祥茂光電 (AAOI) 2014 Q2 法說會逐字稿

Please standby. Well, good day, ladies and gentlemen, and welcome to the Applied Optoelectronics Q2 Financial Results conference call. Today's conference is being recorded. Following the presentation, we will conduct a question-and-answer session (Operator Instructions)

I will now turn the conference over to Ms. Maria Riley. Please go ahead, Ms. Riley.

Thank you. I'm Maria Riley, Applied Optoelectronics Investor Relations, and I am pleased to welcome you to AOI's second quarter 2014 financial results conference call.

After the market closed today, AOI issued a press release announcing its Q2 2014 financial results. The release is also available on the company's website at ao.inc.com.

This call is being recorded and webcast slides. A link to that recording can be found on the Investor Relations page of the AOI website and will be archived for 90 days.

Joining us on today's call is Dr. Thompson Lin, AOI's founder, chairman and CEO; and Dr. Stefan Murry, AOI's chief strategy officer and incoming CFO. Thompson will give an overview of AOI's Q2 results and Stefan will provide financial details and an update on AOI's strategy and markets. A question-and-answer session will follow our prepared remarks.

Before we begin, I would like to remind you to review AOI's Safe Harbor statement. On today's call, management will make forward-looking statements. These forward-looking statements involve risks and uncertainties as well as assumptions and current expectations which could cause the company's actual results to differ materially from those anticipated in such forward-looking statements.

You can identify forward-looking statements by terminologies such as may, expect, plan or believe and by similar expressions. Except as required by law, we assume no obligation to update forward-looking statements for any reason after the date of this earnings call to conform these statements to actual results or to changes in the company's expectations.

More information about other risks that may impact the company's business are set forth in the risk factors section of the company's prospectus and reports on file with the SEC.

Also with the exception of revenue, all financial numbers discussed today are on a non-GAAP basis unless specifically noted otherwise. Non-GAAP financial measures are not intended to be considered in isolation or as a substitute for results prepared in accordance with GAAP.

A reconciliation between our GAAP and non-GAAP measures as well as a discussion of why we present non-GAAP financial measures are included in our earnings press release that is available on our website.

Now, I would like to turn the call over to Dr. Thompson Lin, Applied Optoelectronics president, founder and CEO. Thompson?

Thank you, Maria. Thank you for joining us today. We delivered very strong second quarter results exceeding both current bottom line expectations. We achieved our fifth consecutive quarter of record revenue, maintained gross margin of 34.4% and demonstrated solid earnings growth.

Most specifically, second quarter revenue grew 57% year-over-year and 31% sequentially to reach a record $32.7 million, and we achieved record non-GAAP net income of $2.4 million, up 184% sequentially.

Both our data center and fiber-to-the-home products saw our tremendous growth this quarter. Our data center revenue grew 170% year-over-year and 55% sequentially primarily a result of increased share within our existing customer base, fiber-to-the-home revenue growth of 339% year-over-year and 40% sequentially to reach a record $3.1 million.

As we continue to increase shipments of our WDM-PON products for our foundational fiber-to-the-home customer, we continue to execute on our growth plan in order to categorize on our growing market.

In July, we added a key member to our sales team in Taiwan to help broaden our data center customer base. We are quickly ramping our production capacity in order to keep pace with a strong customer demand for our industry leading products.

And I'm very pleased to announce that in July, we shipped our 1 millionth laser so far this year surpassing this annual milestone midyear. I'm sure many of you have heard me say this before but would be worth repeated because it set AOI apart from the crowd and is helping drive our success.

AOI is one of the few optical component suppliers, focused on the optical access market. Our vertical integration from laser chip to equipment coupled with our design and manufacturing capabilities are key to a continuation of our strong growth.

We are very excited by both our current result and bright future ahead of us. Keeping the strong growth demand within our three markets, we plan to build a new facility in Sugar Land to expand our R&D and laser manufacturing capabilities. We expect to start construction in October and have the facility online by the end of 2015.

Finally, many of you have seen our announcement regarding James Dunn's departure from AOI. As we noticed in the announcement, James will be leaving the company on August 15th in order to pursue other opportunities.

