Good day, ladies and gentlemen, and welcome to the Third Quarter 2011 Marvell Technology Group Ltd. Earnings Conference Call. My name is Regina, and I will be your conference operator for today. [Operator Instructions] Today's event 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 go ahead Mr. Nagesh.

Thank you, Regina, and good afternoon, everyone. Welcome to Marvell Technology Group's Third Quarter Fiscal 2012 Earnings Call. I'm Sukhi Nagesh, Vice President of Investor Relations, and with me on the call today are Sehat Sutardja, Marvell's CEO; and Clyde Hosein, our 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 under the Investor Relations section at marvell.com. 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 risk and uncertainties that could cause our results to differ materially from management's current expectations. The risks and uncertainties include our expectations about sales of new and existing products, including statements about our TD, WCDMA, PON, HDD, and the SSD products, statements about general trends in the end markets we serve, the impact of the flooding in Thailand, statements regarding our financial predictions for the fourth quarter of fiscal 2012, and our expectations about long-term growth. To fully understand the risk and uncertainties that may cause results to differ from our expectations and outlook, please refer to today's earnings release, our latest quarterly report on Form 10-Q and subsequent SEC filings for a detailed description of our business and associated risk. 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 excludes stock-based compensation expense, as well as charges related to acquisitions, restructuring, gains and other charges that are driven primarily by discreet 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 fiscal third quarter earnings 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 third quarter revenues of approximately $950 million reflecting a 6% sequential increase from the prior quarter driven by our mobile and wireless end market. We delivered the following non-GAAP results: gross margin of 56.8%; operating margin of 25%; and earnings per share of $0.40. We generated free cash flow of approximately $239 million, equivalent to a 25% free cash flow margin. In addition and consistent with our plan to return value to our shareholders, we continue to repurchase our shares. In Q3, we repurchased about 50 million shares for a total of approximately $215 million. Now let me summarize our results across our end markets. First, in our mobile and wireless end market, Q3 revenues increased approximately 24% sequentially and represented about 31% of our overall revenues. Growth in this end market was better than our initial expectations with the sequential increase driven by strong growth from our new products that goes into TD-SCDMA smartphones and seasonal growth from our wireless connectivity solutions. Now expanding on TD, this business performed strongly in the third quarter and more than doubled from the previous quarter resulting in TD smartphone market share of over 70%. Mass deployment of TD smartphones started in September and many of these devices are now selling throughout the retail channel. Furthermore, China Mobile is increasing its investment in TD-SCDMA and TD-LTE, and Marvell has a strong product lineup in both technologies to take advantage of this expected growth. While we expect increased competition from followers in TD smartphones, we expect to maintain our leadership position and grow revenues next year. Overall, we have made significant strides in developing and helping commercialize the TD standard in China, which we believe provides an ideal platform for local smartphones. This initial platform…

Thank you, Sehat, and good afternoon, everyone. As Sehat mentioned, Q3 revenues for fiscal 2012 came in at approximately $950 million, representing a 6% sequential increase from Q2 fiscal 2012 and down about 1% from the same period a year ago. We believe our Q3 revenue was impacted by roughly $20 million on account of the flooding in Thailand. Our non-GAAP gross margin for the third quarter was approximately 56.8%, within our earlier projected range, but slightly below the midpoint of our guidance mainly due to continued higher commodity prices. Our overall operating expense for the third quarter on a non-GAAP basis were approximately $300 million, up 5% from the prior quarter and up about 15% from the same period a year ago with the majority of the sequential increase coming from higher spending on R&D for several new products and programs such as TD, SSD, WCDMA and PON. R&D expenses for the quarter were approximately $241 million, up about 6% from the last quarter and up about 22% from the same period a year ago. SG&A expenses for the quarter were approximately $58 million, a 2% sequential increase from the prior quarter and a 5% sequential decline from the same period -- a 5% decline from the same period a year ago. This resulted in non-GAAP operating margin of 25% for the quarter. Net interest expense and other income was about $8 million, while non-GAAP tax expense was $4 million. Our non-GAAP net income for the fiscal third quarter was approximately $244 million or $0.40 per diluted share in line with earlier projections and a $0.02 increase from the prior quarter. The shares used to compute diluted non-GAAP EPS during the third quarter were approximately 615 million, a decrease of roughly 10 million shares from the prior quarter, primarily…

[Operator Instructions] Your first question today comes from the line of Sanjay Devgan with Morgan Stanley.

