Paramesh Gopi – President and CEO
Robert Gargus – CFO
Dan Morris – Oppenheimer & Co.
Allan Mishan – Brigantine Advisors
Christian Schwab – Craig-Hallum Capital Group
Sandy Harrison – Signal Hill Group
Brian Thonn - Kingdom Ridge Capital
Applied Micro Circuits Corp. (AMCC) F1Q11 2010 Earnings Call July 30, 2010 5:00 PM ET
Good morning ladies and gentlemen and welcome to the first quarter 2011 Applied Micro Circuits Corporation earnings conference call. [Operator instructions.] And I would now like to turn the conference over to your host for today, Mr. Bob Gargus, CFO. Please proceed sir.
Good afternoon everyone and thank you for joining today’s conference call. On the call with me today is Dr. Paramesh Gopi, our president and CEO.
Before turning the call over to Paramesh, I want to remind you that the forward-looking statements discussed on this call, including guidance we will provide on revenue, non-GAAP gross margin, non-GAAP operating expenses and certain other financial targets are based on the limited information available to us today.
That information is likely to change. There are numerous risks and uncertainties that affect our business and may affect these forward-looking statements, including product demand and mix, product development and introductions, design wins, manufacturing and supply availability, the impact of workforce reductions, employee relations, and the integration of new or moved operations, risk relating from macroeconomic conditions and markets, and other risks as set forth in our SEC filings, including our Form 10-K for the year ended March 31, 2010.
Our actual results may differ materially from these forward-looking statements. Applied Micro assumes no obligation to update forward-looking statements made on this call. I want to point out that Applied Micro has several analysts that cover the stock and this creates a range of variability relative to the Street financial models. When we say Street estimates, we mean the consensus of 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?
Thanks Bob, and good afternoon everyone. During the June quarter we saw our total revenues increase sequentially by approximately 6%, and product revenues, excluding our licensing revenues, increased sequentially by approximately 7%.
Our opening backlog for the September quarter, meaning orders for delivery in the September quarter, increased sequentially by approximately 17%. We are entering the September quarter with more than 100% of our revenue guidance already in backlog at the start of the quarter and more than half of the December quarter already on the books.
I want to make the comment here that our new products, those introduced since FY08, contributed about 15% of the June quarter revenues and this is expected to grow to over 20% in September and is expected to be over 25% in the December quarter. This clearly demonstrates excellent traction on our newer products.
Total book to bill was 1.3x and reflects robust bookings for both the September and December quarters. Bookings and billing strength was widespread and we showed strength in both processor and transport product families across all geographies and channels. It is important to note that both the processor and transport businesses had book to bill ratios in excess of 1.3 for the second straight quarter. Processor bookings had the longest lead time, approximately 24 to 26 weeks, so these will begin flowing more strongly as supply conditions improve.
I am extremely pleased to note that we have surpassed our guidance and have executed very well in all parts of our business and particularly in growth and profitability as well as the introduction of breakthrough technology and products that fuel our future growth. In fact, we announced three key products this past quarter.
The APM821xx, a high performance, energy-efficient device targeted at multifunction storage applications and media servers.
The APM801xx, a family of energy-efficient embedded processors providing the industry’s smallest form factor Power Architecture-based product. The APM801xx product family enables a broad range of devices, spanning gateways, control panels, and appliances for the smart grid.
Finally, the APM9689x, our Triveni product family, the industry’s smallest dual and quad port, low power, 101,000 10GBASE-T PHY technology. Triveni is the first 10GBASE-T PHY to cancel interference due to walkie-talkies, cell phones, and other wireless devices to solve a critical technology challenge for robust, high-speed Ethernet connections for the data center. I will talk more about this exciting product later in my comments.
Let me now turn to some of the details of the business, starting with transport. Please note that our transport business encompasses two vertical markets, telco and data com. Let’s start with a brief review of our telco business.
I’d like to begin by describing the end markets and their dynamics. While the worldwide carrier transition to OTN is well underway, we see a distinct acceleration of pace in the Americas and Asia Pacific fueled by both legacy upgrades and greenfield deployments. It is important to note that the EU carrier network evolution lags the Americas and A-Pac.
Platform-wise we see two distinct classes of platforms replacing legacy systems: intelligent packet-aware optical switches, and true multi-service super routers. The most common interfaces used in both these classes of platforms today are 10 Gbps and 40 Gbps. The most significant difference in these equipment classes versus legacy is that their overall capacity, and most importantly their support of large port count Metro-class interfaces.
