Scott Wylie - Vice President of Investor Relations Ronald J. Pasek - Chief Financial Officer, Principal Accounting Officer and Senior Vice President John P. Daane - Chairman, Chief Executive Officer and President

James Schneider - Goldman Sachs Group Inc., Research Division Vivek Arya - BofA Merrill Lynch, Research Division Ambrish Srivastava - BMO Capital Markets U.S. Romit J. Shah - Nomura Securities Co. Ltd., Research Division John W. Pitzer - Crédit Suisse AG, Research Division Tristan Gerra - Robert W. Baird & Co. Incorporated, Research Division Glen Yeung - Citigroup Inc, Research Division William Stein - SunTrust Robinson Humphrey, Inc., Research Division Srini Pajjuri - CLSA Limited, Research Division Doug Freedman - RBC Capital Markets, LLC, Research Division Christopher B. Danely - JP Morgan Chase & Co, Research Division Blayne Curtis - Barclays Capital, Research Division Hans C. Mosesmann - Raymond James & Associates, Inc., Research Division Michael Chu - Deutsche Bank AG, Research Division Joseph Moore - Morgan Stanley, Research Division Anil K. Doradla - William Blair & Company L.L.C., Research Division Alex Gauna - JMP Securities LLC, Research Division

Good day, and welcome to the Altera Fourth Quarter 2013 Earnings Conference Call. At this time, I'd like to turn the conference over to Scott Wylie. Please go ahead, sir.

Good afternoon. Thank you for joining this conference call, which will be available for replay telephonically and on Altera's website shortly after we conclude this afternoon. To listen to the webcast replay, please visit Altera's Investor Relations web page, where you will find complete instructions. The telephone replay will be available at (719) 457-0820, and use code 258712. During today's prepared remarks, we'll be making some forward-looking statements. In addition, management may make additional forward-looking statements in response to questions. In light of the Private Securities Litigation Reform Act, I would like to remind you that these statements must be considered in conjunction with the cautionary warnings that appear in our SEC filings. Investors are cautioned that all forward-looking statements in this call involve risks and uncertainty and that future events may differ from the statements made. For additional information, please refer to the company's Securities and Exchange Commission filings, which are posted on our website or available from the company without charge. With me today are John Daane, our CEO; and Ron Pasek, Chief Financial Officer. Ron will open the call with a few brief remarks before turning the call over to John. After John concludes his remarks, we will take your questions. Prior to the Q&A session, the operator will be giving instructions on how you can access the conference call with your questions. I would now like to turn the call over to Ron. Ronald J. Pasek: Thank you, Scott. Fourth quarter revenue of $454 million increased 2% sequentially, exceeding our expectations. We had marginally better growth than planned in a handful of subvertical markets with the wireless market as expected flat sequentially. John will elaborate on vertical market dynamics in a moment. Gross margin and OpEx were in line with guidance for the quarter. 28-nanometer revenue for the quarter was $44 million, an increase of 33% sequentially. 40-nanometer revenue grew sequentially as well, which contributed to a 10% sequential growth in new products. New products represented 47% of revenue for the quarter and 43% of revenue for the year. During the quarter, we were able to increase our share repurchase activity since we had funds available through our bond offering. After repurchasing 4.4 million shares in the quarter, our repurchase authorization sits at 36.8 million shares, so we're well equipped to continue to be opportunistic. For the full year 2013, we returned over $360 million or 61% of cash flow from operations to shareholders, which is within the range of our commitment we made earlier in the year. As for the Q1 outlook, we do see a pause in demand with revenue minus 2% to minus 6%, however, Q1 '14 does show growth over Q1 '13. Gross margin for Q1 '14 will be 30 basis points lower than Q4 due entirely to mix. Gross margin will improve throughout the year as a result of improved pricing practices and continued cost reductions. R&D OpEx will decrease in Q1 mainly due to a decline in tape-out activity from Q4 but will increase in Q2, making the first half R&D equal to what we will see in the second half. Q1 SG&A will decrease from Q4 due to the timing of projects and will remain fairly linear throughout the year. All in, we expect OpEx to be consistent with my November guidance of $750 million. Finally, 28-nanometer revenue will grow roughly 30% sequentially from Q4 '13 to Q1 '14. Now let me turn the call over to John. John P. Daane: Thank you, Ron. Fourth quarter revenue increased 2% sequentially. Wireless was flat and industrial grew for the third straight quarter, both meeting expectations. Computer and broadcast were up shortly and drove the outperformance compared to guidance. For 2013, we were impacted by lackluster end markets particularly in communications. However, 7 of our 11 sub-vertical markets grew in the year as we continue to penetrate new markets and displace ASICs in existing ones. Our automotive, industrial and military sub-markets each grew in 2013. Our computer and networking markets also increased, and we are in the early innings of FPGA adoption and servers. Wireless was down only slightly with LTE in China positively impacting our second half revenue and helping to absorb the ASIC replacement at a large customer in late 2012. Telecom, however, was down significantly as carrier spending was redirected into wireless. For the first quarter revenue, we are forecasting a sequential range of minus 6% to minus 2%. The telecom and wireless category is expected to grow with wireless up sharply on China LTE shipments, somewhat offset by a decline in telecom. The auto, industrial and military category should be flat. Where computer and networking category was up in Q4, with computer growing double digits for 2 straight quarters, we expect the category to decline in Q1. Our computer revenue, while significant, continues to fluctuate based on program buys similar to our military business. I believe in the next 2 years as our computer business grows and matures, that the quarterly fluctuations will attenuate. Our Other category we expect to sharply decline in the quarter. New products will continue to grow. Based on the submarket exit rates from Q4 and our forecast for Q1, we believe that a number of our larger vertical markets are poised for 2014 growth, including wireless, industrial, military and computer. We are in the midst of deploying an expanded product portfolio, including our MAX 10 low-end product, Arria 10 midrange family, both manufactured by TSMC, and Stratix 10 high-end family. MAX 10 release schedules will be announced this year. Arria 10, produced in 20-nanometer, will generally sample in Q1 within the same quarter as our competition. And as the only company with release software for our midrange customer -- for our midrange family, customer engagements are rapidly growing. Design of our Stratix 10 family with Intel's advanced 14-nanometer Tri-Gate technology continues smoothly. And by using process technology one generation ahead of the competition, we expect to secure a vast majority of the high-end, which is over 50% of the FPGA market. In summary, we believe our expanded product portfolio, including industry-leading software, will enable Altera to secure a market-leading position in the next-generation products. Additionally, we plan to further slow expense growth in the years to come to allow our earnings to grow faster than our revenues. Now let me turn this call back to Scott.

