Seeking Alpha
We cover over 5K calls/quarter
Profile| Send Message|
( followers)

Applied Micro Circuits Corporation (NASDAQ:AMCC)

F1Q14 Earnings Call

July 24, 2013 5:00 PM ET

Executives

Michael Major – VP, Human Resources

Paramesh Gopi – President and CEO

Shiva Natarajan – VP, Corporate Controller, Chief Accounting Officer and Interim CFO

Analysts

Ambrish Srivastava – BMO Capital Markets

Patrick Wang – Evercore Partners

Vijay Rakesh – Sterne Agee

Jason Rechel – Oppenheimer and Company

Christopher Longiaru – Sidoti & Company

Hans Mosesmann – Raymond James

Christopher Rolland – FBR Capital Markets & Co.

Krishna Shankar – ROTH Capital

Operator

Good day, ladies and gentlemen. And welcome to the First Quarter 2014 Applied Micro Circuits Corporation Earnings Conference Call. My name is Aisha and I’ll be your coordinator for today’s call.

At this time all participants are in listen-only mode. Later we will conduct a question-and-answer session. (Operator instructions) As a reminder this call is being recorded for replay purposes.

I would now like to turn the conference over to your host for today, Mr. Mike Major, Vice President, Corporate Marketing. You may proceed.

Michael Major

Good afternoon, everyone, and thank you for joining today’s conference call. On the call with me is Dr. Paramesh Gopi, our President and CEO; and Shiva Natarajan, our Vice President and Interim CFO.

Before we begin I would like to remind you that various remarks that we make on this call, including those about our future financial results, including revenues, gross margins, operating expenses, design wins, product plans, our competitive situation, market trends, statements about future developments and adoption of X-Gene and our other products and our anticipated growth and profitability all constitute forward-looking statements for the purpose of the Safe Harbor Provisions under the Private Securities Litigation Reform Act.

These forward-looking statements and all other statements that may be made on this call, that are not historical facts, are subject to a number of risks and uncertainties that may cause actual results to differ materially. We refer you to our most recent Form 10-K and Form 10-Q filed with the SEC, in particular to the section entitled risk factors and to other reports that we may file from time to time with the SEC for additional information on factors that could cause actual results to differ materially from our current expectations. These forward-looking statements speak only as of the date hereof and we disclaim any obligations to update them.

I want to point out that Applied Micro has several analysts that cover our stock, and this creates a range of variability relative to the Street financial models. When we say Street estimates, we mean the consensus of all the major analyst models and not necessarily the guidance that was given by the company.

With that I’m going to turn the call over to Paramesh. Paramesh?

Paramesh Gopi

Thanks, Mike. We had a solid June quarter with revenues of $54.1 million. Non-GAAP EPS was $0.02, ahead of our original breakeven plan and driven by better than expected gross margin. Without the R&D for our pre-revenue X-Gene development (inaudible) our non-GAAP pro forma EPS would have been $0.23 per share. Our backlog as of the date of this call is greater than 75%, and book-to-bill is greater than one. We ended the quarter with $89.7 million of cash on the balance sheet with no debt.

We are pleased to report that X-Gene production rollout momentum continues to exceed our expectations. Multiple Tier 1 customers are actively engaged in product development and are making excellent progress towards commercialization of X-Gene-based server platforms.

I will now discuss our product line starting with our base business. Our embedded computing revenues were $26.5 million for the first quarter and we continue to see strong design win traction. We believe our current embedded computing products will continue to generate solid revenue growth for us in the next 18 to 24 months, as we transition to X-Gene-based embedded products. For the September quarter we expect the embedded computing revenues to decline slightly as our new products continue to ramp.

Our connectivity revenues were $27.6 million. Our base OTN and Ethernet business grew over 30% compared to the previous quarter resulting from a significant uptick in demand for our converged 10-gig and 100-gig carrier Ethernet and OTN devices in the industry leading service provider and data center packet optical platforms such as the Cisco ASR high-density edge router which is the industry’s best selling router platform today.

Additionally, our 100-gigabit per second Ethernet and OTN physical layers are designed into numerous line cards in the high-growth packet optical DWDM systems from leaders in the service provider space, including ALU and ZTE. We expect to see good growth in the September quarter largely driven by the growth of the data center and wireline infrastructures.

A promising development worth noting is that we see early indications of a pick-up in wireline infrastructure spending. As you know this segment has been largely dormant for some time. As and when this trend takes hold we expect positive revenue and margin contributions in our base business. Although we are seeing early indications of an uptick in wireline communications spending we are being conservative and not embedding material growth in our estimates.

Now to X-Gene; X-Gene customers are progressing rapidly with the development of X-Gene based servers as it continues to reach new milestones. To provide some context on this let me walk you through our customers’ typical development cycle for servers. First comes the evaluation phase. The evaluation phase is where the customer will ask Applied Micro to sample X-Gene typically on our X-C1 development platform. Next comes the design phase where the customer creates their own system based on X-Gene. The design phase represents a significant financial and resource commitment by our customer and is not pursued lightly.

Next comes the system validation phase. The system validation phase is where the customer manufactures prototypes of the design and is in pre-release systems validation. A customer will already have invested multiple millions of dollars each to get to this validation phase.

