Ehud Helft - Investor Relations Shaike Orbach - Chief Executive Officer Eran Gilad - Chief Financial Officer.

Alex Henderson - Needham & Company Chip Saye - AWH Capital Edward Balinsky - Segmark International Shawn Boyd - Next Mark Capital Kevin Cassidy - Stifel, Nicolaus & Co., Inc Robert Sussman - Bentley Capital Aria Cole - Cole Capital.

Ladies and gentlemen, thank you for standing by. Welcome to the Silicom’s Third Quarter 2017 Results Conference Call. All participants are present in listen-only. Following management’s formal presentation instructions will be given for the question-and-answer session. As a reminder, this conference is being recorded.

You should have all received by now the Company’s press release. If you have not received it, please contact Silicom’s Investor Relations team at GK Investor and Public Relations at 1646-688-3559 or view it in the news section of the Company’s website, www.silicom-usa.com. I would now like to handover the call to Mr.

Ehud Helft of GK Investor Relations. Mr. Helft, would you like to begin, please..

Thank you, operator. I would like to welcome all of you Silicom's third quarter 2017 results conference call. Before we start, I would like to draw your attention to the following Safe Harbor statement. This conference call contains projections or other forward-looking statements regarding future events or the future performance of the Company.

These statements are only predictions and may change as time passes. Silicom does not assume any obligation to update this information.

Actual events or results may differ materially from those projected, including as a result of our including dependency for substantial revenue growth on a limited number of customers in the evolving Cloud-based market, the speed and extent to which Cloud-based and Cloud-focused solutions are adopted by the market, the likelihood that it will rely increasingly on customers which provide Cloud-based and Cloud-focused solutions in this evolving market, resulting in an increasing dependence on a smaller number of larger customers, difficulty in commercializing and marketing of Silicom's products and services, maintaining and protecting brand recognition, protection of intellectual property, competition and other factors identified in the documents filed in the Company's with the SEC.

In addition, following the Company’s disclosure of certain non-GAAP financial measures in today’s earnings release, such non-GAAP and financial measures will be discussed during this call. Such non-GAAP measures are used by management to make strategic decisions, focus future results and evaluate the Company’s current performance.

Management believes that the presentation of these non-GAAP financial measures is useful to investor understanding and assessment of the Company’s ongoing corporation and prospect for the future. Unless otherwise stated, it should be assumed that financial discussed in this conference call will be on a non-GAAP basis.

Non-GAAP financial measures disclosed by management are provided as additional information to investors in order to provide them with an alternative method for assessing our financial conditions and operating results. These measures are not in accordance with or a substitute for GAAP.

A full reconciliation of non-GAAP to GAAP financial measures is included in today’s earnings release, which you can find on Silicom’s website. With us today on the call are Mr. Shaike Orbach, CEO; and Mr. Eran Gilad, the CFO. Shaike will begin with an overview of the results, followed by Eran who will provide the analysis of the financials.

We will then turn over the call to the question-and-answer session. And with that, I would like now to hand over the call to Shaike. Shaike, please..

Thank you, Ehud. Good morning everyone and welcome to our conference call to discuss our third quarter results of 2017. We are very happy with our results, announcing our highest ever level of revenue for both the quarter and a nine months period ended September 30.

We also announced solid results across the board with strong operating and net profit as well as cash flow. But further than that we have recently demonstrated some very significant strategic progress.

As we move towards the end of 2017 Silicom is in by far the best position it has ever been in and we are very excited as we increasingly realize and further unleash our potential.

From a financial perspective in the third quarter revenues grew by 31% year-over-year to a record of $32.3 million with a net income which grew even more aggressively by 42% to $5.7 million. Eran will provide detailed financial results in a few minutes.

From a strategic perspective, our design win momentum continues and beyond that it covers those technologies in market segments which are required to address the main Cloud related opportunities, legacy Cloud, next-generation Cloud and the Future Cloud. This is best demonstrated by our latest three announcements.

We announced an FPGA based design win with one of our significant long-term customers a networking world leader playing in some of the highest growth segments of the space.

We also won a significant approximately $6 million per year Cloud design win with one of our advanced 10 gigabit per second adapters within a different group of the same giant Cloud customer which awarded us with our largest ever design win back in March.

And then just last week, we announced the additional Tier-1 manufacturer and increased volumes within our largest ever win that being with our Switch-Fabric-on-a-NIC which will be used in our customers next-generation Cloud. These announcements demonstrate how well our portfolio covers the Cloud.

We have broad solutions across the full spectrum, from legacy Cloud where our 10 gigabit per second adapter are used to next generation Cloud with our Switch-Fabric-on-a-NIC and finally FPGA based solutions for the future Cloud.

Our design win momentum confirms the markets demand for our products and solutions in the Cloud and as such demonstrate that the ongoing industry transition to the Cloud continues to represent a fast growing opportunity for us.

Our recent successes demonstrate that we remain perfectly positioned at the epicenter supporting Cloud and data center technology. Our past investments have been clearly executed into the right areas and it is clear that we maintain a continued ability to correctly anticipate and position ourselves for upcoming and future market trends.

Beyond the very impressive success we have already had in 2017 our pipeline represents very significant revenue potential for us and we are working hard to convert those opportunities into design wins.

