邁威爾科技 (MRVL) 2009 Q1 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Q1 fiscal year 2009 Marvell Technology Group earnings conference call.

My name is Tony, and I'll be your coordinator for today's conference.

At this time all participants are in a listen-only mode and we will conduct a question-and-answer session at the end of this conference.

(OPERATOR INSTRUCTIONS) As a reminder, this conference is being recorded for replay purposes.

And I'd now like to turn the call over to your host for today's conference, Mr.

Jeff Palmer, Senior Director of Investor Relations.

Please proceed, sir.

Thank you, good afternoon, everyone, and welcome to the Marvell Semiconductor fiscal first quarter 2009 earnings call.

On the call today is Dr.

Sehat Sutardja, Marvell's Chairman and CEO and George de Urioste, Marvell's Interim CFO.

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.

Before we begin, we would like to remind all participants that this call will include forward-looking statements that involve risk factors that could cause Marvell's results to differ materially from management's current expectations.

These forward-looking statements may include predictions, estimates and other commentary regarding sales trends, revenue growth, gross margins, operating and nonoperating expenses.

Such statements are usually preceded by, but not limited to, such words as: expect, anticipate, believe, should, may, will or other forward-looking statements.

To fully understand the risks that may cause results to differ, please refer to Marvell's fiscal 2008 form 10K and subsequent SEC filings.

Please be reminded that the company undertakes no obligation to revise or update publicly any forward-looking statements.

Throughout the call today we will make reference to certain nonGAAP financial measures which generally exclude the effect of stock-based compensation, amortization of acquired intangible assets and other one-time charges.

Marvell management believes these nonGAAP metrics are useful as they best reflect how the business is internally managed.

While Marvell uses nonGAAP financial measures as a tool to enhance its understanding of certain aspects of its financial performance, Marvell does not consider these measures to be a substitute for or superior to the information provided by GAAP financial measures.

With respect to historic information, the most directly comparable GAAP information and a reconciliation between the nonGAAP and GAAP figures is provided in our [physical] year Q1 '09 the earnings press release which has been furnished to the SEC on form 8K and is available at Marvell's website at the investor relations section at www.marvell.com.

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

Sutardja.

Sehat?

Today we reported fiscal first quarter revenues of $804 million, up 27% from the same period a year ago and a seasonal decline of 5% sequentially.

So we came in well above our guidance which called for a 7% sequential decline in revenues.

I am pleased with our top-line growth in what is usually a seasonally weak quarter.

We met our goals in our core markets and we attribute the higher than anticipated revenues to success with our [eight to 11M] wireless products, the growing adoptions of our network attached storage products and an improved demand for our printer system on our chip products.

I'm also pleased to say that we significantly improved our profitability during the quarter.

The improvement in margins was a direct result of rigorous cost savings initiatives we implemented in prior quarters that are just now coming to fruition.

George will go into greater detail later in the call on the financials and the drivers of the business.

So now I would like to give you an update on new products and an important part of our technology strategy.

In the hard disk drive market we continue to win designs for next generation drives for notebook, desktop and enterprise obligations.

These design wins leverage our hot disk drive system on a chip or SOC products which we confidently believe are at least one and a half years ahead of our competitors.

While we continue to execute flawlessly, lately we have heard our competitors talk about gaining market share at 45-nanometer.

Of course we would like to know how successful they have been at 65-nanometer or even at 90-nanometer for that matter.

There is a reason why Marvell continues to lead the HDD business.

It is because our SOC products enable our customers to be successful in their target markets.

Marvell continues to gain market share because we consistently exceed our customers' challenges.

Our customers tell us that our products help them to introduce new hard drives sooner with better performance, with lower power consumption, and we have improved quality.

These ultimately help our customers to be more profitable.

Now I would like to talk about solid state storage or more familiarly heard as SSD.

As you know, there is a lot of interest in SSDs these days.

We will be announcing our entry into this market at [computechs] in Taiwan next week.

Marvell is leveraging more than 10 years' experience in hard disk drive controllers as we enter the SSD market.

With the introduction of these series of products we will be unveiling our strategy to address the entire SSD market with superior control devices.

While SSDs will not replace hard drives any time soon, we see these as an additional revenue opportunity for Marvell as SSD fine roll in ultra small notebook PCs, mobile internet devices and enterprise applications.

Now I would like to provide an update on our cellular business.

As I have mentioned in prior calls, we have great confidence in our opportunities in the cellular and mobile device market, and we are well aligned with the leaders in these markets.

During the first quarter we achieved what I believe to be a very important milestone as we began volume shipments of our HSDPA communication processor to a key smartphone customer.

We expect a steady ramp to high volume production throughout the remainder of the year.

We also continue to see positive acceptance of our application processors.

