益華電腦 (CDNS) 2017 Q2 法說會逐字稿

Good afternoon.

My name is Devin, and I will be your conference operator today.

At this time, I would like to welcome everyone to the Cadence Design Systems Second Quarter 2017 Earnings Conference Call.

(Operator Instructions) Thank you.

I will now turn the call over to Alan Lindstrom, Senior Group Director of Investor Relations for Cadence Design Systems.

Please go ahead.

Thank you, Devin, and welcome, everyone, to our second quarter 2017 earnings conference call.

With me today are Lip-Bu Tan, President and CEO; Geoff Ribar, Senior Vice President and CFO; and John Wall, Corporate Vice President of Finance and Controller.

The webcast of this call can be accessed through our website, cadence.com, and will be archived through September 15, 2017.

A copy of today's prepared remarks will also be available on our website at the conclusion of today's call.

Before we start, I want to call your attention to our CFO commentary, which was included in our 8-K filing today and is available on our website at cadence.com.

The CFO commentary should be referenced with both today's conference call remarks and the earnings press release issued today.

Next, please note that today's discussion will contain forward-looking statements and that our actual results may differ materially from those expectations.

For information on the factors that could cause a difference in our results, please refer to our filings with the Securities and Exchange Commission.

These include Cadence's most recent reports on Form 10-K and Form 10-Q, including the company's future filings, and the cautionary comments regarding forward-looking statements in the earnings press release issued today.

Also note that this afternoon, we filed our 10-Q for the quarter ended July 1, 2017.

In addition to the financial results prepared in accordance with generally accepted accounting principles or GAAP, we will also present certain non-GAAP financial measures today.

Cadence management believes that in addition to using GAAP results in evaluating our business, it can also be useful to review results using certain non-GAAP financial measures.

Investors and potential investors are encouraged to review the reconciliation of non-GAAP financial measures with their most direct comparable GAAP financial results, which can be found in the Quarterly Earnings section of the Investor Relations portion of our website.

Additionally, a copy of today's press release dated July 24, 2017 for the quarter ended July 1, 2017, and related financial tables can also be found in the investor relations portion of our website.

Today following Lip-Bu's remarks, John Wall will present the financial results and outlook.

Then Lip-Bu, John and Geoff will all be available during the question-and-answer session.

Now I'll turn it over to Lip-Bu.

Good afternoon, everyone, and thank you for joining us today.

We are steadily executing our system design enablement, or SDE, strategy.

SDE offers additional growth opportunity as we expand beyond semiconductors and tap into significantly larger market with system companies and vertical market segments such as automotive, aerospace and defense.

In Q1, we booked our largest design IP contract ever with a major customer in automotive semiconductor sector.

In Q2, this momentum continued with an ADAS system company licensing our PCIe Express Gen4 IP on new 7-nanometer SoC and another major customer licensing Tensilica for automotive radar application.

Also in the automotive space, ROHM adopted our ISO 26262-compliant functional safety verification solution.

And we received ISO 26262 certification for our PCB flow.

Overall, 4 of the top 5 automotive semiconductor companies are now using Cadence IP.

In aerospace and defense, adoption of our Palladium Z1 emulation system increased with purchases by several significant system customers.

Providing integrated system-level solutions is also a key goal of our SDE strategy.

In Q2, we released the Virtuoso System Design Platform, which optimize design integration between our chip, package and board flows.

We also expanded our partnership with MathWorks through a new integration between Virtuoso ADE and MATLAB that will enable customers to accelerate analysis of large data sets when verifying custom RF and mixed-signal design.

Digital and signoff revenue grew 14% year-over-year, driven by a growing proliferation with market-shaping customers.

Innovus is rapidly becoming the implementation solution of choice for CPU, GPU and SoC designs for networking, wireless, consumer, automotive and IoT.

In addition to the momentum on the digital products, there's increasing traction on the adoption of the full flow by digital and mixed-signal customers.

Our full digital flow is now used by over 70 customers performing advanced node design, including more than 20 full flow customers designing at the 7-nanometer nodes.

Next, I want to talk about our IP business, which is the key element of our SDE strategy.

As we have stated, the IP market opportunity remain strong due to growing outsourcing trend.

