Semtech Corp (SMTC) 2017 Q3 法說會逐字稿

Good afternoon. My name is Amber and I will be your conference operator today. At this time, I would like to welcome everyone to the Semtech Corporation's third-quarter FY16 earnings release conference call.

(Operator Instructions)

Thank you. Sandy Harrison, Director of Business, Finance and Investor Relations, you may begin your conference.

Thank you, Amber. Welcome to Semtech's conference call to discuss our financial results for the third quarter of FY17 ended October 30, 2016. Speaker for today's call will be Mohan Maheswaran, Semtech's President and Chief Executive Officer; and Emeka Chukwu, our Chief Financial Officer. A press release announcing our unaudited results for the quarter was issued after the market closed today, and is available on our website at Semtech.com.

Today's call will include forward-looking statements that include risks and uncertainties that could cause actual results to differ materially from the results anticipated in these statements. For a more detailed discussion of these risks and uncertainties, please review the Safe Harbor statement included in today's press release as well as the Other Risk Factors section of our most recent periodic reports filed with the Securities and Exchange Commission. As a reminder, comments made on today's call are current as of today only and Semtech undertakes no obligation to update the information from this call should facts or circumstances change.

During the call, we will refer to non-GAAP financial measures that are not prepared in accordance with generally accepted accounting principles. Discussion of why the Management Team considers such non-GAAP financial measures useful along with detailed reconciliations of such non-GAAP measures to the most comparable GAAP financial measures are included in today's press release. All references to financial results in Mohan's and Emeka's formal presentations on this call refer to non-GAAP measures unless otherwise noted.

With that, I'll turn the call over to Semtech's Chief Financial Officer, Emeka Chukwu. Emeka?

Thank you, Sandy. Good afternoon, everyone.

For Q3 of FY17, GAAP net sales was $137.2 million, an increase of 1% sequentially and 18% from the prior year. GAAP net sales were reduced by $3.7 million of share-based compensation associated with the Comcast warrant, excluding these reductions, net sales were $140.9 million. A 4% sequential increase, a 22% increase over Q3 of FY16 and the fourth consecutive quarter of results above the midpoint of our guidance.

Excluding the impact of the Comcast warrant revenue reduction in Q3 FY17, shipments into Asia represented 77% of total net sales. North America represented 16%, and Europe represented 7%. Total sales to distribution represented approximately 64% of net sales, and direct sales represented approximately 36%.

Bookings in Q3 increased nicely over the prior quarter, and increased by nearly [40%] over Q3 of FY16. The book to bill was again above 1, and the [Touch] bookings accounted for approximately 50% of shipments during the quarter.

Gross margin on a GAAP basis for Q3 of FY17 was 59.1%, down 108 basis points sequentially due to higher share-based compensation associated with the Comcast warrants. In Q4 of FY17, we expect GAAP gross margin to increase between [30] basis points to 90 basis points from the prior quarter as we expect to record a lower net sales reduction from the Comcast warrants.

Operating expense on a GAAP basis declined approximately 36% from the prior quarter. Reflecting the impact of a $25 million gain from the divestiture of our Snowbush IP business, slightly offset a $2.1 million bad debt reserve for a distributor who suddenly filed for bankruptcy protection. In Q4 of FY17, we expect our operating expense on a GAAP basis to increase significantly due to the non recordings of the gain from the divestiture.

In Q3 of FY17, our GAAP tax rate was 15.7% compared to 37% in Q2 as the gain from the Snowbush divestiture was taxed at a favorable rate. In Q4, we expect our GAAP tax rates to be approximately 20%.

Now I will talk about our non-GAAP results. Which exclude the impact of share-based compensation, the gain from the Snowbush divesture, amortization of acquired intangibles, acquisition or disposition related and other restructuring related expenses and reserves for certain liabilities not tied to current corporations.

In Q3, non-GAAP gross margin was 60.4% which was flat sequentially. We expect Q4 non-GAAP gross margin to be flattish, as higher bonus accruals expected to offset a more favorable revenue mix. Q3 non-GAAP operating expense was $51.9 million, up 1% sequentially.

During the quarter, we took a $2.1 million bad debt reserve for a distributor who suddenly filed for bankruptcy protection. The impact of this charge was partially offset by lower bonus accruals. In Q4, we expect non-GAAP operating expense to increase slightly from the prior quarter due to higher bonus accruals offset by the benefit from our scheduled holiday shutdowns.

In Q3 of FY17, our non-GAAP tax rate was 20%. In Q4, we expect the rates to be approximately 20%. For more modeling purposes, we expect our FY18 non-GAAP tax rate to be between 20% and 23%.

In Q3, cash flow from operations increased 24% sequentially and 108% over the Q3 of the prior fiscal year to approximately $39 million or 28% of net sales. We ended the period in a net cash position with $298 million of cash and investments, up from $212 million at the end of FY16. In Q3 of FY17, 77% of our cash and investments were [domiciled] in international accounts.

Our debt balance of the end of Q3 was approximately $248 million. We recently refinanced our credit facility to get a greater operational flexibility, and extend the maturity at almost the same cost as the prior facilities.

