Sukhi Nagesh - Vice President of Investor Relations Sehat Sutardja - Co-Founder, Executive Chairman, Chief Executive Officer, Chief Executive Officer of Marvell Semiconductor Inc, President of Marvell Semiconductor Inc and Director of Marvell Semiconductor Inc Brad D. Feller - Interim Chief Financial Officer, Principal Accounting Officer, Vice President and Corporate Controller

Ruben Roy - Mizuho Securities USA Inc., Research Division Doug Freedman - RBC Capital Markets, LLC, Research Division Harlan Sur - JP Morgan Chase & Co, Research Division Christopher Caso - Susquehanna Financial Group, LLLP, Research Division Quinn Bolton - Needham & Company, LLC, Research Division Daniel L. Amir - Lazard Capital Markets LLC, Research Division Craig A. Ellis - B. Riley Caris, Research Division Glen Yeung - Citigroup Inc, Research Division Blayne Curtis - Barclays Capital, Research Division Steven Chin - UBS Investment Bank, Research Division John W. Pitzer - Crédit Suisse AG, Research Division

Good day, ladies and gentlemen, and welcome to the Second Quarter 2014 Marvell Technology Group Limited Earnings Conference Call. My name is Tony and I will be your operator for today .[Operator Instructions] As a reminder, this conference is being recorded for replay purposes. I would now like to turn the conference over to your host for today, Mr. Sukhi Nagesh, Vice President of Investor Relations. Please proceed.

Thank you, Tony, and good afternoon, everyone. Welcome to Marvell Technology Group's Second Quarter Fiscal Year 2014 Earnings Call. With me on the call today are Sehat Sutardja, Marvell's Chairman and CEO; and Brad Feller, Marvell's interim CFO. We will all be available during the Q&A portion of the call today. If you have not obtained a copy of our current press release, it can be found at our company website at the Investor Relations section at marvell.com. We have also posted a slide deck summarizing our quarterly results in the IR section of our website for investors. Additionally, this call is being recorded and will be available for replay from our website. Please be reminded that today's discussion will include forward-looking statements that involve risks and uncertainties that could cause our results to differ materially from management's current expectations. The risks and uncertainties include our expectations about our overall business, our product and market strategy, statements about market acceptance of our product, statements about general trends in the end markets we serve including future growth opportunities, statements about market share and statements regarding our financial outlook for the third quarter of fiscal 2014. To fully understand the risks and uncertainties that may cause results to differ from our expectations and outlook, please refer to our earnings release, our latest quarterly report on Form 10-Q and subsequent SEC filings for a detailed description of our business and associated risks. Please be reminded that all of our statements are made as of today, and Marvell undertakes no obligation to revise or update publicly any forward-looking statements. During our call today, we will make reference to certain non-GAAP financial measures, which exclude the effect of stock-based compensation, amortization of acquired intangible assets, acquisition-related costs, restructuring costs and certain onetime expenses and benefits that are driven primarily by discrete events that management does not consider to be directly related to our core operating performance. Pursuant to Regulation G, we have provided reconciliations of the non-GAAP financial measures to the most directly comparable GAAP measures in our second quarter earnings press release, which has been furnished to the SEC on Form 8-K and is available on our website in the Investor Relations section. With that, I would now like to turn the call over to Sehat.

Thanks, Sukhi, and good afternoon, everyone. Today we reported second quarter revenues of approximately $807 million, an increase of 10% from the prior quarter and at the high end of our guidance range. During the quarter, we experienced better demand in our storage end market, while mobile and wireless was in line and networking slightly below our initial expectations. We delivered the following non-GAAP results for Q2: Gross margin of 53%, operating margin of 13% and earnings per share of $0.23. In addition, we continued to return cash to our shareholders as we repurchased roughly 7.2 million shares, totaling about $83 million, and paid approximately $30 million in dividends during the quarter. Before I discuss each of our end markets, I would like to first take a moment and provide you a brief update on the overall company as we stand currently. I am pleased to report that over the past few quarters, we have made tremendous progress in executing our strategy. We are focused on growing across all of our end markets and continue to raise the bar for innovation in order to be more competitive. We have always believed that if we continue to push the envelope in innovation, it will naturally lead to commercial success, which is what we are seeing play out today. One recent example of our innovation is the ARMADA 1500 platform that powers the new Comcast device. The Comcast is an innovative new product that allows customers to seamlessly connect and stream video content to any large screen device from any mobile device, including smartphones, tablets and even PCs. Given the rising popularity of Internet video and TV on mobile devices, we expect that the market potential of this application to be significant, especially given its functionality and the extremely attractive retail price…