IP revenue grew 15% year-over-year as our refined strategy gains momentum.

Multiple smart speakers are using Tensilica processor for AI, audio and WiFi.

In Q2, we had multiple Tensilica Vision processor wins for drone, handset and industrial applications.

And we introduced the Tensilica Vision C5, it's the first dedicated neural network DSP IP.

Now for System Design and Verification.

While overall, hardware revenue was less than anticipated for the first half, we expect hardware revenue to be a little stronger for the second half.

Palladium Z1 remains the most advanced emulator on the market.

In Q2, we expanded our partnership with HiSilicon on Palladium Z1 with significant add-on on emulation capacity.

We are also pleased with the early customer reception of Protium S1 FPGA-based prototyping system.

Momentum continue as we received an important endorsement of our technology and integrated hardware approach, with a competitive win at the leading North American semiconductor company that is one of the largest users of Palladium Z1.

Overall, we have several repeat orders and 5 new logos.

Before turning over to John, let me quickly summarize my comments.

Consistent execution drove excellent financial results for Q2.

System design enablement is expanding our opportunity and extending our customer reach.

Software and IP were particularly strong, and proliferation of our digital and signoff solutions is growing with market-shaping customers.

We continue to innovate and introduce new products like Tensilica Vision C5 DSP targeted at neural network applications.

Now I will turn the call over to John to review the financial results and provide our outlook.

Thanks, Lip-Bu, and good afternoon, everyone.

Consistent execution drove excellent financial results for the second quarter, highlighted by revenue near the high end of our guidance range and operating margin EPS and operating cash flow all exceeding expectations.

Specifically, here are some key results for the quarter: total revenue of $479 million; non-GAAP operating margin was 27%; GAAP net income per share was $0.25; non-GAAP net income per share was $0.34; and operating cash flow was $162 million.

Also, please note that the recurring revenue mix was approximately 90%.

DSOs were 31 days, down 6 days from Q1 on strong collections.

Our DSO target remains approximately 35 days.

For geographies and products, Asia continued as our fastest-growing region, with revenue up 18% year-over-year.

As Lip-Bu mentioned, digital and signoff revenue was up 14% year-over-year as we benefit from proliferation with market-shaping customers, and IP continued to rebound from 2016 with revenue up 15%.

Functional verification revenue was down from last year as overall hardware revenue was less than anticipated for the first half.

However, we expect hardware revenue to be a little stronger for the second half.

Now let's turn to our outlook.

We are increasing our revenue and EPS outlook.

For fiscal 2017, we now expect revenue in the range of $1.91 billion to $1.95 billion; non-GAAP operating margin of approximately 27%; GAAP EPS in the range of $0.98 to $1.04; non-GAAP EPS of $1.36 to $1.42; and operating cash flow in the range of $430 million to $470 million.

For Q3, we expect revenue in the range of $475 million to $485 million; non-GAAP operating margin of 26% to 27%; GAAP EPS in the range of $0.24 to $0.26; and non-GAAP EPS in the range of $0.33 to $0.35.

Approximately 90% of revenue is expected to come from beginning backlog.

You will find guidance for additional items in the CFO commentary.

Next, I will take a moment to review our capital allocation priorities.

As we have said before, the company regularly reviews its capital structure, balancing our needs for investment, the appropriate level of risk for our business model and operating environment, maintaining adequate liquidity and the opportunity to return cash to shareholders.

In January of this year, the board authorized the repurchase of $525 million of our common stock.

We did not repurchase shares in the first half of the year, but we do expect to repurchase some shares in Q3.

I also want to take -- I also want to provide a few additional comments before we take questions.

As a reminder, hardware and IP have become a larger portion of our business, which may lead to more variability in our results from quarter to quarter.

Only about 5% of our revenue is in currencies other than the U.S. dollar, primarily the Japanese yen, but about 30% of our costs are in currencies other than the dollar.

So a weakening dollar would generally be a headwind to operating profits and, conversely, a strengthening dollar would be a tailwind.

The dollar further weakened in Q2, but so far, we have been able to manage through this challenge.