We are making progress towards bringing additional design wins from additional Cloud players, telecom, cable companies, SD-WAN companies and other service providers for a variety of products including all types of adapters and Edge devices.

These are regular adapters, Switch-on-a-NIC-Fabric adapters and smart FPGA adapters as well as vCPE and other Edge devices. The potential from many of the opportunities in our pipeline each individual represent significant step up in revenue for Silicom.

More broadly in order to take advantage of the very much increased revenue potential and opportunities we anticipate from all these major players into various market segments in the coming quarters and years we are adjusting our margins accordingly and consequently we anticipate that moving forward our gross margins will then be into the 32% to 36% range.

Bear in mind that the sacrifice in gross margin is a strategic decision that we have chosen to take based on our long track record of generating high and impressive operating margins despite a relatively low gross margin.

Looking at the content of the product that we are planning to sale to the Cloud and its related markets and estimating the revenue that we will be able to generate at somewhat reduced gross margins we fell that being able to propose such lower margins, while maintaining a leverage which significantly increase our operational net profit is actually one of our most important differentiators.

We believe that with such leverage inherent to our business as revenue continue to grow we will be more than compensated by strength on the bottom line as well, that is we expect to see high levels of growth in our net profit and earnings per share.

I would like to now focus specifically on the status and the potential from our largest ever design win with a giant top-tier Cloud player which we announced in March.

In the next few minutes I would like to describe to you both the expanding potential that is now represented by that win as well as the challenges that are still in front of us before we reach a steady state which would enable the full potential.

Indeed as we move forward, it is becoming clear that the potential from this win is significantly greater than we originally had expected.

Last week we announced that based on our customers’ estimates we expect the substantial increase in our revenue from this design win, more than doubling our original expectations from $30 million to an annual rate of $75 million once we reach the full deployment run rate in 2019.

This increase is a result of both increases in our content-per-server as well as increase in overall quantities of servers which are expected to be installed. From the content side, the solution that we provide into every individual server is already expanding. We expect such growth to continue and encompass a significant part of our deliveries.

Obviously this means that you are increasing the future dollar value of our sale into the customer's infrastructure. From the quantity side, the customers' latest estimates indeed indicate a significant growth in quantities over what we predicted earlier.

These higher quantities are also demonstrated by the fact that our customer is now qualifying an additional Tier-1 server manufacturer as second subcontractor for building its Cloud infrastructure, understanding that such huge quantities require multiple sources.

And just as to further demonstrate how huge this design win is while integrating our product into their solution as a part of the qualification process these second server manufacturer is preparing for the demand which it expects and has placed purchase orders with us already totaling to around $10 million.

The fact that our Cloud customer has decided to source its server base solutions from two top-tier server suppliers is a demonstration over the high volume of Cloud infrastructure solutions that it expects to deploy.

I would like to note that given these huge quantities the customers policy of dual sourcing its critical components as demonstrated by bringing in a second server provider will most probably apply to the switch-a-NIC as well.

Our understanding from discussions with the customers is that this is in process and will probably be introduced on the same timeframe at which general availability is planned for or somewhat later.

Our expectation of order volumes that we have shared with you already take into consideration that there will also be a parallel second source supplier of Switch-Fabric-on-a-NIC product to this customer as well as us.

As we have discussed several times since we announced this design win there are challenges ahead which remain, as I'm often asked about those challenges, I would like to spend a few minutes providing more color. Those challenges are divided into two broad categories, customer challenges and deployment challenges.

There are customer challenges because there is no general availability yet to end users. Our customers still need to finalize various aspects of this new Cloud infrastructure. It also needs to build sufficient infrastructure to be ready for initial user demand before rolling out this improved service throughout the full user base.

Our customer believes the risk of not reaching the general availability milestone is negligent to or non-existent. We understand that what remains to finalize is just standard engineering work and debugging rather than any issues which threatens visibility.

However, only when all such debugging is complete the infrastructure is fully functional meaning general availability is therefore reached and the user response is validated the actual level of demand will be known fast.

While our customer may experience delays in meeting the GA date, I think does not yet have a full handle on the exact demands that will be placed on the infrastructure. The expectations of the customer which are based on its current Cloud deployments are huge.

in fact the customer is describing potential scenarios which reach even our current estimations are conservative. The other challenge is the deployment challenge; this is the combination of issues related to engineering, logistics, issues with components, testing, system integration and implement, production ramp up and such.

These challenges are a result of a complex product which was designed and then moved into production in record time; sometimes before it was fully mature. I would like to note that our customer as well as the two Tier-1 server and infrastructure manufacturers that we work with understand that such challenges are a part of an extremely rapid cycle.

Furthermore, here as well the solutions to these challenges are well known by us and we do not have any doubt that all the remaining issues will be result. However, as with all complex projects the above mentioned customer and deployment challenges may cause unpredictable push outs in deliveries and fluctuations in revenues on a quarterly basis.

So, far we have been meeting and surmounting the challenges in line with our plans, general availability is expected in the middle of next year. We are cautious on the timeline but definitely pleased that things are moving ahead well.

In summary, while this design win is significant our Cloud journey has only just begun and it has the power to drive our business to a whole new level.

