Inphi Corporation (NYSE:IPHI)
Q1 2016 Earnings Conference Call
April 28, 2016 05:00 PM ET
Deb Stapleton - Investor Relations
Ford Tamer - President, Chief Executive Officer
John Edmunds - Chief Financial Officer and Vice President
Ross Seymore - Deutsche Bank
Tore Svanberg - Stifel Nicolaus
Brian Alger - ROTH Capital Partners
Dave Kang - B. Riley
Joe Moore - Morgan Stanley
Richard Shannon - Craig-Hallum
Quinn Bolton - Needham & Company
Good day, ladies and gentlemen, and welcome to Inphi's First Quarter 2016 Earnings Conference Call. At this time, all participants are in a listen-only mode. Later, we will conduct a question-and-answer session and instructions will be given at that time. As a reminder, this call is being recorded. I would now like to introduce your host for today's conference, Deborah Stapleton. Please go ahead, ma'am.
Thank you and good afternoon, everyone. We appreciate you joining us today to discuss the financial results for the Inphi's first quarter of 2016. I'm Deb Stapleton, and with me today are John Edmunds, Chief Financial Officer and Ford Tamer, our Chief Executive Officer. John will begin with the Safe Harbor, then Ford will give you an overview of our business. After that, John will provide a financial summary of Q1 and the outlook for the second quarter of 2016. Then we'll be happy to take your questions. John?
Thank you, Deborah. Please note that during the course of this conference call, we may make projections or other forward-looking statements. These forward-looking statements and all other statements made on this call, which are not historical facts, are subject to a number of risks and uncertainties that may cause actual results to differ materially. These forward-looking statements speak only as of today's call. We do not undertake any obligation to provide updates after this conference call.
For further information regarding risk factors for our business, please refer to our registration statements as well as our most recent annual and quarterly reports on Forms 10-K and 10-Q, all filed with the Securities and Exchange Commission, accessible at www.sec.gov. Please refer, in particular, to the sections entitled Risk Factors. We encourage you to read these documents.
Also, during the course of this conference call, we may make reference to non-GAAP financial information. A reconciliation of this information is included in the press release and on our website, which is available at www.inphi.com. This information is not a substitute for GAAP and should only be used to evaluate the company's results in conjunction with corresponding GAAP measures.
Now, to begin our review of the quarter. Let me turn the call over to our CEO, Ford Tamer. Ford?
Thanks, John. Good afternoon. Thank you for joining us for our first quarter earnings call of 2016. Once again, we are delighted to report that for this quarter and for the previous 16 quarters in a row, we delivered strong performance at or above our forecast.
Our revenue in Q1 was $66.5 million. Gross margins were 68.6% and we earned $0.27 per share. This represents 12% and 17% year-over-year growth in revenue and EPS respectively and $0.01 better than street expectations for Q1, 2016.
As I enter my fifth year as CEO of Inphi. I could not be more proud of our consistent deliver at or above expectations. Let me dive right into the product news for the quarter and John will give you the financial highlight. We've made a great deal of progress and had many announcements this past quarter, both in advance off and during the Optical Fiber Conference or OFC in Anaheim.
I will start with the announcements impacting the first of our three major product lines our optical interconnect focused on long-haul and metro customers. The more significant announcement during the quarter, with the introduction of our ColorZ reference design. The industry's first Silicon Photonics 100 gigabit per second PAM4 platform for data center interconnect.
This solution will enable standard [ph] operators to deliver data between switches located 80 kilometers away from each other in different data centers allowing them to operate as one large virtual data center. Our brand name ColorZ is spelled as Color with a Z at the end but the Z is pronounced a plural for Color, so ColorZ.
The Z also stands for ZR, which refers to the 80 kilometer distance specification. ColorZ was Inphi's first announcement of the successful development of our highly integrated Silicon Photonics technology. This platform will, for the first time there are customers cost effectively upgrade to 100 Gigabit per second inter-data center connecting using a low power industry standard QSFP28 form factor.
It would help our customers achieve up to 3.6 terabit per second front panel switch density per 1 RU and fit in the low power envelope of the QSFP28 module. ColorZ in the space and power optimized form factor will speed up the adoption of end-to-end 100 Gigabit per second connectivity between cloud data centers consistent with our vision that the cloud is the network.
