邁威爾科技 (MRVL) 2010 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Q3 2010 Marvell Technology Group Ltd.

earnings conference call.

(Operator Instructions).

I would now like to turn the call over to our host for today, Mr.

Jeff Palmer, Vice President of Investor Relations.

Please proceed sir.

Great, thank you, Anita, and good afternoon, everyone.

Welcome to the Marvell Technology Group's physical third quarter 2010 earnings call.

I'm Jeff Palmer, Marvell's Vice President of Investor Relations, and with me on the call today is Dr.

Sehat Sutardja, -- Marvell's Chairman, President and CEO, and Mr.

Clyde Hosein, Marvell's CFO and interim COO.

All of us will 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 www.Marvell.com.

Additionally, this call is being recorded and will be available for replay from our corporate website.

Please be reminded that this call will include forward-looking statements that involve risks and uncertainties that could cause Marvell's results to differ materially from management's current expectations, including our expectations about product sales and general market trends, expectations regarding new products, statements regarding our financial projections for the fourth fiscal quarter of 2010, our expectations regarding the current economic environment and impacts to industry demand.

To fully understand the risks and uncertainties that may cause results to differ from our outlook, please refer to Marvell's latest annual report on Form 10-K and subsequent SEC filings for a detailed description of our business and associated risks.

Please be reminded that 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 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 Marvell's core operating performance.

Pursuant to Regulation G, Marvell has provided reconciliations of the non-GAAP financial measures to the most directly comparable GAAP measures in our third fiscal quarter 2010 earnings press release, which has been furnished to the SEC on Form 8-K and is available on Marvell's website in the Investor Relations section at Marvell.com.

I would now like to turn the call over to Dr.

Sutardja.

Sehat?

Thanks, Jeff.

Good afternoon, everyone.

Today we reported fiscal third quarter 2010 revenues of approximately $803 million, reflecting a 25% sequential increase and a slight increase over the same period a year ago.

The last year has been a tough one for everyone.

However, Marvell has emerged out of this in very good shape.

Our gross margin during the last quarter was more than 57%, over five points of improvement from a year ago, and amongst the highest in our history.

We also generated almost $200 million in free cash flow for the quarter.

These results reflect the fiscal discipline we have implemented in the Company, navigating through one of the toughest economic downturns of our time.

I am proud of our employees for these accomplishments.

We are very pleased with the revenue growth we experienced in the third quarter, as order momentum improved across all of our addressable end markets throughout the quarter.

The revenue growth reflects both an improvement in the overall end markets, as well as the ramp of our new products.

During the third quarter, the sales of new products was approximately $80 million, or about 10% of our total revenue for the quarter.

This was better than 100% improvement on a sequential basis, a testament to the customer acceptance of our new product introductions.

To remind investors, we are defining new products as complete new devices launched during the last six months.

Looking out into our fourth quarter, we anticipate this trend will continue as new products continue to represent just over 10% of our total revenue.

I would now like to review the performance achieved in our various addressable end markets.

The sale of products into the mobile and wireless end market was up over 40% sequentially and contributed approximately 20% of our total revenue.

About two-thirds of our sequential growth in our mobile and wireless business was due to the strength in cellular market, with the remaining due to continued strength in the embedded Wi-Fi market.

The third quarter was clearly a good -- a very good period for our mobile and wireless business, as we continued to experience positive returns on our long-term investments in this area.

Additionally, beyond our strong results, we have made several key announcements during the quarter.

As an example, during the third fiscal quarter, we announced our Armada family of high performance, low power, embedded application processors.

We have designed the Armada processor family to offer the best combination of price, performance, and features across all end device markets.

The Armada product line addresses a wide range of customer requirements from cost sensitive consumer products, smart books requiring PC level performance, high-end smartphones, and connected full HD consumer video devices.

All the products are based on Marvell's custom CPU cores, which leverage the broad ARM software ecosystem.

Early customer acceptance and interest in the Armada pipeline is very strong across multiple end markets, with more than 50 product design awards so far.

While we are in the early stages of customer engagements with Armada, we believe the early success will translate into meaningful revenue sometime over the next year and beyond.

Within our communication processor portfolio, several notable events occurred during the quarter, with both new and existing customers.

During the quarter, China Mobile, the world's largest mobile phone carrier, launched its OPhone smartphone platform, running the Android-based OMS operating system.

Marvell is pleased to have been selected as first among a small group of strategic silicon suppliers for the launch of multiple new OPhone smartphones from a variety of leading handset OEMs.