The recognition that we continue to receive on the market for our technologies and solutions coupled with a strong demand within fast growing factors such as Cloud is empowering us to build our business to a new level.

Looking at our business from a broader perspective, while Cloud is indeed a very strong growth engine for us, Silicom is much more than Cloud. We continue to invest in and convert a strong and growing pipeline into design wins in some of the hottest networking segments including cyber security, SD-WAN and NSV as well the Cloud.

All these markets depend on flawless connectivity and our product innovations are tweaked towards exactly meeting our customers need. We have cemented ourselves and are increasingly been seen by many existing and potential customers in our target industries as a connectivity partner of choice.

The success we have experienced in recent years demonstrates our ability in projecting, correctly investing in and ultimately capitalizing on the upcoming market trends. Our design wins represent the ongoing fruit of significant plus R&D investment made at the right time.

They are also a clear indication of the superiority of our overall technical and solution capabilities. They are very much aligned with the critical performance needs of both existing and new potential top-tier customers addressing all the high growth markets which we target.

Before handing over to Eran, I want to address our expectations for the fourth quarter. Looking ahead, we believe that fourth quarter revenues will be in the range of $36 million and $37 million. At the midpoint, it represents growth of 29% over the fourth quarter of last year. In summary, we are increasingly excited with regard to our prospects ahead.

We believe that we are better positioned than ever to achieve strong and continued growth in the quarters to come both top-line and bottom line.

With Cloud tailwinds driving demand for our products, additional expected growth in cyber security SD-WAN and other related markets, we believe that our business is now at a financial inflection point in our growth trajectory with a potential to significantly accelerate ahead from this point ahead.

We look forward to reporting on our continued progress over the coming quarters. With that, I will now hand over the call to Eran for a detailed review of the quarter's results. Eran please go ahead..

Thank you Shaike and hello everyone. Revenue for the third quarter 2017 were $32.3 million a growth of 31% compared with revenues of $24.7 million reported in third quarter of last year. Our geographical revenue breakdown over the last 12 months were as follows, North America 73%, Europe and Israel 20%, Far East and rest of the world 7%.

I will be presenting the rest of the financial results on a non-GAAP basis which excludes the non-cash compensation expenses in respect of options and RSUs granted to directors, officers and employees and acquisition related adjustments. For the full conciliation from GAAP to non-GAAP numbers please refer to the press release we issued earlier today.

Gross profit for the third quarter of 2017 was $11.8 million representing a gross margin of 36.4%. This is compared with $9.7 million or gross margin of 39.3% in the third quarter of last year. As we have mentioned in previous conference calls, our gross margin is largely affected by the specific mix of products sold in the period.

However, as Shaike mentioned, looking ahead and given the very strong growth opportunities in our end markets the expected gross margin range will lower from the current 35% to 40% range to 32% to 36% in 2018.

Operating expenses in the third quarter of 2017 were $5.2 million or 16.1% of revenues compared with $5 million or 20.2% of revenues in the third quarter of last year. Operating income for the third quarter of 2017 was $6.6 million, a growth of 40% compared to $4.7 million reported in the third quarter of 2016.

Net income for the quarter was $5.7 million or 17.7% of revenues compared to $4 million or 16.1% of revenues in the third quarter of last year. Earnings per diluted share in the quarter were $0.75, a growth of 42% compared with $0.63 in the third quarter of last year.

Now turning to the balance sheet, as of September 30, 2017 the Company's cash, cash equivalents and marketable securities totaled $36.9 million with no debt. That ends my summary and we would be happy to take any questions. Operator..

Thank you. [Operator Instructions] The first question is from Alex Henderson of Needham and Company. Please go ahead..

Hello guys, Lots of stuff to parse here, I just was hoping we could just split a couple of these into some of these pieces.

So it sounds like on the one hand you are getting substantial increase in demand for the NIC-on-a-Switch card and that has obviously lower margins than your traditional business, which has historically been up in the 40% vicinity.

But it also sounds like you are adding more content to that deployment on a per server basis and that that also has an impact on gross margin.

Can you give us a little bit more clarity around that mix of business within each per server deployment?.

Okay, I will try to do that but even before I do that I would like to mention that the issue of gross margin is not only related to this big win. I mean we have discussed that internally and we have come to the conclusion that we can achieve even more.

I would say - well maybe not that big because that one is really, extremely, extremely big, but very big wins totaling tens of million dollars per year if we reduce our gross margin and obviously calculate accordingly the right price, because we can do that.

we know how to do that, we know how to manage that and that really can become, really big advantage for us over some competitors even in other areas that you are competing. That is also the case here.

That is also indicated within this year, because I would say that we have been in a certain dialogue with our customers and its subcontractors or infrastructure builders for quite some time, as to being able to do more within our solutions because as you know I mean our solutions when we started its certain building block, this building blocks are connected to other building blocks.

Now some of these other building blocks would be even I would say more naturally become a part of our offering; however, I mean because originally they may have been addressed in a different way by our customer and the infrastructure builders and they would like us to do that because once we do that they have a more integrated product which includes all the functionality inside and its makes more sense.

However, they wouldn't do that for the margins that we were originally quoting to them and yes indeed because of that we looked at that contest.

the fact is that in terms our actual labor and complexity remains the same because with the kind of service that we are providing our customers we handle all the difficulties resulting from this additional content anyhow.