As an example during the quarter [Grumman], Trimble and Samsung all introduced several new products which leveraged our high performance application processors.

We have made good progress over the last several quarters to improve the profitability of our cellular business.

We have streamlined the operations and completed the transition from Intel to TSMC for existing high volume products.

We are currently in the process of optimizing our architecture for the next generation of communication processors.

This third generation of products will leverage Marvell's mix signal microprocessor and SOC design expertise along with many of the internal developed intellectual property.

This next generation of devices will achieve even lower power, even higher performance and smaller size than the current generations of products.

We see these new efforts coming to completion over the next 12 months.

I would also like to mention some of the products we introduced during the first quarter, in particular the discovery innovation series of SOCs integrating our latest CPU or microprocessor technology.

The performance of the embedded CPU in these products is above that of any standard [arm] processors out there.

These products are purpose built devices for carrier grade Ethernet networking.

Already we have multiple design wins locked in with major networking OEMs with volume production at the end of the year.

While Marvell has long been recognized as the leader in mixed signal SOC design, we have also achieved leadership in high performance, low power embedded CPU technology.

Marvell has the world's largest development team focused on arm instruction set CPUs.

Nearly all of our SOC products today integrate an internally developed arm instruction set compliant processor.

With the development of a new generation of processors, which we are calling [Sheba], we will further strengthen our technology leadership.

The Sheba CPU technology will be incorporated in several new products in the near future.

We believe this new high performance CPUs will drive a new generation of mobile consumer and enterprise infrastructure equipment.

Now I would like to pass the call to George to provide the financial update and guidance before turning to your questions.

George.

Thank you, Sehat, and good afternoon, everyone.

Today we are pleased to announce fiscal first quarter 2009 revenue of $804 million representing nearly 27% growth over revenue of $635 million for the same quarter in the prior fiscal year.

Our first quarter results showed a seasonal decline of 5% sequentially from the $845 million reported in our fourth quarter.

These actual results were above our Q1 guidance range of $775 million to $785 million.

As Sehat mentioned earlier, we are pleased with the progress we have made toward improving our profitability.

Looking at gross margin, our nonGAAP gross margin for the fourth quarter (Sic-see press release) was 52%, significantly better than our prior guidance range of 48.7% to 49.3% and above our long-term operating model of 50%.

Our gross margin was up over 330 basis points from Q4 and was up over 310 basis points from Q1 of the prior year.

The improvement in gross margin above our Q1 guidance was due to a number of factors.

First, on a sequential basis we realized a 240-basis-point improvement due to better standard costs and wafer yield enhancements.

In addition, through focused inventory management we realized an additional 90-basis point improvement.

Turning to operating expenses, on a nonGAAP basis expenses were $255 million, which was better than the prior guidance range of $290 million to $295 million.

Included in the operating expenses were several one-time charges and benefits.

A one-time benefit includes a $24.5 million payment received from our director and officers liability insurance carriers and had the effect of reducing operating expenses.

This one-time payment is in connection with the previous public disclosure of a tentative settlement of the shareholder derivative litigation.

Separately and also as previously announced, during the quarter Marvell entered into a settlement with the SEC in connection with the previously disclosed investigation of the company's past stock option granting practices.

Marvell agreed to pay a $10 million civil penalty in connection with the settlement.

This payment was charged to operating expenses in Q1.

As we have stated before, one of our goals is to lower our operating costs through intelligent spending practices while simultaneously maintaining our investments in future product development.

In Q1 we have achieved many of our cost management goals and we continue our focus on improving operational efficiency.

We have meticulously reviewed our costs looking for areas which will allow us to be more efficient.

The dominant benefits we realized relative to prior guidance were due to lower legal expenses, lower engineering-related costs such as mask costs, take-out charges, savings on engineering software costs and NREs.

Additionally, we realized modest benefits from refining the accuracy of our expense accrual process.

All of these benefits were partially offset by first quarter seasonal costs related to annual companywide employee performance reviews and the associated compensation increases.

During the quarter we realized an expense in interest and other income equal to roughly $4 million, approximately in line with expectations.

Turning to our tax expense, our effective nonGAAP tax rate was 5.4% for Q1, lower than originally anticipated.

I would like to highlight that our tax payments do not increase proportionately with our profits.

This is a result of our cost plus tax structure.

By way of explanation, it is important to first understand that Marvell manages its tax expense with efforts similar to how we manage our operating expense.

Our tax expense is largely determined by regulations with tax authorities around the world and is derived by what is commonly referred as a cost plus formula.

This means that our tax expense is determined by our operating expenses in different geographic locations.

Because such tax expense does not directly vary by level of income, when Marvell has higher income, the effective tax rate declines.

Conversely, if the company is not profitable, it would still incur a tax expense.

In our first quarter because income was higher than anticipated, the international tax structure caused the effective tax rate to be lower than anticipated.