We are optimistic about the adoption of ColorZ in the month and quarters ahead. And we are on track for delivery of production units to our first customers. On the OFC, we introduced our new Surface Mount Technology or SMT quad linear driver extending our product portfolio for Metro offerings.
Our new dual and single-carrier 45 gigabyte SMT quad linear drivers and transimpedance amplifiers will offer lower cost solutions for 200 and 400 gigabit per second line cards. In the small form factors, these lower SMT drivers enable long-haul and metro network operators to deliver next generation high volume applications with their highly integrated line card designs and advance pluggable modules.
We are excited to be shipping these parts in production volumes this quarter. And there was more. On March 7, we announced that we've shipped more than 1 million 100 gigabit per second coherent amplifiers and linear drivers. I could not be more proud of our key members, who contributed to our attainment of this important milestone.
As a reminder, we first deployed our devices for 100 gigabit long-haul coherent systems in 2010, our 200 gigabit metro coherent systems in 2014 and then our first 100 gigabit pluggable modules in 2015; proof that we have been and continue to be in the right place, with the right product, at the right time, and we're just getting started, which brings me to our second major product line, our networking interconnect business. As you know from previous calls, we have been confident that PAM4 would become a leading protocol as the industry evolves to high speed physical layer standards.
While we're [indiscernible] phase and ready to show at OFC, we announced that Microsoft data centers are planning to install ColorZ modules that incorporate 100 gigabit PAM4 later this year. In addition, five other companies at OFC were demonstrating PAM4 based modules for single-mode and multi-mode fibers for 40, 100 and 400 gigabit per second applications.
These solutions would go into production in late 2016 or in the first half of 2017. We remain optimistic about further PAM4 penetration as the market evolves. In March, we also announced the immediate availability of our second generation highly integrated, lowest power PAM4 chipset solutions, for intra-data center and inter-data center cloud interconnects. These physical layer solutions require 25% less power than their predecessors. They include highly configurable DSP for diagnostics and control.
While it was only last year, we introduced the first generation and pleased that we've already production availability of our lowest power in the industry, second generation PAM4 chipset. To complement our PAM4 solution, we also announced the production availability of the industry lowest power clock and data recovery or CDR, 100 gigabit re-timer for module applications.
Designed for multi-rate operations of 10, 25 and 28 gigabit per second, our lower power CDR delivers industry leading performance with a total power of only 370 milliwatts or about 92 milliwatts per channel. These CDR products are being designed into high density CFP2, CFP4 and QSFP28 module applications.
Finally, let's turn to our third major product line our memory interconnects between CPU and DRAM. In Q1, the revenue grew and now remains at about 20% of our overall business. We're making progress in DDR4. We expect DDR3 to be declining quarter-over-quarter for the balance of the year offset with DDR4 growing quarter-over-quarter with Intel Broadwell. While this may cause fluctuations from quarter-to-quarter, there would be offset by strength in our core communications business. We expect to continue to make progress in DDR4, with Intel Broadwell in 2016 and Intel's Skylake in 2017. So while Q1 was busy and productive quarter for us. No one at Inphi is resting.
At the risk of reputation, the exponential ramp and bandwidth is an ongoing inevitable trend and an enormous opportunity for Inphi. With announcements like these in Q1, we continue to demonstrate that we're strategically positioned both for today and for the years ahead across and within our product segments.
Now let me turn the call over to John to tell you about the details of Q1 and our outlook for Q2. John?
Thanks, Ford. Now let me recap the financial results. As Ford mentioned in the first quarter of 2016 Inphi reported revenues to $66.5 million, which was up sequentially from Q4 by 3.3% and year-over-year by approximately 12%.
Core communications was up approximately 41% year-over-year. We continued to estimate core communications revenue year-over-year growth rate of between 40% and 50% for 2016. We define our core communications as amplifiers, drivers as well as 10, 40 and 100 gig physical interface products, all of which serve the service provider and data center interconnect markets.
Our communications business continued to represent roughly 80% of our total revenues through Q1. The core communications products continue to represent approximately 75% of communications, while the legacy communications products represented the remaining approximately 25%. As expected, seasonality and annual price reset with some customers in transport and legacy caused the revenue to trend down in the first quarter.
