新思科技 (SNPS) 2003 Q2 法說會逐字稿

Ladies and gentlemen, thank you for standing by.

Welcome to your Synopsys second quarter fiscal 2003 earnings conference call.

At this time, all participants are in a listen-only mode.

Later we will conduct a question and answer session and instructions will be given at that time.

If you should require assistance during the call, please press star 0.

Today's call is scheduled to last one hour.

Five minutes prior to the end of the call, I will alert the conference of the time remaining.

As a reminder, today's call is being recorded.

During the course of this conference call, Synopsys may make predictions, estimates, and other forward-looking statements regarding the company.

While these statements represent the best current judgment about the company's future performance, the company's actual performance is subject to significant risks and uncertainties that could cause actual results to differ materially from those that may be projected.

In addition to any risks that may be highlighted during this conference call, important factors that could cause the company's actual results to differ materially from those that may be projected in this conference call are described in the most recent 10-K and 10-Q reports of Synopsys on file with the Securities and Exchange Commission.

In addition, Synopsys would like to advise you that financial and other statistical information to be discussed on this conference call as well as the reconciliation of non-GAAP financial measures discussed on this call to GAAP financial measures is currently available on the company's website.

The web address for such information is www.Synopsys.com/corporate/invest/.html.

Aart De Gues, Ph.D.:

At this time, I would like to turn the conference over to you, Mr. Aart De Geus, Chairman and Chief Executive Officer.

Please go ahead, sir .

Go ahead, sir.

Are we ready?

Operator?

Operator?

Is the conference ready?

Operator?

Operator?

Is the conference call ready?

Okay, I guess we just have a message that the conference is ready.

This is Aart De Geus.

I have with me Steve Shevick, our CFO.

Let me start by apologizing for the delay.

Due to the overwhelmingly surprisingly large number of participants, I think the phone company we've been using has been totally overwhelmed.

And so my apologies for that.

We are clearly shifting suppliers on this one.

With that if you don't mind, let me go straight into the call.

I'll preface it as we have the usual disclaimers that you're well aware of and you're being spared with this time.

I would like to start by briefly reviewing our performance during the quarter.

Then I will give you an update on our end markets, our industry, our platform initiatives, and our competitive positioning.

After that, Steve will cover our financial results in detail, and give guidance for the next quarter and for the year.

We will conclude with Q&A.

Starting with our results, fiscal Q2 what the best bookings quarter in Synopsys' history.

Orders were significantly above plan.

And the resulting revenue of $292 million was above the top end of our target range of 275 to $289 million.

It was also significantly higher than the First Call consensus of $280 million.

The strong revenue results paired with careful expense management, lead to earnings before goodwill of 80 cents.

This was well above our target range of 67 to 72 cents and above First Call consensus of 70 cents.

Our market penetration improved in the quarter as customers purchased technology pools that were typically larger and more inclusive of our full product portfolio.

In Q2, we executed well.

And our results underscore the momentum we have been building with our recent product initiatives.

With this momentum, we expect to achieve results at the top end of our revenue and earnings target ranges for fiscal 2003.

Let's have a look now at our end markets.

Just five days ago, we held our sixth annual executive forum.

This intimate gathering was attended by over 50 senior executives from leading semiconductor and systems companies around the world representing the entire value chain.

This was an extraordinary meeting.

Each of the executives attending was very well informed on the enormous technology and economic changes impacting the present electronics markets.

To summarize some of the insights gained at the forum, cautious optimism best captures the tone of the meeting.

Visibility improved slightly this quarter when compared to a quarter ago.

A poll of attendees forecasted growth of 6 to 8% in the semiconductors for 2003.

Moving up to 11 to 13% in 2004.

With regard to technology roadmaps, a strong consensus emerged.

Forum attendees predicted unabated progress toward smaller geometries.

Learning from last year's tough lessons as the industry introduced 130 nanometers, there was little doubt with that with investments under way, we're on track to deliver the next technology node of 90 nanometers with fewer surprises.

SARS is still viewed as a wild card.

All travel has essentially stopped to and from mainland China, Taiwan, Hong Kong, and Singapore.

But the business impact has so far been mostly manageable.

The perspective of most executives at the forum was to take a wait-and-see attitude for the next three months.

Based on my interactions at the executive forum, as well as many conversations with customers during the quarter, I believe that visibility has improved slightly.

And that our end markets feel a bit more stable today than they did three to six months ago.

Although careful cost management will continue, we believe that companies are starting to refresh their IT infrastructures and we see some signs of life in WIFI, handsets, [MILLARROW] and automotive end markets.

As our customers have not fully adjusted to the post-bubble reality, their philosophy toward EDA vendors has evolved.

As you heard me predict quite a while ago, customers are visibly consolidating their spending with fewer EDA suppliers.

Many customers are turning to more complete solutions, both to save money and to decrease risk.

They're buying bigger pools of software from fewer vendors instead of devoting precious resources to cobble together products from multiple niche suppliers.

Synopsys strategy as a full-line supplier resonates well with them.

This is evidenced by our financial performance.

To sum up my market landscape observations, 2003 will be another tough year for our customers, but will definitely be better than '01 or '02.

The electronics and semiconductor industry feels more stable and buying behavior has begun to normalize.

Synopsys is benefiting directly from these positive trends.

Given this backdrop, let me tell you what we saw in the four major geographical regions.

Overall, our sales pipeline in the quarter was quite large, as we put extra field focus on mitigating the risk of further economic surprises.

Our sales teams did an especially good job.

Deal closure rates were above the historical average, resulting in orders exceeding plan in all four geographies.

In Q2, Japan led the way with a stellar quarter and was a big contributor to our bookings.

All the major contracts in our forecasts for Japan were successfully closed.

In North America, layoffs finally tapered off.

Buying has stabilized, and we have already closed some additional large orders in the present quarter.

Although ahead of our expectations, Europe feels the most low key of all the regions.

Many customers appear to be struggling with the currency effects of a strong Euro.

Lastly, Asia Pac even with the impact of SARS did fine.

