邁威爾科技 (MRVL) 2006 Q3 法說會逐字稿

Good afternoon.

My name is Miles, and I will be your conference facilitator today.

At this time, I would like to welcome everyone to Marvell Technology Group, Limited third-quarter 2006 conference call.

As a reminder, today's conference call is being recorded, November 17, 2005.

All lines have been placed on mute to prevent any background noise.

After the speakers' remarks, there will be a question-and-answer period. (OPERATOR INSTRUCTIONS).

Thank you.

I would now like to turn the conference over to Dr. Sehat Sutardja, Chairman and CEO of Marvell Technology Group.

Sir, please go ahead.

Thank you, Miles.

Welcome, everyone, to our third-quarter fiscal year 2006 conference call.

Weili Dai, Executive Vice President of the Communications and Consumer Business Group, and George Hervey, Vice President of Finance and Chief Financial Officer, are joining me on this call.

Today, I am pleased to report another quarter of record financial results for Marvell.

Q3 revenue increased 34% from the prior year to $426 million.

Also, our sequential Q3 revenue increase of 9% from the prior quarter marked our 32nd consecutive quarter of revenue growth.

We continue to experience strength and growth across all our served markets, and the pace of design activity is accelerating, given our expanding portfolio of products.

With our consistently strong quarterly growth, we are afforded the unique opportunity to more aggressively invest in the research and development of new technologies, which will enable our continued expansion into new, high-volume markets.

With our growing R&D activities, we are developing increasingly sophisticated total system-level solutions combining our world-class analog, mixed-signal, DSP, and embedded microprocessors with our advanced software solution.

By providing our customers with such advanced solutions, we are able to continue and pull ahead of our competitions in our existing markets as well as further drive the adoption of our complete system solution into the new high-volume market opportunities.

As we continue to execute with great focus on our strategies, the breadth and the variety of our customer engagement is rapidly expanding many different end applications and markets.

This growing diversity of our revenue streams is positioning the Company strongly to continue and build upon our track record of consistent long-term profitable growth.

I will elaborate more about our opportunities and our business progress.

But first, I will have George give our Safe Harbor statement and provide more insight into our Q3 fiscal 2006 financial results.

Thank you, Sehat.

Good afternoon, ladies and gentlemen.

I would like to remind all participants that the following dialogue will contain predictions, estimates, and other forward-looking statements covering subjects such as enterprise, consumer, and emerging market trends, competition, customers, suppliers, products and demand, revenue growth, gross margin expectations, operating expenses, other income, accounts receivable, and inventory.

Such statements will be preceded by the words like "expects," "anticipates," "believes," "should," "will," "may," or words with similar import.

These statements include those relating to the pace of our business for fiscal year 2006 and the impact of the continued adoption of our solutions on our revenue growth.

The following factors, among others, could cause actual results to differ materially from those described in the forward-looking statements.

They include the inability to further identify, develop, and achieve success for new products, services, and technologies; increased competition and its effect on pricing, spending, third-party relationships, and revenues, as well as the inability to establish and maintain relationships with commerce, advertising, marketing, and technology providers.

We direct your attention to our annual report on Form 10-K, recent quarterly reports on Form 10-Q, recent current reports on Forms 8-K, and other Securities and Exchange Commission filings, all of which discuss other important risk factors that may affect our business, results of operations, and financial conditions.

Please be reminded that we undertake no obligation to revise or update publicly any forward-looking statements for any reason.

Now, moving to the Q3 financials.

Marvell reports net income and basic and diluted net income per share in accordance with GAAP, and additionally, on a non-GAAP basis referred to as pro forma.

Marvell's management believes the non-GAAP information is useful because it can enhance the understanding of the Company's ongoing economic performance.

And Marvell therefore uses pro forma reporting internally to evaluate and manage the Company's operations.

Marvell has chosen to provide this information to investors to enable them to perform comparisons of operating results in a manner similar to how the Company analyzes its operating results.

Today, we reported that net revenues for the third quarter of fiscal 2006 was a record 426 million, an increase of 34% over the 317.6 million reported for the comparable quarter in fiscal 2005, and a sequential increase of 9.1% in the second quarter of fiscal 2006.

Pro forma net income, which excludes the effect of the amortization and write-off of acquired intangible assets and other, amortization of stock-based compensation and charges related to facility consolidation, was $113.3 million, or $0.36 per share diluted, for the third quarter of fiscal 2006 compared with pro forma net income of 66.6 million, or $0.22 per share diluted, for the third quarter of fiscal 2005.

The 113.3 million pro forma net income for Q3 '06 represents a 70% increase from the Q3 2005 pro forma net income.

Shares used in computing pro forma earnings per share diluted to the third quarter of fiscal 2006 increased to 316.2 million as compared to 300.6 million shares for the third quarter of fiscal 2005.

Net income under generally accepted accounting principles, or GAAP, for the third quarter fiscal 2006 was 93 million, or $0.29 per share diluted compared with a net income under GAAP of 43.6 million, or $0.15 per share diluted for the third quarter of fiscal 2005.

We have provided on our website in the investors section at www.Marvell.com a reconciliation of GAAP net income to pro forma net income for the quarter reported today plus the prior eight quarters.

Now, I'd like to make some additional comments on our Q3 results.

Our Q3 revenue of 426 million was another quarterly record for Marvell, and the 9.1% increase in revenue from Q2 to Q3 reflects strong business momentum and was above the high-end of our guidance of a 7 to 8% sequential increase in quarterly revenue.

