慧榮科技 (SIMO) 2016 Q2 法說會逐字稿

Ladies and gentlemen, thank you for standing by, welcome to the Silicon Motion Technology Corporation quarter two 2016 earnings conference call. (Operator Instructions). I must advise you that this conference is being recorded today, Friday, July 22, 2016.

This conference call contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933 and Section 21E of the [Securities Actions] Act of 1934 as amended. Such forward-looking statements include, without limitation, statements regarding trends in the semiconductor industry and our future results of operations, financial conditions and business prospects.

Although such statements are based on our own information and information from other sources we believe to be reliable, you should not place undue reliance on them. These statements involve risks and uncertainties and actual market trends and our results may differ materially from those expressed or implied in these forward-looking statements for a variety of reasons.

Potential risks and uncertainties include, but are not limited to: continued competitive pressure in the semiconductor industry and the effect of such pressure on prices; unpredictable changes in technology and consumer demand for multimedia consumer electronics; the state of and any change in our relationship with our major customers; and changes in political, economic, legal and social conditions in Taiwan.

For additional discussions of these risks and uncertainties and other factors, please see the documents we file from time to time with the Securities & Exchange Commission. We assume no obligation to update any forward-looking statements which apply only as of the date of this conference call.

I would now like to hand the conference over to your first speaker today, Mr. Jason Tsai, Senior Director of Investor Relations & Strategy. Thank you. Sir, please go ahead.

Thank you, and good morning, everyone. Welcome to Silicon Motion's second-quarter 2016 financial results conference call and webcast. My name is Jason Tsai and with me here is Wallace Kou, our President and CEO, and Riyadh Lai, our Chief Financial Officer.

The agenda for today is as follows, Wallace will start with a review of some of our business highlights, Riyadh will then discuss our second-quarter financial results and provide our outlook. We will then conclude with Q&A.

Before we get started I would like to remind you of our Safe Harbor policy which was read at the start of this call. For a comprehensive overview of the risks involved in investing in our securities, please refer to our filings with the US SEC.

For more details on our financial results please refer to our press release which was filed on Form 6-K after the close of market yesterday. This webcast will be available for replay on our website, www.siliconmotion.com, for a limited time.

To enhance investors' understanding of our ongoing economic performance we will discuss non-GAAP information during this call. We use non-GAAP financial measures internally to evaluate and manage our operations.

We have therefore chosen to provide this information to provide you -- to enable you to perform comparisons of our operating results in a manner similar to how we analyze our own operating results.

The reconciliation of GAAP to non-GAAP financial data can be found in our earnings release issued yesterday. We ask that you review it in conjunction with this call. With that I will turn the call over to Wallace.

Thank you, Jason. Hello, everyone, and thank you for joining our earnings call. I am pleased to be speaking with you again and to report another great quarter for Silicon Motion.

In the second quarter revenue increased 25% sequentially to $141 million. Our [sixth] consecutive quarter of revenue growth. The trends in the second quarter came from all segments of our business, especially from our [client SSD] controllers. Our eMMC controller and SD solution also exceeded our expectations.

Riyadh will go into more detail about our financial performance later on the call. Before I update everyone about our business let me share news of a major milestone that we have recently achieved. We are proud to announce that over the last 10 years we have shipped over 5 billion controllers cumulatively, more than any other company in the world.

5 billion controllers is equivalent to 70% of the world population using our microSD, eMMC, SSD or other type of NAND flash (inaudible) [storage] devices managed by one of our controllers. In 2015 the number of controllers that we shipped is equivalent to 10% of the world population using a storage device powered by our controller.

Let me now update everyone about key products starting with our client SSD controllers. Our client SSD controller sales increased over 30% sequentially this quarter led by very strong sales to our three NAND flash partners. Sales to these three customers exceeded expectations, increased over 50% sequentially as our flash partners' sales of SSD continue to accelerate with PC OEMs adopting more cost-effective TLC SSD.

This strong client SSD controller sales growth will continue in Q3. For the full year we are now expecting our client SSD controller sales to increase by up to 150%, approximately $60 million in 2015.