Next, let's look at nonGAAP net income for the quarter.

Our result was approximately $150 million or $0.24 per diluted share and is up 22% compared to the $123 million or $0.20 per diluted share during our fourth fiscal quarter last year.

This quarter's performance is up 380% compared to the $31 million or $0.05 per diluted share in the first quarter of last year.

Shares used to compute nonGAAP net income were 624 million, down from the 627 million shares in the prior quarter.

The lower number of shares was due to a lower than anticipated share price when using the treasury method of computing diluted share account.

Now -- excuse me.

Now I would like to offer some additional insights on our revenues results during the quarter.

Our performance was better than anticipated considering the seasonality of the quarter.

From an end market perspective, the quarter played out modestly stronger than our original expectation.

Order linearity during the first fiscal quarter was balanced and improved both sequentially and on a year-over-year basis.

Several product areas had strong growth trends either sequentially or on a year-over-year basis.

Revenue from our storage products was flat sequentially as anticipated but grew over 30% on a year-over-year basis.

We remind investors that the first calendar quarter is historically the seasonally weakest period for the hard drive industry.

In our view, our ability to buck this trend is attributable to two factors.

First, our primary hard drive customers continue to gain share within the marketplace.

Secondly, the overall hard drive industry is benefiting from the continued demand for notebook PCs which is creating positive demands for two and a half inch drives.

We see the trend in demand for two and a half inch drives as creating a disproportional benefit to our customers as well.

During the quarter unit shipments of mobile hard drive system on a chips increased on a percentage basis in the high teens sequentially.

This positive trend -- these positive trends were partially offset by seasonal declines in desktop and enterprise SOC and read channel products.

Several of our major hard drive customers experienced sequential growth on a revenue basis with Western Digital being the only customer exceeding 10% of our revenue during the quarter.

Sales of our cellular products were in line with our expectations and declined modestly, but showed growth in the mid-teens on a year-over-year percentage basis.

Demand for our communication processors accelerated during the quarter on both a unit and revenue basis.

However, sales of our application processors were slightly below our expectations during the quarter as one of our older application processor designs came to an end of life.

And as Sehat mentioned earlier we began revenue shipments of our HSDPA products during the quarter.

Sales of our enterprise connectivity products declined sequentially in line with our expectations, but were up over 30% on a year-over-year basis.

International demand for our metro Ethernet products continued to be robust for both new and existing products.

This was offset by expected weakness in demand for system controllers and SMB access products.

We remind investors that Marvell experienced significant sequential growth in our enterprise business during our previous fiscal quarter.

Many of the trends recently highlighted by our peer group in the enterprise networking space are consistent with our market view.

The only difference being the timing of orders among vendors.

Additionally, sales of our PC connectivity products declined sequentially in line with expectations and consistent with normal seasonal patterns.

Sales of our wireless connectivity products were better than we had anticipated, especially demand for our next generation 802.11n connectivity products.

Coupled with the increased demand of 802.11n products were sales of network attached storage processors.

One of our high profile customers was a leading consumer-oriented customer introduced several wireless storage router and access point products.

Lastly, sales of our printer products were better than planned and also contributed to the greater than anticipated revenue during the quarter.

Now turning to the balance sheet.

Cash equivalence and short-term investments were $774 million, up roughly $143 million sequentially primarily due to better total sales, and the balance is up 37% compared to levels of a year ago.

In terms of positive cash flow in Q1 Marvell generated $155 million in cash from operating activities.

This compares to $163 million in the prior quarter and is up substantially over the $54 million generated at this time last year and further illustrates our improving year-over-year trends.

Accounts receivable were $370 million, up $38 million sequentially on strong order linearity.

The sales outstanding, or DSOs were 40, essentially flat sequentially from fiscal fourth quarter and lower on a year-over-year basis from the 44 days reported in the first quarter of fiscal year 2008.

Inventories at the end of the quarter were roughly $370 million, down $50 million sequentially as we focused on improving our operational efficiency.

Days of inventory, or DIO, were 93 days, up sequentially from the 83 days reported in the previous quarter, and equates to inventory turns of roughly four times in line with our historic norms.

Accounts payable were $168 million, down $63 million sequentially.

We would now like to provide guidance for our anticipated performance in the fiscal second quarter of 2009.

Please note we will not be providing full-year guidance on today's call and cannot provide any additional commentary on our full-year expectations.

We would like to highlight that our guidance today is a result of a bottom-up analysis of market demands and actual sales trends in conjunction with ongoing dialogue with our customers about their expectations.

With that perspective as a background, we anticipate fiscal second quarter revenues in a range of $830 million to $840 million, which represents a growth rate of 26% to 28% over the same quarter of the prior year.

To help you calibrate our anticipated revenue, we'd like to share with you our backlog coverage ratio.

