慧榮科技 (SIMO) 2017 Q4 法說會逐字稿

Ladies and gentlemen, thank you for standing by.

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 Exchange Act of 1934 as amended.

Such forward-looking statements include, without limitation, statements regarding trends in the semiconductor industry and our future results for operations, financial condition and business prospects.

Although such statements are based on our own information and information from other [success] we believe to be reliable, you should not place any undue reliance on them.

These statements involve risks and uncertainties, and actual market trends and our results may defer 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 would now like to hand the conference over to Mr. Jason Tsai, please go ahead, sir.

Thank you.

Good morning, everyone, and welcome to Silicon Motion's Fourth Quarter 2017 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 the review of our key business developments.

Riyadh will then discuss our fourth quarter financial results and provide our outlook.

We'll then conclude with Q&A.

Before we get started, I'd like to remind you of our Safe Harbor policy, which is right 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 market 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 on 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 us today.

Let me update you on our business and markets.

Riyadh will then review our financials and provide out outlook later on this call.

2017 was a challenging year, as the NAND flash industry supply tightness and the high NAND prices limited our growth.

These two issues were especially tough in the first half of 2017.

Tight supply of NAND limited our ability to procure components to build Shannon and FerriSSD solution.

More critically, tight supply also drove up the price of NAND.

The high price of NAND, in turn, affected the affordability of client SSD and renewed demand of our SSD controllers.

By the second half of the year, however, NAND was no longer tight, as supply had increased meaningfully from the [RAM] of a new 3D NAND capacity, and the price was softening.

By the fourth quarter, sale of our SSD solutions were continuing to grow, and our client SSD controller had started to rebound.

Overall, we expect NAND supply and lower prices will continue to improve further in 2018.

We believe NAND industry supply will continue to grow in 2018 due to a number of factors: first, our 64- and 72-layer 3D NAND manufacturing yields continue to improve.

Flash makers will further accelerate production and [beat] output through all this year.

Additionally, several NAND flash makers have already started sampling 4 the bit per cell QLC 64-layer NAND, and industry supply will grow as these QLC components enter commercial production by mid-year.

By the end of this year, NAND supply will grow further as new 96-layer 3D NAND entered production.

Recently, as a result of increasing expectation of the industry supply growth, NAND makers have, across-the-board, been increasing their industry [bit supply] growth forecast for 2018.

The cost per bit of new generation of NAND is falling rapidly, as 64-layer has ramped up over the past few months.

We have already seen pricing begin to soften.

We expect that 64- and 72-layer production ramping further supplemented by the introduction of QLC and the migration to 96-layer later this year.

Each of these new generations of NAND will further drive down the dollar cost per bit of NAND, improving affordability and adoption across many end markets.

The price of NAND directly affected the price of client SSD and therefore demand for client SSD, the most price-sensitive of all of the major indications for NAND.

As a revenue point of pricing and adoption, when the OEM price for 128-gigabyte client SSD fell to $28, $25 in 2016, demand for SSD was very strong as client HDDs were rapidly displaced.

In contrast, when OEM prices of these 128-gigabyte SSD rose to at least $40 in May 2017, comparable to the OEM price of HDD, demand for SSD stagnated.

Demand for our client SSD controller contracted sequentially for three quarters, and did not rebound until the fourth quarter when the price of NAND softened and the OEM price of 128-gigabyte client SSD had fallen to $30.

We expect demand for our client SSD controller to grow faster when NAND prices and SSD prices fall further in 2018.

The [rebound orders] in the fourth quarter for our client SSD controller were led by our NAND flash partners.

Based on our [book it] and backlog of [others] we expect our client SSD controller to grow further in the first quarter, again, led by demand from our NAND flash partners.

Our most [volume] of controller sales are to OEMs.

Module makers also generate a meaningful amount of revenue.

Our sales to module makers were flat sequentially in the fourth quarter, as these customers believe NAND prices will fall further and likely not increase their order of SSD controllers until they see lower NAND prices.

We exit 2017 with a large portfolio of design wins at our [3D NAND] flash partners for our second generation PCIe NVMe controllers.

Client SSD using this NVMe controller designed by Silicon Motion benchmark very favorably against SSD using our competitor's controller and place us among the best in class.