We have also started to see the emergence of proprietary 100 Gbps coherent interfaces for long-haul upgrades and expect client-side 100 Gbps interfaces to become prevalent within the next 3 years. To be explicit, these on the client side will consist of configurations that range from 3 by 40 Gbps to 12 by 10 Gbps.
Having shed light on the end market, let me turn to the specifics of our business. Our OTN business was up 50% sequentially, fueled by Cisco, ZTE, and ALU. Our Yahara platform line continued to win key transponder, as well as multiport, multiservice interface cards for various platforms at ALU, ZTE, and FiberHome. Initially, URAL, our leading edge, multiprotocol FEC companion device to Yahara scored major design wins at Fujitsu, replacing legacy comparative solutions that were space and power inefficient.
On the router and packet platform fronts, PEMAQUID, our second-generation, multiport OTN platform, continued a strong ramp at Cisco and NSN. Additionally, RUBICON, our first-generation OTN platform, showed robust demand across our entire consumer base as [unintelligible] Metro platforms at VDE, Tellabs, ALU, and FiberHome were rolled out last quarter. Finally, our breakthrough 40 nanometer PQX Metro platform continued to secure leading-edge designs at market leaders such as ZTE, for 40- and 100-Gbps power efficient multiport Metro platforms. As we’ve stated before, Applied Micro products enable virtually every recently announced OTN platform. Both our current period revenue and our [unintelligible] continue unabated.
Turning now to our data com market. Overall, our data com product revenue grew 40% sequentially but was partially offset by a decline in our legacy switch business revenue. Please remember that our legacy switch business comes in lumps and can easily go up or down quarter to quarter by up to $4 million. Our design build pipeline for our low-power, multiport, 10 Gbps physical [unintelligible] devices, as well as [unintelligible] and SERDES, was robust and we gained sockets at multiple Tier 1 customers, including Cisco, QLogic, and Emulex.
I would now like to introduce to you our newest product platform for the data com segment. Triveni represents groundbreaking deployed development that enables robust transmission of 10 Gbps over unshielded twisted pair or UTB copper cable. Please note that this is a large, emerging market, estimated to be $475 million in 2015 and represents a significant portion of the overall data center physical layer solution space.
To date, 10 Gbps copper solutions have been stymied by poor noise performance and electromagnetic interference susceptibility. Current 10 Gbps copper solutions are unreliable and are unable to perform in the presence of commonplace radio frequency equipment such as walkie-talkies and 3G cellphones. Our pioneering, patented, low-power interference cancellation technology mitigates these issues and represents an enabling step towards robust 10 Gbps copper transmission over unshielded twisted pair copper cable. We have a number of early engagements in this space and hope to be able to share our design-win pipeline with you in upcoming quarters.
In summary, we had an excellent quarter on the OTN and data center fronts, and are excited about the prospects of our new technology and its value proposition to attract the large emerging data center copper solution space.
Now, turning to the embedded processor market, processor revenues grew more than 4% sequentially in the June quarter. Our business growth was fueled by strength in the telco, enterprise, and small-medium business segments uniformly. Specifically, Cisco, Edison, and Brocade showed robust and steady growth. Our power-efficient 460 class product line picked up a number of new control plane and in-line packet processing sockets from Metro and Edge equipment at the aforementioned vendors.
On the consumer and small-medium business fronts, our newest single digit ASP Power Architecture platform [unintelligible] by leading energy gateway vendors for their next generation home automation and control platforms. Additionally, we secured a significant design-win for a managed services gateway that is sourced by a key satellite TV service provider.
Finally, I would like to draw your attention to an important aspect of our business that is gaining traction with our customers. Applied Micro has successfully transitioned into the role of a true silicon solutions platform provider, providing connectivity, computing, and value engineering services to our Tier 1 customers.
We are extremely excited to announce that we won our first WDM platform at a Tier 1 provider, which spanned processors, framers, mappers, and physical layer devices. In addition, we were able to value engineer the solution bundle to provide significant cost savings and energy savings towards our customer’s end platform. The strength of moving all of our IP to TSMC-based process technology, coupled with our converged value engineering approach, have proved to be a winning combination for our customers. We expect this trend to bring us large commercial success.
Now let me turn over the call to Bob. Bob?