So, we have discussed it internally and we came up to the conclusion led by agreeing I would say, not even agreeing, actually promoting even somewhat lower margins because the customer in general is aware with our margins especially for that part, we could increase our contest, increase our bottom line and our EPS and everyone is happy and this is what we did in this case.

Now that…..

Just to be clear here, you are saying that on your base business exclusive of the NIC card product that you are doing for the Cloud customer that's talking about the broader framework excluding that right?.

No, at the beginning, I talked about additional opportunities other than the Switch-Fabric-on-a-NIC. Now I'm talking specifically about Switch-Fabric-on-a-NIC, because you wanted to have more color on that, so I'm giving you that.

And I'm saying that we are getting agreements and understandings with our customer and its infrastructure builders that indeed we will increase the content of what we are delivering to them.

But the overall margin for the full solution would be lower which is something that would probably would start happening now, would be more significant towards the second half of 2018, but moving forward it would increase our content as well..

So, when I look at a per card basis what is the percent increase in the value of the revenues associated with that card as a result of those additional features being added to the Switch-Fabric-on-a-NIC comp?.

There is not just one number in it. Because there are - the card it's going to come in several configurations and it's going to begin I would say with some which would be relatively modest increase but it could come up to being quite significant..

So, 10% to 30% or 10% to 50% I mean what is kind of range on it?.

Well once again, I would say that it begins from 10% but can come up to 50% and sometimes even more..

I see okay.

So, when I look at that incremental volume how delta is the cost or the margin on that, so when you are talking about getting down towards potentially a 32% type gross margin is that a function of the significant increase in the added function or added capabilities on the NIC that are being added which have much lower than the NIC cards margin, is that essentially what I'm hearing..

That's definitely a part of that..

Okay, and then so as we look at the timing of the development of that added functionality, I assume that that feathers in over the time across 2018, so it sounds like with the mix shift to the switch-on-a-NIC card in the first half you see some pressure on gross margins, but as we go through the year 2018 it sounds like you are going to see a lot more of that additional functionality and the impact of some of your design on the core business so should we be thinking about getting down towards the lower part of that gross margin guide as we go out of 2018 and into 2019..

I'm not sure about really being in the lower part, because you need to understand that after all this is not the only part of Silicom, we have other parts of Silicom and therefore I'm not sure about that, but you are right that the impact of this specific issue would be increased as we move towards the second half of 2018..

So is it reasonable to tie an acceleration of the revenues to the deceleration in margins..

In general it's correct, but it not necessarily would be I would say totally linear. There could be deviations here and there and there could be sometimes I mean that we still have POs without the content or with the content and there could be some other design wins.

But looking at the full picture, I would say that moving towards it, yes I mean it's going to move towards the lower end as you mentioned before in the second half of 2018. And going into 2019 I presume it will be there because by then I would assume that we would be selling indeed much more content at the lower gross margin.

Assuming that we are indeed at the 75 million with this specific win, I would estimate that indeed we are around 32% maybe a little higher in 2019..

So, just so that I understand the mechanics of this it sounds like you have got some upfront startup costs in 2018, you are gross margins are coming down, you are OpEx has got to [traject] (Ph) along fairly stable line but it does sound like that could cause your operating margin to cross through that 20% kind of threshold that had been the floor, perceived as the floor maybe temporarily in 2018, but would you expect them to get back 20% type operating margins in 2019?.

Well on that, I didn't make the actual calculation, but in general my feeling is yes.

I mean you need to remember that as we grow with revenues and with the leverage that we have over the operating expenses and with the operating expenses not growing at the rate of the revenues I would assume that this would more or less compensate the reduction in the gross margin allowing us to be in the same area of approximately 20% operating income in 2019..

Alright. So might crack that as you are starting up but ultimately 2019 it’s back into the 20s, can you talk a little bit about the tax line just to try to make sure on the park, we are talking about around 15% on the tax line still..

Tax, okay Eran please..

Yes, quarter three was approximately 14.2%, but in general I can say we maintained original range of approximately 15%..

And then what was the headcount?.

Headcount at the end of the quarter was approximately 260 employees..

And 10% customers..

During the last 12 months we had two 10% plus customers, but the first one is approximately 20% and the second one is approximately 13%..

So does that change during third quarter or is that - or are we just rolling 12 months at this point?.

This is the total of the last 12 months..

We are doing 12 months I believe..

It's not a quarterly number it’s a 12 month..

Right last 12 months..

And then the other question I wanted to ask, when I'm looking at the ramp here the December quarter, historically you have had a pretty significant decline from 4Q into 1Q.

Given the Cloud trajectory does that have a different feel to it on a seasonal basis or should I assume the normal seasonal pattern is held by both your existing customers as well as the new Cloud customers?.

I think that in that case I'm not necessarily saying that this would be the trajectory in coming years of this Cloud win. But I think that looking forward and at what I see right now and Q1 is relatively close, you can assume something very similar right now.

Obviously it would still represent a significant that I believe in general I don’t have the figures right now, but it would still - what I'm thinking about a significant year-over-year growth, but indeed compared to Q4 you would be seeing a decline I presume..