India continued to be a bright spot for growth, as many of our global customers expanded into this region.

Let me now comment on the progress that we made in Q2 from a product perspective.

As you're all aware, early this year we started to roll out our integrated product platforms for design and verification.

Our February 3 -- on February 3, sorry, we introduced the Galaxy design platform.

On May 12, we announced the Discovery Verification platform.

The success of each platform first and foremost depends on the strength of its anchor technology.

In Galaxy, those technologies are synthesis, timing, and physical design.

No matter what the other benefits may be, customers will always choose a platform based on the strength of these anchor products.

Galaxy, which we expect to release in general availability later this year, includes at its core the anchor products of Design Compiler, Physical Compiler, Primetime, and Astro.

We have aligned these products with common timing, constraints, and libraries, and we have opened up the Milky Way database APIs to third parties.

Our Milky Way access program was created to facilitate access to Milky Way licenses and APIs.

If you will remember, it was launched with great interest from the electronics community and now have 28 licenses plus strong support from universities.

One of the initiatives within Galaxy that particular excite me is test automation.

In Q2, our test products had an excellent bookings quarter.

We also announced the addition to our SOC Bis product of a test automation solution for IP core-based designs.

This system on chip test solution comes with strong support from Arm, Agilant, and ST Microelectronics.

In Q2 we saw a significant competitive win for SOC based at ATI Technologies.

ATI is using Synopsys to implement testability within their next generation visual processing unit.

This design has over 200 million transistors.

Another very important initiative for Galaxy is signal integrity.

We have made excellent progress in addressing cross talk, noise, IR drop and electomigration.

Stay tuned for a major development in the next few weeks.

The product highlight for the quarter was undoubtedly our great progress in verification.

I'm talking here about the introduction of the Discovery verification platform.

Discovery successfully integrates capabilities not only to verify the functionality of a circuit, but also to automatically check any restrictions to its usage.

This represents a real step forward in verification technology.

VCS 7.0 to anchorpoint for the Discovery platform, went into general availability earlier this year.

Although the name 7.0 might imply this is just the next version in a long string of successful simulation releases, in reality, it housed a whole new era of verification.

The technology enabling this new era is assertion-driven design.

Just as the field of tests moved to design for tests a number of years ago, VCS 7.0 moves verification to the age of design for verification.

This transition couldn't come at a better time for our customers as they are already spending more than 2/3 of their overall design time on verification.

With the introduction of Discovery, we also announced Magellan, a new product for hybrid formal verification.

With the term hybrid, we mean the combination of two strengths.

Advanced, formal techniques that take advantage of assertions, plus our built-in VCS simulation engine.

With Magellan, designers can uncover complex bugs buried sometimes thousands of cycles deep in a design.

Magellan is already in use by advanced customers such as Invidia and ST Microelectonics.

What gives design for verification even more impact is the rapid emergence of System Verilog.

System Verilog is the new language of choice for RTL and assertion-driven verification.

Complimentary to systems seen at the higher level, System Verilog is 100% backward compatible with Verilog and has picked up the best capabilities of VHDL.

After 15 years of customers having to choose between VHDL and Verilog, System Verilog opens the new door to a new product cycle for users of both languages.

To accelerate the new product cycle for the whole EDA industry, we have put the full weight of Synopsys behind System Verilog.

One of the other strengths of Discovery is the tight integration of our analog and mixed signal products into the platform.

These products, including NanoSim and HSPICE are strong on their own merits and have great results in Q2.

We have been extending the capabilities of HSPICE to address high-frequency design.

This move has been well received by a number of our customers, including Xilinx who recently validated HSPICE for their 10 gigabyte per second serial IO design kit.

We're also seeing increasing usage of NanoSim integrated with VCS for mixed-signal verification.

This combination of strong products and tight integration is what makes Discovery so compelling.

Feedback on our oververification platform strategy has been overwhelmingly positive.

We clearly have a different shaded and demonstrably better solution than anything else in the market.

Three of the top 10 semiconductor companies are adopting this technology and we've already received enterprise-level orders for it.

During the quarter, we also closed the acquisition of Numerical Technologies.

The Numerical products when added to our existing OPC offering strengthen our position in the emerging field of design for manufacturing or DFM for short.

DFM did well for us in Q2 and the sales pipeline for the remainder of the year looks very strong.

Finally, to conclude this tour of our products, I want to highlight the large IP course available in our designware business.

We had a strong quarter in this area, and believe this business offers the potential for growth in the future.

On the competitive front, the main change we see is the continuing consolidation of smaller players.

In implementation, we compete first and foremost with Cadence, who is a large and well established base of software, and with Magma, who has a more narrowly focused product offering.

In verification, aside from the major market players of Cadence and Mentor, the small niche companies appear to be losing a bit of steam.

We find that many customers will try almost any product in the market once, but over the long term, most customers prefer to deal with a trusted partner that has a strong and broad offering.

We're gratified that we're increasingly chosen to be that partner.

In summary, Q2 continued our Q1 position as a solid and steady leader in turbulent economic times.

We moved forward on our objective to have customers choose us as the best overall solution from both the technology and economic risk perspective.

The number of substantial business transactions that we closed this quarter demonstrates our progress toward this objective.

Looking forward, we will continue to be financially disciplined and aim for the high end of expectations with good revenue and earnings growth in 2003.

Let me now pass the mike to Steve Shevick who will give you the financial perspective and guidance for the coming quarter.

Thanks, Aart.

First I will review the details of our Q2 results, and then I will provide our guidance for Q3 and the rest of the year.

Q2 was a very strong year for Synopsys.

We booked more orders in Q2 than in any other quarter in the company's history.

Total orders were well above plan and both up year-over-year and sequentially on an apples-to-apples basis.

Book to bill was significantly above one.

Orders linearity was the best it has been in four quarters.

Revenue linearity was consistent with prior quarters.

Notably, in three of the last four quarters, we have recognized less than 10% of total revenue in the last week of the quarter, demonstrating one of the key benefits of our ratable license model.