This is a significant forecast upgrade as we were previously expecting just 100% growth. This year our client SSD controller and market share will likely increase to somewhere between 30% to 35% from roughly 25% last year as a result of rapid sales growth.

As we had discussed last quarter, PC OEMs were previously hesitant about using SSD base on TLC flash because TLC flash have higher error rates, lower endurance and so are harder to manage and use than MLC flash and SSD. But SSD using TLC flash are roughly 20% to 25% cheaper than SSD using MLC flash. So they have a significant cost advantage.

Earlier this year, after completing general evaluation of TLC SSD, PC OEMs were willing to adopt TLC [SSD] because of improving performance and reliability enabled by our controllers. In the second quarter our flash partners new TLC SSD passed technical qualification faster and smoother than original plans.

This led to early orders and larger orders of our SSD controller driving continued upward revision of our SSD sales forecast this year. The sales volume of client SSD controller that we are now expecting for this year is significantly higher than our own forecast under optimistic scenario at the start of the year.

PCs, both notebook and desktop, are rapidly transitioning from HDD to SSD. The [acceleration] demands for client SSD [mirrors] the deceleration demand for client HDD reported by the HDD vendors. We expect that this year roughly 35% of all the PCs will be using SSD, up from roughly 25% last year.

(Inaudible) variability of TLC SSD, which are higher performance and cost competitive with HDD, is an important factor in driving faster SSD adoption this year. Already almost half of our client SSD controllers are for managing TLC flash and this proportion will increase as this year progresses.

We believe SSD adoption will increase [further] next year when the supply of even lower cost 3D NAND capacity increases. We started initial sales of our 3D SSD controllers to our flash partners recently and expect this controller to be a meaningful part of our client SSD controller sales next year as our NAND Flash partners' 3D capacity expands.

In the third quarter we will also begin mass production of our PCIe NVMe client SSD controller for our flash partners. These new programs are just the first of many more to come in the upcoming quarters.

Our PCIe NVMe customizable hardware (inaudible) [10 gig] controller solutions are the first of their kind enabling our customers to replicate much of the success they have had with our SATA3 controller solution. And extend that into higher performing PCIe.

Let me now turn to our eMMC controllers. Sales of our eMMC controllers grew again in the second quarter following a very strong first quarter. Our eMMC controller sales in the first half of this year grew almost 50% when compared to the same period last year.

This very large growth grows well in excess of overall eMMC market growth and grows well in excess of our own sales projection at the start of this year, is largely the result of SK Hynix, our primary customer, aggressively increasing market share. We have also been benefiting from faster 4G smartphone sales in China this year as the government subsidized consumers to upgrade from 3G phones to 4G phones.

Additionally, we have also benefited from the faster growth of emerging smartphone OEMs such as [Avivo] and OPPO, and the faster growth in emerging markets such as India. We expect our strong eMMC controller sales momentum to continue into Q3.

For the full year we believe our eMMC controller market share will likely increase to roughly 30% to 35%, up from about 25% last year. This quarter we started shipping our eMMC (inaudible) controller to SK Hynix for managing their new 3D NAND. We believe these new 3D NAND eMMCs are targeting at higher end smart phones that will require large storage capacity.

(Inaudible) schedule to launch our [UFS 2] controller in the second half of this year for our first flash partner. We believe the timing of our UFS 2 should (inaudible) with the emerging of larger and more meaningful UFS 2 market next year when UFS 2 is supported by more high-volume smartphone chipset vendors such as [Mediatech] and [Spectrum].

Our UFS 2 solution is designed to be both high performance and cost effective, a deal in helping to bring the solution into the higher volume mainstream smartphone market next year when there is broader chipset support for UFS 2. As you are aware, Mediatech and Spectrum do not currently support UFS 2, but will likely do for next year.

Let me now turn to our highly specialized customized SSD solutions, which are Shannon [Enterprise] SSD and Ferri Industrial SSD. This quarter sales of these SSDs grew by 50% sequentially as a leading Chinese eCommerce customer scales its procurement and will increase shipment to several Japanese OEMs for their commercial equipment including multifunction printer systems and point-of-sale devices. We expect sales of our SSD solution to grow (inaudible) in Q3.