As you know, we have been reviewing the new revenue recognition standard that will -- that we will implement for 2018, and we are confident that we will substantially maintain recurring revenue or revenue over time treatment.

To conclude, we are pleased with our second quarter results, including strong financial performance, software and IP growth and growing proliferation of our digital and signoff solutions with market-shaping customers.

Looking forward, we are excited about the new opportunities resulting from our system design enablement strategy, and we are confident that we will continue to drive strong financial and operating results.

And with that operator, we'll now take questions.

(Operator Instructions) Your first question comes from Rich Valera with Needham.

First question relates to the decision to start buying back stock.

I think as recently as a few weeks ago in public appearances, the management were saying that you guys were going to be doing a 5-year strategic plan, and sort of at the conclusion of that you'd decide whether you'd be starting to buy back stock or not.

So I'm taking, based on your decision, that you've actually kind of concluded that plan.

So I'm wondering if there's anything you could share with us about that plan as it relates to opportunities for M&A on your side.

Presumably that you're buying back stock, I guess that -- you might conclude that you don't see much opportunity for M&A.

But just wondering if you can share anything about that 5-year strategic plan and how it relates to your decision to start buying back stock.

Yes, Rich, this is Lip-Bu.

First of all, I think I'd just want to highlight that last year, we completed our $1.2 billion repurchase program.

And then the board have approved and authorized $525 million buyback in 2017.

And clearly, our approach usually is focused on our business requirement and also the appropriate risks along to -- relate to the business model and operating environment, and we also like to provide the flexibility for us to continue to execute the plan.

And clearly, returning to capital -- return to the shareholder is top priority from my point of view.

So overall, I think we continue to review that with our board.

The board just approved the new purchase starting in Q3.

Okay.

I guess, I'd like to move on to hardware/emulation.

You mentioned it was a little lighter in the first half than you'd expected.

If you could just comment on why it was lighter and if there's been any competitive changes in the markets, particularly as it relates to Mentor's new Strato platform.

And then presumably, something came in stronger than you expected in the first half.

I'm wondering if that's IP or digital that sort of backfilled for that slightly lighter hardware revenue.

Just a comment on that.

So thank you, Rich.

On the hardware side, clearly, we didn't do as well as we anticipated in the first half.

As you know, this is a very lumpy business, and we expect it to be a little stronger in the second half.

And saying that, I think clearly we're still the most advanced emulator on the market.

And our capacity can scale up to 9.2 billion gates and have concurrent 2,300-plus users.

And so clearly, it's still very well received.

16 out of the top 20 semiconductor are using them, our hardware emulation.

9 of the 10 smartphone player are using the hardware Z1.

So overall, I think we are happy with our product offering.

And then now we have the Z1, that is the prototyping FPGA version, and there is a very good and encouraging reception from our customer.

We mentioned about repeat orders.

And also we have 5 new logos.

And using the same compiler so that the customer have the flexibility from FPGA to very scalable hardware platform.

And so I think overall, we like what we have, and we just continue executing.

Just to highlight, it's a very lumpy business.

Saying that, I think Q2 is a good quarter for us.

The software were done quite nicely.

And then just to highlight a few points, the digital side, we are growing year-to-year 14%.

IP, with a renewed focus grew -- we grew 15% year-to-year.

And then the custom/analog is -- we are clearly the leader.

We grew 9% year-to-year.

And then the SPB, 7% growth.

So overall, software and IP is very strong.

And hardware, just a very lumpy business, and we continue to really focus on customer requirement and customer focus.

And so stay tuned, second half, a little bit stronger.

Your next question comes from Jackson Ader with JPMorgan.

Yes, it's Jackson on for Sterling tonight.

If we can just circle back to the buyback that you're anticipating to start in the third quarter, so what exactly is different about the third quarter or the second half of 2017 versus the first half when you elected not to repurchase any shares?

Yes.

As I mentioned earlier, we've -- last year, we completed a very big $1.2 billion of buyback, and the board even though approved for the $525 million for this year.

And clearly, we want to have the flexibility.

Clearly, the timing, you had to reflect on the business market condition, corporate and regulatory requirements and also look at the acquisition opportunity and other factors.

So I think we put that into total picture.