Orders were strong across all of the major product segments.

This is the result of our growing product momentum and the overall strength of our sales channel.

For the quarter, 74% of our product bookings were ratable licenses in the middle of our target range, and consistent with historical levels , and 26% were perpetual licenses.

As was the case in several quarters last year, we booked a number of perpetuals in the quarter that won't turn to revenue until the following quarter.

Q2 bookings included over $15 million of such bookings.

Dealer counted as perpetual licenses in the 74/26 split.

The average duration of the TSLs booked in Q2 '03 was 3.4 years.

A bit of above our target due principally to one large order in which a customer made a five-year commitment to a broad range of our tools, cementing our tools in their design flow for the long-term future.

Total revenue for the quarter was $292 million, $3 million over the high end of our guidance range, and $12.5 million over the First Call consensus.

As one would expect, since orders were well in excess of plan, and the TSL perpetual license split was well within the target range, we had both more TSL orders and more perpetual orders than forecast and the additional perpetual orders contributed directly to the higher level of revenue.

For the quarter, subscription license revenue was $148 million or 64% of total product revenue.

Revenue from perpetual licenses was $82 million.

Services revenue totaled $62 million, down both sequentially and from the same quarter last year.

This was expected, and principally reflects the continuation of the trends I discussed on our first quarter call regarding maintenance, as well as weakness in the training and consulting businesses.

From a geographic perspective, North America accounted for 43.5% of total revenue, Europe for 14.5%, and Asia Pac excluding Japan accounted for 7%.

As Aart mentioned earlier, Japan was particularly strong for us, accounting for 35% of revenue.

Our aggregate EBG operating expenses including cost of goods were $209 million, which was in the lower half of our target range for the quarter of 217 million to $214 -- 207 to $214 million, excuse me.

Commission expenses were higher than expected given the high level of shipments in the quarter.

These were offset by lower expenses elsewhere in the company.

Operating expenses included the expenses relating to employees of Numerical Technologies beginning after the closing of the merger on February 28.

Our EBG operating margin for the quarter was 28.5%.

We expect modest operating margin expansion over the course of the year and continue to expect that we will finish the year with a full-year operating margin of approximately 28% per our guidance last quarter.

Other income was $7.5 million near the low end of our target range of 7 to $10 million.

Other income included one-time gains from the sale of assets of $3.8 million, and rental income of $2.7 million.

We also took a $1 million write-down of an investment in our venture portfolio.

We expect to have investment gains in Q3 but not in Q4 or beyond.

Earnings before goodwill amounted to 80 cents per share, up 105% from the same quarter last year, and 8 cents above the top of our guidance range, driven by orders and revenue in excess of plan.

As a reminder, EBG represents earnings on a diluted basis excluding amortization of intangible assets, in process R&D and other merger-related expenses.

This quarter, net income on EBG basis excludes certain charges relating to our acquisition of Numerical Technologies including $18.3 million in in-process R&D.

Transition expenses relating to Numerical employees who will not become permanent employees of Synopsys are included in operating expenses.

We had no extraordinary legal expenses during the quarter.

A reconciliation of EBG to GAAP appears in our earnings release.

The diluted share count for the quarter was 76.5 million in the middle of our guidance range.

Payment terms continue to be an item of significant negotiation with customers.

Nonetheless, payment terms during the quarter held steady from the last quarter.

Operating cash flow for the quarter was very strong based on better-than-expected collections on all fronts, especially on licenses sold in Japan.

On a GAAP basis, operating cash flow was $120 million.

If you back out $22 million of cash disbursements related to accrued liabilities for various acquisitions and their related costs, the largest of which were $16 million to settle with the Avant!'s landlord and $5 million in fees relating in the Numerical acquisition, operating cash flow on an EBG basis was $143 million.

Cash and short-term investments continue to be strong with our 2Q balance ending at $392 million or $5.12 per fully diluted share.

The net cash cost of the Numerical acquisition was approximately $161 million.

During the quarter, we repurchased approximately 1.5 million shares of Synopsys common stock for an aggregate purchase price of $67.8 million at an average price of $43.86 per share.

As of the beginning of the third quarter, approximately $432 million remained in the repurchase program authorized by the board.

We expect to continue repurchasing this quarter, as long as prices remain attractive.

In addition, the 10-B-51 program we implemented last quarter remains in effect, allowing us to repurchase stock during periods when the window otherwise would be closed.

Q2 accounts receivable totaled $246 million.

An increase of $25 million over last quarter, reflecting our strong orders performance for the quarter.

DSO was 77 days, as expected, which was up slightly from 75 days in Q1, but in line with historical levels.

Deferred revenue at the end of the quarter was $477.2 million, up 20% from the same quarter a year ago, and up $52 million or 12% from the end of Q1, due to stronger orders and better than expected payment terms on TSL's.

As you know, deferred revenue represents only the on-balance sheet portion of our gross backlog.

While deferred revenue is one indicator of our performance, internally we focus more on the overall gross backlog which represents the future revenue from customers that is locked in.

We expect continued growth in the gross backlog, both year-over-year and sequentially throughout the year.

Consistent what we reported last quarter, during Q2, we did not enter into any so-called, non-linear deals or into any agreements in which the total amount of revenue to be recognized from a customer has been reduced or the schedule of such recognition has been pushed out.

Head count totaled 4,353 employees at the end of the quarter, up from 4,161 at the end of Q1 and reflecting our acquisition of Numerical.

At this point, I'd like to describe our expectations for Q3 and the full fiscal year.

We are very pleased with our performance in the second quarter and have increased confidence that we will finish the year at the high end of our revenue and earnings target ranges.

Therefore, for Q3, we expect revenue of 288 to $303 million, total expenses of 207 to $214 million, other I&E of 6 to $9 million including $7.3 million in gain on the sale of investments, outstanding shares of 75 to $79 million, a tax rate of 32.5%, and EBG in the range of 77 to 82 cents per share.

Approximately 80% of our target revenue is scheduled to come out of backlog.