One more question if I could. Can you just update us on the SD-WAN stuff, we haven’t heard much about that lately and it’s a really important piece..

Well okay, Several things, first I would like to draw your attention to the fact that just about three weeks ago it was something like that Intel had a webinar which was discussing SD-WAN and the example that they brought up there as to I would say representing and successful solutions SD-WAN is Silicom's products.

So this was obviously a nice point for us. on top of that I can tell you that we are a - significant part of the pipeline that I was talking about is with SD-WAN. So right now I don’t have any design win that I can tell you.

Yes we have a new design win right now, but we are moving forward with SD-WAN with quite a few customers and I would say definitely making progress.

While sales cycles are not short and some of these guys are very big companies, some of these wins are going into tenders or bids by major telecoms, some of these are $20 million plus per year type of tenders which usually takes some more time. But we are very active on that front.

When exactly that would happen? I don’t know, but this is definitely a major part of our activity these days..

Just to be clear is the SD-WAN also somewhat lower gross margin product as a result of the fact that you are putting stuff in branch obviously that’s going to be a little lower?.

Yes, it's a result of that and it's a result of the strategy that we have discussed before, we believe we can get more by reducing our margins overall into the area that we described before, so it’s part of the strategy, when we continue this strategy it was not only this big win that we are talking about definitely the SD-WAN issue was the part of this strategic discussion..

Okay, I will leave the floor. Thanks..

Thanks..

The next question is from Chip Saye of AWH Capital. Please go ahead..

Yes, thanks for taking my questions.

First on the big Cloud win that was announced in March and then the follow-up announcement recently, are you ever going to identify the customer there?.

That's clearly the question which should go to the customer I believe. I mean we do not have the right to identify the customer..

Okay, I was just checking on that.

Secondly that big Cloud win that we are talking about are all of those products designed on the Intel architecture, you mentioned Intel a second ago?.

Well first of all for this, what we call largest ever win, we are talking about a single product with some variations and indeed the variations are not variations within I would say the Intel part of it. The short answer to your question would be yes, there are all around Intel architecture, if you want me to give you some more color on that I can.

But if that concludes what you wanted me to say, that's okay as well, because IA or Intel Architecture we are doing a networking Cloud which includes some parts of IA of Intel Architecture on the Cloud itself, but obviously it's a part of a bigger server which is also IA which is not ours, so hope that clarify the situation..

Okay, yes that helped and if you don't mind could you continue on a little more because I would be interested, for example the March - the design win you mentioned that was probably based on the Broadwell chipset because the [Parlier] (Ph), Skylake wasn't out, is this a new announcement, is that based on the new chipset or is that still based on the Broadwell chipset?.

So, it seems like I do need to go a little bit more into an explanation of the product. Because its neither of these two because this product, the product that we are selling is not a CPU or a processor product, the CPU/processor is in the server into which we are selling.

So that's I mean these discussions about Broadwell or Skylake or whatever these are questions, which should go to those Tier-1 server manufacturers that we discussed before, which are buying from us our connectivity or networking solutions, put it into their server which I know to be a part of Intel Architecture, but I'm not sure I think its Broadwell series rather than Skylake, but maybe I'm wrong about that.

But in any event our card goes into their, it goes into a PCI Express slot inside their servers providing connectivity of six port, 100-gig, a Switch-on-the-NIC that's why we call it a Switch-Fabric-on-a-NIC, and that's what I was thinking you were referring, the switch that we have on this board has the data path connected to the server which is a Broadwell or a Skylake, but on the card itself we also have an Atom which is for a control of the switch.

So, the switch itself is an Intel Atom 10k which is a switch and a NIC and an Atom that’s what that’s what included and of course optics and other components. That's what is included on the card itself the question as to Broadwell versus Skylake is a question which is relevant for the server which is not what we built..

Right, but I'm saying the server, the CPU like you mentioned the processing the core that's probably still based on working with Broadwell not based on working with Parlier, but that’s probably….

That's what I think, but our card as a card because it goes into a standard PCI express slot would work with both..

Okay, I just didn't know if you have had a chance to work with the new chipset, work with server companies, server manufacturers that are working with the new chipset yet, have you done anything..

Well I can tell you that we are working together with Intel on a reference design which is something that we are doing for Intel and that's around [Sky VE] (Ph) which is the Broadwell DE or the Xeon D version now replaced by Sky VE so this is an area where we work with the Skylake architecture even though it’s the Sky VE which is the version of the Skylake which includes the networking port in the chip itself, in the chipset..

Okay and next question you mentioned the Cloud, the next generation Cloud and future Cloud and when you mentioned future Cloud you tend to talk about FPGA a lot and then there is the design win, the FPGA design win that you referenced, can you talk about what product that is that incorporates that..

Okay, well first of all just to clarify I mean the design win that we had with FPGA is a demonstration of the technology, it's not a design win which is the I would say accurate architecture that would be used in the future Cloud.

This design win that we were talking about is not a Cloud win, it's a different type of a win still the technologies and the IP cores that we are using there are relevant for the Cloud, but right now most of the Clouds not talking about Amazon and Microsoft Clouds, Amazon I think it's public knowledge that they FPGA-as-a-service and Azure is also well known that they are using FPGA.