We discuss it at length in every quarter with our board, and we decided that Q3 is we can -- starting to buy.

Okay.

And then just a quick follow-up.

In -- within functional verification and hardware, which you just mentioned, were there any deals that maybe you expected to close from the first half that slipped into the second half?

Or do you just see more demand building in the second half of the year?

Yes, I think clearly, as I mentioned, it's a lumpy business, and -- but still the most advanced emulator.

And so we continue to work with the customer.

And we basically also want to drive the value and in terms of the customer needs and make sure that we fulfill them.

And meanwhile, we don't want to sell aggressively to clearly want to just meet the customer requirement.

Hardware, we just have to plan.

It's a long-term business.

Your next question comes from Mitch Steves with RBC Capital Markets.

Just had -- just 2 quick questions for me.

So actually kind of first on the strategic side.

I know you guys are looking for a new CFO.

Do you guys might just provide any quick update on the process there?

Sure.

First of all, Geoff committed to March 2018.

So we have enough time to get the best CFO onboard.

And then saying that, clearly, we have strong internal talents, and also we continue looking at very -- we have access to the high qualified external talent.

Geoff commit to me, have a smooth transitions, and so we continue to work on that.

And that is kind of our game plan.

Got it.

And then secondly, I got to circle back to the hardware side.

I think that just generally speaking, Mentor should be losing share it seems like to both you guys and Synopsys.

And my basic understanding is that you guys are actually more comparable in terms of the emulator and what products you compete against, particularly on the smartphone side.

So I guess why would that slow down, I guess, this quarter then reaccelerate next quarter?

Yes.

As I mentioned, it's a very lumpy business, and we continue to drive value and in terms of customer requirement.

And also it's a place that based on their capacity requirement, we want to plan properly on that.

We have a lot of respect for our competitor Synopsys and Mentor.

And so clearly, we keep then a close eye.

Meanwhile, we just continue executing our plan on the hardware emulation and also our new introduced S1 on the FPGA prototyping and using the same compiler, and then we can scale on both sides.

So stay tuned.

And we continue to execute and meet the customer requirements.

Your next question comes from Farhan Ahmad with Credit Suisse.

My first question is on the hardware side.

On the emulation, your first half, still, they're coming in a little bit stronger.

And then I look at Synopsys.

And there, the hardware sales are actually coming in quite a bit stronger.

So is there some market shift going on firstly between you and Synopsys?

And secondly, can you talk about what are the end markets that you saw weak (inaudible)?

And what gives you the confidence that it comes back in the back half of the year?

Yes.

So I think -- first of all, I think the hardware emulation side, as I mentioned, we have a lot of respect for our competitors, Mentor and Synopsys.

Clearly, on the hardware emulation side, we compete more with Mentor.

Clearly, we continue to be the most advanced emulator in the marketplace.

So it's more the timing of the customer requirement.

On the Synopsys side, clearly, a lot of respect for them.

They have the FPGA versions.

And clearly, we have our own S1 FPGA.

And so earlier reception from our customer is very encouraging.

As I mentioned, we have repeat orders and also have 5 new logos that we are very proud of.

And in fact, I mentioned also one very important point, it's a very important endorsement technology-wise and hardware integration approach, with a very competitive win at the leading North American semiconductor company that is very large -- one of the largest user of Palladium Z1, but they endorsed us on the FPGA S1.

And that is a very, very important endorsement.

Hopefully, we will continue that momentum and then scaling it up.

And Farhan, this is John Wall here.

I'd just like to add that -- and remind you that Q2 2016 was a record year for us for hardware, and that we continue to see a secular trend, an increasing customer need for emulation and acceleration products.

But just want to point that out.

Got it.

And then for the emulation business, do you still expect it to be a growth for you this year?

Because if I recall it correctly, you -- at the beginning of the year, you were expecting that the business will grow.

Yes.

As I mentioned, the first half didn't do as well as we anticipated.

But as I mentioned, we expect it to be a little stronger in the second half.

Got it.

And then in terms of the growth that you are seeing in digital and in system interconnect, you have pretty impressive growth there, more than 15% year-on-year.

How sustainable is that?

And how should we think about that portion of the business in the second half of the year?

Yes, very good question.