Thank you, Sehat, and good afternoon, everyone. As Sehat mentioned, we reported revenues of $807 million for the second quarter of fiscal 2014, a sequential increase of approximately 10% from the previous quarter and at the high end of our guidance. In storage, our overall revenue increased by 8% sequentially, representing continued above market performance. Storage represented 52% of overall sales in the quarter. The growth in storage is mainly due to a combination of steady share gains and new product ramps at our customers. In networking, our revenue was down 4% sequentially and represented 21% of total sales. During the quarter, we experienced some softness in demand from our PON customers, which more than offset growth in our core enterprise networking end market. Our networking business is currently strongest within the enterprise portion of the market. Although there have been signs of recovery, we are still cautious. This is consistent with what a large customer indicated recently, stating that they are seeing some inconsistencies in the global recovery. Although we are steadily building the carrier and data center side of our networking business, these design wins typically take longer to translate into revenue than some of our other businesses. Our mobile and wireless end market grew strongly, up 30% sequentially and represented 22% of overall revenue in the quarter. We had solid double-digit growth in both our mobile and wireless businesses due to new product ramps and multiple customers. Moving next to margins and expenses, our non-GAAP gross margin for the second quarter was 53%, which is at the high end of our guidance, mainly due to better mix. While non-GAAP operating expenses came in at $320 million. This resulted in a non-GAAP operating margin of 13% for the quarter. Net interest and other income was $8 million. This…

[Operator Instructions] Your first question will come from the line of Mr. Ruben Roy of Mizuho Securities.

Sehat, wondering if -- I understand you have talked about double-digit growth for both wireless and connectivity. Just wondering if you can give us a little more granularity around those constituents given the execution has been so good into Q2 and, obviously, with the guidance for Q3, please.

So I'll cover this. It's a combination of different things, Ruben. So it's additional devices based on our dual core platform. It's the ramp up of our quad core platform for mobile and then the related wireless that'll go into those mobile applications, as well as the launch of the gaming consoles for the 2 guys that we deal with. Those are the biggest pieces there.

Okay. And given that this year is a little bit different from past years, can you maybe, Brad, talk about how you guys are thinking about seasonality as you look beyond Q3? Is this going to be a typical seasonal year as you look at Q4? Or do you think there are different moving parts this year, please?

Yes. So traditionally for us, our Q2 and Q3 are seasonally strong quarters. Q4 and Q1 are seasonally down. I wouldn't expect anything different this year. So I would expect Q4 to be seasonally down, as well as the fact that it includes January. You should model that down.

Okay. And then just finally on the solid-state drives commentary, year-over-year growth -- year-over-year revenues doubling, where do you think you are in terms of market share? And how much more room is there to grow market share from current levels?

Yes. So we're getting to approach a similar kind of market share as we have on the HDD side of things. So we continue to get traction there, I don't know that we've disclosed a specific number, but we continue to advance our share there.

And if you remember, last year...

I think what we've reiterated, we're talking about in terms of the market there is not proprietary based on, I mean, the merchant market -- segments of the market.

So Ruben, remember, last year, we were maybe about 50%, we exited the year last year about 50% or so, the merchant silicon market. We've been gaining share in Q1. We've gained share again in Q2. And we expect to do -- gain at least 5, if not more than that, point of share in SSDs this year in the merchant market.

[Operator Instructions] Your next question will come from the line of Mr. Doug Freedman of RBC Capital Markets.

