Maxlinear Inc (MXL) 2016 Q4 法說會逐字稿

Greetings and welcome to the MaxLinear Q4 2016 Earnings Conference Call. At this time all participants are in a listen-only mode. A question-and-answer session will follow the formal presentation.

(Operator Instructions). As a reminder this conference is being recorded.

I would now like to turn the conference over to your host, Mr. Gideon Massey, Investor Relations Specialist. Thank you. You may begin.

Thank you, operator. Good afternoon, everyone. And thank you for joining us on today's conference call to discuss MaxLinear's fourth quarter 2016 financial results. And the announcement of its definite agreement to acquire Marvell's G.hn Business. Today's call is being hosted by Dr. Kishore Seendripu, CEO, and Adam Spice, CFO.

During the course of today's conference call we will discuss our financial performance, review our business activities for the fourth quarter and the pending acquisition of Marvel's G.hn Business. After our prepared comments we will take questions. Our comments today will include various forward-looking statements within the meaning of the applicable security laws including, without limitation, statements relating to our current projections, forecasts, and expectations with respect to first quarter 2017 revenue and revenue contribution from key product markets.

Gross profit percentage and operating expenses on a GAAP and non-GAAP basis, the potential impact on our business of recent acquisitions and the acquisitions we may pursue in the future and our current views regarding opportunities and trends in our markets including our current views of the potential for growth in each of our target markets. These projections, expectations, and other forward-looking statements involve substantial risks and uncertainties and our actual results may differ materially from currently forecast results.

Risks potentially affecting these statements and our business generally include substantial competition, in particular, from increasingly large players as our industry consolidates, potential declines in average selling prices and factors that could adversely affect our operating expenses such as litigation, asset impairment or restructuring.

In addition, our target markets, including target markets we may pursue through future acquisitions, including our pending acquisition of Marvell's G.hn Business may not grow in the manner or at the rates we currently expect. We face risks associated with consolidation trends in our operator markets and we face integration risks associated with acquisitions.

For a more detailed discussion of the risks and uncertainties potentially affecting the forward-looking statements we make today, and our business generally, we encourage investors to review the section captioned Risk Factors in our previously filed quarterly report on Form 10-Q for the quarter ended September 30, 2016 and in our Annual Report on Form 10-K for the year ended December 31, 2016 which we expect to file shortly with the SEC.

Any forward-looking statements are made as of today and MaxLinear does not currently intend and has no obligation to update or revise any forward-looking statements. The fourth quarter 2016 earnings release is available on the Investor Relations section of our website at maxlinear.com.

In addition, we report certain historical financial metrics including gross margins, operating expenses, net income or loss and net income or loss per share on both a GAAP and non-GAAP basis. Our non-GAAP presentation excludes certain expense items as discussed in detail in the press release available on our website. The press release available on our website also includes a reconciliation of our GAAP and non-GAAP presentations, which we encourage investors to review.

It is not our intent that the non-GAAP financial measures discussed today replace the presentation of MaxLinear GAAP financial results. We are providing this information to enable investors to perform more meaningful comparisons of operating results and more clearly highlight the results of core ongoing operations in a manner similar to management's analysis of our business. We do not provide a reconciliation for forward-looking non-GAAP gross margin and operating expense guidance, which excludes estimates for stock-based compensation expense, bonus accruals, acquisition-related expenses, and a restructuring charges.

Each of these expense items is explained in greater detail in the press release. The timing and amounts of these future expenses, which we would need to provide a reconciliation of non-GAAP margin and operating expense guidance are inherently unpredictable or outside of our control to predict. Accordingly, we cannot provide a quantitative reconciliation of forward-looking non-GAAP estimates without unreasonable effort.

Material changes to any of these items could have a significant effect on our guidance and future GAAP results. Lastly, this call is being webcast and a replay will be available for two weeks.

And now let me turn the call over to Kishore Seendripu, CEO of MaxLinear.

Thank you, Gideon, and good afternoon everyone. Thank you all for joining us today.

Before delving into the financials we are excited to report Q4 2016 revenues of $87.1 million ,which brings a close to a successful 2016.

In 2016 we not only realized a record revenue of $387.8 million and strong cash flow generation from operations of $117 million but we also made solid progress towards a strategic objective of expanding and diversifying our served addressable market. We achieved this goal by expanding our (inaudible) analog and mixed signal product footprint in our existing broadband operator platforms and by successfully penetrating the large and growing optical and wireless infrastructure markets. Today's announcement of our acquisition of Marvell's G.hn Business is consistent with the stated strategic objective.

The G.hn acquisition further expands the size and diversity of our multi-gigabit access and connectivity footprint to include large Tele Cos deploying X DSL, G.fast and fiber broadband data delivery technologies for the home. 2017 will prove to be an exciting year for us as multi-year product developments (inaudible) sample to are ramped to mass production at key tier one customers in broadband, optical and wireless infrastructure markets.

These product innovations included our next generation multi-gigabit 2.5 SOC our 28-nanometer CMOS, 5 gigahertz to 45 gigahertz single chip microwave (inaudible) wireless (inaudible) RF transceiver and an expanded lineup of 32 gigabaud, 45 gigabaud and 64 gigabaud limiting linear TIAs (inaudible) laser drivers for the long haul metal and data center markets.

In second half 2017 we will sample our first cable infrastructure SOC solution which addresses the emerging full duplex DOCSIS of 3.X (inaudible) and fiber load applications. It should, in the future, enable cable MSOs to deliver up to 10 gigabit data services to the home or the existing hybrid fiber coaxial network.

Our 100 gigabit per (inaudible) high-speed fiber interconnect solution, supporting 400 gigabits per second interconnect speeds inside the data center, is also slated for sampling in the second half. In wireless access infrastructure we are making steady progress towards the development of a fully integrated massive MIMO Rf transceiver solution that addresses the merging 4.5G and 5G wireless network deployments. The foundation of our organic initiatives acrossed infrastructure, broadband access and connectivity marketed is, of course, [Si-MOS] Rf mixed signal communications technology platform.

The success of our ongoing expansion into optical and wireless infrastructure markets is proof of the scalability of our core CMOS technology platform across a range of large and growing communications markets. Moving on to our Q4 2016 results. Our revenue of $87.1 million in Q4 2016 was in line with prior guidance, down 10% sequentially and down 12% year-over-year.