I guess, first question for Sehat. You know, Sehat, you talked about the impact from the Thailand floods, and how it's driving a move to 1 SSDs into advanced HDD technologies like the 500 gigabit platter drive. In your opinion, do you think this is a temporary change, or do you think this is a structural change? And how should we view these -- just the technology landscape going forward based off of the flooding that has happened in Thailand?

I think there's 2 types of changes. The SSD -- the move to SSDs, especially on the high-end products such as Ultrabooks, there is -- will always -- will naturally continue because people can afford to pay for the higher-priced storage solutions. The changes in the adoptions -- the rapid adoptions -- the accelerations of the 500 gigabyte per platter, that's once the transition happens, there is no way back to go to the previous solutions, because in history, it's always -- that's the way it has worked. Every time a new technology matures, the new technology will result in the lowest-cost solution. So traditionally, the 500 gigabytes -- if we don't have the floods in Thailand, one would expect that the 500 gigabyte per platter will mature at a slower rate? Because people tends to be more conservative in terms of deploying new technology. In this case, we have no choice. We've been -- the whole industry -- the drive industry doesn't have a choice, so they have to -- they will have to accelerate the 500 gigabytes per platter. We actually have introduced this technology for more than 1.5 years and maybe even 2 years. So this technology from the semiconductor side is very mature. Our technology is super mature in this area, and we are the clear leader in this area. And so it makes a lot of sense. When you're faced with limited supplies of heads and suspensions, for example, a lot of these components are built in Thailand, when you have these problems, you do want to maximize the number of drives that you can build. So if you can build drives with only one head per drive, so instead of building 500 gigabyte drives, you build 250 gigabyte drives. You can produce twice as…

I guess second question pertains to your mobile and wireless business, obviously China Mobile has been a great growth driver for you in the back half of this year. However, Qualcomm yesterday at their Analyst Day, they unveiled their new 28-nanometer products, and they've talked about support for TD within all their products coming out in the second half of next year. I was wondering, as a competitive landscape in TD specifically starts to increase, can you talk to the dynamics there? And how do you feel about your competitive position and with the emergence of new threats coming into this space?

Yes, based on what we've been saying all this time, is that TD is a real business opportunity. Now the dust has settled, then it's clear that the market for TD is huge, especially China Mobile with more than 600 million subscribers will eventually move the vast majority of their subscribers to TD. It is obvious that more of the existing suppliers will build TD, and it is expected. And we've been saying it all along that everybody will realize that they have to invest in this. So we're happy actually now people now are investing, so just proven -- it just proves that okay, we're on the right track. Now the good thing is, we're ahead. We've been investing. Our device is qualified with all of the base stations. Not just 1 base station, but I mean, every base station in China. So we have at least 1 year, if not 2 years lead in this area. So we're not staying still. We will introduce our new-generation products, and we're also focusing in lower-cost solutions, so not just very high-end products. They happen to have -- very expensive products, they happen to have a TD built-in that's not even proven yet.

[Operator Instructions] And your next question, gentlemen, comes from the line of Harlan Sur with JPMorgan. Harlan Sur - JP Morgan Chase & Co, Research Division: I wanted to focus on the overall market growth in TD. It really looks like the momentum here is starting to pick up given the subsidies on smartphones, the continued strong government support for the network, and also the fact that I think China Mobile's GSM network is starting to get pretty highly congested. So on the smartphone side where Marvell dominates, what's your view, Sehat, on the growth outlook for this segment next year?

So let me address a little bit in and maybe Clyde, you may want to chip in, in this one. So as I said, the 600 million subscribers -- okay, and then maybe a quarter 2 ago, I mentioned about several hundred thousands, 200-plus thousands of base stations on TD base stations being deployed or have been deployed. So that's more than any major carrier in the U.S. have base stations in the U.S., the single base in carrier in the U.S., but they have fewer the base station than that. So we're talking about the like the greenfield, new build-out. So I do believe that this is a huge opportunity. And you've mentioned earlier that it's subsidy. Okay, so we talk about subsidy, that's true about 1 year, 2 years ago. Without subsidy, TD will never take off. Why? Because without subsidy, 2 years ago, the handset was selling for USD $500, USD $600. Clearly, without subsidy, it will never have taken off. But with these new solutions, this platform solution that we provided, low-cost, these phones can be built without subsidy for $100, even less than $100. So clearly, now we're in the deep end phase of the deployment of TD smartphone in China. $100 is not a lot of money to many people in China. Especially, we're talking about smartphones. We're not talking about feature phones here. With many times people buy feature phones for $50, for $60, maybe with, with low-quality LCDs, with terrible softwares, with no storage capacities, with very limited functionalities, no web browsing capability. So this is clearly a huge step up for most people that wants to have to be connected. And I know -- oh you want to add...

