Ladies and gentlemen thank you for standing by and welcome to the Third Quarter Silicon Motion Technology Corporation 2019 Earnings Conference Call. My name is Anne and I will be your conference moderator for today. [Operator Instructions] Before we begin today's conference, I have been asked to read the following forward-looking statements.

This conference call contains forward-looking statements within the meaning of Section 27-A of the Securities Act of 1933 and Section 21-E of the Securities Exchange 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 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 in 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 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 assume no obligation to update any forward-looking statements which apply only as of the date of this conference call.

I must advise that this conference is being recorded today Wednesday 30 of October 2019. I would now like to hand the conference over to your first speaker for today Mr. Chris Chaney Director of Investor Relations and Strategy.Thank you. Please go ahead sir..

Thank you, Anne. Good morning everyone and welcome to Silicon Motion's third quarter 2019 financial results conference call and webcast. As Anne mentioned my name is Chris Chaney Director of Investor Relations. And with me today is Wallace Kou our President and CEO and Riyadh Lai our Chief Financial Officer.

Following my comments Wallace will provide a review of our key business developments and then Riyadh will discuss our third quarter results and our outlook. Then we will conclude with a question-and-answer period. Before we get started, I'd like to remind you of our safe harbor policy which was just 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 U.S. 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 the market yesterday.This webcast will be available for replay in Investor Relations section of our website 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 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 would like to now turn the call over to Wallace..

Thank you, Chris. Hello everyone and thank you for joining us today. First a few financial highlights before I discuss performance of our key products. Our third quarter sales increased 20% sequentially to $113 million with growth led by our SSD controllers EEMC+UFS controllers and SSD solutions.

Earnings per ADS for the quarter were $0.69 up from $0.52 in the second quarter. Later Riyadh will discuss further details of our financial performance and guidance. Let me start with our SSD controllers. In the third quarter SSD controller sales will continue to account for well over half of our total sales.

Grew approximately 15% sequentially and achieved a new quarterly company sales record. Our client SSD controller sales benefited from favorable NAND flash market conditions rapid client still adoption our controller technology leadership and our strong diversified base of customers.

Our pipeline of SSD controller projects with customers suggested that robust SSD controller sales growth that we are seeing this year well extends through next year.All the major NAND flash manufacturing -- manufacturers have been rapidly reducing their inventory levels and the important inventory reduction method has been to aggressively market client SSD especially to OEM for displacing legacy hard disk drives.

Lower dollar per gigabyte NM prices have created more market demand by a more balanced industry supply demand dynamics of help stabilize market condition and visibility for business activity to the industry food chain. At least half of our SD controllers are now shipping into SSD for pc OEN.

And we expect this trend to grow further as our nano sized customers build more high performance and mainstream yesterday using our controller was the audience and our margin maker customer good more valuable audience the also for PC OEM. We have benefited from growing adoption of SSD by PC OEMs from the global brands to China local brands.

With the price SSD decreasing relative to HDD and lower density SSD cheaper than HDD PC OEMs are actively thinking to increase their use of SSDs.

This dovetails with consumers who are increasingly comfortable using low density drives because of the widespread of their ability of fast broadband connectivity to client storage and client applications.According to Gartner Research 55% to 60% of all PCs this year are shipping with SSD up from about 45% last year.

We believe there remain meaningful upside opportunity for us in client SSD market with over 200 million client devices shipping annually that still using HDDs.

We also believe we are well positioned to increase our share of client SSD controller markets from roughly 1/3 today to 40% as we continue to increase market share as existing customers and our customer gain share in the SSD market. Additionally we continue to have an active dialogue with NAND flash makers who are not our customer today.

And remain optimistic that they may eventually also our controller for their mainstream and value line SSD to us. We will also continue to benefit from our client SSD controller technology leadership. Our controller technology for managing QLC 3D NAND is a match in this industry among both NAND flash makers and merchant competitors.

We supply almost all our SSD controller used for managing QLC NAND shipping today including popular notebook PC from the global brands.

QLC NAND as you know is significantly lower cost compared to comparable TLC NAND but much harder to manage their efficiency.By the fourth quarter over 80% of our SSD controller shipping to NAND flash customers will be for PCIe NVMe SSD significantly higher than industry average and the demonstration of our technological competitiveness.

Our position remains strong as the preferred choice of controller technology by our customer for 90x layer and upcoming 100 plus layer 3D NAND. We continue to make progress with our enterprise SSD controllers. Our Shannon open channel SSD for 2 of China BAT hyperscale customers data center are already shipping with our enterprise controllers.