I'm extremely pleased with our topline operating margin resilience even as we navigate through expected (inaudible) legacy and [tropic] products, namely satellite, analog channel stacking and video SOCs. We are now increasingly optimistic that Q1 2017 marks the beginning of the resumption of sequential revenue growth. These legacy products currently represent less than 10% of all our revenue mix. More importantly, the decline in legacy revenues are being rapidly replaced with products with double-digit sequential revenue growth rates addressing the higher quality cable data gateway, high speed optical interconnect, and wireless infrastructure markets.

The improving quality in diversification of our revenue streams is highlighted by our combined optical wireless infrastructure revenues. (Inaudible) they are growing 15% of overall Q4 revenue. Taken together Q4 results closeout a very successful 2016 with our full year revenue up 29% at a record $387.8 million, strong cash generation from operations of about $117 million, gross margin improvement and a proven track record of executing on our strategy goals by a disciplined combination of organic and inorganic growth initiatives.

Now moving to the specifics of our business highlights for the fourth quarter of 2016. Operator revenues declined 9% sequentially and accounted for 76% of total revenue in the quarter. Within operator, cable data shipment rebounded, in particular 24 and 32 channel DOSIS of channel 3.0 solutions. We have a strong demand for cable video as well.

We also saw the beginning of our cable DOCSIS of 3.1 RAM, which grew by more than two times sequentially even though it was a small part of our cable data mix. The underlying market dynamics supporting the DOCSIS 3.1 product cycle are extremely exciting. DOCSIS 3.1 will be a very meaningful growth driver in second half 2017 owing to a leadership position in cable data front ends on our expanded footprint consisting of companion, PGAs and MoCA connectivity solutions.

Satellite digital outdoor units also grew strongly up double digits sequentially, resulting in a near doubling of 2016 annual revenue relative to 2015. The growth in digital outdoor units was offset by the anticipated decline in legacy NOL channel stacking chips. Our satellite digital outdoor (inaudible) products are ramping to full production volume in North America and Europe. We also have a great pipeline of design wins at major operators in the rest the world including Telefonica, (inaudible) and Intellasky that are yet to ramp.

Also in Q4 2016 satellite 4 (inaudible) Rf receiver shipments declined as several tier one (inaudible) operators digest their inventories following a particularly strong Q3. We are really exciting to announce the initial shipment of our first organically developed MoCA solution to a large new tier one wireless Tele Co operator in North America. Our next-generation MoCA 2.5 solution, which is capable of reaching 3 gigabits per second data rates, is also on track to ramp in the second half of 2017.

MoCA originated as an enabler for whole home PVR and video distribution of coaxial cable in cable and satellite subscriber homes. Today owing to it's high-quality of service and robust multi-gigabit band width delivery capability MoCA has emerged as a critical high band with coaxial backbone connectivity solution, essential to enabling robust Wifi coverage throughout the connected home. MoCA technology has also morphed into c.LINK, which is a robust and low cost access infrastructure solution for MDUs or multi-dwelling units.

c.LINK is currently starting to deploy at several cable MSOs in China. Today's announcement of the acquisition of Marvell's G.hn Business, for $20 million in cash, further strengthens our wire connectivity product portfolio. G.hn is truly complementary to MoCA.

MoCA is a broadband connectivity solution for data distribution over coaxial cable while G.hn is ideally suited to deliver broadband data robustly over electrical twisted pair and power line physical media insides the home. As result G.hn allows us to expand our connectivity footprint to include Tele Co MSOs deploying XDSL, G.fast and fiber to homes with no pre installed coaxial cable. The need for alternative connectivity mediums to coaxial cable is especially prevalent outside of North America and parts of Europe.

With additional G.hn we are able to offer OEMs and broadband operator partners multiple wire connectivity solutions each ideally suited for a particular geography or ecosystem. Through the G.hn acquisition we have secured several strategic tier one teleco operator engagements and design wins along with an incredibly capable technology in Valencia, Spain. There have been public announcements of adoption of G.hn by [KD] Telecom in South Korea, China Telecom, Reliance in India and other major Tele Cos in Europe and North America.

As our G.hn design is ramped to production we will be in a strong position to cross sell other MaxLinear products and expand our target and visible market by another 200 million broadband subscriber homes. Moving to our infrastructure and other products, revenue was up 21% of total revenues in the quarter. We witnessed resumption of growth for our high speed fiber integrated products and realized stronger than expected revenue contributions from our broad common mic less MI wireless backhaul access infrastructure asset acquisitions respectively.

For the year our high-speed fiber integrate revenues were at the high-end of the $10 million to $20 million range that we guided to at the beginning of 2016. As we look forward we are excited with by the revenue prospects for our expanding high-speed fiber interconnect product portfolio. In 2017 MaxLinear's portfolio will transform from a single laser driver shipping to two major (inaudible) and OEMs in China to a broad suite of laser driver and TIA solutions shipping into a data center, metro and long [while] fiber applications.

In Q4 we began shipments of two new laser drivers and our first TIA product for 4 by 20 by 100 gigabit NRZ data center and for LR 4 Tele Co applications. We have also diversified our high speed fiber interconnect customer base with shipments to 12 different customers during the quarter. Moving to our wireless infrastructure markets, our acquired microwave backhaul business grew strongly sequentially with contributions from a range of tier one and tier two Tele Co wireless OEMs.

However, we do expect a modest pull back for this business in Q1 primarily due to the strength witnessed in Q4. I would like to note that we are forecasting very strong growth for this business in 2017. We are excited about nearing the initial ramp of our organically developed 5 to 45 gigahertz fully integrated SI-MOS Rf transceiver at a tier one OEM.

The backhaul Rf transceiver front end seamlessly complements the backhaul microwave modem SOC we acquired from Broadcom. This is both consistent with and also validates our strategy of becoming a full platform technology provider in the microwave backhaul market. Our wireless access business also experienced sequential increase in revenue relative to Q3 2016.

As the wireless access market transitions to 4.5G and 5G deployments in order to deliver data speeds over the air the entire wireless network will transform to a massively distributed SDN and NFE network. It will be a cloud-based remote radio access network or CDAN with massive MIMO and beam steering (inaudible) systems, distributed (inaudible) systems, macro cell base stations and small cells interconnected by wireless backhaul and optical fiber front haul solutions to the network core. This will result in a exponential increase in the Rf transever unit TAM.