The increase in Q3 revenue represents the 32nd consecutive quarter of revenue increase.

On our Q2 call, we mentioned that as we enter the second half of fiscal '06, we should begin to benefit from what is normally a seasonally stronger period.

Q3 seasonality was consistent with our expectations, and continued to be below historical norms.

During Q3, we continue to experience revenue growth from a number of our established products, including storage SOCs for both enterprise and consumer products with particular strength in the 1.8-inch market, and Gigabit Ethernet PHY solutions, specifically PCI Express, for the client market.

Continuing the revenue momentum from Q2, our Prestera switching solutions for the network infrastructures showed significant growth in Q3.

Additionally, our wireless SOCs for consumer products continued to expand into early production on an increasing number of platforms.

This broad contribution shows the strength of our core market positions as well as the growing potential of the consumer market.

Given this momentum, we are encouraged as we enter Q3.

For Q2, our enterprise markets represented approximately 75% of revenue with consumer the balance.

We had three 10% customers in Q3 -- Western Digital, Toshiba, and Samsung.

We continue to operate at a very high level of manufacturing efficiency, as well as working closely with our foundry and back-end assembly test partners to reduce our costs, positioning us to remain competitive in our targeted markets.

We were very pleased with our Q3 gross margin percentage of 53.7%, which was 70 basis points above our Q3 guidance.

This focus on operational excellence has kept our gross margins above our long-term model of 51 to 52%.

As we mentioned in our Q2 call, we are focused on increasing our investments in the development of new products, technologies, and advanced process geometries to keep us in the limited leadership position in our markets.

For Q3, pro forma operating expenses grew 5.4%, which was at the low end of our guidance, and our strong revenue growth again resulted in a reduction of our pro forma operating expense as a percentage of revenue to 25.4%.

That was a 90 basis point decline from Q2 and 60 basis points favorable to our Q3 guidance.

Total headcount increased to 2,233 employees.

For the 17th consecutive quarter, we increased our pro forma operating income percentage.

Again, given the strength of our revenue growth, our increasing gross margin percentage, and lower operating expenses as a percentage of revenue, we increased our operating income percentage by 170 basis points to 28.3%, which is above the high end of our long-term model.

Shares used in computing pro forma net income per share for the third quarter increased to 316 million compared with 313 million for the second quarter of fiscal 2006.

The increase in shares was driven by the use of a higher average stock price for Marvell shares in the treasury stock method calculation.

Our balance sheet continues to remain very strong.

As we continue to increase the level of our profitability, our cash flow generation continues at healthy levels.

During Q3, we generated approximately 68 million in cash and exited Q3 with cash and short-term investments of $915 million.

Q3 DSOs increased to 54 days.

The increase was due to the timing of the end of our quarter occurring prior to the end of the calendar month, when many of our large customers make their payments.

We will continue to experience this timing difference for the next few quarters.

As we discussed in our Q2 call, we felt that our days of inventory were below our normal operating range, and we would be adjusting our production levels going forward, keeping us in the most favorable position to respond to increases in demand for our products.

We began that process during Q2, and expected to see the effect of those actions during the second half of fiscal '06.

As expected, we did see an increase in our days of inventory at the end of Q3, exiting at 68 days, which is in the middle of our normal range of 65 to 70 days.

Now, I would like to turn the call back to Sehat for comments on our business outlook.

Thanks, George.

Over the past few years, we have been expanding the breadth of our technology and solutions to address the requirements of the service provider, enterprise, consumer, and digital home.

Today, we're in a strong position to support the converging technical requirements of these entire ecosystems with our ability to provide total solutions which combines our system-level knowledge with highly advanced silicon and software solutions.

Our total solutions offering are now enjoying very strong success as we are able to offer our customers tremendous synergies and prominent (ph) advantages as applications across all these market segments are increasingly becoming interconnected with the wired and wireless transmission of voice, video, and data.

One major building block of our total solution is our wireless LAN technology.

We continue to be very excited about the strong adoption of our wireless LAN technology into the consumer markets.

During the quarter, we expanded upon our growing list of applications shipping with our embedded wireless LAN.

We're now shipping in volume into cellphones, digital cameras, gaming devices, PDAs, and printers.

During Q4, for the first time, we will be shipping to multiple cellphone manufacturers as we increase the number of tier 1 cellphone customers utilizing our advanced solutions.

With these new cellphone, wireless LAN is now migrating from high-end enterprise phones to mainstream consumer and multimedia phones.

Our embedded wireless LAN solutions are enabling this these new phones to enjoy seamless connectivity across the WiFi networks for high-speed Internet access, home media networking, and voiceover Internet calls.

The ability for these cellphones to be able to handle voice over Internet calls as well as cellular calls will create a tremendous opportunity for our embedded wireless LAN solutions.

Additionally, we're very excited about the growing number of opportunities for our VoIP solutions outside the cellphone.

As the VoIP market begins to take off, we are enjoying initial traction in single-mode WiFi phones as these phones enable inexpensive voice calls over the Internet, with services from providers such as Skype and Vonage.

During this quarter, we will begin volume shipments into a number of difference WiFi single-mode phones with our low-power embedded wireless LAN solutions.

Also, we're seeing very strong design activity for either VoIP applications, including gateways and cordless handset with our new highly integrated voice over IP solutions that we announced last quarter.

Building upon our strong success with embedded consumer wireless LAN, we are now further extending our wireless LAN leadership by accelerating the adoptions of 802.11n.

We're very focused on driving the market's adoption of this new, advanced standard, and fully support the new EWC specification.