Thanks Paramesh. For the second quarter in a row, we exited the quarter with very strong bookings and backlog. Book-to-bill was over 1.3 and we now have more than 100% of our revenue guidance in backlog. This does indeed seem strange, but the simple reality is that our guidance is based on what we believe we can deliver within the quarter. This is something that we’ve been pretty good at estimating, I that we have made or exceeded the last six quarters.
Unfortunately, this reflects our ability to ship less than the full demand requested by our customers. We do see tight supply conditions continuing for the remainder of the calendar year, despite seeing some supply conditions improve somewhat in June. However, this improvement in supply was more than offset by increased demand. We continue to proactively monitor the situation and will do everything we can to support our customers. I want to stress that the impact of these supply constraints has been built into the guidance, at least as best we can determine, and we obviously remain hopeful that we can do better.
First quarter revenues were $60.8 million, up $3.2 million or 5.6% compared to the prior quarter, and up $15.8 million or 35% from the same quarter a year ago. Within the $60.8 million of revenue we saw both transport and processor revenues grow, offset by a planned $500,000 decrease from our IP revenues.
The product-only revenues were $57.6 million, and were up 6.9% sequentially from last quarter. The breakdown of our revenues by product family for the first quarter was: processor revenues were $29.5 million, a sequential increase of 4.5%; transport revenues were $28.1 million, a sequential increase of 9.5%; and licensing revenues were $3.3 million, and were down $500,000, or 13.5%.
Sales to North America accounted for 31% of total revenues. Sales to Europe contributed 16% and sales to Asia contributed 53%. There was one subcontractor that represented 10% or more of the June quarter revenues, Hon Hai, at 14%. In March, Hon Hai was 11%. Additionally, there was one distributor that was more than 10%, and that’s Worldwide Avnet and they accounted for 33% of total revenues for the June quarter.
Distributed revenues for the last quarter were approximately $29 million, compared to $27 million for the prior quarter. Distributor inventory measured in days and based on the [unintelligible] June-quarter sell-through, is at approximately 39 days and is below levels of just a year ago. We have lead times lengthening and shortages occurring. We are actively encouraging our distributors to increase their inventory levels and over time would like to see distributor inventories at the 45 to 50 day level. We believe this is prudent for the channel to hold more inventory, particularly when demand levels are increasing and lead times are extending.
Turning to the P&L, our fourth-quarter GAAP net income from continuing operations was $8.3 million or $0.12 per share compared to the non-GAAP net income from continuing operations of $6.1 million or $0.09 per share for the prior quarter. The leverage factor for the quarter was approximately 69%, meaning that we put approximately 69% of the top line increase to the pre-tax profit line.
Our non-GAAP operating margins continue to improve and was 12.2% revenue and improved to approximately 3.8 points from the 8.4% recorded last quarter. We expect operating margins as a percent of revenue to be roughly flat, to up marginally in the September quarter. Our EBDA for the quarter from continuing operations was $9.7 million or almost 16% of revenues, compared to $7.1 million, or 12% of revenues for the prior quarter.
The first quarter non-GAAP gross margins, including licensing, was 67.6%, compared to 65.3% for the March quarter. This exceeded our guidance of 62.5% plus or minus half a point. The improvement in margins relative to guidance in prior quarter will be largely explained by the impact of favorable product mix and improved manufacturing yields and efficiencies.
Looking forward to the September quarter, we are expecting overall gross margins to be approximately 65% plus or minus half a point, and it is down from last quarter due to expected higher processor shipments and specific customer mix. For those building FY11 models, we would guesstimate the gross margins would be 65% plus or minus half a point for the remainder of the fiscal year. The 65% gross margins is 2.5 points better than our previous guidance for FY11.
Non-GAAP operating expenses from continuing operations were $33.7 million, compared to our guidance of $33 million. Our operating expenses include a $1.2 million credit related to a one-time adjustment for certain mass costs that we capitalized and these were offset by approximately $1.5 million in mass costs that we incurred and expensed during the quarter.
Net of the items above, the increase in operating expenses is almost completely related to additional incentive programs that kick in if we exceed certain financial performance parameters that it now appears we will exceed. These increased incentives will cap at $3 million and would result in $750,000 of increased op ex per quarter. This means that our previous estimate of averaging $33 million to $34 million per quarter for the year will now be $33.75 million to $34.75 million per quarter.
