Executives: Rick Muscha - IR Moshe Gavrielov - President and CEO Jon Olson - EVP and

CFO
Analysts: William Stein - SunTrust Robinson Humphrey Gabriela Borges - Goldman Sachs Ross Seymore - Deutsche Bank Srini Pajjuri - CLSA Securities Christopher Danely - Citi Joseph Moore - Morgan Stanley David Wong - Wells Fargo Ian Ing - MKM Partners Hans Mosesmann - Raymond James Christopher Rolland - FBR Capital Ambrish Srivastava - BMO Capital Markets Harlan Sur -

JPMorgan
Operator: Good afternoon. My name is Mike and I will be your conference operator. I would like to welcome everyone to the Xilinx’s Second Quarter Fiscal Year 2016 Earnings Release Conference Call. All lines have been placed on mute to prevent any background noise. After the speakers’ remarks there will be a question-and-answer session.

[Operator Instructions]. Please limit your questions to one to ensure that management has adequate time to speak to everyone. I would now like to turn the call over to Rick Muscha. Thank you. Mr.

Muscha, you may begin your conference.

Rick Muscha: Thank you, and good afternoon. With me are Moshe Gavrielov, our CEO; and Jon Olson, CFO. We will provide a financial and business review of the September quarter and then we will open the call for questions. Let me remind everyone that during our conference call today we may make projections or other forward-looking statements regarding future events or the future financial performance of the company.

We wish to caution you that such statements are predictions based on information that is currently available and that actual results may differ materially. We refer you to the documents the company files with the SEC, including our 10-Ks, 10-Qs and 8-Ks. These documents contain and identify important risk factors that could cause the actual results to differ materially from those contained in our projections or forward-looking statements. This conference call is open to all and is being webcast live. It can be accessed from our Xilinx Investor Relations website.

Let me now turn the call over to Jon.

Jon Olson: Thank you, Rick. Xilinx sales were $528 million, down 4% sequentially and within our guidance. We experienced strong new product growth and strong profitability during the quarter as well as a modest recovery in communications. New product sales increased 11% in the quarter, with increases from our 7 series, Spartan 6 product family and our 20-nanometer UltraScale family which exceeded $15 million in quarterly sales.

Zynq adoption continues to be strong across a broad-base of end-markets, including automotive, wireless, industrial and consumer applications. From an end market perspective sales from communications and data center increased 5%, driven by a modest recovery in Asia and North America. Wired Communication sales were driven by data center access and enterprise applications. For the second quarter in a row wired communications represented the largest contributor of our 28-nanometer sales. The Industrial and Aerospace and Defense segment declined 12% sequentially with expected decreases from Defense.

The combination of Industrial, Scientific and Medical and Test and Measurement declined sequentially with somewhat weaker than anticipated ISM business towards the end of the quarter. Lastly the broadcast, consumer and automotive category was down slightly driven primarily by decreases in broadcast. Gross margin was 70.1% for the quarter, at the high end of our guidance driven primarily by customer mix. Operating expenses, including $2 million of amortization expense were $217 million in line with expectations. Other income was a net expense of $9 million.

Net income for the quarter was $127 million or $0.48 per diluted share. Operating cash flow for the September quarter was $137 million, before $5 million in CapEx. Cash flow for the quarter was impacted by higher receivables, primarily associated with the timing of end of the quarter shipments to our primary distributor partner, who had an extra fiscal week in their quarter. Diluted shares for the quarter were 266 million shares. There was a 6.5 million shares dilutive effect from our convertible notes.

For questions relating to the dilution associated with our comparable debt please visit our investor relations website at www.investor.xilinx.com. We repurchased 2.4 million shares for $100 million during the quarter and paid $80 million in dividends. As a result we returned over 130% of operating cash flow to the shareholders during the quarter putting us well on track to continue to return over 100% of operating cash flow to shareholders in fiscal year ‘16. Let me now comment on the balance sheet; cash and investments were $3.5 billion. We had $600 million in convertible debt and $1 billion in fixed rate debt, resulting in a net cash position of approximately $1.9 billion.

Inventory dollars at Xilinx decreased by $9 million sequentially. Let me now turn to a discussion of guidance for the December quarter of fiscal year ‘16. The guidance I provide will be inclusive of the extra fiscal week in our quarter which we disclosed in our 10-K filing as well as in today’s earnings release. Our backlog heading into the December quarter is up sequentially, both Wired and Wireless Communications are forecasted to be up. We believe our key Wireless customers have reduced levels of inventory and as a result we expect continued improvement from China driven by FD LTE deployment activity.

