慧榮科技 (SIMO) 2016 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by and welcome to the Silicon Motion Technology Corporation Q3 2016 earnings conference call. At this time all participants are in a listen-only mode. There will be a presentation followed by a question-and-answer session. (Operator Instructions).

This conference call contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933 and Section 31E of the Securities Exchange Act of 1934 as amended. Such forward-looking statements include without limitations statements regarding trends in the semiconductor industry and our future results of operations, financial condition 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 pressures in the semiconductor industry and the effect of such pressure on prices; unpredictable changes in the technology and consumer demand for multimedia consumer electronics; the state of (inaudible) any changes in our relationships with our major customers and changes in political, economic, legal and social conditions in Taiwan.

For additional discussion of these risks and uncertainties and other factors, please see the documents we file from time to time with the Securities and Exchange Commission. We estimate obligation to update any forward-looking statements which apply only as of the date of this conference call.

I must advise that this conference call is being recorded today, Friday, October 28, 2016. I would like to hand over the conference to our first speaker today, Mr. Jason Tsai. Thank you. Please go ahead, sir.

Thank you and good morning everyone. Welcome to Silicon Motions third 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 recent business developments; Riyadh will then discuss our third-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 is filed on form 6-K after the market close yesterday.

The 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 enable you to perform comparisons of our operating results in a manner similar to how we analyze our own operating results. The reconciliation of the 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. This year has been an outstanding year for us and the third-quarter is no exception. Revenue increased this quarter by 13% to a record $139 million, our seventh consecutive quarter of revenue growth. Our $155 million revenue is equivalent to a 56% year-over-year growth.

Growth was led by our Client SSD controllers, eMMC controllers and SSD solutions. In addition to record sales this quarter, we also delivered record high quarterly EPS of $1.07, our first quarterly EPS over $1.00. Riyadh will discuss our financial performance in greater detail later on the call.

Let me now update everyone on the progress of our key growth drivers starting with our client SSD controllers, which has become our largest product with sales now even bigger than our eMMC controllers. Sales of our Client SSD controllers increased 25% in the third quarter, our eighth consecutive quarter of sequential Client SSD controller growth. For the full-year 2016, we are on track to grow our Client SSD controller sales by approximately 175%.

In the third quarter, our Client SSD controller sales to our flash partners were especially strong. These sales increased over 50% sequentially with our flash partners accounting for well over half of our overall Client SSD controller revenue.

Sales of our SSD controller to market maker customers despite facing tight supply conditions also increased sequentially though more modestly. We are confident our Client SSD controller sales will ever likely grow further in the fourth quarter with continued growth led by sales to our flash partners.

Our Client SSD controllers have been growing rapidly for many reasons. Demand of Client SSD remains very strong both with PC OEMs and in channel markets. PC OEMs are still increasing their adoption of SSD versus HDD. On the supply side, NAND flash makers are increasing their direct sale of SSD both to PC OEMs and into channel markets versus NAND wafers sales to module makers. However, NAND flash makers are also supplying flash to a select group of leading module makers especially those building SSD for unique distribution channels, all geographic markets incremental to those managed by the NAND flash makers.

Most industry analysts believe SSD adoption in notebook PCs will increase to at least one-third of all notebook in 2016. In addition to notebooks, PC OEMs are also building desktops with SSD, both for mini PC and for high-end machines. A significant number of power users, gamers and system integrators have been building or upgrading PC and game console with SSD already.

Client SSD are also being used in drones, network attached storage, external portable storage and non-mission critical business [interpretations]. Third parties estimate for the size of the Client SSD market this year range from 110 to 135 million units which means already 20% to 25% of our client devices are using SSDs.

Single NAND flash prices will continue to fall rapidly. We believe that over the next three to five years, most if not all of the remaining 350 million to 400 million client devices still using HHD will be replaced by SSDs.

Intel and [WD] recently launched new client SSD with our controllers, Intel's 600P SSD is the industry's first PCI eMMC SSD that is affordably positioned at a SATA SSD price point but with superior PCIe NVMe performance. The client SSD market is beginning to transition from SATA to PCIe NVMe. Drivers of the transition will improve differential products like the new Intel SSD. For us this is also the first major product launch that is using our new series of PCIe NVMe controllers and the first major product managing Intel Microns 3D NAND.