We expect that orders for perpetual licenses will comprise 20% to 25% of our total license orders in Q3, and that the duration of subscriptions will be roughly 3.3 to 3.6 years.

This is because given our visibility into the Q3 pipeline, we think we will close a few key enterprise deals with slightly longer durations.

We're excited about entering into more strategic talks with our customers because it indicates that customers are prepared to commit to our methodology for the long term and to build their design flows around our full suite of products.

In most cases, these discussions involve explicit plans by the customer to reduce their commitment to other vendors.

We believe this demonstrates our ability to gain share of EDA budget within these respective accounts.

As a result, we are slightly broadening the range for our target average duration of subscription licenses for the full year to be between 3 years and 3.35 years.

For the full year, we expect to be at the high end of our previously stated revenue and earnings ranges.

These ranges were revenue in the range of 1.13 to $1.18 billion, earnings before goodwill in the range of $2.95 to $3.10, orders for perpetual licenses between 20 and 25% of total license orders, reported revenue consisting of approximately 21 to 27% product revenue, 51 to 57% subscription revenue, and 19 to 25% services revenue.

In summary, until we see a major upturn in our end markets, we will continue to manage expectations prudently and conservatively.

That being said, Q2 was an excellent quarter.

Our revenue and earnings outlook for the year has improved.

We are now focused on hitting the top end of our guidance range.

Thank you for your attention.

Before we open for questions, my lawyer reminds me that what Aart meant to say was that during this call we may make forward-looking statements.

And you should review our reports on file with the SEC for the appropriate cautionary statements.

Now we can go to questions.

Operator?

If anyone does have a question, you may press star 1.

And I'd like to begin, the first question will be Jay Vleeschhouwer.

You have the floor.

Go ahead, please.

Thanks.

From Merrill Lynch.

A couple of questions for you.

Steve, in fiscal '02, you had one customer that accounted for about 8% of your revenue.

And the question is, do you expect that, with that particular customer you could meet or exceed that same level of revenue which would be about 70 to $75 million this year and again, perhaps, next?

Secondly, product question for Aart.

You mentioned that Galaxy would go to GA later this year.

Is that a change at all in the schedule for that platform?

And also, could you talk about any issues with respect to the timing or handling of the Apollo end of life in a follow-up?

Thanks.

Let me start, Jay, first on the 10% question.

The answer is it's hard to say.

That customer's contract is ongoing.

At the same time, we continue to be exploring new business with that customer.

So whether it remains a 10% customer or not depends on too many different factors at this point.

Aart De Gues, Ph.D.: Jay, regarding Galaxy, yes, it is a couple of months later than what we had projected.

I think last summer, on a fairly sketchy information.

At the same time, though, regarding the Apollo end of life thing, I wouldn't be too quick about that.

No matter what a product stint will last a long time.

We will continue to support all of the users that are using Apollo.

However, the move, upgrade from Apollo to Astro is going extremely well.

And we've now moved over 60%, have already been upgraded.

From that perspective, Astro is our main platform.

Just a follow-up question on bookings.

Were there any customers that were over 10% of bookings in the quarter or perhaps even as much as a quarter of bookings in the quarter as you've seen in some proceeding quarters with other large customers?

There were three customers over 10% of bookings.

Okay.

Thanks, Steve.

Sure.

You're welcome.

Thank you.

The next question is from Garo Toomajanian.

You have the floor.

Actually he just dropped off the line.

We'll wait until he calls back in and then we'll take his question.

The next question is from Sumit Dhanda, you have the floor.

All right.

Congratulations on a good quarter.

A couple of questions.

Steve, first for you, you mentioned that you pushed out about $15 million in orders.

If I heard that right on perpetuals.

What were the percentage perpetuals have been if you excluded that number?

And then on the payment terms, you noted that it's steadied.

Can you talk about what percentage of customers are paying within a year and -- you know how it breaks out for the duration of the license?

And then I have quick follow-on.

Okay.

On the percentage of perpetuals, it would have been lower but maybe by a point or two.

I can get back to you on that one.

I just -- would have been somewhat lower.

But, you know, not significantly below.

The other question was on payment terms.

As I said payment terms have been relatively stable.

In general we've gotten I believe it's somewhere around between 40 and 60% in the first year, on average on our contracts.

And it's been in that range closer to the 60% range.

Okay.

Very good.

And then, I'm just curious, seems like Q2 was obviously a record orders quarter.

Q3 looks good.

The pipeline looks good.

But, you know, essentially you've just moved us to the higher end of the guidance range.

Is it just playing it close to the vest given the tough economic environment, or are you not anticipating the traditional ramp that you see in Q4 every year?

I think that we're taking as a snapshot now, and at this point, does not make sense for us to take the range up.

We did, as you know, put you to the higher end of the range on both earnings and revenue.

But given the environment, taking up the range doesn't make a lot of sense to us now.

Okay.

Thanks again.

Thank you.

We do have Garo on the line.

You have the floor.

Thanks.

Sorry about that.

Aart, one thing that you may have learned a little bit about in your executive forum, and something we've been hearing about is the possibility that customers might be skipping the 130 nanometer node and going straight to 90 nanometer.

Are you seeing ant of that going on and are you ready for that?

Aart De Gues, Ph.D.: I have seen none of that.

This is the first time I ever hear about this.

And by the way, I don't think that is going to happen at all.

The move from 180 to 130 offers many, many great benefits already.

And so we see that a large percentage of the designs that we touch are actually being done on the 130 nanometer, actually about somewhere between 40% and 50%.

So I think that that will fill the fabs first.

We do see a significant number of 90 nanometer designs, but I think they're still really leading state of the art if I can use that term.

Okay.

Also, going into the whole semiconductor downturn, the ED industry did hold up a lot longer than the systems in semispace.

When these customers start to see some improvements, do you think that that will translate into additional order strength for Synopsys in the industry, and do you think there will be a lag between what happens in the customer base and the EDA space?