Separately we already have several design wins with U.S. customers for enterprise class standard NVMe SSD. And a few of these we'll start shipping in the fourth quarter. Our overall enterprise SSD controller sales which had escaped by small steep price mode. Turning to our eMMC+UFS controllers.

In the third quarter our eMMC+UFS controller sales were much better than expected. Sales grew nearly 40% sequentially nearly doubling the growth rate we saw in the prior quarter. The growth was driven primarily by sale of UFS controller to U.S.

NAND flash customers but also by sales of eMMC controller to Chinese module makers.Sales of eMMC controller to SK Hynix remain stable.

Year-over-year our eMMC+UFS controller sales will decline by approximately 1/4 as our Korean customer diversify away from the mobile markets as long as volumes the low end eMMC market and focus on UFS using its own controllers.

Nevertheless, after many quarters of declining sales are EMC plus USA controller have rebounded and we are well positioned for growth driven by growing adoptions of the USS in smartphones, and the more diversified set of EMCs and your customer focus on market. Opportunity under service by the leading them such vendors.

Now I will provide a few comments on our SD solution. Specifically our channel a CD. Shipment of our Shannon SSD fell sharply in the first half of the year.

And as previously communicated our open channel SSD for 2 of China BAT hyperscale customer data center were roughly one year behind schedule due to challenges related to new technology introduction.We have overcome these challenges and in the third quarter Shannon sales started recovering as we began shipping our open channel SSD to our 2 customers.

our open channel SSD are new -- are now commercially deploying in the data center of Alibaba and another BAT. We expect sales of our Shannon SSD to grow further in the fourth quarter and expect gross profit next year to recover to 2017 and 2018 levels as we restructure the business in order to take on less inventory risk.

Riyadh will discuss details of this in his comments. In conclusion over the last 4 quarters growth in our business has been solely depend on the continued strength of our SSD controllers.

Looking forward while we expect SSD controller to remain as a primary growth driver we will also benefit from the rebound in sales of the eMMC+UFS controllers and recovery of our SSD solutions.

With 3 engine of growth all executing we are better positioned for stronger and more balanced growth.Now, I will turn the call over to Riyadh to discuss our financial results and our outlook..

SSD controller sales grew about 15% sequentially similar to the prior quarter. eMMC+UFS controller sales grew about 40% compared to 20% in the prior quarter. And SSD solutions increased about 45% reversing a previous declining trend.

Gross margins in Q3 were down sequentially to 49.8% compared to 51.5% in the prior quarter due to unfavorable sales mix and legacy products.Sales of our low margin SSD solutions increased and legacy controller gross margins declined. SSD controller gross margins were unchanged quarter-over-quarter.

Operating expenses in Q3 were almost flat sequentially. Operating margin in Q3 increased to 22.1% from 18.4% in the previous quarter due primarily to operating leverage. Our effective tax rate in Q3 was 9% compared to 3% in the prior quarter.

Q2 tax rate was unusually low primarily due to a one-time tax benefit which came from the reversal of a previous tax accrual. Earnings per ADS in Q3 increased to $0.69 from $0.52 in the previous quarter.

Stock-based compensation in our operating expenses which we exclude from our non-GAAP results was $2.6 million in Q3 compared to $0.3 million in the prior quarter due to seasonal timing of RSU awards.

We have $333 million of cash equivalents restricted cash and short-term investments at the end of Q3 down from $368 million at the end of the prior quarter.

We paid $10 million in dividends to shareholders the fourth quarterly installment of our $1.20 per ADS annual dividend that was announced in October of last year.Last week on October 25th our Board announced a new annual dividend of $1.40 17% higher than last year's $1.20 and also payable in 4 quarterly installments.

We repurchased $25 million of our shares during the quarter at an average price of $32.82. Now let me turn to our fourth quarter guidance and forward-looking business trends. In Q4 we expect revenue to grow in the range of 17.5% to 22.5% sequentially driven by our 3 key products SSD controllers eMMC+UFS controllers and SSD solutions.

Based on what we currently see from our customers procurement plans we are expecting solid revenue growth next year driven also by these 3 key products. Let me provide color on each of the 3 key products. In the fourth quarter we expect our SSD controller sales to continue growing robustly.

The relatively stable NAND market price that we have been seeing recently has led to better business visibility.Our customers are now more confident in planning business activities and based on their procurement plans for next year we're expecting strong SSD controller growth in 2020 with strong emphasis on PC OEM SSD projects.

In the fourth quarter, we expect our EMC plus us controller sales growth to continue growing strongly.