We would now like to take questions. [Operator Instructions] Operator, would you please provide instructions and poll for questions?

[Operator Instructions] And our first question comes from Jim Schneider with Goldman Sachs. James Schneider - Goldman Sachs Group Inc., Research Division: Related to the China LTE build-out, I was wondering if you could comment on the ratio of baseband versus radio shipments that you're seeing now? And whether you expect that to change as we get into the end of the year? And related to that, have you seen any pickup in terms of orders from China Telecom yet? John P. Daane: We have seen orders, Jim, for China Telecom from some of the vendors to that award, not all of the vendors yet. We are shipping for China Mobile both into radios and base stations. And I think, as you're probably aware, China Mobile has increased the number of base stations it plans to deploy from originally 200,000 late last year to now 500,000 through the end of the year which, combined with the radios, is a very significant increase. I think we'll see China deploy very solidly over a several-year period of time, and hopefully, we'll also see the pickup of some other markets over time, including United States with a couple carriers like Sprint and T-Mobile poised to eventually deploy, plus AT&T and Verizon needing to fill out their network combined with Europe looking like in the next couple of years it will also start 4G deployment. James Schneider - Goldman Sachs Group Inc., Research Division: That's helpful. And then as a follow-up, I noticed that your EMEA shipments were down pretty sharply in the quarter. I was wondering if you can talk about whether that's driven by 1 or 2 very specific customers? Or was that more broad-based across the end markets? John P. Daane: It was a little bit broad-based. There was a communications element of it that was there both in wireline and wireless, but there were also some other market segments that contributed to that as well.

And we'll go next to Vivek Arya with Bank of America Merrill Lynch. Vivek Arya - BofA Merrill Lynch, Research Division: Question, John, do you expect China deployments to be more 40-nanometer based where you benefit? Or 28-nanometers where Xilinx could potentially benefit more? John P. Daane: Yes, as we talked about a couple of quarters now, really what happens is, it's a collection of different process nodes rather than one -- any one process node. As an example, we did not do a low-end product in 40-nanometer. So we sold a lot of low-end product into radios, and 65-nanometers, they're still shipping in volume. So really these deployments will be, today, a combination of 65, 40 and 28. I think what you see is the radios tend to migrate faster, they're high-volume. As such they do look for cost reductions. They will move to the next generation of FPGA very quickly or, alternatively, also move to other types of technology. As you remember, the ASIC conversion we had some years ago was a radio application. The baseband tends to stay in a technology for several years before transitioning. So I think it's going to be a combination. And for us, it is a combination of 65, 40 and 28. Overall, we are #1 in China. We are -- still have in the quarter a 10%-plus customer which is tied to Asia, and we're #1 in radios, we're #1 in backhaul and, this year, we'll be #1 in baseband as well. So we're very confident that were going to do very well and you see that wireless is growing very sharply for us this quarter. Vivek Arya - BofA Merrill Lynch, Research Division: Got it. And as a follow-up, John, your 28-nanometer share has stayed around that 30% or so range for some time. What is a trigger to help you get back to -- I think your goal is to at least get to parity with Xilinx when it comes to 28-nanometer. Is it going to be a specific application? Is it a region? Is it a customer? What helps you recover your share back to your target? John P. Daane: Yes, so our expectation is that our market share in 28, as the revenue ramps, will be at least 50%. As our competition has also commented, the high-end does ramp more slowly. One way to think of this is if you look at the applications in the high-end, their baseband for -- or base stations, they're telecom, military as examples. And those systems do require a couple of years of design and software debug qualification before they're deployed. And we're in that phase right now, where we're seeing more and more of those systems enter production. And therefore, what you're seeing is strong growth from us. I would expect -- our competition has obviously, done very well and we've commented on this in the past. In radios, that business will obviously transition in the next node fairly quickly for cost reductions, as well as some other onetime events, I think. And so we would expect this year that we will catch up to our competition in 28-nanometer revenue. As Ron commented, we expect strong growth again this quarter, second quarter over 30%, and I would not be surprised in Q2 if we were to see another 30%-plus growth quarter as well. So our expectations are we're going to catch this year.