Overall this was a stellar second-quarter for Silicon Motion. Our pipeline of new products and new customers in multiple product segments will make 2016 another record year for us in terms of sales and profitability. And set a foundation for continued strong growth in 2017. I will now turn the call over to Riyadh to discuss our financial performance and outlook.

Thank you, Wallace. First I will outline our financial results and then provide our guidance. This was another outstanding quarter for Silicon Motion. Revenue increased 25% sequentially to $141 million, much higher than guidance because of stronger-than-expected demand across all of our business segments. Our revenue year to date is 51% higher than the same period a year ago.

For the quarter our mobile storage segment revenue grew 25% sequentially. Sales of our embedded storage products grew almost 25% sequentially and accounted for almost 80% of total revenue. Within our embedded storage products our client SSD controllers grew over 30% sequentially, our eMMC controllers grew over 5% sequentially, and our SSD solutions, which are our enterprise and industrial SSDs, grew over 50% sequentially.

Our expandable storage products rebounded in the quarter and grew over 30% sequentially due to greater availability of flash to module makers. Our mobile communication segment also rebounded in the quarter and grew by 38% sequentially and accounted for 7% of total sales driven by new orders for mobile TV SoCs.

Our corporate gross margin decreased to 48.4% in Q2 from 50.6% last quarter, in line with our guidance, as sales from our one-time custom SSD solutions project caused a temporary dip to our gross margins.

This one-time project where we custom designed a high-capacity high-performance hyper scale SSD solution also required our customer to consign to us flash components. But due to how this transaction was structured we had to recognize these components as both part of revenue and cost of sales resulting in lower gross margin for this project.

In Q2 our operating expenses increased to $30.9 million from $26.8 million in the previous quarter primarily due to higher compensation expenses. We ended Q2 with 1,049 employees, 24 more than at the end of the previous quarter. Most of our new hires are R&D engineers.

Operating margin was 26.5% in Q2, slightly lower than the 26.8% reported in the first quarter, but well exceeded our guidance range of 22% to 24% due to much higher revenue. In Q2 we achieved both record quarterly net income of $30.7 million and earnings per ADS of $0.86.

Our EPS was 26% higher than in the previous quarter and 69% higher than in the same period a year ago. Stock based compensation in Q2 was $0.5 million lower than the $2 million in the previous quarter due to the seasonal timing of RSU awards.

I will now move to our balance sheet and cash flow. Inventory days decreased slightly to 97 days in Q2 from 98 days in Q1. DSO decreased to 48 days in Q2 from [55] days in Q1. (Inaudible) days increased to 60 days in Q2 from 57 days in Q1.

Our cash, cash equivalents and short-term investments increased to $219 million in Q2 from $191 billion in Q1. Primary sources of cash in Q2 came from $30.7 million in net earnings, a decrease in receivables of $5.5 million, an increase in payables of $1.5 million.

Primary uses of cash in Q2 included an increase of $8.8 million in inventory, $2.9 million for the routine purchase of software design tools, $2 million for the purchase of additional office space, $5.3 million of quarterly dividend payments.

I will now turn to our guidance. For Q3 we are expecting sequential revenue growth of zero to 5% equivalent to year-over-year growth of 47% to 55%. For the full year we are now expecting revenue to increase 44% to 46%. We foresee seasonal weakness in Q4, consistent with our previously communicated expectations.

We expect our client SSD controllers to grow sequentially in Q3 as we further increase our sales of TLC SSD controllers and continue our initial sales of PCIe SSD controllers to our flash partners.

For the full year our client SSD controller sales will now increase up to 150% year over year. We expect our eMMC controller sales to grow further in Q3, in line with annual peak seasonality patterns. For the full year our eMMC controller sales will grow significantly faster than the smartphone market growth.

We expect sales of SSD solutions to grow sequentially in Q3, primarily because our Q2 one-time custom developed enterprise SSD was successfully received and enlarged by our eCommerce customer and will likely continue selling at slightly higher volume in Q3. We are sure this program will end in Q3.