When I look at the incremental gross margin on the next quarter, pretty low number here. And I'm a little -- I understand the growth in wireless coming at us and that's going to be a contributing factor. When we get into the seasonal quarters, where wireless is not as big a mix of the business, should we be modeling gross margins to recover? Or are there other things taking place here that are going to keep the gross margins under pressure? And I ask that with the caveat that you did just see a 30% growth in wireless. And I would've expected mix in the July quarter to a pressured gross margins and yet we saw a nice number in the quarter just reported.

Yes, that's a good question, Doug. So the way to think about it, you're right, is it's very mix dependent. So in quarters where mobile and wireless and some of the consumer businesses are seasonally down, you should see an uptick in gross margins. Now the reason you didn't see this much in Q2 is because our storage business was so strong that it helped offset some of that decline in margin with most of the -- a lot of the growth in Q3 coming from the mobile and wireless piece of the business. That's what's putting pressure on the margins. But we've talked about the fact that as that business grows, we'll be between 50% and 55%. But keeping it, it will push towards the lower end of that and then it will pull through operating margin leverage for us.

We will also say that in the meantime, when the majority of our wireless business -- our mobile [indiscernible] is still in the 3G side, that where the 3G side is where we're see the margin pressure as -- down the road, as we go through the LTE, the margin will naturally improve back to the more normal range.

And there's mix with mobile as well that should definitely help us out.

Okay, my follow-up is really regarding that LTE and when it becomes the material piece of the revenue mix. Will you start shipping your LTE solution in Q3? And what does that look like as far as a ramp over the next 3 or 4 quarters?

We're already starting shipping the LTE for the -- in the smaller volume, obviously, for the standalone modem. However, for the single-chip integrated quad core LTE, our expectation is the shipment is in Q4, those -- at the end of the year. So that's today's -- based on what we've seen from the customer plan of production. So...

And then I would expect, Doug, that you'll see additional models in early next year to launch from customers as well.

Yes, definitely there'll be more next year.

Your next question will come from the line of Harlan Sur of JPMorgan. Harlan Sur - JP Morgan Chase & Co, Research Division: If we assume that the team continues to be successful at driving your unified platform, obviously 3G today, 4G soon, I know the focus is on the op margins. But could your overall gross margins drop into the high 40% range if you continue to be successful and drive mobile and wireless faster than the overall business?

So as we talked about, Harlan, a couple of things. We have a lot of pieces to our business that allows us to be comfortable. We'll keep them above 50%. Also as Sehat alluded to, the margin in LTE, we should see an uptick there. So it's something we'll have to continue to manage. But we're comfortable we can keep it above the 50% line. Harlan Sur - JP Morgan Chase & Co, Research Division: Okay, great. And then your Q3 guidance calls for a slight decline in your HDD segment, if I look at your HDD shipment TAM for Q3, especially on 2.5-inch, 500-gig per platter, this segment is actually expected to grow about 3% to 5% this quarter. And then on top of that, you've got new game console ramps, which should benefit Marvell, because I think your HDD controller chips are also being used on the hard drives, which are powering the PS4 and Xbox One. So help me understand why your HDD business is going to be down sequentially this quarter? Was there a bit of pull forward of the Q3 business into Q2? Is there some inventory work down that's taking place at the component level? Any help here would be appreciated.

Sure. Yes, so just a reminder that our second quarter includes the month of July. And if we saw the -- the uptick we saw in Q2 and the results was more on the back end side of that. So that uptick in demand that our customers are seeing for their third quarter, a fair amount of that fell into our July month. So it's really the month offset that you see there.

Your next question comes from the line of Mr. Chris Caso of the Susquehanna Financial Group.

I wonder if you could talk a little bit more about the handsets and understand that the both the dual core and the quad core are some of the contributors to growth in the third quarter. Could you give us some sense of perhaps what we're likely to see with regard to design wins as we go out in the back half of the year? I know you guys had some aggressive goals with regard to market share for the end of the year. But what should we expect to see as some of the OEMs begin put out press releases?