This ratio is determined by our 90-day backlog entering the beginning of the fiscal second quarter divided by the midpoint of our revenue guidance.

This ratio is in the mid 60% range and is in line with historical patterns.

This also takes into consideration comparative second quarter seasonality.

With regards to profits, we anticipate nonGAAP gross margins in the range of 51% to 52%.

While this is higher than our previously stated 50% long-term model, we are not changing our longer-term model of 50% at this time.

This is attributable to generally anticipated increases in longer-term manufacturing costs, making it premature to determine whether higher gross margins are sustainable at this time.

We will provide a more detailed update to our longer-term operating model next quarter.

Turning to operating expenses, we anticipate nonGAAP operating expenses in the range of $280 million to $285 million.

With regards to the line item interest and other income, we anticipate a $2 million to $4 million expense.

In terms of our tax rate, we anticipate the effective nonGAAP tax rate to range between 6% and 8% and in review of our share count on a diluted basis, a range of [630 million to 635 million] is anticipated.

Lastly, because Marvell has increased -- has been increasing its cash balances, we anticipate a partial pay-down of our debt by an amount up to $100 million in the second quarter.

Next, and before starting our Q&A process, Sehat would now like to provide an update on the status of our search for a permanent Chief Financial Officer.

Sehat?

Thank you, George.

As many of you are aware, Marvell has been conducting a search for a permanent CFO.

I am very pleased to announce today that search has been completed.

Clyde Hosein will be joining Marvell as CFO on June 23.

Clyde Hosein may already be a familiar name from his prior position as Chief Financial Officer at Integrated Device Technologies.

Clyde has extensive financial management experience in the tech sector as well as strong record of improving operations.

We are very pleased to welcome Clyde to the Marvell team.

I also want to thank George for his excellent service as Interim Chief Financial Officer at Marvell.

I am also pleased to announce that George will continue with Marvell in the role of acting Chief Operating Officer.

Pantas Sutardja, who has been serving as acting COO, will relinquish those responsibilities in order to devote his full attention to his role as Chief Technology Officer.

George has been an important member of the Marvell management team for the past few months and we look forward for his continued leadership.

Thank you, George.

Tony, at this time we would like to invite your questions and request the operator to open the Q&A process.

(OPERATOR INSTRUCTIONS) Questions will be taken in the order received.

(OPERATOR INSTRUCTIONS) Also we'd like remind all participants that you'll only get one Q&A question and one follow-up question.

(OPERATOR INSTRUCTIONS) Your first question comes from the line of Arnab Chanda with Deutsche Bank.

Please proceed.

Thank you.

Two questions.

One's about revenue, one's about gross margin.

About revenues for this year and next year, could you kind of restate what you think are the drivers of revenue growth for this year and next year in your different segments as well as new things like storage, optical storage, solid state drives, power management, etc.?

And I'll have a follow-up, please.

Yes, Arnab, as I had mentioned many times in the past, the revenue drivers of the company for the year, for the current year are the products that we develop many years ago.

So if you -- if I give you an example, the storage products that will drive the revenue for this year, for example, those are the products that we developed two to three years ago, so this is -- okay.

There is very little that is going to make a difference in terms of revenue of this year related to new product introductions.

So when we talk about products like SSDs that we will introduce at the [computechs] next week, those are the products that we will not -- we will not put as a large revenue drivers for the year.

We do believe that those will be important like two years or three years down the road.

So this is the kind of trends that we have seen over the many years that we have done business in building products that typically takes time for design wins and takes time for the customers to develop the software and to get a mature and then once a customer -- a customer go production, okay, then that's the only time that we will be able to recognize the revenue.

The second question.

Okay.

Second question on the gross margin.

Let me see if I can be clearer.

So, the concern was that your Xscale business is not of the margin structure you wanted it to be.

Obviously the overall company's margins have improved.

Is it possible if the Xscale business goes better than expected because of your HSDP, etc., is that going to be -- is mix of cellular versus the other businesses be a driver for gross margins going down or do you think the volume or the cost reductions can offset that?

I'm trying to understand why the margins would go down from here, in other words, like your long-term model.

Thanks.

Okay.

In terms of the Xscale, we did -- okay.

We did mention in the last several calls that, okay, we have been working very hard to move the production from the Intel to the TSMC manufacturing model.

That actually creates, okay, an improvement in the yield and as well as in the cost structure of the product.

So we also been working hard on improving, let's say, cost of testing the product, okay, the test time is an important, important activity that we have taken for many, many of our products.

So in the long run the Xscale -- okay, the Xscale products will move to the -- will be supplemented with the new generations of the processors family that we have been working on that we talk about in this call about the [Sheba] processor technology that will be the -- there will be the migration path for the Xscale to achieve even higher performance, even lower power and significantly smaller dye sizes, so this is a way for us to address the mainstream higher volume cell phone markets.