The scope of this announcement is very broad.

We have approximately half a dozen handset manufacturers currently building OPhone products based on Marvell technology.

Over the next several months, we anticipate additional Po1 handset OEMs with announced OPhone offerings as well, many based on Marvell silicon solutions.

But we have been working with China Mobile for more than two years and are very proud of the outcome of this project.

Additionally, as part of the OPhone effort, we have recently announced the PXA 920, the industry's first single-chip communication processor, which supports both the China 3G, TD-SCDMA, and EDGE cellular portable standards.

We believe the PXA 920 has set a new standard for performance and integration, enabling future OPhone and smartphones to be developed at close to feature phone price points.

We envision the PXA 920 as enabling a whole spectrum of new products.

Furthermore, we have been intimately involved in modem qualification process on the China Mobile network throughout China, as the OPhone platform will leverage our infinitely developed protocol (inaudible), a clear milestone and validations of the maturity of our cellular software.

We look forward to a long and successful relationship with China Mobile and all the OPhone partners around the world.

Also during the third quarter, our long-time customer, Research In Motion, announced its next-generation Bold 9700 3G smartphone.

This platform is based on our latest 3G UMPS communication processor.

Reviews and early market acceptance of the product have been very positive and was one of the contributing factors to the solid sequential growth we experienced in the mobile and wireless segment during the third quarter.

During the quarter, we again experienced over 100% growth in the demand of our 11n Wi-Fi devices in the printer and enterprise end markets.

Additionally, we experienced over 50% sequential growth for our new Wi-Fi Bluetooth combo devices.

Early feedback from our customer on our expanding portfolio of multimedia combo devices is very positive and we would expect to see positive revenue traction starting sometime next year.

Looking to the fourth quarter, we expect continued strength from the sale of applications and communication processors, offset by normal seasonal sales patterns associated with sale of our consumer-grade Wi-Fi products.

Taken together, we anticipate revenue from the mobile and wireless market to decline in a range of mid- to high-single digits, due to typical seasonal ordering patterns, while we anticipate new design wins will continue to accelerate.

Moving on to our efforts within the networking end market, sales increased by greater than 25% sequentially and we presented approximately 20% of total sales.

This is a reflection of both a rebound in the networking equipment market, as well as continued momentum associated with Marvell's specific share gains across multiple customers.

We experienced the strongest growth within the enterprise portion of our business, as sales of enterprise gigabit and 10 gigabit switches contributed the majority of the growth on a sequential basis.

As we progress into the fourth quarter, we believe Marvell's specific share gains in the networking area will continue.

We see our continued success in the enterprise area as being the primary driver of revenue growth in the networking area in the fourth quarter.

Furthermore, we do anticipate modest improvement in our client NIC and Fi business.

Taken together, we believe revenue from our networking business will accelerate in the fourth quarter in a range of mid- to high-single digits on a percentage basis.

Now, last week, the sale of products into the storage end market grew by nearly 20% on a sequential basis, and contributed to just over half of our total revenues.

We believe our share of the hard disk drive SOC market expanded to approximately 60% on a unit basis during the third calendar quarter.

We experienced very strong demand across all drive pipes, but especially within the 2.5-inch drive market.

Within the 2.5-inch drive market, we estimate that nearly 60% of all mobile drive shipped during the third calendar quarter leveraged Marvell SOC technology.

Within the enterprise trend market, we experienced solid improvement, sequentially both on a unit and a revenue basis, as previous ramps of enterprise SOC began to have a positive impact on our growth.

And lastly, within the 3.5-inch drive market, we believe our unit share was nearly 50%, as our customers continue to gain incremental share.

These are clear examples of how Marvell Technology helps our customers to be successful.

Looking forward, we anticipate the sale of products to our storage customers should improve sequentially by mid-single digits on a percentage basis during our fourth fiscal quarter.

This is in line with the projected growth rate of the overall hard disk drive industry.

In summary, we are encouraged by the return of growth rates more in keeping with our historical levels.

As has been the case throughout our history, we have always been a company focused on raising the technology bar in our target markets.

By doing so, we aim to lower the barriers and cost to entry to the next billion consumers eagerly awaiting to connect and join the modern knowledge base global economy.

We will continue our laser focus on delivering innovative products, which will enable our customers to succeed and gain share in this growing global markets.

Now I would like to turn the call over to Clyde to review our financial results for the next quarter and to provide our current outlook.

Clyde?

Thank you, Sehat.

And good afternoon, everyone.