Next comes pilot production, wherein thousands of nodes and many hundreds of clusters are built by our key customers, wherein their workloads in mission-critical applications will be tested in a microcosm. The final step is production release of these devices into data center build-outs.

Several of our early adopter customers have completed both the evaluation phase and the design phase and are now in the process of validating their X-Gene servers. Other customers are in the evaluation and design phases. These customers cover three spheres of demand, hyperscale data centers, OEMs and ODMs. We believe all of these efforts will result in production deployments and commercial sales over the next 12 to 18 months. We continue to see strong demand for samples from prospective customers and ecosystem partners.

We expect to have a very strong position in the ARM 64-bit server market, because of its highly differentiated architecture and unprecedented integration offered by X-Gene. This is evidenced by a unique grounds-up high performance ARM 64-bit set of cores operating at up to 2.4 gigahertz, specifically designed for cloud servers; enterprise class features including ECC and RAS; highly integrated mixed signal I/O features including 40 gigabits per second of Ethernet; Generation-3 SATA; Generation-3 PCIE and workload accelerators; as well as a high-performance on-chip fabric with software defined networking capability.

When you roll up all of these capabilities you get something that is comparable to the industry standard x86 server platform that is fully ready to power the cloud at a significantly lower total cost of ownership. We expect X-Gene to be the key enabler for the essential software component for the ARMv8 ecosystem to drive commercial deployments.

Recently Red Hat demonstrated its Fedora operating system paving the wave for RHEL and enterprise customers. Linaro hosted the world’s first demonstration of a Kernel-based Virtual Machine running on X-Gene, an important step toward enabling low TCO platform-as-a-service offerings for data center customers.

Now on the customer front, we are very gratified to see that there were a number of public demonstrations of customer platforms based on X-Gene at events such as HP Discover in Europe and America and Computex in Asia. As we continue to lead the market with our first-generation X-Gene products, moving into the commercial deployment phase I’m excited to share with you that the second generation X-Gene development in 28-nanometer is well underway in line with our own and our customers’ expectations.

A noteworthy milestone in this regard is the completion of the 28-nanometer X-Gene test ship in our labs. Based on the measured results we expect a significant power reduction as compared to the first generation X-Gene product. In addition, integrated support for scaling out to 100 gigabits per second will be part of the X-Weave framework that we announced this morning. We expect to tape-out our second generation production part in late 2013 and to sample the customers in the first half of 2014. Based on this schedule, our customers should be able to design in the second-generation X-Gene seamlessly well ahead of any competitive offerings.

With that, let me turn the call over to Shiva. Shiva?

Shiva Natarajan

Thanks, Paramesh.

First quarter revenues were $54.1 million and in line with our guidance. Sales to North America accounted for approximately 55% of total revenues, sales to Europe contributed 14%, and sales to Asia contributed 31%.

Consistent with last quarter, we had two direct customers that accounted for more than 10% of our business; global logistics support vendor, Wintec, and a distributor worldwide, Avnet. If we look at where our parts end up, which includes product that moves through distributors and logistics support vendors we believe that Cisco and Alcatel-Lucent would each exceed 10% of our revenues for the quarter.

Distributor revenues for the first quarter were approximately $31.2 million. Channel inventory days excluding non-cancelable, non-returnable orders were at 66 days, and this is in line with our expectations.

Turning to the P&L, we continue to deliver on our commitment of achieving breakeven or better. Our non-GAAP net income for the quarter was $1.4 million, or $0.02 per share, compared to the non-GAAP net income of $0.1 million, or $0 per share, for the prior quarter. Our non-GAAP operating margin was a positive 1.7% of revenue and improved 2.5 points from the negative 0.8 percentage achieved in the last quarter. Our non-GAAP EBITDA for the quarter was a positive $4.5 million, or 8.4% of revenues compared to a positive $2.9 million, or 5.1% of revenues, for the prior quarter.

The first quarter non-GAAP gross margin was 59.2%, which is much better than our guidance of 56.5% plus or minus 0.5%. The single most important driver for this was product mix. Our key 10-gig and 100-gig Ethernet/OTN PHY products for data center and carrier infrastructure did better than expected.

Looking forward to the September quarter we expect this trend of favorable product mix to continue. We expect no licensing revenues and expect gross margins to be approximately 61.5% plus or minus half a point. First quarter non-GAAP operating expenses were $31.2 million, which is at the higher end of our guidance of $30 million to $31 million. This includes almost three weeks of operating expenses relating to TPACK that we sold in April and a couple of smaller one-off items that hit expenses.

For the September quarter we expect our non-GAAP operating expenses to be in the range of 31.5 to 32.5 million. This marginal increase is due to the timing of some tape-outs. Our non-GAAP interest and other income was $0.8 million. Interest and other income is expected to be approximately $0.3 million for the September quarter. Our non-GAAP taxes were approximately $0.3 million and we expect to see this continue at this level for the next several quarters.

The share comp for diluted EPS purposes was 70.2 million shares. Looking forward to the September quarter we expect the share count to be approximately 73 to 75 million shares. The increase is primarily due to the timing of the grant of shares related to Veloce post-closing earn-out milestones.

