Good afternoon. My name is April and I'll be your conference operator today. At this time, I'd like to welcome everyone to the MaxLinear 2016 Second Quarter Earnings Release Conference Call. [Operator Instructions] Thank you. And now, I’d like to turn the call over to our host, Mr. Gideon Massey, Head of Investor Relations. Sir, you may begin your conference.

Thank you, operator. Good afternoon, everyone. Thank you for joining us on today's conference call to discuss MaxLinear's second quarter 2016 financial results. Today's call is being hosted by Dr. Kishore Seendripu, CEO and Adam Spice, CFO. During the course of this conference call, we will make projections or other statements regarding future conditions or events relating to our products and business. Among these statements, we will provide information relating to our current expectations for third quarter 2016 revenue, including expectations for revenue trends in our cable, terrestrial, satellite, high-speed interconnect, wireless infrastructure and other target markets; gross profit percentage and operating expenses; on GAAP and non-GAAP basis; the potential strategic and operational impacts of our completed acquisitions from Microsemi and Broadcom 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 statements are forward-looking statements within the meanings of the Federal security laws and actual results may differ materially from the results reflected in these forward-looking statements. We are subject to substantial risks and uncertainties that could adversely affect our future results. Our business and future operating results could be adversely affected if our current target markets, including the terrestrial, cable, satellite, high-speed interconnect and wireless infrastructure markets do not grow as we currently expect, or if we are not successful in expanding our target addressable markets through acquisition or the introduction of new products. Acquisitions, including our recently completed acquisition of Microsemi wireless access business unit present particular risks relating to our ability to integrate the acquired business and maintain relationships with key employees, customers, suppliers and other third parties. In addition, substantial competition in our industry, consolidation among broadband operators in our principal target markets, potential declines in average selling…

Thank you, Gideon, and good afternoon, everyone. Thank you all for joining us today. Before jumping into the details of the financial results, we are very excited by the increasing scale of our strategic footprint in the wireless access and backhaul markets, resulting from our recent acquisition of Microsemi and Broadcom wireless infrastructure assets. We also continue to aggressively pursue organic and inorganic growth opportunities in wireline network infrastructure, consisting of high-speed fiber-optic interconnect for metro long-haul and data center markets. The fundamental underpinning of our infrastructure market initiatives and our operator broadband access and connectivity markets, either core CMOS RF, mixed-signal SOC technology platform. This scalable technology platform has enabled MaxLinear to address what is now a growing, sizable and diversified served available market of approximately $4 billion in 2020. Our revenue of $101.7 million in Q2 of 2016 is consistent with the prior guidance range. It is down 1% sequentially and up 44% year-over-year. These relatively flat sequential revenue results were achieved owing to the continued strong momentum in our high-speed optical interconnect solutions, along with seasonal strength in our cable, DOCSIS data and CPE markets. These growth drivers overcame significant expected declines in legacy video SoC shipments and a larger than previously anticipated sequential decline in legacy satellite analog channel stacking outdoor unit shipments. Our ability to maintain a relatively flat top line in the face of large declines in legacy Entropic products, namely video SoC and analog channel stacking outdoor unit products, highlight the strength and increasing diversification of our revenue streams. Our relentless focus on product engineering and supply chain optimization as well as the favorable trend of infrastructure revenues in our product mix has resulted in another sequential expansion of GAAP and non-GAAP gross margins to 61.9% and 63.8%, respectively. Expanding gross margins and…