We are very proud of our innovation approach to our digital flow.

As I mentioned in my remarks and the Innovus, it become the platform of choice for the place and route for multiple application.

And they are a very big platform.

And so we are very excited about that in terms of the performance, area, power and then also the run time.

You clearly see the big improvement and the customer adoption is very rapidly with the market-shaping customers.

And so we are very encouraged on there.

And the second part is, clearly, our other part of digital flow, the Genus on the Synthesis side, and our Voltus on the timing, the Tempus on the signoff and the Power.

So I think those all come together with our new tool, Pegasus.

We have a really end-to-end full flow.

And we are very excited.

70 customer are endorsing us and using us, improved rating on the full flow.

And we are especially excited, 20 of them for the 7-nanometer.

So overall, I think we are excited about digital flow.

In the custom and the lock side, 9% is very, very strong for the analog and custom flow.

And clearly, we're the leader on that.

And then the -- even with a little bit lesser than anticipated in the hardware side, our formal verification Jasper had been great.

And then our Xcelium, that is a simulation with the Rocketick integrated new tool, we have more than customer -- 20 customers adopting, and we are scaling.

And we'll continue to drive the performance and scalability and stability.

And so I think overall, I think we have pretty good solution for the most advanced nodes for the digital and the mixed-signal customer.

And so we're excited about the overall portfolio and the [soft] win, and also the IP also coming up very strong.

Your next question is from Jay Vleeschhouwer with Griffin Securities.

A couple of short-term questions to start, first for Geoff and John, then a more strategic question for Lip-Bu.

For Geoff and for John, you reiterated your cash flow guidance for the year.

Yet, through the first half of the year, you've already done about 60% of the low end of your guidance range for cash flow.

And I'm wondering what you're seeing in the second half of the year that perhaps cash flow might be less than in the first half, hence, the iteration of the current range.

And then secondly, back to the emulation question, just to clarify what you mean by stronger.

Is that an increase in absolute terms first half to second half?

Or are you talking about year-over-year increase in the second half versus pretty easy comp versus second half of last year?

And in order for you to grow your emulation business for the year, it looks like your second half would have to be up by about 1/3 in total or more.

If you could comment on that.

So Jay, this is John.

I'll take that first question on the cash flow.

But I'd just point you to the DSOs that's -- we had a 6-day decrease in DSO from Q1.

A number of large payments came in after the end of the first quarter, driving a more favorable comparison between the quarters.

So Q2 ended up being a very strong quarter for us for collections, but -- so you'll see a little bit more shift through the first half for cash collections.

But our guidance reflects our confidence in the business and takes into account everything we know at this time, so we reiterated the cash flow guidance.

Yes.

And then, Jay, on the hardware side, as I mentioned, the first half is lower than anticipated.

As I mentioned, it's a very lumpy business, and we expect to do a little stronger in the second half.

That's what we see in the -- and again, we want to really preserve the value, and we want to make sure that we provide and protect the value rather than just selling for whatever price.

So we want to keep the value and then drive a more manageable growth.

For you, Lip-Bu, a couple of market and customer questions.

Number one is about the length of your current product cycles or adoption cycles.

We've seen now, for the past roughly 2 years, that the implementation business, both yours and Synopsys, had done well.

And I'm wondering how long you think this cycle might last.

The reason I ask you that way is when we look at the last big product cycle for you in PCB, that lasted for about 3 years, 2012 to 2015.

And I know it's not the same technology or even the same customers necessarily, but do you think that there may be just another year or so to go to fulfill this adoption or upgrade cycle and implementation then you're going to have to see some other category pick up?

Yes, it's a good question, Jay.

And clearly, as you know, the complexity of the design has increased substantially when you move down the geometry to 10 to 7 to 5. And we are very aggressive on the 10 to 7 and 5 and beyond.

And so clearly, this is -- we are very committed to drive the technology leaderships, and so we continue to drive the innovation.

I'm very proud of my team that in the last 3 years, we have 23 organically developed products that are very disruptive.

And that's why we continue to drive the success in the implementation.

As I mentioned, competitors increase, so I don't see any slowdown.