We expect our EMC and us controller sales growth to continue through next year, driven by both accelerating UFS adoption by smartphone OEM with recent availability of application processors that support UMCP and going strength of our diversified base of module maker customers.

We believe growth from us adoption and module maker customers next year will more than offset further downside risk related relating to our Korean customer. In the fourth quarter we expect our SSD solution sales to continue recovering with growth from both Shannon and Ferri.

While Shannon gross margins will continue to increase next -- gross profits will continue to increase next year as new programs with Alibaba and other customers enter production.Revenue growth will be more limited as we start transitioning our business arrangements with Alibaba to a consignment model in the first half of next year.

Separately we are excited about our Ferri opportunities next year especially for automotive applications. In the fourth quarter gross margin is expected to be in the range of 48.5% to 50.5% comparable to Q3.

There could be some downward pressure on our gross margins in Q4 because of increasing sales of SSD solutions and we also expect Shannon gross margins to remain under further pressure in Q4 before improving early next year when we start transitioning Alibaba sales to a consignment arrangement.

Fourth quarter operating margin is expected to be between 21.6% to 24.6% comparable to Q3. While we are expecting meaningful gross profit upside in Q4 from higher levels of sales improvement in operating margin will be limited because of higher operating expense in Q4 due to R&D tape outs for new controllers.

Stock-based compensation expense in the fourth quarter is expected to be in the range of $7 million to $8 million.Effective tax rate in the fourth quarter is expected to be approximately 15% our model tax rate. Let me now update everyone about the restructuring of our SSD solutions product group.

Last quarter we disclosed that sales outlook for Shannon had been deteriorating which led to a $5 million write-down of NAND flash components and SSDs in inventory and the risk that a significant portion of Shannon's $34 million of acquired goodwill in intangible assets may be impaired.

This quarter after completing our impairment evaluation we wrote down Shannon goodwill by $16 million because the present value of our expected Shannon cash flow was determined to be less than the sum of our purchase price plus subsequent investments.

We have also been restructuring our Shannon's operations to achieve better operating integration and efficiencies.And have changed our business arrangement with Alibaba to where NAND is consigned. And with the transition to this new arrangement which will. Start in the first half of 2020.

We will reduce our exposure to NAND flash volatility and capital tied up financing inventory. Separately, we have been organizing the operations of our small big Tara Software Defined storage product line and in q3 recognize the $2.7 million bills returned from customers.

This follows a complete $4 million write down of our big Tara goodwill and intangible assets at the end of 2018.

With these asset impairment issues behind us and with our core SSD controller sales well positioned to continue growing robustly our eMMC+UFS controllers making a comeback and our SSD solutions recovering and casual visibility -- cash flow growth and visibility improving our Board announced last week a new annual dividend per ADS that is 17% higher than last year's.

This concludes our prepared remarks.We will now open the call to your questions..

[Operator Instructions] First question comes from the line of Anthony Stoss of Craig-Hallum. Go ahead..

Where you guys, congrats on the old SIMO on SSD starting to show its head again.

Riyadh probably for you I'm curious if you can quantify how much the net loss would be for 2019 from the Shannon Systems group what the gross margins would be? Just so we get a clearer picture on true profitability on Silicon Motion? And then if you could take a stab at after the consignment transition is complete what would be kind of an expected range of gross margins for all of SIMO? And then I had a follow up..

First of all, on the first part of your question our Shannon product line the gross margins have come down significantly. Last year our business was roughly about 20% type gross margin. In the most recent quarter we were down effectively to 0.

And we expect to stay at that level in the fourth quarter; before the profitability of this business starts recovering in early next year when we transition to a consignment business model. But until then our business is -- our Shannon business unit is not profitable we are making a bit of a net loss of roughly around $5 million..

And then what would you expect the gross margins to be after the consignment is complete the transition?.

Unfortunately, I can't give you more clarity on that because this is still taking place. We're going to be working one step at a time. First, we're rolling out the consignment for our Ali projects.

We have roughly half a dozen projects with Ali going into next year and we're going to be rolling out the consignment but there's still a lot of implementation that needs to take place. So this will -- the pace of how this is rolled out will determine how what sort of improvement.

But without NAND passing through for example on our Ali our product sales to Alibaba are -- without the NAND the profitability the gross margins of this business will improve significantly.

Because as you know in a typical SSD NAND accounts for roughly 80% of the bill of materials and without the NAND flowing through while our revenue will come down, we -- we're not -- without having to carry the NAND costs our gross margin becomes so much higher..