James' service to AOI has been very helpful during a very exciting time from the company. We thank him for his contributions and we wish him all the best.

Dr. Stefan Murry, current AOI's chief strategy officer, have been appointed by the board to serve as AOI's CFO upon James' departure. Stefan has been with AOI since the start and has been actively involved in all aspects of AOI's management throughout the years.

His in-depth knowledge of our business and he has demonstrated commitment and leadership capabilities positioned him very well for success in his new role.

With that, I would turn the call over to Stefan to present the details of our Q2 performance and outlook for Q3. Stefan?

Thank you. As Thompson indicated, we achieved our fifth consecutive quarter of record revenue with a very strong gross margin profile. Total second quarter revenue grew 31% sequentially to reach record revenue of $32.7 million.

Looking at revenue by customer, 88% of our second quarter revenue was derived from our top 10 customers with data center revenue providing over half of that total.

Looking at our growth by market, revenue from CATV products in the second quarter was within our expectations of $10.6 million, an increase of 1% from the $10.5 million reported in the same period last year and up 9% sequentially.

We see two near-term catalysts that will drive revenue growth for our CATV broadband products in the back half of 2014 and into 2015. First, the impending DOCSIS 3.1 upgrade cycle, and second, the move to Fiber Deep architectures.

Both of these are driven by increasing consumer demand for more and faster bandwidth and by competitive pressures CATV providers are facing from new market entrants like Google Fiber and other 1G fiber-to-the-home initiatives.

Looking into these growth drivers in more detail, recall the DOCSIS 3.1 is the new high-speed data over cable standard that will allow cable operators to deliver next-generation broadband services over their HFC networks.

As part of the DOCSIS 3.1 upgrade, MSOs will need to deploy new equipment in the head-end or central office, the node and in the customer premise. As the largest CATV ODM equipment supplier, we are actively involved with many of our key CATV equipment customers in developing the head-end and node products that will form the basis of their DOCSIS 3.1 networks.

In Q2, we were awarded four key development contracts by one of our key cable TV customers, specifically related to DOCSIS 3.1 node and head-end equipment and related accessories. We expect to finish most of this development by Q4 of this year.

We believe some cable operators will look to Fiber Deep architectures to improve the customer experience by reducing network congestion. In a Fiber Deep architecture, fewer homes are serviced per individual node and fewer nodes are connected to each head-end transmitter. As a result, cable operators will need to significantly increase the number of nodes in their network compared with their current architectures.

For example, we are starting to see new provisioning strategies that reduce the number of homes served per node from the typical 400 to 500 homes to 150 to 200 homes per node resulting in more than double the amount of nodes in their networks.

Looking at the incremental revenue from this upgrade cycle in CATV, we currently project a $430 million total expenditure in head-end equipment and nodes from North American MSOs over the next four to five years.

We derive that estimate by using the approximately 54 million homes currently using CATV broadband in the United States and 200 homes per node along with the ASP associated with our new DOCSIS 3.1 CATV products.

The revenue we derive from this upgrade cycle will be additive to our CATV revenue from international markets which continue to be strong particularly in China and South to Southeast Asia.

Additionally, I would like to comment on the recent announcements made by some CATV providers that they intend to deploy fiber-to-the-home services in select markets.

From a technology standpoint, there are a number of options for CATV providers to offer FTTH-type services. Some of these include RFoG, GPON, GEPON, Ethernet over coax and, of course, WDM-PON.

AOI is the provider of components for RFoG, GPON, EOC and WDM-PON. So from a technology standpoint, a movement towards fiber-to-the-home deployments by CATV providers should be an incremental growth catalyst for AOI.

Based on all these growth drivers, we expect our CATV revenue to grow in excess of 20% in 2015 compared with 2014. And as we have mentioned on previous calls, we continue to expect CATV revenue growth for 2014 to be in the single digits as MSOs in North America await commercial availability of DOCSIS 3.1 equipment.

Looking now at our data center business, revenue was again very strong and exceeded our own expectations. Data center revenue was $17.9 million, nearly triple the revenue in the same period last year and 55% higher than in the previous quarter.