Turning to the balance sheet, our cash and investments totaled $89.7 million or approximately $1.28 per share at the end of the first quarter, an increase of approximately $4.3 million from the March quarter. Our inventory turns for the June quarter were 6.9 X, slightly higher than the 6.3 that we had achieved for the last quarter.

Veloce payments: During the June quarter we paid out a total of $25 million in cash and approximately $10 million in common stock for a total of $35 million. Cumulatively to-date we have paid a total of $68 million, out of an assumed maximum merger consideration of $178.5 million. During the September quarter we expect to pay a total of approximately $40 million, which we expect to be roughly equally divided between cash and stock. Assuming no other changes to our cash balance this would leave our cash balance at the end of the September quarter an estimated $70 million.

To summarize, through September we expect to have paid a total of approximately $108 million of the Veloce consideration.

Now turning to GAAP. Our non-GAAP financials exclude certain items required by GAAP. Our net income on a GAAP basis was $10.9 million or 0.15 per share, $0.15 per share versus a net loss of 17.6 million or $0.26 per share last quarter. A complete reconciliation between the GAAP and non-GAAP financials is set forth in our first quarter earnings release which can be found in the Investor Relations section of our website. Please note that there is no reconciliation relating to forward-looking non-GAAP measures.

That concludes my remarks, let me turn the call back over to Paramesh. Paramesh?

Paramesh Gopi

Thanks, Shiva. Finally, I’m very excited to mention this morning’s X-Weave announcement. We believe our next-generation family of terabit connectivity products will revolutionize the industry, providing unprecedented levels of high density low TCO integration and performance for both service provider and data center applications. X-Weave was co-created with top tier service provider and data center customers and uniquely leverages our established position in the service provider market to address the rapidly growing public cloud, private cloud and enterprise data center markets. The convergence of X-Gene and X-Weave will offer a full portfolio of data center components for the new cloud infrastructure and positions us well to execute on our growth strategy.

In summary we had a solid quarter where our revenues met guidance and we exceeded non-GAAP profit expectations. Our base business remains strong and we continue to execute on our market expansion strategy with both our X-Gene and now our X-Weave products focused on huge growth opportunities for our products that will transform our business model in the future.

With that let me turn the call over to Shiva. Shiva?

Shiva Natarajan

Thank you, Paramesh. Just before going to Q&A let me address our guidance for the September quarter. Total revenues are expected to be in the range of $54 million to $56 million with a midpoint of $55 million. Total gross margin is expected to be 61.5% plus or minus half a point; OpEx roughly $31.5 million to $32.5 million; interest income, $3 million and taxes of about $3 million.

Now let me turn it over to Mike. Mike?

Michael Major

This concludes our formal remarks. Aisha, please provide instructions to our listeners for the queuing process.

Question-and-Answer Session

Operator

(Operator Instructions). Your first question comes from the line of Ambrish Srivastava with BMO. You may proceed.

Ambrish Srivastava – BMO Capital Markets

Hi. Thank you, Paramesh and guys. Pretty good execution all around, including on the balance sheet. I had a couple of blocking and tackling and then a little bit bigger picture competitive outside of ARM for you, Paramesh. So on the guide for OpEx beyond the September quarter, how should we be thinking about it? And then beyond that, as you ramp up 28, how does that change the OpEx? So that was one.

And gross margin for the remainder of the fiscal year given that September you will see upside as well. And one number Paramesh, that you guys give us which helps us gauge the vitality of the core business is the new product, so if you could please give that out and then I had a follow-up on the server space.

Paramesh Gopi

I’ll wait to answer your question. I think Shiva has a comment.

Shiva Natarajan

Yeah. So when I read out the guidance, I think, I mistakenly said $3 million for interest income. We’re not expecting $3 million. It is $0.3 million. And likewise, for taxes, it’s $0.3 million number. And I’ll turn it over to Paramesh.

Paramesh Gopi

Yeah. So, Ambrish, let me repeat your questions and make sure that we address them each one at a time. The first question was OpEx and what is our view of OpEx given the next generation of X-Gene. We continue to maintain our stance that we generally develop two processor generations at once, and the OpEx that we have disclosed and we have talked about includes all of the 28-nanometer OpEx. So there is no change to our OpEx going forward.

The second you asked, the question you asked was gross margin. A little bit of color on that as we told you on the call today the increased gross margin is directly a result of our new products ramping. There are three major categories of new products that are ramping.

The most important category is the connectivity category where we are sole sourced on 100-gig and multi-port 40 and 10-git on the industry’s leading router and optical, packet optical platforms. Those really drive our gross margin and will give you the positive evidence that our base business and from a new product perspective is really, really driving our ability to do what we are doing here relative to moving the business up and investing in the future.

And then the third part that you asked was essentially I believe relative to what we should look for gross margin for the rest of the year, is that correct, Ambrish?

Ambrish Srivastava – BMO Capital Markets

Yes.

Paramesh Gopi

So I think – let me answer that as a follow-on to my previous answer which is I think we see all the signs of the carrier infrastructure market in the wireline space starting to recover. We will definitely benefit from that recovery going forward. So from our perspective the carrier recovery will directly benefit our gross margin profile going forward.

Ambrish Srivastava – BMO Capital Markets

And then the new products metric that you gave out, Paramesh?