And also some of the new exciting application in term of machine learning, deep learning, cloud infrastructure changes and then the whole opportunity in autonomous driving in the automotive side, and so I think are going to be driving new run time, the new performance, power, going to be a big challenge.

And then we continue -- we're going to innovate, continue working closely with the customer, providing the best tool and solve their first time pass requirement.

So I think overall, the adoption cycle, I don't see any slowdown because this are new requirement, the complexity and also the more deeper, advanced nodes that need double, tripling, triple tripling patterning.

So I think all in all, I think I'm excited about the future.

And I think most important for Cadence is to drive the leadership in technology.

And then with that, we're really, really focused on customer, make sure that we are the trusted partner for them to go forward, they can count on us to go forward.

Your next question comes from Krish Sankar with Bank of America Merrill Lynch.

I had a few of them.

First one, either for Lip-Bu or John.

I don't want to like -- I have the same question on buyback, but I'm trying to figure it out a different way, your capital allocation or capital return policy, in the last 6 months, you didn't do any buybacks.

And there's always speculation that it's a takeout or M&A happening or something.

So from your standpoint, why won't you consider doing a dividend so that you don't have to worry about the timing of the buyback?

Yes.

So Krish, I think clearly, capital allocation is a topic that every board meeting, I discuss with my board.

Geoff and I and John, this is a topic that we review quarterly, and so -- in terms of buyback, in terms of dividends.

And all this have been discussed with our board.

But so far right now, the board authorized for $525 million buyback.

And we review that, provide a flexibility, we look at our business requirement.

And then after that $1.2 billion buyback we completed last year, we decided that Q3, we're going to start buyback again.

Got it, got it.

All right.

And then on your hardware business, can you roughly say the split between emulation and FPGA prototyping?

Is it like an 80-20 split?

Or am I in the ballpark?

Yes, we don't provide a breakdown.

As you can tell, I mean, we don't want to give the information to our competitors.

But clearly, hardware emulation is our current volume productions.

And then the S1, we just announced recently.

And then we are excited with the earlier reception from our customer of repeat orders, and then with 5 new logos.

And then one very important and very strong competitive win that they never used our FPGA and they adopt us.

And that will carry out the momentum going forward.

And Krish, this is John here.

I just want to add that we have great products in the hardware space.

And note that we delivered excellent financial results for Q2.

Software and IP were both strong, and we increased our outlook for the year.

So software hardware -- our softer hardware in the first half is only part of what is a really good story.

Got it, Got it.

And then a final question for Lip-Bu.

If you look at the market, obviously, you have all these like exciting technologies like AI, deep learning, autonomous driving.

Is there a way you can quantify what the EDA or Cadence opportunity is, either as a percentage of the market size for AI or a percentage of -- or the dollar value, the percentage of what the chip market could be for those?

Is there any way you can help us quantify the EDA opportunity in those trends?

Yes.

It's a good question.

It will be difficult for us to quantify it at this moment, but I can share some of my excitement.

Clearly, the machine learning, deep learning, AI, going to be changing our semiconductor industry.

And then the application is very broad.

And autonomous driving is just one of the application.

Machine learning, deep learning and -- that can be a very huge impact to the data center and cloud infrastructure and then also for the industrial IoT.

And so -- and even in medical, genomic sequencing.

And then there's a lot of compute that our brain can't function that fast and -- with so much data.

And as you know, the big data is massively -- a lot of data from not just compute and also from video is massive.

And then clearly, the next 10, 20 years, the data is going to be hard to manage, data is going to be the biggest opportunity and challenge.

And so there can be a lot of implication to our customer -- end customers.

And they had to design based on that to do the data analytics from training to inference.

So they're going to be a new requirement for a lot of our EDA solution to provide them to optimize the calculation and the data management.

So we are excited about it.

And it's very hard to quantify it.

And stay tuned.

And then if we had the insight, we'll share with you.

But right now, it's just an exciting opportunity.

It's just an emerging new market.

The application is going to be very broad, and it's very exciting.

Your next question comes from Tom Diffely with D. A. Davidson.

Just following up on that last question.

When we get into an environment when guys like Nvidia are building these huge chips, is this a bigger opportunity for you on the hardware side?