We'll go next to Ambrish Srivastava with BMO. Ambrish Srivastava - BMO Capital Markets U.S.: I had a question on share buyback, Ron. If share count has barely budged, and I understand the convert was done later in the year, but what's the right way to think about when you say, opportunistically, you have the potential to buy back a lot of stock? What -- how should we be modeling share count for the remainder of the year? That's my first question. And then on my follow-up, back to 28, John, history shows and it has shown in the nodes time or time again that share count -- share basically normalizes back down to somewhere in the 45%, 55% range, whether it was 65 or -- in 40, 45 the same thing happened. So the question is we've heard Xilinx's perspective and you have laid out your perspective which is in the high-end. Can you give us some guide posts that help us understand why the high-end should enable you to get back, i.e., whether it's applications that you have design wins and something to help us get more convinced that, that's what's going to happen and not what Xilinx is saying? Ronald J. Pasek: Ambrish, this is Ron. So to answer your question. First of all, we didn't do a convert. We did a straight bond, a 5-year and a 10 year. Obviously, when we did that in Q4, we couldn't be in the market repurchasing shares at the time. So after that went out, we executed some repurchase later in the quarter. Think of mostly December as when we did most of our repurchasing. So therefore, you don't see a huge movement in diluted share count because it's back-end loaded. What I said at the investor meeting in New York, you should expect over time that we will lie back in a linear fashion. Within a quarter, we will be opportunistic when we think there's a dislocation. So not an accelerated share purchase but you will see us be very aggressive when we can. John P. Daane: On the 28-nanometer, one thing to remind everybody is our overall market share compared to our largest competitor in FPGAs is 40%. So anything that we do in the node that's north of 40% actually means that we capture market share. And I think that's important to highlight because I think people -- or in the past where our competitor has said they plan to win 50% of the 28-nanometer node. Well, at 50%, they obviously would lose market share to Altera. And that we've had over 50% or, excuse me, over 40% market share of every node since 130. And that's also important to remember in this process as well. A second point I'd make is, as we have commented and our competition has also commented, the high-end is over 50% of the FPGA industry, and we expect that to continue in the next several nodes. And in the high-end, the 28-nanometer, we clearly had an advantage in that we used a higher-performance process, which gives us a much faster fabric. We had a superior routing architecture, which is ultimately, I believe, why you're seeing changes from our competition in 20, in routing, that allowed us to do more complex designs and close them both from a timing and from a design complexity perspective. We also had lower power. So for many reasons, I think we had a superior footprint for the high-end versus our competition which really had a superior product for the midrange. And since we won a majority of designs at the high-end, and you think of, again, telecom in many of these other markets that we highlighted during our November analyst meeting, those are our strong suits for Altera in particular. So we're seeing those high-end applications ramp now, that's the strength you're seeing from our 28-nanometer business. And as I mentioned earlier, the midrange and low end is a little bit more susceptible to moving to the next node because applications like radios or consumer, which can ramp very quickly and, therefore, give you early revenue, are also applications which tend to migrate to the next node as quickly as possible, whereas the high-end tends to stay in a node for quite a long period of time due to the complexity of the design. So hopefully, that answers a little bit of the question on 28-nanometer market share.

Our next question comes from Romit Shah with Nomura. Romit J. Shah - Nomura Securities Co. Ltd., Research Division: John, your revenue outlook for March is a little bit below consensus and you highlighted weakness in telecom, the networking computer segment, as well as Other. The weakness you're seeing in these areas, do you believe that it's kind of a 1-quarter phenomenon and that these segments will improve as we get into the second quarter? John P. Daane: It's difficult to project out a quarter. I think we've been telling everybody that for many years, and so we'll stick with a 1-quarter guidance. What I would tell you is that 2 categories that are really down are Other. We had strong growth in broadcast in the fourth calendar quarter. We expect that, as well as a couple of the other markets within Other to be down in Q1. We also expect computer to be down. As I mentioned earlier, the computer market tends to be a little bit more tied to programs today and like military, have some program buys that will come 1 quarter and then go away for the next quarter. Wireless, in particular, is growing sharply. And then again, the auto, industrial, military category is expected to be flat. So that's kind of the way it rolls out. Again, if you wanted to categorize it, computer has been a little bit more volatile than other categories. It grew double digits for 2 straight quarters. It really is a primary driver for the downfall this quarter. We would hope it recovers, but again, difficult to project exactly what's going to happen 1 quarter out. We would tell you that, so far, our book-to-bill is good and that's a good sign. We are seeing people book out further even though our lead times haven't changed. That's a good sign. But we're going to be cautious and just see how things develop this quarter and ultimately next. Romit J. Shah - Nomura Securities Co. Ltd., Research Division: Okay. And then on 28-nanometers since we're all keeping scores, Xilinx said they recorded about $100 million in revenue in the fourth quarter. Could you give us your number for... John P. Daane: As Ron mentioned in the earlier script, it's 44 end of quarter.