But other than that I mean this is for the future, so the design win was just a demonstration of the technology we are well prepared and we are having discussions with FPGA solutions for the Cloud but the design win that we had is not a Cloud design win..

Okay and I guess the future would be incorporating FVGA in your NIC, or switch-on-NIC is that right..

That's suppose to be a part of that because the next generation we are not the only FPGA, we may use different optics, but yes definitely I mean the FPGA I mean the current FM10K is still limited to a certain extent with the number of rules et cetera of what it can process, our thinking process is that indeed this FM10K will be replaced by an FPGA..

Okay and that will be a situation where you have an ASIC that's $4 million, $5 million dollars and then you have an FPGA so you can offload - use the core for the main computing needs and then do whatever your special application or need is with FPGA with your server adapter or what..

Something like that but there is one nuance in there.

I mean we expect that - I mean some parts of the IP or the offloading as you mentioned would be something that we do, but in our architecture of FPGA is we also have some what I would say some sort of a sandbox or closed area that maybe our customers would like to use or potentially even our customers' customer would like to use just like whatever Amazon is proposing right now..

Well it’s enough. Thank you very much for the answers and its very exciting stuff. Thank you..

Thank you..

The next question is from Edward Balinsky of Segmark. Please go ahead..

Hi, I'm interested in the competition of the contingent consideration. As I understand that the requirement to pay contingent consideration ends at the end of the next quarter.

Is that correct first of all?.

Yes, that is correct..

I'm not sure how you arrive at your figures, they seem to jump all around from a high of 5.54000 to a low of 4,642. Are those numbers which are derived by a formula or do you calculate directly what is owed to the customer? Towards the support of the personal supplied the equipment..

First of all, the number during the last few quarters was more or less the same. There are no fluctuations. And counted with a $4.6 million and currently this is $4.8 million and it is connected also with making it present value. So actually this is the same number.

And as you said the next quarter will be the last quarter in which we will have the contingent consideration..

I appreciate you. Thank you then, that takes care of my question..

Thank you..

The next question from Kevin Cassidy of Stifel. Please go ahead..

Thanks for taking my question and congratulations on the great result. I wanted to understand a little more too about the FPGA card.

You have a design win that was just a proof of the technology, can you say what kind of pipeline of interest you have with all different industries but particularly the Cloud?.

Yes, I mean there are certain things, I cannot be very specific about these opportunities, but I'm just selecting by words, in order to make sure that I'm responding to you with information that I can't really share.

But I would like to tell you that we have several fronts where we are dealing and having discussions both with partners as well as customers which could be very significant in cooperating our FPGA solutions.

I would like to tell you for example that we are going to release pretty soon a card, with FPGA of course, which would include a mechanism which would shorten the time to market to go with such FPGA card, because typically FPGA cards are relatively complicated for integration.

I would also tell you that we have discussions and cooperation discussions with I would say almost every side that you can touch upon with respect to the FPGAs.

Some of the major Cloud companies in the world are talking to us about FPGAs, so it's a lot, but I cannot be very specific right now, but the one thing that I would like still to say is that we believe that's in order to get wins with FPGA it will take time, it's not going to happen soon, and that's because of several reasons.

First of all, the technology needs to be completely developed, it's not there yet. Secondly the evaluation period and proof of concept et cetera are lengthy and third, the integration of these products into the Cloud in the architecture is again something which is relatively complex and take time.

So, it will take time but I think that it's going to be extremely significant for Silicom once it happens, I don't think anything significant is going to happen this year and probably not even in the second half of next year with any of these significant wins.

I'm not talking about small wins which could happen any day, but I'm talking about a significant Cloud companies committing to solutions which are based on FPGAs. So, I know that I didn't give you too much color, but I can't give you any more..

No that’s a great color.

Just to get a reference of when to expect some inflection point in revenue, and maybe it's too early also to know what the effect would be on gross margin?.

Well okay, as to what the accurate effect would be I cannot say, but we definitely believe that with FPGAs, the gross margin is going to be better for us even when we are going for the Cloud, and that’s because within the FPGA its not only hardware rather IP and IP course, and that has a price tag to it, even if it comes at the part of supporting the customer around these IP.

So we believe it's going to be better, but I would like to tell you something that I think no one knows as of yet, because in terms networking this may not be the case in other market segments but in terms of networking until now FPGAs has not been deployed in these quantities that are now expected to be deployed in the Cloud.

I mean the current FPGA solutions that we are selling they have a much higher gross margin then even - I'm not even talking about the Cloud cards, but rather our so called traditional products.

We are talking there about much more higher margins at 40%, but the quantities on the other side are not that high, when we are going to go into tens of thousands of FPGA based cards we believe that the margins will need to be lower than they are today. Where exactly they will end up? I'm not sure.

I'm sure that it will be higher than the I would say non-FPGA Cloud based products..

Okay great that’s helpful.

And maybe just for your revenue for the third quarter do you break that out by end markets or actual cards that drove the revenue?.

We have some analysis related to that. Eran do you have that? I mean I'm not sure it's quarterly maybe it's a nine month or whatever..

Yes, we have such an analysis that can be seen on our quarterly for the picture, which will be uploaded to the website in a few hours. Generally speaking over the last 12 months the percentage of revenues to service providers or cloud players was about 36% for enterprise is about 47%, for government 13% and the rest is many others 4%..