Separately, WD recently launched their first set of Client SSD following their acquisition of SanDisk. So WD (inaudible) SSD is using our new SATA controller that we recently developed for WD. This product is a continuation of our previous (inaudible) SSD controller design work with SanDisk prior to WD's acquisition which involves developing controller solutions that do not require expensive DRAM components but deliver performance comparable to SSD that use fast [data] embedded DRAM.

We are entering the fourth quarter of this year with a larger pipeline of Client SSD controller design wins than last year. Most of these projects are for 3D NAND. Already our SSD controller have been designed to support all of the 32, 36 and 48 layer 3D NAND that are being produced by our flash makers. By the middle of next year, our SSD controllers will be in volume production for managing Hynix, Intel, Micron, SanDisk, Toshiba and even Samsung 3D flash 3D NAND. No other SSD controller supplier whether captive or merchant has this breadth of 3D NAND support and capabilities.

All of the NAND flash makers are in commercial production of 3D NAND though other than Samsung, volume from other flash makers is still limited. Seeing the CapEx of the agreement to manufacturers 3D NAND is very expensive, we believe flash makers have been cautious in building capacity as they go through the learning curve on current generation of 3D NAND. They will become a lot more aggressive next year in building capacity for 54 and 72 layer 3D NAND as the economy of this higher layer of 3D components will be significantly better than (inaudible) NAND.

Furthermore, we believe the economies of producing 32, 36 and 48 layer 3-D NAND isn't very different from current 16 and 40 nanometer (inaudible) NAND. So the incentive to produce more current generation of 3D NAND may be limited. We believe the industry if it grows will start accelerating beginning May 2017 when commercial production of 64 and 72 layers 3D NAND increases.

Many of our new SSD controller programs will start mid-next year in line with the ramp of this next-generation 3D NAND. We believe NAND supply will grow rapidly in the second half 2017 and will scale further in 2018.

Most of our pipeline of projects for next year are for PCIe NVMe Client SSD. We are confident that with the ramp of our new PCIe NVMe projects and the continued market share gain with SATA SSD all with existing NAND flash partners, we should be able to comfortably grow our clients SSD controller sales at least 20% to 25% next year on the top of the 175% growth that we are expecting this year.

Now let me turn to our eMMC controllers. The sales of our eMMC controllers grew again in the third quarter as we benefited from peak season demand. Our third-quarter year to date eMMC controller sales are over 50% higher than the same period last year. The strong growth is a result both our market share gains and expansion of the transport market.

We have been gaining market share because SK Hynix, our primary eMMC customer, has been gaining market share and reporting very strong eMMC sales. We believe our eMMC market share this year is growing to 30% to 35%, up from about 25% last year. The eMMC market has been growing much faster than overall smartphone market because of a large portion of ultra low cost smartphones and China white box tablets were previously using bare NAND. This product has now transitioned to using eMMC solutions.

Looking ahead, the segment of the eMMC markets are gradually transitioning to UFS beginning with high-end flagship smartphones. We are working with several flash partners on UFS controller solutions and as previously communicated, we have already secured design wins with one of our flash partners. Our UFS controller for this partner will go into production in the fourth quarter which is positioning us well for further adoption of UFS next year when MediaTek introduces chipsets that support UFS.

We will be bringing to our flash partner a UFS controller that is both high-performance and cost-competitive and we will be supporting our flash partner in their smartphone OEM customer with our extensive field engineering resources. We are progressing well and interacting with our primary flash partner for our cost-effective UFS solution. This solution is designed for their [UNCP] platform that will combine our controllers, NAND flash and mobile DRAM into a single PGA package. This will be complementary to their internal solution and will help them to quickly extend UFS adoption to more mainstream devices.

Let me now turn to our highly specialized customized SSD solutions which are Shannon Enterprise SSD and Ferri Industrial SSD. This quarter Ferri SSD solutions grew further led by increasing sales of our leading Chinese (inaudible) Internet customers.