For those looking to understand this, it’s simply that we had a plan to increase EPS from about the $0.16 that we had last year to approximately $0.46 and we established a bonus program that would only begin above this level and would result in the excess above this level being split approximately one-third to the incentive program and two-thirds to the shareholder, but with an incentive cap of $3 million.
Our interest income was $1.2 million and is lower compared to the prior quarter due to the impact of approximately $800,000 in realized losses from sales of certain investment securities.
The diluted share count for EPS purposes was 68.7 million shares. The diluted share count increased by approximately 1.1 million shares compared to the prior quarter, mainly due to the impact of our higher stock price during the first quarter and the effect of shares granted last quarter pursuant to performance programs related to last year.
Going forward we expect to continue to repurchase shares under our board-approved repurchase program, and we expect to keep the share count flat at around approximately 68 million shares. At the quarter-end we also had two share repurchase contracts for $5 million each outstanding. One of these contracts settled for shares in July and the other contract is expected to settle in August, and this one could settle for either cash or stock, depending on the stock price. We will likely be doing open market purchases during the September quarter as we are committed again to maintaining our share count at approximately 68 million shares.
Interest income is expected to be approximately $1.2 million for the September quarter. Our tax rate continues to be projected at 3% for the next several quarters. The guidance numbers we have provided here, along with the revenue guidance that Paramesh will provide shortly, should result in non-GAAP Street models that generate approximately $0.13 in EPS for the September quarter, compared to the current Street consensus of $0.11. This means that we are basically pulling the Street models in a full quarter, meaning that the previous Street consensus for December is now our guidance for the September quarter.
Turning to the balance sheet, our cash and investments totaled $222.5 million, or $3.24 per share at the end of the first quarter, an increase of approximately $16 million from the March quarter. The increase is largely due to higher cash generated from operations, the proceeds from the sale of our equity investment in Intrinsity, and somewhat offset by an increase in our investments and working capital. The $222.5 million in cash balance excludes the $10 million that was related to the two unsettled share repurchase contracts outstanding as of June 30.
Capital expenditures for the quarter were approximately $1.5 million. You can refer to our cash flow statement in the earnings release for more information. Our working capital was approximately $238 million, an increase of $10 million, and we have no long-term debt. Our DSO is at 30 days and we expect it to range from 30 to 42 days going forward. Capital depreciation for the quarter was $1.9 million.
Our overall inventories increased by $2 million and were $17 million at the end of the quarter compared to $15 million at the end of last quarter. Our inventory terms were 4.5. Given the industry-wide lengthening of lead times and the various part shortages, we will plan for our inventories to increase in order to provide timely support for our customers. We expect inventory turns to be in the 4 to 4.5 range for the next quarter or two.
Turning to GAAP. As you know, our non-GAAP financials exclude certain items required by GAAP, such as amortization or impairment of purchase intangibles and good will, items related to other than temporary impairment charges on our investment portfolio, impairment of strategic investments, stock-based compensation expense, restructuring charges, and non-cash tax adjustments. The timing, occurrence, and magnitude of such items can be difficult or impossible to estimate for future periods.
Our net income on a GAAP basis was $1.4 million versus a net loss of $100,000 last quarter. The difference in our first quarter GAAP net income of $1.4 million and our first quarter net income from continued operations of $8.3 million is a delta of $6.9 million. The $6.9 million is primarily comprised of two items: one, $3.9 million of stock-based compensation and $3.6 million of amortization of purchase intangibles respectively, a total of $7.5 million, and two, a combination of changes to our investment impairment reserve and restructuring charges that netted to a $600,000 benefit.
Looking forward to the September quarter, we can expect certain known GAAP charges, such as the amortization of purchase intangibles and the stock-based compensation, to continue at approximately the same levels recorded for the June quarter. A complete reconciliation between the GAAP and non-GAAP financials can be found in our earnings release, which can be found in the investor relations section of our website. Please note that there is no reconciliation for forward-looking non-GAAP measures.
That concludes my remarks. Let me turn the call back to Paramesh.
Thanks Bob. Bob has already provided you with second-quarter guidance for the op ex and gross margins. Our revenue guidance for the September quarter is to sequentially increase product revenues by 5% to 9%. The midpoint of this is a 7% sequential increase and will target total revenues at approximately $65.1 million, plus or minus $1.2 million, an intent to shift the current Street models forward a full quarter such that our September quarter guidance is now equal to or greater than the previous Street consensus revenue and EPS numbers scheduled for the December quarter.