No, I think Sehat, you said it well. So it's going to grow significantly. I think the estimates on the smartphone space, which where we play, is 10 million to 12 million this year growing to 20 million or 30 million next year. So that portends to a steady otherwise, doubling of growth in the space. We expect more competition, as Sehat pointed out earlier, but we have the lead, we address in areas where other people could come after us. So we still expect to grow in spite of competition. Harlan Sur - JP Morgan Chase & Co, Research Division: And then my follow-up question, on the pull in of your 500 gig per platter to your customers, I think the team was expecting to ramp-up more aggressively in the first half of next year. So are you seeing the 500 gig per platter ramp starting this quarter, or is that more likely kind of a Q1 ramp?

As we indicated earlier in our prepared remarks, it was going to ramp this quarter anyway, putting aside the flood. The rate and pace is increasing because of the component shortage that Sehat alluded earlier, as well as in his prepared remarks. So short answer to your question is starting this quarter, it's doing better because of the floods. And then we think the adoption rate is probably going to be even faster when people get the yields improved on this new technologies, they're going to continue to use it. So this is going to be very good for Marvell.

Your next question comes from the line of Srini Pajjuri with CLSA. Srini Pajjuri - Credit Agricole Securities (USA) Inc., Research Division: Clyde, on your guidance for the storage business, you said you expect it to be down about 20% to 30% and that's actually significantly better than what some of your customers are talking about. I'm just trying to understand where the difference is coming from, if it is share gains or if it was something else?

So there's a couple of things. You got to look at the timing of the recovery, obviously the timing of the flood affected us in all of Q3 versus some other people because we have October quarter. Similarly, our quarter ended in January, the significance of that is we expect to see some production, a restoration starting in the December month. So we'll get some benefit off that, and that's going to mitigate that. There's also -- one of the good things about Marvell is we serve many customers in many different countries, and so they also shift in capacities to other customers and to other countries. So you won't see it singularly as you might see it in any particular customer. Srini Pajjuri - Credit Agricole Securities (USA) Inc., Research Division: Okay, great. And then just as a follow-up. Sehat, you talked about progress in WCDMA at other customers? Just trying to get some more color. Could you talk about how many customers you have or how many design wins, and how should we think about the ramp over the next few quarters?

Clyde, why don't you answer this?

It started off -- as we said before, it started off in our Q3, it's modest. It's a few customers. We expect the revenues from that to grow next year both in terms of revenues from the customers that already started, as well as some new customers, and we'll provide more updates as we go through.

Your next question comes from the line of Uche Orji with UBS.

My first question is about networking, Sehat. Your flat guidance, does that imply that the inventory correction is complete? And if so, can you just give us some parameters to reach that -- is allowing you to reach that conclusion?

If you notice in our prepared remarks, okay, I did mention that okay, we do expect to have continue increase in the PON deployment. So in certain areas, we'll continue to have increase in revenue. But in other areas, there we'll still see some downside on -- maybe through the inventory maybe -- correction may still be lingering a little bit. So that's the reason why we are forecasting flat as a whole.

I'm going to just follow up on this, and the reason is when I look at your commentary last quarter, we were very confident this business will grow, and obviously, it didn't. Can you just give us a little bit more insight in terms of things like bookings. Is there enough bookings coverage to feel comfortable with on the PON side to offset the areas where there is continued debenture correction? And outside of PON, what else is growing? And in terms of what will drive growth going forward from here? Can you just help us understand your confidence of the growth drivers for your networking business, that's it?