And then one quick follow-up. Can you quantify the Q4 or any tapeouts how much it was? And maybe go a step further and give us a ballpark of what the March quarter OpEx should look like.

It's -- your OpEx is up about $5 million from Q3 and Q4 does it revert back to kind of a Q3 level in Q1?.

So tapeouts in Q4 are going to be meaningfully higher than in prior quarters including in the third quarter. We're going to have roughly about $4 million of incremental tapeouts for new controllers in the fourth quarter. But as we go into Q1 our overall tapeout and operating expenses should revert back to more normal levels..

Thank you. Nice execution, guys..

The next question comes from the line of Craig Ellis from B. Riley FBR. Thank you, please go ahead..

Yes, thanks for taking the question. And guys congratulations on the good execution and on the nice dividend increase last week. So Wallace I want to start off by following up on some of your prepared remarks as it relates to the SSD controller business.

As you work with your NAND OEM customers and interact with them about what their end customers are seeing.

Can you give us a sense for all the benefit to SIMO next year if we get an increase in PC SSD penetration that would be perhaps similar to this year? And if the company's market share can rise from 33% to 40% when do we see that impact the business? Does that flow into the business in the first half of the year? Does it go into the back half? How should we think about the linearity of that increased penetration in the market share gains of the company?.

We -- thank you for the questions. We cannot comment on the detail number how much we can grow for PC OEM for next year. But from current design pipeline we have very high confidence we definitely will grow at least 10% to 20% as the market grows. Because we currently have 4 to 5 customer support top 3 PC OEM. Next year we probably will add 2 more.

So I think from current design pipeline we are very confident we can -- we're going to gain market share our customer will gain market share in PC OEM..

But Craig, let me further add the 40% market share target is not a target that we've set for ourselves for next year but rather a target that we can achieve in a few years' time. We believe our market shares are going to continue to inch up for the many reasons that we talked about but likely we will probably not hit 40% by next year..

And then Riyadh I wanted to follow up on one of the fourth quarter guidance points around gross margin. So if I rewind the clock to three months ago the implied guidance for fourth quarter would have had a gross margin decrease towards 46%.

But now clearly the outlook for the fourth quarter is much, much stronger at 49.5% we're in the middle of the target range.

Is the company now confident that they can execute the business to target gross margins through calendar '20 with the improvements you're seeing in the way you're managing the SSD solution business?.

It is possible we're quite confident about how our controller gross margins are -- have been performing how we have been able to deliver on stable gross margins.

We have the benefit that now the NAND price NAND supply demand dynamics are now a lot more stable a lot more favorable to all the parties in the food chain the pressure on us and other participants in the food chain have now alleviated right.

So we have less pressure on our controllers plus we're moving our SSD solutions business arrangement to a consignment model much faster than we had originally anticipated. Earlier this year we thought that as we moved to a consignment, we probably will not be able to implement until the end of -- towards the end of next year.

But it looks like we're moving a lot faster and we'd be able to -- we should be able to start having the benefit of this sometime in the first half of next year. So ne net these are all very favorable trends for our gross profitability going into next year..

And if I could just ask one more. Very encouraging to see that eMMC and UFS business start to bounce back.

The question is since that business is running at mid-20s to mid-40s quarterly revenue range over the last 10 quarters is the visibility that the business can grow through the low end of that range towards the middle of the range? Is there potential for the business ever to get back to the mid-40s? If you could help us understand the potential of that business over the next 4 to 6 quarters it would be helpful..

Well, with the current design market trends we do see a strong demand for UFS transitioning from premium line and expand to mainstream smartphone so they help us to grow UFS with one of our major U.S. NAND customers.

And in addition I believe because there is -- we have quite a lot of the eMMC design for value line for smartphone we do see tremendous growth in design pipeline from our China module makers. So that really encourage us to see the strong momentum to growth that's why we're seeing from Q3 we see the momentum will extend to the Q4 and to 2020.

On current forecasts we do see because the market transition from eMMC to UFS they helps -- they really help for us -- help us to see the future and growth momentum and design pipeline through our major partner. And we believe we might add 1 more major partner for 2020.

But then looking forward we believe our mobile sale revenue will back to our highest level like at 2016 and 2017 level..

Thanks, Riyadh..

Thank you. The next question is from the line of Karl Ackerman of Cowen. Please go ahead..

Hi, good morning and good evening. First, I want to just touch on mobile. It has been a follow-up to Craig's comment -- question then.

How the same multi CIG mobile controller business will be into the fourth quarter and into at least the first half of 2020? And I guess I asked that because it would appear there might be a build-up in UFS inventory that was destined for Huawei.