Wired communications continues to be impacted by a weak FX environment, but we are expecting growth during the quarter from OTN and cable infrastructure businesses. We expect our Industrial and Aerospace and Defense segment to be up sequentially driven by return to growth in our space and defense. Lastly we expect all components of the Broadcast, Consumer and Automotive category to be approximately flat. As a result we are expecting total sales to be up 3% to 7% sequentially. The midpoint in this guidance is predicated on [turns] [ph] rate of approximately 48%.

Gross margin is expected to be approximately 69%. Operating expenses for the December quarter are expected to be approximately $230 million, including $2 million of amortization of acquisition related intangibles. Other income and expense for the December quarter will be a net expense of $7 million. The share count is expected to be approximately 265 million shares. There will be an adjustment to our full year forecast on CapEx.

We are raising our forecast to approximately $40 million for the year up from previous guidance of $25 million as a result of building renovations on our San Jose campus. The tax rate for the December quarter is expected to be 12%. Let me now turn the call over to Moshe.

Moshe Gavrielov: Thank you, Jon and good afternoon. Despite the anticipated challenges on the September quarter revenue I'm pleased that Xilinx continued to deliver strong profitability.

Gross margin was 70.1% facilitating operating margin of 29%. This is a testament to the strength of our business model, the diversity of the markets we serve, the impact of several corporate initiatives focusing on productivity improvement. Revenue from our new products category, which delivered robust 11% sequential increase is expected to further accelerate in the December quarter. This gives us growing confidence that the September quarter represents the bottom of the current cycle for Xilinx. As forecasted, last quarter 28-nanometer revenue growth was driven largely by increases in wired communication industrial applications and defense.

Special mention is due to our Zynq product family, the industry's first all programable SoC which increased sequentially by nearly 40%, repeating a similar increase in the June quarter. Our overall 28-nanometer revenue is expected to grow significantly in the December quarter, with sales driven by a majority of our end-markets. These include communication, industrial, automotive, defense and test and measurement. 20-nanometer revenue in September quarter comfortably exceeded $50 million forecast. Majority of the growth continue to be driven by Virtex UltraScale, the industry's only ASIC class 20-nanometer high end product offering.

In the December quarter we are expecting 20-nanometer sales to again post strong growth and exceed $20 million led by wired communications and test and measurement applications. The 16-nanometer node we shipped the industry's first all programmable MPSoC. This product has been already delivered to customers. This is a quarter ahead of schedule highlighting the third consecutive generation of product leadership enabled by stellar execution, by our foundry partner, TSMC, in conjunction with our engineering and operations teams. This exciting new family targets application in embedded vision including ADAS, industrial internet-of-things, 5G wireless systems and will facilitate Xilinx’s expansion into the domain of ASSPs.

Finally, last week we announced the strategic partnership with Qualcomm to deliver industry leading heterogeneous computing solutions. These solutions will deliver new levels of efficiency and performance through FPGA-enabled dynamic work load acceleration on Qualcomm's ARM-Based Server platforms. As I mentioned at the beginning of my comments, top line 3% to 7% sales growth expected in the December quarter is again being driven from our new product category. This is forecasted to deliver another healthy double digit percentage increase comprised by both our 28-nanometer and 20-nanometer leadership portfolios. We now turn the call back to the operator for Q&A.

Operator: [Operator Instructions]. Please limit your questions to one to ensure that management has adequate time to speak to everyone. Your first question is from William Stein from SunTrust.

William Stein: Great, thank you for taking my question. I'd like to first ask about the extra week in the quarter, and your expectations for that effect on the P&L.

Should we expect that to be sort of a linear view and maybe add roughly 7% to sales or is that an overly optimistic view as to what that adds?

Jon Olson: Hello, it's Jon. I don't know about being overly optimistic or not. But because the way this is a holiday, end of the year kind of thing and what the week's work out they would definitely, kind of piles up towards more of the day of the holiday which are typically not good production time period. I mean the end of our quarter ends up being, I think like January second or something like that. So I mean we are now overlapping both Christmas and the New Years' time.

So I don't think, thinking about it from a 7% or arithmetic way is necessarily the right way. However there is obviously some more shipping days in the quarter as a result of it. And of course we get a full week of expense no matter what.

William Stein: Understood. Thank you and then maybe I can just tackle the cyclical commentary.

I think you are pretty explicit that you feel September, Moshe, I think you said September marks the bottom in the cycle for Xilinx. Perhaps, you can help us with the strength by end market when we look out to see what the next cycle, what the upcoming let's say up turn should bring. Would you anticipate that the comment for structure end-market would be the greatest driver of that? And are you seeing -- I missed whether you said you're seeing signs of life in wireless infrastructure in China, whether that's sort -- whether those are -- that spending has started to come back. Thank you.