Yes, so as we've been saying over the past couple of quarters, we started with the dual core design wins. And in the last quarter, also, a lot of the [indiscernible] designs and heavy wins start moving to the quad core solutions. And we also stated that the quad core solutions is just starting to ramp up in productions. So I think that these will not change, it will continue -- we'll continue to be the same. The design wins, not just in the smartphones, but also in tablets, because as we say in the last couple of quarters, we are focused on the single platform architecture so that our customer can build either tablets or smartphones based on the same investment that they have in engineering. So that's -- that will lower the cost of R&D from our side, as well as from our customer side. And the same thing with LTE side. Our LTE is the same space on the same platform as the 3G. So the same strategy again there. The investment that our customers have put into their 3G platform with us, okay, will directly translate into easier transitions on the 3G platform to the 4G platform. So a lot of the higher end design wins are also moving to the LTE platform. Including to people that don't even plan to have -- to use the LTE in certain markets, but they still want to have the LTE devices as a check point in the marketing that it is a 4G phone. So we are very excited that the progress that we have made in the last couple of quarters is finally bearing the fruit.

Yes. And Chris, to touch your question about the 10% market share, I think you're alluding to, is we've seen strong design traction, we'll see additional models in Q3 and Q4. It's really going to be dependent on how successful those models are in the market. But we continue to see the traction we like to see and we continue to watch it closely.

Okay. And I guess as a follow-up to that, perhaps you can talk to the sustainability of the design wins that you've won and your mobile and wireless business, it's kind of grown in a bit of fits and starts. With the success that you guys are having right now, what provides you the confidence that these design wins will be sticky and we'll see continued momentum as we go into next year?

Yes. I think the answer to that is the LTE. Our early investments in LTE, as well as our decision to move to LTE, second-generation LTE, as our launching point of our LTE's 4G solutions is or that's the second. Third, our continued investment in next-generation LTE, and finally, the fact that we are quite ahead compared to the other companies outside the first one, they're already in the market today, they have LTE. That's the reason that we believe that the design wins that we have with our customers will continue to increase, to improve, over the next year, and then -- plus the continued improvement in building next gen -- the next generation devices. So as we said earlier, as long as we continue to invest in this -- to innovate in developing chips data, solutions data, they are more cost-effective, more innovative performance in power dissipations and overall cost, bond cost-reductions. We believe we will make this business to be successful.

Your next question comes from the line of Quinn Bolton of Needham. Quinn Bolton - Needham & Company, LLC, Research Division: Sehat, I just wanted to follow-up on the LTE modems. You had mentioned in the prepared comments that you're looking to get certification by the end of 2013. Wondering if that's for FDD or TDD market or both.

Yes. So the full certification there we talked about for the FDD for the North America market. The vast majority of the North America market is still FDD at this point. We are actually, our LTE modem is fully certified in the TDD market. But since those markets are mainly in Asia, those certifications are in Asia, specifically more in China. But in the long run, both FDD and TDD solutions actually -- I shouldn't say in the long run. Our TDD and FDD solutions are already unified solution today. But in the long run, the certifications will be FDD and TDD, both for the North America and in Asia. So there will be no difference in the certification in the long run. Quinn Bolton - Needham & Company, LLC, Research Division: So just to clarify, you already have the TD-LTE certification for Asia today?

Yes, correct.

Yes. That's correct. Quinn Bolton - Needham & Company, LLC, Research Division: Okay, great. And just a follow-up question on the 3G market...

Excuse me, Quinn, but it was not just FDD LTE. We have also demonstrated the voice of TD-LTE, like a couple of months ago in Asia, if you noticed the announcement with ZTE on the trial runs of the voice of the LTE with the TDD networks in China. Quinn Bolton - Needham & Company, LLC, Research Division: Okay, great. And then the follow-up question is just on the 3G TD-SCDMA market, it sounds like there's been some accumulation of inventory, at least in that low-end smartphones segment. Wondering if you're seeing that impact your business or whether that's really a function of a lot of inventory for single core smartphones and you're coming in above that with the dual core and quad core platforms?

Yes, Quinn, I think on the inventory side, remember, we are traditionally more closer to Tier 1, Tier 2 OEMs as opposed to the white box market, and you may be hearing about more inventory accumulation from some of our competitors in the white box area. But I don't think it particularly impact us.

Your next question comes the line of Daniel Amir of Lazard.