Today the existing Xscale, because due to the larger dye sizes, but with super high performance, those solutions are suitable, very attractive for the very high-end market for the smartphone.

So my belief okay, my belief is that, okay, as soon as I can get these products to be made to be smaller and lower, lower cost, I will -- even our existing customers they have been using the Xscale products for the very, very high-end markets will be able to move on to address the higher volume markets, so this is an important area for us to execute.

And I believe, okay, we are now on our way to do that.

And Arnab I would be happy to add on another perspective in terms of the question relating to our longer-term operating model and why have we not chosen at this time to change our 50% gross margin target longer-term operating model.

Undoubtedly as everyone is aware on the macro economic scene, there's a certain volatility in worldwide costs and many of these costs are set at market rates that have been recently subject to extreme volatility, whether it's energy, gold, the cost of transportation, etc.

We feel at this juncture it's prudent for us to do a deeper analyses of the longer-term impacts on our business, and hence we will come back at the end of this quarter and provide a detailed update to our longer-term operating model.

Thanks, Sehat, and congratulations for hiring the CFO, and thank you, George for all your contributions.

You're welcome.

Next question.

With UBS, your next question comes from Uche Orji.

Please proceed.

Yes.

Can you hear me?

Yes, Uche.

Yes.

I actually have a couple of questions.

One main question a follow-up.

In terms of the volume of Xscale, how much more of the Intel base Xscale do you still need to work out of inventory at the moment?

And in terms of your ability to expand the communication processor business just as the smartphone or the categories like mid-range and low end, can you give us an update if there's anything you're doing on that side?

So Uche, I will address your question in terms of volume of Xscale to be worked out of inventory.

You may recall that from our earnings conference call in -- last March we mentioned we have completed the transition of our current generation of high volume communication processor products to TSMC.

With respect to our 3G HSDPA products, these are only being sourced from TSMC for high volume manufacturing, so no transition is required for such products.

As well as any new orders received for these products will be fully sourced from TSMC.

We continue to maintain a level of application processor inventory which we believe has a long shelf life and as such expect to burn off that inventory over the next several quarters.

Okay.

That's great.

Thanks.

Could you kindly repeat the second part of your question?

I was just trying to understand the -- why the strategy for the CP business, if there is given that -- the size of Xscale business now and the fact that it's purely for smartphones this way the strategy will be focused on, or do you have plans to address other categories of phones, a mid-range or even low-end phones, just to be able to diversify yourself away from the [rain] business?

Right.

Okay.

So I've mentioned this several times in the past.

I just want to clarify again that our strategy is to make sure that our customer, our existing customers to be super successful.

They are the number one priority.

Okay.

The second priority is for us to develop the longer-term technology that will allow our products to be made significantly to be smaller in dye size to address the higher volumes, so I also talked about this just earlier.

So this is the long-term sustainable strategy to go after very, very high volume markets.

Whether these products will be used by our existing customers or new customers, I am confident that we will be able to see good results if we have the right cost structure for this product.

So, okay, we will -- as we get these products out, we will give you more updates later.

All right.

That's great.

Just one last question on the solid state drives.

Can you just update me on what the opportunities you see in the SSD market and you also tell me how you view as ideal partners, the traditional or nonflash manufacturers?

What is the opportunity and how do you plan to address this market?

Yes.

There are two types of markets that we believe is important to address in the SSDs.

One is that we talk about is the products that we have to introduce next week.

This is the market -- this is a product that is targeted to address the ultra mobile PCs as well as the companion, okay, solid state drive for traditional laptops.

So I will look at that as the biggest opportunity.

These are the devices that will have reasonably -- reasonable capacity in terms of the gigabytes, but low price, low cost enough.

And then the other market opportunity is the very, very high end enterprise class SSDs.

These are multi -- these are devices that can store 64 gigabytes, 128 gigabytes, and even down the road to be even higher capacities, and these are the kind of device that will be sold at very, very high prices, but people will still pay for it because it comes with absolutely the best performance.

Those are the two areas that I believe that will be an important areas for us to address.

Now, in terms of the long term, I am optimistic that this market could be big as people are starting develop -- people are successful building smaller, higher capacity flash chips and also down the road as 415-millimeter nanometer -- 450-millimeter wafers comes out on the line, this could be a very, very interesting market for us to address.

Okay.

That's great.

Thank you very much.

Thanks, Sehat.

Sure.

Operator, our next caller please.

Your next question comes from Craig Ellis from Citi Financial.

Please proceed.

Thank you for taking the question.

The first question is for George.

George, you identified a number of items that were tale ends to gross margins in the current quarter and it sounds like there's also operating expense management initiatives underway.

Can you identify what's in play in the outlook that could be a tail wind to both gross margins and operating expense control?

