Rick Muscha Jon A. Olson - Chief Financial Officer, Principal Accounting Officer and Senior Vice President of Finance Moshe N. Gavrielov - Chief Executive Officer, President and Director.

Romit J. Shah - Nomura Securities Co. Ltd., Research Division Ryan Carver - Crédit Suisse AG, Research Division Glen Yeung - Citigroup Inc, Research Division Ambrish Srivastava - BMO Capital Markets U.S.

Vivek Arya - BofA Merrill Lynch, Research Division James Schneider - Goldman Sachs Group Inc., Research Division Ruben Roy - Mizuho Securities USA Inc., Research Division Srini Pajjuri - CLSA Limited, Research Division Anil K.

Doradla - William Blair & Company L.L.C., Research Division William Stein - SunTrust Robinson Humphrey, Inc., Research Division Christopher B. Danely - JP Morgan Chase & Co, Research Division Tristan Gerra - Robert W. Baird & Co. Incorporated, Research Division David M. Wong - Wells Fargo Securities, LLC, Research Division Hans C.

Mosesmann - Raymond James & Associates, Inc., Research Division.

Good afternoon. My name is Rachel, and I will be your conference operator. I would like to welcome everyone to the Xilinx Second Quarter Fiscal Year 2014 Earnings Release Conference Call. [Operator Instructions] I would now like to turn the call over to Rick Muscha. Thank you. Mr. Muscha, you may begin your conference..

Thank you, and good afternoon. With me are Moshe Gavrielov, CEO; and Jon Olson, CFO. We'll provide a financial and business review of the September quarter, and then we'll 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 Olson..

test, defense and ISM. Lastly, we expect broadcast, consumer and automotive to be slightly down as consumer decreases offset AVB and automotive increases. Lastly, we're expecting the other category to be up.

As a result, we are expecting total sales to be approximately flat sequentially, plus or minus 2 percentage points, with sales from APAC and Japan up and sales from North America and Europe down. The midpoint of our sales guidance is predicated on a turns rate of approximately 53%, consistent with our 4-year average.

Gross margin is expected to be approximately 69%. Though we are not revising our long-term gross margin target of 64% to 66% at this time, we believe gross margin will continue to be around 69% for the next few quarters as a result of product mix and continued focus on margin improvement across our product families.

Operating expenses in the December quarter are expected to be approximately $225 million, including $2 million of amortization of acquisition-related intangibles. We expect our R&D and SG&A guidance for FY '14 to be approximately $490 million in R&D, approximately $385 million in SG&A with $10 million of amortization.

Other income and expense is expected to be a net expense of approximately $9 million. The share count is expected to be approximately 291 million shares. The tax rate for the December quarter is expected to be approximately 13%. Let me now turn the call over to Moshe..

Thank you, Jon, and good afternoon to you, all. I'm pleased that we delivered revenue at the high end of our guidance, coupled with record gross margin. Overall, new product sales were truly outstanding, increasing 22% over the June quarter, doubling on a year-over-year basis. They now represent 36% of our total sales.

This growth was driven by exceptionally strong sales of our 28-nanometer product generation, which comfortably exceeded $80 million, totally eclipsing both the previous quarter's results and our original $60 million target. Sales were driven by old 28-nanometer families.

These include our midrange Kintex-7, old members of our high-performance Virtex-7, our high-volume Artix-7 product family and Zynq, the industry's first All Programmable SoC. This tremendous growth of 28-nanometer revenue continues the momentum from the June quarter, where we already had delivered more than 70% of 28-nanometer PLD share.

In the December quarter, we expect 28-nanometer sales to continue to ramp at an accelerated rate, reinforcing our confidence that this node will be by far the most successful in our history. We're confident that our 28-nanometer revenue will exceed $90 million in the December quarter.

The widespread customer adoption and aggressive revenue ramp are clear examples of the transformation we've made from a supplier delivering programmable logic to one that is enabling All Programmable and smart systems.

These results unequivocally demonstrate that in 28-nanometer, we're a generation ahead in terms of silicon, design tools, IT and the commensurately higher value we deliver to our customers. In parallel, we have already delivered several industry firsts at the upcoming 20-nanometer node.