As Sehat mentioned, fiscal Q3 revenues came in at approximately $803 million, representing a 25% sequential increase over fiscal Q2 2010, and an increase of about 2% from the same period a year ago.

As Sehat mentioned, our overall revenue performance was better than we had anticipated in our initial guidance, as well as we had updated on our guidance on October 26.

Our non-GAAP gross margin for the third quarter was 57.8%, an increase of 244 basis points from the second quarter, and up 546 basis points from the same period a year ago.

This was higher than the midpoint of our earlier projected range of 54% to 55.5%.

The sequential improvement in our gross margin is a reflection of improved product mix, additional cost reductions we implemented during the quarter, combined with ongoing improvements in operational execution.

Our overall operating expenses for the third quarter on a non-GAAP basis were $231 million, which was in line with our earlier projected range of $225 million to $235 million.

R&D expenses for the third -- for the quarter were $187 million, an increase of about 9% on a sequential basis, and a reduction of 8% from the same period a year ago.

On a sequential basis, R&D increased due to lower customer funded R&D projects, which are typically lumpy, and lower vacation usage during the quarter.

As compared to a year ago period, the 8% decline was due to lower head count, lower stock-comp, and related expenses.

SG&A expenses for the quarter were approximately $44 million, down nearly 22% sequentially, and a decrease of about 31% year on year.

SG&A expense during Q3 were lower than anticipated, primarily due to unforecasted insurance reimbursements and lower than forecasted legal expenses.

On a year-over-year basis, SG&A was down due to the combination of lower head count, reduced legal expenses, lower consulting expenses, and lower sales commissions.

This resulted in non-GAAP operated margin of approximately 29%, up over nine points from the approximately 20% operating margin reported in the prior quarter, and an improvement of better than 10 points from the same period a year ago.

Net interest expenses on other income was an expense of approximately $1.4 million, slightly higher than our prior expectations, primarily due to currency impacts on foreign denominated tax reserves.

On a non-GAAP basis, we realized a tax benefit of approximately $350,000 during the third quarter.

During the quarter, we had approximately $33 million in tax benefits, which are not included in our non-GAAP presentation.

These related expiration of statute of limitations on the tax contingency reserve and an adjustment of a prior year deferred tax asset.

We did not reflect the benefit of these tax items in our non-GAAP results, as we view these items as nonrecurring, primarily related to items we typically exclude from our non-GAAP results.

Our non-GAAP net income for the fiscal third quarter was approximately $232 million, or $0.35 per diluted share, an improvement of almost 100% versus our net income of $119 million, or $0.18 per share, we delivered in the prior quarter.

During the same period a year ago, we earned $145 million, or $0.23 per share.

The share used to compute diluted non-GAAP net income per share during the third quarter were approximately 664 million, up from 652 million shares in the prior quarter, and higher than the 633 million shares reported in the year-ago period.

Changes in diluted share count are primarily due to shares issued in the periods, along with variations in average and ended share trading prices in the reported periods, which impact the treasury method of computing diluted share counts.

Let me summarize our quarter results on a GAAP basis.

We generated a GAAP net profit of approximately $202 million, or $0.31 per share, in the third quarter, as compared to the $0.09 per share we experience, we reported in our second quarter fiscal 2010 in and higher than the $0.11 per share we reported in the same period a year ago.

Sequential improvement in our GAAP earnings was a result of higher revenues, improved gross margins, continued operating expense control, and nonrecurring tax benefits.

The difference between our GAAP and non-GAAP results during the third quarter of fiscal 2010 was due to stock-based compensation expenses of approximately $34 million, or $0.05 per share, amortization of intangibles representing $26 million, or $0.04 per share, and one-time tax benefits of approximately $33 million, or $0.05 per share.

Now I would like to review our balance sheet as of the end of our fiscal third quarter.

Cash, cash equivalents, and short-term investments were $1.46 billion, up approximately $185 million sequentially, and up $428 million from the same period a year ago.

Cash flow from operations for the third quarter was approximately $204 million, as compared to $182 million reported in the second quarter, and down from the $258 million reported in the same period a year ago.

Free cash flow for fiscal third quarter defined as cash from operations, less CapEx, and purchases of ID licenses was approximately $196 million, representing a 24% free cash flow margin.

A 12% sequential improvement from the $175 million in the second quarter of fiscal 2010, and a 20% decline from the $244 million in free cash flow reported in the year-ago period.

Accounts receivable was $394 million, up about $66 million sequentially, reflecting the higher revenue levels, and essentially flat as compared to the same period a year ago.