Of the SSD market transition from SATA, where we have a disproportionally-large market share, to faster NVMe, where we have secured our disproportionally-large share of new merchant programs, we believe our overall client SSD controller market share will likely increase further this year.

A material portion of our new NVMe controllers are for managing the upcoming QLC 64-layer 3D NAND.

Our flash partners plan to start shipping by mid-2018, client SSD using their QLC NAND and our controller.

Based on the strong performance and lower cost characteristics of QLC NAND, we believe this new QLC SSD should be very well-received by OEMs and the market.

Additionally, given the continuing shortage of DRAM supply and very high price of DRAM, many customers have been actually seeking DRAMless SSD controller solutions.

We are the leader in DRAMless SSD controllers, and as many of you may have seen, third-party industry reviews demonstrate that SSD with our DRAMless NVMe SSD controller benchmarked very favorably, even against SSD with DRAM.

Typically, DRAMless SSD have lower [bill of materials] compared with SSD using DRAM, but are inferior in terms of performance.

Through the use of our unique and proprietary controller technologies, with RAID protection, we have been able to quickly reduce the performance handicap of DRAMless SSD to where they can compete not only in the low-cost SSD segment but also increasingly in the mainstream segment.

In the last few years, we have been devoting an increasing portion of our SSD engineering resources towards the data center SSD market.

As previously discussed, we are working on data-center-grade open-channel SSD controllers for two hyperscalers.

We delivered our first pre-production commercial sample to our first hyperscale customer at the end of 2017, and anticipate our open-channel NVMe SSD to enter commercial production by mid-2018, which will likely be the world's first open-channel NVMe SSD controller to enter commercial production.

Separately, we are on track to begin commercial sampling with our second hyperscale customer by mid-2018.

We believe we are able to very effectively bring this new open-channel NVMe SSD controller to our customers because of our unique hardware plus firmware turnkey capabilities, integrated SSD solution design expertise, deep understanding of NAND characteristics, and tight collaboration relationship with our hyperscale customers, especially their storage infrastructure teams.

We are custom-designing silicon firmware, driver software, and SSD solutions for our customers' specific requirements and believe our skillset is unique in the merchant market.

We believe open-channel SSD technologies will be broadly adopted by most, if not all, of the hyperscalers over the next few years, and we are well-positioned.

Let me now turn to our eMMC business.

In the fourth quarter, revenue was stable sequentially, better than originally expected.

Our relationship with SK Hynix remains strong today, and we continue to very effectively support SK Hynix with our very competitive, low-cost, high-performance controllers.

As we have said, we are complementary to SK Hynix' own internal controller program, and nothing has changed.

Their internal program remains focused on high-end, high-density solutions, while we provide a competitive solution for them to address a mainstream, high-volume segment of the mobile embedded memory market.

SK Hynix started 2018 with at least 70 design wins at a large number of smartphone OEMs for their EMC with our controller and their ND and 3D NAND that will [run] throughout this year.

Another 10 design wins were added in January.

For the related UFS segment, we believe the market as [instilled] in the near term will remain small, and expect SK Hynix to continue using their own UFS controller while we will continue to support our U.S. flash partners for their UFS embedded memory roadmap.

Overall, we are confident that our EMC plus UFS sales will remain stable this year, similar to the overall smartphone market.

The global market for smartphones has increasingly matured.

Market conditions in China are weak, and replacement cycles are getting longer, especially for high-end smartphones.

So, the overall market growth should be flat this year.

Now, turning to our SD solutions, revenue increased again this quarter, driven by stronger-than-expected demand for our Ferri industrial SSD.

In 2017, the sale of our SSD solution grew stronger than expected in spite of higher NAND prices, because NAND supply was better in the second half of the year.

As you may remember, demand for our Ferri and Shannon SSD solutions are not particularly price-sensitive, since our OEM customers are more focused on performance, product customization and dedicated customer support.

Looking to 2018, we continue to see strong demand on both our Shannon and Ferri customers.

For Shannon, we are expanding both our product portfolio as well as our customer base, with strong sales growth at Alibaba and other customers.

Our Ferri products are seeing increasing adoption into new markets like automotive, gaming, and telecom server systems.

Based on our growing customer base and increasing design activity, we are confident that our Ferri and Shannon SSD solution revenue will both grow strongly in 2018.