Moshe Gavrielov: Okay.

So if you look at it market-by-market there is some recovery in communications, not necessarily from huge, snap backs, in particular and not on wireless. So this does not assume there is a tremendous improvement in wireless. We are seeing improvement in wireless. It has come off the trough. The trough I believe was in this quarter before last and it is improving slowly from that.

But there is no huge rebound. We are seeing strength in wired communications that is growing nicely and some of the other markets are growing at a nice level and are driving this. It tends to be a little broad, but Jon do you want to share your specifics?

Jon Olson: Yeah, so I think if you look at our new product growth rate of 11% and an expectation that we're going to have a nice growth in this next quarter as well on new products, I think what we're seeing is bouncing off the bottom. We're seeing some of the designs that we've really started to grow pretty significantly in wired and across automotive and industrial, when we look at our Zynq product family starting to grow very significantly now. And that's a good sign.

And the two drags that we had going for us through the earlier part of this calendar year was our aerospace and defense business declining from the standpoint of the Joint Strike Fighter Program finally shipments being done. So that is essentially aerospace and defense especially hitting bottom for us in this past quarter. And then the fact that inventory needs to bled off in China particularly. And we believe our customers have reduced their inventory levels of FPGAs that they had in their stores and we're starting to see some signs of more reordering and some more incremental deployments, as Moshe said but not a giant snapback. So those are some of the factors that I think have contributed to our view, in our confidence level that we're going to start moving back up now from the bottom.

William Stein: Great, thank you.

Jon Olson: Next question please?

Operator: Your next question is from Jim Covello with Goldman Sachs.
Gabriela Borges : Great, thanks so much for taking the question. This is Gabriela Borges on behalf of Jim. Maybe just a follow-up on the earlier comment on inventory levels and the supply chain.

If you could share with us any metrics specifically on inventory at wireless OEM customers and then to the extent you could talk about inventory and distribution more broadly across the end markets and where those levels are compared normal levels that will be very helpful. Thank you.

Jon Olson: I think just as a general comment, inventories across most of our end markets are pretty lean and including in distribution, despite the fact that we had a buildup in distribution right at the end of the quarter, but again that was timing from when Avnet's quarter end because they had the 14 weeks last quarter which was after our -- a week after our quarter end. So there was some build up there but that really isn't a sign of any sort of long-term business situation. So I think in distribution we're in pretty good shape.

With respect specifically to wireless it's always difficult to know the true story about what the status is exactly. I think as you recall we've talked about this before. There are base stations that our customers have on loan to the carriers. There are base stations that we're in inventory completed and then there were situations where we know FPGAs were built up in their factories and what we can see the most obviously is the FPGA inventory and that's where we feel we work that down at major customers and there should be more of a demand pull kind of a situation now on a go forward basis, that's our view on that.

Gabriela Borges: Appreciate the color thanks very much.

Operator: Your next question is from the line of Ross Seymore with Deutsche Bank.

Ross Seymore: Hi guys. I wonder if you could talk again on the wireless infrastructure side, just conceptually as you look forward. I think you said that it was down about 50% year-over-year a couple of quarters ago. If you think about where it could go in the future do you think that, that business for you can exceed the prior peaks or are there some things that would limit it from ever reaching those peaks just conceptually trying to think if we are at the bottom what does the next top look like and where does it compare to the prior?

Moshe Gavrielov: So if you look at wireless, there is various elements and there's 4G deployment and the next major surge will be 5G deployment, which were already designed into a lot of the prototypes which will be demonstrated in the next few years.

But reality there is long time before there is the massive growth in 5G. So if you are looking for the short-term we do expect for it to grow it’s unlikely in the short-term to get to the previous peak. That peak was driven by a surge in China. What we’re seeing now is deployment in India but that is viewed as a much smaller market and then other markets are deploying at a low rate. So I don’t see us returning to that previous peak, which I believe was in the March 2014 quarter any time soon.

I do expect us to see a recovery from the trough of the previous peak it’s going to take time and it will probably require a combination of elements to return to that level.

Jon Olson: I would just point out that that previous peak that Moshe referred to was an artificial peak in retrospect meaning there was a lot of inventory build -- being built up by our customers who all thought they were going to get a higher share, it was earlier in the China LTE rollout of TD and I think we had many customers thinking they were going to get a bigger share out of that. And so in retrospect that’s why I called it the artificial peak.

Ross Seymore: Great, thank you very much.

Operator: Your next question is from the line of Srini Pajjuri with CLSA Securities.

Srini Pajjuri: Thank you. Jon. Just a question on OpEx then I have a follow-up. So obviously the extra week is, I think impacting the OpEx in the December. How should we think about it as we head into the March quarter and beyond?