This new high-performance standard will improve the performance, range, and features of both fixed and mobile devices as well as expand the total size of the wireless LAN markets, as it opens up many new high-volume usage models for the digital home, such as enabling the transmission of multiple high-definition video streams.

Last month, we announced that we are the world's first company with chipsets that comply with the new EWC standard.

We're very excited about the timing and performance capabilities of our highly integrated solutions.

As we share our solution with our customers, the response has been very positive.

We plan to aggressively leverage our first-to-market position for designs in the enterprise, laptop, and digital home applications.

Additionally, we're working closely with leading handset manufacturers and carriers to provide optimized products for the mobile handset with integrated software and hardware support.

By combining our new 802.11n solution with our expanding portfolio of products, including our gigabit switching, voice over Internet, and gateway solution, we are in a strong position to drive this technology as the wireless backbone of the digital home.

Also, given the throughput of the 802.11n is well in excess of fast Ethernet speeds, this new standard will be yet another technology that further advances the market productions of Gigabit Ethernet technology in the home.

In regards to the Gigabit Ethernet market, we continue to enjoy strong growth during the quarter from the ongoing market transitions from fast to Gigabit Ethernet.

On the client side, our PCI Express gigabit products shipping Intel are experiencing strong growth for the PC, laptop, and server applications as these solutions continue to ramp (ph) many new designs into production.

In the infrastructure market, we are also seeing tremendous growth, as the Marvell total solution, which combines both software and silicon, is gaining rapid market acceptance ranging from sophisticated metro equipment to high-volume applications used by small businesses and the digital home.

We are particularly excited to see the adoption of Ethernet technologies in the metro infrastructure is clearly beginning to take place.

While competitors begin to announce their initial silicon for the metro, we have been shipping our complete system-level solutions, which meets the most stringent quality-of-service requirements of the service providers for the past four years.

Today, across the metro, enterprise, small and medium business, and digital home, we are enabling our customers to develop equipment with the highest levels of future integration and ease-of-use with the ability to combine switching, routing, security, voice, and wireless capabilities all in one system.

Now moving to the hard disk drive market, we continue to enjoy solid growth from the strong adoption of high capacity storage in a growing number of consumer and mobile devices.

We also remain focused on continuing to expand our market share within the hard disk drive market through additional design wins for our storage SOCs as well as increasing our silicon content within the hard drives with our preamplifiers, motor controllers, power management, and our newly acquired enterprise hard disk drive controllers.

Earlier this month, we were pleased to close the purchase of the QLogic hard disk drive and tape drive controller business.

We're also very excited to have this group of very talented employees joined the Marvell family as part of this purchase.

Now with these established operations in southern California, we look to grow this site into another large design (ph) center for Marvell by hiring additional talented engineers from the local area to help fuel our aggressive research and development activities on a broad range of products.

I would now like to turn the call back to George for additional comments regarding our financials and guidance for the next quarter and --

Thanks, Sehat.

We are pleased with the consistent performance of our business.

And the large number of the emerging market opportunities has positioned Marvell to achieve another record year.

We continue to make significant progress in penetrating a large consumer opportunity that we are addressing, with products utilizing our analog, mixed-signal technology at a high level of integration.

We believe that we are on track to achieve a revenue balance of 50/50 between the enterprise and consumer end markets by the end of calendar 2007.

Additionally, our entering into other large markets such as digital power management, optical storage, and voice over IP will continue to provide us with an increasing TAM to sustain our long-term revenue growth.

Before getting into the detailed Q4 guidance, I would like to update you on the purchase of QLogic's hard disk drive controller business.

We completed the purchase on November 4, 2005, and our Q4 guidance and actual results will reflect the partial quarter.

As we indicated at the time of announcing the transaction, we expected the purchase to be immediately accretive and have a positive impact on all of the major metrics of our income statement.

Now moving to the Q4 guidance, our business momentum in Q3 was quite strong.

And we remain encouraged by the continuing order patterns from our customers resulting in good visibility entering Q4.

As I mentioned in my Q3 commentary, we experienced a number of positive trends that resulted in good breadth of products in end markets contributing to our Q3 growth.

These trends continue into Q4, and we are excited about the opportunities.

Considering the above, we have targeted Q4 fiscal '06 organic revenue to increase approximately 8% from Q3 fiscal '06.

Additionally, we expect the Q4 partial quarter revenue dollar contribution from the hard disk drive controller products to be 25 to $28 million.

Combining the organic growth and the hard disk controller products, total revenue growth from Q3 should be 13 to 15%.

At the midpoint of the Q4 guidance, total fiscal '06 revenue would increase approximately 36% from fiscal '05.

Continuing the trend from the prior quarters, and the positive effect of the hard disk drive controller products, we expect our Q4 gross margin percentage to increase by 70 basis points to 54.4%.

For Q4, operating expenses will increase approximately 12%.

The larger-than-normal quarterly increase is the result of including the operating expenses for the hard disk drive controller business.

Operating leverage should continue in Q4, with operating income percentage increasing by approximately 100 basis points.

We expect interest income to decline by approximately $500,000, reflecting the cash payment to QLogic for the hard disk drive controller business.

Our balance sheet indices should be consistent with the last several quarters.

And shares used in computing pro forma net income should increase to 322 million.

I'd like to turn the call back to Sehat.

Thank you, George.

That completes our commentary.

Miles, would you please poll for questions?

(OPERATOR INSTRUCTIONS).

Cody Acree, Legg Mason.

Thanks, guys, and congratulations.