Let me conclude by noting that this year should be a much better year than last year.

We are no longer facing the twin issues of NAND supply tightness and the rising NAND prices.

Today, we are no longer facing any NAND supply constraints.

NAND prices have started coming down but still remain elevated.

We believe that as long as the NAND industries remain highly competitive, NAND prices will quickly trend down, similar to the falling cost of new NAND components, and our business will accelerate, especially our client SSD controllers.

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

Thank you, Wallace, and hello, everyone.

I will summarize our financial results and then provide our outlook.

Before I begin, I would like to reiterate that our comments today will focus primarily on our non-GAAP results unless otherwise specifically noted.

A reconciliation of our GAAP to non-GAAP data is included with the earnings release issued yesterday.

Our Q4 revenue increased 7% sequentially, in line with expectations.

For full-year 2017, revenue declined 6% year-over-year.

For Q1 2018, we expect revenue to decline 3% to 7% sequentially as growth from both of our SSD and eMMC controllers are more than offset by the seasonal decline of our SSD solutions.

For full-year 2018, we expect revenue to increase 5% to 10%, led by strong growth from both our SSD controller and SSD solutions, and stable eMMC controller sales, offset somewhat by declining memory card and flash drive controller sales.

Let me reiterate that while we expect NAND supply to continue increasing, and NAND prices to continue falling, we do not have good visibility on how fast and how much prices will fall.

Our revenue guidance for Q1 in full-year 2018 is based on today's NAND flash price.

If NAND prices were to fall, our SSD controller sales could be much stronger.

We will revise our guidance when we have better clarity on this.

I will now walk you through the performance of our key products, before talking you through key elements of our P&L and the rest of our financials.

In Q4, sales of our SSD controllers grew over 15% sequentially as several new programs at our flash partners ramped.

For full-year 2017, our SSD controller sales declined a little over 10% and accounted for almost 1/3 of sales.

In Q1, we expect our SSD controller sales to continue to grow as our flash partners' SSD programs scale further.

For full-year 2018, based on today's NAND and SSD prices, we believe our SSD controller sales will likely grow roughly 20%.

If NAND and SSD prices were to fall further, our controller sales could be stronger than currently anticipated.

While we believe price elasticity of demand for client SSDs is high, we do not believe it is prudent today to speculate and try to quantify potential upsides from falling prices without hard facts.

In Q4, sales of our eMMC controllers were stable sequentially, as SK Hynix began rebuilding of their depleted eMMC controller inventory earlier than expected.

For full-year 2017, our eMMC controller sales declined about 5% and accounted for almost 1/3 of sales.

In Q1, we expect our eMMC controller sales to increase sequentially, as SK Hynix accelerates the rebuilding of their eMMC inventory.

For full-year 2018, we believe our eMMC controller sales should be flat, in line with 2018 smartphone industry forecasts.

In Q4, sales of our SSD solutions grew over 15% sequentially because of strong Ferri industrial SSD sales.

For full-year 2017, our SSD solution sales grew over 15% and accounted for almost 20% of sales.

In Q1, we expect our SSD solution sales to decline meaningfully quarter-over-quarter, largely because Shannon's new 2018 data center SSD projects will restart from a low base.

For full-year 2018, based on our current portfolio of Shannon and Ferri customers and projects, we believe our SSD solutions should be able to grow roughly 20% with strong growth from both our Shannon and Ferri products and with seasonally strong sales in Q2 and Q3.

Our Q4 gross margin increased to 46.6% from 46% in the previous quarter, largely due to gross margin improvements of both our controllers and SSD solutions.

We expect our Q1 gross margin to improve further to within a 46.5% to 48.5% range, as we anticipate selling more higher-gross-margin controllers, and less lower-gross-margin SSD solutions.

For full-year 2018 we expect gross margin in the 47% to 49% range, in line with our 48% gross margin in 2017, as we expect our 2018 mix of higher-margin controllers and lower-margin SSD solutions to be roughly similar to 2017.

Q2 and Q3 gross margins could be towards the low end of the range, due to seasonally-strong SSD solution sales.

Our Q4 operating margin increased to 24.1% from 20% in the previous quarter.

Operating expenses in Q4 were lower, primarily because of lower R&D expenditures.

