark all you want to, but until the existing
solution no longer does the job, or is so slow as to be unusable (like 10X plus
slower), they will continue to use that old tool..
Now many of us go and talk
to companies and say, “Don’t you want to improve your designers’ productivity?”
and we show all these measures and these design gaps and so on. But I am convinced that the management of
electronics companies believes that their designers are available to work on
weekends and late at night, and don’t need to have their productivity
improved. As long as the job gets done,
and the productivity improvement is expensive to implement -- i.e. a tool
purchase -- it’s a hard sell against the thinking that “all I have to do is
make the people work harder.” The way
to sell a tool is to step beyond the productivity space into the space of “What
can my current methodology not do?”
“What do I have that’s broken?”
“What do you have that actually gives me something I cannot do any other
way?”
I would categorize it into
three areas. One is that you solve
problems before they become critical, before something is broken, and you’re
there with solutions when the problems become killers. I offer as some examples, physical
verification -- what you saw over the last few years was a case where nobody was
expecting it, but all of a sudden, the physical verification job got longer and
longer and longer. Until finally with
the big chips, you told your boss that the design was done, but you have it
there running on the servers, and you haven’t finished the design rule
checks. All of a sudden you realize you
may never finish, and that’s when the discontinuity occurs. The big one that has occurred is design rule
checking and layout-to-schematic verification, but there are many more in the
analysis area, as the panelists have noted, including extraction, power
analysis, inductance, etc.
Hardware/software co-verification has been around for years; the telecom industry finally adopted it, but it’s sitting there waiting. It’s still a small business, but it’s the kind of thing that can create large markets, especially when you deal with the software creators and the embedded software industry, which is a large industry -- a big one that only a few EDA companies are in.
Another opportunity is to
grow into adjacent markets. You’ve
heard about one of them -- services -- I believe there are some others. Printed circuit board design used to be a
zero-growth market; in fact, a good
year was zero growth. Now, all of a
sudden, it’s one of the rapidly growing markets. It’s the kind that’s growing into adjacent spaces, because it’s
become a major part of the design cycle time.
Now it’s moving into the mechanical realm, with the tools for implanting
wiring in plastic cases, designing the cabling, and many more
applications. As Buno highlighted, a
major market emerges when you move into the mask-making process; inexpensive capacity is created through
software by optical proximity correction and phase shifting of photomasks. And other growth opportunities come from
attacking more of the design flow. For
example, FPGA designers haven’t traditionally spent much on tools, but the
growing size of FPGAs means they need ultra-easy-to-use tools in a complete,
tightly integrated flow.
As an aside, I frequently
have to note the innovation of my competitors as well. When we talk about getting into new
businesses, it was pointed out to me today by Penny Herscher that the bottled
water industry is larger than the EDA industry. Well, I have to hand it to these people at Cadence. Would you look at the bottles of water on
the tables with Cadence labels on them?
They saw the opportunity and they bottled their own water!
The third opportunity space
is new forms of distribution. Aart noted, we had a lot of hype around the
Internet, but not much happened. And
that’s true with EDA on the Internet as well.
One thing that did happen, though, was remote use of emulation -- something
where there’s a big capital expense that can be reduced by remote use. That’s an example of distribution that has a
lot of leverage. And there are other
opportunities. You saw Internet usage
drive design reuse with the availability of data repositories and design
software that can access IP remotely.
The next questions that John posed related to the semiconductor industry and its cyclicality when mapped to the outlook for our industry. If you look at the EDA industry versus the semiconductor industry, you notice several things. There are some distortions in revenue recognition in the data in 1999 and 2000, so you have to look at aggregate data. First of all, the volatility of the EDA industry is very small compared to the semiconductor industry. People who deal with the semiconductor industry are fond of noting that since 1978 the semiconductor industry has only once grown annually at a rate between 10 and 24 percent -- otherwise it has been either greater or less. That is a tough business -- let me tell you, I lived it for 21 years. EDA will fluctuate 15 percent plus or minus a couple of percentage points when you pull out the extraneous factors in the data. But you can’t help but notice that whenever the semiconductor industry goes down, the EDA industry goes up. Now unfortunately, the other side of that is that when the semiconductor industry goes up, the EDA industry goes down. But admittedly, it’s a small effect.
Now, why has this happened,
cycle after cycle, year after year? I
have to tell you, I think it is a real
phenomenon. I went through all this
year after year. When there is a
recession in the semiconductor industry, it’s a price recession. That is,
the unit volume never declines in the semiconductor industry. It has grown nearly every quarter in
history. So you’re building more, but
you’re enjoying it less. Basically, the
prices drop, but units continue to grow.