Just shifting gears here towards the networking market, I mean, this was down a little more than you expected this quarter and it should be up next quarter. Can you just talk a little bit about the dynamics. What you're seeing there? What's the opportunity there in the networking market? Is it really just the PON market recovering for you and the NPU market growing? Or is there other things going on? Because you did mention some share gains as well.

Yes. So the softness we saw, as we mentioned, was on the PON side. But going forward, the opportunity we see is both with the continued growth of the NPUs, as well as some of the design wins on the carrier and telco side of things, right? We've talked about a focus on that business for a while now. Those design wins, unfortunately, take longer to get to market. But when those do get to market, we expect some nice growth from those.

And in terms of the NPU market, I mean, should we expect that business to continue to grow, the double-digit quarter growth rate that we've seen here in recent quarters?

Again, I don't think we've given an actual -- exact growth rate for Q3. But at least for Q2, it grew double digits, and we do expect that to grow in Q3.

Okay, great. And then the final question, just on the storage market. Can you just give us an idea how you view the hybrid market? You did say that should be a growth over the next coming years. I mean, so do you think the focus right now should just be more on just the SSDs rather than hybrids? And kind of in terms of Marvell, are you indifferent if it's a hybrid or an SSD?

Yes, it's every -- all of the above. So today, in the hybrid space, let's talk about our expectations. My expectations is, down the road, the hybrid drives will be a significant portion of the market for hard drive. Now I say down the road, what does it takes for the hybrid drives to be a significant portion of the hard drive market, that requires cost reductions, and I'm talking about significant cost reduction. Today, hybrid markets consists mainly of 2 different devices, one device for the hard drive and one device for the SSDs. Mindful that the SSD is a small SSD. So whether it's a device, there is packaged as a single hybrid drive or 2 devices in the laptop, they are just -- they are basically 2 devices. Now as a result, the cost structure of hybrid solution today is not that attractive. So what's the solution? Well, the solution is to integrate the SSD technology into the hard drive SoC. By doing so, we could drastically lower the cost of a hybrid drive, and this is exactly what we have been doing over the last few years. And recently, we just sample our first generation hybrid single chip SoC into our customers. And we believe that this is the trend in the -- that's, again, to play in next year or 2, where the truly high-volume hybrid drives will be based on single chip SoC controller.

Your next question comes from the line of Mr. Craig Ellis of B. Riley.

Switching gears, just to the balance sheet, Brad, with the increase in inventory in the quarter of 25%, where would you expect inventory dollars or days to be exiting the current quarter?

The days will probably drop a bit when we get to the end of this quarter. As we talked about, it's really building forward to meet the demand for Q3. And with Q4 being seasonally down, I would expect those inventory levels to be down a bit.

Okay. And then on the cash generation and cash used side, the buyback was more equilibrated with the cash generation in the quarter. Is that how we should think about the buyback going forward? Or what kind of parameters can you give us in terms of company's excess cash return?

Yes, to your point, Craig, we bought back more than our free cash flow this quarter, again. And we'll continue to be in the neighborhood of our free cash flow, but a little more opportunistic than we've probably been in the past to watch what the stock does. But it's -- we remain very committed to the plan.

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

I wanted to ask about your sense of the opportunity that you see ahead of you in the TD-LTE market. And part of this question, I'm thinking about what happened TD-SCDMA, where initially you kind of focused on the high end of the market. The market, I think, came in at a little bit lower end, and I'm not sure you were that well positioned for that. So one, what's the opportunity in TD-LTE; and then, two, where in the market do you think you will be focusing?

TD-LTE is going to be, in the long run, is going to be the majority of the market in the world. And the reason to that is because a lot of frequency bands, new frequency bands, they are being allocated for LTE. They are inherently TDD in nature. So for example, the 2.3 gigahertz bands across the world, whether it's in the Asia or in North America, they are TDD based. So 2.5 gigahertz is the same thing. Not to mention that there are new bands that will be freed up across the world in the 700 megahertz bands, and those are TDD bands. Now at the same time, while we're excited that the TDD LTEs will be the driver of LTE, it does not mean that our LTE solutions are locked into TDD LTE only. As we mentioned many times in the past, our TDD LTE is a actually a superset of FDD as well, FDD LTE. So the solution is exactly the same, whether you are from us, whether it's FDD or TDD. They are the same device and the incremental cost from -- to address TDD is insignificant compared to the overall solutions. So this is different than in the case, in the past, where we're addressing the TD-SCDMA. That solution is completely a different solution than the W-CDMA. So we are not locked into just a certain set of customer base as we move to the LTE space. The LTE solution is -- our LTE is -- device is to address every customer in the world.