So I'm wondering if you could provide any color on how you think sales to OEMs today perhaps even on a broader basis across your own customers? And I have a follow-up..

I believe Q4 our mobile sale revenue is very solid from current designing and the pull and the procurement from our customers. We do see the design win is beyond Huawei, it has expanded to Xiaomi, Oppo, Vivo and Samsung Mobile.

And we do also have the confidence because China module makers start to win value line smartphone because the majority of NAND makers are moving away from low density eMMC. And that help us to see the market growth on China module makers.

And we do see the open trend is favorable to us because there are more transition in smartphone from eMCP to uMCP in 2020. That are really helping us to see the visibility of our design pipeline and sale forecast is very, very strong. We believe we will continue growth on mobile storage product in 2020..

I guess given your unique position within the supply chain what are you seeing in terms of 128-layer NAND development? The integrated Korean NAND provider is already there.

But what do you think the industry transitions to 128?.

It's a pretty good question. I think now all NAND maker except China maker all have 100-plus layer TLC or QLC sample. We have been tested and provide and saw more development for all the customer. It all depend on the NAND maker the capital investment and how they view about their total output and planning.

I think different NAND maker have a different strategy toward 2020. But frankly speaking we believe from May 2020 majority NAND maker will transition from 96-layer 92-layer into 100-plus layer that's between 112 to 144-layer TLC QLC production ramp.

But the percentage depend on each NAND maker capital investment and how quickly they will ship to more from 30% to 100-plus layer product..

Anne, can you go to the next question please?.

Yes, thank you. Our next question comes from the line of Mehdi Hosseini of SIG. Go ahead..

Yes, sir. Thanks for taking my questions. Couple of follow-ups. I want to find -- a better understand how you see your current revenue mix by channel versus OEM versus NAND manufacturer. Any color here will be great. And I have follow-ups..

We see Q3 our PC OEM and channel are roughly 50-50. But we see the Q4 PC OEM portion will grow stronger than channel because we see our Q4 will maintain a good momentum for clients the controller growth..

And then what about as your Korean customer -- the mix of Korean customers reduce should I assume your exposure to the NAND manufacturers would be less? Or any color there?.

If you're talking specifically about our eMMC controller within that space with our Korean customers coming down we also have our U.S. customers going up.

And so this is a leading to a situation where if you were to look at our eMMC UFS revenue stream for next year we believe that there could still be certain -- some downside risks relating to our Korean customer. But at the same time the growth momentum that we have from our U.S.

customers and from our China module makers these upside opportunities are more than sufficient to cover the downside risk that we anticipate from our Korean customers for next year..

And then moving on to Shannon; when you talk about Shannon seeing growth into December and growth system into 2020 is that driven by the open channel or is -- or other parts of Shannon? And if it is open channel how should I think about the opportunities in 2020?.

See primary growth for Shannon moving from Q3 to Q4 or 2020 is based on open channel SSD solution. And we have multiple program with Alibaba and the other customer from BAT. And we do believe because the new business model with a NAND consign, we will -- can be more aggressive to engage business with Alibaba.

Because as you know this year due to NAND price decline sharply really, we hesitated to take a more business with Alibaba because the -- because margin is almost close to 0. But if I think for next year we're willing to grow as fast as we can with Alibaba. And from other customers I think we probably sell the solution.

So from net-net we see the overall Shannon business going to grow much healthier than this year and much stronger than 2019..

And is this going to be a consignment of the raw material NAND? How are you going to be able to leverage your existing relationship? In other words would you facilitate or have your U.S.

based customers to supply the necessary NAND to Alibaba and this way there will be a win-win situation for everyone?.

So as we said before in China enterprise SSD solution is always selling solution. For U.S. enterprise we only focus on enterprise SSD controller business. We don't have value to sell enterprise SSD solution in U.S. to compete with Samsung Intel and other NAND makers.

But I think the -- through the collaboration engagement with Alibaba we gained much broader and wider potential solution not just open channel and other in -- such as the key value. And also hybrid SSD this all potential development and help us have a much better opportunity to engage progressively into U.S. customer..

And just one more follow-up.

How do you see the mix of NAND used in 2020 between the QLC and TLC? And how do you see it changing from this year to next year?.

As you can see QLC today primarily produced is for Intel and Micron. And moving to 2020 I believe all NAND maker going to produce QLC. However the total output probably is just between 10% to 15%. And -- but we do have really good news because PC OEM it takes a long time to accept QLC based SSD.