Our strong data center growth in Q2 was primarily within our existing customer base and consisted mostly of our 10G products although we are pleased to report that 4-G products represented 13% of our Internet data center revenue.

Based on current customer expectations, we believe that 40G products will reach 50% of our data center revenue by the end of 2014, which is earlier than we had previously expected. We believe that this demonstrates both a strong demand for 40G products and AOI's capability to rapidly expand production capacity to match the demand while maintaining the discipline at manufacturing process needed to generate solid gross margin.

Additionally, the advancement of 100G products in the data center is moving faster than originally anticipated. Based on customer requests, we are now working on developing 100G solutions and we expect to complete the development of these products in early Q2 of 2015 with customer qualification to follow.

This is fully in line with customer expectations as is our plan to be in volume production of 100G products towards the end of Q2 of 2015.

Given current market dynamics and customer forecast, we believe that we will maintain our Q2 data center revenue level in Q3 and Q4. While we currently do not see any indication of a deceleration in port growth, we believe that if there is a slowing of data center port deployments in 2015, this effect will be moderate and will be compensated for by the increased ASP of our 40G and 100G products.

We are making good progress in our discussions with our new data center customers as a result of the sales channel incentive that we announced last quarter and our recent hire of a new VP of sales in Taiwan dedicated to the data center market.

With continued growth from our existing customers and the addition of new customers, we believe that data center revenue will grow by more than 45% in 2015 compared with 2014.

Turning to our fiber-to-the-home market, revenue grew 339% year-over-year and 40% sequentially to $3.1 million. During Q2, our first production line in Sugar Land was operating at full capacity and we continued the process of bringing up our three additional lines in Taiwan.

This capacity is needed because we expect to see initial field deployments of our WDM-PON OLT by our foundational customer in Q3 and larger deployments in other cities in early 2015.

However, we are expecting a slower than originally anticipated ramp in Q3 and Q4 due to a one to two-quarter push out at our customers' deployment schedule and we will accordingly delay the build out of the additional three lines in Taipei that had been planned for Q4.

We believe the slight push out in the schedule does not reflect a fundamental shift in our customer's plan. In fact, we have been contracted by our customer to form additional long-term reliability testing as part of their ongoing deployment plans in order to verify long-term reliability prior to larger scale deployments.

We believe this additional testing is a strong sign that our customer continues to make progress towards deploying WDM-PON in new markets so we view this as a vote of confidence by our customer. This testing is already ongoing and we will need to build a meaningful number of units for this testing, so our production lines will continue to operate at close to their full capacity.

Regarding our ONU tunable transceiver, we continue to make progress toward qualification and have had new inquiries about our WDM-PON technology from several potential customers. We believe that AOI is very well-positioned to take advantage of the ever-increasing bandwidth growth within the optical access market.

Moving down the income statement, we are very pleased with the gross margin expansion we've seen in the last two quarters, given the strong revenue contribution from our data center and fiber-to-the-home products.

Q2 total gross margin remained relatively consistent with the previous quarter at 34.4% and well above our 2013 annual corporate average.

Turning now to operating expenses, overall operating expenses were in line with our expectations and while Q2 operating expenses were up on a dollar basis, given our top-line growth, operating expenses as a percent of revenue declined, bringing us closer to our target model.

R&D expense was $4.0 million or 112% of revenue, up $0.5 million from the previous quarter. Consistent with our plan, we continued our investments in 100G data center products, our fiber-to-the-home ONU transceiver and DOCSIS 3.1 technology.

We expect R&D investment to remain at this dollar level throughout the remainder of 2014 with R&D expense continuing to decrease as a percentage of total revenue. We will continue to balance R&D investment appropriately in order to capture market share and promote growth at both the top and bottom line.

Sales and marketing expense was $1.5 million or 5% of revenue, up $0.2 million from the previous quarter primarily due to increased commission cost and the additional sales channel incentives.

Looking at the balance of the year, we plan on maintaining sales and marketing expense at approximately $1.5 million per quarter.