Same question as last quarter, George -- is it now time to raise your long-term gross margin targets above that 51 to 52% range?

Well, Cody, I have to give you my consistent answer, which is no. (laughter) We're obviously in a very good position at the moment with a lot of positive things happening, including the new business that we just purchased.

We'll look at that and probably give you an update when we report Q4.

But obviously, given where we are at right now, we are operating significantly above the model.

And we'll give you an update on the Q4 call.

And George, just a follow-up there -- with the hard drive controller division now, what do we look for for growth of this division?

Is it something that we have to wait for an integrated SOC to come about before we see much growth, or is this something that will grow organically?

Well, you know, again, we're not going to break out that -- we've given you the starting point.

And that will be the last time that we give you any specific detail revenue about that business.

We don't have it for a full quarter, because we didn't close it as of the first day of the quarter.

But that business does have growth prospects.

And we're looking forward to achieving those next year.

And of course, the long-term objective is to go to integration.

And it will take about at least a year for us to make any changes in that roadmap for that business to begin with.

Ambrish Srivastava, Harris Nesbitt.

Question on WiFi-enabled handsets.

I guess you guys have talked about this market and -- myself included has been pretty skeptical about it.

But a little bit more clarity -- are you guys designed into the Nokia N80 that was recently released?

And then I have a follow-up.

The answer is yes.

Okay.

My follow-up on the gross margin, George -- is it a combination, or what is the main driving force here?

Is it the investment you guys made a few quarters ago?

Because there's not a lot of capacity out there.

So is it a combination of that and some design node related -- or what exactly is guiding this upward trajectory?

Similar, Ambrish, to the last quarter.

We know that we just began to see the effects of our longer-term fab strategy coming into play in Q2.

And that continued also into Q3, and will continue going forward for a number of years now.

So I would probably say that roughly half the improvement comes from that.

The other half is a combination, again, of our excellent yield performance coupled with, as we said, the move to the network infrastructure starting to gain momentum -- by its nature of the end market is a pretty profitable business for us.

So there's a little bit of a mix shift going on there as well.

Just one last one, and I'll run away after that (multiple speakers) For Sehat, this is more on the wire LAN side.

What's your sense of where we are in the mix between embedded and not embedded?

And given the fact that you guys have probably the best design-in in this segment, what is your sense of how the market would look like a year from now?

Well, it will be hard thing to predict -- to try to come up with any kind of (ph) precision about when the market -- the embeddeds are going to be significantly bigger than the PC markets.

But we do believe that the embedded markets in the long-term is going to -- by far will be the major markets for wireless LAN.

And we've been working on this for -- obviously, as you know for the last three, four, five years, addressing specifically the embedded market.

We believe probably by sometimes the middle of next year or so, second half next year, the embedded markets will be bigger than the PC market.

Michael Masdea, CSFB.

Thanks a lot.

I guess, Sehat, as you're getting bigger and bigger and you're adding on more and more growth, we've seen a couple acquisitions from you guys.

But the question really is do you need to start doing more acquisitions to fund the growth, or can you handle it from your own internal resources?

Yes, this is something that we obviously ask ourselves all the time.

What the trade-off is in organic growth and acquisitions.

We've been -- in general, we always made an assumption that we want to grow organically as much as we could.

We do have product -- R&D developments, four or five years -- as of today, we have -- R&D is there -- reach all of the way four, five years or even more farther away than that.

But -- a think that we also continue to look at the option of acquiring technologies from some startups out there.

And we look at that on a continued basis.

We have bankers that talk to us all of the time.

And unfortunately, you don't see too many of those things to be attractives (ph) for the last several years, because when we compare against our internal development, a lot of times, our internal developments are ahead compared to some of the developments by these startups.

But you never know, okay?

So maybe some of them are ahead of us.

And we will quickly move to take advantage (ph) -- to take that opportunity down the road.

I guess that brings up the issue also of -- as you get bigger and bigger, and potentially do you some M&A like we just saw, how do you keep the organization kind of feeling like the Marvell that everybody is attracted to?

Are there changes you make to the organizational structure?

Or what's the key to keeping that together?

My concern is actually to -- in terms of (ph) internal organization is to give challenging projects to my guys.

And as long as they have challenging projects, whether things are structured in one division or another division, I don't think there will be an issue at all to our guys.

So more important to us is really to have exciting projects, and in large market opportunities for our people.

And also, I think I'd like to add on top of what Sehat said, in the last 10 years, based on Company's growth and how we adjust scalability so far as being -- doing very well.

And of course, we will continue with the recipe.

And we expect to continue to do very well and understand how to continue to scale the Company.

David Wu, Global Crown Capital.

My congratulations for a very well-executed quarter, again.

I was wondering, Sehat, you mentioned about 802.11n.

And some of the things that I believe you folks are working on relate to cellphones.

And can you give us a perspective of how you see the world of -- I guess this popular thing is called (ph) -- where does NAND stop and these small drives start?

And the other side is how do you -- do you need -- do you feel you need to bundle baseband with power management with your wireless LAN and Bluetooth -- all in these so-called superchips in the next several years?

(multiple speakers) A lot of questions there.

I get the important part is that -- out of those questions, my understanding is -- the 11n portion is probably the biggest part of the questions.

We believe the fact that the digital home is going to be a reality, the adoptions of networks in the home is an important target for us to achieve.

And 11n is the premiere of -- it's the leading technology that will be adopted in the home for the backbone connectivity of all the equipment -- the TVs and the HD DVD players, the music players inside the home.

So 11n is a very important technology for the home.