Or do you think the core EDA software side is the bigger opportunity on an incremental basis?

Yes, very good question.

And clearly, we have a lot of respect for Nvidia and what Jen-Hsun had done is fabulous for the industry, continue that innovation engine that they have.

And so saying in general, I think clearly, the machine learning, deep learning, as I mentioned, is a huge impact to our semiconductor industry beside the memory that we are right now seeing a lot of impact into the memory side, again, tying into that whole data management and the storage-related area.

So I think on the -- back to your question in terms of Nvidia as an example and then broadly into the whole AI, machine learning, not just the hardware.

Hardware is just for the verification, whatever the design.

But a lot are going to be driving the IP, like our Tensilica will be a very great platform because it's programmable, low power, and it's very good for a lot of industrial application for autonomous driving that I highlight in my remarks and also the whole data center cloud infrastructure.

So I think they're going to be require more optimization from a tool point of view and also the speed and the run time of our design flow and the most advanced node.

And so those massive parallelism are going to be critically required.

And that's why we're excited we are so well positioned in terms of many of our new tool, how competitively we were able to scale to massive parallel in term of processor core to optimize a solution that the customer is looking for.

And that is something that we are very well positioned.

Okay.

And I know we spend a lot of time talking about the leading edge.

But as we look into things like IoT and the pervasiveness of sensors and the like, on the low end, how are you positioned to benefit from that in terms of the unit growth and the design growth at the low end?

Yes, a very good question.

So I think on the low end, like the IoT and also the devices that can collect the data, and then we also have a very unique offering because the Tensilica is very low power, programmable.

And then the other part is one of our expertise is the low power into our flow.

Either it's a custom flow or digital flow.

That's one of the area that we are driving.

Low power is going to be the key for some of this industrial IoT that should -- that will be able to collect data.

And in some ways, they can replace the batteries so that you can last longer.

And then some of the technique, some of the approach, that is a very strong Cadence offering we have in our tools.

Okay.

So when you, I guess, put this all together, is your expectation then as this starts to develop over the next few years that the growth in your served market actually accelerates from where it is today?

Yes.

I think we're kind of looking forward to that.

And one thing that I wanted to highlight is our analog mixed-signal.

And that is a lot of this IoT or data collection on the end products, that is critical for the low-power analog mixed-signal side.

And that is our expertise.

We're going to be offering that to the customer.

And in fact, we -- a couple of our important customer in those area have been adopting us and then using our flow -- full flow for doing that.

Okay.

I guess then moving over to -- obviously, the Asian market was very strong for you over the last year.

Curious though, was most of that growth from core EDA, the hardware or the IP side of the business?

And do you think that trend continues, whatever it is?

Asia Pacific is a very important region outside the U.S. And a couple of areas that we really like and we really focus on like in Japan, the system company and then the automotive semiconductor area, they are doing very well.

We are very well positioned with those customers.

And so we doubled down on that.

And then the other part of the Asia, clearly, Korea is a very important market for us.

We have a lot of success in that in digital flow, hardware emulation and IP front.

And then the other part, a big engine, is China.

We are very well positioned in our China position.

Clearly, the government have a very big initiative in term of driving the domestic semiconductor industry, with a massive, massive investment in the hundreds of billions building that.

We are very well positioned to partner and then collaborate and support.

That is very broad through the digital flow, the EDA flow and also IP and also hardware emulation.

So these are all total solution that we can provide to some of these leading company.

I think I mentioned on the hardware emulation side, with HiSilicon in terms of the add-on capacity that they require, and they're one of the winning company.

And then in the past, we also mentioned a couple of others.

And then there's quite a few Chinese company that's going to be a world-class company, and we want to be part of that and support them globally, just like any of our U.S. company and European company and Japan company.

Okay.

And then just finally, going forward, do you think the FPGA prototyping product will be sold or packaged with emulation?

Or are those separate sales?

Yes, depends on customer requirement, hardware emulation that have a scale or capacity like 9.2 billion gates.

And that is very attractive for many leading large company because the hardware emulation is the most accurate and -- in term of whatever you design, you want to verify that.

Then secondly, I think FPGA for prototyping, and that is a very good way to do that.

And we -- also both, using the same compiler that is very attractive for customer.