We expect Q1 operating margin to decline to 21% to 23%.

Operating expenses in Q1 will increase largely because of higher R&D expenditures.

For the full year, we expect operating margin to be in the 23% to 25% range, above our 23% operating margin in 2017.

We expect operating expenses for the full year to increase less than our 5% to 10% revenue growth guidance range.

Total headcount at the end of Q4 increased to 1,250 employees, which was 36 more than at the end of Q3, and 128 more than a year ago.

We are planning on adding fewer headcount in 2018.

Our effective tax rate in the fourth quarter was 17%, lower than the 23% in the previous quarter.

We expect our effective tax rate to continue to decline this year, the benefit of recent corporate restructuring activities.

Average effective tax rate for the full year 2018 should be roughly 15%, lower than 2017's 21% effective tax rate.

Our fourth quarter EPS was $0.79, higher than the previous quarter's $0.57.

Full-year 2017 EPS was $2.81, lower than 2016 EPS of $3.64.

To help you reconcile our non-GAAP results just discussed with our GAAP results, I will highlight two key differences.

In Q4, stock-based compensation in our operating expense, which we exclude from our non-GAAP results, was $7.9 million, higher than the $3.3 million in the previous quarter, due to the seasonal timing of RSU awards inconsistent with timing of past years.

Full-year 2017 stock-based compensation in our operating expense was $15.2 million, lower than 2016's $17 million.

For Q1, we expect stock-based compensation of $3.2 million to $3.3 million, and $15 million to $17 million for full-year 2018.

In Q4, we took a $10.3 million goodwill impairment loss, which we excluded from our non-GAAP results.

We regularly test for impairments relating to the value of all of our acquisitions and recently determined that our FCI business is valued lower than our carrying cost.

Let me now talk about a few key items from our balance sheet.

We had $366 million of cash, cash equivalents and short-term investments, [$34 million more] than in the previous quarter, and $89 million more than a year ago.

In November 2017, we paid $10.7 million of dividends to shareholders, the first $0.30 per ADS quarterly installment of our annual $1.20 per ADS dividend that was announced in October.

For full-year 2017, we paid $32.1 million of dividends to shareholders.

In Q4, we did not repurchase any shares related to our $200 million share repurchase program.

This share repurchase program was authorized in July 2017 and is set to expire in July this year.

As you may remember, we intend to only opportunistically repurchase our shares when we believe they are significantly undervalued.

In line with this principle, we put in place a trading program to repurchase a maximum percentage of our daily trading volume if our share price were to fall to a level around our 2017 Q2 52-week lows.

Depending on how our share price performs over the life of our 12-month repurchase program, we could be utilizing a significant portion of our $200 million, buying nothing, or some amount in between.

This concludes our prepared remarks.

We will now open the call to your questions.

(Operator Instructions) We have the first question from the line of Mike Burton.

You mentioned you expect client SSD controllers up 20% in 2018.

Is that your expectation for the market growth as well, or do you anticipate any share gains?

And on that competitive front, have you seen Marvell or any other internal solutions begin to affect pricing?

The 20% client SSD gain is based on our client SSD growth.

We believe the market rate will probably be something similar.

This is market growth as we currently expect it, based on today's NAND flash pricing.

As you know, (inaudible) really enough comparable competitors in the market for client SSD controllers.

And then, can you also update us on your progress with enterprise controllers, on any sampling activity, or any chance you'll begin to see that ramp this year?

Or, is it more of a 2019 thing?

So, our open-channel NVMe data center SSD controller strategy remain on track, and we began assembling with our first hyperscale customer last quarter.

We will begin assembling with our second customer in the next few months, it's in the middle of this year.

Initial feedback on our first customer has been very good, and we are confident that our second customer's sampling will also progress [mostly].

We are very likely to be the first in the world to begin commercial sale of open-channel NVMe SSD controllers.

We believe open channel is a high-priority technology for the most hyperscalers, the technology overcome the current [back file SSD] technology and enables active and direct management of storage and latency.

Open channel enables host devices, for example, a server, to manage individual NAND components in SSD, including where data should be stored, and enable SSD to be optimized for high-performance computing requirements.

This year, we do not expect meaningful revenue contribution from our first open-channel project, as it is small-scale.