We see supply constraints continuing into early calendar year 2011 at a minimum. Despite these supply constraints, we see our year-over-year growth rate to be 25% to 28%, and we will work hard to do even better. This growth rate is in excess of our 20% goal and is fueled by the secular growth of our chosen markets. I want to stress that we are executing to our plan and we are excited about our growth opportunity.
We recognize that we still have a ways to go to get to our profitability goals, but we believe we are making excellent progress and we are improving at a much faster rate than expected. With that, let me turn the call over to Bob for the Q&A session. Bob?
Thank you Paramesh. That concludes our formal remarks. Operator, please provide instructions to our listeners for the Q&A process.
(Operator instructions.) And your first question comes from the line of Dan Morris of Oppenheimer. Please proceed.
Dan Morris – Oppenheimer & Co.
Talking about supply constraints, Paramesh you mentioned you expected it to continue into early next year. Is that affecting both transport and processors, or is it more on the processor side?
So Dan, we see it on both sides. It’s impacting both. We are in a regular mode of having calls with customers or with key customers asking to have joint calls with us and the suppliers to try to accelerate demand and in a lot of times even the customer recognizes that they’re asking for stuff inside of lead times, but their demand is better than they forecasted. So it’s still a problem and it’s something that is consuming a lot of management time and it’s just tough. We can’t seem to get ahead of it because I can’t build any inventory because as fast as we can build it it’s going out the door right now.
Yes, so Dan, this is Paramesh. I think to kind of take a step based on Bob’s comments, we are seeing very robust demand in markets, especially the market that we are pretty much driving and are single-sourced in as well as systems relative to optical networking, wireless networking, and overall Metro networking, doing really well. So everything that we build is gone before we know it, and so this is going to be a, I think, to really catch this and really go on top of it is going to take a huge amount of work on our part and you know, I think we’re gearing up for it. We want to build inventory. We want to make sure we’re in good shape, but it is a tough problem.
Okay, and maybe if you could update us on your plans for ramping up TSM, when that can occur and how quickly that can happen?
Well, TSMC has a $9.5 billion expansion program, but that will take some time, and what’s happening right now is they are being nice enough to, on occasion, find wafers for us or to help us, but it’s kind of like you’re borrowing from the next quarter. So they help you suck 400 wafers or 500 wafers forward from October into September, I’m making up an example, but in some ways that just makes the hole that you have in the December quarter potentially bigger. So we’re continuing to struggle with that.
In the case of IBM, they have informed us that they bought additional equipment, but the lead times for that equipment is long, and so based on what they’re telling us it’s probably the March quarter before they start to show any appreciable input from that perspective. Now I’ll also remind you that we said back even in January when we first highlighted that we could see this condition happening, that we thought we might get a break in the second half of the calendar year, where either the PC or the cellphone market might slow down a little bit and that might free up some capacity such that we could help get additional supply.
So depending upon whose numbers you want to look at, there are a number of people that are somewhat arguing that some of that is happening, and there are others that are arguing that it isn’t. Right now I’d say, again, we saw a little bit of supply improvement in June, but unfortunately my demand is going up faster than the supply is improving.
Okay. But at this point TSM is still a very small percentage of your revenues, right?
No, not at all.
No. IBM is probably in the range of about 50% to 55% and TSMC is the other 45% and in a simple way, I mean it’s a little bit of an oversimplification, but you can almost look at it as the transport stuff comes out of TSMC and the Power PC comes out of IBM.
Okay. That’s helpful. And in terms of your gross margin, obviously a big uptick here. Sounds like you feel things are sustainable at a much higher level for the next several quarters. Could you just talk a little bit about what gives you that confidence?
Well, we made a number of improvements where we’ve gotten improved yield. Because demand levels are so high the concept of having excess and obsolete reserves is almost out the door. So this thing has been creeping up. I think we’ll have a little bit of a downturn here in the September quarter, mostly because for the last two quarters we’ve had strong bookings inside the Power PC business but because of the 24, the 26 lead time, that hasn’t really started to flow to the P&L. We’re expecting to see the largest portion of that in the September quarter and historically the processor business has about 10 or so lower gross margin points than the transport business does.
Okay, great. And last one from me. Just in terms of op ex you mentioned that the new level is what, $33.75 million, $34.75 million that we should be thinking about. Is that, are you going to try to hold that level for the next few quarters or will it grow a little bit more with revenue growth?