Aart De Gues, Ph.D.: Well, I think it's fair to say that EDA has always been relatively smooth compared to the mega ups and downs of our customers.

And I would expect that to continue.

Having said that, though, it is clear that if our customers feel better, they will be more interested about investing for the future.

And there's quite a number of customers that see the opportunity to, for lack of better word catch up with the opportunities of 130 nanometers and do some really exciting new products.

And so all of that feels pretty good to us.

The customers are cautious because they're coming out of two really tough years.

But we are seeing a few little rays of light.

Okay.

And lastly, I know that the subscription rate meaning the bookings less subscriptions was pretty much on plan, and you also mentioned that Japan was particularly strong.

I'm wondering if there's been sort of a change in the sentiment there, where they have historically preferred perpetual licenses.

Are they doing more time-base and TSLs, as well?

Aart De Gues, Ph.D.: You know, I think for most of the world, people have become totally accustomed to the notion of time-based licenses and then from case to case, they determine based on their own needs what is most appropriate.

And so in many ways, I think we've found sort of a point of stability that in my own mind I give a pretty big range to, you know, somewhere between 15 and 30% were perpetuals.

In reality, it's been sort of fairly narrow range that seems to be working well.

Great.

Thank you very much.

Okay.

Thank you.

The next question is from Bill Frerics.

You have the floor.

One moment.

I'm sorry.

An expansion of the one that you just answered for Garo.

There was the fact that you had hit almost 100% of your major bookings targets in Japan, the fact that perpetuals were so strong, was that a reflection of that geography?

Aart De Gues, Ph.D.: Not only.

I think Japan did well on every account.

It was clear going into the quarter that we had a number of big opportunities in Japan.

It was also clear at that time, remember, we're now talking early January, that the economic landscape was filled with a lot of questions.

And so I think our sales team executed particularly well to make sure that if there were more economic surprises, we would be well covered.

The reality is they executed especially well in Japan, and all of these transactions came in.

And so we feel pretty positive about how Japan is looking at Synopsys right now.

Okay.

And did Q2 borrow any budgeted business from what you might have envisioned for Q4?

In other words, when you got to the table, with certain customers, did you essentially manage to contract for what had been maybe a larger, more extensive budget?

No.

There were no full forwards.

We had a great Q2, and the pipelines for Q3 and Q4 both look strong.

Okay.

Thank you.

You're welcome.

Great.

Thank you.

The next question is from Raj Seth.

You have the floor.

Go ahead, please.

Thank you.

Aart, I wonder if in the implementation space where you mentioned Cadence and Magma as your primary competitors if you could characterize the competition there little bit further.

Perhaps where is that you're seeing strengths, where is it that you see Lava and Cadence strong?

What's happening in that space?

Everybody's sort of claiming to be gaining share.

Can you help us there?

Aart De Gues, Ph.D.: Yeah.

I think everybody makes bold claims.

The reality is things change gradually.

We do see benchmark.

We see a lot of interest in the new technologies.

We see a lot of interest in some of the capabilities necessary to go to 90 nanometer.

But maybe the strongest effect to our advantage is the fact that a number of customers are looking at us as a broad portfolio supplier, that on one hand can absolutely win in many situations on the strength of technical benchmarks.

On the other hand, clearly provides a much lower risk, strong solution.

So you will not hear too much clarity from them as to who they're picking, because as they shift some of their business toward us, they want to make sure that these other suppliers keep supporting them.

So at the end of the day, you know, market share shifts are measured in revenue.

And that takes a little bit of time until it's visible.

At last year's DAC, I think it was last year, you put up a 2 x 2 when describing Magma.

I think it was size of the design, speed of the design, and didn't put them, or didn't view them as competitive in the upper right-hand corner of that box.

Has that changed at all because Magma obviously has some momentum, you know, what's going on there?

Aart De Gues, Ph.D.: We see them as a competitor.

That is good for the field.

And we are taking them on at every opportunity.

And so in that sense we never discount any company that is successful.

At the same time, we do see that Synopsys has been doing quite well.

Okay.

And last question for Steve.

Steve, in the second half, how do I think about cash flow given the changing terms, et cetera?

Yeah, well, cash flow was obviously excellent in Q2 based on strong collections.

And payment terms were pretty much ahead of our forecast.

I think at this point, we're not ready to extrapolate for the rest of the year.

And so the way we model will continue to model relatively conservatively because the pressure on payment terms does remain.

If we continue the same performance, I think we'll have a great cash flow year.

If it goes down to where we've modeled, we'll still have a good cash-flow year.

What's a good cash-flow year for the year?

I'll tell you at the end of the year.

Okay.

Thanks.

Okay.

Thank you.

We have Jay Vleeschhouwer again.

You have the floor.

Go ahead.

Thanks.

Another follow-up on the product side.

For Aart or Steve, can you describe in some more detail how PC is doing either as a stand-alone product or in association with Astro as a bundled sale?

You've had comments along those lines in preceding quarters such as Q4.

Also, Aart, what's the update on your having a new and improved floor plan compiler?

Is that still an important objective for you for the year?

Aart De Gues, Ph.D.: Okay, let's start with PC and Astro.

Both of these products had their second-best quarter ever, actually close to their best quarter.

So in that sense, we are moving forward, well with them.

Regarding the floor plan capabilities, we've made good technical forays moving forward.

We've also learned that both of the products that we have have a strong capabilities that appeal to different market segments.

And so our pathway will be to proceed more towards an increased integration.

In all cases, the product bookings are up year-over-year.

Okay.

For Steve, earlier in the year in January, you recall that you gave a presentation in which you talked about the percentage of the subsequent three or four quarters revenue then in backlog, you said that Q3 has 80% of revenue in backlog.

Can you talk about the percentages subsequent to third quarter?

I don't think we're going to give out a number on rolling four quarters basis.

But you know, in general, the backlog tends to amortize at a relatively consistent rate no matter which quarter you're in.

Operator, next question, please.

Okay.

The next question is from Jennifer Jordan.

Jennifer, you have the floor.

Go ahead, please.