And that is just to be clear that's an estimation because some of that just like our largest ever win we sell directly to cloud guys and that we know but there is a part of that which we sell indirectly into the cloud through our OEM customers. So that includes both of these 36% that Eran mentioned includes both of these..

Okay, great thanks congratulations again..

The next question is from Robert Sussman of Bentley Capital, please go ahead..

Thank you, could you talk about what the challenges will be in scaling up your business, listening to what you have in house and design wins and the quantity of potential nuance and those volumes, you could easily at some point next year be going at over $50 million quarterly rate.

Can you tell us what the challenges will be in scaling up and your ability to do it?.

Okay well first of all, there is definitely a challenge that we have I would say began to encounter even with that largest ever win when we started with that and I would say that the major decision that we have taken is to shift more and more I would say, that's a gradual process, but that's happening towards I would say contract manufacturers allowing the scalability in manufacturing as well.

Of course I'm not concerned about the scalability and whatever needs to be done with our R&D and support and the scalability issue there would be mostly due to the huge quantities we would need to be able to scale up in manufacturing and operations and that's why we are now working with one of the world's giants in terms of manufacturing and we intend to go into that direction.

Because I'm not sure whether you are aware of that or not but traditionally we have not been using turnkey manufacturers but rather we were making all the preparations and just sub contracting the SMPs into some sub contracting taking back everything to us et cetera.

We are moving more and more towards the turnkey as I mentioned before it’s a gradual process but that's our plan to address scalability..

So assuming more significant design wins which between now and sometime mid next year sound highly probable would you be able do you think you will be able to handle $50 million plus in quarterly revenues..

Well I don't think that we're going to be there in mid next year first of all, but there is one thing I would like to tell you and that is that I think that this largest win that we are talking about that was and forgive my language that was a crazy project and that means that the ramp up rate was such that I don't think it's going to happen with other wins.

So what I'm saying is that either - and we have the plans on the bids in the other pipeline opportunities that we're having we looking at that.

We have the plans clear in front of us, it's not just like it happened in this project that I would say, I mean of course I'm exaggerating but just explaining one day after the award we need to deliver tens of thousands of cards that’s not going to happen with the others.

It's going to be in much more gradual process and it would give us time to get ready for that. So we know that’s the way that it's going to happen. It's not going to be a that kind of a step function that we had with this largest ever win.

And I said, we have a general plan as to how to address this scalability issue which is mostly by working more and more with turnkey contract manufacturers and I feel confident, I'm not saying there wouldn’t be any challenges. Whenever you get a new huge project there are challenges.

But I think we are thinking ahead about these things and we are ready..

Last question.

Do you see any areas of the Company where you need to hire a significant number of people either research and development, working with customers on software or any other areas like that?.

Well significant is something that needs to be defined.

So I don’t think it’s going to really be significant in terms of our OpEx, but in general, generally speaking yes, we are hiring more people related to FPGA, and when I'm talking about FPGA it means both FPGA guys developing or familiar with IP course which goes into FPGA software related to FPGAs and hardware related FPGAs..

Okay. Thank you..

The next question is from Aria Cole of Cole Management. Please go ahead..

Good morning and thank you for doing the call and best of luck to the coming year. A question related to your number one customer in the data center markets.

Just to ask a naïve question I just want to double checks that they are not an exclusive customer so that the products they are going to be shipping to them can be offered and sold to other data center potential customers as well.

And then the second question is given the performance and improvement and value, you are delivering to the new data center customer that you are hoping to achieve and be offered on the GA basis next year, how are the potential or other data center customers in industry responding.

Because clearly potentially this next generation data center architecture is something that need to look at very seriously I would assume.

So how are they responding in terms of just waiting and being patient or knocking on the door with you and saying we need to talk because if there is going to be improvement in performance and data center as we want to be there quickly..

Okay. So first of all yes, we can sell our cards and mostly I would say because whatever we sell right now is very specific, but versions of that we can definitely sell to other customers. But there is one challenge that I would like to explain and also explain what is coming with it.

The current card that we are selling to this big customer is related to the architecture that this customer develop and the architecture that this customer develop is a unique architecture in terms of the software.

And that means that someone to be able to use the card yes it is, right now it needs to have the same software in order to be able to get the same performance. So there are many customers who wouldn’t have this kind of software.

So they would need to decide as to whether or not they are investing in that or they are waiting for someone else to develop that for third-parties and they are by the way some potential third-parties that we are talking to, to be able to propose this solution together with these third-parties. So this is a possibility as well.

So there is a challenge with respect to the software to as part of it, now what do I mean by to a part of it that I would say that the card that we are delivering through a wide performance advantageous in two areas, one area is in the basic architecture, which is indeed where our current customer has an advantage over other Cloud players and the other area I would say in general is the offloading area.

So in terms of offloading we have solutions ourselves that we are proposing to other customers and there is definitely an interest, so we are talking to several customers about that. However.

that being said and taken into consideration the timeline, some customers are saying okay we would like to take what you are doing right now and use it with an FPGA in order to get some more offloading capabilities because they will be focusing on the offloading rather than offloading and architecture.