(inaudible) customers design Enterprise SSD have been favorably received by our customers. We have been engaged for a new project for delivery next year. Our Enterprise SSD are designed for our customers' specific requirements relating to performance, capacity, power, latency and other features. We have also seeing very strong growth this year for our Ferri Industrial SSD which primarily targets Japanese OEMs for their commercial equipment including printer systems point of sale sales devices. Our Ferri SSD are industrial grade, single-chip SSD customer design for OEM specific requirements. Over the last few years, we have been advocating a large base of Ferri SSD projects that generally have long production cycles. The visibility from our project bookings suggest strong Ferri SSD sales growth next year.

Overall this was an outstanding third-quarter for Silicon Motion. We are entering the final quarter of the year with a strong pipeline of Client SSD controllers, eMMC and UFS controllers and SSD solution projects which will position us for further growth in 2017.

I will now turn the call over to Riyadh to discuss our financial performance and outlook.

Thank you, Wallace, and hello, everyone. First I will outline our financial results and then provide our guidance.

Before I begin, I would like to reiterate that my comments today will focus primarily on our non-GAAP results unless otherwise specifically noted. A reconciliation of our GAAP and non-GAAP data is included with the earnings release issued today.

We continue to deliver on sales and earnings growth significantly above semiconductor industry average by focusing on unique market opportunities relating to NAND, flash control technologies. Our business continued to deliver results that can sustain long-term gross margin of around 50% and gradually move our operating margin towards our 30% target.

Wallace just updated everyone on the progress and prospects of our three big market opportunities, client SSD controllers, eMMC and UFS controllers and our highly specialized SSD solutions.

This quarter revenue increased 13% sequentially to $159 million, 66% higher than the same period last year. Our revenue year to date is 56% higher than last year. Sales of our embedded storage products which include our client SSD and eMMC controllers and our industrial and enterprise SSD solutions grew over 15% sequentially and continue to account for about 80% of total revenue.

Within our embedded storage products, our Client SSD controllers grew 25% sequentially. Our eMMC controllers grew over 15% sequentially and our SSD solutions grew less than 10% sequentially.

Our expandable storage products, our card and USB flash drive controllers declined about 10% sequentially due to tightness of flash availability. This quarter we had three 10% plus customers, Hynix, another NAND flash partner, and a China VAT Internet company.

Our gross margin increased to 48.9% in Q3 from 48.4% the previous quarter due to sales of a more favorable product mix. Our operating expenses increased 4% to $32 million from $30.9 million in the previous quarter. Most of the increase in operating expenses was R&D related. We ended Q3 with 1094 employees, 45 more than at the end of the previous quarter. Most of our new hires continue to be R&D engineers.

As a result of strong sales growth, stable gross margins and a modest increase in operating expenses operating margin increased to 28.7% in Q3, a significant improvement from the 26.5% reported in Q2 and 24.5% in the same period last year.

Our effective tax rate in Q3 was 16.8%, lower than the 18.5% in Q2. This quarter we delivered a net income of $38.3 million and earnings per ADS of $1.07.

Our earnings per ADS were 24% higher than in the previous quarter and 88% higher than in the same period a year ago. Total stock-based compensation in Q3 was $6.5 million, higher than the $0.5 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 to 91 days in Q3 from 97 days in Q2. DSO decreased to 38 days in Q3 from 48 days in Q2.

Payable days decreased to 49 days in Q3 from 60 days in Q2. Our cash, cash equivalents and short-term investments increased to $269 million in Q3 from $219 million in Q2. Primary sources of cash in Q3 came from $38.3 million in net earnings, $15.1 million from net working capital, $10 million of bank loans net of restricted deposits. Primary uses of cash in Q3 included $2.8 million for the routine purchases of software and design tools, $5.3 million of quarterly dividend payments.

We continue to generate strong cash flow and as well as had just talked about, we feel pretty good about our sales visibility for next year because of our pipeline of projects that we have built so far that will generate sales next year.

As a result this week we announced our new annual dividend of $0.80 per ADS to be paid in quarterly installments. This is a 33% increase from the $0.60 declared last year.