The forecast for the balance of the year is expected to be relatively flat, with small quarterly declines. Sales of high speed memory were up in the quarter and remained at about 20% of the business in Q1. Gross margins on a non-GAAP basis came in at 68.8% for about where we expected, based on slightly less favorable mix than Q4 and manufacturing year-end favorable variances from supplier contracts that reset at the beginning of the year. Based on the current forecast of mix for Q2, we expect the gross margins in Q2 to be in a range of 68% to 68.5%. This is down about 50 basis points from Q1.
With a mix of new products coming in the second half and depending on in initial manufacturing yields, we currently estimate the growth margins in Q3, may improve by approximately 50 to 100 basis points from the Q2 level.
On a GAAP basis, we had GAAP positive operating income in Q1, at $2.5 million as compared to a $2.6 million operating income in Q4 and compared to a $5.5 million operating loss in Q1, one-year ago. The delta between GAAP and non-GAAP operating income is approximately $11.2 million, made up of several basic recurring adjustments.
As discussed previously, the GAAP measure includes stock compensating expense of $7 million. Purchase accounting related adjustments totaling $3.5 million as well as a final amortization of acquisition related retention bonuses of $700,000. This all nets down to $11.2 million of additional operating experience in the GAAP books.
Finally, we also have $2.4 million of non-cash debt cost amortization associated with the convertible bond issue. The related tax effects of all of this additional GAAP expenses totals to about $2.1 million additional tax benefit. The Q1 GAAP result was net income of $200,000 or $0.01 per share, which compares to a net loss of $9.7 million or $0.26 per share in Q1 of 2015.
The Q1, 2015 GAAP net income was more negative mainly as a result of purchase accounting adjustments primarily the $6.2 million amortization of the step up in inventory value related to the Q4, 2014 acquired Cortina inventory.
In addition in Q1, 2016 we also early adopted accounting standards update 2016-9, the change is accounting for tax benefit associated with stock compensation. In short, additional tax benefits from appreciation in previously granted stock-based compensation can now be benefited to the GAAP P&L rather than both as an additional credit to be paid in capital as was done previously, as the so-called windfall tax benefit.
This change will more closely aligned GAAP and non-GAAP tax revision and should allow us to more consistently show GAAP profit going forward. Although, we'll have no effect on non-GAAP net income, we estimate the effect to be roughly $5 million benefit per year, to forecasted GAAP net income in 2016.
The downside is that we have fewer unrecognized tax benefits that can theoretically be used to buy back shares under the treasury stock method resulting in approximately 900,000 additional weighted average shares outstanding and the denominator of our EPS calculations. We estimate, that this is costing us about $0.01 in non-GAAP EPS in Q1 and will cost us some additional $0.01 spread through the balance of 2016.
Since we have to adopt the change in 2017 anyway, we decided the improvement and aligning GAAP to non-GAAP reporting in the near-term was worth going ahead to early adopt.
Now for future clarification and analysis of the operating results, let's turn to some additional non-GAAP measures in comparison. On a non-GAAP basis, net income for the first quarter of 2016 was $11.7 million or $0.27 per diluted share. And it was up in consistent with non-GAAP net income of $9.3 million or $0.23 per share in Q1, 2015. Non-GAAP operating expense for the quarter totaled $32.1 million. This was up $200,000 from the previous quarter. While the product, which was the product of several offsetting changes, primarily driven by higher Q1, payroll taxes.
Overall in Q1, we were able to deliver non-GAAP operating margin of 20.6%, which was lower than the 21.8% in Q4. But the difference only represents about $350,000 change. Mostly, a $200,000 increase in operating expenses and $150,000 in lower aggregate gross margin. In Q2, with higher revenues and slightly higher operating expense we anticipate an improvement in non-GAAP operating margin which will bring it into the range of 20.5% to 21.5%.
GAAP interest expense of $2.487 million when reduced by non-cash debt cost amortization of $2.378 million gives you non-GAAP interest expense for the quarter of $109,000. Included in this number is accrued cash expense for the convertible debt of approximately $647,000. Less interest income of $478,000 for a net carry associated with the convertible debt of $169,000.
The delta event to $109,000 in expense is comprised with some other miscellaneous income. With regard to the Q1, non-GAAP tax provision. We're currently using a non-GAAP rate of 14% and based on current forecast, this would be used for the balance of the year.