These include the tape-out of the industry's first 20-nanometer All Programmable device back in the June quarter; the development of the industry's first ASIC-class programmable architecture called UltraScale; and the capability to enable devices co-optimize with our Vivado Design Suite that enables another 1.5x to 2x system-level performance and integration value.

This past quarter, we were very encouraged by numerous successful competitive customer valuations with our 20-nanometer UltraScale devices. Coupled with a significant time-to-market advantage, we clearly have another generation of technology leadership, both in the midrange and, in particular, the high end of our 20-nanometer product offering.

In addition, our FinFast program extends our world-class partnership with TSMC to create the fastest time-to-market, highest-performance FPGAs built on TSMC's 16-nanometer FinFET process. The 16-nanometer FinFET combined TMSC's multi-year lead intensity with double patterning with the advantages what FinFET transistors for performance provides.

We're combining the 16-nanometer process with our new UltraScale architecture, Vivado tool advantages, our multiyear leadership in SoC and 3D IC technology and our SmartCORE IP to provide the best solution for programmable systems integration, our customers' mandate for the FinFET node.

Let me now turn the call back to the operator to open it up for the Q&A session..

[Operator Instructions] And your first question comes from the line of Romit Shah..

Jon, I just wanted to get some more color on the decline. I guess it's an older product revenue stream that's coming down this quarter. Could you just -- any color you can provide on which vertical or customer -- why the decline is happening now.

And does this headwind go away after the December period?.

So, Romit, let me make sure I understand your question.

This is about the decline from a large customer of older products on the wireless segment?.

Correct..

Okay. Yes. So I just want to make sure that I had it straight. So, yes, what's happening here is we are seeing the pickup of LTE business in China. As we talked about this in the past, we thought it would begin in September, October.

And in the September quarter, we did see, I would say, the beginning of it, and then we do believe we'll have acceleration of that as we go into December and then follow on in March.

The offset to that is the business transition that's going on at the large European wireless customer where we had been shipping 65-nanometer products in large quantities into the LTE designs there.

And as we've said in the past, we've lost the socket at 40-nanometer and there's a transition that's going on at that customer away from us for certain designs and towards our competitor. And so, what we're signaling here is over this past quarter and into next quarter is that, that revenue is starting -- is coming down some.

And we believe by the time we get to the end of the December quarter, we'll have essentially bottomed out, for the most part, in that transition because we didn't have -- we relate with the product family of 40-nanometer and we didn't win those sockets as we did from the previous generation..

Your next question comes from the line of John Pitzer..

This is Ryan Carver for John. Just to kind of follow up on that question a little bit.

I mean, is this decline at this other large European customer potentially impacting your attach rate to the LTE build-out in China? I guess, maybe broadly, I mean, how should we look at sort of the rollout in China? Is it going to be more 40 than 28 in terms of the process? Or maybe you could just sort of help us out with -- or sort of, from a node perspective, what kind of stories we should expect? And when the 40 to 28 transition should happen in terms of the China LTE rollout?.

Yes. Well, this decline that we're referring to from the European customer will not have any significant effect on the China position for Xilinx at all. I mean, that -- the number of the allocation, as best as we can tell, to the European-based manufacturers is pretty small in China.

And so, we continue to be really bullish, very bullish on our position in China. The initial rollout of technology in China LTE at some customers, some manufacturers, primarily one, is a combination of 40 and 28. But that transition -- that large customer transitions to 28 pretty rapidly.

And pretty much from the rest of the Chinese manufacturers, the other 4 manufacturers are coming out of the sheet with 28-nanometer products, which is largely ours.

So we still believe we're going to have, by far, the most significant PLD share of the China LTE rollout, and this phenomenon that I'm referring to with the European manufacturer doesn't really impact that..

Your next question comes from the line of Glen Yeung..

Jon, I'm just going to keep popping away on the same issue. I'm wondering if you could give us a sense as to the dollar value of the loss in Ericsson.

And then, as you say it bottoms out in the December quarter, does that tell us something about the March quarter? And as much, is that no longer a drag on the business? And you said that China will continue its momentum.

Do you expect China to actually -- China LTE to actually be up quarter-on-quarter in the March quarter based on what you can see at this point?.