I would like to highlight that we recently borrowed $35 million from two banks. While we continue to operate a business that generates very strong free cash flow and have not in the past needed external financing and continue to be able to finance our operations while internally generating cash flow, we have determined that the current very low cost of bank loans is comparable to the cost of using our own internal funds.

For internal intercompany loans, we need to account for unnecessary tax payment to government authorities on interest income and the cost of documentation and transfer of pricing studies. We expect our strong operating cash flow to repay the loan within 12 months.

I will now turn to our guidance. Consistent with the expectations that we had communicated during our last earnings call, for Q4 we are expecting revenue to decline 9% to 14% sequentially which is equivalent to an increase of 39% to 47% when compared to the same period last year. Based on our expectations for Q4 for full-year 2016, revenues should grow 52% to 54% to somewhere in the $548 million to $556 million range.

Our Q4 sales will include the following key drivers. We expect our client SSD controllers to continue growing sequentially as we ramp our new PCIe SSD controllers. For the full-year, our client SSD controller sales will now grow approximately 175% from the 60+ million last year. Next year you should expect our client SSD controllers to grow at least 20% to 25%. We expect our eMMC controller sales to be stable in Q4, better than past seasonality patterns due to continued market share gains.

For the full-year, our eMMC controller sales will grow approximately 50%. Next year you should expect our eMMC and UFS controllers to grow at least in line with overall smartphone growth of 5%, a more sustainable long-term growth rate. eMMC and UFS will remain a high priority project of our flash partners.

We expect sales of SSD solutions to decline sequentially in Q4 because of project timing. For the full-year, our SSD solution sales should be more than triple. Next year you should expect our SSD solutions to grow more moderately in the 15% to 20% range. We expect sales of our expandable storage products to decline in Q4 as flash tightness continues to impact module makers' ability to procure flash components. While we do not have good visibility on our card and USB flash drives, sales of these products will likely decline again next year as these products have been in secular decline for a number of years and are not a high priority market for the NAND flash makers.

In Q4, we expect our gross margin to be in the 48.5% to 50.5% range. We expect Q4 operating margin to decline to 25% to 27% but meaningfully higher than the 24% in the same period last year. Lower operating margin in Q4 is primarily due to lower revenue. The impact from slightly higher gross margin and stable operating expenses are less material.

For the full year, our operating margin should be roughly 27, much higher than the 24% in 2015. Stock-based compensation in Q4 in our GAAP operating expense is expected to be $7.9 million to $8.3 million, higher than the $6.3 million in Q3 due to the seasonal timing of RSU awards. Our RSU awards are generally granted in the second half of every year. Our model tax rate remains at 18%.

We will now open the call to a few questions.

Ladies and gentlemen, we will now begin the question-and-answer session. (Operator Instructions). Mehdi Hosseini, SIG.

Thanks for taking my question. Just back on your commentary on client SSD, it seems like you can grow revenues on a flattish -- sequentially it could be flattish and you still hit the 20%. I just want to make sure I understand the dynamics that are driving the client SSD and assume that you would be shipping to multiple flash centers also continue to increase direct sale to year-end.

These are assumptions and then to that extent, what will be the dynamics that would drive the market growth above 25%? So just to summarize, what are the key drivers that would give you confidence that at the worst-case, client SSD comments will be flattish on a sequential basis and the dynamics that will drive the climate SSD growth above 25%? And then I have a follow-up.

I think for our current SSD design pipeline from both the SATA and the PCIe, we have confidence to grow. Although in my (inaudible) supply in the first half of next year because a majority of design is from the NAND makers [upon it]. So we have confidence sooner forecast and penetration with the PC OEM with the channel market and we believe we should be able to grow 25% from this year.

Okay, so during your bookings and your relationship, you are confident that the 20% is the worst-case scenario, is that how we should think about the 20% minimum growth driver?

That's correct. From a client SSD controller.

Okay and then, Riyadh, you talked about you upped the margin target of 30%. Are you going to be a multiyear target or -- because you're going to continue to invest in your enterprise SSD or is that target going to be a hit sooner than later?