We will be entering a transformative market from a position of strength in both optical and wireless RF transceiver technologies. As we look forward we are excited by our progress towards accelerating the scale leverage of our product portfolio that grows both wireless access and back hole infrastructure platforms. We are pleased with the customer channel and technology platform leverage and synergies we are deriving from our wireless and fiber infrastructure acquisitions in the wireless infrastructure space.

Lastly, legacy video associated revenues derived from the Entropic acquisition declined $2.1 million in the quarter or 2% of total revenues versus 7% in the prior quarter. As we expected this weakness owed to last time buys in the prior quarter where cable, HD DTA deployment and satellite IP client devices. We expect the legacy video associate business to be at an eligible level of 1% of our total revenues in 2017.

Before I turn the call over to Adam Spice, our Chief Financial Officer, I would like to reiterate that we are extremely pleased to have delivered a strong fourth quarter and fiscal year 2016. We are pleased with the increasing diversification of our revenue streams across broadband access and wire and wireless infrastructure markets. Entering 2017 we also feel optimistic about the prospect of resumption of sequential growth starting in Q1.

As we navigate through the final phase of our declining Entropic legacy businesses we are excited about our products expansion, strong customer and end market diversification potential and the sustainability of our core brand infrastructure technology platform. We look forward to sharing more information regarding the progress of our development initiatives in the coming months.

Now let me turn the call over to Mr. Adam Spice, our Chief Financial Officer, for a review of the financials and our forward guidance.

Thank you, Kishore.

I will first review our Q4 2016 and full year 2016 results and then briefly discuss our outlook for Q1 2017. As Kishore noted, our Q4 revenue was $87.1 million, consistent with our guidance.

GAAP and non-GAAP gross margins for the fourth quarter were approximately 57.8% and 63.9% of revenue respectively. At the high end of our prior guidance of 57% to 58% for GAAP and 63% to 64% for non-GAAP gross margin. The delta to the mid point of our guide was primarily driven by favorable mix within our operator and infrastructure revenues.

This compares to GAAP and non-GAAP gross margins of 57.6% and 63.1% respectively in the third quarter of 2016 and net GAAP and non-GAAP gross margins of 56.4% and 58.1% respectively in the year-ago quarter. The delta between GAAP and non-GAAP gross margins in the fourth quarter was primarily acquisition-related reflecting the amortization of $5.2 million of purchased intangible assets and inventory step-up and a lesser $100,000 of stock-based compensation and stock based bonus accruals. Q4 GAAP operating expenses were approximately $42.1 million, $400,000 below guidance with a delta primarily related to lower prototyping. facilities and restructuring expenses, partially offset by higher payroll and patented related spending.

GAAP operating expenses included accruals related to stock-based compensation and stock based bonus and incentive plans of $5 million and $2 million respectively and $3 million for the amortization of purchased intangible assets. GAAP operating expenses also included $1.3 million of restructuring related to the closure of our Atlanta and Austin facilities, commensurate with the rolling off of the Entropic Video SOC revenues and related support overhead and $600,000 in acquisition and integration-related expenses.

Payroll under our -- sorry payouts under our second half 2016 performance bonus plan will be settled primarily in shares of MaxLinear stock, which are expected to be issued in Q1 of 2017. Net of these item non-GAAP OpEx was $30.1 million or $400,000 below our prior guidance of $30.5 million and $1.5 million lower than in Q3 2016 and up approximately $2.6 million from the year-ago quarter. Fourth quarter GAAP OpEx attributable to R&D was down approximately $1.9 million quarter-on-quarter and up approximately $1.4 million year-on-year to $24 million.

Which included stock-based compensation of $3.3 million, $1.3 million related to the second half 2016 stock based bonus plan and accruals and the final incentive award compensation related to our five speed acquisition and $100,000 of amortized purchased intangibles. Excluding these items fourth quarter non-GAAP R&D was down approximately $1.2 million quarter-on-quarter and up approximately $2.2 million year-on-year to $19.3 million. Within the sequential R&D spending decrease there was $600,000 in lower payroll and headcount-related expenses, $300,000 in lower occupancy costs and $100,000 in lower tape out expenses.

These sequential decreases were partially offset by a $100,000 increase in design tool spending. Fourth quarter GAAP OpEx attributable to SG&A was down approximately $900,000 quarter-on-quarter and down $1.2 million from the year-ago quarter to $16.7 million. GAAP SG&A expense included $2.9 million for the amortization of acquired intangible assets, $1.7 million in stock-based compensation, $700,000 in stock based bonus plan accruals and the final incentive award compensation related to our five speed acquisition and $600,000 in acquisition and integration costs primarily related to our announced acquisition of Marvell's G.hn business.

Excluding these items, fourth quarter non-GAAP SG&A was down $300,000 on a quarter-on-quarter basis and up $500,000 from the year-ago quarter to $10.8 million. With sequential decrease driven primarily by occupancy and other overhead related savings, only partially offset by higher sales commission expenses, patent expenses and professional fees. At the end of the fourth quarter of 2016 our headcount was 553 as compared to 571 at the end of the third quarter of 2016 and 500 at the end of the fourth quarter of 2015. We continue to evaluate our staffing levels globally, particularly following our recent acquisition activity, to strike a balance between driving near-term bottom line operating leverage and staffing key long-term growth initiatives.

GAAP income from was $8.3 million in Q4 compared to income from operations of $10.7 million in the prior quarter and a loss from operations of $8.7 million in Q4 of last year. GAAP income from operations was $63.1 million for the full year 2016 versus the GAAP loss from operations of $43.6 million for 2015. GAAP earnings per share in the fourth quarter were $0.12 on fully diluted shares outstanding of $68.4 million. This compares to GAAP EPS of $0.14 in the prior quarter and a loss of $0.14 per share in Q4 of last year.

For the full year 2016 GAAP earnings-per-share were $0.91 compared to full year 2015 GAAP losses per share of $0.79. Non-GAAP earnings-per-share in Q4 were $0.38 on fully diluted shares of $68.4 million compared to $0.43 per share in Q3 of 2016 and $0.46 per share in Q4 of last year. For the full year 2016 non-GAAP income per share -- earnings-per-share was $1.79 compared to full year 2015 non-GAAP EPS of $1.28 on a fully diluted basis.