We'll go next to John Pitzer with Crédit Suisse. John W. Pitzer - Crédit Suisse AG, Research Division: John, I guess I want to go back and kind of compare this wireless CapEx cycle to kind of the '09, 2010 period. If memory serves me correct, from September of '09 over the next 4 quarters, you effectively doubled your revenue coming from the telecom and wireless end markets. I'm just kind of curious as you think about the China LTE build-out. Granted you're starting from a higher base today than you did in 2009, what kind of sequential growth or growth over time should you expect? And if you think about a peak quarter, what kind of revenue numbers should we be thinking about? John P. Daane: Wireless is a difficult area for us to predict. The customers are very large. Obviously, we have direct contact with those customers but they do change their forecast and change their order patterns quite regularly on us. And so it does make it difficult to project out over time. Ultimately, I think that's because the carriers order at the last minute and will push out or pull in deployments depending on what their short-term plans are. I'll give you a negative and a positive if you compare back to history. The negative was during the '09-'10 cycle, which ultimately went into '11 as well, the first half of '11 in particular, we were deploying both not only to China for both 3G and 2G, but we were also deploying into North America for the LTE builds as well. Right now, where there are builds going on around the world, the most significant one is in China right now. Hopefully, we'll see some of the other major markets start to pick up, which would mean wireless has a lot of room for growth for us and for our industry. The positive is that if you look at the content that we have in 4G, it is higher than we had in 3G or 2G. And I think that's some of the reason that you're seeing, the business pick up right now, even just based on shipments to one geography. So it's a little bit different -- difficult to compare because the time frames are different, the geographies were different, the content was different. But again, as we look at wireless, the positive overall is the increase in smartphones is saturating the network. Verizon has announced that they need to add equipment as an example. There is a need to buy the equipment and I think over the next few years, you'll see a couple more geographies other than China pick up, and all of that is obviously extremely positive for us. John W. Pitzer - Crédit Suisse AG, Research Division: And also, John, as my follow-up, as you think about the strong growth in 28 sequentially expected in the March quarter, and I guess more importantly as you close that gap with Xilinx, what end markets are driving the predominance of the growth in 28 for you? John P. Daane: It will be broad. So you want to think of it as really a lot of the markets. Because we did a low-end, a midrange and a high-end product. Generally, I think we're probably splitting the low-end with the competition. Midrange, they clearly are doing extremely well. High-end, we're doing extremely well. I would say it's most every application that we had on that pie chart that we showed in November. John W. Pitzer - Crédit Suisse AG, Research Division: So John, is the mix of 28 revenue similar to the mix of the overall business? Is that how we should think about it? John P. Daane: Probably -- I haven't looked at it, so I couldn't say for sure. I just don't know and I'd hate to guess.

We'll go next to Tristan Gerra with Baird. Tristan Gerra - Robert W. Baird & Co. Incorporated, Research Division: Given that Stratix V seems to be about 70% of your 28-nanometer mix currently, should we assume that you have a greater mix into base station and radios currently at the 28-nanometer node? John P. Daane: I think at base stations yes. Radios, I would say, mixed. Maybe our competition at 28-nanometer is better than us, although we do have 40- and 28-nanometer business that's still there. I would say that it depends on the rest of the market. So let's say, generally, in telecom, we've been ahead. Base stations, generally, we're ahead. I'd have to go to computer, I believe we're well ahead of the competition storage ahead of the competition as examples, and a lot of that stuff is 28. We're seeing industrial ramp in 28 now. So it's fairly broad. Tristan Gerra - Robert W. Baird & Co. Incorporated, Research Division: Okay. So when I combine, you and your competitor in terms of some exits, still looks like there is a majority of revenue today at least, which is about 10 quarters into the ramp of 28-nanometer from the midrange as opposed to the high-end. And that seems to be different from what we seen at previous geometry nodes. Is it the design wins that you have in place that gives you confidence that you can still get -- that the high-end will be more than a half of the total node in future quarters? John P. Daane: Yes, I think it's the market segments. And I think if you look at a product long term, as we've talked about before, the Prototyping business will move to the next node to get the higher-end devices, so that's always transitory. It's in a node about 2 years. As we've mentioned before, Xilinx obviously enjoys that today because they have much higher density parts than we do. And then the second thing is, some of the consumer and particularly the radio business will move and migrate. And so the stuff that will remain will be a lot of the other market segments for which we're strong and those market segments are ramping right now. So we're very confident over time that we have over 50% of this node. Plus, if you look at our 20-nanometer introduction roughly right at the same time as our competition on silicon for general sampling, but we're advantaged in that we're the only ones that have shipped software for customers to use for 20-nanometer design. And so we're perfectly poised with a lot of those base station accounts -- excuse me, radio accounts, to compete and take the business with our midrange offering of Arria 10 for that product line. So those are some of the reasons. But again, I think if you look at the high-end, that's an area that we've done extremely well.