Paramesh Gopi

I think what we’d like to do is get you right after the call as we work the numbers, Ambrish and give you that number.

Ambrish Srivastava – BMO Capital Markets

Okay. And then maybe the bigger picture question Paramesh, there’s a lot of what I would call a [waferware] talk out there on the ARM side so I don’t want to get into that because you guys, at least, to my humble opinion, are way ahead of the ARM, I’ll call some of them pretenders.

But Intel, you seemed to have wakened up the big giant, not to assume that they were ever asleep. Can you please help us understand what you’ve seen from them in terms of whether it’s PR or their product releases? And how does that impact or when you compare your roadmap with them, where do you see opportunities or threats for X-Gene? Thank you.

Paramesh Gopi

I think, first I want to say that the notion of a server-on-chip, we created the category. We moved the category forward and I want to reiterate what we said on the call. Given the – given what we’ve achieved in to our first generation product and what we talked about, at least, what we publicly disclosed with our second generation product. We have changed the effective metrics for a cloud server by doing what we’ve done, relative to integration, the notion of an SoC, which is a Server on a Chip and the notion of monolithic integration.

So I think, given the recent announcements that we’ve seen from Intel, it heartens us relative to being strategically on the right path to solve the problem for datacenters. And I believe that we have done the – we’ve broken new ground and it’s heartening to see the x86 industry standard server component vendors starting to acknowledge the new ground that we’ve broken.

Operator

Your next question comes from the line of Patrick Wang with Evercore. You may proceed.

Patrick Wang – Evercore Partners

Yeah, hi guys, how are you doing?

Paramesh Gopi

Good.

Patrick Wang – Evercore Partners

And so, my first question, Paramesh, I think you had talked about coverage. Could you just go through that number for us again real quick? It seemed like it was, I think I remember 75% but I just wanted to double check that?

Paramesh Gopi

Yeah, the backlog is 75% and our book-to-bill is in excess of one, yes.

Patrick Wang – Evercore Partners

Okay, great. So to follow-up on Ambrish’s question about gross margin. I mean I hear you that the product mix sounds like it’s moving in the right direction, so it’s – one, is it fair to assume that 61.5 is a nice starting point going forward? And then I guess this is related to my next question, you talked about Green Shoots and Wireline. We’ve been waiting for this for a long time now. I was hoping if you can just maybe read that into some of the milestones or key programs that you’ve got on your mind in terms of that business improving?

Paramesh Gopi

Yeah, so let me start by saying that we have had probably the best OTN/Carrier Ethernet quarter to-date in the last 2.5 years in our company. That’s a huge, huge milestone for us. If you look at that and translate that to where that’s coming from, that’s coming from fundamentally because of two pieces. One is leading packet optical platforms. I mean, we are in every line card, virtually every line card of all the leading packet optical systems including things like the Cisco ASR 9k which has been mentioned on Cisco’s calls over and over again over the last 18 months, right?

So our growth is now directly manifesting from the new products that got designed in almost a year and a half ago into all of these leading packet optical platforms going into the two major markets; Hyperscale data center connectivity that is between data centers and basically Wireline service provider connectivity. For instance, guys like Comcast, guys like Verizon, guys like AT&T starting to really bridge private clouds and service provider infrastructure together. We are extremely excited to see that, that gross margin is essentially really manifesting our market traction and our design win pipeline evidence in the marketplace. All right.

I can tell you for a fact that we’ve been waiting for this for a long time. And as opposed to most of our service provider based design wins ramping it’s great to see that both the data center infrastructure as well as the service provider infrastructure, which I think according to most analyst estimates who are looking at guys like Cisco and people who supply in there are growing between 25% and 35% year-on-year. We’re directly going to benefit from that.

So from my perspective, I think, the gross margin is directly tied to our traction and our pipeline in that market.

Patrick Wang – Evercore Partners

Okay. That’s helpful. So I mean it feels like the way that you’re talking about it that you are seeing acceleration perhaps into the December quarter. Is that fair?

Paramesh Gopi

Yeah. I think I would be conservative with that. I think I want to come back and tell you that we see a lot of the signs of it. If you think about it, if I take the derivative of how much we’ve grown especially on the OTN and carrier Ethernet side, as I said it’s been best quarter-to-date, right? We see signs of it. But then again what I want to tell you is that the datacenter component has a strong, robust, linear growth vector, but the service provider component is extremely lumpy, right? So while we see service – I would say, if we have five service providers. We see it starting to materialize. Whether it gets a phased over two quarters, over three quarters is what we are not able to look at right now.

Shiva Natarajan

It’s too early now to project out into the December quarter, Patrick, but I think we will definitely benefit from the tailwind of the trends we are seeing.

Paramesh Gopi

Absolutely.

Patrick Wang – Evercore Partners

Okay, great. And then also I guess I was hoping to talk about some of the synergies you’ve got between your 28-nanometer X-Gene part and your 40-nanometer part. I feel like you’ve talked about how you’re able to leverage the design and some of the learnings, perhaps maybe bug fixes and things like that and 40-nanometer is out there sampling today, how does that help you guys in 28-nanometer and then can you also remind us a little bit of the timing in terms of when you get your second Gen part out?