So I'm saying there is a potential there for a card as it is, we are trying to sell it to other customers.

I would say that it would take time as well, because of the need to have the architecture ready in order to incorporate these cards as to current solutions and once the customer doesn't have that it would need some time for this customer to build that architecture for that and if it is not close to that point in time yet they may look at FPGA as well.

So, the closer the customer is to getting to a similar architecture will be able to sell and we have some relatively close customers who are interested in that. If the customer is not there yet, would need more time, then he would be looking at FPGA because in that case it would take more time..

Okay. Thank you..

We have a follow-up question from Alex Henderson from Needham and Company. Please go ahead..

I wanted to talk about the second sourcing commentary that you had mentioned relative to your Cloud customer. I assume that they would bring in and I can think a number of companies, Company like a Mellanox or something like that, that's already in the NIC card space.

On the other hand, don't know of many companies that have a gross margin structure that looks anything like yours.

So I was wondering if one you could talk a little bit about how much of an advantage you have in terms of time to market as a result of working directly with Intel who is obviously the prime supplier to a lot of these Cloud companies and two to what extent you would see yourself as a prime supplier with a smaller second supplier in a competitive environment, because of your ability to deliver product at lower margin.

How should we think about the sustainability of this Cloud business overtime as your strategy is obviously quite differentiated from most other people?.

I would saying in general that I believe that we would have certain advantage. I don't know exactly how big this advantage is going to be. On the other side, I would say that the numbers that we shared with you are based on the assumption that we would get 50% of the expected quantities..

And relative to the timeline around….

Oh I mean we have definitely - we are there. The second source card if everything goes well, I mean best case scenario is supposed to be mid-year more or less around the time of GA from what we are hearing but these are rumors so I cannot be sure about that.

Is that there are delays and probably would not happen before the end of Q3, but I cannot be 100% sure about that, these are rumors that we are hearing.

Needs to understand that this is quite a complicated card, it's not I would say an invention but in terms of depth of technology which is needed to be able to build this card with perfect signal integrity and so on and so forth it’s a complicated card. It’s not an easy card to build..

Going back to the original question, how much of an advantage are you delivering as a result of your tight relationship with Intel..

Well okay in that case I will respond to another part of what you said before. I think we have a very big advantage due to the relationship that we have in Intel, but I would like to say that I mean any second source card here would have to use an Intel, Silicom as well.

It's not going to be anything other than the FM10K it must me FM10K, because all the software structure of the customer is around FM10K, so the second source would build the card with FM10K as well..

I see and relative to the cost structure, obviously you guys are very differentiated on the business model, being able to deliver you know at these type of gross margins.

Do you see other companies able to do that structure and still sustain profitability or do you think it’s going to be hard for people to match your performance and be profitable and then similarly along the same lines if that other customer as you just mentioned has difficulty delivering a second source will the Cloud customer continue to go ahead with the project with a sole source until they can get their architecture and their cost structure and their product performance to where it needs to be..

Well so first of all I do believe we will have an advantage in pricing, but I do believe again that's what they are telling me, it’s not something that they are telling me as a commitment of course I'm not sure whether or not there was an agreement there or not.

But our understanding is first of all they would go with us even if we were the only source for that. I believe that we would have an advantage in pricing, but I think that having a second source, I mean they would continue with a second source even if the second source is more expensive.

Just to give you an example, I’m not sure at all that what they are getting from both of their Tier-1 suppliers for the servers at this point is at the same price, not sure about that at all. So I think we are going to have an advantage obviously in time to market because we are there the second source is not there as well.

I think we would be better positioned to get bigger quantities, but I'm not sure exactly what they will do. As I have said I mean our estimations are based on 50% only. By the way just to have you know, it is not something new for our customer to discuss this second source.

As a matter of fact when we started the development of this card there was another Company which was supposed to be the second source. Only they failed, they simply were not able to deliver a working product which is why our customer eliminated them.

So one company already failed, we were successful now they are working with another company that maybe able to deliver, I'm not saying they will not be able to deliver but this is again this is a complex product not a simple one and we definitely have an advantage. However, we would like to be conservative, but we wanted to give you the full picture.

I mean the plan is to have a second source and to be on the safe side, we just in our estimations we took 50% of the quantities and calculated based on that..

So just to be clear to the extent that you are looking at $75 million plus in 2019 that’s based on the 50% market share, if other vendor stumbles or is late and you end up with the $75 million number of 75% share that would increase that $75 million to well over 100, is that correct way to think about it?.

Definitely. Yes, that’s the right way to think about that..

So the total demand they are talking about is 150 million and….

And it could be assuming that the customers' estimations are indeed materializing, of course..

Of course I understand, but just wanted to make sure I understood the mechanics around it..

Yes..

Thank you.

There are no further questions at this time. Before I ask Mr. Orbach to go ahead with his closing statement, I would like remind participants that a replay of this call will be available by tomorrow on Silicom's website, www.silicom-usa.com. Mr. Orbach would you like to make your concluding statement..

Yes, thank you operator. Thank you everybody for joining the call. We look forward to hosting you on our next call in three months time. Good day..

Thank you, and this conclude Silicom's third quarter 2017 results conference call. Thank you for your participation. You may go ahead and disconnect..