We'll go next to Glen Yeung with Citigroup. Glen Yeung - Citigroup Inc, Research Division: John, if you would have looked at your non-China business, and sort of today looking at the 2014, would you characterize your level of confidence in growth for 2014 higher now than it was at this time last year? John P. Daane: For China? Or for non-China? I'm sorry... Glen Yeung - Citigroup Inc, Research Division: Non-China, non-China. Anything outside of China. John P. Daane: Well, again, I think wireless grows this year, or at least that's our expectation at this point. China will be a contributor to that on a year-over-year basis. All of the rest of the growth in areas like, as I mentioned, military, computer, industrial predominantly will be non-China growth. Glen Yeung - Citigroup Inc, Research Division: I'm sorry, just to clarify, your confidence in that relative to this time last year, greater or less great? John P. Daane: Ron and I are looking at each other trying to remember. Ronald J. Pasek: I'm trying to remember how we felt last year. John P. Daane: I don't know that I can answer that because I don't remember what we were thinking. But our -- we did have a plan for growth last year, no doubt. We did not achieve that. But we do expect growth this year. And I'm -- last year, I don't think we necessarily commented on a lot of these verticals or framed it the way we do today. So I think we do feel good that there our industrial did bottom out in the first calendar quarter of 2013. It's grown for 3 straight quarters. That's good. There doesn't seem to be any macro reason that, that would slow down, or even if it flattens out, we're still up year-over-year. Military's based on programs that we know that are going to go into production this year. We feel very good about that. Computer, obviously, you've seen that business grow very strongly for us in 2013. Early innings in that category and we've mentioned wireless. So I think generally, -- and those account for, I should mention, segments probably for about 60% to 65% of our overall revenue. So I think we feel good. The things can happen. We'll just have to see how the year plays out but we are feeling confident that we will see, certainly, growth, and certainly growth in those verticals. Glen Yeung - Citigroup Inc, Research Division: Okay. That's helpful. And then just as a follow-up. I'm not sure if someone asked this and you couldn't answer, so I'll just ask it again. Do you expect that your -- ultimately your business that comes from China for the TD-LTE build will exceed the amount of business? And you can look at it however you want, whether on a quarterly business or in its totality, will exceed the business that you could have gotten or you did get in the TD-LTE [indiscernible] come out? John P. Daane: I think over the long term, the deployment for TD-LTE will be greater. Two reasons for that though, Glen, that are pretty easy to triangulate on: Number one, you're going to see China Telecom do some TD-LTE deployment; and then number two, China Mobile obviously, is making a very heavy push to use TD-LTE, whereas the TD-SCDMA network was not as heavily utilized by their customer base. And so by nature, I think they will spend more to build out the network and some of the higher frequency spectrum that they're getting will require more radios just by nature of what the higher frequency requires. So I would expect the overall spend to be higher for TD-SCDMA. Certainly, the units will be much higher.

We'll go next to William Stein with SunTrust Robinson Humphrey. William Stein - SunTrust Robinson Humphrey, Inc., Research Division: I'm wondering if you can comment as to when you expect to start seeing measurable revenue from your 20-nanometer and ultimately your 40-nanometer product. John P. Daane: 20-nanometer will start to sample if you look at in Q1 -- excuse me, if you look at the typical pattern you see it takes about 1.5 years for it to be about 5% of revenues, maybe a year to be 5% of revenues in a quarter, and then it grows from there. Typically the year for strong growth is the third calendar year that you'll see for a product line. And then the other thing is we will be shipping our 14-nanometer in the first calendar quarter of 2015 based on the fourth calendar quarter 2014 tape-out that we've talked about before, and then you can kind of follow the math on that one as well. William Stein - SunTrust Robinson Humphrey, Inc., Research Division: That's helpful. Maybe one other way to kind of go out this TD-LTE rollout. Do you anticipate this getting to a point where you see meaningful increase in the overall telecom and wireless market relative to where you are today? John P. Daane: It's possible that telecom inventory or wireless in particular, will grow as a percentage of business. I would just bet that it would, based on the fact that we expect it to be one of the bigger growth drivers this year. But it's hard to significantly change any of the percentages within 1 year because we have so many vertical markets that are so large, our business is fairly distributed that you do not see typically a 5% change in overall as a percentage of our overall business in any vertical in any year. Just -- that would be very, very difficult to do. And generally where you saw the strong growth from our company as an example, on 2010, much faster than the industry, just to remind everybody, that was very broad-based growth. Wireless grew very strongly that year. But just about every market for the company grew between 55% to 75% over the prior year 2009. So again, that's I think why you typically don't see significant shifts as a percentage of our overall business just within 1 year. Wireless will grow as a percentage, but probably not 5 points, probably more like 1, maybe 2. William Stein - SunTrust Robinson Humphrey, Inc., Research Division: If I can just squeeze in one more, in what inning would you estimate that we are in terms of the rollout of TD-LTE in China relative to component vendors? Because I understand that there may be some inventory at either the OEMs or at the carrier. John P. Daane: It's -- that one is extremely hard to predict because if you go back 2 quarters ago, China Mobile was saying that they were going to deploy 200,000 in Q4. Obviously, that shifted out a little bit. Then they were making a comment that there would be perhaps an additional 100,000 to 200,000 this year, or maybe slipping into 2015. And then they recently came back and said they're going to do an additional 300,000 this year. Effectively pulling in a lot of shipments that were expected out over another couple of years. And so what you're seeing is carriers regularly change their mind and, therefore, very difficult to predict when this peaks, how long it lasts, how much they'll deploy. And again, it will also be based on what other carriers do in China and what other carriers do around the world and we do expect that in the next few years. We're going to see more in the U.S. and certainly a start in Europe. And I would say probably 3G will pick up again because some developing countries like India will start deployments en masse. And all of that would be great for us because of our strong content within wireless.

We'll go next to Srini Pajjuri with CLSA Research. Srini Pajjuri - CLSA Limited, Research Division: John, the Other segment it's a decent chunk of your business and it grew nicely. I'm wondering what all is included in that? And then what's driving that growth or what drove that growth in Q4? And looks like it's declined pretty meaningful in Q1, just trying to understand the different dynamics there. John P. Daane: Yes, it includes consumer, test, medical and broadcast. As I mentioned, at the beginning of the call, broadcast was up strongly in the fourth calendar quarter and that is a market that we expect to be down. And then a couple of the other markets like, for instance, consumer, which is tied to seasonality of holidays, is expected to be down this quarter. And so that's what's driving the overall category down. I wouldn't say that there's any particular trend in that business as in you should -- we don't expect it's going to trail off. That business probably, in some quarters, some years, will be up, some years will be down. But overall is -- we just call it Other because it's hard to lump those 4 verticals into something that make sense. So it's just Other. Srini Pajjuri - CLSA Limited, Research Division: Okay. And then, Ron, on the gross margins, you said you expect margins to come up in the second half. I'm just curious, you guided for 69% at the Analyst Day, so do expect for the full year, 69%? And then historically, when China was strong because of larger customers, your gross margins actually were negatively impacted. So if China remains strong, can you still improve your gross margins? Ronald J. Pasek: Yes, so I'm confident that you'll see margin improve throughout the year. It is being affected by the strong quarter in Q1 for wireless particularly in China. I mentioned in my script some pricing practices and cost reductions. There are some things we're doing that are different than prior years. So I expect that to kick in in the last 3 quarters. So I'm still confident of the guidance I gave at the November meeting.