Paramesh Gopi

Yeah. So I wanted to make sure that we talk about this. Number one, X-Gene is the only ARM server part in the market. Therefore the entire ecosystem revolves around it. Therefore all of the learning in terms of mission-critical, first customer deployments are only privy to us because we are driving the tip of this peer, number one. Number two, the amount of – the millions of dollars of just pure manpower, software and open source investment that have gone into the first product are directly paying benefits to the advent of the second product.

And let me give you one example that would be illustrative. When you go and design a server and somebody puts millions of dollars of commitment from a customer point of view on to your platform they look for at least two generations of roadmap. Why, as with every server generation in industry-standard x86 world you usually have a value generation and you have a volume generation. The value generation is for different tiers of the infrastructure.

For instance, if you need a very large scale-out mem cache tier that would be more suited to a value, but to a high performance generation rather than a high volume generation. Whereas, if you need web scale, you would use our volume generation. So it’s the same. We’re not changing the server cadence of the industry. We are providing a much lower TCO shelf and following the same cadence.

The amount of software that has been written for X-Gene bar none is foremost in the entire ARM camp, right? Nobody else has our software advantage. Therefore from a timing perspective, we see roadmap – we – our customers are talking about how to tier their datacenters, both with our first-generation products and our second generation products at the same time. That’s the way the server market has been cadenced and that’s exactly the way we want to adapt our product strategy to make sure we match the cadence.

Patrick Wang – Evercore Partners

Okay. Can you also address on when we might be able to see the 28-nanometer part in production ready for customers?

Paramesh Gopi

Yeah. I think what we said on the call today was, we have test chips back. We are fully now set in terms of achieving the performance expectations that we thought we would. And remember, for us, it’s an SoC; it’s not a chipset, number one. Number two, all the mixed signal, if you noticed today, we talked about X-Weave, X-Weave is going be such critical part of the 100-gig scale-out piece. Right. Both X-Weave and X-Gene will be available to customers in the first half of next year.

Patrick Wang – Evercore Partners

Perfect. Thanks so much. Congrats.

Paramesh Gopi

Thank you.

Operator

Your next question comes from the line of Vijay Rakesh with Sterne, Ag. You may proceed.

Vijay Rakesh – Sterne Agee

Yeah. I guess, good job on the quarter here. I was looking at the OTN commentary that you gave. You said that it’s tracking very well, how much was it in the June quarter and how do you see that tracking to end of the year?

Paramesh Gopi

I think what we said was that we had a 30% quarter-on-quarter growth. While we don’t breakout the numbers strictly, I can tell you for a fact that the number of ports were in far excess of 30%. And if you look at the optical platforms that we were going after they’re not purely OTNs. So I want to be very clear. OTN is becoming a standard port protocol for 10, 40 and 100 gig Ethernet type physical layer devices, for any type of distance.

So I’m glad you asked the question because the one thing that we want to emphasize is that while we say OTN, we mean packet, optical or packet data center. So OTN has now, thanks to our innovation, OTN has become a standard for all 10 gig, 40 gig and 100 gig Ethernet ports in the data center and the service provider edge. So that’s the key.

Vijay Rakesh – Sterne Agee

Okay. And on the 38-nanometer X-Gene also I know mentioned it’s in the lab. When do you start sampling it. I know it has a pretty good performance increment over the 40-nanometer but when do you start sampling that?

Paramesh Gopi

Yes, so we said both X-Gene 28 and X-Weave 28 will sample the first half of next year.

Vijay Rakesh – Sterne Agee

Got it. Great. And last, just briefly, I know Intel announced the Rangeley and then you are hearing a lot of your peers talk about ARM servers, so obviously they are far behind, but how do you see the competitive environment, let’s say, into first half next year?

Paramesh Gopi

Let me – when you say competitive environment, are you talking vis-à-vis ARM or vis-à-vis Intel?

Vijay Rakesh – Sterne Agee

On the ARM side, there are a lot of other entrants who want to come in, be it AMD or Intel or guys up and down the street from you. Just wondering how that environment looks as you look at first half next year.

Paramesh Gopi

Well, I want to once again, mention, right? We created the category. We spent three years developing two generation of industry-standard x86 class enterprise-quality cores running at 2.4 gig. We also spent three years developing all of the proven IO, the same OTN and 10-gig and 40-gig and 100-gig Ethernet IO that is shipping, that is helping our base business today is part of X-Gene and now X-Weave.

When we build, when somebody talks about ARM servers and they talk about A57, A57 is a tablet core. It has nowhere the performance that we are talking about. Neither does the server-on-a-chip strategy have the full portfolio of IP that we talk about. So from our perspective, the pure time aspect, the investment and the asset portfolio creates a hugely differentiated barrier to entry and really helps us ascertain our lead relative to customers.

Vijay Rakesh – Sterne Agee

Got it. Great, thanks.

Operator

Your next question comes from the line of Rick Schafer with Oppenheimer and Company. You may proceed.

Jason Rechel – Oppenheimer and Company

Yeah. Hey, guys. It’s Jason Rechel calling in for Rick. Thanks for squeezing me in. Paramesh, you did a good job, I thought, of kind of breaking out the five different design stages of X-Gene. And I guess I’m just curious if you have a couple customers in the validation phase today, how long are each of those phases?