The balance on the outstanding stock repurchase authorization stands at approximately $62.2 million. The primary use of cash continues to be to pay down our debt, to repurchase our shares and make strategic investments. And Q4 of FY17, we expect to purchase our office building in Burlington, Canada which houses a significant portion of our signal integrity products group's operations for approximately $12 million.

Accounts receivable for Q3 of FY17 increased 4% sequentially, in line with net sales growth, and our days of sales outstanding increased by two days to 38 days and remains below our target range of 40 to 45 days. Net inventory in absolute dollar terms in Q3 FY17 was flat with the prior quarter, while days of inventory declined four days sequentially and represented 101 days of inventory, just above the target range of 90 to 100 days. In Q4, we expect our net inventory and our days of inventory to increase as we build ahead for the Chinese New Year.

In summary, we're pleased with the FY17 strong performance to date and the better than seasonal expectations we have guided to for Q4. If we achieve the midpoint of our Q4 guidance, FY17 net sales would have increased approximately 11% and operating profit would have grown a corresponding 53%, which we expect would soundly outperform our peers in the industry.

This leverage is enabled by a stable gross margin at the high end of our 55% to 60% range, and operating expenses that are remaining in check. We believe we're making solid progress towards our non-GAAP operating profit target range of 25% to 30%.

I will now hand the call over to Mohan.

Thank you, Emeka. Good afternoon, everyone. I'll discuss our Q3 FY17 performance by end market and by product group, and then provide our outlook for Q4 of FY17.

In Q3 of FY17, we achieved non-GAAP net sales of $140.9 million, which was an increase of 22% over Q3 of FY16 and an increase of 4% over the prior quarter. We posted non-GAAP gross margin of 60.4%, and non-GAAP earnings per diluted share of $0.37.

In Q3 of FY17, demand from the high-end consumer end market increased from the prior quarter and represented 29% of total net revenues. Approximately 22% of high-end consumer net revenues was attributable to handheld devices, and approximately 7% was attributable to other consumer systems.

Demand from the enterprise computing end market decreased from the prior quarter, and represented 27% of total net revenues. While demand from the industrial end market was flat with the prior quarter, and represented 26% of total net revenues. Finally, demand from the communications end market increased from the prior quarter, and represented 18% of total net revenues.

I will now discuss the performance of each of our product groups beginning with our signal integrity product group. In Q3 of FY17, net revenues from our signal integrity product group declined 4% sequentially as anticipated but grew 15% over Q3 of FY16 and represented 43% of total net revenues. Net revenues from the enterprise computing end market decreased sequentially as we had expected, as demand for our data center CDR products softened slightly from a record Q2 performance.

In addition, our PON business remained soft during the quarter. Net revenues from the communications end markets increased as we saw a modest increase in the demand for some of our telecom backplane products. Revenues from the industrial end market were approximately flat.

We have recently seen a pickup in activity from both the PON and wireless space station markets, and expect to see both these businesses start to recover in Q4 and continue into the first half of FY18. We also expect demand for our 100 gigabit per second CDR platforms used in cloud and hyperscale data center platforms to increase nicely in Q4, and potentially set another quarterly record.

Semtech's signal integrity product group is benefiting from several main drivers that should result in continued growth and record revenues over the next several years. First, Semtech is one of the leading providers of clock data recovery circuits, or CDRs, and physical media devices, or PMD's, used in 1 gigabit per second, 2.5 gigabit per second, 10 gigabit per second, 25 gigabit per second and 100 gigabit per second optical modules to the data center, PON and wireless base station markets.

Second, the increase in use of 100 gigabit per second modules by the cloud and hyperscale data center markets is leading to higher demands for Semtech's Quad 25 gigabit per second CDR solutions, and contributing to higher Semtech content in these optical modules. Semtech's CDR platforms offer best-in-class performance with a lower overall solution cost.

Third, our signal integrity product group continues to expand its optical portfolio, including single and quad channel optical solutions with integrated PMD functions. We recently announced the GM28 L95 integrated laser driver and limiting amplifier platforms, targeting asymmetric 10 gigabit per second PON applications.

Our roadmap includes our higher performance single lambda 100 gigabit a second PAM4 platform. This program placed on modified DSP architecture is progressing well, and we anticipate sampling first silicon in Q1 of FY18.

We have a number of strategic engagements with tier 1 customers for our new single lambda 100 per second PAM4 platform. And we believe once the platform is available, we will gain a lead position in the emerging 100 gigabit per second single lambda optical market as well as the quad lambda or 400 gigabit per second optical market for cloud and hyperscale data center solutions.

We believe our signal integrity product group is very well positioned to continue its double-digit revenue growth which it has produced over the last five years. For Q4 of FY17, based on strong bookings and a healthy starting backlog, we expect net sales from our signal integrity product group to increase led by increasing demand from the data center, wireless base station and PON markets.