No, I’m going to try and hold that for the remainder of this year and again the only change there is the $750,000 and that’s all related to the fact that the higher revenue growth margins and stuff we now are exceeding the full-year plan that would kick in this bonus program. And I actually am pretty proud of the way we did the bonus program because we set a goal to get to a number that we thought was very important to the shareholders and we said we wouldn’t pay anything to management or anything else until we exceeded that number. And then above that, with the cap, we split one third going back to management and two thirds going to the shareholders.
All right. Thanks guys.
And your next question comes from the line of Allan Mishan of Brigantine. Please proceed.
Allan Mishan – Brigantine Advisors
Hey guys. A couple questions. So just to clarify, on the op ex front, for the coming quarter you expect to be in the $33.75 million to $34.75 million range.
And for the balance of the year.
Okay, great. And then a question on two of your growth areas, OTN and the 10-gig PHY, were both up very strongly. I think you said OTN was up 50% quarter on quarter and PHY was up 40% quarter on quarter. Do you think that that resets the run rate level there and then you can grow off that base, or do you think that maybe it has to come down a bit because maybe you know you were able to ship some of those products this quarter and that’s why it went up. Just trying to understand, you know, the characteristic around that jump.
So yes and no, meaning that the OTN number I think it sets a new base level, and I expect the OTN number to keep growing from here. The data com piece had a little bit of catch up which was stuck in backlog last quarter because of supply constraints that we were able to catch up on a little bit more this quarter. So that one will come back a little bit. But in terms of order of magnitude, we’re probably seeing like $1 million to $1.5 million.
Yes I think to further on Bob’s comments, the OTN stuff is pretty much a new step and a new base. The data com stuff we are, believe it or not, since we’re so sole-sourced and we have one of the few interoperable multi-rate, multi-protocol physical layer devices, we can’t build enough to sell them, right, so still there’s a little bit of exchange between demand and booked forecast here. So that’s kind of where we are.
Okay, and then on the Titan products, when do you expect to ship production revenue. I know that that was originally supposed to be happening sometime in the second half of this calendar year. I don’t know if you can get more specific on that.
So we had said all along that we didn’t think it would start shipping for revenue until the December quarter. I think it’s still the December or March quarter roughly before we start to see revenue from that.
Okay. And then, on which of the parts, either by product line, or application, are you the most supply constrained at this point?
It’s fairly uniform, I mean it’s a really hard call to make because I can tell you that across our product lines and families today, Allan, trust me, if we could we would build out significant inventory across all of our product lines, across all of our application platforms. So it’s hard, that’s part of the reason we are actually indicating the sensitivity of supply to our business looking forward to you guys. Because it’s not just secularly involved in, okay, these two sets of products from TSMC or these three sets of products from IBM, right. I mean, fairly uniform.
And whenever you get supply constraints like we’ve got, it’s never as simple as take the wafers you’ve got and go maximize revenue, or take the wafers and go maximize gross margin. I mean we have to consciously take into account key account relationships, even if it means that we have to burn a whole boat of wafers on that particular account or that particular part for that particular account, knowing that part of those parts will just go into inventory for a quarter or two. But if it’s critical to the account, and we need to do that to support our customers, we will. So like the increase that we had in revenue quarter to quarter, that $2 million, was primarily related to excess wafers associated with critical demand that a customer wanted, that we had to go ahead and build even though it wouldn’t ship that quarter but it will ship over the next couple of quarters.
Okay. Thanks very much guys.
And your next question comes from the line of Christian Schwab with the Craig-Hallum Capital Group. Please proceed.
Christian Schwab – Craig-Hallum Capital Group
Great, thank you. Great quarter guys. Is the OTN absolute dollar amount big enough now that it sounds like since it was up 50% that you can break that out as an absolute revenue number? I mean, how much revenue exactly was OTN-related in the quarter?
Or, let’s put it this way, I guess I should say I’m not willing to break it out yet separately. It’s part of the transport business and I’m just going to report it that way without getting into more detail, at least for this fiscal year.
Great, and then as we look to your guidance of 25% to 28% revenue growth for the year, given the significant backlog, it kind of, in essence fair to assume you just kind of expect continued modest revenue growth sequentially as you remain, it looks to be chipped and strained, possibly as long as March?
Yeah, you do the math and it’s going to kind of imply a wimpy third and fourth quarter, being blunt…
Yep, kind of up 3%, give or take, 2% to 3% sequentially.