Open channel concept from controller to SSD, whole device and starting infrastructure still need to be proven before they will scale.

We have the next question from the line of Rajvindra Gill.

Congratulations on rebounding the SSD business.

A question, though, on the eMMC.

You mentioned it was stable in Q4 and expected to rebound a little bit in Q1, as SK Hynix rebuilds inventory earlier than expected.

Can you talk about their procurement plans, how it differs perhaps from say other of their competitors, given that we've seen a pretty steep inventory correction in China handsets at least this quarter and maybe going to the second quarter?

How does SK Hynix' procurement plan differ from other flash vendors?

Yes, we cannot comment on individual NAND flash makers, their procurement strategy and plan, but a lot of it relates not just because of the smartphone market, it also relates to their NAND output allocation from customer to customer.

So, they have several tier 1 customers, and also relates to their allocation between 2D NAND and 3D NAND.

So that's why at certain times we'll get a procurement earlier than the market trend, sometimes it's how they balance their really total NAND output and fully leverage their equipment between 2D NAND and 3D NAND.

And on the NAND pricing, I think it was very prudent of you to say, look, we don't have a great visibility into NAND pricing, although you're seeing signs that it's dropping.

I was wondering if you could somehow quantify it a little bit, like what do you see today in terms of the pricing declining?

And, under kind of what scenario would you say, where you would increase your client SSD forecast if NAND pricing were to fall further?

If it were to fall even further, how much upside do you think it would be on the client SSD?

Is there a way, some sort of correlation, that you guys are thinking about in terms of a drop in NAND pricing leads to X percentage increase in the attach rates for client SSDs?

I think we start to see the NAND price softening, from late Q4, but due to I think NAND makers definitely will try their best to hold the price so the decline rate is really still small.

But, we believe probably in late Q1 or Q2, the decline in price will be escalated.

And you see, [NAND interest industry]is highly competitive, with six vendors from Korea, Japan, and U.S. Soon, there could be #7.

In our competitive industry, other than temporary periods of a shortage in supply, suppliers have consistently passed on to their customer rapid cost savings by introduction of the new technology.

So, we believe the NAND supply will be sufficient, and we'll probably see more competition in the second half of this year, and we should see a bigger [max deal] price drop in the second half of this year.

Raji, Let me also add that there is tremendous dollar-per-bit cost savings, from moving from 2D to 3D NAND, from TLC to QLC, from 48- to 64- to 96-layers, and higher-layer-count NAND, and historically as you know, dollar-per-bit would fall 20% to 30% year-over-year on average.

So, this should lead to some rapid continuing price reductions of NAND prices when prices converge with the cost reductions.

More specific to our client SSD controllers, as you know, we've been seeing improvements in client SSD pricing as OEM prices for, say, the benchmark 120-gigabyte SSDs are now down to $30 from at least $40 mid-2017.

OEM pricing for the next category up, the 256-gigabyte SSDs, are now about $60.

With the greater availability of new and upcoming lower-cost NAND components, there is the possibility that NAND prices could fall by a third within the next 12 to 16 months.

And if that were to happen, the price of 256-gigabyte SSDs will be down to like $40, comparable to OEM prices of HDD, and when that happens, we could see very rapid displacement of remaining client HDDs by SSD.

We have the next question from the line of Charlie Chan.

My first question is, (inaudible) the prior [50 equals assumption] because you seem to assume NAND price will be flat or the same, right?

We all know that the price elasticity, if NAND flash keeps the same why would you assume your [control SSD shipment] can grow 10%?

It should be also flat, right?

So, are you gaining market share, or what's the thoughts behind?

I think we definitely see the market trends, NAND price will decline.

But, I think we have a much broader range of client SSD controller today and are engaged in more projects to date.

But, as I said before, we have to see much better visibility of the NAND price decline, because there's many different [known] factors in the market today.

So, example, second half it might be some smartphones, [top of the dance] to consume more NAND.

So, I think we want to be cautious at this moment, but when we see better NAND supply in the channel, and most of the feedback sees the NAND price might go down from second quarter.

And when we have better visibility, we will revise our guidance.

Charlie, let me also add, the 20% client SSD growth we're seeing today is based on today's NAND prices, and this is coming entirely from the new programs that we are ramping today and will continue to ramp as the year progresses.