Sure. Last quarter, we had steady growth from new products. But if you look at the competitive space in networking, whether it's FPG or other products that's been fit to the space, you saw some fairly significant reductions, mid-single-digit type decline. You follow that up into the December quarter, you're looking at double-digit, 10%, 15% decline from many of the products. So Marvell was down slightly. I think down 1% in October quarter and flattish next quarter. The end market for our products resemble what you see in industry. However, some of the new products are beginning to kick in to improve on that. And so PON, 10 gig, 5, which is new for us in the last few quarters. And some switching design wins is also beginning to kick in. So that's the 3 areas that tend to buffer it. On the line customer tightening up of the inventory. So the backlog supports the flattish forecast. And, yes, it's better than what most people are seeing, but which if you look at the backlog I'm fairly confident that at this point in time that, that would support it.

Your next question comes from the line of Craig Ellis with Caris and Company. Craig A. Ellis - Caris & Company, Inc., Research Division: Guys, first time gross margin, Clyde, I think at the 55% level, sequentially down about 180 basis points or so. Can you just help us understand how much of that would be mix, and how much of that would be some of the input cost headwinds that you identified?

From which period? Craig A. Ellis - Caris & Company, Inc., Research Division: Fourth.

From a year ago, we have about 4 points of decline, which has led people -- a lot of investors do. But 1/2 of that is input cost, primarily gold. And then the remaining is about equally divided between new geometries 40 and 28 which we have put into production right now. And the third piece of it is foundry prices. Let me address what we're doing about that. On gold, we are moving to copper. We should ship our first copper products this quarter. We were arguably late versus a lot of people. Our customers were a lot more reluctant because of some the quality issues in copper. We feel we got that addressed. So we should start our copper transition now. A year from now, if you jump ahead, Craig, we should expect more than half of our new products coming out with copper. So it's a very fairly aggressive ramp-up to copper, and that's going to address the gold even at turn prices. And the foundry issue, we have developed a multiple source foundry strategy that has a number of advantages. One of them is to address the lower pace of cost reductions that we are currently experiencing that would help accelerate that and make us more competitive. But it also address a number of risk profiles for our customers especially with the 2 big disasters this year, customers are beginning to resonate with that. So the foundry thing, we think, is going to start getting addressed in a quarter or two as we move it over. These things are obviously slow to move. And the third element, which is about a point from a year ago is our new product. And that's simply accounting. Once the production product is ready, you get to start amortizing the assets. But as you know, the product tend to be back-end loaded as customers take about 6 to 12 months to bring that. So you will get the benefit of that probably in 6, 9 months, you'll start seeing that benefit. So all of these areas, we feel are very addressable. And the last thing I'll say is, even with our current guidance, in our peer group, we still have the best gross margin in the industry. So we have a lot of room to transition these issues and we still have the best margin. And with the actions I outlined, I think we will improve that as time goes on.

Maybe I'll just add a little bit on top. Okay, when we also do multiple sources, okay, they will increase our mass cost as well, as a result. So they will also have some impact on the gross margin in the short term, which -- the benefit comes later when those products go into production. Craig A. Ellis - Caris & Company, Inc., Research Division: Okay, that's helpful. And then, Sehat, you're getting close to the end of the current authorization off of what was yet another real good free cash flow quarter for the company. So, Clyde, can you just help us understand how you're looking at your excess cash alternatives from here?

I think we have shown our hand, we believe that our stock is undervalued given our plans and strategies. Obviously, you know a lot of about that than people in the investor base. We still believe that. We bought back, I think, 50 million last quarter. We still have a couple of hundred million, and we had to actively participate in the market. We have another board meeting in December. We'll address these same issues. We have a very good balance sheet. We need something like that in this economy, but we're generating a lot of cash as we did last quarter. So our intention is to return that to shareholders in some which way shape or form. And our current preferred methodology is share buybacks, we intend to continue that.

Your next question comes from the line of John Pitzer with Credit Suisse. John Pitzer - Crédit Suisse AG, Research Division: Clyde, you talked about the recovery in the hard drive space one quarter out. I know there's a lot of variables. I think you said industry recovery of about 10% to 15% in the March timeframe. Just out of curiosity, is that coming from production coming back online in Thailand in your opinion? Or excess capacity outside of Thailand, which will be better utilized in the March quarter? What's kind of the variables that go into that statement?