It can go one way or the other depend on the application, depend on their needs.

And then they can scale it or they have to do it separately.

They can do it all in the same compiler, and that is very, very compelling.

So I think we're going to provide that.

And we'll continue to scale, continue to drive the mix and return for that.

And then we will announce when we are ready and then continue to drive the capacity, lower the power and the scale for the FPGA and the emulation side.

Our final question comes from Monika Garg with KeyBanc's Pacific Crest.

The first is on R&D.

If I look at R&D as a percentage of revenue, first half of this year, it's somewhat [37.3%], which is higher than previous year.

So is it the normalized R&D level to think about going forward?

Or how should we think about what is R&D level?

Yes, I think let me start and then John can chip in.

So first of all, as you know, there's not much out there to buy for EDA tool.

And so a lot of customer are shifting to us with the assumption that we will double down, triple down on R&D to continue to drive the solution that can help them to design the most complex chip.

So we basically had to continue investing.

And especially right now, we have couple of area we try to drive leadership like in digital, some of the digital flow and some of the verification flow, we want to drive the leadership.

And so -- and then also the proliferation to some of the leading customer, we need to invest and work with them.

And then some of them are game-changing.

And then 7-nm nodes and 5-nm nodes are going to be defining going forward who own the biggest flow for the customer.

So I think those are critical in this period that we have to continue investing and then the -- like in the past, Geoff and I, we talked about every time we put investment, we're also looking at ROI, in terms of customer commitment to us, to our flow before we put resources behind.

And so besides doing that innovation for the product leadership, providing the proliferation on the customer to win, and then thirdly, until the customer commit to us and make the commitment before we put the money behind in term of R&D and FAE support.

So I think so far, we continue to drive efficiency.

So it's not just continue increasing, each of the R&D team know very well, we also ask them to drive the efficiency, drive cost reductions.

And so we are mindful of the overall return.

And that's why we continue to guide the operating margin from 26% to 27% this year.

And so far we are on track on that.

Yes, Monika, this is John.

I'd just like to add that the R&D investment that we're making now is for future years.

But our strategic priority is to develop innovative products and help our customers be successful, capture market segment share and bring our solutions to market-shaping customers.

But we're continuing to build on our innovation and take action to drive growth and deliver results for our customers, all of which should enable us to deliver value to shareholders.

And Geoff, I don't know if you want to add anything?

Yes.

Just from an expense perspective, in the first half of the year, we have higher social security payments, both for the employees that are on the company's behalf and less vacations and more vacations in the second half.

So usually the expense moderates in the second half.

Got it.

And then you talked about $525 million share repurchase authorization from the board.

Maybe can you discuss how you're thinking about share repurchase in the second half of this year?

Yes, that's -- so as we said, we like to choose -- we like to maintain more flexibility with the most recent authorization.

The company regularly reviews its capital structure, balancing our needs for investments, the appropriate level of risk for our business model and operating environment, maintaining adequate liquidity and the opportunity to return cash to shareholders.

The board and the management make their decisions through the lens of shareholder value.

But we didn't repurchase shares in the first half, but we do, obviously, expect to repurchase shares in Q3.

And we're not specifying the amount of time nor the method at this time.

Okay.

And just the last one -- 8, 9 months since Siemens announced acquisition on Mentor, closed almost 4 months.

Have you seen any change in competitive dynamics in the market?

Sure, Monika.

First of all, I think the Mentor products remain competitive under Siemens' ownership.

Siemens is a very big company, $100 billion company.

They have tremendous resources, a lot of respect for them.

And so we continue monitoring the product competing with us, and we don't anticipate any less.

And so we treat them the same and then maybe even more.

This concludes the question-and-answer session of today's call.

I will now turn it back over to President and CEO, Lip-Bu Tan.

In closing, through innovation and execution, we are well positioned to build on our success and to further proliferating our solutions with market-shaping customers.

I would like to thank all our shareholders, customer and partners, Board of Directors and very hard-working employees for the continued support.

Thank you for joining us this afternoon.

Thank you for participating in today's Cadence Design Systems second quarter 2017 earnings conference call.

This concludes today's call.

You may now disconnect.