Yes. So again, if you look at our -- the fact that our communications business increased this quarter, and we actually -- most of -- the biggest part of the increase was in the wireless segment. So even with the decline of the European customer, we still grew wireless quarter-on-quarter.

And again, we're forecasting the wireless to be up again a little bit in the December quarter.

And having worked through most of the decline from the European customer, then that would indicate that -- assuming there's China business growth going on in March, that would be reflected more fully in terms of a bigger growth percentage in the wireless business for us in the March quarter. So I'm kind of putting all these things together.

Really, it's kind of -- turns out to be around a 2-quarter transition of what we're experiencing here..

But, Jon, can you quantify how much it was?.

I'm not going to get into the quantification of that now..

Your next question comes from the line of Ambrish Srivastava..

Jon, you guys have been signaling about -- talking about this share loss at 40. But beyond that, focusing on LTE, and we always see the fits and starts, you said that you expect the China LTE to -- can accelerate through December and March.

What gives you the confidence that, that will happen? And then, my quick follow-up is, we've been picking up that lead times had been stretching out for some of your -- actually for the Kintex-7 and Virtex-5.

Could you please comment if that's true or not? And where are the lead times?.

Yes. So the rate -- Ambrish, the pace of the rollout is obviously dependent on, one, the OEMs, how fast the OEMs are going to be shipping into China Mobile and how fast the other providers come online as well.

So we kind of have a 1 quarter at a time visibility, and the current forecast for this current -- this quarter that we're in right now and the following quarter do indicate a nice healthy ramp for us. So I mean, we're just -- we're going to base on what they're telling us.

And the story seems to be consistent from all 5 of the major suppliers that account for 90% of what's going on. The Chinese manufacturers account for essentially 90% of that. And I'm -- and we're talking [indiscernible] in Shanghai Bell [ph] is 1 of the Chinese manufacturers where we have very strong position as well.

So the forecasts are all very consistent, and they're all -- I mean, these are all up into the right for the next couple of quarters. So that's the basis with which I'm making that statement.

And then, with respect to lead times, there's no doubt that, as Moshe just talked about, we are definitely doing well in the 28-nanometer revenue at a higher rate and it's ramping faster than we had even anticipated. And that has put some tension in the supply chain in terms of lead times.

But at this point in time, we don't see anything ahead of us that's going to be able -- that's going to restrict us from hitting the numbers that we need to hit for 28-nanometer, although it is a little on the tight side..

So you're getting enough capacity from TSMC on that?.

We are definitely getting enough capacity from TSMC. But the question is, is how fast -- how good are we at forecasting what part in ordering ahead of time enough that we are getting really good support from our foundry..

Your next question comes from the line of Vivek Arya..

Jon, first, now that there is more news around China LTE and some scope and timing of base station rollouts, is it sort of coming in line with your expectations or you still think it's a quarter-by-quarter kind of visibility?.

Yes. So we've all -- there's been lots of things written about the pace of how fast this is going to happen and the rates to 200,000 base stations installed in a very short time period. We have never held that position.

We have always held the position that this is going to be a multiple quarter rollout, and it was going to take the neighborhood of anywhere from 4 to 6 quarters to roll out this first phase. And we still maintain that it is -- it appears to be on a pace like it's more like a 4-quarter or so than it is. It's all going to be a rocket ship tomorrow.

And a lot of that has got to be related to the fact that all of the front-end logistics requires that ordering patterns of all the manufacturers for silicon and other things that go into it. And so, we think it's on the pace that we would have -- we've been thinking about it and we've been forecasting, which is a multiple quarter rollout..

Got it. And as my follow-up, perhaps....

Your next question comes from the line of James Schneider..

Jon, I will need to get you to maybe provide some quantification of what revenue you saw from the China Mobile rollout into the September quarter and then roughly what you're expecting in terms of magnitude for the December quarter..

So, Jim, I'm not going to get into any specific dollars per se. But I mean, again, we've been -- we have been -- we've had, I would say, a relatively low level of shipments over the previous several quarters, a little choppy, as they were doing some of their prototypes.

And you've -- you look at the fact that our comms business overall is about the same level that it has been, so 43%, 44% kind of a range. And so, you can imagine that we're not talking about $30 million, $40 million kinds of growth sequentially quarter-on-quarter; it's a much smaller level than that, again, with this slow ramp.