That is correct. Improving our operating margin is and will continue to be a multiple year project. Over the last few years, we have been improving our operating margin. Last year our operating margin was roughly 24%. This year we will probably finish the year around 27% and over time as we continue to scale our revenue and keep our gross margins fairly stable and grow our operating expense at a low rate compared to revenue, we should continue to be able to deliver on improvements in operating margin towards our 30% target.

Thank you.

Anthony Stoss, Craig-Hallum.

Two-part question. First off, did I hear you correctly, Wallace, that you expect Samsung to come on as a new customer in the SSD side in the middle of 2017? And then Riyadh, if you wouldn't mind just talking about your OpEx thoughts kind of going forward throughout 2017.

Okay, so while we did not mention exactly on the current SSD customer. We said we are able to support all NAND flash makers 3-D NAND for their SSD solutions and we cannot comment any NAND makers who want to sell their (inaudible) into channel or model makers or OEM customers, we are ready to support that.

Let me also add to what Wallace just said, what Wallace had mentioned in his prepared remarks is that by mid-next year our SSD controllers will be in volume production for managing Hynix into Micron and Toshiba and even Samsung 3D NAND.

Okay, then on the OpEx question, Riyadh?

Can you rephrase your OpEx question again? I'm sorry.

Maybe a little bit more detail regarding what you expect in OpEx next year. I know you expect revenues to grow faster but any more help on OpEx would be appreciated.

Yes, we will continue to build our operating infrastructure as we have a lot of projects relating to all of our growth opportunity relating to client SSDs, relating to eMMC and UFS controllers, relating to our Shannon and Ferri SSD solutions. And also [off-prem] is going into the enterprise SSD solutions. So we will continue to be growing our operating expense so we should be expecting our operating -- for us to continue to invest in R&D and to grow our operating expense gradually in order to meet the demands of our customers. But we will be increasing at a slower rate compared to revenue.

Great, thank you.

Charlie Chan, Morgan Stanley.

Thanks for taking my question and congratulations for great results. So first of all, you have started to ship PCIe comparisons. So what is that for? Is that for client SSD or maybe for enterprise SSD?

The product is primarily for client SSD, for channel as well as PC OEM.

but I think our customers might also use for very low-cost comp hyperscale SSD solution.

Okay. Wallace, can you give us some sense about PCIe penetration in the client SSD and the product SSD today and maybe next year?

(inaudible) now [income], Samsung really pushing PCIe and become from premium line. They would like the try to become mainstream but however I think the momentum is growing for client SSD next year. The penetration rate would be probably at least 10% or 15% bigger than this year. However I think for enterprise side, primary solutions to SATA but PCIe is manufactured to try for the cloud, for cloud SSD and we do see follow-on new designs in the cloud side, the PCIe and in (inaudible) solutions. That becomes the standard and that is why all the NAND, all the flash makers should do more NAND to enterprise and the call side because it has a more profit in that area.

But PCIe eventually we think is not just a NAND 0.2 or [user] 0.2 but there be many different form factors eventually. You're going to see PCA PCIe. solutions come to market within one or two years.

Okay, so for your Company, what would be the PCIe product mix next year and would that improve your plan to ASP?

I think next year our primary shipment SATA is bigger than PCIe. However PCIe momentum is growing. And we see more demand of PCIe from PC OEM as well as all other applications. So I think the PCIe eventually will become our major growth driver in the next few years.

Okay, thanks. So my next question is regarding your module business. I will look into 2017. I know this year you have some one-time like rolling in procurement so it could not be no more new scale so for 2017, it should mean no more further one-time procurements. What would be the revenue size for both enterprise SSD and for industrial SSD.

Charlie on this question, you asked why our SSD solution this here is turbocharged if you will related to the inclusion of the NAND component into our cost of sales and so as a result, our SSD solutions revenue grew rapidly. We're going to be growing our SSD solutions this year in the tune of 3x compared to what we had last year. Last year SSD solutions delivered revenue about $25 million. So going into next year, we expect -- because of the turbocharged nature of our SSD solution this year, we are expecting more moderate rate of growth for these sort of products, moderate rate of around 15% to 20% for next year.