Moving on to our protection product group. In Q3 of FY17, net revenues from our protection product group increased 10% sequentially and increased 21% over Q3 of FY16 and represented 29% of total net revenues. Demand increased from the high-end consumer, enterprise computing and communications end markets, while the industrial end market declined from the prior quarter.

Demand from our Korean smartphone customers increased from the prior quarter, despite the cancellation of Samsung's Note 7 smartphone. While the impact of the Note 7 cancellation is significant, our diversification and growth drivers across our protection business and our overall business provide enough balance to minimize the impact. Demand from our Chinese smartphone customers was also very strong in Q3, further diversifying our protection consumer business and positioning us well for sequential growth in Q4.

Our protection product group continues to benefit from several industry trends. The most significant trend is the transition by mobile system manufacturers to more advanced chip lithographies, which results in lower levels of on-chip protection as protection circuits typically require larger geometries to be effective.

These mobile system manufacturers are turning to Semtech's innovative Z platform to ensure their systems are protected against ESD and surge events. This trend is also driving the need for more advanced protection to be added in communications, computing, industrial and automotive systems where once good enough discrete protection devices are no longer sufficient to protect sensitive electronic systems. This is particularly evident in the automotive and enterprise computing segments, as both these segments are increasingly using advanced chip lithographies.

Our protection product group continues to focus on expanding its platforms to address broader vertical markets. During the quarter, our protection product group introduced new platforms targeted at interfaces covering a broad range of applications and end markets.

We introduced the RClamp 2594M, a full-line gigabit ethernet protection device for telecom and industrial systems. This device provides protection for 35 amps of surge current on quad 2.5 gigabit per second ethernet lines. We also introduced the EClamp 2374KQ, a full line integrated ESD protection and EMI filter for automotive infotainment applications.

We believe our strategy of focusing on advanced lithography protection and diversifying our protection business is paying off, and should help drive growth in our protection product group over the next several years following a return to growth this year. In Q4 of FY17, we expect our protection business to be approximately flat from the prior quarter, as lower seasonal demand from our Korean smartphone customers should be partially offset by growth from our Chinese smartphone customers.

Turning to our wireless and sensing product group. In Q3, net sales from our wireless and sensing product group grew 16% sequentially and grew 45% over Q3 of FY16 and represented 17% of total net sales. This was a new quarterly net revenue record for our wireless and sensing group, led by record quarterly revenues from both our LoRa platforms and our proximity sensing platforms.

During Q3, demand increased for our LoRa platforms, as global LoRa deployments continue to accelerate. In Q3, several trials of deployments using a LoRa based low power wide area network for IoT and M2M applications were announced.

These announcements included Comcast announced an agreement where LoRa technology enabled platforms will be used in their machineQ trial venture service to provide IoT connectivity to enterprise and industrial customers. This service is expected to start with two cities, and can be expanded up to 30 cities within a period of 30 months.

Comcast is the first major carrier in the US to announce a LoRaWAN platform, and we see this as a key milestone for LoRa in the North America, market. It is early in the program but the feedback from Comcast to date on the technology and its value has been very positive, and they're on track to complete their initial network trial deployments by end of March 2017.

SoftBank of Japan announced that they would be deploying a LoRaWAN based network that is expected to provide services for applications such as automated gas and water meter reading, tracking for senior citizens and children, monitoring highway and railroad infrastructure and managing of logistics and fleets of cars, trucks and buses. ZTE in China announced the launch of their first IoT demonstration base dedicated to LoRa technology in Nanjing, ZTE communicated initial plans to deploy LoRa in over 30 cities.

South Korea Telecom announced the completion of their nationwide LoRaWAN network, and are now actively driving IoT and M2M use cases in South Korea. Digital Catapult in the UK launched a three LoRaWAN network across London intended to foster IoT innovation by providing small businesses with free access to a LoRaWAN network for testing IoT applications. The initial network is expected to consist of 50-base stations, making it the largest LoRaWAN network currently in the UK.

IoT solution providers [Kottai, HetNet ] and [Loriot] announced the deployment of a LoRaWAN smart-sensing network that covers more than half the population of New Zealand, and allows businesses, local governments and conservation groups to collect and analyze IoT data. And Pervasive Nation announced an IoT network rollout in Ireland based on LoRa.

These are just a sample of some of the 50 LoRa-based networks in trial or full deployment across various regions of the world. It is becoming clear that the low-power long-range and low-cost of a LoRaWAN network is a true enabler to many emerging IoT applications. We continue to expect that most if not all major regions of the world will have some form of LoRaWAN network deployed or in the midst of deployment by the end of calendar year 2017.

The global success of our LoRa technology continues to be driven by our Partners in the LoRa Alliance. Which now represent over 400 companies that are involved in everything from developing LoRaWAN sensors, developing LoRaWAN gateways, deploying LoRaWAN networks, providing LoRaWAN cloud-based analytics and offering LoRaWAN-based IoT services.

To further expand the number of LoRaWAN IoT use cases in the market, we just announced the availability of our LoRaWAN Picocell gateway reference design. This LoRaWAN Picocell will have a USB interface and enable consumers and small enterprises to create their own private LoRaWAN network at very low cost. This indoor Picocell gateway can be used to connect thousands of sensors across a several mile radius for less than $100, making it perfect for smart building, smart enterprise, smart campus and smart retail environments.