Yeah, and I’m not saying that we’re, we’re not trying to telegraph that demand is slowing down, we’re just trying to say that given the supply constraints we’re not exactly sure, and so the guidance we want to give is we expect to grow, but because of the supply constraints, we’re not willing to project out more than about the 90 days.
Great. Perfect. No other questions. Thanks.
[Operator instructions.] And your next question comes from the line of Sandy Harrison of Signal Hill. Please proceed.
Sandy Harrison – Signal Hill Group
The two products that you were talking about today, I guess specifically the Trivini and the 10-gig. Is that just going after cable replacement in data centers for the installed copper? Does it address the short reach? What are some of the benefits? Does it allow for lower power? Just kind of, outside of the interference which it sounds like you guys have addressed with it, sort of a little bit more on that family because it sounds like it’s a pretty interesting group.
Yes, so let me kind of tell you about that market. I mean for a long time, the standard has been alive for about five six years, this whole 10 gigabit over twisted pair, lots of people have announced lots of products. If you really look at the net amount of shipments and actually enabling platforms that are in production or in deployment, they’re very, very, very, very small. In fact I think it’s almost like 90% plus is fiber today at 10 gig in most of the data centers.
Copper is basically, has been primarily shut out because of the just poor, poor, poor reliability in the power. So today, in order to have an enabling technology you’re going to have to create something that is essentially interference-robust and very, very power efficient. So if you think about, you know, GigE and if you look at 10 GigE, we want to be within striking distance of GigE from a power perspective and striking distance of fiber, so that’s a big, that’s a huge, it’s almost a huge, huge, huge, huge barrier in terms of that piece.
More importantly, if you don’t have a reliable technology that, you know, emergency equipment, planes flying overhead of data centers, people, you know, guards with walkie-talkies and all, you know, the new 3G and 4G cellphones. They just tend to really corrupt the link and you just don’t get 10 Gb. So we have, we believe we have, right now arrived at a point where we’ve got enabling patent-pending technology that both takes care of the power per port problem and most importantly solves the reliability problem.
And from a range perspective it’s a 100-meter technology, which is what IEEE specifies, but we have modes to do things like 30 meters, 15 meters, and shorter ranges with appropriate power proportionality.
Yeah, Sandy this is Bob. The inside of the data center in many ways they prefer a copper connection because it’s more of a physical connection and if it snaps into place and locks it works good whereas the optical connectors sometimes they’ve got to make sure the light beams and everything are perfectly aligned and it’s a little bit complicated. And about two years ago when some of these 10 gig copper solutions started to appear, the first problem was that they were requiring 16, 17 watts or so to drive the cable, which was power prohibitive. Now the solutions are down in the 4 watts or sub-4 watt kind of category, and as they did that and started to go longer cable lengths, then this RF interference popped up so the ability to solve both is, I think key and if you believe the forecasts, ideally if you went out a few years it’s about 75% in the data center that would be copper and 25% would be optical in rough terms. So it’s a nice little market for us to go into here.
And sort of keeping it in the new product sphere, the new WDM platform you talked about where you’re integrating most if not all of your major products onto a platform, how are you going about marketing that? Is that going after replacing internal capabilities of your box customers? Is it hey, let us help you help yourselves? How is that being marketed? And then that’s a little different than typically what you’re seeing and how’s it being received by the customers?
So, really good question. I think the big thing there is value engineering. That’s what drives this, which is, I have seven skews of a card in a platform with multiple port densities, multiple control plane architectures. Can you guys engineer one skew for me that’s basically I can populate it or depopulate it with my IP based on your chips or with [unintelligible] that I have my IP in and I can intentionally blend the personality of those seven cards into one card that I can product manage as a one-card platform with different, essentially, population skews rather than seven or eight different actual line cards, right? So that’s the big, so it involves software on the control plane front, it involves essentially drivers that we supply from a BSP perspective for platforms such as PQX and Yahara, and PEMAQUID. It involves potentially firmware for PHYs that have both the back plane features, like the KR pieces, and the front plane pieces like your WDM control pieces. And so you’re talking about essentially a platform that sits in a box and the right way to think about it is a reference card design that has all of our stuff in it along with our customer’s IP, so it solves two pain points. It reduces their overhead in maintaining eight different variants, gives them a scalable platform that they can evolve over time with their IP and our IP, and most importantly, saves software effort, so that’s the big deal.
And then you talked a little bit about your capacity and your fab and who’s doing what and with the current profile of fab from TSM versus IBM. You know, a year from now what is your capacity breakout look like. Is IBM going to below 25% or are they going to remain a significant supplier for some time?