15% to 20% year-on-year?

Year-on-year.

Lastly on your 28 nanometer products, when would you start to recognize the follow mix costs because the R&D seems over the last in the second half or that (inaudible) cost would happen in 2017 and when will start to shift that 28 nanometer products?

We believe our 20 nanometer product in SSD and our UFS will be shipping by middle of next year, second half of next year.

Okay, so will the tape out cost reflect this year's earnings or next year?

The tape out will be [prepared] for this year earnings. I think Q3 and Q4 timeframe.

Okay. Got it, let's see if I may for UFS, what would be the ASP difference versus the eMMC?

I apologize, we cannot comment on the price but definitely UFS has a higher performance and then more sophisticated design and that has much bigger than EMC so ASP will be higher.

Okay, okay, got it. Okay, thank you very much.

Rajvindra Gill, Needham and Company.

Thanks for taking my questions and congrats on good results. A question on the UFS adoption next year. Can you maybe talk a little bit about the transition to UFS across the handset OEMs, how fast do you think that transition is happening and what do you think the main growth drivers or catalysts for UFS adoption? How do you think you are positioned competitively versus, say, internal solutions that are out there?

It's a very good question. I think from our position for 2016 this year in the first half the momentum for UFS adoption is strong. However, go to the second half, it's slowing down. The primary the smartphone adopter is Samsung. We do see there are about six other smartphone makers but they only use one model use a small volume for UFS.

But according to like Huawei, they really like the way for see the really performance benefit on UFS and many others and smartphone makers putting for media tech their new solution which can support UFS next year, Q1. We believe for UFS the account won't become the mainstream product in mobile. It had to be UNCP which means UFS controller and mobile DRAM and NAND to get in single PGA package and this trend probably won't be able to happen until late 2017 or 2018 timeframe.

Okay, thank you. That's helpful. And Wallace, you mentioned the client SSD penetration rate some numbers. In terms of the notebook PC market if I recall you said one-third of the market is SSDs and then you're also seeing growing traction in desktop PC SSDs as well. Can you maybe talk a little bit about how your market share or your competitive position is in the notebook PC and desktop PC? The attach rates are accelerating. Your market share is fairly good at a lot of the flash OEMs outside of Samsung of course which does internally.

How do you look at the market share dynamics going into next year as the attach rates accelerate further? Can you talk a little bit about how you are positioned to continue to grow that business relative to, say, Marvell or even other competitors?

Okay, I think it's a very good question the part issuing regarding (inaudible). Yes, we believe the notebook, one certain notebook has adopted SSD as of today. Of course commercial notebook has a higher percentage than consumer product line for notebook.

Regarding the desktop, really primarily it's two factors, mini PC also the game PC. The game PC user try to adopt SSD because they are much faster playing the games but normally the users do drive the (inaudible) SDD for data storage for the SSD as a gaming playing so achieve higher performance. We also observed the channel and CC shop and do it yourselfer PC they adopt SSD instead of a HDD. Regarding the overall market, as you know NAND supplies today are tight so NAND makers try to maximize the profits for every bit they can produce for the NAND. So from that angle, some major NAND makers shipped the NAND more SSD solutions into enterprise data centers so I think (inaudible) for example, they aggressively transitioned from client SSD into enterprise SSD. So we see that as a tremendous opportunity for SMI, for SIMO to grow and to work with other NAND makers and other model makers to fulfill this spot and we believe our alternative solutions can help all the NAND makers quickly enter into the market with their new generation 3D NAND.

This is the advantage we have and we have a good track record [come here] with Marvell or other SSD controller competitors. And today I think we are well known by all the PC makers and all the NAND makers and our products are consistent and cost-effective with a very decent performance and we continue to put R&D investments into new technologies especially in the 3D NAND either fresher algorithms, special rate designs, and to maintain high retention and better quality and put more value to consumer end-user. We believe we have very good and bright position today in this Class [E] market and we work with all the major NAND makers and continue to grow the market.