And that's about as much as I think I want to say about it. I don't want get into start talking about individual shipments to individual customers..

Your next question comes from the line of Ruben Roy..

Jon, on the 28-nanometer product related to the China LTE build-out, would you say that most of the equipment that you're selling into today related to China LTE specifically is 28-nanometer? And then, as a follow-up to that, in terms of the customer that you talked about that was shipping some 40, some 28.

And you guys are expecting a rapid transition to 28, do you expect that transition to happen in this 4- to 6-quarter first phase of rollout?.

Yes. So pretty much everything we're shipping into the China LTE business is our 28-nanometer product, and it's actually a combination of Kintex and Zynq in some situations. So those are the 2 leading product families. So pretty much 100% of that -- of what we're shipping in there is 28-nanometer.

On the transition for the -- to the large manufacturer that's going from 40 to 28, we would expect that to happen probably within the first -- within the next quarter or so, 2 quarters -- or probably 2 quarters more of those sockets starting to move to the 28-nanometer designs..

Your next question comes from the line of Srini Pajjuri..

Jon, just to follow up on that. You said that you're shipping Kintex and Zynq. My understanding is that some of these base stations actually use higher end parts. I'm just wondering if you're shipping Kintex and Zynq because the initial rollout is going to be only radios as opposed to base stations..

Well, I mean, the value proposition that we are delivering in the base station arena for both radio and baseband cards and backhaul is in focus a lot on the performance of the fabric.

And so, we are -- since we have a high-performance fabric on our midrange product, which is a differentiator, and it has a lot of DSP intensity, which is what's required to do all these waveform processing, if you will, really, Kintex is ideal. We've designed it -- that part to be -- to hit the sweet spot of what was required.

The reason Zynq is in there is because Zynq also carries an additional processing capability in addition to DSP, and that's obviously being used, in some cases, to even enhance the offering from certain customers. So when you think about high end, you need to think about what attributes of the high end are important.

And the most important attributes of the high end to the base stations are the DSP capability and the overall fabric speed. And the SerDes capability is, I would say, lower speed SerDes required. So this is why we decided to make Kintex focused on that -- those particular attributes and therefore that application.

And that's where we differentiate ourselves from our competitor primarily, where they end up having to use higher-end parts..

Your next question comes from the line of Anil Doradla..

When I look at your 28-nanometers, can you share with us how much revenues you got from prototyping? And as we exit 2014, if you look at China build-outs, what would you say your market share would be on the LTE rollout when it comes to 28-nanometers?.

Okay. So let me take that, give Jon a little well-deserved rest. As you look at our offering this past quarter, prototyping was just about 10%. It's great business, we love it and it's recurring business in that we have the biggest designs, the biggest parts. And as long as we have the biggest parts, that continues because there is even a market for it.

Having said that, it's now a small part of our business and has been a small part of our business for several quarters now. By far, the largest portion is Kintex and the entire Virtex product offering is catching up at this point in time. As you look at LTE, we designed Zynq particularly for that product offering.

And as you look at the midrange past quarter, we actually had, I believe, 90% shipments. And we would be very disappointed if it drops below 80%. And so -- and that's what is the best product for LTE out there. So that's how we look at it..

Your next question comes from the line of William Stein..

I'm wondering if we can talk about gross margins for a sec. You have pretty consistently been beating the target range, not taking that down but highlighting you'll stay in the same high range for a few more quarters.

What would potentially bring that back down to the range? In other words, why not up the range to approximately where you've been recently and where you expect to be in the next couple of quarters?.

Yes, so -- well, this is really around -- one around visibility and the trends in our business of where we're competing at. So in the past, we've talked about the successes we've had against not only ASIC written displacement, but ASSPs. And as we get more familiar with working in -- through some of those sockets.

It does have some slightly different business characteristics relative to margin profile, depending on the sockets. In other words, the value proposition for an FPGA is worth something -- worth a premium, though maybe not as much of a premium in certain kinds of situations.

So we really are -- we're not foreshadowing anything of that going down because we're in this. It's really more one around visibility of what we can see ahead of us. We are not intentionally planning on reducing the gross margin. We don't have anything, "Oh, it's going to drop off the table in 2 or 3 quarters." That's not what we're trying to signal.