Now turning to the balance sheet, overall cash was up $1.5 million in the quarter from $326.7 million to $328.2 million. We continued to have strong cash flow from operations in Q1, at $13.1 million, which was similar to the $13.5 million from one-year ago. Q1 cash flow from operations of $13.1 million was down however from the comparable amount in Q4 of 2015 of $25.8 million.
The $12.7 million decline is largely due to a net expansion of working capital or use of cash in Q1 of about $4 million compared to a contraction of working capital or source of cash in Q4 of about $14 million. In addition, the Q1 P&L before changes in working capital was generating about $4 million more cash flow than Q4.
These three pieces that accounts for the majority of $12.7 million delta. The company also had Q1 capital expenditures of $7 million, which was up from the $3.4 million one-year ago and up from $4.4 million in Q4, due in part to additional expenditures to support our new ColorZ product development.
The net then allowed us to generated $6.1 million of free cash flow in Q1, which was down from the $10.1 million in Q1 one-year ago and down from the $21.4 million in free cash flow, we generated in Q4. Again the delta Q4 is based on the Q1 expansion of working capital rather than the Q4 contraction.
We continue to expect approximately $5 million of cash expenditures in Q2 of 2016, again driven in part by the need to invest in the ramp of our new ColorZ products. In addition, the company invested approximately $2 million cost-based investment in a private company and we made about $2 million net use of funds in the finance inspection driven mainly by paying tax withholding on restricted stocks [indiscernible].
Accounts receivable came in at $33 million or 45 day sales outstanding in March compared to $30.4 million or 43 day sales outstanding at the end of December. Inventory on the balance sheet at the end of March was $17.3 million, which did not have any acquisition related step up in inventory value. As compared to $17.8 million at the end of December, which included $200,000 of step up in inventory value.
Excluding the Q4 step up, inventory days were down to 75 days or 4.8 turns at the end of March compared to 86 days, were about 4.2 turns at the end of December. So the inventory balance is down, we continue to work on being efficient with our inventory, while also having enough inventory available to respond to demand.
Payables increased to $11.4 million or 50 days in March from $8.4 million or 41 days outstanding at the end of December. Due largely to the timing of inventory received in the quarter. Now let me recap the outlook for Q2.
I remind everyone again that the following statements are based on current expectations as of today and include forward-looking statements, actual results may differ materially. We do not plan to update nor do we take on any obligation to update this outlook in the future. Our forecast of Q2 revenues is up, in a range of approximately 2.8% to 5.8% or $68.4 million to $70.4 million.
We expect non-GAAP gross margins to be in the range of 68% to 68.5%. We expect non-GAAP operating expense to be in the range of $32.5 million to $33.1 million. We are currently estimating the non-GAAP effective tax rate to be 14% for 2016. We are confident these components should then align, resulting in a non-GAAP operating margin above 21%.
This should also lead to non-GAAP net income of between $11.9 million and $12.8 million. Which on approximately $44.2 million estimated diluted shares would result in estimated non-GAAP earnings of between $0.27 and $0.29 per diluted share. We also estimate stock-based compensation expense to be between $8.7 million and $8.9 million.
In addition, we expect about $3.2 million from purchase accounting and we expect $2.4 million in non-cash amortization of debt expense. Finally, these increases should also generate related tax benefits of approximately $2.4 million. This would imply, a GAAP result in the range of $100,000 loss to a positive $800,000 of net income.
GAAP net income per share would then be in the range of a breakeven to $0.02 per share. A more complete reconciliation of Q2, GAAP net income forecast is attached to the last page of the press release. We will not update this outlook during the quarter until the time of the next quarterly earnings release unless Inphi publishes and notice stating otherwise.
So please ask any questions you may have today, during the general Q&A period and now we'll be happy to take your questions. Soniya?
[Operator Instructions] and our first question comes from Ross Seymore from Deutsche Bank. Your line is now open.
Congrats on the strong quarter and guide. I guess one accounting question first, before some more longer term technology questions. John, in the guidance that you said, the tax impacts without hitting into too much of the minutia. Is that one point hit to your or $0.01 hit to your EPS guidance for June as well as it was in March or is that $0.01 kind of spread over the final three quarters of the year?