At the same time, once you develop this technology, you also want to have the technology to be adopted, not just for the consumers, but also for the laptops and for the cellphones.

But the driving applications will be initially just for the home, and then moving to the cellphones quickly after that.

The other one was the NAND versus small --

Okay, the other one is the NAND versus small form factor.

I guess NAND -- we're referring to the flash memory. (multiple speakers) So again, I'm very consistent with my answer to this.

I always say from many, many years that there are always applications for both for the NAND flash, as well as the small form factor storage.

I always say that if the requirements are sufficiently handled by flash, flash will be the major solution.

But for those applications that requires more capacity for the same price, the micro storage, small form factor storage will be the dominant solution.

So anything -- basically, what it means that anything requires video storage, lots of music capacity -- multimedia storage, pictures, will always be mainly addressed by micro storage.

Now we're talking about, we're talking about down the road will be several hundred million units of opportunities for the micro storage, if not that they -- initial (ph) now could be one billion units of those applications.

But in the meantime, there is some confusion in market, which always happen whenever two competing technologies trying to address the same market.

But we are always focused -- we're focused on making sure that our micro storage business will continue to grow.

Anything on cellphones?

Yes, micro storage is applicable for cellphones as well.

Sorry;

I was speaking about other (ph) chips that you might offer on cellphones like (multiple speakers)

We have chips for wireless LAN for cellphone.

But that is as much as I can talk about at this point.

Srini Pajjuri, Merrill Lynch.

Just a quick question on the QLogic business.

It looks like they reported about 88 (ph) million last quarter.

And I understand it's only a partial quarter.

But I was wondering, is there any seasonality to this business that we need to look at?

There is a little bit, Srini.

Normally, Q4 is the weakest quarter historically for them.

Again, we have to sort of at this point just listen to what they told us about their business.

We tend not to see as much swings in our own business.

And one of the things we hope to do moving forward is to make this thing much, much more consistent, typical Marvell type of performance.

But historically, it has had some seasonal effects in a predominately -- that normally occur in the fourth quarter.

Okay, and George, as we look into the next few quarters, do you expect to -- any change to the expense structure for this business in terms of adding some back office or maybe taking some back office cost out of the system?

No.

We basically purchased engineering assets here.

We have a very good infrastructure already at Marvell.

So it was a fairly seamless moving of the products from QLogic to ourselves.

So there was just a nominal number of sort of G&A-type folks that came with the purchase.

So no, we're not anticipating needing to do anything.

We've actually done everything we believe we need to do.

What we need to do is step up the R&D investment there moving forward.

Fair enough.

And then just one question on the GigE for the client side -- it looks like it has been growing pretty strongly, and it looks like it's been gaining share.

Just wondering, how long do you think these share gains will continue?

Or do you expect the growth to moderate and trend in line with the seasonal PC spend as we look into Q1 and Q2 of next year?

Well, in terms of gigabit, actually, we've been talking the last few years.

I think the gigabit in terms of cost shipment (ph) is still continuing to grow.

And we see another segment going to the consumer in the past as cost (ph) PC.

Now, the consumer platforms also -- some of the high-end also converting to the gigabit.

So we believe that business will continue to do well.

As I think you're probably aware, Srini, we're addressing this market in two ways.

One, obviously, through our part of the ship with Intel, which we already -- Sehat already made commentary on the strength of that, as well as our Yukon products, which are our own branded products that we are selling.

So we have a dual sort of product attack there.

And both are doing well right now, as (multiple speakers) already said.

Jim Liang, SG Cowen.

Can you give us an update on the optical storage business in terms of traction with potential customers -- kind of your thinking here in terms of timing of revenue ramp and any gating factors -- what are the gating factors for the revenue ramp?

Sure.

So the development of the optical storage products has been going on very, very well.

We have been engaging with several early customer engagements.

A lot of activity -- a lot of work that's been -- that we're certainly working on with all of these early customer engagements, mainly in the areas of software development.

We have a lot of softwares need (ph) to be completed.

The silicon is working extremely well.

So we expect that once we finish all of the software -- the funding (ph) software development, we will see initial revenue.

So my guess is by second half of calendar '06, will be the time that we will see this revenue becoming more material.

So is there any gaining factor from a timing of your customers introducing new platforms?

Or is this -- they can just swap in your silicon, and you can start to ramp into their existing platform? (multiple speakers)

There will always be some of those.

But most of the gating items really is getting all of the firmware to be completed and tested again -- against all these things -- system out there, and there has been in the market.

Remember, there are so many chipsets out there that we must make sure that we are fully compliant with.

So really, it's time to -- going through all of the grunt work similar to any product that we built in the past.

So nothing that didn't -- that we know of that didn't surprises at this point.

I just have a follow-up question.

Sehat, you mentioned earlier on metro infrastructure or metro Ethernet.

Can you just give us some more color as far as the drivers of growth in-market and sort of your view of timing of a significant volume ramp in that market?

Yes, we always know that Ethernet is always going to be the higher-volume business compared to any other communication network technologies.

So sooner or later, Ethernet will get into the -- inside the core of the networks -- in this case, the metro networks.

So we've been seeing these activities for the last four years.

And we are seeing that more and more of our customers are deploying Ethernet in their core infrastructures.

Our belief is that the adoptions off in the future about (ph) video, VDSL2, 3G wireless, voice over IP, will push more and more of the service providers, the carriers to adopt the metro Ethernet because of the significant cost savings to them.

So do you see that as more of a -- kind of an '06 are calendar 2007 type of --?