G&A expense was $3.4 million or 10% of total revenue, up $0.4 million when compared to the previous quarter, primarily due to increased personnel cost in Taiwan and additional rent expense associated with the transition to our new facility in Taiwan.

The move from our existing facility in Taiwan is underway and progressing well and we expect the additional rent expense to subside by the end of the year.

Non-GAAP operating income in Q2 was $2.3 million or an operating margin of 7%. And in Q2, we achieved EBITDA of $3.7 million or 11% of revenue. Non-GAAP net income after tax for the second quarter was $2.4 million or 7.2% of revenue as compared with only $0.8 million in the previous quarter and a loss of $0.1 million in the same quarter of last year.

We generated non-GAAP net income of $0.15 per share and GAAP net income of $0.12 per share using a weighted average fully diluted share count of approximately 15.6 million shares.

Overall, we are very pleased with the direction of our business. We've made some strategic investments in R&D and sales and marketing that we expect will help expand our growth and drive leverage to our bottom line.

The target model we outlined about a year ago provided for gross margin of 33% to 35% and operating margin in the range of 17% to 20% on a $60 million quarterly revenue run rate.

Given the current growth trajectory of our business combined with above 34% gross margin, we currently expect to achieve our target operating model in early 2016.

Turning now to the balance sheet, we ended Q2 with $43.0 million in total cash, cash equivalents and short-term investments compared with $61.1 million at the end of the previous quarter.

During the quarter, we repaid $10.7 million in debt, invested $7.5 million in capital investments and increased inventory by $7.3 million.

Accounts receivable increased by $1.2 million to $25.0 million consistent with growth in revenue. Inventory increased to $31.2 million mostly related to our data center and FTTH products.

We expect to maintain inventory at this level as we increase safety stock to meet rising product demand. The aging of our accounts receivable and inventory remained consistent with prior periods.

Consistent with our plan, we made a total of $7.5 million in capital investments in the quarter primarily to expand production capacity for our data center transceivers and WDM-PON transceivers in Taiwan.

This brings our CapEx for the six months period to $13.1 million. As we previously announced, we expect to spend a total of $16 million to $18 million this year on machinery and equipment to expand production capacity.

Including capital expenditures related to our facility expansion in Taiwan, we expect total CapEx for the year to be approximately $26 million.

As Thompson mentioned, in July, we shipped our 1 millionth laser for the year. This is significant in its own right but we believe it reflects a deeper trend towards more and more optical devices being used in higher bandwidth systems.

For example, our WDM-PON OLT transceiver uses 16 individual laser diodes in each unit thus each unit sale of an OLT transceiver represents 16 laser chips manufactured by AOI. Similarly, 40G and 100G data center transceivers will feature multiple transmitter lasers in each module.

As a result, our fab must increase its production rate faster than might otherwise be apparent by a cursory analysis of unit transceiver sales. We believe this trend is fundamental and points to why AOI's vertically integrated model is a competitive advantage compared to companies that rely on external sources for key components like laser diodes.

In order to support the rapid growth of our data center and fiber-to-the-home products and to support our longer term growth plan, we have begun planning an expansion of our R&D and laser manufacturing facility in Texas.

Our plan includes moving out of a nearby rented space and constructing a new wafer fab and R&D center on our existing property directly adjacent to our current building thereby consolidating our Sugar Land operations back into one location.

The new facility will provide enhanced wafer fabrication capacity and expanded laser component manufacturing capabilities. We have designed the facility to provide sustained R&D and manufacturing capacity that we need today and will service well in the future.

We expect the new facility to cost approximately $30 million and we plan to finance it through traditional commercial mortgage lending. We expect construction to begin in October and expect to complete the facility by the end of 2015.

Moving to our outlook for Q3 of 2014, we are entering the quarter with very strong bookings and forecasted demand and we expect to achieve our sixth consecutive quarter of record revenue. Therefore in Q3, we expect revenue between $35.2 million to $36.6 million, representing an impressive 70% to 76% year-over-year growth rate and 8% to 12% sequential growth.