We'll go next to Doug Freedman with RBC Capital Markets. Doug Freedman - RBC Capital Markets, LLC, Research Division: Now that you're sampling your 20-nanometer solution, can you give us an idea of where you think that, that'll find traction? John P. Daane: We'll be sampling 20-nanometer in the March month. The 20-nanometer initial designs are broad-based because it is a midrange family that actually has higher performance than the current Stratix V series and has some higher density as well. So it's a pretty good series with high-performance transceivers as well. It covers a lot of markets. We're seeing a lot of traction in wireless. We're seeing a lot of traction in military, telecom applications. So it's -- and those are just a few of them. It's computer -- it's pretty broad-based at this point. I have a lot of customers designing. And again, if I were to say what is our advantage today, it's having the software out, and that's allowing us to engage with customers and allowing them to do the design work. And that's really allowing us to go to our market much earlier than the competition. Doug Freedman - RBC Capital Markets, LLC, Research Division: Great. And if I could, for my follow-up, one for Ron. On the OpEx, you said you're sticking with your 7 50 for the year. We're starting off well below that trend. You're calling for an uptick in the June quarter to have the first half and back half equal each other. It would assume that I have to have a downtick to have that balance. Is that downtick in September or December? Ronald J. Pasek: No, I'm not going to say. What I'll say, remember in the November meeting, I mentioned that there were 2 things that were incremental in spending from '13 to '14. One was the annualization of the 2 acquisitions we did mid-2013. And the second one, I showed on a chart a $32 million for increase in mask and wafers and tools, tape-outs supporting the product plan we have. So those, as you well know, are very lumpy. So that's what you're seeing, you're seeing no lump in Q1, no lump in Q2 and you'll see a lump in the second half at this point.

We'll go next to Christopher Danely with JPMorgan. Christopher B. Danely - JP Morgan Chase & Co, Research Division: Just a follow-up on the previous question. Ron, I know this is longer term, but you said the expense growth will slow markedly next year in 2015. Is it possible that OpEx could be flattish or slightly up? Or R&D could be flattish or slightly up for next year? Ronald J. Pasek: It's a good question, Chris, and John talked about this in his script as well. I don't want to comment on what we think it will be other than our plan is to grow OpEx slower than revenue. We haven't done the plan yet. Again, we try to be competitive with salaries. So you're always -- you always have a little bit of a headwind for raises and things like that. So there's got to be some increment. But again, I think what I talked about in November was an operating model that encouraged minimal OpEx growth and we grow revenue faster than OpEx for several years. Christopher B. Danely - JP Morgan Chase & Co, Research Division: Great. And then my follow-up question probably for John. John, I think you mentioned in the past that you've used 20-nanometer node as, I guess, transitional or a little bit smaller. Is that -- does that still hold? I guess what will be your view on 20-nanometers? What other can we expect in a bunch of other product families to come out at 20? Or were you just going to make the leap directly to the Intel at 14? John P. Daane: Chris, the reason we're doing only a midrange in 20 is, when we looked at the high-end, we originally had higher density members of the family planned and power consumption was just too high. And it's because we're using a planar process. And in many of the applications that we're in, the applications or -- they're not in applications or boxes which have cooling or fans. They're in things like radios or base stations or telecom systems or military where power is a very significant issue. And we found with the higher density or high-end members using 20-nanometer, the power consumption simply was too high for the applications and not usable. And so with the FinFET availability, the high-end really needs to be a FinFET, because that allows you to get the power consumption down for the much higher density fully featured devices. And so for us, we'll move it to '14. I would expect the competition will probably do something within what is called 16FF with the competition, which is more of a 20. And I think overall, because we're a generation ahead in process technology, that will allow us to have much higher density, much higher performance, as well as lower power, which is why we believe we really capture a vast majority of the high-end when we introduce the 14-nanometer. But again, the reason for only a midrange in 20 really came down to power consumption, and it just wasn't doable within a planar process.

We'll go next to the Blayne Curtis of Barclays. Blayne Curtis - Barclays Capital, Research Division: Maybe just the high-level -- I'm just trying to understand the mechanics on the Guide-by segment. You're guiding the whole business down. It would suggest some pretty material double-digit declines in networking, computer, Other, and only a modest increase in Telecom and Wireless. I just want to make sure I have those mechanics right. John P. Daane: Yes, there's a modest increase in telecom and wireless because where wireless is up sharply, telecom is down. Then you get the automotive, industrial, military segment, as we coached, was flat. Networking and computer are both down, but it's really driven by computer. And as we mentioned, it's the program timing situation that we're in because of the data center builds that tend to build for a couple of quarters and then pause and again, as our business in computer matures, we would expect those fluctuations to attenuate. And then the Other is down both because of broadcast and consumer. So -- and between Other and computer and networking, those groups are down quite significantly, and you're right, that does drive the overall business down. Blayne Curtis - Barclays Capital, Research Division: Just for a follow-up, if the telecom -- wireless is up modestly, wireless up sharply, obviously, that means wireline is down pretty materially. Could you just explain what's causing that? It was weak in December and then now much weaker in March. John P. Daane: Slower deployments in optical and access are the primary reasons for that. Remind you that within the telecom space is traditional telecom equipment, and generally, what's in our networking space, which is combined with computer, is more of the traditional networking products such as routers and switches that would be sold to corporations rather than service providers.