Ross, I expect the spread over - probably the final two quarters of the year, given the guidance, were not given. So it doesn't affect the guidance and it basically, I think will affect rounding in Q3 and potentially in Q4.
Okay and then, switching beyond that to some longer term dynamics. Ford, obviously a lot of excitement about the ColorZ when and the Microsoft adoption etc. Can you talk a little bit about your confidence in the numbers that you gave for that, I think $8 million in the fourth quarter is what you said before and I remember right about $30 million next year? Talk about your confidence in that and is there any aspect of that because you expect other customers to start ramping that ColorZ product line as we get to 2017?
Ross, you're referring to $8 million in the second half of this year and $30 million in 2017 calendar year?
We're on track and as far as other customers. There's strong interest from other customers in the solution. We're not at liberty at this point to discuss, a name. But, the ability of deploying this big bandwidth between data center 80 kilometers apart and creating this virtual data center is showing a strong interest from other cloud operators as well.
Would you expect any of those to be able to turn it around and actually ramp in 2017 or is that an item that there's too many variable to knowing that strong interest, as you described it with turning to revenues?
We should be able to have a second customer in 2017 and maybe more, but I think we'll take a step at a time. We're very focused right now on delivering the units into production to our first customer and ramping that and taking a step at a time. We're under evaluation with others right now.
I think as my last question, quickly be on the memory side of things, that roughly 20%. I know you've talked in the past DDR4 goes up and DDR3 goes down. I think last I remembered, you said that'll be relatively flat for the memory portion in this calendar year. As long as that's still the case and then given the Broadwell and Skylake in DDR4 traction, would you expect 2017 to return to being a growth year from your memory business?
Yes, Ross this John. I think, right now we had a good quarter of growth in Q1 and we're just taking the balance of the year quarter-by-quarter as Ford alluded to in his comments. We anticipated decline in DDR3 and we do and we're working toward growth in DDR4, which we think we're making progress with.
So for us, it's just about kind of moving through. We definitely helped to have growth in 2017 beyond, what we're achieving in 2016.
Great. Congrats, again.
Thank you. And our next question comes from Vivek Arya from Bank of America Merrill Lynch. Your line is now open.
Hi, this is Shankar [ph] on behalf of Vivek. I've a question on the PAM4 plus DSV. Could you give us a status update on where that product is in terms of qualification and when is it ramping and what kind of revenue profile can we see for this year and then next year for that product line?
Thanks, Shankar [ph]. The product is on track to go to product in the third quarter. We expect the first revenue for 100 gigabit per second to come from our ColorZ reference design that will also be shipping in the third quarter. Beyond that, we expect 40 gig to go first followed by both 100 and 400 gig, as I said we had a few customers demonstrating the solution at OFC. And so we expect further revenue in the fourth quarter and then ramping further into 2017.
Got it. And just one follow-up on the memory side. So, how much of benefit are you getting from Broadwell? Because on the prior call you had mentioned that LRDIMM adoption is better and then the 8 gigabit DRAM started ramping. Is that still the case as you move along with Broadwell? Are you seeing traction in LRDIMM side as well?
So, our register share is higher in Broadwell than it used to Haswell. And we expect our share to increase into Skylake compared to Broadwell. We're still, we're not a big player in buffer. So the impact of this 4 gigabit to 8 gigabit on us is a minimal impact.
Got it. And one follow-up on the margin side. You mentioned the gross margin can go 1,500 bps in Q3. Is that driven by the ColorZ or is that anything else that's driving the margins up?
Actually, we have a number of new products. Amplifiers, drivers that we'll be shipping in the second half in increasing volume. So it's a combination of all the new products including ColorZ.
Thank you. And our next question comes from Tore Svanberg from Stifel. Your line is now open.
Ford, can you talk a little bit about what you learned from OFC on the competitive landscape especially in relation to PAM? Where are things sort of as you expected or any color you could share with us will be great? Thanks.
Thanks, Tore. First Tore, as you know we are - we have both PAM and as well as CDR. So maybe first I'll talk about our CDR. We are pretty excited about the CDR being the lowest power CDR in the market. In the press release about our CDR, we had a customer quote [ph] in there. That shows traction with that solution. And so, we do expect our CDR to generate revenue and significant revenue later this year, into next year.