Do you have a follow-up Glen?

And just to be clear though, that every customer at every sort of price point, is that the way to think about it?

Well, when you look at LTE -- okay, let's address that -- I think you're alluding to this questions about low-cost TD-SCDMA in the past. TD-SCDMA -- the ultra low-cost TD-SCDMAs were addressing very low end market. Those are the markets where the throughput requirement is only a few megabits per second. When we moved to LTE, there's no such thing as a few megabit per second markets. The minimum requirements of a TDD LTE today is 100 megabits per second. And the market is going to disappear very rapidly. The vast majority of the market is going to move to second-generation LTE. We mentioned about 50% higher throughput of 150 megabits per second throughput. Once you've moved to that segment of the market, the requirements for -- the memory bandwidth is different. The requirements for the screen, the high-resolution screen, is different. So we -- at least, at the very least, for the next several years -- there is no such thing as ultra low-cost $50 LTE phones. At most -- all phones that will be introduced in the next several years will be at least, will start with the 100 if not more. Now not saying that we will not address the lower cost markets further down the road, but their timing is not next year. That's more like 2, 3 years down the road.

Just one other question, then, which is you were about to qualify at AT&T for LTE. And obviously, are already qualified in China. So as we see the Chinese handset OEMs look to move their product into the U.S. market, can you talk about the opportunity that, that presents you, given that we've seen now an intersection of your qualification both at those OEMs, but also on the U.S. networks?

I think the only thing I can, I think, that say is this, the U.S. market tends to be a heavily subsidized market. It tends to be driven with higher performance, higher end market. So the move to -- our move to LTE is very, very suitable for driving the North America market. So we are, of course, excited, because now it allows our customers in China to be able to address the North America market using our solutions as we move to the LTE side. So anything on the LTE is going to be positive for us. It was only just 3G, I think, and a lot of North America carriers are not so interested in delivering the market. When they advertise the market they say, faster is better. 4G is better than 3G. So it means that they cannot sell a 3G solution in North America any longer.

Can I ask one quick one?

Sure, go ahead.

Briefly? On connectivity with respect to phones. It would seem that there's sort of some market share movement in handset connectivity today, we see that between, for example, Broadcom and Qualcomm. I wonder if that represents an opportunity for Marvell to improve its position in connectivity for handset, particularly given the opportunity you see in baseband and AP?

Sure. If you look at, in the past, the LTE -- I'm sorry, not LTE, the modem side and the connectivity sides are independent. So somebody may have used a Broadcom connectivity together with the modem side, a different modem side from, say, Qualcomm, somebody else's. But moving forward, a lot of this solution that we are driving is a platform solution. So in the platform solution, we provide our modem and a Wi-Fi as -- it's almost like a turnkey solution, but it's fully baked what we deliver to the customer. So as a result, the opportunity for us to capture the connectivity as we increase the modem business, it's real. Today, the attachment is 100% and we don't expect the attachment to change moving forward. It's going to be continued to be 100%, and further down the road, it could be -- maybe even integrated as a single device.

So Glen, one thing is we haven't -- we rarely come across any customer on the handset side that wants to use somebody else's connectivity, very rare. They pretty much ask for our connectivity with our solution. So the trend that's...

So there's really one neck to choke. So if there's any questions about what's wrong with the connectivity or with their modem, with the throughput, it's only just us to choke.

Your next question comes from the line of Blayne Curtis of Barclays.