Jon Olson: Yeah, so we provided full year guidance for R&D and SG&A and we do still expect to be in that range and maybe SG&A a little bit towards the bottom of that range quite frankly but there's still a couple of puts and takes going on, on that.

But we’re very comfortable with the previous range. We can get you those numbers if you need them it’s we have…

Srini Pajjuri: I think I have them too, yeah, that’s great. And then maybe for Moshe. You put out a press release a couple of days ago on the Qualcomm partnership. I guess one of the reasons that Intel has been talking about for acquiring Altera is I guess they want to integrate FPGA into the x86 die.

I am just curious if this partnership with Qualcomm first, is it exclusive to Qualcomm and then I think you also mentioned that down the road you plan to offer integrated solutions does that mean that Qualcomm would be offering fully integrated single die solutions with Xilinx FPGAs on it? Thank you.

Moshe Gavrielov: Okay. So let me first answer with regards to Qualcomm. So we will have a range of solutions that we work on with Qualcomm, start from cards and then move to higher level of integrations we haven’t defined what those higher levels of integration are but we are working closely with them. And this is a program that started two years ago and what was shown at the rollout was that the first server that they were providing to their software developers and to their customers as a development platform and there was in parallel a board which had our device on it which connected with that and provided the 10 to 50x acceleration for certain sets of applications.

And so that's the situation there. We have a very tight relationship with Qualcomm. It's not an exclusive relationship. One of the opportunities we have is we believe that the data center market is an emerging market. There are three elements to it.

There is a networking part, there is a storage part and there is a acceleration part, which is the one which is getting the most praise. But isn't the only one. So the first two are obviously we are in a good position there. With regards to the third what we are seeing is there is a lot of competitive architectures, x86, ARM and others which are pursuing that market at this point in time. Clearly Intel has the highest market share.

But it is expected to have more than one set of solution, that is driven by the customers who are very large independent companies and they desire to control their own destiny. So we do believe that that will over time provide interesting growth opportunity for us and we have a strategy that enables us to participate at different levels of integration with all of the players in that market.

Srini Pajjuri: That's great, and when do you expect production revenues from this partnership, Moshe? Thank you.

Moshe Gavrielov: Well, with the Qualcomm I can't tell you without disclosing when they are going to ship the product hence, I am precluded from answer that, but other partners are in different stages of deployment. A few are a little further ahead, a few are further behind.

And so over time we do expect this to be a significant growth opportunity for us. But this is not measured in months. This is couple of years out at the earliest before it starts becoming visible.

Srini Pajjuri: Great, thank you.

Moshe Gavrielov: Thank you.

Operator: Your next question is from Chris Danely from Citigroup.

Christopher Danely: Hey, thanks guys. Just one question on the call end [ph] market. So if you look at the spending patterns from the China carriers and looks like it's got pretty strong in the second half of this year. So how much of this are you guys seeing and then conversely getting into the next calendar year, it looks like spending is down.

So any thoughts you have on the market for next year would be appreciated as well.

Moshe Gavrielov: Yeah, Chris, on the spending pattern out of China and just generally Asia, as we have been communicating here, we do see positive signs. But we don't really see -- we see the positioning that there could be brisker growth out there in the next two quarters for us, the rest of our fiscal, but our customers are certainly not committing to that level at this point in time. So it's one of those situations again where you don't know whether it's a false signal that they are saying 'hey, it could be bigger than this or may not be bigger than this. So we put a number in that we think is more in the center of likelihood there.

And relative to next year it's very difficult to forecast for the whole year on the wireless segment. But I don't think we are looking at it, based on how poor this year is. I don't think we're looking at it as a down year for us next year, that's about all I'll say on that.

Christopher Danely: Okay, thanks guys.

Operator: Your next question is from Joe Moore from Morgan Stanley.

Joseph Moore: Great, thank you. If we go back to the server acceleration business, can you talk about how big you think that opportunity ultimately can be, Altera had talked about it as a $1 billion type of long-term opportunity, Intel made it the focal point of the transaction. Do you agree that it could be that big and then I have a follow-up?

Moshe Gavrielov: Well, it's an emerging market and it's a little too early to figure out how it will turn out. But it is clearly in the hundreds of millions of dollars. Could it be a billion dollar? Yes, it could be a billion dollars.

But I think it's pretty mature to sort of count on it being that big because it is an emerging market and we expect it to be a very bifurcated market. Again it's sort of important to look at the end customers and these are the big data center company in North America and then there is the big data center companies in China. Each of them has an ecosystem and software environment that they totally control. And they tend to have an application, which but their applications are different and they tend to do it their own way. So we do expect that to come out with a vast platter of solutions and we believe we'll be in a good position to provide that as opposed to big data small number of solutions which we believe will be the competitive approach.