We expect sales of our expandable storage and other products to decline in Q3. We expect Q3 gross margin to decline slightly from 48.4% in Q2 to 46% to 48% as a result of one more quarter of higher sales of our very low gross margin one-time enterprise SSD project. We expect Q4 gross margin to recover towards 5% -- to 50% -- we expect Q4 gross margin to recover towards 50%.

Our Q4 gross margin will benefit from not having this one-time project, but we will also face in Q4 seasonal weakness relating to our client SSD and eMMC controllers, products with above corporate average gross margins. And seasonal strength of our card and flash drives products with below corporate average gross margins.

For the full-year our gross margin should be in the 47.5% to 49.5% range. We expect Q3 operating margin to remain stable in the 26% to 28% range. For the full year we are now expecting operating margin in the 25.5% to 27.5% range, higher than last year and what we had previously planned because of stronger sales and better operating leverage.

Stock-based compensation in Q3 is expected to be $7 million, higher than the $0.5 million in Q2 due to the seasonal timing of RSU awards. For full-year 2016 our stock-based compensation should be approximately $19 million to $21 million. Our model tax rate remains at 18%.

Looking into 2017, we continue to see a number of growth drivers for our business. Our client SSD controllers should grow at least in line with 20% to 25% market growth for client SSDs. Our eMMC controllers should grow at least in line with the 5% market growth for smartphones.

Relating to our SSD solutions. The market for industrial and enterprise SSDs are expected to grow by 15% to 20% next year and China may grow a bit faster. Our expandables are expected to decline 10%. Everything else should remain stable.

We will provide additional color as we get closer to next year. We will now open the call for your questions.

(Operator Instructions). Anthony Stoss, Craig-Hallum.

Hi, guys. A three-part question. Riyadh, I don't know if I missed this, but how big or how much revenue from the custom SSD final on the one time in Q3, what is the revenue impact in Q3?

The revenue is fairly material. The gross profit, however, is only a couple million dollars. As you know, for these SSD solutions, NAND Flash is a significant part of BOM costs, somewhere in that magnitude of 80% of BOM cost is NAND Flash. And so as a result the gross profitability of these sales are fairly low when we have to account for NAND.

Originally this project was designed where our customer would consign flash to us and we would not have to account for these costs. But as it was subsequently structured the structure was unfavorable and we had to account for the component cost resulting in what should have been high profitable business to a much lower profitability business. So without this business we would still have grown our business quite significantly in both Q2 and Q3.

Okay. And then next two, on the client side SSD market, what share do you think you have right now and where do you see that going? And then lastly, on the eMMC share side you talked about your share increasing. Is that from your one primary customer or are you bringing or expect to bring on additional customers? Thanks.

For the client SSD market we predict around between 30% to 35% market shares today.

And for eMMC we are expecting our market share to the somewhere around 35% for the full year.

Is that (multiple speakers) from your primary customer? I'm sorry, go ahead.

That's correct. Our eMMC is -- most of our eMMC sales are to one flash partner. Hynix is somewhere around 95% of our sales last year.

Okay, thanks, Riyadh.

Mehdi Hosseini, SIG.

Can you elaborate on the mix of your client SSD of between the portion to be sold to OEMs and the portion that basically goes into the [after hour] market or consumer SSD?

And then second question, for 2017 revenue forecast, how do you account for the 3D NAND manufacturing view? And what happens if the yields do not improve or how much of a yield overhang is embedded in your forecast for 2017?

So, regarding our client SSD controller shipment today, about 2 to 1 ratio to PC OEMs versus retail margin makers. Regarding 3D NAND yields, we really very hard to comment how, but what I can assure you all that NAND makers work very hard to improve the yield.

And we believe (inaudible) they will reach the target. And we don't expect higher volume from (inaudible) Samsung NAND maker output for this year. But we believe all the makers will be ready in the middle of next year with a volume ramp for 3-D NAND.

Great. Just one follow-up, again looking into 2017. How does the mix in the client SSD change? Do you expect that the 2 to 1 ratio remains unchanged?