Sehat, I was wondering if you could walk through your goal is 10% of the W-CDMA market. You mentioned you're shipping a couple of Tier 1s. I guess, the Samsung device is more of a data device. So I was wondering if you could share a little bit more color on your traction with Tier 1s. And then when you look at the trajectory of 3G and then moving to LTE, if you could just talk about the timing of LTE. I mean you talked about getting some U.S. carrier approvals, but to ship on a global platform, you obviously need all the carrier approvals. So if you could just give a little more color on your traction with the other global carriers, that would be helpful as well.

Yes, we are still committed to the, obviously, we're advising to get a 10% at the end of the year. So as we said many times in the past, this is not an easy number to achieve, but on the other hand, we also believe that with our unified solutions now, with the fact that we are the only second supplier in the market that have a 4G platform that is compatible with our 3G platform. We're seeing a lot of our customers are starting, like several quarters ago, designing our 3G solutions with -- anticipating that when our 4G solution is ready at the time, they could easily move to the 4G solutions to upgrade the device to the 4G. So we haven't seen any changes in the behavior in the customer. They continue to be -- wanting us to be successful in this market because they know that, in the long run, there's going to be a very few customers -- I mean, suppliers in this market they could deliver the technology that could both support -- that support the LTE. And we are working hard on -- okay, you are talking about just the data? It's not just for data. It's data and voice. So all our customers -- not a single customer just building, let's say, our device for data. What you see is -- if you see a data only, that's only just for initial ramp. Every single customers are driving voice and data. The same thing is plain with LTE. If you see the first -- some of the customers driving the data-only first in the market, that doesn't mean that's the only thing they're working on. They're all working on voice as well. In fact, in LTE, there is a need for voice over LTE, which means voice -- even more advance, more advance voice over the data networks. So we'll see that data market -- that will continue to play out in that direction.

Do you have a follow-up Blayne?

Okay. And then maybe just shifting to hard drive, maybe I don't think I understood the explanation exactly, the uptick in hard drives in the July quarter versus the flat TAM, I think you're just saying that you had an extra month and that was up. Can you maybe just, one, just clarify that? And then on the enterprise side, a huge uptick there, you've been gaining on that platform. Any color as to what stage you are as far as that ramp?

Yes, so you're right in the characterization of the growth in Q2 was back-end loaded, driven by storage. So the upside that you don't see in Q3 is really in the July month for us, which falls into our Q2. In terms of the enterprise growth at that customer, it did grow more than 50%. It's still relatively small though. So as we talked about before, the timing to transition that can be tough to figure out exactly. We expect it to be fairly steady though, over the rest of the year.

So Blayne, the other thing is you can listen to the HDD OEMs, they're increasing their TAM by a few million units for Q3, not a whole lot, right? That was coming off of a flat TAM in Q2, and we grew 8%. So I think if you net that all out, we're still growing -- gaining share, but it probably all evens out.

The next question comes the line of Stephen Chin of UBS.

First is on your SSD business. Sehat, you mentioned positive comments on SATA and also PCI Express interface. I was wondering if you can help give us some of your thoughts on the SaaS interface for enterprise SSDs? And also for the PCI Express interface SSDs or are these the NAND-cash cards that you guys have had in the market for a while now?

The vast majority -- the volume part of the market for SSD initially started from the SATA side. And then more recently, we have moved that market -- we also move into the PCIe base for the client type of the market. So in the PC -- client type of market, the market will continue to be dominated by SATA and PCIe base -- low-cost PCIe base. So if you talk about SaaS, the SaaS market is extremely -- is an extremely small market. So we don't talk much about it. We do have solution for the SaaS market, but we don't talk much about it, because there's really no point to talk about it when the volume is quite insignificant. And what was the other question?

Yes. How -- I guess how your PCI express NAND solution is performing.

It's doing very -- extremely well. We're probably the only one who have this client base PCI Express solution in the market today.

And we have about at least a 9-month lead on the PCI Express SSDs...

For the client market.

For the client market.

And my follow-up question is on your operating expenses. So I understand you guys are probably investing or, yes investing in the 4G LTE bands and just wondering from an OpEx perspective, should we expect spending there to continue ticking up on the R&D side? Or do you think the current spending levels is enough to meet your deliverables over the next year?