10% to 15% was more in our current quarter. And in the current quarter, it's going to probably outside of Thailand. And then in time, they'll restore it back. So as you get out to other quarters, it'll be more in Thailand. But in the near-term, it's obviously outside.

To give you an idea of their capacity in Thailand. There are people who have capacity in Thailand. There are people who have assembly capacities in Malaysia. People have capacities in Guangzhou and Guangdong, and they have capacities in Wushi [ph] in China and in Japan as well. They used to be quite a significant capacity in the Philippines. So okay some of these things could also be brought up too. They might not be there yet anymore or maybe a small percentage, but it could easily be brought up as they are -- suppose to the buildings are still there. John Pitzer - Crédit Suisse AG, Research Division: And then, Clyde, how do we think about the inventory in the hard drive controller space? I guess, I'm curious as to how much of your inventory do you think was damaged by the Thailand flood? How do you think customers are trying to manage inventory of controllers through this? Are you likely to get hit worse as they figure out where to find capacity and then see a snap back as you start to replenish inventory, or how do I think about that?

Specifically for hard drive controller, our inventory was not affected. And we've got all of our inventory secured. Our operations team did a fantastic job. And now we have full support of our customers. So it's not an impact directly to Marvell. In terms of broadly speaking in the channel, John, as you know, the PC space is the primary space. I think they'll use up most of their existing capacity to support that space and true-up a lot of their existing inventories. So from a broad channel point of view, I think you'll see a fair amount of drain especially for the Christmas season, which is one of the bigger buying seasons for it. And then in Q1, Q2, you'll see a replenishment. So obviously Marvell's results or Marvell's forecast reflects the near-term effects of all of that, but there's a fair amount of upside next year as people restore the broad channel inventory.

Your next question comes from the line of Mark Lipacis with Jefferies & Company. Mark Lipacis - Jefferies & Company, Inc., Research Division: The question is, Sehat, I want to make sure I understood on the TD side, your comments about the low-cost solution. Could you just clarify for me, does this mean -- is this a new set of products that you're introducing to the market that's additive to the smartphone products? Or is this what drove the growth so far?

As mentioned earlier, about 2 years ago, when we talk about smartphones for TD, we're talking about $500, $600 phones. So now we're talking with a single-chip smartphone solution that we provide. So many of our customers, I mean, basically all the -- more than a dozen of customers, shipping products, shipping smartphones that sells in the retail approximately $100, give and take. So that's what we call low-cost, unsubsidized. With subsidy, if the carrier want to have subsidies, they can easily make it for free, if they wanted to. But there's no need actually because the price will naturally go down from $100 to $90 to $80. As people will have improved the manufacturing cost of the overall phones. After all, the silicon cost is not the majority of the cost of building the $100 phones. So there's a huge -- again, when we call it low-cost, we truly mean because it is low-cost. Anything that sells for $100 or less, it is low-cost. Mark Lipacis - Jefferies & Company, Inc., Research Division: Okay. And this is what drove your 20% growth this quarter?

What do we say? The -- yes. Mark Lipacis - Jefferies & Company, Inc., Research Division: And then a follow-up question on the solid-state drive of market. I'm trying to understand the trade-off. Say, hypothetically speaking, assuming all market shares stays the same. Let's say Ultrabooks take over in solid-state drives cannibalize 10% of the notebook market, can you help us understand how does Marvell -- can you give us a framework if you're thinking about how Marvell benefits or loses in that scenario?

On SSD, our business model, and we have delivered this although investors will see more of that incoming. We have more share in SSD than we have in HDDs. We have about 62% share in HDDs today, we have more share in SSD. That price point today is higher, although that's going to compress as time goes on, but it's higher. And the gross margins are better. So as this transition happens, pick 10%, pick what you want, we expect this to be accretive to Marvell.

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

Maybe just a follow-up on the last question. We saw one of your competitors LSI buy Sandforce this quarter and I wonder if you can talk about whether or not you feel Marvell needs to acquire similar types of correction technology, or whether or not you feel you can develop that technology in-house?