As you may recall, 2015 losses were heavily influenced by the significant purchase accounting impacts of the Entropic acquisition that closed in May of 2015. Moving to the balance sheet and cash flow statement. Our cash, cash equivalents and investments balance increased $26.6 million from the end of Q3 2016 to approximately $136.8 million. And increased $6.3 million as compared to $130.5 million in Q4 of last year despite having closed the micro (inaudible) wireless infrastructure asset deals for a total of $101 million.

Our cash flow for operations in the fourth quarter of 2016 was approximately $27.6 million versus $18.4 million generated in the third quarter of 2016 and $24.6 million in the year-ago quarter. Our days sales outstanding for the fourth quarter was approximately 53 days or six days more than in the prior quarter and 14 days more than in the year-ago quarter. The increase in days sales outstanding is primarily a function of changes in shipment linearity as well as a general lengthening payment terms granted to some of our largest direct customers.

As a result we only recognized revenue on a sell-through basis and, as such, we're not subject to revenue fluctuations caused by changes in distributor inventory levels. Our inventory turns were five in the fourth quarter compared to 5.7 turns in the third quarter and 5.1 turns in the year-ago quarter. As disclosed in our press release, and mentioned earlier in the call, we have executed a definitive agreement to purchase Marvell's G.hn business for $21 million in cash and expect the acquisition to close early in the second quarter of 2017.

As such our Q1 guidance does not include any operating contributions from the pending acquisition during the first quarter. Nor does our guidance include any potential acquisition-related charges including those for amortization of purchased intangible assets. We expect the G.hn business to be dilutive to our GAAP earnings in 2017 primarily due to acquisition-related purchased accounting impacts and to be roughly neutral to our non-GAAP earnings in 2017.

We will provide more color on the expected financial impact of the deal post-closing. That leads me to our guidance. We expect revenue in the first quarter of 2017 to be in the range of $86 million to $90 million.

Built into this range we expect seasonally strong operator revenues to account for roughly 80% of overall revenue and infrastructure and other the remaining 20%. More specifically, within operator we expect growth to be driven primarily by strength across our cable data, cable video and MoCA cable products combined with terrestrial set-top box, tuner to modulator SoC shipments, supporting analog to digital broadcast conversions in South America and a return to growth in our satellite 4K gateway business.

Within infrastructure and other, in which we are now including the increasingly negligible Entropic legacy video SoC revenue, we expect growth from our wireless access products to be offset by step back in wireless backhaul after a strong fourth quarter as well as a step back in high speed optical revenues and also by seasonal weakness in TV and consumer terrestrial set-top box.

Once we get beyond Q1 2017 we expect revenue in unit shipment growth in our 100 gig laser drivers in China as fiber upgrades expand to tier two cities and dense metro regions. Additionally we are ramping new solutions across both drivers and TIA addressing 32 gigabaud, 45 gigabaud and 64 gigabaud coherent data center, metro, and long haul applications across an increasingly diverse set of customers and geographies as referenced by Kishore earlier. That leaves me with the remainder of our Q1 2017 guidance.

We expect GAAP gross profit margin to be approximately 60% of revenue and non-GAAP gross profit margins to be approximately 62% of revenue. The sequential lift in GAAP gross margin is due primarily to the roll off of the amortization of acquired inventory step-ups in Q4. The forecast in sequential decline in non-GAAP gross margins is primarily due to the mix influence of growth in terrestrial receiver shipments supporting the South American analog digital broadcast conversions, which is expected to be a temporary one or two quarter mix effect.

As a reminder our gross profit margin percentage forecast could vary plus or minus 2% depending on product mix and other factors. We continue to fund strategic development programs targeted at delivering attractive top-line growth as we look forward into 2017 and beyond with a particular focus on infrastructure initiatives and our goals of increasing the operating leverage of the business. As such we expect Q1 2017 GAAP operating expenses to decrease approximately $1.1 million quarter-on-quarter to approximately $41 million.

With the largest decreases coming from the removal of restructuring expenses, lower tape our expenses and lower patent spending, partially offset by seasonal step ups in payroll-related expenses, trade shows and CAD tools. We expect that Q1 2017 non-GAAP operating expenses will increase $900,000 sequentially to approximately $31 million driven by the earlier referenced items. In closing we're pleased to report the close of a very eventful and successful 2016.

Our progress in acquiring and integrating new growth initiatives and diversifying our product portfolio and end market exposure have enabled us to deliver significant revenue growth and expansion in both gross and operating margins and more than a doubling of our annual cash flow from operations. Also, as reflected in our guidance, we feel increasing confident about the resumption of our sequential revenue growth starting in Q1 2017. We continue to expand our served addressable market organically and through strategic acquisitions furthered by the Marvell G.hn acquisition announced earlier today. We believe these inorganic diversification initiatives position us to uniquely exploit a benefit from the growing demand for bandwidth across consumer, operator and wired and wireless infrastructure platforms.

And with that I would like to open the call to questions. Operator?

(Operator Instructions). Our first question comes from the line of Tore Svanberg with Stifel. Please proceed with your question.

Yes. Thank you. First question in the past you have given us targets for the year, especially with some of your newer businesses like last year you gave us the fiber optics target.

If we look at 2017 do you have targets at this point for the fiber optics business and also wireless infrastructure?

So hi, Tore. This is Kishore. You know, the last year was -- we gave you targets which we exceeded to be closer to the high-end of the range for the fiber optic business.

This year I think we were alluding to the previous calls we expect the revenue to be somewhere between $20 million to $30 million hopefully closer to the $30 million range with optical fiber. Within that context the revenue increases would be primarily driven by new product launches in linear TIAs and linear drivers for the 32 gigabaud, 45 gigabaud market sand those -- and then for LR 4 applications which we've got design wins, we have started initial shipment across a broad range of customers. So as the TIA product portfolio expands and captures momentum in the later half of second -- of 2017 we expect the revenues to grow and so that's where we view the range to be for the fiber optic business.

On the wireless, you know, the wireless business as a whole I think we have given -- we have -- I think we earlier talked to you about -- correct me if I'm wrong, Adam -- you have given guidance in the $35 million to $40 million range of revenues for 2017. We reinforced the guidance on those -- on those as well.

Very good. And with the acquisition of the Marvell G.hn Business, first of all does it come with any revenues this year and can you also talk a little bit about what this does as far as expanding your -- your [SAM]?

Sure. I'll take the financial questions and Kishore can talk about the SAM expansion. So yes it does come with revenues.