Well, we think the PC OEM portion will continue to increase. However, I don't know whether you -- you're probably also aware, but today the desktop PC, [60%] of DIY market. And DIY market tends to move (inaudible) from HDD to SSD. This also increased the -- our potential business from module maker to enter the desktop and game PC market

Mehdi, let me also add we are seeing very strong growth from both our OEM customers as well as the module maker customers. Furthermore, in terms of overall trend, last year we had shipped a lot more to module makers than OEMs, but this year but that has reversed.

We are shipping a lot more to OEMs than module makers, we are seeing very strong growth from both segments. But specifically with the flash customers, [ours], we are seeing very, very strong growth in the most recent quarter that just ended.

And how do you see that changing into next year?

They should maintain the pace because we see not only the notebook and three-in-one notebook but also the desktop and including game PC is converting quickly from [HDD] to SSD.

Okay, great. Thank you so much.

Mike Burton, Brean Capital.

Congrats on the very strong results. First how do you see the Chinese handset OEM market shaping up in the second half? There has been some suppliers that have voiced some concerns about a potential slowdown coming.

Well, regarding China and our market, we see the demand stream in(inaudible) is strong because China government and the carriers subsidize from 3G to 4G, about 200 million subscribers they're going to complete for this year. So the transition of the 3G to 4G is very strong.

That benefits our EMC controllers because traditionally in the 3G smartphone you use [roaming] and for 4G they need eMMC or eMCP. So it is (inaudible) whether your smartphone maintains single-digit growth or flat, but we see the demand for eMMC controllers maintains still very strong, especially in Q3.

Mike, let me also add, specific to us, a lot of our eMMCs are being sold -- or rather Hynix's eMMC with our controllers are being sold into the Chinese marketplace. And for us we are seeing very strong growth in Q1, Q2, our growth will continue into Q3. But we should see our typical seasonal downturn in Q4.

Great, thanks, that is helpful. And then also on the UFS side of it, I will stick with mobile I guess, how many partners/customers now do you have for your UFS 2.0 solution? And you talked about an initial customer spending ramp in the second half, is that customer currently a large customer of yours or is it a new flash OEM? Thanks.

Well, we can say currently we (inaudible) with one NAND Flash partner, we going to start to ramp in second half this year. We also in the [progress] engaged with another very important NAND maker. And we believe we will have vendor production for next year, first half next year.

Great, thanks.

Rajvindra Gill, Needham & Company.

Thanks for taking my questions and congrats as well on good results. Question on the competitive landscape in the client SSD controller market. Clearly you guys are benefiting from an acceleration in attach rates, but you have also been able to gain some share according to some of the math that you gave.

Can you talk about whether you will see more competition next year as the attach rates start to accelerate? And if so, kind of what is the competitive positioning going into next year?

Well, with [flash] becoming more difficult to manage and SSD performance becoming better I think the barriers to entry are getting higher. I think with our strong relationship with other flash OEMs and without really (inaudible) knowledge for the new NAND, it will be very difficult for any competitor to compete against us effectively.

Also I think the large and low R&D investment (inaudible) controller platform and leading edge processor [geometry] can cost a lot of money today. We are in 28 nanometers moving to 16 nanometers next year and it's a lot investment for any newcomer to compete.

Our [turnkey] solution is very hard to replicate and without our long track record, without the flash OEMs, the (inaudible) collaboration, our (inaudible) office mutual collaboration it is very difficult for any [emerging] controller supplier to provide such a structure of this model turnkey solution to compete against us.

(Inaudible) we integrate all of the flash makers' specific needs including other future 3D NAND into our controller. So it is a very strong platform, strong solution provide to meet and fill the gap for our NAND maker. That is why we believe we have a strong position moving to next year and beyond.

And, Riyadh, and terms of the annual guidance of 45% at the midpoint, that implies that Q4 will probably be down about 11% sequentially from Q4 versus Q3. Can you talk about kind of what are the reasons for the sequential decline?

Obviously some of it is going to be seasonal. But maybe if you could elaborate a little bit further on that Q4 drop based on the implied guidance.

Sure, sure. Yes, you should expect seasonal downturn of our overall sales in Q4. Specifically about the key parts of ours you should expect seasonally down of our eMMC sales as well as our client SSD sales. And our other product, our SSD solutions, we should -- as our big projects taper off in Q3 our SSD solutions will be at a slightly lower level in Q4 too.