Yes, Jay just asked a couple of my questions.

But I just want just to first -- when you're defining book to bill and you say you have a positive book to bill, how are you defining it?

Just the standard way.

Well, is it six months or is it all your bookings total over the bookings for the single quarter?

It's the quarter.

Okay.

Quarterly basis.

Quarter bookings, quarter bill.

Okay.

Then the second question is you mentioned a little bit the floor plan compiler and Jupiter and what you need to do with those products to bring them together and bring out a new product.

What can we anticipate as we look to the Design Automation Conference for new product announcements?

Aart De Gues, Ph.D.: Well, you know, to anticipate that you sort of have to go there and actually we have a session set up specifically for the analyst community.

And so, give us the room to keep that quiet until we're there.

Okay.

You're going to hold your fire.

Aart De Gues, Ph.D.: Trying to.

Thanks, Aart.

Thank you.

The next question is from Rich Tortoriello.

You have the floor.

Go ahead please.

Thank you.

I just wanted to go back to this guidance question that's been asked a couple of times.

I guess I'll just try a different way.

It seems like you're basically guiding for a sequentially flat EPS in the fourth quarter.

So really no seasonal uptick.

Is there anything structural that's changed about the business that would preclude a seasonal uptake, or is this really just conservativism because it's a tough market?

Good question, Rich.

As I think you might have heard, what I said was we would have investment gains in Q3 but not in Q4.

So other INE will actually be down significantly in Q4 from Q3, which obviously affects earnings.

On a -- however, on a raw earnings basis, I'm sorry, on a raw revenue basis, we will see revenue uptick sequentially from Q4 to Q3, and then operating earnings as well, will increase from Q3 to Q4.

Okay.

That's very helpful.

And Aart, I was wondering if you could just talk about, amongst the different sort of, classes of your customers if you think of IDMs, maybe Asic, and then COT, is there any differences amongst the strength of those or any of them seeing that you're seeing that are maybe slightly stronger than others in terms of the rays of light that you're seeing in terms of recovery?

Aart De Gues, Ph.D.: I don't think that we can actually point to very big changes.

The reason why I'll get into end market specific instead of how the designs are being approached.

I think it is clear that earlier last year a number of the COT companies had a little bit of a tough time because the execution on 130 nanometer didn't quite come about as easily as they had thought with their suppliers.

It seems that those problems are out of the way and that is moving forward again.

If I were to rank order the strength, I would put IDM first, COT, second, and Asic third.

But that's just a gut feeling, I don't really have very hard evidence.

One final one if I could.

It sounds like there's a lot of enthusiasm there for the VCS 7.0 product.

I am wondering if you can comment on specifically head to head with Vericity who would seem to be square in the target of this, if you could comment on if you have been able to displace it or sort of competitive environment relative to simplicity now that you have VCS 7.0 as part of the portfolio.

Aart De Gues, Ph.D.: Regarding replacement, the answer is in some cases, yes.

I think it's a good company that has provided some powerful technology.

At the same time, what we're talking about here is a much, much broader value proposition.

That encompasses not only the test bench features, but many aspects of formal verification, coverage metrics, and most importantly, the full simulation engines.

So the combination of all of these things in isolation is by no means as powerful as when they are brought together under the umbrella of design for verification.

And that is really where we're heading.

And I believe that message has been extremely well received by people that have huge investments in verification.

And that's why we are seeing such a rapid adoption of these technologies.

Okay.

Thank you.

Aart De Gues, Ph.D.: You're welcome.

Okay, thank you.

The next question is from Alexander Guana.

Alexander, you have the floor.

Go ahead, please.

Thank you.

Aart, I was wondering if you could give further color to the positive comment you said about IP.

Is there any particular type of IP that's particularly a powerful for you right now?

In an industry shift right now that you -- with the tough times in semiconductors they need more help?

Could this possibly be a precursor to more of the service revenue coming back, as well?

Aart De Gues, Ph.D.: Well, I think you know we've been talking about the IP concept for quite a number of years as being a necessity to do large designs in any form of reasonable time frame.

The hurdles have been, are the tools ready, are the IP formalism ready, so on.

And Synopsys has invested in this design methodology for a ling time under the general label of designware.

Now the blocks that are in designware have gradually become more and more powerful and broad.

And the IP successes that we see right now are very much focused around standard blocks that everybody needs, that are relatively difficult to do, and that can be utilized on a broad set of products.

And so USB, PCI, are sort of the names that come to mind because those are the interface blocks that I used in products I need to communicate with each other.

And we provide many of these interfaces, and they are at a very high level of quality.

So I think that will continue as we see these interfaces improve and evolve over time, and as new ones are showing up.

Is there any particular process geometry that you're providing the IP at right now, and any particular emphasis in terms of migrating it to a finer geometry?

Aart De Gues, Ph.D.: I would say that's probably the bulkiest is 180, 130, and now moving rapidly down to 90.

And the reason for that is, as you move to the smaller geometries, you simultaneously move to much more complex chips.

As you move to much more complex chips, the need and the desire for IP reuse goes up dramatically and quickly.

And in addition, if we can provide sophisticated blocks that work really well at these geometries, it decreases the risk to our customers substantially.

As all of that is coming together rather well.

And I was wondering if you were to hazard a guess as to where you might be pleasantly surprised on the upside in the back half of your year here, what area's really showing the greatest untapped upside opportunity right now?

Is it in the verification, is it in the analog mixed signal, is it improvements to floor planner?

Where is that you'd like to really hit the home run?

Aart De Gues, Ph.D.: Well, in many ways, the home run has to be the macro picture because although we are very excited about some of the new adjacent markets such as design for manufacturing and ID, although we're very excited about some of the new capabilities in test, clearly the whole story about verification I think is very fundamental to the future.

The more fundamental macro effect is the fact that our customers look at our complete portfolio and say, this is a portfolio that on average is very strong from a technology point of view.

A company that has provided good support over many years, and therefore, by definition, is lower risk from a technical and an economic point of view.