DSO was 41 days, down about three days sequentially, and down nine days from the same period a year ago.

Net inventories at the end of the third quarter were $239 million, up from the $211 million reported in the second quarter; 13% sequential increase, as we managed to improve areas where we experienced tight supply.

Net inventories declined $100 million, or 30%, on a year-on-year basis.

Days of inventory were 60 days, down six days sequentially from the 66 days reported in the previous quarter, and down 20 days from the year-ago period.

Accounts payable was $317 million, up $49 million sequentially, due to increased unit volumes in the quarter, and up $93 million on a year-on-year basis.

Now I would like to provide an update on our current projections for the fourth fiscal quarter of 2010.

We currently project fourth quarter revenues in the range of $820 million to $850 million for sequential increase of 2% to 6%.

We currently anticipate revenue from our network and storage addressable end markets to grow sequentially, while we anticipate a modest, but normal slowdown in our mobile and wireless end market, primarily due to seasonally lower sales of our embedded Wi-Fi products into the gaming market.

We currently project non-GAAP gross margin in the range of 58.5%, plus or minus 50 basis points, approximately 70 basis points of improvement versus the previous quarter.

We currently anticipate non-GAAP operating expenses to be approximately $235 million, plus or minus $5 million.

We anticipate R&D expenses to be approximately $185 million and SG&A expenses of approximately $50 million.

At the midpoint of our range, this should translate to an operating profit of approximately $254 million, or about 30% of operating margin.

This is a 9% increase in operating profit on a sequential basis.

Greater than twice our revenue growth rate on (inaudible) testamented operating leverage Marvell instructed to deliver.

The combination of interest expense and other income together should net out to about $1.5 million.

Non-GAAP tax expense should be approximately $12 million, or tax rate of about 5%.

We currently believe the diluted share counts will be approximately [670] million.

Taken together, we currently project non-GAAP EPS to be in the range of $0.33 to $0.39 in comp per share.

At the midpoint of our guidance, this is about -- represents about $0.36 per share, representing a sequential increase of about $0.03 operationally, offset by about $0.02 of increased taxes.

This EPS level is the highest quarterly earnings per share in Marvell's history.

On the balance sheet, we currently expect to generate about $150 million in free cash flow during the quarter.

We anticipate our cash balance to be about $1.6 billion, excluding any special items, or M&A activity.

We currently expect our non-GAAP -- our GAAP EPS to be lower than our non-GAAP EPS by about $0.09 per share, plus or minus a $0.01.

About $0.04 of this difference is related to amortization of intangibles and $0.05 in stock-based compensation expense.

Now I would like to turn the call over to the operator to begin the Q&A portion of the call.

Anita?

(Operator Instructions).

We're now ready for our first question, and it comes from the line of James Schneider from Goldman Sachs.

Please proceed.

Good afternoon.

Thank you for taking my question.

Congratulations on the strong results.

Maybe we can start out on the hard drive market for a moment.

I think there's been a lot of speculation about what the true dynamics of that market are.

But there's also been a lot of talk about potentially better seasonality heading into the first quarter of 2010.

Can you tell us from what you are hearing from your customers, what you expect in terms of seasonality going into the first quarter?

Yes, sure.

I think the continued lower cost of PCs, laptops, and netbooks over the last several quarters really actually has been very good to the drive industry, especially the 2.5-inch end markets.

So I think this is really driven by the elasticity of the end markets and, of course the new application that emerge such as gaming, and PVRs for the drives.

So overall, I think the industries are very optimistic.

That's the way I -- that's the way we look at it.

Our next question comes from the line of Allan Mishan with Brigantine Advisors.

Please proceed.

Hi, guys.

Nice job.

As I look at the gross margin improvement that you experienced here in Q3, and when I look at what happened to your mix, I actually would have thought that that would have penalized your gross margin.

Can you just walk us through exactly what kind of mix improvements you saw?

Thanks.

Thank you.

Unfortunately we don't provide gross margin by end market segment.

But if you look at -- I think this might be some perceptions about mobile and wireless end markets and the margin related to that.

That experienced very good growth and that's probably where the models might be off.

But this is continued execution by Marvell management team.

They have done an excellent job at managing.

We designed for cost and designed for gross margin.

So it's very broad and I think very good execution for us.

Alan, did you have a follow-up?

Our next question comes from the line of Sukhi Nagesh of Deutsche Bank.

Please proceed.

Thanks for taking my question and congratulations for a strong quarter.

A couple of questions I had on the comp processor side.