So if we’re in the system validation phase today, how long until we start to see an actual build-out? And likewise, if you’ve got multiple customers in the design phase today, how long does that on go before you start to enter the validation phase?

Paramesh Gopi

Yeah. So let me take one step back and walk you through kind of timing. And related back to what we’ve told you guys and in terms of our path. When we sample something it’s on our platform. It typically takes three to four months for customers to start to run software on our platform. And the next three to four months after that is basically getting their platforms up and running, right, that’s the second phase.

And then this following three to four months, think of it as a third quarter after we sample our platform is when they get their platforms up and running and start to run real mission-critical loads on their platforms. After that we fully expect them to tweak their platforms and to change whatever configuration they need to on their platforms based on our back and forth with us and it takes roughly about the next quarter after that for them to build out a cluster.

And there’s three tiers of customers, there’s hyperscale data center guys whose build plan is a function of scaling their data centers. For instance, a guy like an Amazon or a Facebook or a Google they have predetermined plans to scale their infrastructure and they’re on their schedule. The other piece of it is the industry standard server OEMs like HP and Dell. As you know they basically sell to a plethora of Tier 1 and Tier 2 enterprises and cloud hyperscale guys.

They are a lot more predictable because they are portfolioed across 20 Tier 2 guys and their volume equals all of the three big cloud guys, right? So the top three are the most heterogeneous. The next top tier server OEMs are more homogeneous in terms of the way they ramp. And the last are the networking companies like the likes of the Huawei, the Cisco, the ZTEs they are a lot more enterprise focused, right? So those are the three categories but typically from the sample to first revenues meaningful revenues is about a year and we’ve always said that expect revenue in our fiscal – calendar 2014, fiscal 2015.

Jason Rechel – Oppenheimer and Company

Okay. Got it. Thanks. And I guess moving to the Veloce payments, as you exit September, I guess, there would be potential for 70 million in incremental pay outs, could you just walk us through your assumptions in terms of what you expect to pay over what time period of that remaining 70 million?

And then, second part is, I guess, if I back out your core X-Gene spending, it looks like R&D spending for the core business is running a little less than $10 million per quarter today. Do you think that that’s a sustainable level of investment or a requisite level of investment to grow that business over the next couple of years?

Paramesh Gopi

Let me answer the second question first; R&D spending for the base business, right? So if you look at our base business, all of our connectivity assets, we spent a lot of – a lot to get our connectivity portfolio reoriented towards the datacenter. Scaling the connectivity portfolio is going to be cadenced with scaling of X-Gene, which is why we announced X-Weave and X-Gene as two major products going forward; one for 100-gig connectivity and 240-gig connectivity, both within the rack and outside the rack in the datacenter, and X-Gene at 28.

So a lot of the IP is being leveraged between both product families, I want to be very clear about that. So when you look at a $10 million per quarter kind of R&D spend for the base business, that’s all going towards high-speed connectivity and today we are looking at things like finFET for the next generation, right? So, first question.

The second question is Veloce and the September quarter. As we’ve said before, I think we’ve been very transparent in terms of how we plan to how we’ve paid Veloce. The payment is contingent on certain milestones being met, number one. Number two, the payment has two components. Obviously, milestone component and a retention component, that will be over the next 12 to 18 months going forward.

Jason Rechel – Oppenheimer and Company

That’s good. Okay, got it. Thanks. And then, lastly, I think you did a phenomenal job on margins in June and it seems like some of the 10 gig, 100 gig stuff was perhaps a little bit unexpected. So, I guess, I’m just curious what may be underperformed relative to your expectations in June and do expect that to continue into September?

Shiva Natarajan

Yeah, so the margin mix is clearly as you can see from the revenue numbers also skewed more towards connectivity. And so we’ve talked about what sort of drove that, right. And so that is the trend that we are going to continue to see that we are seeing for the September quarter and hopefully for the quarters beyond that, that was the driver for the better margins.

Paramesh Gopi

So the other thing I want to talk about, to follow-on to Shiva is, and we mentioned this on the call. There is a little lumpiness in our embedded computing business, which is typical of enterprise cycles, nothing that we don’t expect. But I think the one thing that we’ve always talked about is that amount of design win pipeline that we have in leading packet optical data center and service provider infrastructure platforms. It was not on accident by any means. I can tell you that for a fact. It was very planned and more importantly, it was deliberate because a lot of the work that we did for the last two and half years is going to pay off in large dividends over the next two years in terms of the key connectivity portfolio that is in the top grossing, top ranking router packet platforms that are both pervading the data center and service provider space.

Jason Rechel – Oppenheimer and Company

Great, thank you, guys.

Paramesh Gopi

Thanks.

Operator

Your next question comes from the line of Christopher Longiaru with Sidoti Company. Please proceed.

Christopher Longiaru – Sidoti & Company

Hey, guys. I’ll add my congratulations on the execution.

Paramesh Gopi

Thank you.

Shiva Natarajan

Thanks.