Thank you, Kishore. On Q2 revenue of $101.7 million, GAAP and non-GAAP gross margins for the second quarter were approximately 61.9% and 63.8% of revenue respectively, versus our original guidance of 60% to 61% for GAAP and 62% to 63% for non-GAAP gross margin. The upside in gross margins was in part due to the mixed effect resulting from the receipt of $1.3 million in revenue from the sale and concurrent license of certain legacy video SoC intellectual property in the quarter. This compares to GAAP and non-GAAP gross margins of 59.6% and 61.3%, respectively in the first quarter of 2016, and GAAP and non-GAAP gross margin of 38% and 58.4% respectively in the year-ago quarter. The delta between GAAP and non-GAAP gross margins in the second quarter was primarily related to the amortization of $2.1 million of acquisition-related purchase intangibles and inventory step-up. Q2 GAAP operating expenses were approximately $40.5 million, $500,000 below guidance. The delta due primarily to higher purchase price accounting charges and acquisition and integration cost and expenses associated with our recent acquisitions, which were largely offset by lower than forecast operating expenses. GAAP operating expenses included $1.3 million in acquisition and integration fees and expenses related to our recently announced acquisitions, $200,000 of restricted merger proceeds and contingent consideration related to our Physpeed acquisition and $800,000 for the amortization of purchased intangible assets. Accruals related to stock-based compensation and stock-based bonus and incentive plans were $4.8 million and $2.6 million respectively. And we incurred $200,000 of professional fees related to the Cresta Technology's patent litigation. Consistent with 2015, payouts under our 2016 performance bonus plan are expected be settled primarily in shares of MaxLinear stock with the first half 2016 awards being made in mid-August. Net of these items, non-GAAP OpEx was $30.6 million, $1.4…

[Operator Instructions] And our first question comes from the line of Anil Doradla. Please go ahead with your question please.

Hey, guys. Just a couple of clarifications here. So Adam, when we look at the guidance, clearly, there seems to be some near-term pullbacks from the two things that you talked about. Now, can you give us a little bit color as to how much of it is driven by your lack of visibility and conservativeness versus how much of it, you've got a lot of visibility and that's clearly happening. Obviously, there are a couple things going around. So can you help us appreciate some of those finer points?

Yes. There are a lot of moving pieces and prior to - similar to prior quarters, there's a range on our guidance. I think what we’ve tried to do and we always tried to do is set the midpoint where we think there's a very high confidence of us getting to. And I think what the guidance really reflects, again, is we've had a tremendously strong first half of 2016 in our operator business. The cable CP platforms have done very, very well. And what we have seen in prior years and will we see playing out today is a return, a consistent return to volatility at some point in the year almost always in the second half of the year and in prior years, it's either been a Q3 or Q4 or the end of Q4. And right now, I think what you're seeing is that we have enough visibility obviously for Q3, given where we are in the quarter that we just feel that we have certainty that the volatility - sorry the seasonality is certainly there for a portion of our business. And part of it also is the fact that we, again, we had a very, very strong Q2 also in our high speed optical interconnect business, because in prior quarters, as that business was ramping, we were really supply constrained. And so what you saw a very, very strong Q2, where it more than doubled from Q1. So I wouldn't say there's - for example, the software guide isn't so much a function of all these businesses being necessarily weaker, but the fact that you have some kind of growing pains in some of these new businesses and they’re naturally kind of sporty [ph] and choppy and I think that we’re seeing this certainly in…

When you talk about the hard cutoff on the analog to digital stacking, you talked about some near-term growth there, so is it fair to say that Q4, we'll see this hard kind of handover between analog to digital or is that something that you just are not able to predict at this stage?

This is Kishore. That would be a logical conclusion. And we have been informed by the primary customer that they are contemplating a hard cutover and to the extent that we are seeing in Q3 some more analog ODU, it's hard to bracket that as an end of life ordering or Q4 and Q1 would present the end of life. But it's very clear now that the analog ODU at this principal customer operator in North America has intimated us that they’re going to go a hard cutover. Having said that, a natural question would be, do you see a stronger amp in digital ODU, but as things go in life, the bad news comes first and then the later [Technical Difficulty]. So our backlog does not indicate the sort of the reciprocal to the analog ODU going away, the digital ODU improving in a lockstep in time. But as we said, we're very optimistic and actually we're positively disposed that our ODU business has long-term growth legs and it's good to do very well because we have the most comprehensive and the superior technology platform. We noticed there is another way to look at all these issues, well there is a video SoC going away legacy products going away. Once they're done, you're going to see a good strong gross margin expansion and at the same time, all the products in our portfolio will be growth portfolio of products. So we're looking forward to this, to put this behind us. Regarding which, we have been telling everybody buy side and sell side analysts about disappearance of the legacy analog ODU and video SoC business over the several quarters we have been in communication with you guys.