In Q3, we expect non-GAPP gross margin to be in the range of 34% to 34.5%, reflecting a similar revenue mix compared with the first half of the year offset by a slightly higher percentage of CATV revenue.

Non-GAAP net income is expected to be in the range of $3.2 million to $3.9 million and non-GAAP EPS on a fully diluted basis between $0.21 per share and $0.25 per share using a weighted average fully diluted share count of approximately 15.7 million shares.

To provide further visibility for the second half of 2014, we expect total revenue for 2014 to be in the range of $131 million to $135 million and non-GAAP EPS on a fully diluted basis between $0.70 per share and $0.76 per share using a weighted average fully diluted share count of approximately 15.7 million shares.

With that, I will turn it back over to the operator for the Q&A session. Operator?

Thank you, Dr. Murry.

(Operator Instructions)

We'll go first with Simon Leopold with Raymond James.

Thank you very much. A couple of things I wanted to check on. You made a couple of comments that maybe are a little bit conflicting. You indicated you were making an assumption of a similar product mix but then as you spoke about the various line items, I had the feeling that cable TV would be up in the third quarter, data center relatively flat, and the fiber-to-the-home up but on a more shallow lamp than prior expectations. So if you could help me compare the similar mix to the commentary on each of the individual segments if I summarized this correctly.

Hey, Simon, this is Stefan. So what we see basically is data center, you know, being relatively flattish like we pointed out, so you're correct on that.

Cable TV we think will be -- will be up slightly perhaps but not -- maybe not a whole lot. And then FTTH will be again, you know, relatively flat over the next two quarters as we indicated. So you've got that part pretty much correct.

But when we're talking about, you know, a relatively consistent mix, I mean these are not big changes from the last quarter is what we were seeing.

Okay. And then you talked about a little bit of a push out on the fiber-to-the-home business. You mentioned one to two quarters. Do you have insight into the cause of that push out? Is it complications with permitting, delays in construction activity, other resources? Help us understand the why in terms of the push out?

Well, I mean we have had a lot of conversations about that obviously but I can't really give you a whole lot of insight into the nature of the push out. What we know is that -- what I can say for sure is that it has nothing to do with AOI's product, it was a technology, they're moving ahead, you know, pushing our products out.

As we mentioned in the field trials, we've had an internal field trial, if you will, within our customer on their -- within their buildings. That was operating for the last couple of months, absolutely rock solid.

So new don't see any technological reason or availability of products or anything related to the products that we're manufacturing as the reason for the push out. I think -- you know, I'll leave it to you, guys, to speculate on what might be causing the deployment push out but it's nothing related to our products.

Okay. And you mentioned a number in your discussion on the data center outlook for 2015. I believe you said expectations were greater than 45% growth in '15. Did I get that number right?

That's correct, 45% or greater year-over-year 2015 compared to 2014.

Okay. And then one last one, somebody has got to ask about the departure of the CFO, so pretty surprising, stock was trading somewhat weakly today. So if there is anything more you can help us understand why this is occurring, it's a surprise generally to investors and analysts to a bit of a yellow flag. So we'd like some more color.

Sure. Well, as we indicated, you know, James is leaving for personal reasons to pursue other opportunities. I can't obviously elaborate on what those opportunities might be. That's for him to disclose when he feels appropriate.

What I can say is that it doesn't have anything to do with any, you know, disagreement about our financials or any strategic disagreement among the management team, nothing onerous like that.

Okay, thank you for taking my questions.

Thanks, Simon.

Troy Jensen with Piper Jaffray has the next question.

Yes, congratulations on the nice results, gentlemen.

Thanks.

Thanks, Troy.

Hello, Stefan. I'll start with you, you know, on the DC side, data center side. Any new significant wins on the quarter, any new customer wins?

We don't have any new wins in data center in the quarter. We do have a number of ongoing qualifications though with the new customer that we hope to talk a little bit more about on our next call.

Okay. And then just a follow-up on Simon's call or Simon's question, when you say, you know, predicting 4% growth next year, can you just help us with the visibility that you're getting from these data center customers? Most, you know, other optical companies don't have visibility behind kind of the quarter, so just a confidence to say such a big growth rate?