Christopher Longiaru – Sidoti & Company

So I guess my first question comes with X-Weave and how that plays into your retention strategy, what does it mean going forward from an opportunity perspective? I think you said it’s a $1-billion addressable market. And what’s the timing on infiltration of that, and what you think the cost is from an R&D perspective going forward?

Paramesh Gopi

Okay, good question. First of all, we’ve always said that we wanted to take our company and expand our SAM and TAM by almost 10x. That’s why we started X-Gene three years ago to create a new category and really break new ground for a new market. And the other – the big part of X-Gene was X-Weave. Why is X-Weave a $1 billion? Because if you look at the amount of data center connectivity that goes out whether it’s between data centers in the form of large optical links, within data centers in terms of clustering servers or within racks in terms of clustering blades. X-Weave applies to all those three markets, right? Number one.

Number two, X-Weave R&D has already been factored into all of our OpEx. So there is no extra R&D for X-Weave. X-Weave, as you can see is contemplated as part of our entire R&D spending and is completely included in our OpEx. Okay. So one of the things I want to say is that with X-Weave we now have the entire portfolio from the time a fiber gets into a rack to the final fiber leaves the rack to supply all of components to connect servers, to connect racks and to connect data centers together. And it represents the most complete offering of connectivity, analog, DSP and processing in any type of data center installation.

Christopher Longiaru – Sidoti & Company

Thank you for that. I also wanted to ask, you guys said that you have 75% of your backlog booked in for the guide that you gave?

Paramesh Gopi

Yes.

Shiva Natarajan

Yes.

Christopher Longiaru – Sidoti & Company

Okay. Could you just refresh my memory, how much was booked in when you give the guidance for this quarter at last quarter’s quarterly conference call?

Paramesh Gopi

I think it was close to 77% if I remember the numbers right.

Christopher Longiaru – Sidoti & Company

Okay. That’s all I needed. All right, thank you very much guys. I will jump out.

Paramesh Gopi

Thank you.

Operator

Your next question comes from the line of Hans Mosesmann with Raymond James. You may proceed.

Hans Mosesmann – Raymond James

Thanks. Hey, Paramesh, now that you have a little more understanding of what the market dynamics are, can you remind us the opportunity for X-Gene in terms of the size of the market and how it relates to the size of the market that the other ARM players are going to have over the next several years given that their approach seems to be – to take the Cortex-A57 class product and which is probably more low end, that would be very helpful? Thanks.

Paramesh Gopi

Yeah, so really good question. So let me just talk about market first. Today cloud server market is roughly a third of the entire server market. It is the fastest growing segment of the server market. That market today is exclusively serviced currently by x86 Xeon class family of processors and chipsets. Therefore from our perspective if you don’t have a custom 2.4-gig core with fabric with the requisite type of between 40 and 80-gigabit – gigabytes per second of memory bandwidth, plus all of the integrated 10 and 40-gig and 100-gig I/O, you don’t – there is no way you can play in that market. This is a very, very key technical, I’ll call it, step or hurdle that you need to cross, number one.

Number two. All the other ARM players from best we can tell, using a mobile tablet core may build an appliance to service a small niche in that market, but cannot effectively service the entire market. Did I answer the question?

Hans Mosesmann – Raymond James

Yes, and so, what size of the market would, so you’re going to be addressing all of that market, one-third of the server market and what part of the market will the others be addressing? Would it be half of that one-third or a fraction?

Paramesh Gopi

I would say – yeah, I would say, it would be a fraction and the reason I say that is, if you were to take the cloud today and boil it down to – there is basically the largest number of servers in the cloud is doing web scale stuff. The second largest server pool in the cloud is doing things like Hadoop and large storage. And the third is basically doing mem cache type workloads, right. Those three workloads. Well, you may get a tiny sliver of one-third of one of those markets, like you may be building a purpose-built appliance for some small storage type category. But once again, let me tell you that industry standard cloud servers today run a 64-bit enterprise-class open source stack. If you don’t have enterprise features, there is no way you can service that market, because nobody is going to change their code base for an appliance.

Hans Mosesmann – Raymond James

I see. That’s very helpful. Thank you.

Paramesh Gopi

Thanks.

Operator

Your next question comes from the line of Chris Rolland with FBR. You may proceed.

Christopher Rolland – FBR Capital Markets & Co.

Hey guys, good quarter. So you guys talked about hyperscale customers versus call it micro server system OEMs. So what do you think the breakup is right now in terms of trial boards or trial engagements in each of those segments? Just on a percentage basis, call it.

Paramesh Gopi

Are you saying from our perspective, right, from what we are looking at right now, correct?

Christopher Rolland – FBR Capital Markets & Co.

Exactly, are you shipping 30% to hyperscale guys and 70% to micro server OEMs? What’s that breakdown look like?

Paramesh Gopi

It’s actually pretty equal. It’s almost 50-50, right. So, in other words, I think both segments of the market and I will add the third segment which is networking and core networking type segment, is the other market that we are trialing things into right now. But it’s equally spread across both those segments, Chris.

Christopher Rolland – FBR Capital Markets & Co.

Sure. Maybe you can talk a little bit more about that networking segment and that opportunity there?