And if you don't mind me sneaking one final clarification. So DOCSIS 3.1 ramps, I mean, they're different operators with different strategies. How would you characterize where we are in the DOCSIS 3.1 ramps? Thanks.

I would say that we are still, in my calculus, they are - the North America and Canadian operators have always been more aggressive in the DOCSIS data deployment. So we have some backlog and deployment, for Rogers, in Canada and we have some preliminary backlog deals for a major cable operator in North America. But if we just were to pattern out all these guys function, I'll just say that we are at e=0 and I would say we will ship some in Q1, not something that would meaningful impact our guidance. No, it could go up strongly, these guys have done pretty strange things, just when we think things are down, they come and place large orders for new transitions. But at this point, Q4 is far away. But I really think that 2017 will be a good product cycle for us and we are very, very well positioned. You would also see later at IBC announcements regarding new DOCSIS 3.1 cable app certified products being announced with particular OEMs, this time for some major operators all over the world.

And our next question comes from the line of Gary Mobley. Go ahead.

Just a point of clarification. The analog swim business, correct if I'm wrong, but isn't that approximately 20% of the total revenue?

Well, we said that in Q1, it was about $17 million, Gary. So it was about 17% of revenues roughly, a little bit less than 70% in Q1. Q2, we said it was going to be down as it was in Q2. So if you think about right now, if you think about what it means for Q3 as a percentage of total revenue, it's probably more in, call it may be in the 12%-ish kind of range for total revenue in the quarter. So it's not 20%. So it's been coming down, it’s come down from 17% to let's call it, you know, call it around, it’s around 12% - 11%, 12% in Q2. And it's kind of flat to uppish a little bit in Q3, but kind of treading water, if you will.

I realize it's difficult to predict the transition between the analog swim modules and the digital swim modules with your one main customer, but do you have any assessments of what your market share can be in digital swim, I know you'll be splitting the business with Broadcom, but do you think it will be a 50-50 split?

Gary, this is Kishore. At this stage, I'm very happy to tell you guys that the digital ODU market, the split is maybe 90-10 in MaxLinear's favor for the ones that are deployed. Even if it be even 10, I don't know that. I cannot point out very clearly. However, the analog outdoor unit product, this particular customer get 100%. And now, that's going to transition over to Digital outdoor unit and when the transition complete, we expect that we would be splitting maybe 60-40, in that range for the digital outdoor units or half. Let's call it - just keep it simple, 50-50. And so that's the landscape, which implies that globally, when you look at digital outdoor unit markets, we should have substantially the greater majority of the sockets worldwide. And as we speak, we are really doing a good job in winning the sockets worldwide. And the other thing on the longer-term, you need to understand that the kind of stuff we’re doing with the really, really high frequency integration of Ku-band, down conversions and the stat to IP conversions, it's a strategic platform for us and we believe we will be longstanding winner in this marketplace.

Along the lines of competition from Broadcom, can you talk a little bit about what you feel in terms of Broadcom's commitment towards DOCSIS, specifically DOCSIS 3.1 and in MoCA 2.5 and beyond?

Gary, I cannot speak for Broadcom's strategic plans, but I can speak for MaxLinear's plan. I would say where we stand. I would save DOCSIS 3.1, we work with Intel. It's a very great partner for us. We are pretty much close to offering all the mixed signal in front of the Intel baseband processor. And I would like to say that our platform is, by far, the most superior platform out there. Broadcom does have an offering. We don't know the state of its maturity, however, I would say in North America, with the major operators who are planning the deployment, we're extremely well positioned. With regard to MoCA 2.5, I don't think there's going to any player who is going to have MoCA 2.5 anytime soon, given that even in MoCA 2.0, the bond at MoCA 2.0, bond to Channel 1 is still being qualified or certified and they're not very strongly working solution, except MaxLinear's MoCA 2.1, because now the only fully integrated solution including the LNA, the transmitter, PA and full modem all integrated in a single chip. So I don't think we should think about competition in any manner. You should look at us as the leader in MoCA and we'll continue to be a leader in MoCA and DOCSIS 3.1, we hope to continue to preserve or grow our market share as time proceeds.