So, I see IBM being a significant supplier for the next 24 months at a minimum. I mean the Power PC stuff that we’ve got now, we have no intention of moving that to TSMC. We just plan on introducing new products that’ll be in TSMC, so we’ll be shipping stuff out of the IBM foundry for 4, 5, 6, 7 years still. It’s just over time, as the new products ramp, the two lines will cross, but honestly that crossover point is 3 years, 4 years out, somewhere in that timeframe.
Gotcha. All right, thanks for taking my questions guys.
And your next question comes from the line of Brian Thonn with Kingdom Ridge Capital. Please proceed.
Brian Thonn – Kingdom Ridge Capital
Paramesh, in your prepared remarks you talked about the OTN rollout accelerating in North America and AP and maybe how Europe lags. Can you just kind of talk about some of those dynamics, both regionally and if you think about Core, to Metro, and then to the Edge? And then also kind of give an aggregate estimate on like, what [unintelligible] in terms of the deployment of OTN.
Sure, I think, let me start by answering the first part of the question. If you look at most Metro area and if you look, let’s take one step back. Look at new data plans in the world, right, for mobile phones. And the data plans for mobile phones directly scale with business and GDP output. So if you take that and if you look at the highest rate of change of GDP output and business output, the two geographies that are changing the most are basically, are producing the most, are United States, roughly, and Asia, in terms of net rate of change. In terms of data plans that’s related to, again, businesses and enterprises. This is very well correlated, guys like Cisco on the data center and enterprise side and guys like Ciena and Nortel on the optical side. If you look at both their products and where they sell them those are the geographies that are the hot geographies for them.
Now, from a service provisioning perspective, there is a huge transition, and I will say a sort of interesting set of dynamics that is going to occur over the next probably three to five years which is how do you get service providers to be able to dynamically add bandwidth on demand in incremental and meaningful chunks? And that aggregate bandwidth addition has to be cleanly done, and done in a swift and nonintrusive fashion. So any bandwidth addition infrastructure upgrades are all begin done via OTN and I’ll give you an example. If you’re going to add things where you’re going to, next you’re going to add base station capacity for 3G or LTE, most geographies today who are doing that are all moving to an ODU pipe, which is a GigE pipe that essentially gets added to a large metro box. So it’s no more, oh I need a DS3 or T1, right? It’s basically a GigE pipe.
So geographies that are dynamically adding bandwidth for mobile users and geographies that are adding bandwidth for entertainment are all moving to changing their existing equipment to be more OTN-centric in the Metro and in the Core. And pursuant to that, they’re going to do the following, if that’s the way we look at this. The next big build out is going to be the Metro Edge piece. And the Metro Edge piece as I indicated before is going to be everything from panoptical devices all the way to base stations, all the way to large demigod boxes that sit on the edge, on the Metro Edge.
So that’s the way we see the transition occurring, and the right way, if you want to look at proxies for that, is to look at where the new Metro Edge platforms from Nortel and Ciena are going, where the new ASR and TSR families from Cisco are going. They’ll give you polar views of the way networks are being upgraded going forward. So keeping [unintelligible] any new and fractional service additions in the Metro are going to be only done with OTN. Any new or fractional services that are going to be deployed in the Core are going to be done be OTN. All the Edge, Metro Edge services will evolve later in time.
I don’t know that I answered your question. I tried to paint some color though.
That’s okay. And just kind of the aggregate what inning are we in, or how do we think about the inflection?
Okay, so I think that we’re, I think as I said earlier on, I think the Core is probably about 50% to 60% built out in most of these geographies. And I think over the next two years you will start to see the Metro starting to evolve, both the Metro, the Core transition, and the Metro to Edge transition. In two to three years. So I will say, if you look at Core as being innings one we’re at least 60% of the way through innings one, and if you look at innings two being the Metro to Core, we’re probably 20% of the way through Metro to Core. We won’t even get to Metro to Edge for the next two years.
Got it. Okay, thank you very much.
With no further questions in queue, I would now like to turn the call back to Bob Gargus for closing remarks.
Thank you. We’d like to thank all of you 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 calling 888-286-8010 and entering the reservation number 65996548. We will also file a copy of this script in an 8K with the SEC in the next few days. Please feel free to call me if you have any additional questions, and again thank you for your participation on the call today and have a nice evening.