We'll go next to David Lewis [ph] with Indaba Global Research [ph].

I just have one question, a clarification. In the Telecom and Wireless business, what approximately was the mix in the fourth quarter between the 2 categories? John P. Daane: Wireless is about 2/3, Telecom is about 1/3. And you contrast that with probably 1.5 years ago, we used to say 60-40. So Wireless has grown a little bit more, Telecom has shrunk. Obviously within the 2013 numbers that I mentioned where wireless was only slightly down, telecom was down significantly for the full year. That's what's caused that slight shift.

Oh, I see. I was trying to reconcile your slower numbers relative to Xilinx. I guess the category's a little differently defined. I get that.

We'll go next to Hans Mosesmann with Raymond James. Hans C. Mosesmann - Raymond James & Associates, Inc., Research Division: John, TSMC has recently kind of indicated that, at 16FF, the transistor density is not as poor as some had thought and it approaches the transistor density of Intel at '14 FinFET. Can you weigh in on that commentary? Maybe they're not as far behind as some had suggested? John P. Daane: So we used only publicly announced material when we talked about that. And so if you go back in time, there is material from TSMC that shows that the scaling or -- which is density, between 20 SoC and 16FF they're the same, there's no scaling. And obviously, we used the Intel material and then both on wafer prices being higher in 16 that came from both conference calls from Xilinx, as well TSMC. So we were completely using what is information provided by the vendors themselves. Now I did hear within the TSMC conference call that they commented that their 16FF may have scaling today and maybe old material was wrong, I can't comment on that. That could be true. There was a second element that they discussed which is the fact that the transistor is faster, therefore you're going to need fewer gates to achieve a design, and that will also mean the dye size is smaller. That would be true with any vendor of a FinFET. And so when you compare technologies, you're not going to give one vendor that benefit and not the other vendor that benefit. So you really need to pull that out and just look at the pure scaling from the design roles. And I think from that perspective, we see that the company that we're engaged with, Intel, is still a full generation ahead, density, performance, power. And we're absolutely convinced that we're going to have the best product for the high-end space.

We'll go next to Ross Seymore with Deutsche Bank. Michael Chu - Deutsche Bank AG, Research Division: This is Mike Chu for Ross. A question about telecom and wireless revenues. If I look back to the '09 through 2012 timeframe, I see a lot of volatility in the quarterly or the sequential growth rates of that segment. Then I noticed in -- just from the past few quarters, that growth rate has sort of moderated smaller up, smaller downs. I was wondering what is driving that and I guess could we see a quarter with the pull in of the base stations and the 500,000 number, could we see a quarter where we see kind of that double-digit kind of growth again? Or has something materially changed in that end market segment for you guys? John P. Daane: I'm just looking at -- I don't have the full history of several years and quarters in front of me, so I don't know that I can really comment on that. And yes, I don't know if I could really offer any clear guidance there. I mean, there have been times in prior years that you've seen the telecom and wireless segment, as well as many of our other market segments grow double digits in a quarter. That is entirely possible given any quarter. You see that... Michael Chu - Deutsche Bank AG, Research Division: Yes, for example, in the second quarter, third quarter of 2010, you saw 20% increases, and then maybe the June sort of '12 period, you saw 33% increase sequentially. I'm just wondering, is that possible again? Or did you just have more diverse customer base, more diverse products into that end market where you wouldn't see that kind of volatility? John P. Daane: I think you'll always see some volatility within these markets depending on what's happening. I think Telecom and Wireless is a segment that has a little bit more concentration in terms of customers, which is why we tend to have larger vendors as a percentage of our overall business. It's still very possible for us to see very strong growth quarters. I think we tend to -- as we've discussed in the past, underperformed during a downturn probably because people cut inventory on PLDs. We outperform on an upturn probably because we've been in a situation where they've overly depleted our inventory need to buy. Combined with the fact that 2010, '11 were really infrastructure spend years, '12 and '13 had been more in the consumer area. It feels like there's a little bit more of a shift back to the infrastructure area. Infrastructure benefits us. We're not a consumer player. So we could see those. I just -- it's impossible for us to predict or say for sure. Michael Chu - Deutsche Bank AG, Research Division: Okay. And then just a quick follow-up on gross margins. You guided gross margins is down a bit because of mix. I was just wondering if that's mainly driven by the strength in wireless? Or is there also the fact that broadcast and computer are down that's also driving that slight decline in gross margins? Ronald J. Pasek: Exactly. It's a combination of a strong wireless quarter, a weak Other, a weak computer and a flat aim, yes. John P. Daane: Telecom -- telecom's down.