You talked about China Mobile in particular, Sehat.

Talking about a 30 million unit TAM or three years.

Is that the market that your comp processor will be focused on, and can you talk about what your share will be over the next few years in that market?

Thanks.

Okay.

So I assume you are referring to the PXA 920 device that we just introduced last quarter.

So this device is a very specific, high volume, high performance combination processors that will -- that serves the PDS-EDMA specific China 3G standards, plus the compatibility with existing EDGE networks.

So from the, from the performance point of view, this is truly is the state-of-the-art smartphone solutions, from the cost point of view, this will address the high volume markets.

So the way we look at it, while initially this is going to be addressed for the higher end smartphones, but very rapidly, very quickly over the next year or so, this device will also be used in main stream high volume feature phone replacements.

So our aspiration is significantly higher than any number that anybody will throw in, but this is our expectations.

We expect specifically next several years to continue to ramp on this product, as well as our future derivatives of these products.

Sukhi, did you have a follow-up?

Our next question comes from the line of Craig Berger of FBR Markets.

Please proceed.

Hi, guys.

Great job, and thanks for taking my question.

I guess I want to understand the -- what you view as the sustainability of gross margins.

We've seen, obviously, a huge ramp in two quarters and I know investors wonder about the sustainability.

And then I have a follow-up.

Thank you.

Okay, so in my opinion, the gross margins in semicon business should be on the high end as the price of semicon devices get lower.

To offset the increased cost of R&D, to offset the increased costs of litigations, the offset and increased costs of sales and marketing expenses.

So this is a fundamental shift that you will see over the next several years for any company to be perceived to be doing well in this business.

So the kind of numbers that we have I think is a fair number, and our plan is to maintain around that number for as long as we could.

And I think this is what I think any best in class semicon companies will have to achieve as well.

Clyde, do you want to add onto that?

No, I think, Sehat, you said it right.

Just to add one piece -- the cost of bringing out new products is increasing to the barriers -- increasing and what Sehat was describing as the barriers to entry gets raised and our ability to design at those geometries gets us best in class and margins reflect that.

Craig, do you have a follow-up?

Yes.

And then the follow-up is on storage where guidance is basically gets you above prior peak shipment levels.

I know I'm going to get asked this from investors.

Do you think your customers are building inventory downstream?

Do you think you're shipping over PC drive consumption rates?

And lastly, do you have any thoughts about paying a dividend?

You're generating a lot of cash.

Clyde, do you want to -- why don't you take over this question.

So, it's -- we -- particularly in this business, a storage business, we are on a consignment basis, so customers call when they need.

That implies a couple of things.

One is they don't need to build inventory.

Certainly not of our chips-- and so that's part one.

Part two is, as Sehat mentioned, I think there's a transformation that I think, Craig, it's good to mention to investors that price of PCs -- full-featured PCs, laptops are now sub-$500 levels, and that's creating demand, particularly in the BRICK countries, that people might not (inaudible).

So short answer to the inventory question is while we are not experts in the channel, it doesn't appear to us that there are inventory buildups.

As to the second part of your two-part follow-up question, I think that's a consideration, as dividends for a future period of time.

I think right now, I think all of us would agree, we are not out of the economic downturn.

Yes, models, financials are very good and excellent, but that reflects our engineering and operational excellence and our team success.

So I don't think it's -- I think it's too soon right now to be talking about distribution of cash.

Fair question, but I don't think we are out of the woods economically to be discussing those things yet.

I want to add a little bit on that second part.

My opinion is, whenever we need money, the banks will not be on our side, so the banks will never be on our -- our friend, when we need money, when we have plenty of money we have lots of friends.

So my, my bias is we need to build a cash modem to weather any possible uncertainty in the future, I mean I'm talking about the Company, talking about the economy.

Thanks, Craig.

Operator, next question, please.

Our next question comes from the line of Glen Yeung of Citi.

Please proceed.

Thank you.

My first question is with respect to the business in China, the TDS business in China.

If you look out over the course of the next four quarters, is there a point in time where you expect there to be an upwards inflection in that business?

And then, I'll ask it now, my second quarter really pertains in a big picture sense.

When you think about your reported quarter, when you think about your guidance for the out quarter and kind of a sense -- your level of visibility today versus where it was a quarter ago and versus where it it is normally.

Can you mainly give us a sense as to what kind of visibility you think you have?

Thanks.

I'll take that first part and Clyde, you can take the second part.

So on the first part, the PDS DMA, we are really optimistic, we are really glad that we have been working with China Mobile for two years, more than two years on this.