Well, I think there are two things. Number one, you know, as we've indicated previously, I think we do -- tend to have better visibility than most companies in this business because again, we're dealing directly with the large data center operators that are our customers.

So the other thing as we kind of indicated in the remarks, the technology transition that we're seeing from 10-gig to 40-gig and then 40-gig to 100-gig in the -- in the middle part of next year, that's significant because they need to give us a lot of visibility into those transitions so that we can make sure that we're supporting them as they -- as they make those transitions, in other words, making sure that we've got, you know, the production capacity in place and what have you.

So we've had quite a few detailed discussions with the customers about their plans for -- you know, for 2015 and that's why we feel fairly comfortable in talking about it.

It's worth pointing out too that -- you know, I want to emphasize here that every time that we undergo a transition in data rates from, you know, a slower speed to a higher speed, that tends to have an impact on the ASP, in other words, we're getting a higher ASP for these newer generations of products.

So again, we're not necessarily counting on, you know, a whole lot of port growth although again, we don't necessarily see any deceleration in it. But even if that were to occur, the increase in ASP will allow us to grow the revenue at a very quick rate even without, you know, say, a lot of growth in the port count itself.

Okay, now, the last question for me then, you talked about some cable TV design wins. Can you just help us out with the timeline for when those kind of start to monetize and maybe the magnitude of the opportunity here?

Well, as we indicated, what we -- what we have our awards of basically development contracts, if you will, in other words, we've been given the green light to go ahead and develop these products for the customer.

We expect to finish that development by the end of the year. And then revenue should start to flow fairly quickly after that. Now, not all four of those would be exactly at the same time obviously.

So there'll be -- you know, some of them will be earlier into Q4 and some of them will be a little later in Q4 but we should start to see revenue flow fairly quickly after that.

Perfect. Good luck in the second half, gentlemen.

Thanks very much, Troy.

Thanks.

Moving to Paul Silverstein with Cowen.

Several questions if I may. Can you, guys, hear me Okay or so?

Yes, we can hear you.

Yes.

Yes.

First -- so first off, any 10% customers?

Yes, we had one 10% customer in the quarter.

And do you have any thoughts how large a contribution that made?

It was substantial contribution. I think it was 47%.

47%, Okay.

Yes.

Would it be fair to assume that was in data -- that had to be in data center obviously, I mean...

Correct, that was right.

Okay. And in terms of the four large data center customers, are they just large in terms of the size of the customers, not necessarily what they've been buying to date? Were all four -- do all four contribute to revenues in the quarter?

Yes, they all four contributed.

Did all four grow and more importantly when you look out to the future, are you expecting meaningful growth from all four as you go forward?

Let's see. Yes, they all -- they all grew I believe. At least the two ones that contributed the most, grew. Obviously one grew, you know, quite a bit in dollar terms more than the others. We do moving forward continue to expect to see them all grow and, again, you know, we do have some additional qualifications that are undergoing right now.

All right. And the 10% customer, was it the same one as the one from last quarter?

Yes, it was one of the two...

All right. And Stefan, going through your comment about port speed versus more ports.

Yes.

Assuming that you continue to penetrate and if we assumed away any new wins and just focus in the four large data center customers that you have and one were to assume that you continue to gain footprint, correct me if I'm wrong, but in two of the cases, the relatively new relationships and I assume that the port count that you currently have relatively small but growing in the third case while a little bit older vintage, there again it seems like you're still early in the ramp.

So, I guess, where I am going with this is that it seems to me you should be gaining a significant number of ports over time over and above the shift to higher line rate. But I want to make sure I understand it correctly.

Yes, I think that's -- I think that's a fair assessment. What I was trying to get at with the line rates is that when we talk about, you know, 45% growth year-over-year, for example, that's not necessarily implying that there is, you know, exactly a 45% growth in port count or a higher growth in that if you count for maybe some price erosion or whatever. I just want to -- I don't want you to extrapolate the number of ports directly, that's all I'm saying.