And our next question comes from the line of Quinn Bolton. Please go ahead.

Just a couple of clarifications. Kishore, I think in your comments, you talked about an imminent decline in the legacy video SoC and obviously you talked about the hard cutover to digital channel stackers at your largest data analog stacking customers. So just kind of I hear there was comments and it sounds like both of those product lines, sounds like going to zero relatively quickly. Is that the right way to interpret your comments?

I would say that - I wish I could give you more color, but these guys always drag their feet more than not. But they are very good at giving you warnings, just to prepare the supply chain. I would say at this point, it’s a good assumption to work with but when you say two quarters, you have to take into account what Adam already guided for Q3, so within the context of Q3, Q4, Q1, it's a good assumption to make.

And then, I don't think I heard but obviously, if your analog channel stacker customer has notified you of a hard cutover, I assume that you've now seen orders on the books of that customer on the digital stacking side. Can you confirm if they placed those orders?

We are shipping digital audio pretty robustly, but it's not to this particular operator, right. We're shipping to a major operator in North America. But with this particular operator, we still do not have the backlog that we would like to see that gives you the signal that this is a hard cutover but they're still within lead times. So we could - orders could develop for digital outdoor unit because it's a zero-sum game to some extent, right, between digital audio and analog audio for this operator, even though this shares will be split between our competitor and us. But we have not seen the backlog to indicate the converse of what they're been telling of analog audio hard cutover. But it's going to happen. We're just trying to be upfront and honest about this and giving you information as much as we have and as well as we have at this stage.

Quinn, to give you a little bit more color too, I think when we talk in these terms, sometimes there can be confusion in where people put their guidepost. But we're not - I think we should not take away from the discussion is that the revenue goes to zero in Q1. There's no scenario where the revenues go to zero in Q4 and go to zero in Q1. That's not on the table. I would think that - so what you see now is a declining, what we think as a declining forecast starting now for some of these businesses, particularly in Q4 for analog ODU. And naturally these things would have tails on them. And then the question becomes, how do you manage that tail? Do you, for example, do you put last time buys in place so that we can continue to get efficiencies of our supply chain and not be distracted by smaller volumes as they tail out over time. For us, it might be a better decision for us to basically put last time buys in place, pull some of that revenue from later in 2017 into either Q4 or Q1 of 2017. So it's really hard to draw a very specific conclusions on what the profile is going to look like. But I think it's clear that it would be safe to not have any real significant revenue for those two product lines beyond, I would say, the Q1 period. So for Q2, it would be safe to taking it down to some negligible amount, but I wouldn't do that before Q2.

Okay. And then obviously you talked about the optical business, obviously you take a step back into 3Q as you caught up to demand in the second quarter. Based on sort of your visibility, would you expect that business to recover back in the fourth quarter? And if so, do you think it has a chance to get back to the second quarter levels or would it be a more modest recovery? I know Q2 seems like with the doubling in that business, it was a particularly strong quarter.

So Quinn, I think there are two levels to that answer. I know you put out an article on optical super cycle, but if that doesn’t happen obviously, we would benefit from that. But at this stage, we do not - we only have anecdotal information of good things happening, but no indications of real volumes because the carriers in China, for example, have not yet given out the bids. So in that context, we don't see going to Q2 level. However, we are launching new products and they could ramp in Q4, so it's early to say that and if good things were to happen, I do not see any reason in Q4, Q1 window why we won't start seeing some more good outcomes. So we are cautiously waiting to see if the super cycle pans out, but we will really thrilled if that does pan out.

And your next question comes from the line of Ross Seymore. Please go ahead.