It has been an extensive engagement with China Mobile in qualifying our portable stacks in the network.

PDS DMA is the crown jewel of China 3G, so this is their, the pride of China technology.

We have strong belief that PDS DMA will be a dominant solution in China moving forward.

So right now all of our partners are working day and night to build various (inaudible) phones utilizing this technology.

So we hope to see second half next year will be a very good ramp for this product.

What I meant, single-chip highly integrated PDS DMA smartphone technology that we just introduced last quarter.

As to the second part of the question and visibility, Glen, I would say compared to six months ago, better.

Compared to nine months ago, obviously much better.

So I think there is, I think part of it is inventory.

I answered earlier, I think it's probably more leaner than probably back then.

We talked about PC demand.

I think most people don't see the demand coming from lower priced PCs.

And then from time to time, we've had shortages, so people tend to now given longer lead times.

The foundries, Glen, I'm sure you know and probably have written about the foundries that tighten up supply.

So we tend to get longer lead times and longer visibility today than we did six ,or nine months ago.

Great.

Thanks, Glen.

Let's take the next caller, operator.

Our next question comes from the line of Sumit Dhanda of Banc of America/Merrill Lynch.

Please proceed.

Yes, hi.

Couple of questions.

First, just so I'm clear, Clyde, you and Sehat mentioned that the 58.5% gross margin level is sustainable.

How -- that seems to be a little bit at odds with what you had said on the last call, where you thought the mid-50s was a more sustainable model.

Could you just clarify what the expectation is on a sustained basis going forward?

I, I think I was trying to avoid then giving a new update on our model.

We've been historical, given updated models at the end of our fiscal year, which is our next earnings call and I would like to defer to that.

We've seen improvements.

Part of it, I described before, is make continued improvements.

Fair point that we did not forecast to you those improvements, but sometimes those improvements may not be as certain at the beginning of the quarter.

Second thing is our revenues improve by about $100 million since our last forecast, allowing us to flush out all our inventory at higher costs and start shipping inventory at low cost.

So all those things combined would show the sequential improvement.

Fair point on giving better visibility, but I'll defer the model.

We'll give you a formal update in our model next quarter.

Okay, great.

Do you have a follow-up?

Yes, I did, actually.

Great.

On the OPhone success that you talked about, Sehat, can you clarify whether you have any exclusivity on the silicon or what kind of share you anticipate given that you're getting a little better visibility, about half a dozen OEMs starting to design in your silicon, and along those same lines, I'm assuming that even if that grows to be a big business sometime in the future, you'll be able to target margins in line with your corporate model there?

Sure.

So the PXA 920 is a -- is our first highly integrated PDS DMA -- plus high performance application processor, graphic engines, and so on.

So this is a very high end device we designed with China Mobile, but China Mobile does not build phones, so they work with their partners and our partners to build phones that they will sell.

So this is not specific to any customer.

This is a device that will be used by half a dozen to dozens of cell phone manufacturers to build their own version of phones.

And some of these phones, a lot of these phones will be OPhones, but some of these phones will not be OPhones, but they will still have PDS DMA, they will have the same PDS DMA, they will have the same chip.

In fact, we forgot to mention, actually, it's not just the chip.

It also comes with its companion silicon.

It will come with our own RS technology.

It will come with our Wi-Fi, Bluetooth, FM radio combo technology.

It will come with our PMIC, Power Management technology, battery charger technology.

So a lot of these things actually comes, starting from this point, it comes with our chipset solutions.

So this is very actually exciting for us because we have been working in a lot of different areas.

We've been working on PMIC for many years to address this market.

It just happened -- everything aligns up.

Now we can start shipping these products and then moving forward, more and more of our products will have this chipset solutions available.

Clyde, did you want to take the question about the margins being in line with corporate average, the OPhone?

We don't disclose that, Sumit, so --

Should be fine.

I think we just -- we're comfortable.

We should be fine.

Great.

Thanks, Sumit.

Operator, we'll take our next call please.

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

Please proceed.

Sure, thank you very much.

Can you hear me?

Yes, Uche, we can, thank you.

First of all, congratulations on a very good set of numbers.

Clyde, when you were talking about R&D and some customer-funded R&D and the (inaudible) nature of that, can you please just explain what it means?

Yes, from time to time, we have an array of nonrecurring expenses that sometimes are funded by customers, particularly in some areas of our (inaudible) business, that tend to fluctuate quarter over quarter, somewhat lumpy.

.

We had more credits in the Q2 period and the Q3 period.