We'll go next to Joseph Moore with Morgan Stanley. Joseph Moore - Morgan Stanley, Research Division: In your comments upfront, you talked about the computer business being lumpy and having kind of program buy similar to military. I guess that characterization kind of surprised me. So I don't really think of it being that project-oriented. So what are the lumps that you see there? And... John P. Daane: So a lot of the early revenue that we have has been from companies who build their own servers and, therefore, are populating their own data farms or their own systems and farms. And so as our business starts to penetrate the traditional server manufacturers themselves, I think that's where you'll see the business become a little less volatile. But today, you do see a number of the search companies, social media companies that build large data centers that will do heavy build-outs 1 quarter and then maybe stop for a quarter and then start again because they're doing it location-by-location. And that's why our business is slightly more volatile right now than it otherwise would be. Joseph Moore - Morgan Stanley, Research Division: Okay. That's helpful. And then is there any update on the stock silicon interconnect roadmap for 3D? What are your plans there? John P. Daane: Yes, so our plans are to utilize that technology for building complex system products. We have not announced anything specific as yet, so I will let the marketing group announce that before we talk about it here.

We'll go next to Anil Doradla with William Blair. Anil K. Doradla - William Blair & Company L.L.C., Research Division: Couple of questions, I mean, China has a checkered past in terms of announcing how many base stations they're deploying, how many eventually works out. So John, you made some comments about 500,000 base stations. So I have 2 questions around this. One, based on your historical knowledge of China, what do you think will play out? I mean, there are many puts and takes? And second, when you look at China, clearly, that's going to be a big driver for your growth this year and next year. What do you consider a good kind of build-out with respect to Altera? Are you talking about, if there's 150,000 base station rollout that's good enough for you guys to do a 10% growth on the top line for 2014, assuming we don't see any deterioration of other segments? Can you give us some qualitative commentary? John P. Daane: Okay, the second part of the question, I really can't comment on. We haven't provided any detail of the ASP and numbers of units per system or percentage of business that we would get from X base stations or Y radios. So I've got nothing to add there. As to China, I think China and the carriers are very similar to really what we've seen with carriers elsewhere in that what they announce and what they actually do in the short run does change. They could push out, they could push in. We've seen that also from the major carriers in the U.S, where they may announce an increase in CapEx, and you see it's a couple of quarters before that actually happens. However, what we have seen from China is they tend to deploy more equipment than is actually publicly announced within the bids. And that's always been a great benefit of ours. And to some extent, I think that additional equipment does go to some of the local vendors. And obviously we're very strong in China and we benefit from that. So I think over time there'll be more equipment deployed than you're seeing in the numbers that are announced. Certainly, we saw that in all of the 2G and 3G systems and bids. Anil K. Doradla - William Blair & Company L.L.C., Research Division: And as a follow-up, do you guys worry about CapEx allocations from infrastructure to handset subsidies? John P. Daane: It's always been curious. And so I would think it would be a tremendous positive if those subsidies were to lighten. Because in talking to some of the carriers, they will spend as much sometimes on the subsidies as they do on CapEx. And so that's why it's curious to me. I think if they were ever to change that policy, then there would be a significant amount of cash available to improve and build out their networks. And I think that would be very positive for us. Not something obviously we control but it is something I do monitor.

We'll go next to Alex Gauna from JMP Securities. Alex Gauna - JMP Securities LLC, Research Division: John, you talked about experienced softness in networking related to some of the enterprise demand situation that's presumably a result of cloud migration. On the other hand, you also mentioned exposure to lumpy compute build trends from some of the larger build-your-own server guys. To the best of your ability to see those 2 forces working against each other, how are you faring in that? Meaning, what projects or what products might you be related to on either side? And what does that mean for your growth prospects in those areas? John P. Daane: Yes, in computer it's all high-end, so it's all going to be our Stratix technology. And really what they're doing is using the product for acceleration. CPUs generally, whether -- you can choose any instruction set, were ultimately not designed to handle mathematics algorithms. And the FPGA having the DSP structures embedded in them with the logic that allows you to do a lot in terms of shaping and preprocessing traffic allows an acceleration of algorithms-like search, which is a math algorithm or compression, as an example, which is used for pictures. And so a lot of the companies are finding that using an FPGA, you can accelerate your algorithm at a fraction of the power consumption of what a CPU would take. And as you're probably aware, power is the #1 spend for a data center, not the CapEx itself. So using an FPGA to lower power is actually a double win. So it's our high-end product, again, we have seen this happen to us. I think it was, I want to say, the first calendar quarter of last year may have been soft in computer as well. Some of that is because the first quarter is always soft in computer generally. But additionally, just -- there's been some volatility in the past in the computer area in particular. It could be that Q2 is also soft, I don't remember. Alex Gauna - JMP Securities LLC, Research Division: Okay. And then also relating to your comments earlier on what TSMC is saying about their 60-nanometer FinFET versus what you're doing with Intel at 14-nanometer? Do you anticipate Intel is the only roadmap you need in FinFET for the foreseeable future? Or would you imagine broadening your product mix to incorporate what others in the foundry business are doing? John P. Daane: Yes, at 14- or the 20-nanometer node, because you've got 2 companies that are doing 20-nanometer FinFET, you've got Intel doing a 14. Within those, 14 with Intel is all we're going to do. When it goes to 10-nanometer, that's where we will obviously look and see what's the best technology both from a capability of performance, density, cost, availability and we'll make a decision at that particular time.

Operator, we have time for one more question, please.

We actually have no other questions at this time.

Well, listen, with that said, why don't we wrap up today? And as a final comment, as to conferences this quarter, we'll present at the Goldman Sachs Technology and Internet Conference in San Francisco on February 12. And then on March 3, we'll be at the Morgan Stanley Technology, Media and Telecom Conference also in San Francisco. This concludes Altera's earnings conference call. Thanks for your interest and participation.