The way we look at it is more of a -- continued adoptions of Ethernet as more and more people see the success of Ethernet, as more and more IT managers seeing that Ethernet is just as good as the SONET technology, but at much, much lower-cost, more and more people are going to just adopt this -- there won't be this hockey stick type of deployment.

It will just be -- for the next five years, you will see the continued market share gain of Ethernet.

Arnab Chanda, Lehman Brothers.

Anything what we could is said a lot louder, I guess, by your revenue performance.

So maybe I could just ask a couple of questions.

One, Sehat or Weili, if you could just address a little bit -- you were talking about your voice over IP business.

Is that separate from your wireless LAN business?

And you were also talking about your 802.11n offerings.

It seems like the specs are not set there yet.

So when do we think that you could see revenues from those two businesses -- if you could talk a little bit about that?

VoIP is an application that fits on top of our wireless LAN solutions.

Just like -- well, let me go back a little bit there.

We've been thinking all along that our wireless LAN solutions are a lot more sophisticated than the traditional wireless LAN solution.

And the reason for that is because we have a complete CPU subsystem integrated with our wireless LAN solutions.

It was designed that way because from the get-go, we wanted our wireless LAN solutions to be able to go easily into consumer devices where they do not have powerful postprocessors -- like PC processors.

And because of that, because our wireless LAN (ph) solution is really a system solution, a complete system solution, we were able to quickly deploy a VoIP application simply by just porting some software onto our wireless LAN chipset.

So this is just an example of how -- why we could quickly get into the VoIP market in such a short time.

Now, regarding 11n, 11n is -- yes, there's a lot of I think debate in the standard committees about how to finalize the 11n.

The debate has been going on for too long.

It's to the point where we felt -- ourselves as well as a few of our competitors in the wireless LAN business felt that we need to do something to accelerate -- to put feet on the ground and say, enough is enough.

And we need to deliver the 11n solution to the market, because the demand is there in the market.

But too many people -- too many cooks in the kitchen wanted to have a say on what 11n is gone to look like.

So this is what the EWC committee has been trying to do is trying to come up with solutions that will accelerate the 11n adoption, the de facto 11n solution to the market.

So we are ready basically.

If there's any changes beyond our expectations, we could quickly adopt those changes and update our solutions.

Just one last follow-up, please.

One of the other areas that I think Ethernet has potential to go into is storage.

You obviously talked about the opportunities in metro.

Could you talk a little bit about if you are involved with iSCSI -- since you are obviously very strong, now you have a full portfolio in the -- maybe in the drive side.

Could you describe that a little bit?

And if so, when should we see that in your revenues?

Thank you.

Well, I don't want to talk about things that appear that that we're not ready to talk yet.

Obviously, we do believe storage and Ethernet are synergistic.

So we do have some activities in these areas.

But we want to delay -- yes, the specific discussion -- maybe on the next quarter conference call, maybe.

Adam Benjamin, Jefferies.

Can you talk a little bit about -- as we look at handsets and the adoption of different technologies within handsets, it appears that Bluetooth is gaining the initial traction ahead of, say, wireless LAN or WiFi.

As we move forward here, how do you see that playing into the dynamic with your design win traction in the handset as to whether that's a technology you need to have or how you see that playing out over time?

Okay.

The way we look at it, Bluetooth and WiFi will always be -- will coexist in the cellphone for a long time.

Bluetooth will be used mainly for very, very short-range accessories types of connectivity, meaning like cellphones to ES (ph) headset, while WiFi will be used for connectivity to the network.

That's from the application side.

From the silicon side, the integration of Bluetooth with WiFi will increase the size of the WiFi by like 10% or so.

So it's a minor increase in cost in our -- by having to adopt Bluetooth into our WiFi silicon.

So in the long run, both of them will be supported.

And just a follow-up on the optical storage.

You talked about material revenue in the second half.

Can you quantify that a little bit more specifically?

We didn't say material.

We said small revenue in the second half, Adam --

We may have said that, but let's call it --

It's going to be the beginning -- initial -- let's put it this way.

It will be initial revenue in the second half of next year.

And we're not yet ready to give a quantification of that, just because, again, there's still a lot of development going on.

And we want to make sure we have some good information before we share that.

So no revenue until calendar Q3 of next year?

That would be a good assumption.

Bill Lewis, J.P. Morgan.

I'd like to ask about the storage business.

It appears that the disk drive customers, particularly in enterprise and desktop, are seeing a little bit worse than normal seasonality right now.

Could you give a view on what you are expecting in your storage business in this current quarter?

And then, is there an opportunity in your April quarter, maybe given some of the weakness in those market in the second half that's actually a little better than seasonal?

Well, as I kind of said in my prepared comments, 2005 is playing out to be a year where seasonality, which does exist, is below historical norms.

It was less downward pressure in the first half of the year, and the drive guys spoke to that on their individual reports.

And then, as we moved into Q3, they also commented that while yes, they had a positive seasonal impact, it was less again than maybe what the market had traditionally seen.

And of course, we always take a very conservative view of this.

So in our expectations, we are always taking a very conservative view of seasonality.

And in this case, it would be positive seasonality.

So I don't think we have an exposure -- not an exposure, but I think we are consistent with what we believe at this point the fourth quarter would look like.

If this is true and plays out that way, then it could -- and no one knows yet.

But it could certainly make Q1 maybe again a less than historical seasonal pattern.

Okay, and the second question, if I could -- recently, your prime competitor in semiconductors for disk drives made some significant management changes.

Have you heard any response?