Just sticking in that infrastructure category. To get it down to 16%, it seems like that that fall off in high-speed interconnect has to be large, but you should have somewhat of an offset from the two wireless acquisitions in there. Can you just talk about kind of relative movements in the three main pieces in that infrastructure and other segment for 3Q?

So we'll give you some color on that. So again, we mentioned there's going to be some kind of, I would say, smaller declines across TV and terrestrial set-top box and those are really driven more by just there's one of those older markets where you got some ASP compression that play, not a lot of unit growth and so forth. So that’s really not what’s moving the needle much, you're absolutely right that we've got a pretty significant falloff from Q2 to Q3 for high-speed optical. And that's the biggest kind of headwind, if you were to keeping the revenue relatively flat quarter-over-quarter. So if you look at the step down from HSI, there is a more than offsetting step up from the wireless assets that we bought, but there's also another step down, which I mentioned earlier in the call which was related to a sale of some IP and license that was nonrecurring in Q2. So there was about $1.3 million that we got in the quarter that goes away. So that's part of that sequential decline is you've got to step down HSI, you don't have the recurrence of the $1.3 million from the sale of the IP and all of that's being offset by the introduction of the wireless infrastructure revenue and it basically goes to basically keep the business roughly flat sequentially. So those are the moving pieces.

So that IP revenue was in the infrastructure and another category, I take it?

Yes. We just call it other, because it is what it is.

Because it’s other, exactly. And then getting back to your answering the questions about how fast stuff goes to zero. Can you remind us in the analog channel stacking side versus the legacy video SoC, rather than when they might decline. Is there things that would offset that or the glide path down to a low amount be different, you know, different customers in analog channel stacking, single customer and legacy video SoC so that the probability of going to zero might be different between the two?

So Ross, this is Kishore. As you pointed out, the cable video SoC ship to cable operators and analog channel [indiscernible] ships to a major operator in North America. So they are two, these are diversity of two. And to that extent, they could split a bit in terms of time. However, on the video SoC, the consolidation between - the primary customer was, the remaining customer was the Time Warner charter configuration and their consolidation has created a pause or the indication that they may not be interested to further prolong this platform. I've talked about end of life in the SoC platform, so that's what caused these abruptness in my statement about imminent decline. On the analog [indiscernible], I think Adam’s comments are still valid, there will be a small tail. As Adam responded to Quinn, it's not going to go to zero, but it's good idea to assume that it will be a very small tail in Q1 and could be zero in Q2.

I guess there's one final question for Adam. You guys obviously do a good job on controlling the OpEx side of the equation. Off of that base in the third quarter that you guided to, the $32.5 million, how should we think about that changing going forward?

I think it's relatively - we normally don't give guidance beyond the current quarter, but I can say that we're relatively flat, I would say Q3 to Q4 right now is what we're looking at.

And our next question comes from the line of [indiscernible]. Please go ahead with your question.

I appreciate the [indiscernible] and obviously it’s a lot of moving parts and pieces here. As we look at - it seems like working the band aid off where we knew some businesses were going to decline eventually and we’ve kind of come to that day. Can you maybe frame up for us what, absent these next two to three quarters where we have the end of life of other unit and obviously, the SoC, what we're looking at and in aggregate shape in terms of our growth engine for the infrastructure and the remaining businesses, the digital stacking and what not, maybe to framing up in that sense as opposed to what’s going on here in the near term with the noise, if you will.