The Picocell gateway reference design has already been picked up by several OEMs and will be marketed under their brand names. We will demonstrate the Picocell gateway reference design along with other gateway reference designs at the Consumer Electronics Show in January together with our LoRa Alliance Partners.

In Q3, we also announced that we received our first LoRa geolocation related license and royalty revenues. While small today, we do expect these revenues to increase as the use of LoRa for geolocation and geofencing increases and the availability of LoRaWAN networks increases. Our GPS free geolocation capability that allows IoT operators to add asset tracking and geolocation to any LoRaWAN sensor connected to a LoRaWAN network is creating additional revenue opportunities for LoRaWAN operators and service providers.

We have four signed agreements in place with an increasing number of others in the pipeline. As our geolocation accuracy improves and the global use cases increase, we will provide more details on this very exciting part of our business. Out LoRa business continues to exceed our highest expectations, and we believe we are well positioned to become a leader in the global IoT market.

In Q3, our proximity sensing platforms also delivered a quarterly net revenue record, as mobile device manufacturers use our proximity sensors to help minimize harmful RF emissions. Historically, tablets have been the primary target for our proximity sensing devices, but we're seeing increased demand from smartphones and wearables as these mobile devices integrate high-powered radios and as human body protection regulations increase.

Driven by these trends, we expect the demand for our proximity sensors to continue to grow. For Q4 of FY17, we expect net sales from our wireless and sensing product group to decrease slightly from Q3, as lower seasonal proximity sensing demand is slightly offset by what is expected to be another record performance from our LoRa business.

Turning to our power and high reliability product group. In Q3 of FY17, our power and high rel product group increased 5% sequentially and increased 18% over Q3 of FY16 and represented 11% of total net sales. Demand increased from the prior quarter from our communications and industrial end markets, while demand decreased from the high-end consumer and enterprise computing end markets.

Our power and high rel business remains focused on delivering differentiated platforms for the automotive, industrial and wearable segments. Semtech's wireless charging platforms deliver flexibility and versatility by offering a programmable, multi mode and scalable power range from 100 milliwatts to over 20 watts. Ideal for wearable and infrastructure applications.

We recently announced the link charge CT, a dual-mode wireless charging system for use in public, enterprise and consumer furniture countertops. Our solution is compatible with current industry wireless standards, enabling interoperability between furniture infrastructure and a diverse number of mobile devices and wearables. Our wireless charging business has a number of design wins with a broad group of customers in fast-growing markets that should start to move to production over the two to four quarters.

During Q3, demand increased nicely for our isolated switch platforms. Semtech's solution is optimized for low-voltage switching applications such as smart thermostats and other home automation systems where customers are looking to replace older mechanical relays with smarter, quieter and more reliable solid-state technologies.

We recently announced two new Neo-Iso products, the TS13 102 and the TS13 103 that expand the Company's Neo-Iso platform, and enable self-powered control systems in IoT applications such as smart thermostats, alarm panels and factory automation systems. We expect this business to continue to grow as more devices transition from older mechanical relay solutions. In Q4 of FY17, we expect sales from our power and high reliability product group to decline consistent with normal seasonality.

In Q3, the total Company distribution POS increased 16% from the prior quarter, and achieved a new 13-week record. Distributor inventory declined by three days from 73 days in Q2 to 70 days in Q3, and remains at the lower end of our targeted range of 70 to 80 days. Our distributor business remains very well balanced with 52% for the total POS coming from the high-end consumer and enterprise computing end markets, and 48% of total POS coming from the industrial and communications end markets.

Moving on to new products and design wins. In Q3 of FY17, we released 24 new products and we achieved 1,912 new design wins.

Now let me discuss our outlook for the fourth quarter of 2017. Based on a strong starting backlog and strong current bookings trends, we are currently estimating Q4 non-GAAP net revenues to be between $134 million and $142 million.

To attain the midpoint of our guidance range or approximately $138 million, we needed net terms orders of approximately 36% at the beginning of Q4. We expect our Q4 GAAP earnings to be between $0.13 and $0.17 per diluted share, and our Q3 (Sic-see press release "Q4") non-GAAP earnings to be between $0.33 and $0.37 per diluted share.

I will now hand the call back to the operator, and Sandy, Emeka and I will be happy to answer any questions. Operator?

(Operator Instructions)

Cody Acree, Drexel Hamilton.

Thanks guys for taking my questions, and congratulations on the continue progress. Mohan, could you maybe give any more specifics about the sequential impact or maybe what you expect the sequential impact to be from the issues with the Note 7?

All the impact is built into our guidance, Cody, so we had content both protection and sensing content in the Note 7. So it was a disappointment to see it canceled, but the impact as you can see is fairly minimal. We have in Q4, the projected revenue was about between $4 million and $5 million from the Note 7 phone which is not there any longer, and you can see the strength of our guidance and the demand is fairly well balanced because a fairly minimal impact for us.