So in terms of operating expense control, we have a -- shall we say a heightened focus of attitude of being cost conscious within the organization worldwide.

That in conjunction with our internal approval processes enables us to carefully review any significant expenditures before they're actually approved.

We also have benchmarks where we have our annual operating plan and our forecast, so every business decision requires careful justification.

The coordination among and the collaboration among employees continues to create the tail wind, if you will, for having this heightened focus on cost management.

In terms of the gross margin, we have been successful in gaining better price agreements with our various manufacturing operations, and we expect those agreements to sustain themselves through the second quarter, and that becomes the basis of the tailwind into Q2.

Okay.

That's very helpful.

And then given that those initiatives are, I would expect, gathering steam, I would expect an offset would be something like incremental hiring and that's why operating expense is guided up modestly?

Yes, that would be one contributing factor, that's correct.

Okay.

And then switching gears, Sehat, I wanted to ask a product-related question.

As I look at the storage system it seems like iterative technology in desktop and notebook, the enterprise system on chip transition and NAN flash controllers are three product revenue drivers this year.

Can you rank those in terms of which will contribute most to growth this year and then also rank them for next year, please?

Yes.

I mentioned earlier that really that the revenue driver is always not the chips that we develop that we introduced last week or next week or next month.

Really the chips that we develop two, three years ago.

So this is the nature of a lot of the products that we built, especially the disk drive market.

So if you look at the products they are going into production with our disk drive customers, many of those chips were developed two, three years ago.

And when -- so -- and when we talked about SSD, the reason we want to be early in the SSD market is because we want to put our foot in the door, so that when the price of those SSDs, okay, the chips, sorry -- the flash -- the price of the flash chips, individual flash chips, they will go into the SSDs becoming very attractive, we want to have our devices to be -- to have early traction into the market.

So most contribution will be more of a year from now, if not a year and a half, two years from now.

Okay.

So the iterative technology and the system on chip transaction are the drivers this year and the SSD controller starts to gain revenue traction next year?

Yes.

Thanks, Sehat.

Your next question comes from the line of Craig Berger with FBR Capital Markets.

Please proceed.

Good afternoon.

Nice job, guys.

On the gross margins is it fair to assume that the cellular gross margins are over 40% now?

Yes.

And is that sustainable here?

Is there further improvements in that line item?

Yes, especially, in the long run.

As I say, our goal is to go to after the -- to develop the next generation device which is going to be significantly smaller in dye size.

Once we achieve that, that's a time when we can make the gross margins to be in line with our traditional business.

Can you guys talk about inventories that your customers or inventories in the PC supply chain, what you're hearing from customers?

Dell's inventories are up, Ingram Micro, IBM, Sun, Seagate, obviously.

Can you just talk about channel inventories?

You have heard some comments in these areas?

Tell us a little bit more about what you've heard.

Well, I'm just wondering if you guys think the PC supply chain is lean right now, normal, maybe a little more than -- more inventory than needed, what you guys are seeing out there in terms of inventory levels at your customers.

Okay.

May I invite Jeff to answer to that as you've got your good finger on the pulse of trends in the marketplace.

Yes, so Craig, in terms of inventories we're actually seeing inventories look fairly in line for the seasonal period we're in.

We are seeing our customers gaining share, also combined with increased demand for two and a half inch drives which is kind of offsetting some of the normal seasonal weakness you see here in the Q1, Q2 time frame.

Last question.

On the cell phone chips, obviously, RIM is a customer.

Who else are your current customers in that business, if you could?

We have -- okay.

In this business we have two types of customers.

One is this fully integrated base [ban] plus application processors, and we have customers that only buy our application processors in the cell phone.

So if the majority of our customers, okay, are the one they are buying stand-alone application processors, combining with our -- somebody else so those are the majority of the business to date.

And which customers are those?

Craig, as I -- we mentioned in the prepared remarks, this past quarter we also saw three new customers, we saw Samsung, Trimble and [Grumman] all introduce new products based on our application processors.

Do you envision RIM becoming a 10% customer this year?

We're not making forward-looking statements statements out past the third quarter.

Yes, we're not doing that.

We're not -- when it happens, okay we'll talk about it.

Thanks a lot, guys.

With Merrill Lynch, your next question comes from Srini Pajjuri.

Please proceed.

Thank you.

George, just one clarification on the OpEx front.

As we get into the second half of the year, guiding for $280 million to $285 million for this quarter, I'm just wondering what are some of the puts and takes as we look into the next couple of quarters?

Well, Srini, I'll be happy to comment on the second quarter, but I'll have to be consistent with Marvell's policy to not take it beyond that.