Yes. So we've talked about the focus on keeping OpEx at current levels and working very hard to look at the prioritization of our spend and making sure we're investing in the right projects. So we're comfortable we can keep it at these levels for the rest of the year.

Operator we'll take one last question.

Next question comes from the line of John Pitzer, Credit Suisse. John W. Pitzer - Crédit Suisse AG, Research Division: Brad, just going back to the mobile side of the business, so that I understand the gross margin dynamics. I appreciate the fact that 3G is what's ramping in volume and that's probably driving the incremental gross margins, which are lower than, I think, some of us would've liked. But I'm kind of curious, gross margins within the 3G business, have they been relatively stable, so that really is just a mix issue? When does LTE become meaningful enough volume that it starts to offset some of the gross margin headwinds you see as 3G ramps? And I'm assuming LTE is probably losing money right now. So any volume is fairly leverage-able event, is that a fair assumption?

Yes. So your question about 3G. The pricing has been fairly consistent or the margin profiles have been fairly consistent there, obviously, as Sehat alluded to, lower than some of our other businesses. In terms of when the LTE crossover will be, it's probably sometime next year, I would guess, where you'll start to see the impact of LTE start to counterbalance some of the 3G work. What was the last part of the question? John W. Pitzer - Crédit Suisse AG, Research Division: LTE, assuming it's mostly expense and not lot of revenue right now, so any revenue becomes fairly leverage-able, is that a fair assumption?

Yes.

Yes, definitely. In fact, if it's not most or all of our R&D that we are doing for the basic R&Ds in the businesses is LTE, not just the current LTE, but next-generation LTE. John W. Pitzer - Crédit Suisse AG, Research Division: Okay. Then Sehat as my follow-up, just going back to the SSD controller market. I'm kind of curious as to what you think the ultimate split will be between merchant and captive. And the reason why I asked is, as I talked to the raw NAND suppliers, they continue to make the argument that building the NAND gives you a leg up on the controller, especially as you move from planar to 3D and, clearly, a lot of those NAND providers are trying to move up the value chain by doing more on a solutions-level basis. So help me understand the ultimate break you think between merchant and captive, and as the market moves to 3D, does that change the dynamic at all?

Yes, okay. We have been consistent in saying this, okay. If you look at an SSD solution, 90 -- let's call it 90%, if not more than 90% of the cost to build an SSD, is the NAND, not the controller. So what's the reason -- question is, what's the reason any NAND controller companies want to build the control themselves? Well, if they think they can build a better controller, they can make their NAND to work better. Yes, of course, they will build it themselves. However, we have proven that, as a company, we know how to build SSD controller better than the guys internally could build. Not just building things to work, how do we make -- how do we -- how could we make the controller to make the error rate, the lifecycles of the flash to be longer, to double, triple, quadruple the lifecycle of the flash chips? So these are the reasons why those companies, okay, the people they traditionally use to build their controller have realized that our error correction technology that we have developed over the years, specifically in the LDPC technology that we have invested over the last 13, 14 years in hard drives, are now starting to pay off. The NAND chips, the lifecycle is dropping like rocks, okay. As the NAND chips suppliers move from 40-nanometer to 30-nanometer to 20-nanometers and to 1x-nanometer they're finding out that the lifecycles are crashing. And they realized that -- they're starting to realize that only the best controller could solve this problem. And since 90 plus -- 90 or 90 plus percent of the cost to build an SSD is the flash chip. But if they could move to the smaller geometries, chips on the flash side, they get to win, even if they have to pay somebody like us to deliver the controller. So I don't think this is going to change anytime soon.

John, the other thing is you just look at the history of the hard drive industry itself, the hard drive OEMs initially had their own chip development as well -- controller development internally as well. And as we add more and more features and functions and patents and IP, they started to outsource this. You're seeing that right now on the SSD front, we expect that to continue. It's not a [indiscernible] control matter. Tony, I think we're done?

Yes, sir.

All right. Thank you, everyone, for listening to our second quarter call and we'll talk to you soon.

Thank you.

Ladies and gentlemen, this concludes today's conference. Thank you so much for your participation. You may now disconnect, and have a great day.