Was it similar for Q3?

Q3 was smaller, it was about a $1 million impact.

Okay, great. Thank you for that. Could you maybe, Mohan, just rank order through FY18 maybe just your views of from a dollar growth contribution, whether it be by product application or by product or application, where you expect to see the most growth?

Well in terms of growth rates, obviously our wireless and sensing business is growing phenomenally well. Obviously the LoRa story is gaining great momentum, also the sensing side as well because the proximity sensors that we have now, as I mentioned on the call, getting into the mobile devices, both smartphones and wearables, is going to drive the growth there.

On the single integrity side, clearly data centers is going to continue to grow nicely. We believe that we have a very good position in 100 gig CDRs, the quad 25 gig CDRs, and as I mentioned on the call, PON and base stations coming back a little bit. PON for sure, 10 gigabit PON we have a very strong position, I think that will grow nicely. So that signal integrity product business should grow well, as well.

On the protection side, we think that is now that the Note 7 issue is behind Samsung that they will see a little bit of recovery next year. But regardless of whether they do or not, we feel good about our position in China, protection and smartphones, and also our diversification strategy.

We're getting more protection into automotive applications, into computing and communications applications. So we think our protection business will show growth. And then on the power and high rel side, really the wireless charging, it is still early days but we still think that, that is a segment of the market that is going to at some point take off here as more and more devices require wireless charging and infrastructure gets more deployments.

And then maybe lastly, could just talk about the competitive landscape for LoRa? Obviously a lot of different standards out there for low-power wide-area networking. Could you just talk about what you are seeing as far as maturation and maybe where you see the most competition?

Well that is the beauty of our LoRa. So on the low-power wide-area network, there really isn't that much competition.

From our standpoint, the cell phone technologies like LTM and NBIoT which supposedly play in the same space but are higher power and then higher cost technologies. So we don't really view them as competitive, and they are higher bandwidth technologies so we view them as more of a complement to LoRa.

And then the other competitors we have are very specific to one vertical, or they are more service providers. So competing against the likes of Comcast and SKT and those guys, not really competing against us. So from that standpoint, we feel like we have a pretty good position with where LoRa is today.

Congratulations. Thank you.

Steve Smigie, Raymond James.

Great. Thanks a lot. I'll just I guess follow up on that last question.

With regard to LoRa, I think you had a product out today taking you beyond [on laces] and spectrum for LoRa into licensed. And I was hoping you could talk around the dynamics around that a little bit. Obviously, it seems like the carriers are hoping to have some advantage in the license spectrum, so I just curious, any color you could provide there.

Yes, so today's release was not anything related to that, Steve, maybe you misread that. What we announced yesterday was the Picocell Gateway. So this is a -- actually it's a LoRa gateway essentially for indoor usage.

So what it does is essentially what we've had up to date our carrier grade gateways, these are gateways that deploy on towers or strands and essentially cover a city. So what we've done with the Picocell Gateway is just provided an entry point into the consumer and small enterprise space by allowing anyone, a consumer or a private enterprise, to put a LoRa Gateway in their home or in their building and connect to a bunch of sensors to create their own private network at a very low cost. So, it certainly -- it's still using unlicensed sub-1 gigahertz technology.

Okay. Could you talk about where you think Chinese is as a percentage of handsets now? So is Samsung maybe 5%ish of revenue? It's back to where Samsung -- and if you added up all your Chinese guys, Huawei, et cetera, what would they be as a percent of total revenue?

Yes, Steve, this is Sandy. So if you look at where the Chinese are, they're in that mid-single digits right around there depending upon any particular quarter. That's basically where we're seeing them come in at.

Okay, great. (Multiple speakers).

And just to add some color, Steve, for the next year, we anticipate that our China handheld business will be probably about 40% to 45% of the total handheld business and Korea will be about 55% to 60%. So certainly China for us, the China handset business has come on nicely and is offsetting to some extent some of the issues we've had from our Korea business.

Okay, great. And then in the 100g space, I was hoping you could address a couple topics. One of which would be just where you see your market share at this point in 100g data center.

And then I was hoping you'd talk a little bit more about DSP and multi-fi and progress you're making there. Obviously, there was an acquisition announced by one of your competitors, so I was hoping you could talk relative those guys and also versus an infi as well versus the non-single end approach.

Yes, so today, obviously our approach is selling CDRs. We sell CDRs and we sell PMD devices into the 100 gig market but also 25 gig, 10 gigs, and lower bandwidths and that is where the revenue is today, and that's what we're doing. So most of our 100 gig revenue is coming from quad 25 gig CDRs modules, and we think we have a very good position there.

Certainly we're having record quarters. We anticipate this year will be a record growth from last year, I think some 250% growth versus last year's 100 gig CDRs sales. So certainly we feel we have a very strong position, and I think that we anticipate that will continue next year, and that is both data center and metro sales.