Here you are with this
wafer fab, it’s producing more than it ever has, and you’re getting half the
revenue for it. What do you do? You redesign. You shrink designs. You get
more revenue per wafer. You introduce
new products. You do whatever you can
to bulk up the amount of revenue you can get for that given amount of
manufacturing capacity. I can assure
you that after many, many layoffs in the semiconductor industry, I can never
remember laying off designers. You can
shut down manufacturing facilities, you can lay off all sorts of people in
other functions (all of the administrative people can go), but the designers
stay.
Now what happens when we’re
at a peak? When you’re at a peak, the
fabs are loaded. The assembly sites are
loaded. You go in to the manufacturing
manager, and he is saying, “We can sell everything I can make,” and you say,
“Well, just let me start running this new product.” He says, “New product! Why would you want to run a new product? We’re selling everything we can produce! If you put a new product in my manufacturing
line, you’ll mess everything up; the yields will go to hell. Keep away the new products, and I’ll
generate more revenue for you.” And
that in fact is what happens.
Semiconductor companies are not quite so anxious to buy tools when
everything is great. They buy when
things are not great. It’s happened
every time in the past; it’ll probably
happen every time in the future.
John Barr, Moderator: Thank you,
Wally. To close the panelist
presentations, we have Ray Bingham. Ray
is the President and CEO of Cadence, and he brings a great outsider’s
perspective to our industry, having spent some time at Marriott and Red Lion, and
has some great stories about Africa and Europe for a company that was in the
chemical business. He joined Cadence in
1993 and became CEO in 1999.
Ray Bingham, CEO, Cadence Design Systems
Thank you. John and I have worked and struggled
together for some time. What a
difference a year makes. A lot lighter,
a lot more fun talking about clearer horizons than last year. Remember last year? It seemed to me that we spent the entire
evening dredging through licensing models.
It’s no wonder that people went into dog food. There was little to celebrate.
It is a huge difference now.
It’s easy to lose track of the horizon when you’re looking down. It’s easy to forget why you were there in
the first place, and it’s no wonder that people wander off and do other things
when they look back at what got them there in the first place. This is a much more fun, optimistic evening,
and I think that it’s a lot easier to circle back to the center of what we do
and the value we can create. We’ll talk
about how we can leverage it in a much more valuable way for ourselves, for the
people that we’re responsible for, and certainly for our customers.
This is a great time to
be talking about this. We’re smack dab
in the middle of earnings season. The
FD (full disclosure) will have more of an effect on some than others. And obviously it’s a fantastic time to be in
this business. The speakers before me
have really made the case for the value of the basic science that we
pursue. The arguments are all
there. The conclusions are a little
different, and that’s not too big a surprise if you think about the variety of
companies that are represented here this evening. One thing we can be convinced about, though, is that there are new
technologies that are creating new drivers for us as an industry, and for the
customers that we serve. I think at the
core of the wrap-up that I can give you is to talk about the opportunity where
we can play new roles that create huge leverage for each of us, each in our own
way, depending on the technology that we bring and the resources that we have
at our command.
Now, it wasn’t so long ago that we used to talk about the fact that 80 percent -- some large fraction -- of what we design went into the computer. Back at the time I came into the industry, seven years ago, it was probably a PC. Well, it was no wonder that there was so much talk about our being tied to a single industry, and our vulnerability to the fluctuations of that particular industry, whether it was the semiconductor industry or perhaps the PC industry. Well, that’s changed a lot. Any industry that serves a single customer is vulnerable. Any industry that serves multiple interests has more opportunities to develop its own mission in the marketplace.
Source: Cadence Design Systems
So where does it go from
here? What does this mean for us over
the next year? Or the next ten
years? It would take a Ray Bradbury
rather than a Ray Bingham to predict what might happen on the basis of this over
the next ten years. But look around you
today. You’ve all heard this. In fact,
most of you are in the middle of living it.
I arrived this morning on
a plane from Asia, where it seems that everyone you see is living the
electronics dream, is living some of the new drivers that create opportunities
for us in the marketplace.
In Taipai, for instance,
you can read the morning newspaper and check your bank balance on your cell
phone. I saw a lot of that in Tokyo as
well. In Paris, waiters in those great
restaurants take your order, communicate with the kitchen, and call up your
bill on a PDA (personal digital assistant).
In Zurich, an E-pass will send you through Customs; it will also put you
at the head of a ski lift line in the Alps.
In Rome, I’ve seen a city guide you can view from your cell phone, which
will give you access to the best and the most entertaining that the city has to
offer, in whichever language you would like.
You understand the picture.
You’ve seen it, you’re living it, but it creates incredible
opportunities for us that, given the things that we develop and produce, give
us an opportunity to participate.
The question for us is
how we participate, and how much will we participate. As the place where electronics begins, we offer the customer
something very incredible. So what
drives our opportunity? What makes our
success possible, and how good will it be for each one of us?