The business is coming with -- I would say it comes with nascent revenues but certainly a strong pipeline of design wins and platforms that are already ramping so this isn't where we have to wait for the revenue to start to develop. It's already in place. It's growing.

It's de minimis in say for example in the Q1, Q2 time frame that we're talking about, the close that hand in Q2, you know, it's in the low single digit millions of dollars of revenue and that's -- when I he mentioned in the prepared remarks that we expect it to be essentially neutral to non-GAAP earnings in 2017, you know, of course it assumes probably maybe could be a little bit dilutive in the first quarter and then kind of a little accretive in the Q4 period, it could, but it's kind of in that neutral boundary for 2017. And the margin profile of the business is very consistent with the rest of the MaxLinear business so we're pretty excited about adding this to the mix. And, Kishore, do you want to talk the SAM?

So you know in connection with the G.hn platform Marvell's G.hn platform is the most advanced platform for the power line multi gigabit home connectivity using the power lines and twisted -- twisted pair. And so this nicely complements our very, very strong position MoCA and together we will be the undisputed technology leader for wireline connectivity. You know, MoCA as a coaxial connectivity platform is prevalent primarily in North America and Europe and a few other parts of the world for distribution of data inside the home.

However there are large telecom operators, especially those that serve data (inaudible) DSL or fiber, you know, and G.fast that would like to use the twisted pair on a power line -- a wireline connectivity and today we don't address those customers platforms. So if you really add up the major telecoms in China loan on the telecom side, China Telecom alone has 200 million subscriber homes, but if you add up all these countries and homes where there's no coax pre installed insides the home the subscriber base that you can address on the connectivity for broadband is quite substantial. So at this point I would say the -- the wireline connectivity TAM is at least doubled for what it was from let's say 100 million subscriber homes to now anywhere between 200 million to 400 million subscriber homes.

So that's how I would characterize the expansion in the Sam. So is that -- is that okay, Tore?

Yes. Yes. No, that's great. Last question is on MoCA. So you mentioned you are already shipping the MoCA 2.5 solution.

I'm just wondering as we look at the whole year does that mean MoCA could actually grow this year since you already starting shipment now or will it still be some puts and takes with the older version of the MoCA product?

I think MoCA is -- is a very promising one. We expect that compared to 2016 we have -- you know, the experience of growth and especially this particular design with a large wireless operator for the latest offering in MoCA is very promising and we expect to see some growth. Having said that, on the existing cable platforms the older Entropic MoCA solution, 2.0 solution will be the substantial revenue driver for us and that will remain so until we get through 2017 on a run-rate basis.

However, there is great news for us in terms of our product that we just launched getting good traction design win and very, very large US wireless operator and we are very excited about that.

Very good. Thank you so much.

Thank you. Our next question comes from the line of Gary Mobley with the Benchmark Company. Please proceed with your question.

Hi, guys. Thanks for taking my question. I wanted to ask about 2017 growth.

I'm assuming it's not going to be a growth year for you guys, but maybe if we're so lucky you can give us some general parameters, you know, as far as growth considerations and if you're not willing to give that, make you could speak to when we might see the first positive year-over-year comp. Might it be in the third quarter of 2017?

Yes. You know, it's pretty early in the year, Gary. As far as kind of put a definitive line in the sand to say when the sequential or when the year-on-year turns positive.

I think to your point right now we're not pointing to a positive full year-on-year increase. I think that would be -- if we got there, that would be a huge accomplishment but that's not currently in our planning scenario today. That said there are -- there are opportunities in the second half of the year certainly to reclaim a positive year-over-year compares.

Okay. And 2017 OpEx, can you give us some general some of your general budgeting considerations and a way to think about OpEx.

Yes. Our view on OpEx really hasn't changed. Of course, when we tuck-in the Marvell G.hn Business we'll update you guys as far as what that means for spending but right now I think we -- we have said earlier that we expect to be able to roughly keep our OpEx flat year-on-year and, you know, there will be some movements in there other things in the mix and is there's timing of tape its that can make it a little bit lumpy quarter-on-quarter.

For example this quarter it was a step-up, still within the range, I think last year when people were asking me a similar question I would say that in 2017 we would probably expect to see OpEx anywhere between, call it a number, would be $29 million plus to maybe as high as $32 million in a given quarter. So I think you have got about a $2 million or $3 million range within any quarter for OpEx, but I think we're going to be able to contain it to within that range and, hopefully, get pretty close to flat 2016 to 2017 for the full year.

Okay. Given the geographical differences in use cases between MoCA and G.hn I'm assuming there's perhaps limited customer overlap, but in those cases where you do have overlap or, I guess asked differently, have you had any discussions with the different telecos you queued up to deploy G.hn for Marvell and how you guys might benefit from this or augment any future roadmap plans?

Gary, it's a very, very good question and that's why I went fairly deep in to explain the differences between MoCA and G.hn general. The general operator situation is quite -- you know, quite separate. It's a very complementary acquisition for this reason.

The telecom operators that subscribe to the (inaudible) standards do not use MoCA for the distribution site. They like to use anything that's twisted pair and that's why into the home they use G.fast or X DSL or fiber ethernet into the home. So in those circumstances outside of North America there's no corruption between the G.hn MoCA situation and it's very, very large market and for those telecos G.hn will primarily, most likely, proprietary valid as distribution standard. Where it is slightly different is in North America coax -- where coax is totally prevalent in the homes and there MoCA wins outright because if you compare MoCA to G.hn, MoCA is still a far superior connectivity medium because you're sending data over a shielded copper line. So MoCA is far superior to G.hn however the G.hn wave two technology from -- from Marvell accomplishes on twisted pay -- or gigabit data rates that otherwise would not pass positive in the past. So what it offers for us now is to go and sell the product to all the telecos that are deploying XDSL, DSL, G.fast or fiber outside of North America where the ITU standards lineup and offer them a connectivity solution called G.hn and with that entry into that platform we can now start cross-selling other technology portfolio IP element that we have in the MaxLinear capabilities.

So that's the game plan here to expand the TAM, enter the SAM and expand the SAM.

Okay. All right. Thank you, guys.

Thank you. Our next question comes from the line of Ross Seymore with Deutsche Bank. Please proceed with your question.

Hi, guys. Thanks for all the end market color that you give. I guess looking back on the other side of legacy you talked about the $2.1 million in the fourth quarter from the legacy video SoC.