The question is really on the timing of single Lambda 100 gig programs, and we obviously are one of the first ones to get there and then talk about a strategy and approach with that with the multi-fi deal we did. Good progress, as I mentioned, we should see silicon in the Q1 of FY18. I would anticipate by the second half of next year that things will start to clear up a little bit about which customers and which technologies are going to win, but certainly revenue probably is not going to be significant I think until maybe even the 2019 -- 2018, early to 2018 to 2019 timeframe.

But we think we are well positioned with that platform. But for today, all the revenue is coming from CDRs and PMD devices.

So obviously our competitors are Macom and Infi and then they have their own acquisitions and their own DSPs, and I think there is enough room for several players. I don't know exactly how it's going to play out, but today we believe our position is very good.

Thank you.

Craig Ellis, B. Riley & Company.

Thanks for taking the question. I'll just start with congrats on what has been very good execution, and to Emeka and Sandy, thanks for all the transparency with the GAAP and non-GAAP translations.

Starting off with protection, Mohan, you're clearly seeing very good diversification in that business. So taking a step back, where are we in the longer-term perspective of your desire to diversify that business? Are we 80% of where you would like to be, something less than that, just give us a sense for how you feel about the diversification big end markets and customers within some of those end markets?

Yes, it's a good question, Craig. I think it's early days in that diversification. The first, obviously we had many quarters of sequential decline in this business and the goal was to start to grow the business again, and we've now had two sequential quarters of growth which is very encouraging.

But the charter was, okay let's diversify away from handhelds and Samsung, and try to get some more automotive and industrial business and communication business. And we have new products that are coming out in those areas, so I feel good about the diversification and strategy there. But the other aspect of it was to crack other handheld guys other than Samsung and we've done that with the Chinese.

We're doing very, very well, particularly in China. So I would say it's early though, I would say we're 30% of the way there. We still have a ways to go before I would say that this is now a fully diversified business so that really mitigating the risk against downside from any of the handheld guys.

But we do like the handheld space still. I still believe that it is a good space for us. As I mentioned, the smartphones across the globe are moving to more advanced lithographies, and that really fits our technology very well.

So I hope that we'll continue to see that trends in all of the markets. And if that trend continues in the other markets, I think then protection you're going to see is -- gets back to the double-digit growth, which is really what we want.

Just following up on that point, if we see as I think there are indications that three of the four baseband manufacturers that could go to 10 nanometer in the first half of the year are from China, and are going there. Mediatech from Taiwan, but HiSilicon and Spreadtrum are all I think planning to release 10 nanometer baseband product. If that happens and given that those are likely to be deployed into China handsets, what does that do for protection uptake versus the last transition at 16, 14 and the one before that at 20?

It just makes -- it just means that what happens is that all of your interfaces, your antenna interfaces, your memory interfaces, your touch screen interfaces, your microphone, all of these things have to be protected because of the risk of damaging your main baseband processor or some other key component in the device. And doing it at those at the higher speeds with the more advanced lithographies is just very, very difficult to do with traditional protection, and that is why we've always said, we said on our analyst day and we've said for many years, that as the trends go in our favor, more companies will -- more devices will come to Semtech for their protection.

Then the next question is with regards to LoRa. Clearly deployments are moving along at a torrid pace, and I'm not asking for a guidance question on 2017 and I think you clarified some of the relative growth dynamics in response to Cody's question.

But can you help us with the key milestones as you look at that business for 2017, whether it be material licensing revenues or deployments in regions where they may not be at a level where you would you like right now? What are some the key things that you would like to see with that business as you go through the coming year?

Yes, so I have gone out and said that we will get this business to $100 million within a few years. This year, we'll end up somewhere between $20 million and $30 million of LoRa enabled business, and then next year I hope that will double to about $40 million to $50 million and then double again in FY19. But obviously, that revenue is based on several factors.

One is more and more gateway deployments, second is more end nodes connecting to those gateways, and then the third is geolocation and other types of revenues coming from them. The good news is that we're seeing a lot of gateway deployments. And so across the globe, we've mentioned Korea, I've mentioned China, I've mentioned, obviously the Comcast deal, we're hopeful that, that is going to translate into a nationwide deployment in the US and a lot of connectivity based on that.

So the way we view it is that as the networks get deployed, it's just a question of timing. Whether how quickly the use cases come on board, and so we are looking at that more and more.

The nice thing is really about this IoT space, it is really a new industry that is being created and therefore the opportunity is huge and the applications are diverse. It's not just one application space. But those are the things we look for, gateway deployments across the different regions and then end node deployments and use cases.

Thank you. And then the last question is for Emeka. Emeka, beyond the fiscal fourth quarter $52 million midpoint operating expense number, if we see the business grow and grow solidly in the first part of FY18, what does that do to operating expenses from that $52 million point? Can they be sticky or do you need to add talent to fund some of the growth that is possible longer term?

No, Craig. I think like I indicated in our call last quarter, the expectations that we have now and what we're planning to is to have the next fiscal year essentially be flattish in terms of OpEx on a non-GAAP basis to FY17. So really, driving a lot of leverage off the expected top line growth in FY18 is really a focus of this new management team and this company, so I would expect that as we go into next year that operating expenses should be running at the current levels.