Well, understood, but maybe I misunderstood your earlier comment because I thought your comment was that that was mostly just from an increase in the line rate as opposed to much of any growth in port count, but maybe I misunderstood you.

No. Again, I'm not trying to -- I'm not trying to draw any conclusion about whether our ports are actually growing or not. I mean, in fact, they have been, Okay? So -- but what I'm trying to say is that for 2015, going into 2015, we have two effects.

We have whatever happens to the actual number of ports that we sell up, down or sideways, but keep in mind that we also have this transition from, you know, more and more 40-gig and then in the middle part of next year, we expect to see 100-gig layering in, and those will have, you know, a significantly higher ASP. So that effect in and of itself can account for quite a bit of growth regardless of the number of ports. Does that make sense?

Understood. But just to be clear, Stefan, it sounds like there are two major drivers. There is ongoing growth in port count as you more deeply penetrate the customers you have and perhaps any additional customers and there is the increase in line rate.

Yes, that's correct. That's absolutely correct.

All right. And to -- I think it was Troy's question, but to your statement that you have another customer in qualification, were you referring to a new customer that you haven't yet announced or is it one of the existing customers that's in qualification?

A new customer -- a new customer that we haven't announced.

So that would make your fifth major data center customer.

Correct.

All right. And on the gross margin front, the small gross margin performance, is that primarily or exclusively a function of the mix or is there something else going on?

No. It's -- I mean it's the same kind of things that have been driving our gross margin before, so the mix is the major factor there. Obviously, you know, an overweight of data center revenue tends to be accretive to gross margin relative to, for example, cable TV.

All right. And then one more if I may. On the visibility question which has been asked by others but, you know, I apologize but I've got to come back to it.

Sure.

I understand you gave a forecast, I assume the rolling forecast for data center, from your data center customers, from your fiber-to-the-home customer, and I understand that gives you some visibility, you know, on one quarter.

But given what just happened as an example of how things can change and unexpected in a very sudden manner where you think you have visibility but it's not visibility that you actually saw, and correct me if I am wrong, but on the fiber-to-the-home side, I thought historically you cited that you're building the capacity against firm orders all be it perhaps new orders really firm and so you shifted.

But the obvious question is, you know, how often do you get these rolling forecasts? How confident can you truly be, whether it'd be data center or fiber-to-the-home in terms of the visibility and predictability of the business? To some extent perhaps it's a rhetorical question but if you look back historically, what type of confidence level can you and we have?

Yes, it's a very fair question. I think it's worthwhile to start by touching a little bit on the differences between the fiber-to-the-home and the data center deployments, Okay?

Fiber-to-the-home, as you know, is a very new -- new technology, it's new deployments, and frankly certain aspects of those deployments are not, you know, within the control of our customer.

It's a lot more complicated in that sense. It's not -- I mean it's not -- it can't be completely forecast even by our customer no matter what they would like to do, right, in fiber-to-the-home because there are other extrinsic things that come into play there. And Simon mentioned some possibilities for those types of things in his question earlier, right?

On the other hand, data center is very much in the control of the customer. They have, you know, data centers that they have built out, that they're upgrading to higher data rates, for example.

They have, you know, new data centers that they're building and building online and all of those things are very predictable and very -- you know, very much under the control of the customer.

So I think there is a -- I think we're getting very good visibility both in the fiber-to-the-home and data center from our customers in the sense that they're telling us very -- in a very open and -- way what they plan to do.

Now, you know, to your point, certainly those plans can change from time to time and in the fiber-to-the-home sense that perhaps it's a little bit more likely because of the, you know, extrinsic factors that affect those types of deployment.

But we feel very strongly that the data center business is more predictable and certainly our past history indicates that the forecast that we get from the customers tend to be very accurate, if anything, a little bit of an underestimate.

And then the final thing on the data center market that I want to say is, for the growth rate that we are talking about, we're not only talking about the existing customers that we have but bringing on new customers as well.

And so we're not only relying on, you know, forecast from one customer or two customers but we are layering and diversifying our customer base and that also will contribute to growth and lessen, you know, our reliance on any one particular customer's forecast.