Can you size for us where the analog channel stacking was and give us an idea of how you expect that to fall-off as we go forward?

Yes. Ross. This is Adam.

Look, on the analog channel stacking we're telling people that they should expect somewhere in the range of call it $7 million in Q4 and that's exactly where we landed. So I think it was very consistent with our expectations and the expectation for Q1 we were telling folks to expect somewhere between say $4 million and $5 million, you know, in the first quarter. And I think we'll actually do better than that.

So I think the fall-off sequentially Q4 to Q1 is probably a little less or more subdued than we were thinking a month or so ago but it's still very much -- for the year nothing has really changed. It just seems like there's a little bit more gradual decline Q4 to Q1 and maybe even Q1 to Q2 but when you look at the year in its entirety nothing has really changed. Just a little -- .

And then as is we look into the guidance for the March quarter and you talked a little bit about what the gross margin puts and takes were can you go into a little bit more color about that terrestrial business? I assume that must be in the operator side of the equation to get you to your 80% of sales. I'm surprised it has that much of on impact on business. So talk a little bit about the leverage on that gross margin line if you could please.

It's actually a fairly stark step-up because there really -- the revenue really wasn't there in Q4 at all. It's really a kind of a Q1 Q2 and it's very operator specific deployment cadence that's really, again, first half of 2017 focused. So it's kind of a bursty, happen for a couple quarters and then go away, it's kind of a one time revenue event. You know, it's leveraging a very old product that we have in our portfolio. It's our ISDBT 200 DMOD SoC that we have been shipping since I think about 2011 and so I would say it's -- for this particular opportunity it was a fairly competitive process because again its lagging technology and other people could throw their hat in the ring.

We threw our hat in the ring and we were pretty aggressive to try to drive the remaining contribution margin dollars from those old products on to our side of the ledger and we are successful in doing that, but it did come at a cost as far as overall gross margin optics or mix for the Company.

And it's really not that much of a mystery, right? Because if you look at the HDTA deployments that Brazil was promising to deploy before the Olympics and then the World Cup and they just department delaying and then the political issues and the currency issues and final they pulled the trigger and they placed for the year about 4 million to 5 million units of HDTA (inaudible) government subsidized through the operators and the bulk of that is going to be shipped in the first half of 2017. And so if you look a part that's got no cause optimization for the last six, seven years, you know, we're not going to do anything about it and so it was -- it was a perishable revenue and we decided to take it basically.

The overall margin -- Ross, to give you an idea I mean it's actually from a -- there's not a lot of support or other overhead kind of tied to this particular project but it is a -- it's quite a bit lower than like -- many times lower than our corporate average. It is -- that's why it has such a diluted impact. It's pretty posterior gross margin but it's not bad when you look at it purely a contribution margin dollars, right?

And-- but the good news is that -- so the margin will bounce back higher once these shipments have happened through Q1 and maybe early part of Q2. So we are not overly concerned. It was tough for us to not walk away from the contribution margin.

I guess the final question I would have is the -- it sounds like the majority of it is that business going up, but it also has the wireless infrastructure and the high-speed interconnect businesses sounds like they're going down sequentially. So talk a little bit about the reason for those falling off and I assume that that's a negative mix implication for the gross margin as well that might reverse when those two businesses start growing again.

Absolutely. You know, in the wireless business, we -- this is our first year in it really speaking the beginning of the last year at this time of the business we didn't have a business in wireless so if you really look at that from that perspective we are experiencing what may be a seasonal or CapEx spending step down accept regulations for our buckets out place by telecom operators but we are very strong Q4 for both our wiring access and wireless backlog whicker very, very pleased because at the time of these two acquisitions we got a little anxious about maybe there was some -- we had estimated things wrongly so we're very happy about that.

On the second piece on the optical side once again it's our first whole Q1 with no distortions and we are being cautious, China went to the Chinese New Year and we had a little bit of a backlog coverage issues we go through a distributor and so we take a cautious approach but generally we think we have consolidated our market share in the laser driver market with two main Chinese OEMs while our TIAs are just beginning to ramp.

So still the bulk of the revenues in the laser drivers is with the two Chinese OEMs and so -- and we are waiting to see if the promised increase in the market size of 30% to 35% that we are told by our customers materializes, so we're just being careful about it as well and obviously there's some forward pricing of the products as well in preparation for taking the big share in the market that we expect it will increase.

Great. Thanks, guys.

Yes.

Thank you. Our next question comes were the line of Anil Doradla with William Blair Investment Management. Please proceed with your question.

Hi, guys. Just a couple of clarifications. So on building up on Ross' question, you know, were there some strategic reasons beyond just the contribution margin or getting that topline for pursuing these (inaudible) sales? And going forward would this be -- could we see such events these one-off events playing out in the course of 2017 and I have a follow-up.

Okay, Anil, I think you asked a very, very good question. If you recall what Adam said, it is an operator business so it is a major operators in Brazil, we have our satellite front ends -- satellite digital outdoor unit shipments. These designs will run, believe it or not, three to four years ago.

If you look at our earnings transcripts from that point in time we keep talking about this and finally it happened when we had completely given up. So there was no way to walk off from that because these designs were done and our baseline partners on the side for the MPEG decoders, they are committed so you don't walk away from the situation. So you call it strategic, yes.

The actions were strategic to remain in -- in the socket, but the business itself is not strategic for us and we have continuously emphasized that, but to the extent there's a broadband operator and it is on our broadband platforms we will support them. Secondly, none of the other businesses outside of the terrestrial DTA stuff are these well below corporate margin products, so hopefully it is a one time blip and we have noted that last year we have steadily every quarter improved our gross margin, in fact Q4 was around 64%, and we hope to get back there as rapidly as possible. So I wouldn't look at it as a systematic pattern at all, it's a one time blip and normal DTA operator businesses out there that we are now serving.

So it's fair to say, you know, Kishore, what you're saying if this opportunity would have been presented to you today, you would not have -- you would have walked away from it.

Absolutely.

Wonderful. And on the channel break down for cable can you help us understand the break down between 16, 32, 24 today an as you exit the year 2017 how should we be looking at that.

Yes I guess that's quite a bit of detail. I think that if you look at certainly the trend has to consolidate and 24 channel so the bulk of our volumes now are our 24 channel in the cable data DOCSIS market. You know, there is an increasing amount of 32 channel there as well.