It's really one around visibility as the composition of the new business we've been winning..

Your next question comes from the line of Christopher Danely..

I know you're not getting into the specifics around any of the customers but -- so just to clarify, in terms of the China build, did the product transition at the large European wireless OEMs -- I'm just trying to dance around the name here.

Did the product transition there, did that completely negate any growth in China in Q4? And then, when would you expect the revenue to peak from the China LTE build-out?.

Yes. So either quarter, actuals of September and our forecast for December, the decline in the European customer does not negate the increases from China. We are growing China faster than the decline. And so, I think that answered your question.

And the next one is, when do we think the peak is going to be? I really don't know how to model the subsequent wave beyond the first 200,000 base stations.

If I think about the 200,000 base station number, I mean, it kind of feels like the peak is going to be somewhere in the middle of next year, but I don't have a really solid view of that because, again, we're just starting it out now and I don't really have as much customer feedback of how things are going with a customer's customer at this point in time.

Maybe in another quarter, I'll have a better view of that..

Your next question comes from the line of Tristan Gerra..

Going back to a prior question, do you expect Virtex-7 to contribute meaningfully in any 4G whether it's in China or elsewhere? And as such, do you see the market moving to the high end of the 28-nanometer range for communication application, or do you expect the current mix to be prevailing into next year?.

We expect the Virtex to start growing at a faster rate. It already is significant. If you look at all of the Virtex product families. But Kintex is still larger. Virtex primarily is targeted at non-wireless applications, so -- and the wireless applications tend to happen a little earlier and a little faster.

The Virtex ones, we expect in 2014 and relatively early in 2014 to start to accelerate, and those are just -- it's a broad set of applications, and Virtex addresses them both. So military is another one which tends to take a little longer.

The high-end applications generally have a long qualification time, whereas the midrange, which is the wireless, tend to happen faster. Virtex, we are on plan and doing actually very well there, and it will continue to accelerate.

That's why, if you look at the overall numbers, we beat every prediction that we put in place and every target significantly. And you can see that we grew from significantly over $50 million to comfortably over $80 million. And now, we're predicting $90 million for the next quarter.

And in 2014, we would expect growth beyond that, and that will have a significant Virtex component to it..

Your next question comes from the line of David Wong..

You mentioned 20-nanometer and 16-nanometer technology.

Can you refresh our memory on which families you're sampling -- you've sampled in 20-nanometers? Or if not, if you haven't sampled yet, when do you expect to be sampling silicon in 20, when you expect to have first silicon on 16 for various families?.

So the -- we were the first semiconductor company to tape out a 28 with TSMC. We are the first semiconductor company to tape out a 20. We taped out in June. We believe we have a multi-year lead -- sorry, a multi-month lead over the competition at 20-nanometer. Actually, it's a bigger lead than we had at 28.

The first product offering there will be Kintex-oriented, shortly followed thereafter by Virtex product line. And the generation beyond that, we are working closely with TSMC. We expect to have first chips [ph] and then tape-out in 2014 with TSMC. That's the current plan.

We haven't announced which product, what -- the product tables are still something which we have not disclosed, so I'd rather not comment on that -- on 16. But on 20, the first Kintex shortly thereafter, there will be the Virtex product line on 20-nanometer..

Your next question comes from the line of Hans Mosesmann..

Can you guys give a little more detail on the legal action there on that impact to quarter in terms of what the issues are at hand and why you lost the judgment?.

Yes. So we -- there was a case in the most popular place for -- to take large companies to, which is an East Texas jurisdiction, and we lost -- we got a judgment by a jury in Q4 of fiscal year '12.

And after the jury gets done, then the judge gets an opportunity to review the decision and decide whether he wants to add any multiples to any of the aspects that were decided where the jury said we infringed on a few patents.

And it's taken that judge from Q4 of FY '12 until this quarter in order to decide what to do, and that's when he added an additional, I think, $28 million of judgment on top of the $15 million for a total of $44 million. I guess, there's $1 million [ph] around in there somewhere because the total is $44 million.