Or what has been the response from your customers given this change?

Are they concerned or not?

And what actions at Marvell are you considering -- maybe both to take advantage of any potential disruption if this management change precedes some sort of more structural change or divestiture at that competitor, and are there any defensive moves that you are considering?

I think it's inappropriate for us to make any comments about our competitors' problem with their management or internal issues that they have.

It's important then for us to continue -- regardless of if they have a problem are not to continue to work harder to develop better products so that we can continue to gain market share against our competitor.

So I think that's as much as we could talk about.

It's a continuation of what we've done for the last five years.

So why would we change now?

We're in a very good position.

Seogju Lee, Goldman Sachs.

Thanks.

George, you talked about seasonality a little bit.

I just wanted to explore the new products and some of the emerging growth drivers that you have had like WiFi into the consumer devices.

What sort of penetration rates do you think you are at currently?

And how do you think those progress as we go through the balance of this year and through next year?

Thanks.

Well, I think one of the most exciting things we've been doing stock (ph) -- and we've talked a lot about this when we were -- and in public forums and so forth, is just the vast number of platforms that our embedded wireless LAN is being designed into.

And I think in Sehat's prepared comments earlier, he talked about half a dozen different major categories of consumer products that we are now shipping volume production to.

So our view is that where we were in the middle of the year as far as platforms in production is going to increase severalfold to the number of platforms that are in production as we exit Q4.

And they are in those areas that Sehat described such as cellphones and cameras and games and printers and PDAs.

We have multiple design wins in each of those categories, and, say, every month or so, more and more of the customers we're engaged with are bringing their products to market.

That doesn't stop at the end of Q4.

That is something that we still believe we're in the fairly early stage of the adoption of embedded wireless LAN into these types of products.

So we consistently list this as the top growth drivers in revenue opportunity for the Company for '05 and certainly would be the case in '06.

Sandy Harrison, Pacific Growth Equity.

Real quickly, could you perhaps give us kind of an update on how your ARM strategy is going -- it's something you've talked about in prior calls and something that's been an exciting opportunity as you built it into some the other products and so forth.

Just kind of, Sehat (ph), if you could, give us an update on some of the things you're doing there and some of your success stories?

Sure.

I guess the question is related to the Feroceon ARM embedded processors that we have been developing over the last several years in-house.

So just to recap, this is -- this is a processor that we developed under the ARM architecture license that we have with ARM.

So we would -- I think that we would -- second company in the world that has developed the arm processor in-house.

So it's the -- our strategy is to develop the embedded ARM processor in-house to support all the different varieties of consumer products that we have in the Company, from wireless LAN, storage, any other embedded -- printers, anything that requires high performance processor embedded into the products.

So we announced several products that we have developed and we continue to develop additional families of processors to address different market segments from ultra, ultra-low power and high performance to high performance where power is not a concern.

So this strategy is an important (ph) for us as we get further into the consumer space.

Is there some other products or some other processor technologies out there that would be complementary to you ARM (ph) that you're looking at?

Or is ARM basically the platform of choice and the one which you will continue to work with?

There will always be something that we'll be using.

But ARM will be the major -- majority of our platform base, because by doing so, we could leverage the software R&D development that we have done in-house to all the different platforms, as well as leveraging all of the softwares that have been -- that are available out there in the market -- dropped (ph) by hundreds of different software providers out there that already support ARM as their major platform.

Alex Gauna, UBS.

Sehat, you mentioned the increased momentum behind converged cellular WiFi for you.

I'm only at this point aware of your opportunity at NTT DoCoMo.

Is there any way you can give us some color on some of the carriers that you are employed with or some of the geographies?

And also, would you characterize these opportunities right now as still trial-stage, or have they really started to migrate into volume production ramp?

This is George.

As we indicated already, we are now shipping production volumes to multiple cellphone manufacturers.

Probably the one that is most known or noticeable -- our awareness is with Motorola, which is the DoCoMo that you're referring to, although there are multiple Motorola models now that are in -- that have our WiFi in them.

And then we did -- in response to Ambrish's question indicated Nokia as another one.

There are others.

But those are the only two that we are prepared to speak to.

But clearly, our WiFi adoption in cellphones has moved well beyond just the qualification phase.

We're now in production.

And has it moved beyond NTT DoCoMo, or is that the principal --

No, there is -- like I said, each phone -- now there are quite a few -- have a different service provider attached to them.

And again, can you give us some help on those service providers at present?

I think it's more you need to talk to the cellphone manufacturer.

Okay, and then in the first half of calendar 2006, from what you are seeing, is the larger opportunity in converged cellular WiFi, or is it in voice over IP -- home network to WiFi with the telephone opportunity?

I think they're both large (multiple speakers).

We're not going to -- you know us.

We don't get to that level of detail.

They are both exciting opportunities, Alex.

We're well positioned in both.

Both are important.

We are not so concerned about which ones are going to come up ahead compared to the others.

They both will be equally large.

Now of course, the cellphone will be even bigger.

At the end (multiple speakers), it will be bigger.

Yes, it will, but the adoption rate has to take hold there before that can turn into those kinds of -- we still, Alex, consider embedded wireless LAN to be the single largest growth driver for the Company going forward.

Okay.

And within the hard disk drive segment, there are a lot of evolutions -- SATA1 going to SATA2, SCSI going to SAS, and fiber channel still in the mix.

Is it a "don't care" for Marvell which -- what the enterprises go with?

Or would you prefer to see SATA2 be more successful or fiber channel?

Is there relative positioning within hard disk drives for you?