Thank you for that. Just last question from me, on eMMC, you had mentioned that year-to-date it's up 50% year-over-year and you are expecting the growth rate next year to grow in line with the market at 5%. That's kind of the minimum case. Can you talk about some of the reasons why that growth rate might be higher next year and why was the growth rate so high this year? You mentioned market share gains with your customer, SK Hynix, do you foresee more market share gains with your customer, you are diversifying your eMMC customer base which could potentially add more growth next year as well? Just wondering if you could talk about some of the potential upside catalysts to that 5% number next year, given the fact you grew 50% this year.

What happened this year, why we gained market share quickly and grow very fast with eMMC controller. This year I think there are two major factors and because everybody knows smartphones only grow single digits, 3% to 5%. And why EMC can grow is almost 40%, 50% this year because China subsidized the 3G phone to 4G so-called 4G phone. So that's why the 3G phone normally uses raw NAND, bare NAND and goes to 4G adopt eMMC or eMCP. That's why they demand about 200 million units and transitioned this year from 3G to 4G. That has created a strong demand for eMCP.

The reason I think I cannot speak for our customers but I think normally eMCP with a mobile DRAM, that has a better profit than a pure NAND solution in other areas. That's why our major customers prefer to grow eMCP allocate more NAND into eMCP solutions to gain the market. This way financially we do gain market quickly in 2016.

For 2017, next year, because the transition from 3G to 4G is down, so the smartphone itself, we are going to continue growing in line with the market. However, eMMC solution is a [change out] standard that being extended to other areas such as set-top box and smart TV and car navigation systems. So I think we are going to grow potentially upside for others not smartphone areas. And there's also UFS, we can also good UFS as part of the smartphone. So we believe we will be in line -- this is in line with the smartphone market growth and within the other side will from other segments, other application areas.

Thank you Wallace.

Mike Burton, Brean Capital.

Thanks and congrats on the strong results. A two-parter on mobile storage. What does a typical Q4 look like for this segment? You have a couple years recently where I believe that business declined 20% plus. Is that typical in your expectation for this quarter?

And then second, we've get a lot of questions from investors concerned about the tightness in the Chinese handset OEMs as a precursor for an inventory correction. Are you seeing any of that in Q4 or do you think we should be thinking about Q1 in 2017?

I think for this Q4, our mobile -- normal controllers for EMC still maintains strong. They could be similar [rates] like a Q3. So we do not see the over inventory in the channel and we do not see double booking at least from our customers and while we can reach assets from the smartphone makers.

Okay, and then regarding Hynix, you mentioned in your comments making progress on UMCP packages with them. I'm just wondering if there is any change in the timing of when you expected your ramps in UFS for next year. I believe you had talked about your first customer being in the first half and then your second, your largest customer in the second half. Is that still kind of the timing for next year? Thanks.

That is why we believe it's a target we are looking for.

You're correct, Mike.

Suji De Silva, ROTH Capital.

Congrats on the strong results here. On the hyperscale side, I'm trying to understand the customer base. Do you have really one or two large customers that are driving the bulk of the revenues there or have you diversified the exposure there to several customers to understand that segment?

Suji, we have a lot of customers but our customer concentration, our customers are in fact quite concentrated with one particular customer and this is one of the big BATE Internet companies out of China.

Is there prospect there, Riyadh, for more customers to join into that along with the one you have now?

We are working on that but so far our visibility going into next year stands with just this one customer.

Fair enough there. Then on UFS, I know a lot of questions have been asked here, but I've just want to understand how many partners, customers you think you'll be ramping with in 2017, just to understand the breadth of your penetration there?

So as we mentioned in our prepared remarks, we have one NAND flash vendor which we're going to be going into production by the end of the year and we have another one that we are working with that is more going live sometime mid next year.

Okay, great. Last question, what's going on with the ASP and SSD market? I know they were around $5, so what are the trends you are seeing as you see that segment growing in the mix between enterprise and client? Thanks.

We aren't going to comment specifically about the actual ASPs of our products, our controlled products, given that we don't have too many customers and for competitive reasons we are not going to be going into that level of detail. But as a benchmark, our ASPs for these products are very roughly around $5.