Paramesh Gopi

Yeah. So one of the things that we’ve talked about for a long time is the ability to have micro servers and things like service provider equipment like a micro server blade. That model exists today. So obviously, when you have a service provider box and you have a micro server blade in the service provider box, the biggest thing is service provider box is the density per unit power. So given our low TCO footprint and open source software compatibility, those service provider boxes that contain micro server blades are extremely interested and have a lot of traction with us in trying to move their current industry-standard server blade offering to an X-Gene type product.

Christopher Rolland – FBR Capital Markets & Co.

Okay. And then between 40-nanometer and soon-to-be 28-nanometer, do you think there are significant number of guys waiting on the sidelines before they start the design cycle waiting for 28-nanometer or is there really no benefit to waiting? How do you guys sort of view the ramp in terms of design or trials from 40 to 28? Thanks.

Paramesh Gopi

So a really good question. The 40-nanometer piece is so critical because it’s the cornerstone of the first real, I’ll call it, industry-standard compatible chipset for an ARM-based server with all the TCO benefits. Right. So the one thing that we’ve consciously tried to do is, again, follow the same server cadence, volume generation/value generation versus technology generation.

So if you look at the cloud today, the volume – if you were to take the three tiers, which is the web scale tier, the storage tier or large distributed file system tier and then the mem cache tier, it’s like a pyramid with mem cache being at the top, those smaller number of servers requiring the most high performance. Right.

So if you look at it, where do people want to slop things in? Well, they are going to follow exactly what they’ve done before. They take this generation, put it where the volume is and as the volume starts to move up – servers get basically re-tiered based on generation. That’s what happens today. They are going to do nothing different, which is re-tier their server allocation for the next generation.

The one thing that the 28-nanometer allows you to do is scale out to 100-gig-type fabrics, all right, with X-Weave as being a key part of it. So the ability for you to combine both those to have for instance a cluster-based aggregation server would be the key.

So nobody’s going to wait for 28 to get the TCO benefit that they can get now. However, if you want to scale out and scale out to a Next Generation set of applications that is higher in the tier you would use 28. But I’ve got to tell you that one of the big things that we have learnt over the last two years is no customer wants to commit millions of dollars of NRE unless they see a two generation tangible roadmap.

Christopher Rolland – FBR Capital Markets & Co.

So how do you guys view those 28-nanometer potential customers, is it a big step up from what people are doing at 40 now in terms of number of trials?

Paramesh Gopi

Actually it will be a much easier smoother design cycle because the software is seamlessly portable. All the learning, all of the ecosystem seeding, when you move from X-Gene to X-Gene it’s going to be probably as opposed to a year and a half to go and seed an ecosystem it’s going to be months, okay, for transitioning code from one generation of X-Gene to another generation of X-Gene. So from our perspective the customers in their mind it’s a clean, easy roadmap move for them to re-tier their servers based on our roadmap.

Christopher Rolland – FBR Capital Markets & Co.

Okay, all right. Great, thanks, guys.

Paramesh Gopi

Thanks.

Operator

(Operator Instructions) Your next question comes from the line of Krishna Shankar with AppliedMicro APM. You may proceed.

Krishna Shankar – ROTH Capital

Yes, congratulations on the good results and progress folks. With the X-Weave product line that you announced would that be into, as you start to sell X-Gene in high-volume would you need to bundle the two together, in other words, X-Gene and X-Weave would go together for Next Generation data centers or could you have standalone sales of the X-Gene with third-party networking components?

Paramesh Gopi

Short answer is, you could have essentially the latter, which is, you could have standalone X-Gene without X-Weave. But when we look at racks and when you build a rack upon rack upon rack upon blade servers, we want to make sure that people have both components from us. The whole intention was to be able to provide a full portfolio.

So having said that, X-Weave and X-Gene will work seamlessly together, things like latency for large clusters, things like quality of service, for reprioritizing traffic between clusters are all contemplated between the portfolio of products. So while you can do it with some other vendors’ product, you would get a great deal of performance and power reduction when you bundle the two together.

Krishna Shankar – ROTH Capital

Okay. And then with the X-Gene and the 28-nanometer product line have you given us any details of how it will be different or enhanced relative to the current 40-nanometer version? What types of new features will be on it?

Paramesh Gopi

We have not publicly disclosed those. The only thing that we publicly disclosed is that it will have support to scale out to 100 gigabit per second backbones and scale out Ethernet networks so...

Krishna Shankar – ROTH Capital

Great. Thank you.

Paramesh Gopi

Thanks.

Operator

There are no further questions in the queue at this time. I would now like to turn the call back over to Mr. Mike Major for closing remarks.

Michael Major

Thank you all for your participation today. There will be an audio replay of this call available on the Investor Relations section of our website. You can also access the audio replay of this conference call by dialing 888-286-8010 and entering the reservation number 19311239. We will also file a copy of the transcript of this call and an 8-K with the SEC in the next few days. Please feel free to call me if you have any additional questions. Again thank you all for your participation on the call today and have a nice evening.

Operator

Ladies and gentlemen, that concludes today’s conference. Thank you for your participation. You may now disconnect. Have a great day.

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.

THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.

If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!

Source: Applied Micro Circuits Corporation (AMCC) CEO Discusses F1Q14 Results - Earnings Call Transcript

Check out Seeking Alpha’s new Earnings Center »

This Transcript
All Transcripts