Now all PC OEM they can take QLC SSD based for the value line. So we see the client SSD controller will grow not only for the high end and the mainstream but also value line. And for the long-term we see QLC become very important role and product mix for both client SSD as well as datacenter.

Because for datacenter storage SSD QLC probably with the best quality package solution to meet the customer needs. And with the hybrid solution and equated with a persistent memory SSD and then QLC based SSD probably will be long-term solution for datacenter to balance the cost and performance..

Would you actually be able to get a better pricing on QLC versus TLC for the same product or for the similar application?.

I cannot comment for that. But I think that we all based on customers their volume and their end customers and to do the pricing. And this is really all customer driven and pricing is sensitive so I cannot comment for that..

Sure. Okay, very helpful. Thanks..

Thank you. Our next question comes from the line of Rajvindra Gill of Needham & Company..

Yes, thank you. Congrats as well on the good results. Well the first 1 wonder if you could kind of elaborate further on what you're seeing in the NAND environment? You'd mentioned some sort of stabilization in the pricing dynamic. And your PC OEM customers are more confident and business procurement.

So I was wondering if you could maybe discuss how that has changed over the last say six months --.

I believe -- go ahead..

And how do you see the NAND supply demand dynamic going forward into the first half of next year?.

We see after the CQ shots exit Toshiba Memory the Yokkaichi fab power outage and NAND makers start to leverage the reasons and to really moving the price up. Now from September moving to -- and into Q4 the price -- NAND price is relatively stable that is helping all the player in the food chain to place the activity in the business.

And we see that's why the demand is getting stronger as the NAND is much stable. Into Q1 2020 due to seasonality and -- see the NAND price could be flat or slightly decline this all depends the demand and supply. Because most of the NAND maker today the inventory is back to normal level.

So there's no need for them to aggressively pushing the price or moving around the extra inventory. So we believe the price will be stable and -- into Q1 as well as the whole '20..

And you mentioned Riyadh an improvement in the profitability for your client SSD controllers. In the past you had mentioned that a lot of your customers -- your flash maker customers were buying cheaper DRAM-less SSDs to clear out that excess inventory.

And there was about a 20% discount in your ASPs relative to those 2 products the DRAM-less versus the DRAM-SSDs. Can we talk about -- I would assume logically that as you start shipping more PC OEM business that the ASPs for your controller -- client controller SSDs should start to kind of move back up to more normal ranges.

So just can you remind -- could you let me know if that's the case.

Should we start seeing more stable ASPs in client SSD specifically and hence a return to more normalized gross profit from that business?.

Our control ASPs for our products -- all our products including our SSD controllers are generally stable over time but they do fluctuate somewhat quarter-after-quarter up or down based on product mix and other factors. And we saw some of this in the previous quarters of this year.

Our ASPs are generally higher for newer or more advanced SSD such as PCIe NVMe versus legacy Shannon our ASPs are also higher for full feature SSDs such as with performance oriented with DRAM SSDs versus lower cost DRAM-less.

We also offer our module -- our flash -- NAND flash makers the module makers a broad range of SSD controllers from high performance to mainstream and value line and procurement patterns could change from time to time. In Q2 we saw strong sales of controllers for DRAM-less SSDs. But in Q3 sales were much stronger for DRAM with DRAM SSDs.

Our sales mix is also quickly migrating to PCIe NVMe and 1/3 of our sales in Q4 last year were NVMe. But by the fourth quarter of this year at least half of our controllers -- SSD controller sales are going to be NVMe. Hope that's helpful..

The last question on Alibaba. You mentioned moving to kind of consignment a business arrangement where you would have lower revenue but higher margin because you're not selling the raw NAND.

Where are we with that kind of process do you think that's in the works and Alibaba is going to agree to that? And I think in 2018 or 2017 you were doing about $100 million of Alibaba and Ferri revenue combined should we expect revenue contribution overall but higher margins.

Just want to get a sense of where we are in that process and how we think about the economics of that?.

We see in this year because we suffered a sharp NAND price decline this is several businesses and we cannot deal with Alibaba because other NAND makers are much more aggressive in pricing because they try to move their excess inventory. Now we believe Alibaba have agreed will consign for nine consign this model.

And we probably going to use at least 2 different kind of NAND that's selected by Alibaba. Because I think what's happening in first half of next year and -- but we just don't know exactly the date because there are certain procedures and the procurement process and legal document need to be signed.

And -- but that would really help us because Alibaba is a major business for our Shannon business. And when it became consigned the gross margin and all the business, we can be much more flexible aggressive to take because -- and there is guarantee we make a profit. It doesn't matter NAND is excess of the supply bigger in demand or shortage.