So we mentioned some of the reasons why we're able to realize some significant savings on OpEx in Q1, for example, the legal expenses starting to trend down, and with regards to a more careful analysis of our accruals to make sure expenses are not over accrued, and we also have the benefit of some engineering costs that were either saved or deferred.

I would expect that, for example, some of the masks costs and the take-out costs of engineering that occurred in the first quarter will to some degree come to realization in the second quarter, so hence we would expect the total operating expense in the second quarter to be modestly above that in the first quarter.

And with regard to your question on the remainder of the year, we'll comment on that at the second quarter conference call.

Okay.

Fair enough.

And then just on the business side of things, Sehat, looks like if you look at the last couple of quarters, you have probably outgrown your customers on the storage front.

And I'm just wondering, looking out there again in the next couple of quarters, what are the puts and takes?

And assuming that your customers will grow after whatever rate that they are they'll grow, do you expect to grow in line with your customers or do you think there are any puts and takes that will help you or hurt you as you look out to the next few quarters?

Yes, I'm optimistic that the technology that we develop over the last several years, okay, resulting, okay, in products that a year and a half, maybe even more ahead of compared to our competitions will allow our customers to win market share away from our known customers.

So from that point alone, okay, I am optimistic that we will -- okay, we will be able to gain market share as an indirect result of the success of our customers.

The other part that I'm also optimistic is the emergence of HDTV, the -- all these DVRs also will create new opportunities, well I have to maybe combine those.

The fact that the costs of the pair of byte drives are coming to the point where it's become so affordable, the -- this emergence of this HDTV DVRs will also create new opportunities, even bigger opportunities for the storage products.

So I think the market will -- as the economy, if the economy continues -- I mean the economy is really -- the world economy -- the world economy is doing fine, okay, I think we'll be fine.

Okay.

And then you did mention that you're seeing some price pressures, obviously, you did a pretty good job on the gross margins.

So I'm just curious as to where you are seeing this price pressures and if you can provide any more color on the front.

Yes.

In terms of price pressures, we talk about price pressure because this is the day-to-day life that we have to deal with as semiconductor suppliers.

Our customers will continue to press us to lower the prices and our job is to build better and better products with more features to offset some of those price pressure.

So in general, okay, we have been successful in maintaining this balance of lowering the price and improving the features.

Okay.

And just the last question.

Sehat, again, it's been more than a couple of years since you announced the optical products I'm just wondering if you could provide us any time line or any update as to when you expect to see meaningful revenues from that segment.

Thank you.

Yes.

Okay.

So I -- we -- I don't know whether we talk about this last time, so we did -- okay, we did, okay, work on the HD DVD, so we had a fully committed customer working on HD DVD last year, and unfortunately the HD DVD, okay, standard was -- Toshiba decided to get away to stop working on HD DVD.

That impact our market entry into the timing of the market entry into the optical.

But it's not -- on the other hand, it's -- we also see it as a positive.

Now we only have to develop Blu-ray versus before we have to develop HD DVD and Blu-ray.

So we do have several customer engagement in this area, but we will -- it's a little bit -- slightly early in the Blu-ray side to get to talk about it so we will talk more about it in the quarter or two.

Do you expect to see volume ramps this year, or is it more of a 2009 story?

It's more of a 2009.

Calendar 2009.

Yes, calendar, yes.

Thanks.

Operator, next question, please.

Your next question comes from James Schneider with Goldman Sachs.

Please proceed.

Hi.

Good afternoon.

Can you talk about in terms of your July guidance, does that contemplate the full ramp of your enterprise hard drive design wins for the SOCs one customer?

Yes, we have -- yes, we do have -- we're talking about for this year?

Yes, okay for --

Our second quarter.

Second quarter guidance.

Yes, that includes that.

Jim as we have said before, the enterprise drive will most likely ramp over a four to six quarter period.

It starts ramping here in the second quarter.

Okay.

Perfect.

And then secondly, can you talk about in any potential upturn, I think you talked directionally about some of the puts and takes in OpEx, but do you think there is opportunities for further reductions in OpEx from the levels you guided to in Q2?

Currently we're comfortable with the guidance that we have spoken earlier on this call for the second quarter, and we are all -- we are continuing our reviews of operational efficiencies and looking for opportunities for improvements, but the guidance that we have given you is our most realistic anticipation of the second quarter expenses at this point in time.

Okay.

Great.

And then lastly, in terms of the base processor business can you give us an update on when you expect that to be profitable on the operating line?

I don't think we will even talk about that, okay, but in terms of -- I mean, with the detail, but, okay, I will say it will be quite a while, okay, maybe a matter a year or so before becoming to be materially positive on the whole -- as a whole.

All right.

Thanks very much.

We have Jefferies & Company, your next question comes from Adam Benjamin, please proceed.

Thanks.

You called out the wireless connectivity 802.11n being better than expectations.

Can you comment whether that was up sequentially in the quarter?