Okay, great. Thanks, guys.

Jaeson Schmidt, Lake Street Capital.

Thanks for taking my questions. Most have been asked already but just curious, I know visibility is cloudy on the core business, the USB and card side. But is a 10% annual decline going forward kind of a good base case?

Yes, it is. Given that we don't have much visibility into the segment of our business, visibility is quite murky, annual decline of in the teens or of at least 10% is probably a good estimate.

Okay, then the tax rate in 2017, is that Q4 rate a good number to use?

You should stick with our model tax rate, our long-term model tax rate is 18%.

All right. Thanks a lot, guys.

Mike Crawford, B. Riley & Company.

Regarding this new flash manufacturer, Yangtze River Storage Technology, what if any support resources do you expect to apply to this entity?

We cannot comment too much. I think the Governor in China have a China play they had -- they really are recruiting more people and looking for the strategic honor and the license but then you license on technology. But we keep as a [prioritization] with them. Where ever they do have NAND coming within two years, we will support them. And although everybody understand the challenging, overall I think more flash is better for us and we have a broader market to play.

Okay, thanks, Wallace. Then just to be clear on eMMC/UFS, if you look out a period of time, whether it's two years, three years versus today when I believe almost all of your eMMC revenue is coming just from Hynix, do you expect -- what kind of mix would you be targeting for other flash partners to be part of that eMMC/UFS business for you?

I think the today for eMMC primarily come from one NAND maker. We also have two to three multi-makers starts to grow in China channel areas. We believe they are about 10% of our current revenue so we are seeing their continued growth in the next few years.

In addition for UFS, we engage with two NAND makers while we are being productive in Q1 and the other is our primary transponder for mobile. That will be in production in the second half next year.

However for the mobile space, it's not down yet but between EMC and UFS and this highly possibility, high possibility that PCIe one day will move to mobile space if you look at Apple today, they use the PCIe-based solution for iPhone. And AFS PCIe will become really the mobile solution. We think we will have a broader opportunity to engage in multiple NAND makers and for mobile applications. That's how we play.

Okay, thank you.

Tom Sepenzis, Northland.

I was just curious, I know that the cost of the debt is very low but is there any specific thing that you have earmarked that money for? Is that just to hire more people or are you looking at potential acquisitions? What would be the use of that cash?

We have a lot of operating entities in many different jurisdictions. Some of our operating entities are making a lot of -- making strong cash flows while others are not. So in certain places we may need to bring cash into the entities that require investments and generally in the past, we have been using internally generated cash flow for this.

This time around given the very low-cost financing environment, we have determined that the use of external financing is comparable to the use of our own internal funds especially when it's treated as an intercompany loan where you have to arrange it arm's-length basis. You have to account for interest income and the income is then taxable.

So when you factor in all these expenses plus the transfer pricing studies and the documentation, it works out that the use of very cheap bank loans, the cost of these bank loans are comparable to the arm's-length costs plus all the other related costs that we need to build into it. The two costs are quite comparable.

Great, thank you. I was wondering if you could just provide us specifically on the Shannon business in terms of customer wins, just number or who and what type are you seeing in the Chinese market for that?

We have quite a lot of customers, though our customer base is concentrated at one, one large customer accounting for a good majority, a large majority of our total sales in Shannon and this one customer is one of the BAT Internet companies out of China.

Great, thank you very much.

There are no further questions at this time. I'd like to hand the conference back to our presenters. Please continue.

I would like to thank all of you for joining us today and your continuing interest in Silicon Motion. We will be at the following conferences this quarter: [Stephens] Next Generation Storage, NAND working company in New York; [First and] Technology Innovation Summit in New York; Mizuho Investor Conference in New York; JPMorgan Investor Conference in Hong Kong; UBS Global Technology Conference in San Francisco; Morgan Stanley Investor Conference in Singapore; Bank of New York and Jefferies ADR Conference in New York; (inaudible) industries US Technology Conference in Phoenix. Details of this event will be available on our website. Thank you and goodbye for now.

Ladies and gentlemen, that does conclude our conference call today. Thank you for participating. You may all disconnect.