And the other part of it is we had a shutdown scheduled in our July, Fourth of July, and some other countries, this quarter obviously we didn't have that scheduled.

I think we were fairly busy enough this quarter.

So the net result is we used up less of that vacation accrual, resulting in an increase.

There's no structural change in R&D.

We still intend to continue to manage tight expenses, as you can see from our forecast for

Also can assume that most of our R&Ds are using advanced process technologies, so significant costs of our R&Ds are because of that mass cost.

Right.

Okay.

I see.

Just a -- I had another question I wanted to ask you.

In terms of the new revenues, new product new revenues, the 10% of your revenues, now which are new products, what is the implied implemental cross margins on these new products, and how much of those new products is China Mobile OPhone?

None in -- I mentioned earlier, 10% of our stuff in Q3, none of that is OPhones yet.

It's still early, I think he answered earlier, we expect to ramp up in the second half last -- next year.

And we don't disclose by-product segment -- for competitive reasons of gross margins in any given --

Sure, but should we -- from the way the gross margins going up, from the comments you made, looks like the products are coming in at higher or at least come in average gross margins.

Is that how we should think about it?

Very comfortable gross margin across our product line, and we will provide our guidance -- further business model updates to you guys next quarter.

Our next question comes from the line of Adam Benjamin of Jaffray's.

Please proceed.

Hey guys.

Just a follow up on the commentary in the Bluetooth Wi-Fi combo part that was up 50% in the quarter.

You guys have one Tier 1 gaming win there.

I'm just curious, outside of that gaming win, were there any other Tier 1, either in the handset space or gaming market that are contributing to that growth?

Yes, so it's a little talk about all the specific market segments that were targeting, so -- but we are very happy, our tractions in the combo device is taking -- is finally taking off.

Remember, we have never invested in Bluetooth until about -- a couple years ago.

So now we already -- that was our first generation device, the one that's ramping now.

We have second tier device in the pipeline and we'll have the third generation device next year.

So the beauty is we have all the software, we have all of the qualification done, so the next -- the rest is just putting the different combinations for different market segments.

So, you ask which type of markets?

We'll say anything embedded that requires Bluetooth and Wi-Fi is a fair target market.

Do you have a follow-up?

Yes, just a follow-up on the gross margin.

I know it's been asked a couple different ways, but, Clyde, as you think about the gross margin going forward from these levels, clearly you got a benefit of some older higher priced inventory rolling off, and then you have potentially mix, and then just some cost efficiencies that you may have locked down during the downturn.

If you think about those three dynamics going forward, how far -- if you can maybe break it out in percentages, how you should be thinking about it, that can contribute to either this level sustained or even higher, that would be helpful.

Clearly, maybe quantifying where you are in terms of rolling out the higher priced inventory first, that would be great.

Thanks.

I will cover this a little bit -- let me give you a hint of why this is happening.

Maybe the Clyde will follow me -- follow up on this.

In our business, in our semicon business, it's a thing that several things happens.

Higher integrations, the high integration using more advanced process technology will allow our customers to reduce costs.

Our customers will have fewer chips on the board.

They have lower power.

They have fewer layer of PC boards, less time to assemble the product, shorter time to test the product.

So more reliability, but overall, improved performance and lower costs.

So while we continue to increase the integrations, we actually give our customers lower overall costs as a result, so what happens is, the customer is seeing benefits from this high integration levels, and at the same time, we are also getting incremental benefits in the profit -- in the gross margin, so this is what is happening.

This is my opinion.

This is what we are seeing.

And I think for as long as we continue to do high integration, or higher integration, and using more advance processor technology, this trend will continue, or at least be sustainable.

Yes, Sehat, you are absolutely right.

To answer your question, there's obviously some improvement we forecasted and another 70 basis points in the January quarter, so there's obviously some improvement to be had.

I think you guys know us by now.

We always driving to improve better, but I just want to go back to what Sehat said so you guys understand this.

Marvell has been a great engineering powerhouse.

This is the benefit of giving customers better products and billing them very efficiently.

So that -- number of great engineers do that.

That's the effect of it.

And hopefully that keeps going, but right now we forecast one quarter out and give you guys another 70 basis points-- we'll update that in three months.

Our next question comes from the line of Nicholas Aberle of Caris & Company.

Please proceed.

Hi, guys.

Thanks for taking my question.

Just another question on the wireless business.

So do you guys -- since you guys purchased the Intel Wireless assets back in 2006, we've been waiting for you guys to layer on another top tier wireless customer in addition to the Legacy REM account.