And so as we move to the smaller geometries, the combination of all of those things become more important, and although that makes my answer to you a little diffuse, the reality is overall menu here I think is rather complete.

Well thank you and congratulations.

Powerful quarter.

Aart De Gues, Ph.D.: Thank you.

Thank you.

The next question is from Wiliam Frerichs.

You have the floor.

Go ahead, please.

This is a follow-up.

From time to time, you've given a metric of the value of renewals over the deals they replaced.

First of all, I noticed some of the deals were longer, and so it may not be possible to do, you know, time-based apples-to-apples comparison.

Were the new deals significantly up?

And secondly, were you able to take advantage of the respinning Avant! deals that may have expired?

Bill, let me answer that two ways.

So on the renewal deals, I think if you have to compare really the deals that we do this quarter versus the base deal for the customer.

Which may have happened a few years ago.

And in every case, those are up significantly.

What happens between the last deal and this one is that the customer buys incrementally for shorter term need, and then it gets a lot harder to figure out which deal belongs to which time period, et cetera.

But in general, the deals we did were significantly bigger than the deals they replaced.

In terms of the length of the deal and how that relates to size, to the extent that you're getting into an arrangement where you're providing a significant amount of the customers design flow, you would tend to have a larger deal because you're taking a larger chunk of their budget.

I don't know that the duration itself has much to do with it.

It's really the degree of importance that the customer is putting on you in their design flow.

Okay.

And have you seen opportunities to -- when you are about to hit a termination of an original Avant! deal to open that back up, and add a lot more product and maybe negotiate a deal that will -- where all products regardless of where they originally came from would expire, say, in three years or so?

Sure.

The Avant! contracts have been a great source of opening and extending relationships with customers as they've come up or as the related Synopsys contract has come up

Okay.

And so they give you more to talk about.

When have you been through the fat part of those opportunities, or are there more to come?

I think, you know, there are more to come out there.

We've been through a number of them, but there are plenty to go.

Okay.

Thanks.

Okay.

Thank you.

The next question is from Mehdi Hosseini.

Go ahead, please.

Thank you.

I have several questions.

The first question I have is more macro related.

Aart, if you could help me try to understand some of the value propositions that have gone into the 90 nanometer technology, you know, provided.

I hear from fab that they are just improved the yield at 130 nanometer.

Some of the economic benefits would just begin to realize at that technology node.

So what is it for some of your customers that would make them to go beyond 130 nanometer, and move into the 90 nanometers?

Aart De Gues, Ph.D.: Well, fundamentally, the drivers to go the next generation have not changed in the last 20 years.

Which is, you know, Generation N-plus-one, always brings some additional risk, but it brings at least two or three benefits.

One of the benefits tends to be that you can have higher speed products.

The second benefit is that because you can do more complex chips, you can integrate more functions than before.

In other words, things that on the board may have taken three chips before now take only one chip.

And last but not least, and this is very important for all the people that do very high volume, they are extremely sensitive to the area of each chip because the cost is so related to that.

And so cost is a strong driver for people that have high volume.

Now, it is absolutely true that when you have a generation, there's a whole bunch of people that say, you know, let me use the existing one, not the most advanced one because I don't want the risk and I want to just ride the experience curve of the existing generation.

So a market tends to split up in the most advanced kinds.

The mainstream and the laggers, and they all have sort of their own technical and economic combination of drivers.

Sure.

Now, would it be fair to say that a number of Asic designs has actually been coming down?

And that's one area that you will see more need for moving into 90 nanometer?

Aart De Gues, Ph.D.: No, I would not correlate to two, I think the number of Asic designs as coming down.

But the number of transistors on a given Asic design has gone up dramatically.

And so the most -- the hardest driving people on technology are really the people that have very advanced chips.

Be it processor cores, ESB cores, graphics cores.

Those have the characteristic of on one hand taxing the silicon technology to the max, and on the other hand, being high volume to boot.

So, you know, that's a good recipe for wanting to be at the leading edge.

Sure.

And then, you know, Japan performance so great in the second quarter.

But you can't ignore the fact that U.S. nearly accounts for half of your revenues.

So would you expect that with this kind of a bullish outlook for the third quarter, would you expect the booking or business from U.S. to pick up in the latter part of '03?

Aart De Gues, Ph.D.: Every time we prepare for the earnings release, we debate if we should give any geographical data anyway because part of the problem is that these things vary greatly from quarter to quarter as a function of what deals are closed.

And so I think the macro picture here is to more look at a four or eight quarters running tally, and there you are absolutely right.

The U.S. as it should accounts for the largest piece of market, with a bit over 50%.

Japan, Europe, and Asia Pacific split the rest.

And the variations from quarter to quarter can be huge.

Having said that, yes, one wouldn't expect Japan to be, you know, one of the lead contributors two or three quarters in a row.

And so ergo the other geographies will make up for the flat.

Sure.

And one last question.

I realize that revenues from consulting services is down to like 1 to 2%.

Is that -- is it the bottom of -- for that part of the business, would you expect any pickup to be just for any color on that?

I would appreciate it.

Aart De Gues, Ph.D.: Well, by definition, 1 to 2% sort of is the bottom, it's hard to go much lower than that.

And, you know, it's really an not surprise given the economic landscape around us where in the last two years, many of our customers were down in '01, 46%, in '02 about 33%.

So they were not looking at buying a lot of services when they had to lay off their own people.

Having said that, I think there's a piece of our service business that is actually very good.

Which is that we have consulting services around methodology that really help our tools.

And you know, we think that this will be a steady piece of revenue for us.

Thank you.

You're welcome.

Okay, great.

The next question is from Erach Desai.

You have the floor.

Go ahead, please.

Hi.

I had a couple of questions.

One was, Steve, can you give any color or any granularity to what Numericals revenues were in the second quarter?

Numerical's revenues contribution was very small, as we had expected.

Well, I understand very small relative to your larger numbers.

But --

So small as consistent with what we had expected, yeah.

But small, absolutely, as well.

Okay.

Five million, 10 million?