And so, now, we're in a phase now where the judge has asked us to spend a little more time with the other party and try to work out some sort of a royalty scheme to that. And then, there's a point in time where we get to appeal to a different jurisdiction, to whatever the next highest level court action would be.

I think it's back in the Washington, D.C. circuit, which deals with all patents, and our intention is to do so. So at this point in time, this is really a part and parcel of the same decision that was made back in Q4 FY '12 on the same set of patents, and we have not gotten an opportunity yet through the legal process to appeal, which we plan to do.

So there's no -- there's a P&L charge right now for $44 million, and there may be some sort of a bond posted in the future and those kinds of things, but we have not really stopped our fighting or negotiation in this particular situation because we think that the court did not find the right answer in this case..

Your next question comes from the line of Christopher Danely..

So -- well, hopefully, I can ask 2 follow-ups since I think I'm going [ph] again. One thing, so Intel talked about some delays at their 14-nanometer process yesterday.

Do you think that impacts your competitor in the 14-nanometer market? Does that make you feel better about how you are looking in terms of your process technology versus theirs? And then, my second question is just on OpEx. We've seen it creep up here over the last few quarters. We're getting closer and closer to the next fiscal year.

Any predictions on how OpEx looks going forward either next fiscal year for the rest of this?.

Let me answer the first question, Chris. With regards to process technology, we feel incredibly gratified due to the quality of the relationship we have with TSMC. As the new kid on the block, we were last to start on 28 and first to deliver on 28, and that clearly has manifested itself extremely well in terms of market share.

The revenue numbers are here. They're large, they're growing, they're accelerating. The proof is in the numbers. Similarly, in 20, we actually think we have the bigger lead in terms of time-to-market and we have a much broader product portfolio, broader and deeper one. And so, we feel very confident there.

And the FinFast program we have with TSMC at 16 and the fact that they already will, at that point in time, have learned the ins and outs of double patterning and will move onto FinFET makes us very comfortable with our decision.

So the -- there are always other alternatives out there in terms of foundry partners, but we're delighted with the one we have. And we intend to prove that, as we have in 28 and as we feel very comfortable with the 20-nanometer position, too. And with that, I'll hand it over to Jon..

Yes. On OpEx, what's happened in terms of our OpEx from our original prediction from last Analyst Day is that we've had an increase because the share-based expense has cost us more because the stock price has gone up quite precipitously in that time period and our profitability has increased faster than we had anticipated.

So our bonus plans are tied to profitability of the company. And so, that's part of the reason of why we've seen an increase over that original projection.

From inside the projection, we are starting to spend more money on mass and wafers in the second half of this current fiscal year than the previous first half of the fiscal year because we are starting to tape out some additional products, as we've been talking about and Moshe referred to earlier on 20-nanometer and a few things on 28-nanometers as well.

So there is some added costs relative to that. Going into next year, while it's next fiscal year doesn't start until April, so we're still in the middle of our planning process.

We do believe mass costs will go up next year because it will be -- we'll be starting to tape out more 20-nanometer products and then onto 16-nanometer activity that we talked about. We're also very mindful that the business model needs leverage in there. We need to be growing cash flow and earnings on a year-on-year basis.

And so, that's how we're thinking about it. We are thinking about growing the company and putting leverage in the company.

I'm not going to give you any sort of a spending growth rate percentage at this time, but we are definitely mindful of both the things we have to do from a technology perspective and the financial aspects where we need to manage the P&L for growth in terms of return to shareholders..

[Operator Instructions] And there are no further questions at this time..

Okay. Well, we'll wrap up. Thanks for joining us today. We have a playback of this call beginning at 5 p.m. Pacific time, 8 p.m. Eastern Time today. For a copy of our earnings release, please visit our IR website. Our next earnings release date for the third quarter of fiscal year '14 will be Tuesday, January 21, after the market close.

This quarter, we will be presenting at the Crédit Suisse Annual Technology Conference in Scottsdale on December 3 and the BMO Capital Technology and Digital Media Conference in New York City on December 10. In addition, please do save the date for our 2014 Analyst Meeting on February 11 in San Francisco. More details to follow. This completes our call.

Thank you very much for your participation..

Ladies and gentlemen, this does conclude today's conference call. You may now disconnect..