And then comment as to your strategy going forward there.

You have historically focused more on the embedded market and been not a significant player on the router and PC end.

So can you comment a little bit about your strategy there?

Okay.

Adam, in that regard, the first part of your question, Jeff has some information at his fingertips.

Jeff, would you kindly respond?

Yes, Adam in general, 802.11n was up sequentially as -- wireless was in general was better than we anticipated going into the quarter.

Yes, also, I want to take the second question.

You're right that we have been focusing in the past heavily on the embed market.

That was the recent way we have devices in the printers, in the cameras, in the cell phones, in the gaming devices, so there was a reason why we were strong there.

But the side effect was that we were not focusing on the PCs.

So, yes, we understood that and we are now looking into also addressing that PC market as well.

When do you expect to have a single chip end solution for the PC market, Sehat?

We are not -- okay.

I don't think we will talk about that, okay.

Okay.

We'll talk it when we are ready.

Okay.

And then just one last question, maybe Jeff, just a follow-up on the enterprise SOC to Seagate.

It's ramping in July as you indicated.

You'd previously talked about finishing the year at 50% of the mix.

Do you think that's still on target?

I think that's still reasonable.

As we said last quarter, Adam, a four to six quarter ramp so assuming by the end of this year about half the ramp is complete.

Right.

It's really not at our requirement.

Yes, I don't think we can even precisely even predict that.

Right.

It's up to the customer's ramp.

Okay.

And then just one more follow-up.

Your competitor there has talked about garnering the next generation wins that would ramp in 2010.

Do you have any comment as it relates to that?

I'll be glad.

Yes.

So our competitor has been talking about this many years.

Okay, every time, every time we have design win, they always talk about their future technology.

So when we were -- today we are -- today they are talking about 45-nanometers solutions, while they have even had a hard time getting the 90-nanometer, and 65-nanometer and 90-nanometer products to work.

To give you an idea, our products today in the desktop consume about half the power as compared to our other competitors.

That translate to much lower operating temperature on the silicon, on the SOCs.

Our competitor device typically has to run around 125 degrees centigrade on the desktop.

Okay.

Imagine, okay, an enterprise that perform -- the speed has to be higher, you need to integrate two micro processors, you have to integrate double the bandwidth, double DVR bandwidth for the [DM] interphase and you have to double the seal fast or fiber channel interfaces.

I couldn't imagine what kind of temperature they are talking -- they are going to talk about.

So and I'm not worried about it.

All right.

So it's fair to assume that those "wins" for 2010 are rather suspect?

They have to lower the power dissipation by a fact of 3X to make the device to be viable device.

Got you.

Thanks a lot, guys.

That's all I have.

I think, operator, we have time for one more question and that will be it for today.

Okay.

Your next question comes from Shawn Webster with JPMorgan Chase.

Please proceed.

Great, thank you for squeezing me in.

On the OpEx front, as far as your guidance, can you give any clarity as to where it will land between SG&A and R&D lines?

Yes, do you have that number?

I would say just generally consistent with the percent allocation over the last two quarters.

So nothing marketably different from what the recent trend.

Okay.

And what was your head count at the end of the quarter?

About 5,300 or so.

5,320, I believe.

About 5,320 at the end of the quarter.

Okay.

And do you expect to continue to add head count over the course of the year?

We are going to be -- we are -- we understand the sensitivity of this, so we're going to be very mindful in terms of only adding the absolutely necessary head counts to fill in any gaps that we have.

Okay.

And then as the Q2 revenue guidance, you were talking about turns of required of about 35%.

Is that about what it was in Q1, your actual turns?

Shawn, I'll take that one.

Actually, it was actually lower turns required in Q1 than in Q2, but a historical seasonal perspective, it's very much in line with historical seasonal backlog percentage in Q2.

Sorry, I'm just trying to get grounded on it.

What were the actual turns in Q1?

I don't know that we have that exact figure for you right here.

Okay.

In Q1, at the last call, we said we were booked to greater than if 70% of the mid point of our guidance when we held the conference call on March 6th.

Right.

Okay.

And orders for Q2, can you talk about which product categories the orders are the strongest, or what you expect to have the best sequential growth for Q2?

I think same as last quarter.

Yes.

Yes, I don't see any major difference compared to last quarter.

Largely a continuation of the trend of the prior quarter.

So the same areas you saw strength in Q1, you expect those to continue to be the strongest?

Yes.

Okay.

Thank you very much.

All right.

Operator, at this time I don't think we have any more time.

We would like to thank everyone for their time and interest in Marvell, and we look forward to speaking to you at our next conference call at the end of August.

Thank you very much.

Thank you, gentlemen.

Thank you.

Ladies and gentlemen, thank you for your attendance in today's conference.

This concludes your presentation.

You may now disconnect.

Good day.