So that is where makers are very helpful to balance the other SSD solution sale to other customer in China..

Okay, thank you..

Thank you. Next question is from the line of Gokul Hariharan of JPMorgan..

Hi, thanks for taking my questions. First of all could we talk a little bit about the progress on enterprise controllers both for open channel SSDs as well as for other enterprise customers I think you started shipping something in Q3 and potentially increase in Q4.

Could we take a stab at when does enterprise controller revenues really become material let's say 5% of your total SSD controller business or something like that? And I had a follow up..

Yes, our enterprise controller began shipping in third quarter and through our channel SSD to Alibaba and other customers in China. But in Q4 we expect to begin shipment to U.S. customers for enterprise SSD. Volume will be very small this year.

I think you know our enterprise SSD controller it can be used for both function open channel SSD and standard NVMe. We also start to add more features to provide key value into the NVMe solution. That help us to customize the solution tailored for specific customers. That can differentiate on standard NVMe.

And we believe that will increase the demand for our enterprise controller for the U.S. customer also help our SSD enterprise SSD solution sales in China for 2020..

Let me also add we are not sure when we can have our enterprise, we can grow our enterprise SSD controller sales to where it accounts for say 5% of our total sales. We have been making very good progress with our customers in China through Shannon as well as our U.S. customers where we will begin initial shipments in the fourth quarter of this year.

So we are making good progress but we still are unsure how quickly we can scale our business. But that outlook is certainly very optimistic for us..

So traditionally, Q4 is not a peak season I think we are seeing a pick-up in every business. Could you characterize a little bit in terms of what is the driver for demand? And are you seeing any kind of pull ins because of tariffs or anticipation of tariffs etc.

given that also your other supply chain peers have kind of commented as that potentially is the reason for some of the pull in that they have seen in Q3 as well as in Q4..

I think there are a couple of reason behind it. The first we do not see any pull ins from customers from Q1 to Q4. Because their design pipeline is very strong and the demand is more globally not just 1 or 2 regions from our sales side. And the main reason because the demand naturally is overall the channel demand is not that strong.

So while the demand is very strong for our controller? We believe the main reason is because the NAND price becomes stable the inventory back to the normal that's why the more -- it's more business involved in the business build up.

And as you can see the PC market this year that has not declined it's become flat or 1% growth from 2018 because new smartphone model for several China customers start to ramp. And we benefited from some transition from eMMC to UFS. As well as our end customer might gain market share in the client SSD.

That's why we do see the strong demand for our controller in all sectors..

So one last question I think if you could take a stab at, I think 2020 I think you mentioned that client SSD you have confidence of growing about 10% to 20%.

Should we expect client SSD controller is still the fastest growing business next year or should we see faster growth for eMMC and UFS given the dynamics in mobile?.

Well, if you are looking at it from a percentage year-over-year percentage growth we also have to deal with a lot of large numbers. Our client SSD controller sales are already well over half of our total revenue accounts for well over half of our revenue.

So it will be growing at rocket rates that we saw in the -- in past years is going to be increasingly difficult because of how big it has become. But to continue growing rapidly for the foreseeable future it is still possible given where we are in the displacement cycle of hard disc drives by SSDs for our client devices right.

So as you know there are about 500 million client devices and approximately half of these devices have already converted from hard disc drive to SSDs. But we still have the remaining 200 plus million units of client hard disc drives that are still shipping.

And these are still opportunities for us to grab for our customers to work with OEMs to transition their products from HDDs to SSDs so there is still a lot of growth opportunities upside opportunities for us for the next few years. So we feel fairly good that our SSD controllers will continue to do very well in the years to come.

But at the same time we have the benefit of our eMMC UFS rebounding because of our position with -- on the UFS side as well as Chinese module makers that Wallace talked about plus, we have the recovery of our SSD solution; specifically Shannon. And the benefit of us going to Ferri automotive applications..

So let me add one comment here; the reason we have such a strong confident to grow our client SSD moving into 2020 that because as you know three years ago, we see the more NAND maker develop their own internal controller for client and enterprise.

But now we see more and more NAND maker come to us because they want to outsource the mainstream and the value line SSD controller to SMI. So this has become more opportunities and much more interesting for us because sometimes we don't even have enough resource to handle multiple project.

Because as they see the mainstream value line SSD become commodity there is really no need for them to develop controller to compete in such a commodity market because the margin will be low. They rather outsource to third party like SMI to do so. So we definitely see the strong momentum from our current existing customer.