So, Stefan that said, your 45% growth forecast for next year for data center, in that forecast, are you assuming more than five customers i.e. new customers on top of the customer that's coming on or is that just from the five customers that you have?

It's substantially from those five customers although we do expect, you know, that we will continue to undergo qualification activities with additional customers. And so, you know, you know, there may be some upside to that.

Okay, I'll pass it on. I appreciate it. Thank you.

Thanks.

We will now hear from Richard Shannon with Craig-Hallum.

Hello. Richard Shannon?

Hey, Richard?

It's actually -- hey, guys, this is [Jorge] actually. I'm here for Richard. Can you, guys, hear me?

Yes, [Jorge], just fine. How are you?

Good. How are you?

Thanks.

Congrats on the results. So, just a couple of questions from me. One, on the -- on the ONU product, since your FTTH customer has pushed out its plans for a couple of quarters, do you have any update in which we can expect the initial shipments of the ONU units?

Not much update on -- other than what we said in the remarks that we continue to make progress talking to them about, you know, qualification and things like that, so not much else I can say beyond that.

Okay. So I think initially we were expecting the ONUs to begin shipping this quarter. So that's not the case anymore then?

Yes, that's correct.

Okay. And then last question on the data center, you know, based on your comments, that 40-gig would probably reach -- will be close to 50% of the sales. That seems to imply that, you know, 10-gig shipments -- unit shipments would be falling down in the second half.

I just would like to get more color around that. Is this just based on your two customers that are probably shifting to higher speeds a lot faster than the rest of the industry or is there broader indications?

Yes. I think that's pretty accurate. And in fact, you know, one of the customers is moving a little earlier than the other ones but that's exactly what's going on. It's a customer-specific kind of thing.

Okay. All right. Thanks, guys. That's all for me.

Thanks, Jorge.

Yes.

(Operator Instructions)

And we'll now hear from Krishna Shankar with Roth Capital.

Yes. When you look at the cable segment of your businesses, the growth in the second half of this year is going to be driven by mostly the U.S. DOCSIS 3.1 and, you know, deeper fiber deployment or can you talk about the international cable markets and how they may contribute to growth?

Yes. So we do continue to expect that the growth for most of the second half is going to continue to come from sort of international markets. As we talked about, our DOCSIS 3.1 products wouldn't be really available until -- you know, towards the end of the year.

Some of them may contribute a little bit in Q4 but mostly not. So most of the growth is going to come from, you know, the traditional sources that we've talked about in the past and a lot of that is international, in South and Southeast Asia like we mentioned.

Okay. And then similar to data center where you would expect 45% revenue growth next year, can you give us some range for what you think the cable part of your business that could grow in 2015 given the likely scenario for DOCSIS 3.1 and continued international growth?

Yes. So I mean as we talked about in the -- in our remarks earlier, the cable TV business, we expect to see revenue growth in excess of 20% next year year-over-year.

Okay, Okay. And then I guess it's still, you know, very early to talk about the fiber-to-the-home deployment but if your sort of core customer starts to deploy next year and extends its traditional cities, what could be the range of revenue growth for FTTH next year you think?

Yes, I mean we're not -- we're not really able to take too much this far in advance. I think, you know, as you can appreciate, it sensitively depends on, you know, when they get you started and how fast they continue to ramp.

I would expect to see -- you know, I mean, you know, it's not going to be a step function, right? I mean they're not going to go from pre-deployment-type quantities to instantly large quantities. So there will be a period of ramp, and we think that that ramp basically or the start of that ramp, if you will, is probably delayed by one to two quarters.

Okay. Thank you.

All right, Krishna.

And this concludes our Q&A session for today. Dr. Lin, I'll turn the conference back to you for closing or additional remarks.

Okay, and thank you for joining us today. We are very pleased with our Q2 results and we believe that we are on track our key growth initiatives. As one of the few optical providers focused on the access market, we believe that we are very well-positioned for continued growth in all three of our target markets. We look forward to continue to surpass at both the top and bottom line. Again, we thank you for your support.

And again, ladies and gentlemen, that does conclude our conference for today. We thank you all for your participation.