I know you are in all of them.

Yes, we are actually neutral on the interface side.

We build all the interface technologies that our customers are requiring.

So we do not dictate the evolutions of the storage interface, because -- to tell you the truth, it has nothing to do it technology.

It's more of market positioning by our customers.

And they have their own strategies and we just have to follow their strategies.

Okay, and lastly, if I could, you mentioned Western, Toshiba and Samsung as greater than 10% customers.

And we get -- I doubt you'll break them down individually.

Can we get the aggregate of what they represented as a percentage of sales?

No, we don't do that.

But clearly, they're at least 30%, right?

Fair enough.

Congratulations.

Powerful quarter.

Ross Seymore, Deutsche Bank.

Thanks, guys, and congrats, as everyone else has said.

Just a little more depth on the QLogic acquisition.

George, in the 70 basis point increase in gross margin going forward, roughly speaking, what's organic versus that that the QLogic acquisition brings?

Again, we gave the breakout on the revenue just to give you guys a starting point.

Beyond that, we're not going to get into the detail of -- we said it was accretive, and that it would be -- have a positive impact on all of the major metrics.

So clearly, some portion of the 70 basis points is attributable to that.

And just to help you a little bit, it's an enterprise kind of product.

So you can kind of speculate a little bit about the kind of gross margins that enterprise products have.

Okay.

And then similar sort of thing on the tax rate.

Can the QLogic revenues get taxed at the same rate as the 10% that you guys get?

They shouldn't have any impact at all, no.

We're going to stick with our 10% going forward.

Got you.

And then the last question just more for Sehat as far as strategic decisions going forward.

When you enter into brand new segments, is that the time when you typically have the sales and general and administrative category increase?

It seems like you guys have done a great job leveraging existing technologies.

But I'm just trying to see when there might be some stairstep action on the OpEx side of things, if ever.

We're trying to -- as Weili likes to use the term, we're on a scalability plan here at Marvell as we go through the billion -- heading to 2 billion and then on.

So we have to scale the entire company to address -- becoming a much larger concern.

And that is not just in the engineering, which obviously is our focus.

But we have to do that across every function in the Company.

So we hopefully won't have to do any stairstep at any one point in time, then we'll be able to sort of do it seamlessly in growth over the next couple of years.

But our focus is everywhere in the Company.

And to address the timing of how we scale, that's leverages (ph) and efficiency (ph).

So that's I think where we have done a good job in the last ten years, as I mentioned earlier.

And a lot of these organically developed new markets do take time.

So typically, it takes three, four, sometimes even five years for us to see the significant revenues from these new activities, because we have to do -- we have to develop this thing organically.

And of course, the easiest -- to have instant revenue increase is just to acquire a bunch of companies all the time.

But we do not believe that as a general practice, because we do want to make sure that we get our employees exciting projects in-house.

We trust -- okay, we bring the best of the best people.

And we want them to be proud of what they are working on.

So we cannot just offer (ph) profitability (ph) and then continue to out of the blue, just acquire companies and say hey, what they are -- what internal has been developing is not so important after all.

So we don't do things like that.

So we have got to take the hard routes.

And hopefully, and then we'll build a more solid organization by doing so.

What you're doing currently is working very well.

So I would continue doing that.

The last question is more on the COGS side of the things for George.

Quarter over quarter, was there anything going on in the back end that made the costs rise?

I know that some of your competitors have talked about the prices increasing and the capacity tightening.

Is that something that was a little bit of a drag that could actually be an opportunity going forward as capacity comes on back there?

Well, actually they do -- there are some --

(multiple speakers) costs -- we are not immune to that.

We're a major player, obviously, now in that whole segment.

But we are able to -- so yes, we are seeing some nominal increases on some things.

But we are more than able to deal with that as obviously post --

The more higher value-added products compensate for that.

Right.

So I think, Ross, if in the -- once the substrate issue gets resolved, and maybe there's an opportunity to reduce costs, we should benefit from it.

But I think at this point, we're pretty comfortable with where we are at.

Ladies and gentlemen, we have almost reached the end of the a lot of time for our Q&A session today. (OPERATOR INSTRUCTIONS) Shaw Wu, American Technology Research.

Thanks.

I just want to revisit Bluetooth.

So far, you have been able to make these big tier one embedded WiFi with tier one cellphone manufactures.

Just wondering, strategically, is Bluetooth a business you have to be in?

Yes, as I said in a previous question -- so we believe Bluetooth will coexist with WiFi.

And I also mentioned that the Bluetooth integration into WiFi silicon is not a big deal.

I said that of course (ph), that maybe a little bit more than 10% now.

But you get the idea -- it's going to be small enough that we are able to -- it makes sense to do so.

So, we were not a Bluetooth player in the past.

But because now we are a major player in WiFi, in the cellphones, because Bluetooth will always be there, we will be a player in Bluetooth in the cellphone as well -- but on a stand-alone basis, because we do not believe a stand-alone Bluetooth will exist for too long.

Yes, or another way of saying it is one should not assume that we don't have Bluetooth.

Okay, do you need to make any acquisitions in that space?

Or do you believe your internal expertise is adequate?

No.

We basically will do it internally.

That does conclude our Q&A session for this afternoon.

I will now turn the call back over to management for closing remarks.

Thank you, everyone.

This complete our Q3 fiscal 2006 conference call.

Looking forward to updating you next quarter.

Thank you very much.

Ladies and gentlemen, thank you for joining us today.

This does conclude our conference call for this afternoon.

You may now disconnect.