So, on top of what we're seeing with our order patterns, with our flash partners, there is potential for higher client SSD growth.

But, we currently just don't have the visibility and believe it's prudent for us to not bake it in at this point and time because we don't know -- to be frank, we don't know when prices are going to fall, and how much they're going to fall.

So, based on that, we think that it's prudent just to bake it, to assume our demand just based on today's NAND flash prices, demand based on today's NAND flash prices.

My next question is on eMMC.

So, you used to have a 4-year visibility.

So, does the customer, is it Hynix, give you a 4-year forecast already?

And, does that come from older?

And any risk that Hynix would switch to in-house during the mid-year?

Let me start off with our answers.

As you know, we have a very strong relationship with Hynix.

Our relationship has not changed at all, and we're starting this year, 2018, with at least 70 design wins at a large number of smartphone OEMs.

This is SK Hynix, starting the year with at least 70 design wins at smartphone OEMs, with our controllers paired up with their 2D and 3D NAND, and these will -- these 70 programs will ramp throughout this year.

In addition to these 70 programs that we started the year, in January alone we've added -- or rather, Hynix has added another 10 programs that will use our controllers and their NAND.

I think to answer your question, it's a pretty complicated equation because for any NAND maker, they're trying to fully utilize their NAND output.

So, it depends how they allocate different generations of 3D NAND and 2D NAND, so that will give us a forecast and allocate certain to the different market sectors.

So of course, other NAND makers would like to produce more for ship a higher density for high-end, but that would depend on their capability and winning the design.

The mature mainstream [handset] market, Samsung [hand is] leader and they are very stable for that market, and we have the market.

So, we can only base on what the forecast we prepared for 2018, for eMMC.

Okay, that's why it's a –

(inaudible) Charlie.

Yes.

Their focus generally is on the higher-end, higher-density products, whereas our solutions are targeted at the mainstream segment of the market where the majority of the volume lies.

So, we help Hynix stay competitive, and we see our resources as being complementary to their own resources.

Our R&D resources help them expand their product reach, and since the number of end markets and applications for NAND continues to grow, their internal bandwidth is also limited.

And so, by working with us, they're able to extend their reach and be able to react quickly and cost-effectively.

Last one, if I may, this will be the strategic -- so, China is now building up its own NAND flash supply chain.

So, how would Silicon Motion fit in this China NAND flash supply chain, and is the company open for a position, an opportunity from China?

And is that [legitimate per tower all]?

We are in regular contact with the Chinese NAND makers and have an excellent relationship with all of them.

We are monitoring their progress closely, and when they begin production we will have a controller ready to support them for a variety of products including eMMC and client SSD.

I believe they are saying that they will be in full production by 2019, are demanding more supplier [mean] more competition, lower price and faster supply growth, positive factors to us.

And by the way, our Cayman company and they're regulated, is not the same as the Taiwan [open company], just in Taiwan.

We have the next question from the line of Mike Crawford.

Shifting gears a little bit, what targets does your software-defined storage platform Bigtera need to hit this year to trigger the $1 million incentive payout?

It's a small matter, so, the incentive program I believe was for last year, 2017.

And for 2017, given that it's essentially a very small organization with limited capabilities, their original expectation was quite low to start from.

And so, when the year had closed, the Bigtera group were not able to achieve their objectives and so the earnout was not paid.

As for the Ferri controllers, you talked about an expanded market with more automotive, gaming and telecom customers.

Do any of those newer applications have the potential to be more significant?

Yes, we do see there's a stronger demand for automotive.

However, the [design] cycle is longer.

We do win several major designs from leading automotive brands, and also the server side, we also see the [boost] storage for servers becoming popular and a stronger demand from a couple of key major customers from China and the U.S.

We have the next question from the line of Suji Desilva.

On the SSD controllers, not sure if you covered this, but what trends have you seen in pricing there?

Are those stable, or any uptick based on mix?

You're talking about client SSD controllers, right?

Correct.

Prices have been very stable, continues to be roughly, on a rough basis, about $5, with the combined trends of our legacy SATA controller pricing beginning to come off.

But, our new PCI NVMe controllers are coming into our product mix at a higher level.

Okay, and then my other question is on the SSD solutions, and maybe the NVMe controllers.