Joseph Moore: Okay, great. And then following up on that, to the extent that you've talked about other architectures other than x86 in that market, can you talk about your approach within x86, you sort of assume that the lion share of servers is going to remain in x86. What's your ability to compete, how important is that, that you're not in the co-packaged part with the processor? How did you compete external to that?

Moshe Gavrielov: Well, the biggest challenge to integration is power. And if you look at this market it tends to use a pretty high performance processor and a relatively large FPGA. You need those in order to get the acceleration.

If you look at those, those tend to be relatively power hungry devices. So there is a limit to what you can do by integrating them to one package. So even if it's just interfacing to an Intel device we believe that we will be able to provide a companion device which will enable a differentiated solution for several of the players in that market. So we do expect to play in that market too, and definitely in any non-Intel market we expect to be to have a very strong position with regards to servicing the requirements we are the primary provider of FPGA solutions and the largest provider of FPGA solutions in the market.

Joseph Moore: That's very helpful thank you very much.

Moshe Gavrielov: Well, thank you.

Operator: Your next question is from the David Wong from Wells Fargo.
David Wong : Thank you very much. Can you tell us a bit more about your plans for 16 nanometer? So you've got your first product, will there be a rollout of a large number of more products in the near future? Or will be -- these be spread over time?

Moshe Gavrielov: So what typically happens is when we rollout the first device that it goes through at length the evaluation phase. And as soon as we get to a clear understanding of the level of functionality we then do a rapid rollout of the full on devices.

The functionality of this initial device has exceeded all expectations. We got it back faster and it operates much, much better and hence we already shipped it to customers just to sort of demonstrate the level of functionality. So I would expect the rollout of the next devices to be as fast as we can. We do need to complete the evaluation of first device. That takes significant amount of time.

Typically this process takes about a year, to year and half from the first device until you have most of the family rolled out. And given the success and the fast turnaround we got from TSMC and the functionality, both on the manufacturing side, on the engineering side, we should do better than we have in any previous generation. And it is a very broad and deep product offering. It has a very high performance processor sub-system with significant number of embedded dedicated, accelerators. And the combination of those will enable us to take it and compete, overtime against the ASSPs in more and more markets.

So it is, we think this is a watershed moment for us in terms of the rollout. David Wong : Great, thank you.

Moshe Gavrielov: Thank you.

Operator: Your next question is from Ian Ing from MKM Partners.

Ian Ing: Yes, thanks.

Question on one of the sub markets. You said that ISM was weaker than expected especially at the end of the quarter. Perhaps kind of what’s driving that and does that continue into this quarter for some period?

Jon Olson: Yeah, at this point in time we don’t think it’s necessarily a bellwether to any sort of cycle. It’s just was, I would say consistent with maybe what we’ve seen in some other component companies, a little softness in that market for a short time period. So it’s the matter of returns profile.

We still -- we are expecting continued strength on our new products in that particular area, some of the older designs, just we’re taking a pause probably some inventory adjustments there. But the new products continue to be strong in that particular end market. So I think it was more -- it’s just more of a -- just the situation for how things played out at the end of the quarter than it is any sort of a forecast for the future.

Moshe Gavrielov: Ian if I can just add color, it’s a market that develops very, very slowly and as a result to the extent there was a downturn, it was the older technologies right. But what we have seen and this was driven by Zynq, is in 28-nanometer we actually saw significant and vibrant growth.

So and this is an expansion play for us because the Zynq product offering is different from anything we’ve had in the past. So we are now seeing that enable us to grow into new areas and so we fully expect ISM even if there are pull backs from time to time, we expect to see that to be one of our more promising growth markets going forward. And it's Zynq and it’s the MPSoC again given the nature of the design wins some of those design wins even on 28-nanometer five or seven years old and are just now starting to hit the, and become visible on the Zynq side and that will grow to be a bigger part of our business going forward.

Ian Ing: Thanks Moshe. And then longer term question here.

I mean how much manufacturing cost improvement is left? I see you have now the second highest gross margins in all of semiconductors I mean do you eventually become like a linear which is very high margin but low growth or are you getting some cost improvements as you migrate to 20-nanometers and 60-nanometers et cetera.

Moshe Gavrielov: This is a never ending process and it’s I have often heard the terms that we reached saturation in terms of our capability to improve in terms of density, cost et cetera and as soon as hear that we set a new bar and surprisingly we hit the new saturation point. So I think the opportunity for improvement while not trivial is very significant and we continue to do that on numerous fronts. And it requires a lot of engineering efforts and both in-house and together with our foundry partners. But I don’t think you should sort of assume this is as good as it get.