And last question more kind of big picture, Wallace, in terms of -- we were mentioning -- talking about a competitive landscape, but what are your thoughts in terms of the flash vendors continuing to outsource the client SSD controller technology to merchant suppliers like yourself?

Do you think that trend continues next year? I mean clearly you do. But what are some of the reasons why the flash vendors will continue to outsource going forward on the client SSD (multiple speakers)?

So in our knowledge understanding our turnkey solution becomes very efficient to align with NAND makers' next generation NAND flash. It is very important for NAND flash makers to benefit for their own (inaudible) the new NAND to achieve a cost benefit.

So I think the NAND makers (inaudible) R&D invest in the longer term and bigger profit [product] to the enterprise SSD and then for high-end.

For certain premium line [client side], some NAND makers will also develop for themselves with internal solution. But we see moving forward into 3D NAND and our NAND partner going to award more projects for client SSD to us, their internal resources will focus on enterprise SSD development.

Thank you.

Next question, please.

Mike Crawford, B. Riley & Company.

So, imitation is a form of flattery and on the competitive front Marvell recently acknowledged a change in its own SSD controller strategy where they want to start incorporating more firmware with their hardware designs. Do you -- have you seen that as well?

We also heard about some, they tried to increase more firmware engineers and tried to provide solutions. By seeing, as you know, the -- we are moving very fast in (inaudible) development cycle. And we will (inaudible) separate independent team can support 12 major OEM projects simultaneously and for all different types of NAND across the board.

So I think that definitely there could be some -- they have their strategy to come in the market. But our relationship and our business model and market position is very broad, very strong for (inaudible) today.

Okay, thank you. And then on this hyper scale data center custom designed SSD solution, so it sounds like that has been successful for eCommerce partner. Riyadh mentioned $2 million of gross profit, is that for the life of the project or just in Q3?

And also, I think last quarter you said that you thought it was going to be -- I thought you said a couple million of revenue. And I am wondering if you could quantify the revenue in Q3 that been obviously wouldn't continue in Q4 related to that.

Mike, let me put this into perspective. Each quarter originally this product -- this project was to last for one quarter and the gross profit for this would be in the range of a few million dollars.

But since we have to also account for the NAND component cost and when use look at the BOM cost of an SSD NAND component cost accounts for in the ballpark of 80% of the overall BOM, right.

So if you were to scale up then the revenue of this project becomes a lot bigger relative to the gross profit base. And when you sort of extend that for one more quarter that gives you a sense of how big the project is.

So from a revenue perspective it is actually quite chunky. But because of the need to account for the cost, the gross profitability is a lot lower. And the overall gross profit is a lot smaller than the revenue.

Yes, Mike, let me add some comments. Hopefully you can look at the bigger picture why we made the decision to take business. Because the end customer is the number one worldwide leading e-commerce customer there is tremendous new technology and [potential] new solution.

So it goes through the (inaudible) engagement it provides or opens the door for us to engage with tremendous future opportunity with all types of different solutions. So, this is not only do we need [gain] through the business model, that we can moving from client side our channel system can moving quickly to enterprise solution.

Right. So given that there is this flash -- originally this flash was to be consigned and now you are actually -- you have to account for it on a pass-through basis. But so it sounds like the project has worked and your customer is happy with it.

So then what is the next step from there? It just becomes more of a typical Shannon enterprise SSD sale and that is part of the -- of what drives that business towards $75 million in a couple of years?

That is correct, Mike. We have these huge chunky projects and we have more routine custom projects. And so, it is a combination of these projects will lead to our overall price targets.

Specifically for this project, while from a P&L perspective it doesn't look -- superficially it doesn't look very pretty, but from a investment or return it is actually quite attractive. So that is why we are pursuing it from a financial perspective.

And as well, as talked about in quite some detail, the strategic benefits are also quite significant in allowing us -- enabling us to pursue more unique custom projects with this Internet company.