Is the competitive landscape there different than the client SSD, just to understand?

I know some of the products are custom for hyperscale, but if you could walk us through that, that'd be helpful.

Thanks.

Can you repeat your question again, Suji?

You're talking about our open channel?

Yes, SSD solutions, the competitive landscape there versus the SSD client solutions, and also, on open channel, the competitive landscape on those two areas?

So, for SSD solutions we have a two-pronged line.

One is the Ferri, the other is Shannon SSD solutions.

For Ferri, the competitive landscape comes from the NAND makers.

However, that portion is not so attractive to the NAND maker.

For example, for automotive, the design cycle is very, very long and the volume won't be very big.

You need very intense training and support.

So, some NAND makers only want to capture very large programs from leading models, and so that becomes very favorable for our business.

For Shannon business, for enterprise, definitely there are many different competitors.

The majority also comes from NAND makers such as Intel, Samsung, and WD, and [Micron coming].

For China, we don't see a really visible player in the market, although some play a very, very low price.

With this channel, have very good position to build a strong relationship with Alibaba, extend to the second hyperscaler customer, and we believe by end of this year our open channel will grow to a much broader space and heading for 2019.

And 2019 could be the really visible year to grow our open channel business.

(Operator Instructions) We have the next question from the line of Mehdi Hosseini.

I have two, and one for Chairman Wallace.

You talked about NAND pricing trend.

I want to better understand your thought process on availability of NAND.

Most of the manufacturers have talked about a 40% to 45% NAND bit growth.

Is that what you're expecting?

Is that what is behind your ASP assumption?

And if not, how should we compare that to your thought process?

And I have a follow-up for Riyadh.

Our estimation about the big growth for 2018, they're aligned with the NAND makers.

They're just about the big growth, 40% to 45%, that's correct for 2018.

Great, sure, because the reason why I asked that is they also have talked about the pricing trend that you mentioned, kind of a decline in the first half, but then, the increased prospect of prices flattening into the second half.

Is that why there is a lack of visibility that you mention, that has gone into your revenue guide for 2018?

That is also correct, and we do believe we should start to see NAND price decline from late Q1, moving to early Q2.

But, if some NAND makers really position, because there's a certain smartphone maker who would like to double their storage density in the second half new models.

So, that increase the uncertainty, uncertain factor regarding NAND price trend and supply.

However, I think there's just really a two-sided equation.

If the NAND price continues to stay high, and it's really impossible for many makers to double density, NAND prices have to go down, and go down further, in a smartphone with double density.

Frankly speaking, many, many consumers, they are satisfied with the 64-gigabyte or 128-gigabyte.

Not many really want 256-gigabyte, right?

So, we -- but however, NAND makers, they really position that way so we have to comply with their direction while they state, so we try to take a more conservative way to provide the guidance for the whole year.

Got it, very clear, and then for Riyadh, there was a gain from disposal of other assets.

Was that the reason why R&D went down?

How should I think about it?

No, Mehdi, R&D is different from the disposal of asset.

The disposal of asset relates to a piece of property that we sold off, and we had a gain on that disposal of that property.

Separate to that is our expenditures relating to our R&D.

They vary quite considerably sometimes quarter-over-quarter, and for certain quarters we may have a higher expenditure related, and they're generally related to the timing of tape-outs.

Our headcount expenditures generally are fairly smooth quarter-by-quarter, but they may spike up or spike down, depending on the tape-outs of our projects.

So, those are the key movements resulting in the operating expense changes we've had in recent quarters.

Very quickly, what is the pro forma R&D and SG&A for Q4?

We don't provide operating expense, per se, but it's implied from the operating margin difference to the gross margin, the difference between the operating margin and the gross margin.

So, I'm afraid you'll have to work out your math.

At this time, there are no further questions.

I would like to hand the conference back to Mr. Wallace Kou, President and CEO for any closing remarks.

I would like to thank all of you for joining us today, and your continuing interest in Silicon Motion.

We will be attending several conferences in Asia and the U.S. in the first quarter.

Details of these events will be available on our website.

Thank you, and goodbye for now.

Thank you, sir.

Ladies and gentlemen, that does conclude our conference for today.

Thank you for participating.

You may all disconnect.