Now with regards to moving to future nodes the good news is that we are reaching the old historic, so called saturation densities with new nodes faster than you reach them with old nodes even though the new nodes are significantly more complex and have more manufacturing layers. So I think in particular if you look at it see their ability to continue to improve and the volumes they run are very helpful to help us improve on those fronts and we continue to push to achieve those regardless of the node both on old node and new node it’s something that we need to do. So the strategy is both to continue to improve in that regard but also to provide market growth and the market growth comes through expanded product capability. And that sort of, I will admit that of the past few years, I have two years of not delivering that but we are intent on delivering that going forward, for sure much better than we have in the past two years.

Jon Olson: I'll just add one comment on that as we -- if there was a question behind your question around are we sacrificing growth to protect margins, no, that's not what we're doing in this.

We have here is a situation where we have customer and end market mix that's pretty favorable for us, that we're at the high end of our 68% to 78% stated model for the current period and situation. And there are -- we aren't tied by accepting deals at lower than that margin if need to. We just have a portfolio in the balance and we drive cost reduction where we think volume is going to be and we put more focus on that. And I think you're just seeing a disciplined efficient situation that we put in place in the company and 110% agree with all the things that Moshe just said.

Ian Ing: Okay, thanks Jon.

Operator: Your next question is from Hans Mosesmann with Raymond James.

Hans Mosesmann: Thanks. Hey Moshe wanted to go back to the many solutions versus few solutions type of approach and your wording. Is it is since, are you getting questions in sense of urgency from customers directly regarding that many solutions approach as a result of the Intel and Altera announcement earlier this year, thanks?

Moshe Gavrielov: Hans, just to clarify you're asking generally or are you asking with the data centers in particular?
Hans Mosesmann : Actually yeah data center acceleration, that aspect of potentially why Intel may have wanted the Altera asset?

Moshe Gavrielov: Okay. So with regards to datacenter and the acceleration that subset of applications, as we're going through cloudification or whatever the appropriate word is, and that's a major growth element.

A significant portion of the CPUs that are being deployed and now being deployed in these large datacenters in -- and the typical usage is very different from what the general purpose server did. As they build up these big data centers, in particular when you look at the massive players in that market, they tend to have very specific applications. And those applications tend to be massively parallel. Right, so if you think of things like search et cetera you can see how you could benefit from massively parallel solutions. And what sort of happens is for better or for worse, if you go to just making the CPU faster, the return there is not significant.

And if you just go to multi-core CPU, there is a limit to what you can achieve there again without blowing the power and the blow [ph] out of the water or out of the data center. And so what has happened is the biggest limiter is power and the performance per power configuration that you get from an FPGA or programmable solution is superior for most of these applications. And the factor that we are seeing consistently is anywhere from 10 to 50. And it depends on what application, and exactly -- and it can change even between in the same company as they shift things around. So there, is if you believe, if you look at that then that implies that the traditional ability to address the customers' requirements through selling them faster CPUs or more parallel CPUs is limited.

Hence the desire to have a programmable solution and the combination is very beneficial for what is becoming a very large path of the, what used to be a traditional server market. And that's best we can tell, that's what the customers are telling us, that's what they are excited about. We do believe that the Intel acquisition of Altera was driven by a desire to control that and to make sure that the roll out is driven and not in any way compromised by the CPU power requirements. And hence the rational for doing it and these same customers tend to have a very independent desire to control their destiny, which probably opens up opportunities for other non x86 architectures, for other players if there is a competitive x86 shown at the right point in time together with FPGAs. Now that maybe more than you asked.

But that's the comprehensive answer. Hans Mosesmann : No, that was very good, exactly what I was looking for. Thank you.

Moshe Gavrielov: Thank you.

Operator: Your next question is from Chris Rolland with FBR and Company.

Christopher Rolland : Hey, guys, thanks for the question. Just about the geographic breakdown, seems like you guys saw maybe a little bit of North American weakness but APAC was strong. And you guys are not the first ones to talk about this today. This is probably different than what we've been reading in the papers. I was wondering if maybe you had some sort of theory on why that might be and why there is a such difference in perception or reality.

Jon Olson: I don't know if I can comment on others as compared to us. I mean I can certainly comment on our business. I mean our strength in Asia Pacific was wireless business for us a little better China and then the India exposure that we had mentioned earlier, I think in answer to one of the question, those are really the bigger drivers there, with some wired business going on. Moshe talked about 5G prototyping, obviously the Chinese manufacturers are involved in that as well and as well as the other Asian manufacturers as well. So we have been participating in all that.