Hi Brian, this is Kishore. I think I agree very well to the categories here, right? So I would like to still focus first on our operator business, it’s still a substantial part of our business. We have cable data, which is a very setting platform with all the top stuff over-the-top stuff. And the DOCSIS 3.1 would present a great product-cycle growth story moving forward as, if you follow-up Comcast for any of these guys, the subscribers and the data consumption, subscription is really increasing, and they're very excited about increasing the data raise and they will be - and then the ASP improves on these things. Okay, that's one part of it. Second is, they're launching these new MoCA platforms. Obviously, there is some timing risk on the deployments of these new platform, what we call, [indiscernible], which is call MoCA 2.1 for about 1 gigabit per second and MoCA 2.5, what we call [indiscernible] that has 2.5 gigabit per second data backbone for precisely the DOCSIS 3.1 kind of deployment inside the home. So as a platform category that's a pretty good solid growth vehicle, right? And then we move to satellite. We got our 4K satellite gateway front-end. They're doing very, very well. We shipped to the major tier-1 operator in Europe and in North America, and they will prove to be a very strong vehicle for us as well. As the world is now, indeed, switching over ultra HD platforms from all the video and over-the-top operators and their deployment. And then we want to talk about our offerings in infrastructure and other categories, which will very much infrastructure at that time. So that's a growth vehicle right? Our high speed internet connect products for telecom, which is basically long-haul and metro, are really doing…

So if I can summarize, there is obviously a lot of growth drivers that are a bit on the comp. And in the meantime, over the next two to three quarters, as we're winding down these businesses, those growth drivers won't offset the headwind. But it sounds as, though, by the time we get into Q2, I know you were talking about the second half of next year, but it seems like in terms of inflections with seasonality and the end of purchasing, shouldn't we be hitting that inflection point more around Q2 than the second half of next year? Or is it something that takes a little bit longer for the growth engines to kick in?

I mean, sitting here at this point, I want to be conservative in my tone and my approach to next year, naturally, because we need to regroup in how we look at the outward years and the timing of each of these units. But yes, you could be right. But at this point, our second half would be a safer inflection point to think about.

Yes, I think Brian, also is that I would say that - just want to reiterate, too, that we are seeing these - we're seeing the declines in the - declines are in front of us for analog ODU and for the video SoC. But at the same time, I think part of the challenge that we faced that it contributed to our guidance in - for our guidance for Q3 is seasonality in our core business, right? And again, that's something that's new. People that have been familiar with MaxLinear for an extended period of time will realize that, that the seasonality presents itself. And I think that last year, the seasonality was there, and we talked about it, but it was really kind of swamped and overcome by the Entropic accretion and the synergies that we opt in that deals so aggressively. So I think a little bit - 2015 was a little bit of a head-fake, right, because it was there, but people didn't pay attention to it even though we spoke to it. And I don't think the seasonality in the core business across cable in particular shouldn't be a huge shock to investors because it's been a perpetual piece of your business. But just legacy, but there's also core seasonality. So I think that's what people need to think about. I think the big question that we have for ourselves right now is what is the extent of that seasonality. I've said in my earlier comments that it's been there every year. It's either been Q3 or Q4 or Q3 and Q4, and there's been combinations where you had Q3 - like last year, Q3 cable was down strongly - down sharply in Q3 and then basically flattish to Q4. So we don't yet have enough visibility to know whether that's going to play out again this year, as it did last year, or whether you've got even more seasonality on top of that weather could see a little bit of a balance. So it's too early to call Q4, I don't think you can take seasonality for Q4 off the table at this point either.

And our next question comes from the line of Tore Svanberg.

So obviously, you've talked about these legacy declines and other here. So maybe to ask or frame Brian's question a little differently, once you get through this sort of transition period, what type of a growth rate do think you will be seeing from MaxLinear going forward?

Now that's a tough one, Tore. I think that - again, I think Kishore addressed it the right way, which is, if you look at the various pieces, or if you really look at our infrastructure and other, which is now - it is more infrastructure than other, right, and that's where the focus of the investment is, that's where growth of the revenues are coming from. I mean, that we are very confident. You got solid double-digit growth coming in the foreseeable future for that business. And I think that's one that even, again, should be a grower sequentially from Q3 to Q4. That's not going to suffer from the - kind of the, I would say, the legacy headwinds. When you certainly have the operator business, I think the operator business is one where historically we've talked about that business being a 10%-ish grower, plus or minus, depending. It could be a little bit less in the year where there's not a lot of new deployment news. It's going to - should be able deal better when there is a heavy new deployment cycling like DOCSIS 3.1 or 4K satellite gateway front-ends. And certainly, we're seeing that very nice growth on 4K satellite front-ends right now. We had a very good first half year in 2016 driven by the 24- or 32-channel adoptions with DOCSIS market. So yes, I think that when you strip out the analog headwinds - the analog ODU headwind, you should look at our core operator business, which is comprised of all the front-ends, the MoCA plus the digital channel stacking that should be a double-digit growth business. I don't think there's any reason to change from that framework of view. I think that the infrastructure should be multiples of that as far as…