All right. So I guess, maybe I need to -- let me answer it this way, we've been investing in SSDs, as we say it, for many quarters already, actually quietly for what, maybe 6 years or more, way before anybody realized that there is an opportunity to utilize flash memory to build solid-state disk. This was a time when nobody knew what good we could do or how to solve this problem, so we have amassed significant technologies to make it happen. And this is not surprising that we have the vast majority of the design wins of all of the merchant market SSD controllers in this business. So clearly, we have a comfort zone in competing against anybody who wants to go into this business. Whether it be new suppliers or somebody who just showed up a couple of years ago. So the answer, we don't have to buy anything. We have our own. We have stuff that people don't know yet, what's coming. So we have a lead, a huge lead in this area. And in terms of the opportunity, again, I'd say, this is just really a function again, the availability of the flash that's going to be out there, that's going to be available in the market. We'll be -- we have nothing to worry about.

I also wanted to ask about where -- if you can just give us more specificity on where Marvell is with respect to TD-LTE, when you think that's going to ramp? How ready you feel for that transition? And whether or not you think you have that same lead in TD-LTE as you do in regular TD?

Sure. So TD-LTE is an important technology to address 4G, the next gen high data rates, especially as we move to beyond 100 megabits per second types of throughput capability. So TD-LTE is a natural migration for TD-SCDMA, just like the TD-SCDMA is the natural migration for EDGE. Again due to the vast majority of the usage model that goes into these smartphones or laptops or tablets, we'll be detail-oriented downloads applications. So we do expect TD-LTE will be deployed as soon as the cost structure of the base stations make sense, meaning in the next initially small money initially in the next year, but more rapidly 2 years from now, especially in the subsequent or next situation, you need to invest in the base station first before you have the volume of the handsets. So the handsets will come behind it. I don't know, okay if I have to guess, maybe 6 months, 9 months, 12 months behind the deployment of the base stations. So this is the way we look at it. And we already sample our TD-LTE in case you want to understand how we position on the silicon, on the handset side. So we should be -- again, the short answer is, we should be on time on the ramp on the TD-LTE on the handset side.

And we'll have it backward-compatible today.

Yes. And not to mention TD-LTE is only used for data downloading, for voice communication. Until or probably for the next 3 to 4 years, you will not see voice on LTE until LTE is widely -- until the base stations are completely built out. This is all mainly for data downloading. So backward compatibility is extremely important at least for the next 4 to 5 years.

Your final question comes from the line of Craig Berger with SBR. Craig Berger - FBR Capital Markets & Co., Research Division: I just wanted to get an update on how you see market share in hard disk drives. Are you still gaining at Seagate? Where are we there? Are you still gaining at Hitachi? Where are we there? And I'm asking more on the client-side.

Sure. The Hitachi, we've gained pretty much today on most of the mobile space. So that continues, but that's done. And then with other customers, I don't think we have any other changes than the last time we gave you, which they should be ramping up next year. Craig Berger - FBR Capital Markets & Co., Research Division: I see, next year. Okay. As a follow-up, you guys clearly have done well in solid-state drive controllers. I'm still trying to learn more about that market. And so I'm just surprised to hear that your market share is greater in solid-state drives given guys like OCZ, Indilinx, SanDisk, Toshiba, Samsung, Sandforce, do you supply the captive flash producers, and which -- are any of those your customers? Can you just give a little more on the competitive landscape?

We won't name names, but many of the names that you talked about are customer of Marvell. When we talk about SSD, we're talking about PC-type replacements were 6 gigabit speeds or faster, which is still fairly small. We acknowledge that. Since the revenues have doubled from the last year, still small, but we think many people think it's poised to growth. So those customers, many of those people you mentioned are customers of Marvell already.

Even for some people that used to build their own inside controller, many of them are realizing that our controllers are superior. So, okay, at the very least, on the high-end products, okay, they are now moving to using our technology. Craig Berger - FBR Capital Markets & Co., Research Division: One point of clarification on TD-LTE, are you guys doing field trials there yet?

I don't know how to answer that one. I think we just recently sampled. So we -- okay we tend to work very closely with the carrier. So I am sure there's some kind of trial that we work on with carriers. So I don't know how expansive it is. There will be more, of course, as time progresses.

Ladies and gentlemen, this concludes the question-and-answer portion of today's event. I would like to turn the call back over to management for some closing remarks.

Thank you, Regina. I'd like to thank everyone for their time today and your continued interest to Marvell. I look forward to speaking with you in the coming months. Thank you, and goodbye.

Ladies and gentlemen, this concludes today's presentation, and you may now disconnect. Have a great day.