So we've seen really good new product and new development product as well as some of the legacy LTE design that we have won become stronger. I would say industrial was solid or okay for us in Asia as well, but not knocking [ph] in other parts. The North America phenomenon for us was really more about the aerospace and defense business declining than it was anything else. The decline was really driven mostly by the tail end of shipping of the joint strike fighter and the rest of our businesses there were good and in fact the large [ph] business was even up a little bit in North America as well. So I don't know if that help you at all Chris.

But...

Christopher Rolland: Very helpful, thank you. Just as a follow up, you just mentioned India and Moshe mentioned it earlier, I think he has mentioned on the last call as well. Moshe, this time around you said that it was a smaller market. Obviously from a population standpoint they are just slightly behind China there.

But how do we think about it here, are we thinking about India as fewer base stations or fewer FPGA content per base station when compared to, let's say a Chinese base station. What are the moving parts here and why is it so significantly smaller?

Moshe Gavrielov: Okay. So sadly it is smaller because we believe that if you look at the Chinese approach they invest very heavily in infrastructure and they do it on a broad level, and they do it quickly and they try to make sure, maybe not at the same level you see in the Western world that there is good coverage. India culturally it's sort of incredibly cost focused, and the challenge that exist today in India is even though there is over a billion people there and it's definitely a large country, the problem is that there are a lot of carriers, probably far too many and I think very few if any of them are making any money. So the approach is to do the minimum, right? And as a result the deployments are likely to be in aggregate quite a bit smaller than what has been done in China.

And that's -- it is what it is, maybe overtime it will improve but you go there it is a still a very poor country with a still small middle-class. And if you go to China you can see the infrastructure there is phenomenal and the middle-class there is much larger. And so I believe that that's the major driver for the difference between those two deployments.

Jon Olson: From a content perspective, there is no significant content differences. I mean our anticipation is just gaited by our penetration in to certain manufacturers who have won the contracts for various carriers.

So where we have more content at a given manufacturer then we'll -- and they win that the business in India then we'll participate. But there is nothing about the structural content of an Indian base station that's really different than any other rest of the world.

Christopher Rolland: Great, thanks a lot. Thanks and good luck guys.

Operator: Your next question comes from Ambrish Srivastava with BMO Capital Markets.

Ambrish Srivastava: Hi, thank you Moshe and Jon. I had a question on the size of the business. You have laid out what seems to be a pretty realistic scenario. 4G is seems to be very long in the tooth or deep. And then 5G is way out there and I think Moshe you have been very consistent about the size and the timing of the server opportunity.

So in the last three years where you had a big 4G opportunities in front of you which followed rather quickly after the 3G. So the question is and you also gaining share at 20 nanometer. So the question that I have is, is the business sized appropriately for a much lower build for the next two three years? Or do you look at what's going on in this space and look at M&A and say what maybe we need to think outside the box to grow and do something unconventional like EPLD companies sort of being affiliatory and you turn and say maybe there are some synergies that we can go after and be acquisitive. So just wanted your thoughts on that. And one real quick longer term, you haven't talked about autos.

I know last year we had a conversation about A-Dash [ph] and how you compete with Mobileye. If you could expand on that that will be great. Thank you Moshe.

Moshe Gavrielov: Okay. So maybe Jon can do the automotive one and then I'll give the very loaded or the answer to the very loaded questions you asked at the beginning.

Jon Olson: Yeah, just at the high level on automotive business for this fiscal year we're expecting a 30% year-on-year, so full year FY '16 -- fiscal year '16 versus full year fiscal '15, we're expecting 30% growth in our automotive business. And that's based on from the product perspective it's based on penetration of Zynq into a variety of applications led by the advanced driver systems capability. Where we complete with Mobileye, by the way, we do we have other design wins that aren't Mobileye oriented. They could to be in radar, distance sensors things like that at the various different FPGAs in that. But relative to the direct completion with Mobileye it's around forward-looking cameras.

So again our business is broader than forward-looking cameras but that's where we do go head on with Mobileye. Our strength has been in the dual or stereo cameras, where there is two cameras looking forward versus one mono camera, that's where we've got our original set hold. And we're very confident about our ability to compete with Mobileye across the broad set of platforms. Some of our customers actually believe they are competitors to Mobileye because they're doing the full solution and trying to win designs at a variety of platforms at various automobile manufacturers. So as we sit here today looking about this year's growth over last year and next year's growth and near after, we can articulate to ourselves pretty much every platform that we won in this regard and tend to see the kind of growth level that is potentially ahead of us again begin continuing this double digit growth on a go forward basis.

Because we know we won these designs and the designs are very sticky for a multiple car year models. So with that I'll hand over the rest of it to Moshe.