That is great, thank you. And then last question on UFS. So, with your eMMC business with Hynix, I think you supply the vast majority if not all of the controllers that Hynix uses for eMMC. But do you think that share shifts with UFS and UFS2?

So let me comment, for UFS market today, because there is -- currently adoption is still small compared to eMMC, and as we believe in next year with Mediatech and Spectrum, their chipset supports the UFS2, the UFS2 embedded solutions, our so-called [UMCP] with DRAM solution, become popular the volume will start to increase.

And we support Hynix UFS by eMMC solutions today and, as you know, Hynix also developed UFS three years ago, I think however we have opportunity to engage with any other NAND maker beside our current NAND partner. We believe we will continue to grow eMMC next year, we also will start to grow UFS next year in 2017.

Okay, thank you very much.

Tom Sepenzis, Northland Capital.

I am just curious, both Intel and Micron are set to introduce 3-D [Xpoint] at the end of this year. And I was wondering if you would be involved in that. And if so how that might impact ASPs and margins?

We cannot comment how Micron and Intel introduced the 3D XPoint and we cannot comment our involvement in 3-D XPoint this moment.

Okay, thank you.

Charles Chan, Morgan Stanley.

So, my first question is regarding your fourth NAND Flash OEM customer. So it seems to be delayed for several quarters. Can we get any idea the reason behind? And Wallace just said that maybe there will be some production in first half. Is that for this fourth customer?

Regarding our fourth potential customer, NAND flash partner, while we have been working with them for quite some time, the time when we can start shipping to them in commercial basis looks like it has been pushed out a bit.

And so, even if they were to have started production this year the meaningfulness of this revenue stream this year would have been fairly small. And the revenue stream next year would also have been fairly small. We are still engaged with them and so we will continue to work and see how this project evolves.

So sorry, let me add a comment, Charles. Maybe I didn't state properly. Our first NAND maker for USF 2 (inaudible) production this year. And we mentioned our second NAND maker partner will enter production the first half of next year for UFS2.

Okay. And on your enterprise SSD business, besides the Shannon system, what is your exposure to the enterprise SSD controller [IC]? So you mentioned now for your SSD controller IC, majority of them are either for PC OEM or retail vendors. Is there any shipment for data center usage for this SSD controller IC?

Today we're already seeing some of our flash partners repurposing our client SSD controller paired with higher-quality NAND to address a non-mission critical hyper scale data center market. We also have our channel enterprise (inaudible) SSD business that targets very high-end segments of China.

However, the volume remains small today. We will continue to monitor the market and when the volume becomes bigger/meaningful then you should expect us to have controller solution targeting this sector.

Right. So apparently the (inaudible) already reached the sweet spot, right? So at what points or what do you think would trigger a major adoption of the SSD in data centers from your perspective?

We would think that, as you know well, Intel is pushing PCIe NVMe very hard. And we see the adoption rate for future data centers moving to PCIe NVMe. However, in the very [low cost] portion we will state with the SATA.

So the solution I think for SATA needs to be a cost sensitive solution for PCIe NVMe moving from 2 terabytes and 4 terabytes. We believe maybe second half next year we should see the volume moving with bigger scale compared with this year.

Okay, is there any -- sure, please.

In the enterprise HDD market, especially the mission critical transactional type of HDDs, the high-performance HDDs, a lot of those are already moving over to SSDs, especially the PCIe NVMe SSDs. And you should continue to see that trend accelerate over the course of the next few years.

Okay, okay. Yes, so lastly just want to follow up a previous question regarding the competition, because your competitor (inaudible) said publicly next year they will also use 28 nanometer. And they are going to lower the SATA (inaudible) price to $3, PCIe maybe to $5. Do you think that would impact your pricing for the controller IP business?

I think you know we have been in controller business for more than -- almost [15] years. Competition (inaudible). A lot of competition cost. But we have a very strong product portfolio. We position high-end mainstream (inaudible).

And we cannot comment specific to our competitor. But I think every player has their strategy. But we have very high confidence maintaining leading market for both SSD eMMC. And we will definitely also see the potential growth for UFS2 next year.

Okay, great. Thank you very much.

(inaudible), Nomura.