Under five.

Okay.

And related I guess in a sense, I'm trying to -- I don't have my model in front of me because I'm traveling.

I'm curious you did something like -- correct me if I'm wrong, $148 million in subscription revenue in F2Q.

What was the number in the first quarter, was that 143?

141.

141.

So part that because Numerical was essentially at a subscription model, parts of that was Numerical?

Part of the increase?

Correct.

A very tiny piece of the increase.

Remember, actually none of the increase I'm told.

So none of the increase was Numerical.

The Numerical was all turns orders?

Okay.

The contribution from Numerical was terms orders, yeah.

And -- fair.

Second question is a clarification on what Jennifer asked.

I just want to be sure that I understood it correctly.

When you defined book to bill and you talk about bookings in the quarter, I just wanted to clarify.

Bookings that are going to be recognized in the quarter, or going to be recognized over multiple years?

It's total bookings in the quarter versus total revenue in the quarter.

But bookings that can be recognized from --

Whenever.

Whenever.

Exactly.

Just like Cadence does.

Okay.

And then, curious.

Some mixed sort of data points coming from one particular large semiconductor customer in Texas.

And this -- apparently Magma got a pretty significant win, and also expanded beyond their sort of traditional Asic focus.

I guess there's also rumblings that you guys had a good quarter over there.

I'm just trying to reconcile all the different set of data points that are coming on that front.

Any color you can provide?

Aart De Gues, Ph.D.: Well, we never comment about individual customers unless we have explicit approval.

But it is fair to say that the Texas contribution has been very, very strong.

Okay.

And Steve, separately on sort of development with the SEC settlement with the major banks, I was just curious, Synopsys was brought up as one of the companies with respect to maybe one investment bank issuing some research that might have been tainted.

Just curious, have you -- has Synopsys been contacted with the -- by the SEC?

No.

No, no Synopsys executive to my knowledge was ever interviewed.

There has never been any suggestion of any Synopsys involvement in anything related to CSFB.

You suggesting something?

No, I'm just asking.

Can you, now when you guys had the $37 million fee, one time fee with regards to legal and banking et cetera, did you ever disclose what was paid to CSFB for the Avant! deal?

Not specifically.

I mean, I can tell you it was a standard investment banking fee.

Okay.

Thanks.

Thank you.

The next question is from Arnab Chanda.

You have the floor.

Thank you.

A couple of questions.

First of all, one of the things that was going on when Avant! was independent was a lot of customers were a little bit reticent in signing up full course.

I'm wondering if some of the, you know, the good growth that you've seen is coming from sort of expanding that base within the implementation area, which maybe helps you kind of outperform the industry.

Do you have any comments on that?

Aart De Gues, Ph.D.: Well, I think one of the premise, of course, behind the merger was that we are putting together a set of technologies that are extremely important for advanced design.

And we are executing well on that premise.

And I think customers have recognized that.

And so in that sense, yeah, are we positive about the acquisition?

Absolutely.

Do we have a lot of opportunity and work ahead of us?

Both of those are true, as well.

But all in all, I think that is going very well.

Great.

Next question about that, if you don't mind.

Now, on the other hand, you obviously had a great market share and have a great market share in this sort of standard synthesis area.

You have seen Cadence buy a next generation synthesis company, obviously, they've made an effort in the past.

You know, is there -- is -- are there some technologies that you think are, you know, you see a kind of transition as you've seen with physical synthesis where you've seen mergers between kind of old placement and route from higher up in the design cycle?

In other words, is there a technology node that allows the entrance of a competitor, would you say?

Aart De Gues, Ph.D.: Well, you know, I have never been a believer that technology nodes themselves bring discontinuities.

And the reason is that even if new technologies demand some new capabilities, customers, even like to adopt revolutionary technology in an evolutionary fashion.

If one maintains a strong set of products and builds from there, preferably in an adjacent fashion, that's actually a good strategy.

And we follow this literally for the last decade.

And so in that sense, I don't think that 90 nanometer or so drives any major discontinuity.

I do believe that some of the new techniques that we have in our existing products are a necessity for 90 nanometer, and we're doing especially well with those customers.

Great.

Last question.

If you could talk about, you know, what -- quantify the investment gains that you had this quarter so we can get a sense of sort of what will not be there, and in terms of modeling.

Thanks.

Sure.

This quarter, it's I think I say in my script, $7.3 million in gain on the sale of investment.

The overwhelming amount of that, if not all of it, is from the sale of our Artisan stock in their secondary offering.

After that sale, there's really no more equity investments that we have to sell.

So other IME will go down to essentially just interest income offset by effects and whatever other things flush through there.

So it should go down dramatically.

Aart De Gues, Ph.D.: Operator, at this time, we would like one more question and then we'll wrap it up.

Okay.

The last question will be from Jennifer Jordan.

Jennifer, you have the floor.

Go ahead, please.

Yes, Aart.

Just to go back, you mentioned earlier in the call that you have now upgraded about 60% of the customers who were using Apollo to the Astro next generation router tool.

At what point do you think or what kind of time frame do you give yourself to accomplish the remaining 40%?

Aart De Gues, Ph.D.: Oh, you know, the reality shows that the 40% takes a long time because some customers will remain on support, on Apollo for quite a number of years.

Having said that, though, we're really excited about the fact that the move is so rapid because Astro has many powerful capabilities, it's only for advanced users.

This is of high value.

So, you know, there's no such thing as 100%.

But close to that probably will take an overall average 45 years based on experience we've had from past products.

Do you have a sense of what the rate is to get like the next 10 or 20% done, just out of curiosity?

Aart De Gues, Ph.D.: Oh, yeah, I think that's going to be absolutely done within the next two to three quarters.

Okay.

Thank you.

Aart De Gues, Ph.D.: You're welcome.

With that, we'd like to conclude the conference call.

Thank you very much for attending.

We're looking back on a strong quarter, and we're looking at good visibility going forward.

And as usual, Steve and I will be available for questions.

Have a good rest of the afternoon.