Sounds good. Thanks, guys

Harsh Kumar, Stephens Incorporated.

Yes, hey, guys. Congratulations, tremendous execution, Mohan, Emeka and the team.

Mohan, a just had a quick question on optical. On the PON side, I noticed that I think your commentary said that you are expecting a little bit of a pickup. I was wondering if you could distinguish between the parts upon that you're seeing a pickup? Are you seeing a pickup in 2.5 gig versus 10 gig or are you seeing a universal lift up here shortly?

Mostly 10 gig, Harsh. We anticipate the 10 gig was going to start to emerge, but it is real now. We're seeing good orders in that space, and so that is very encouraging. We think our position is very strong in the 10 gig PON space, so that is where most of the strength is.

Okay. And then I'm looking at your guidance, and I think historically, I think this time you said you needed 36% turns to make the midpoint of your guide. I believe historically in the past you have needed higher turns?

Am I accurate in that assessment, and your backlog is just pretty strong going forward you're just keeping room? Is that how I should look at it, or is this a normal level of turn that you assume at this time of the year?

No, your commentary is very accurate, Harsh. Typically we would require a much higher number of turns. The backlog was very, very strong for us in Q4 and so the turns requirement is lighter, particularly in the signal integrity product group and wireless and sensing product group.

Great. That's all I had, guys. Congratulations.

Rick Shafer, Oppenheimer.

Thanks, guys, and I'll add my congratulations on a nice quarter. I guess just one more quick follow up on LoRa.

You guys obviously announced the Comcast milestone in the quarter. Since then, it has been a month or two, have you guys seen any noticeable impact or acceleration in terms of adoption or interest from some of the other players now that they have bought in, a big name like Comcast has bought in? Any commentary there?

Yes, I would say obviously Comcast is a big name globally actually, and we did an announcement recently with Dr. Peng in China which is one of the Comcast equivalents over there. We are seeing cable operators around the world start to look at this very closely and say hey, we really don't need a mobile system in place, network in place, to be able to do this. Let's take a look at what Comcast is doing.

One of the reasons why we did the Picocell Gateway announcement just yesterday was to demonstrate that actually we can create a very powerful LoRaWAN private network throughout the globe with a small gateway and end node devices. And then if you add to that the likes of a Comcast or a big network M&O like an SKT, you can get pretty good nationwide coverage from end to end quite quickly at pretty low cost.

And so that's somewhat a game changer. So Comcast obviously with their namebrand and what they are doing and how they are thinking about this is a very powerful message, and it's certainly getting a lot of -- we're getting a lot of leverage from that.

Thanks. And maybe just switching gears to wireless charging. Any update there, maybe if you could frame how big that business is for you guys now? And how big does that business need to be or what contribution does it need to be making for you guys to hit that billion dollar target you guys have talked about?

Well today it is really small, Rick. I think it is still in its early stages, and we're playing mostly in the transmitter side. But we think it is one of those industries that is going to become much bigger quite quickly once all of the smartphone manufacturers and all of the wearable devices adopt wireless charging, which they do not have today, but once they do that, I think it will stimulate more demand for wireless charging. It's quite a challenging technology to implement, and so we think we have a good position there.

We believe it will be $100 million business for us sometime down the road, but I think it's quite a ways off from that. I think initially it's going to probably take a three, four years before we get to a point where we could say it's on its way there. I think that's a 10-year journey.

Okay. That's clear. And then my last question maybe is for Emeka. How would you guys frame your legacy business at this point? Are we at a point where revenues are stable, do you feel like there is any risk there or any future impact or drag on growth, or are we past all that now?

I think we're past most of it. A lot of the decline in our legacy business is coming from the old long-haul [certies] and stuff that we got out of a few years ago, and I think that business has been down significantly to basically noise level at this point. So my expectation is that on the legacy side that we're past most of the decline and anything else that it should be maybe flattish to slightly down going forward.

Got it. Thanks a lot.

Ian Ing, MKM Partners.

Yes, thanks for taking my questions. Emeka, debt level question here. You're net cash turned slightly positive in the quarter.

Could you remind us how much operating cash you need to run the business? And if your net cash does cover operating cash, would you retire the debt immediately or keep some leverage for other reasons? Thanks.

Well the problem here with regarding the debt immediately is that a lot of the cash that we have, 77% of that is still offshore. So -- and we can only service the debt with our domestic cash. But going back to your question of how much cash to we need to run the business, that is always a tough question to answer but I've always pegged that in the $50 million to $100 million range.

Okay. Thanks. And then in your prepared commentary, you talked about seeing a pick up in the wireless base station business. Should we just think of that as year-end CapEx budget flushes and at the China carriers, or is there some other things going on like inventory replenishment or other things?

I think that's the way you should think about it today, just the end of year thing. Ian, we don't have that much visibility into beyond Q4 for base station. I think PON clearly we are seeing first half of next year also be strong as well as data center, but for the base station stuff I think it's more of a end of year maybe a Q1 thing as well.