Thank you for taking my question. My first question is regarding to your client SSD control market share. If I hear correctly, you said about a 30% to 35% market share this year. And I wonder is it only above your market share in OEM -- PC OEM sector or it is including PC OEM and the retail SSD (multiple speakers) shipments?

(Inaudible) worldwide SSD shipment from all the (inaudible) report and survey this year total SSD unit shipment will be between around 130 million units is the [optimist] best case. So we believe we will be around the range from 30% to 35% range for the total worldwide shipment percentage.

Because previously you mentioned about your roughly ASP per client SSD controller. So in my roughly calculate is that you since have even higher control shipment this year already.

I'm sorry, Donny, could you repeat your question again?

Yes. Because you said about your client SSD controller business sales can grow by up to 150% this year. So I roughly calculated that if you -- if I use SP let's say about $5, it seems like that your market share is not that high to 30% to 35% yet. So I am wondering if you have -- you should more -- even more client SSD controllers already this year.

Well, we are looking at the numbers based on research reports that we have received indicating potential size of the SSD market. Potentially the SSD market could be bigger or smaller than what these analysts are suggesting.

And this is simply just the implied market share based on our numbers versus these numbers. And so that may not be the real market share, but it is an approximation of where we may be.

See 150% is our sales revenue growth on last year. But the units -- the market share is based on unit shipments for controllers.

Okay, got it. Thank you. And so, what kind of market share you are expecting to reach in the longer term?

We believe we are going to reach a minimum of 40% market share in the next few years. We believe just like with Samsung, the above 40% to 45% range for the SSD market share today.

Got it. And my second question is regarding to SSD penetration rates. So, in your opinion what kind of penetration rate we should expect this year and probably next year?

I think this year based on third-party research that that we just mentioned, the market (inaudible) client SSD will be around 130 million units. And this is expected to grow more than 20% for the next few years.

So this year is only about 35% of all PCs are using SSD, that's up from 25% last year. Today we think they're about close to 500 million storage devices being shipped annually. So if we ship only one quarter SSD, we believe over the next few years the majority of these storage devices will be using SSD.

So, we can envision an annual SSD market of well over 300 million units in a few years. So, penetration rates, I think Riyadh and during the statement, will be around 20% growth rate in the next few years.

Okay. And my last question is regarding to your operating leverage. So, seems like you have improved your operating efficiency a lot due to sales scale improvement. So -- but you also mentioned that you have some product in 28 nanometer already.

And I am wondering when your products are migrating -- I mean more products are migrating into a 28 nanometer, so will; it increase your OpEx going forward?

Well, Donny, we are scaling our revenue quite significantly, which means we are going to be able to afford more of these high dollar tape outs. And so, the number of tape outs that we can do I think is these finer nodes that are very expensive will relate to how fast we are able to continue to scale our business.

The other part of how we are able to get operating leverage relates to how we use our headcount. And as you know, we remain resource constrained and we are still trying to invest and hire people to address all the additional (inaudible) with a drive to continue and expand our operating margin.

So you should expect us to grow our operating expenses related to both headcount and also tape outs, project tape outs. And we will try to do these -- try to increase our operating expense, grow it at a slightly lower rate than our total revenue so that we can continue to enjoy operating leverage.

Got it. Thank you so much.

Thank you very much. We have no further questions from the phone line. I would like to hand the call back to management for closing remarks.

I would like to thank all of you for joining us today and your continuing interest in Silicon Motion. We will be the following conference this quarter. In August we will be presenting at the Pacific Crest 18th Annual Global Technology Conference in Vale; Oppenheimer 19th Annual Technology Conference in Boston; Jefferies (inaudible) Asset Day in Chicago.

In September we will be presenting at the Citi Tech Company in New York; Brean Capital (inaudible) in New York; JPMorgan Asia CEO/CFO Conference in New York; (inaudible) Technology Conference in Taipei and New York; Deutsche Bank Annual (inaudible) in Las Vegas; and CLSA Investor (inaudible) in Hong Kong. Details of these event will be available on our website. Thank you and goodbye for now.

Ladies and gentlemen, that does conclude the conference call for today. Thank you for your participation. You may now disconnect the line.