So you're really seeing the -- the -- we never had much of a 16 channel presence. It was really 8 channel, then the market moved on us pretty quickly where we got share in 24 and now in 32 and of course now we're facing the DOCSIS 3.1 ramp, which we talked about before is really more of a second half phenomena. So I would say that it's -- it's to our favor that we have gotten higher ASP volume transitions to a higher channel count and I don't see anything kind of upsetting it, the next wave is just kind of going then from 24 and 32 to operators that are being the earlier ones to deploy DOCSIS 3.1, which right now we have been talking about Comcast being the tip of the spear pore DOCSIS 3.1 deployments.

And just a clarification, Adam. Did you quantify the impacts of this (inaudible) revenue was this 100 bps or 150 bps on the gross margin? I missed that, I think.

We -- we haven't -- we haven't broken that out, but I would say -- based on the guidance for our Q1 guidance had it not been for this particular thing it probably cost us I would say probably around 100 -- I would say it's probably 100 basis points is probably a reasonable estimate for that.

Okay. Wonderful. And looking forward to 2017. Thanks.

Great. Thank you.

Thank you. Our next question comes from the line of Quinn Bolton with Needham & Company. Please proceed.

Hi Adam and Kishore. Just want to maybe just answer it. Did you say that the impact from this terrestrial satellite business in Brazil was about 100 basis points of the sequential decline in gross margin?

That's about right.

That's correct.

Okay. So I mean these are -- these are parts (inaudible) they actually higher priced parts in that market. You said something about four million or five million units.

Should we be thinking at a price about a buck -- maybe two half million across those few quarters is that about the right.

You could have been our sales or marketing in charge, you know.

Okay. Okay.

You're pretty close. You're in the ballpark.

Okay. Great. Second question on the G.hn, you said I think Kishore, over twisted copper pair, the Marvell solution can get a gigabit per second. Surprised me because I didn't realize the G.hn was up that fast. I guess what's your thoughts on the competitive outlook versus Wi-Fi in those market today that are served by power line or twisted pair?

You ask a very -- you ask a very complicated question. When I talk gigabit plus -- G.hn, every standard and every connectivity standard talks about gigabit plus as the Holy Grail but actually what is really the most important criteria is whether it's Wi-Fi or G.hn or MoCA is how reliably the bandwidth you get in a particular distance per user reliability. These are all peak data rate statements and that's really a few hundred megabits, right? And that too means that it's really -- if they are doing that by adding more bandwidth on G.hn and likewise it's true for Wi-Fi as well.

Wi-Fi is even worse with the coverage and the robustness as you move the whole. So what it's playing out into is that -- and you are seeing an announcement for example of retail products where they are improving Wi-Fi coverage by using G.hn or power line connectivity as the backbone, but on the operator side they are (inaudible) their Wi-Fi coverage by using these wired power line connectivity as the backbone and have local adaptors between the power line connectivity and the Wi-Fi so that you can use Wi-Fi locally at the distribution point. In the United States MoCA -- MoCA is the preferred one by all the cable operators and because it's the most robust medium of connectivity, maybe the coaxial cable just shielded and so nothing comes close to it.

So I think that G.hn is really going to be more a non-US, non-North America parts of Europe phenomena and, you know, and it's going to be complementary in terms of, just like MoCA will be in the United States, for Wi-Fi coverage.

Okay. Great. Thanks for that color.

Then just lastly in the prepared comments I think you said now you now expanded the five (inaudible) customer base to 12 different customers. Your two lead customers in China were kind of just trying to long haul. Of those 12 that you started to ship to this year could you give us some sense of how many of those would be coherent applications, how many of those would be data center application and, perhaps importantly, how many of those are outside of China just to give us a sense of the geographic diversification of that customer base.

I would say out of the 12 about 8 -- 8 will be outside of China and the one new one in China will be a small player and then, you know, two will be in Japan and the rest of them will be both Japan and North America combined. So -- so they're in both locations. So put it differently all the diversification is coming outside of China.

However, most of these diversification is coming at least half or more in the TIA space and the shipments into China for TIAs happen from outside of China. China does not do sourcing from their suppliers inside China. The really none Chinese players who supply (inaudible) into the Chinese market because of the sophisticated technology in terms of photo detector and alignments with the TIA to ensure that maximum sensitivity.

So put it simply the end market may be China but all the customers are primarily non-Chinese.

Okay. Any sense data center versus coherent.

Okay. And the data center piece, obviously, on the -- I would say -- I don't have a breakdown by that, but I would say that the same players who are there in the (inaudible) are also in the data center and those are two or three of those big module guys and there is one guy who is PO data center.

Great. Thank you.

Yes.

Thank you. Our next question comes from the line of Brian Alger with Roth Capital Partners. Please proceed with your question.

Hi, guys. Thanks for squeezing me in. A lot of questions have been asked.

Just a couple follow-ups. I think you mentioned that the satellite digital outdoor was actually up double digit. I'm just wondering about that was an expected growth or if that was better than expected.

So I would say it was an expected growth though it's lumpy and as we have always with these broadband operators cable and satellite their behavior is quite sporadic in the Q4, Q1 window. Some more in Q1 some will not order in Q1 and will order in Q4. So it really going through the what I call the vacillations in order patterns from these satellite and cable operators.

So really I wouldn't read much into that as more than expected. If you integrated over the two quarter window it's probably as expected.

Okay. And as we make the transition away from the analog stacking and move towards digital, do we expect that volatility in the order pattern to dissipate?

I don't think you should expect when it involves operators the -- the seasonality in the year to disappear. You can call it volatility or seasonality, I don't know, but generally these operators (inaudible) Q2 is strong. If you just go with the cable operators and Q3 is reasonable, Q4 they may go all the way down and then Q1 they come backup or they pick up in a hurry at the end of Q4 and they go down in Q1.

So really there's a two quarter window of good nice LDP order and after that it's, you know, so so. So I think that kind of behavioral pattern I expect it to remain.

Okay. And just a quick I guess inquiry with regards to this G.hn business. I think you guys mentioned the development teams out of Valencia.

Curious as to how many people are part of that team and what kind of OpEx level, you know, we're looking at. Obviously you described it a little bit as being earnings neutral plus or minus. Personnel wise what are we looking at.