Great. And then last question, your CapEx in the quarter, uptick here to $8.4 million. Is that a discrete event or is that more sustained going forward? Thanks.

I think it's probably -- I think is not a discrete event just for the quarter. We do expect typically we plan our CapEx to be at 5% of net revenues. I think, Ian, to finish that, we do have a whole lot of exciting new platforms that continue to come out all the time.

And sometimes some of the test and configurations that we need to bring this to marketplace is a little bit on the more complex side and we need to replace those. So the $8 million a quarter, [at a noise] is going to fluctuate from quarter to quarter, but I think if you are thinking of CapEx just think about 5% of net revenues.

Okay. Thanks, Emeka.

(Operator Instructions)

Steve Smigie, Raymond James.

Great, thanks for the follow up. Emeka, it seemed like you backed off some of your previous targets on LoRa growth here a little bit. I was just curious if I'm reading that right, or just is it a wider range or how should we think about that?

No I don't think we are backed off anything. I think we talked about having a $20 million to $30 million revenue this year, and seeing that double next year. And then the following year getting to $75 million to $100 million of revenue. So I think those are still the targets that the Company's marching towards.

Steve, we said $100 million in three years. We are making really good progress towards that.

It's, I think next year, next fiscal year, is going to be a very important year for us. We are projecting essentially a doubling of the revenue for LoRa enabled revenue, and that is going to be the key milestone. I think if we achieve that next year then getting to the $100 million I think is within the next fiscal year after that is going to probably be easily doable.

Okay. And, Mohan, can you talk a little bit about what does the dollar content on a gateway look like?

Well it depends on the gateway. The carrier grade gateways which are typically in the $500 range to $1,000 range; I think the content for us is about $20 to $30. In the Picocell gateway, it's closer to the $10.

Okay. And just for the PON, I just wanted to go into that a little bit more. Obviously it seems like it fell off. Was it a little bit softer this quarter than you expected and then, it sounds like that is coming back. I think [falling off] was more about inventory and that the overall demand pie hasn't changed very much. So that point, are we set to see you back ultimately to levels you were at before and maybe even a tad better?

Yes, Q3 wasn't a surprise. We knew it was going to be soft this quarter. And, but we know bookings are strong now, and we did anticipate by the end of the year it was going to start picking up.

And as I mentioned to Harsh, it's really the 10 gig PON that seems to be doing quite well. So this is indicative of the trends in the marketplace, higher bandwidths, higher bandwidth fiber to the home, a higher bandwidth to the enterprise. It's just going to be a continuous trend we think.

So when you look at the trends in data centers and when you look at the trends in fiber to the home and PON, and you look at trends in base station, they are all increasing in bandwidths. So they're all hitting our sweet spot.

All of the higher speed signals will need CDRs, so in some cases PON, which historically we just sold PMD type of devices in, will also require in some applications CDRs. And the same true of base station, same is true of the data center side. So we just get more content, right.

Okay. If I could just add just the -- it sounds like you got a win on the geolocation, which seems a lot sooner than I would've expected. What was that first win?

And then also in proximity sensors, you talked about the potential of getting into handsets. It sounds like it's definitely there, it sounds like it could go into Chinese handsets. So that seems like a bigger opportunity than I was thinking at one point, so I was hoping you'd talk a little bit about that too. Thanks.

So the proximity sensing, we've obviously had good success in tablets and some laptops and things like that. But we've been trying for a while to get into smartphones and wearables, and we now have success in smartphones. And so that's good news for us, obviously Q4 will be a little bit seasonally down.

But as we see next year, that should grow nicely. One of the things that's helping us there is just the regulations on manufacturers having to control radio power and human contact. So that's a good thing for us. And then your other question was -- ?

Your first win there on the geolocation.

Yes. So we've got good momentum on geolocation, and you will start to see some press releases and some news from us as we get that progress. Bear in mind, the accuracy it still needs work and we're still developing the technology. But there are some applications where the ability to not use GPS, so having a sensors that don't have GPS means that you can have a sensor that runs for five to ten years without changing a battery on the end node. And then if the asset is something that where the accuracy doesn't need to be better than 50 meter range, then I think there's many applications for that.

So those are the wins we are currently seeing. They're small today, and I don't think I want to indicate that there is going to be huge revenue. But I think it's a good indicator that the technology works and is valuable, and so as we see more progress there we will start to communicate more on it.

Okay, great. Thank you.

There are no further questions at this time. I'll now turn the call back over to the presenters.

Thank you. In closing, Semtech has now delivered four quarters of sequential performance at the upper end of our guidance, and we are forecasting a better than seasonal decline for Q4. We have focused on divesting non-core assets and diversifying our protection business while investing in exciting high-growth markets such as cloud and hyperscale data centers, optical connectivity, IoT and wireless charging which we believe positions us well to outperform the industry this fiscal year.

We have also maintained our spending discipline, which has enabled us to deliver earnings growth well in excess of our revenue growth, demonstrating the leverage in our model. With that, we appreciate your continued support of Semtech and look forward to updating you all next quarter. Thank you.

This concludes today's conference call, and you may now disconnect.