That's helpful. Then coming back to the outdoor unit business. So you mentioned you kind of foresee a pretty steep decline in analog but you haven't really seen the backlog yet in digital. How do you think that's going to play out? Or, I guess, the question is, if digital would potentially ramp for some of your customers in Q1, when would you typically get that backlog or with that visibility?

Tore, this is Kishore. I think that we do get visibility quarter out for sure. And at this point, we have a major operator in North America outside of these analog ODU operators that's ramped very nicely. We got some operators in Europe just ramping, little sputtering. But those should be in solid ramp phase in Q1. And then coming back to this North American operator, there is analog ODU base today and wants to switch over to digital outdoor unit, we expect a 50-50 split. We're going to share the split from 100%. And then the ASPs themselves are down another 30% to 40% relative to analog outdoor units. So all in all, the amount of uplift you get from transition to digital ODU from this primary operator is decent but nothing significant relative to the analog ODU. So you have to keep that in perspective. So if you think about this and, say, fast forward three to four years from now, can satellite be $100 million business? Yes, absolutely. But right now, on a run-rate basis, it's going to be well below that. That's the bottom line.

Sounds good. And on the Cable business, I know you typically don't guide more than a quarter out, but it does seem like per-items mentioning of seasonality in the second half. It does look like that business is maybe correcting a little bit earlier than usual. As you look at Q4, it does look like you also have the growth of the D1s - I mean 3, DOCSIS 3.1 starting in Q4. So is there a chance for Q4 to be flattish in the core Cable business? Or do you think if it's to be down sequentially?

It's very hard to say, because - I'll tell you one thing right, the previous two years, the cable data side, we have seen some very schizophrenic ordering patterns. We thought things are going to go down in Q3, and we guide so. And they started placing orders in a hurry in Q4. And then - and there was a two years before that, typically doing exactly a similar tradition to 3.0, we saw a situation where we actually preannounced a negative outcome, and then at the end of the quarter, they started ordering and we felt like fools, right? So given that situation, we are trying to be at a place where we don't make that mistake. So how Q4 would turn out? Hopefully, it's better. But at this stage visibility, even it's available, is really not good visibility in the sense that it's absolutely volatile and not dependable. And if you add on top of that the DOCSIS 3.1 transition dynamics, I think I am not willing to measure how it's going to play out. But hopefully, it's better than what I'm telling you today.

Fair enough. Just one last question for Adam. Adam, your inventory days came in at 63 days in the quarter, which I believe is an all-time low. Is that just simply a function of some of these end-of-life products that you have with the legacy businesses? Or is it anything else going on there?

No, I think, certainly, we depleted our inventory significantly of our high-speed optical interconnect, and we were supply-constrained pretty severely in Q2, even - we basically shipping out pretty much everything that we had orders for. So I think that we certain step that we have very lean inventories in interconnect products. We're basically building to backlog for these end-of-life products because we don't want to be left with inventory in hand. So I would say it's fair to say that the end-of-life for legacy products is having a pretty big an influence in how we manage our inventory levels right now, and it's all building to backlog. Very little reliance on any forecast at this point, because we ought to be left for stuff on hand.

And there are no questions at this time. Do you have any closing remarks?

Thank you very much, operator. As a reminder, we'll be participating in the Roth Capital's Second Annual Data Center Technology Conference on September 7 in San Francisco, the Deutsche Bank Technology Conference on September 13 in Las Vegas, and 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 to you in the next quarter. Thank you.

This concludes today's conference call. Thank you for your participation. You may now disconnect.