We are also looking forward to more customer coming to SMI including NAND makers..

Got it. Thank you..

Thank you. Next question is from the line of Suji Desilva of Roth Capital..

I react with the shift in the SSD solutions business model away from the -- toward the consignment can you give us a sense of what the realistic revenue peak range might be versus past? Like give a sense of the delta there..

Suji, one way to look at it is based on the projects that we have on hand we believe we can scale our SSD solutions business back to what we saw in 2016-'17 type levels.

So when we -- if we were to -- if you were to look at our business activities at those years and to just look at the gross profitability when we move to a consignment business, we should be able to enjoy that sort of level of gross profit levels those type of gross profit levels.

But the revenue associated with that is going to be a lot less because we are going to -- we won't have the pass through of the NAND components to our revenue when we move to a consignment model. But the gross profit level should be comparable to what we saw in 2016-17 type -- those years..

Let me comment beside Alibaba all other customer in China will still sell the enterprise SSD solution. But in 2016 and '17 we don't have a Baidu as our customer we don't have a Baiden right. So we are gaining more diversified customer in China. So they -- that portion will continue growth for sale revenue.

For Alibaba we believe we want to focus on profit and it will help us to aggressively engage much multiple business because then we don't need to worry about the NAND inventory, we don't need to worry about the NAND price. And we don't want to commit with a NAND maker either.

So this would help us to focus on true business technology and make it pure profit from Alibaba such as NAND consignment business model..

Let me also add a bit to what Wallace is saying. Now when we are talking about our consignment business model, we are taking baby steps working first with just 1 customer Alibaba moving into consignment model. For our other customers it could be more limited because of this level of sophistication and size..

And then the SSD controller business it sounds like the 4Q is shifting to a stronger OEM mix in the 50-50 you typically see OEM module maker.

What's the trend into 2020? Is the OEM now growing to a point where it can sustainably be a larger part of the SSD controller market than the module maker or will it be still sort of mixed?.

We believe -- so we believe entering 2020 our OEM portion client SSD controller will be higher than module makers. Because we have more customer provide solution to PC OEM customers..

And then, lastly UFS sounds like you're doing well with the U.S. memory customer.

And you talked about leads with an -- additional customers what timeframe would those potentially come to market if they happen?.

Probably second half of 2020..

Great congratulation on the progress guys..

Thank you. Next question is from the line of Weston Twigg of KeyBanc Capital Markets. Please go ahead..

Hi, thanks for taking my question. I just wanted to follow-up on this consignment business and the revenue impact in the solutions business overall. So what it sounds like is your margins will improve but your revenue growth I thought at one point you said it could get back to that kind of $100 million level that we saw before.

But in the last comment it sounded like maybe that's not the case.

Can you help us just to understand what the revenue impact of the consignment model will be?.

For our Shannon products if we were not to move to a consignment model our revenue levels next year will likely revert back to what we saw in 2016 and '17. But as we move to a consignment model the revenues are going to come down but the gross profit dollars the gross profit dollars are going to be comparable to what we saw in '16-'17.

Without the NAND pass through our revenue will be affected but at the same time our gross profits are going to be protected..

But also, I want to -- I just want to remind you like our sale revenue with Alibaba this year is very small so you can do the math..

And then the other question I have is just on the Ferri business automotive growth in 2020.

You mentioned it earlier in the call but could you help quantify that opportunity for us?.

Yes because Ferri NAND price decline we suffer quite a lot this year. But when NAND price become stable, I think we have a much better opportunity to grow in 2020. So overall, I think we focus on more profit and not just the sale revenue growth. But we focus on more profit growth. And that is our main strategy for our SSD solution business for 2020..

Is it possible to help bracket the type of growth we may see on that Ferri side or is it just hard to say at this point?.

I think Ferri we will -- I can only say it will grow very well in 2020..

Yes, this is one of the means for us to get back. If we do not move -- any how let's say back to the hypothetical that we were not to move to a consignment model which is one of means for us our Ferri product line to bring our overall revenue back to the 2016-'17 type range.

And so it's an important factor for us to drive the -- our business back to where it should be. So it's not just Shannon and there it's about moving the consignment but also contribution from our Ferri products..

Understood, thank you very much..

Thank you. Ladies and gentlemen that concludes our Q&A session. And I would like to hand the conference back to Mr. Wallace Kou, President and CEO..

I would like to thank all of you for joining us today and your continuing interest in Silicon Motion. We will be attending several investor conferences in Asia and the U.S. during the fourth quarter. Details of this event will be available on our website. Thank you in joining us..

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