Moshe Gavrielov: Yeah now I'll try and respond to your very big question in a succinct way because otherwise we won’t get home for dinner tonight but the my perspective and this is no longer in dispute is that the semiconductor world is going through a significant structural change and it’s being driven by economics and as a result there is actually less and less players, I think that, that is no longer a bone of contention. You just need to count and see how much smaller it is. It’s significantly smaller.

There is fewer and fewer, to the point of almost none getting refresh through the venture world and the cost of moving forward from generation to generation is actually going up and there is fewer and fewer players who can do that. So if you step back and look at that entire trend that clearly is an opportunity for companies who can move forward and we are one of those companies that’s why we were first 28 with SMC first 20 with TSMC as a semiconductor player and that’s why we got the 16 out way ahead of the competition. The reason that we do that is not just for bragging rights it’s actually because once we have that leadership we can expand markets we address. If you look at what has happened to Xilinx over the past 10 years the role our devices play and generally the role FPGA devices play in a customer systems is changing. And as a result I will concede that the historic [indiscernible] logic FPGA market is not growing.

But what is happening and you could argue that that is reason not to invest but what is growing is the SAM in terms of new markets we are addressing. If you sort of looked at our business you can see that even though this is not happening quite at the rate we had hoped we are displacing ASSPs. So if you look at the Zynq product offering that required new investment, a big investment but it’s now starting to pay off these past two quarters both grew by around 40% and this is just the beginning. And that enables us to play in markets we have not played before. So that’s the strategy we have.

We believe that yes the results are not there in terms of the overall numbers but if you look at the change in the markets and the competition then actually there is an opportunity and in order to do that we are continuing to develop and be at the leading edge. Having said that, we can still do it and deliver profitability which was sort of in the higher echelon. So our strategy is to doing that because we believe that the opportunity is there the markets are developing and unfolding. And with just to take one example automotive was market which was very, very low single-digit percent for us and now is we expect for that to grow over the next few years. Industrial control, again another slow market is a market where we can displace a lot of solutions where in the past the customer would go to Freescale and TI et cetera to a large extent a lot of those historic players are no longer around.

So that’s the strategy we have and part of that strategy requires us to continue to provide leadership also, it also requires us to invest heavily in tools so that our customers can exploit the opportunity that our silicon provides them. And that’s an ongoing element and that’s the strategy at this point in time. The challenge on the M&A side is that there is very few companies out there that we can buy that fit into our business model and actually augment our offering and hence you are right in pointing out that we haven’t done much there. But I do believe the investments we are making while continuing to deliver best in class profitability will over the time enable us to continue to grow.

Rick Muscha: Operator we have time for one more caller.

Operator: And your last question is from Harlan Sur with JPMorgan.

Harlan Sur: Hi, thanks for taking my question. So on the subject of new opportunities and the importance of tools on the Zynq and Zynq MPSoC product families which are in reality just full blown SoCs. I mean there is obviously a lot of software and firmware development, that's being tied to these programs. You've got your SDx development platform to kind of help to facilitate much of the software and the firmware development.

So I guess the question is what's been the adoption curve of SDx, how important has the SDx portfolio been in terms of helping customers get into market faster and potentially capturing more design wins with the Zynq platform?

Moshe Gavrielov: Harlan, thanks for the question. So it has been very successful. There is tiers of products that we are providing, there is high levels interphase rate which is something we have had for close to five years and there is the SDx families of product that going out they have SD XL I believe has thousands of users at this point in time. SD SoC is starting to be used broadly. It's a big enabler to grow the population of users beyond the hardcore hardware FPGA designer.

So it is essential and technology is doing well and the market deployment is doing well. The reason that the financial implications aren't yet visible in a big way is that the initial targets for those markets were industrial control and automotive and those two market move at the slower cadence. With the new generation we are actually expanding the number of markets we are addressing and we expect to see increased adoption of them. So yeah it is essential and it is something we need to do in order to displace ASSPs and to address the bigger SAM and so far in terms of technology and the deployment it's going according to plan.

Harlan Sur: Very helpful.

Thanks Moshe.

Moshe Gavrielov: Yeah, sure. Thank you.

Rick Muscha: Thanks for joining us today. We have a playback of this call beginning at 5 PM Pacific Time, 8 PM Eastern Time today.

For a copy of our earnings release, please visit our IR website. This quarter we will be presenting at the NASDAQ Conference in London on December 1st, the Credit Suisse Annual Technology Conference in Scottsdale also on the 1st and the Barclays Global Technology Conference in San Francisco on December 9. This completes our call. Thank you very much for your participation.

Operator: This concludes today’s conference call.

You may now disconnect.