Looks like the China Mobile relationship's going to help you break into the top tier.

Do you expect that foot in the door, at China Mobile, to allow you to extend the relationships with top tier guys to non-China GSM EDGE opportunities?

Yes.

Is there any --

Any color?

No.

Here's the key.

The key is technology.

We need to prove the technology is superior, we need to make the chips to be small die size, low cost, low power, long battery life.

So people can use cheaper batteries, tinier phones.

Tinier phones mean cheaper and sexier, so -- software, portable stacks, interoperability in any given network in the world, whether it's China, US, or Europe.

So we have gone through this exercise.

So once we have all of these things, the rest is just business, business transactions, timing when any -- of the first tier of waiting to ramp up, introduce new products, there's nothing unique in this business versus any other business.

So for people, they will, let's say delay the introductions, maybe on the next-generation versions, so for as long as we continue to build better products over the next -- over the next year or two, we will continue to gain -- so that's just as simple as that.

Do you have a follow-up, Nick?

Yeah, one last one just on Armada.

Sounds like it's getting pretty good reviews on the trade racks.

You guys talked about 50 wins.

Sehat, what end markets do you think you're going to see acceptance with initially?

And then in your opinion, what's the biggest end market possibility over the next, say, two to three years for that product line?

Thanks.

Yes, okay, sure.

So it's across the board.

If you look at the Armada family of devices, we have -- if you look at the announcement, we have a number of devices from the lower end, lower cost, to the super high end, with dual channel HD, HD 264, VC-1, high profile, HDTV decode -- and common to all of these devices, they come with powerful embedded custom processors that we build internally that's unique to Marvell.

But they run on standard ARM instruction sets, so they run all the ARM softwares in the market.

So if some devices have single processor, some devices have dual processor -- dual core ARM processor that we built, so the market is all -- is any consumer devices from connected, HDTV recorders, Blu-ray player, to high performance digital picture frames, you can show pictures as well as gateway portal to the Internet.

The smartbooks, what do you want to call it, netbooks or smartbooks -- thin clients, in fact we have several thin clients already in the market.

So it's pretty much across the board.

You can put it in the automotive consoles, you can put it in digital cockpits.

Anything that requires display is a fair target of this device.

Question is what type of display and what price points?

And then you pick the right device for that market.

And the beauty is now we have complete, complete, complete segments, targets complete segments for all of these markets.

Great, thanks Nick.

We'll take one last caller, operator?

Our next question comes from the line of Quin Bolton of Needham & Company.

Please proceed.

Thanks for squeezing me in, and congratulations, especially on the good margins.

Just wanted to come back to the drive side of things.

Number of the drive manufacturers over the last quarter have talked about component shortages, including silicons.

Just wonder if you could talk to your ability to supply.

Are you seeing any tightness in the foundries or have you seen your lead times stretching out on the drive side?

Well I guess, I want those people that have shortage of components to start talking to me -- especially those people -- no, okay.

Anyway, so there were a time when during the upturn, obviously people did not forecast their -- do not broadcast the size of the upturn, But the beauty is, our customers -- we do not shortchange our customer, even during the tough time.

This is a testament to our -- Marvell's solutions, capability, as well as other operational capability.

So while we can, we can overnight be able to improve the throughput of the products and lines to capture those unforecasted demand.

So as a result actually, we have happy customers and this is, again, reason why over the last -- besides technology superiority, we've been talking about our storage, we have the lowest power chip in the world in storage.

We have the most -- the highest performance, we have the highest yield, we can give -- allow our customers to build drives with the highest yield, meaning lower cost to themselves.

This is one of the reasons that we continue to gain market share in this business.

Quinn, did you have a quick follow-up?

Just a quick follow-up, I know you guys have talked about pretty impressive market share numbers, both in mobile and desktop, just wondering if there's been any update on the timing of the HGST and the Seagate ramps, for mobile and desktop?

No updates at this time, but we are very happy.

So moving along very well.

Great, thank you.

Thanks, Quinn.

Well, great.

Thank you, everyone.

I think at this point, we're going to conclude the call.

We appreciate your interest in Marvell and we look forward to seeing you at future investor conferences.

We will be at the Needham Growth conference in New York City on January 14.

We look forward to seeing you then and at other investor events.

And we thank you for your interest in Marvell.

Thank you.

Have a very good day, everyone.

Bye-bye.

Thank you, guys.

Thank you for your participation in today's conference.

This concludes the presentation.

You may now disconnect and have a great day.