I think the nut of it,
and it’s been said in a number of different ways here this evening, is how do
you leverage the things that you bring to market? It’s about leveraging your tools on behalf of your customers and
leveraging services, whether they are consulting services, methodology
services, or the outsourcing of all kinds of design services. Taken together, these solutions represent a
differentiated capability, something that makes immeasurable difference and
value to your customer, rather than just selling those same guys your
products. That’s the difference, and
that’s the opportunity.
To achieve the twenty
percent growth opportunity, if you believe any of the things that were
presented by the previous speakers this evening -- from Moore’s Law to what
happens when you go into deep submicron, and you look at things like OPC
(optical proximity correction), the growth of the end markets, the kind of
prosperity that’s created by semiconductor manufacturers, twenty percent growth
ought to be child’s play. It ought to
be very easy if we participate in any way in the examples that have been given
here this evening. Now “child’s play”
is an exaggeration; it’s meant to get attention. But I think that we can deliver our tools and our services in new
ways that change how we participate in this very exciting world.
Let me just mention a
couple of ways we here at Cadence have developed our business to participate in
a different way. This is not unique to
Cadence, but think of the solutions that are being delivered, where before it
was an array of tools. Think about what
we call “synthesis place and route,” the integrated physical design capability
that is being delivered to customers and is leveraging additional value. It starts with that kind of thinking, where
you attack what formerly were clumsy ways of doing designs. Layer on top of this new design process
methodologies. Think about what you’re
doing with your customer as a process issue, not just a tool issue.
Moving on further, one of
the things that we talk a lot about is being responsible with our customers for
creating the design environment. Not
just the tool; not just a service; not just a solution, but a comprehensive
design environment. We do business with
a few customers on the basis of what we call a “virtual design environment,”
where we create an environment and a system to support the operation of that
environment front to back -- the whole process. Supplying the process, and helping to be responsible for the
output of that process. It’s a
different way of looking at things.
We’ve developed a considerable interest in it, and are being asked to
replicate it.
A third area is adjacent markets, and I think adjacency really means getting more involved with your customers and the things that they’re trying to accomplish. Think about it as a supply chain problem. Think about joining with your customers and looking upstream to their vendors, and downstream to their customers, and connecting with them in a way that matters and is relevant to them upstream and downstream from their business. Now all of this requires different capabilities, and there’s a different role implied in each of these things for each one of us depending on the technologies we bring, depending on the capabilities we bring, depending on the resources that we have available to us and the customer relationships that are there. There’s a bigger role in each of these if you think in a very different way about these drivers, this increased design activity, and these new kinds of demands.
Source: Cadence Design Systems
The final question that was asked has already been answered. These are the same Dataquest numbers across a slightly different scale. I think the bottom line has been stated in several different ways. Our customers are in the business of designing new products. Good times or bad, unit volumes or not, prices or no, the design work has to be done.
The more leverage that we
can bring with the design capabilities that we bring, the more we understand
their businesses, the more value we can bring.
And indeed, I’ve had the same experience that Wally Rhines has had -- I
don’t see designers being laid off.
It’s other parts of the supply chain that go first.
Question and Answer Session
John Barr: We have time perhaps for a couple of
questions from the audience if there are any.
Seeing none, I would like to ask just one question of the panel. Let’s have a number for growth for the industry
for 2001 -- percentage growth. Buno,
your thoughts on it?
Buno Pati: It’s a lot better than it is now.
John Barr: Okay, Wally?
Wally Rhines: 14 percent, and remember,
that’s an incredibly strong growth rate when you consider license model effects
averaged in there. You grow 14 percent,
that really is on an absolute basis, a very very strong year for the EDA
industry. But just put it in
context. I think 2000’s going to be
around a ten percent year.
John Barr: Ray?
Ray Bingham: I think that even
acknowledging the licensing model differences and where each company is in
that, we’ll see higher growth than that.
Higher than 14 percent for the industry as a whole, approaching 20
percent.
John Barr: Steve?
Steve Carlson: I’m not sure how I feel about
it. I actually agree with Wally that 14
is probably realistic number for this year.
John Barr: And Aart?
Aart de Geus: Well, the head of our legal
staff just shook his head and said, “Don’t answer!”
John Barr: I want the number!
Aart de Geus: It’s going to be, uh,
significantly better than last year.
John Barr: Okay.
With those insightful numbers, we will come back to you next year and
we’ll see how we did. I want to thank
the panelists for their participation tonight.
I think it was a message of growth beyond core businesses. If I look back last year, it was license
models. If I look back on how I spent
my time in the last year, it was trying to explain those licensing models to
investors, and you spend a half hour doing that, and you’re kind of numb. We’re glad to be beyond that, and look
forward to a great year. Thanks to
everybody.