So the team today is a team of about 77 people in Valencia and then there's a handful of people outside of Valencia in the US. So I think that, you know, the -- one of the benefits that we're seeing from picking up a team in Valencia is actually the cost is -- is actually pretty reasonable compared to US development efforts. So it essentially helps us give a broader kind of diversification of resourcing, now we have Burnaby, British Columbia, we've got a very big design center in India, we have got a nice sized design center in Herzliya, Israel, in addition to Irvine and Carlsbad and also (inaudible), Taiwan -- we're getting a pretty global footprint and this just expands that actually gives us one of our lowest score locations.

So we're not at that point yet, we'll give you more color as we get closer to the deal close because we have really just -- we have not really completed our -- our integration planning for the -- for quite yet.

It's really interesting, right? It dawned on me thinking about this question is we will have close to about 200 R&D design engineering pace with Valencia's addition to MaxLinear's already pretty big R&D team and that's quite exciting because it means they are optimizing our R&D OpEx quite across the best talent in the world geographically while taking advantage of the differences in cost structure.

All right. All right. Good. Well, appreciate all the detail, guys. Appreciate you answering all the -- all the nuances of the business. Thank you.

Thank you.

Thank you. Our next question comes from the line of Eric [Rasmussen] with Stifel. Please proceed.

Yes. Thanks, guys for allowing me to get a follow-up here. Just on the tax rate, Adam, what should we be modeling for 2017?

You know, I think most -- if I look out at where -- where this -- your colleagues are at and kind of where we have been directionally pointing people we have said that the tax rate will kind of migrates its way up towards 20% as we progress through the year and I don't think anything has really changed off of that you should be getting to an overall -- right now based on everything we know of course tax is a difficult noise that's coming out of Washington, but right now nothing is causing us to change our outlook and I would say if you have got a blended tax rate that's in the mid-teens on a non-GAAP basis or -- for 2017 I think you're in a pretty good spot. Really no need to change where you're at.

Okay. Thanks. And just one last one. And it's on the DOCSIS 3.1. It sounds like things are on track for more of a bigger ramp in the second half of the year. Two things. Anything that could potentially push that out a little bit that you might foresee now and then, also, any sort of size of how big that business could be whether it's in, you know, revenue or penetration or those sort of things?

So, Eric, you ask a very, very good question and if you recall even in last year -- beginning of last year in the first half I was saying that having seen the patterns ever DOCSIS 3.0 deployment I expect the revenue ramp for DOCSIS 3.1 to happen in the second half of 2017 and it seems that, you know, that's how it's playing out. Is there risk for the rollout, yes, sure in the sense that, you know, it's not as big as they want it to be at this stage, but it's -- in indisputable that by Q4, Q1 the following year is going to be a massive one. So I think that, you know, it is going to be a ramp in the second half and how it starts in Q3 versus Q4 I cannot speak for.

It will be a big ramp. And I do expect that there will be -- there will be more growth in data subscribers for at least a big player like Comcast and with the -- with the way things are playing out you have heard about syndication of the Comcast X1 platform to other cable operators, the uptick in DOCSIS 3.1 will happen much faster because it could spread to multiple operators very rapidly so it could be yet one of the biggest ramps for the DOCSIS platform. Secondly, you know, the -- for us there's an increase in the (inaudible) and as always new standard deploys and we had you had benefit from that. So we expect growth coming over the DOCSIS 3.1 for sure.

Thanks so much.

Thank you. Our last question is a follow-up from Ross Seymore with Deutsche Bank. Please proceed.

Hi, guys. One housekeeping item and one more real question, I guess. Adam, the stock based comp and share count I know this is he when you compensate people in bonuses with shares, not money. So what is the answer to those two for the first quarter, please?

We have been running about $5 million a quarter in stock based comp and a little over $5 million in stock based comp and about $2 million per quarter in the stock based bonus, which I don't think you should assume any real deviations from that. And then on the share count I think on a fully diluted basis we have been -- I think we're modeling around -- probably around 69 million shares.

Got it. And then the less housekeeping related question. If you go into your high speed interconnect business, Kishore, you talked a lot about the number of customers increasing and geographies increasing as well. If my math is right it sounds like you ended the year at about a $5 million quarterly run-rate and you talked about potentially getting to a $30 million number.

When the first quarter is down that kind of implies there's going to be a big pop somewhere and I know that business is always lumpy, but any help you can provide on either seasonality or timing of new product ramps or new customer launches that gives you confidence in that hockey stick at some point during the latter three quarter of this year.

The clear answer is that -- I just want to make this kind of clear that, right, a lot of attention to these higher gigabaud 32 gigabaud, 45 gigabaud, 64 gigabaud linear products but really that's a small percentage of the big deployments that are happening in China will be the case for the -- because they want cost to be low and for the next two years 100 gigabit NRC is going to be the big -- the hugest portion of the deployments.

So having said that, you know, we are launching the TIAs and the ramp of these TIAs is what is going to drive the growth and that will be the second -- and I expect the pops to start happening the second half the year sometime. Now, the -- the good thing about TIAs are they tend to be margin rich unlike drivers which tend to be module and but the negative thing about them is they take a longer time to of qualify because people (inaudible) very good -- complicated optical alignment issues and then people buy (inaudible) directly into the main boxes.

So -- so you have this dichotomy. The cycles stay on track, the ramps happen in second half and can you get pushed a little bit, yes. Therefore, I'm being cautious when I give you the range of $25 million to $30 million of guidance for optical revenues for 2017. And at the OFC, which is in March sometime, we will be announcing a bunch of new products and there will be demos and I think from that point there will be good visibility of all the new products in the linear market that we will be launching which is primarily the 32 gigabaud, 45 gigabaud and 64 gigabaud products but 45 gigabaud and 64 gigabaud products will not generate any meaningful revenue until 2018 or beyond and that is -- that is just a fact for the industry.

Great. Thanks for the detail.

Yes.

Thank you. Ladies and gentlemen, that does conclude our question-and-answer session. At this time I will turn it back to management for closing remarks.

Well, thank you, operator. And as a reminder we will be participating in the (inaudible) Financial Group's sixth annual Semi Storage and Technology conference on March ninth in New York and the 29th annual Roth conference held March 12 to 15 in (inaudible) in California. So we hope to see many of you there. With that being said we thank you all for joining us today and we look forward to reporting on our progress during the next quarter. This concludes today's conference call.

Ladies and gentlemen, thank you for your participation. You may disconnect your lines at this time.