Moving on to PAM. I think is as expected, we had different vendors announcing single LAN [ph] 100 gigabit per second PAM. We do believe that we're still correct in our timeframe, which is short-term. What we'll be shipping first into 2016 and 2017 really is going to be the two LAN [ph] 100 gigabit PAM, which is what we have and what another major competitor has in the market.
Beyond this, [indiscernible] objectives for 50 gigabit, 100 gigabit, 200 gigabit and 400 gigabit that again mostly used this two LAN [ph] 100 gigabit PAM, which is again what we have and this other competitor has, which gives us a clear lead. Especially given this new second generation product that we believe is the lowest power in the market.
So we're still very positive about what PAM4 could do for us both inside the data center and between data centers.
Very would. Would you be able to share with us, if you're benefiting from any specific metro deployment this year? I mean, I know your business is broad based, but you know if you could point anything that would be helpful?
Yes, Tore. So we, first we started getting orders for CFP2-ACO, which is the analog CFP2. And so that's a proxy for that metro ramp starting for us in the module side of the business. On the driver side, we have made some very significant progress. We had a very strong growth in Q1, after a spectacular growth in market share gain in 2015. The growth is driven by continued strength in the gold box, but also while us introducing this SMT driver.
And, so we believe we're the first vendor or the only vendor today that successfully released two volume production coherent [indiscernible] driver and free form factors namely the gold box, the SMT or surface-mount technology package and the single-chip solution. All optimally to meet and exceed the different requirements for performance prior cost to different segment. So ultra-long-haul, long-haul, [indiscernible] metro and DCI. It's hard for us to tell you exactly the breakdown of long-haul versus metro versus DCI because it's the same product that we ship in all three spaces.
But we do believe the significant piece of this is, metro system they are shipping along with our amplifiers and driver. So we believe we're well on track for both the module as well as the system. The one last thing I would say on the module, is that we're also are designed into CFP DCO digital equivalent. The DCO has about a year head start on the CFP2 ACO. And I think for us it will be hard to say, what the spread is, DCO versus ACO?
DCO has strongly favoured in China and that's a large market. CFP2 ACO are starting and again, we're neutral. We're supporting the different vendors in that space and so - believe metro is on track to grow nicely for us in quarter-on-quarter for the remainder of 2016.
That's very helpful, Ford. Just one last question for John. John, you talked about the accounting impacting the share count for the June quarter or I mean, March and June. Is there any impact after that or is the sort of the new run rate and we'll just model is from there?
No, it's basically the new run rate, Tore. That's why it doesn't impact the Q2 guidance that we gave because we already built it into the guidance. So it's really just affecting around in mostly I think in Q3.
Okay, very good. Great jobs guys. Thank you so much.
Thank you. And our next question comes from Brian Alger from ROTH Capital Partners. Your line is now open.
I want to follow-up, the products. Ford, you just mentioned the low power CRD, that's an interesting design. One of the conversations I had with CEO was that, you guys were developing that on CMOS, which I thought was a little different from the competition. Can you maybe talk about the benefits not just from a power standpoint, but also potentially from cost of working in components on CMOS?
I guess where I'm going with this, I understand you also have amplifiers running on silicon instead of maybe the more exotic substrates. What advantages and disadvantages did you have in that as, you know NRZ platforms, seem to getting really price competitive?
Thanks, Brian that's a good question. I think we've always said, that having both the 35 and the platform technology expertise and scales and offerings would help because the technology start on the long-haul and metro on the 35 material. And then, as it migrates to inside the data center CMOS is favored because of lower power and high volume ability to scale up at large fabs such as TSMC.
So, yes, at OFC we did announce the lowest power CDR targeted 400 gigabit NRZ application, as you mentioned. At 370 milliwatt per part that CDR is a 40% improvement in power over the closest competitor we believe and as you mentioned because it's an CMOS process. It would have cost advantages as users of the mainstream high volume CMOS TSMC process. Did I answer the question or was there another part of the question that?
That was a lot of it. I guess, in the CDR. Obviously you're giving success there and we have that design announced and I was just curious with other components that you guys are targeting specifically amplifiers and we know of course that the Silicon Photonics that you're using to ColorZ to use that substrate as well. I'm just trying to understand how much of an advantage, the expertise in CMOS may actually play out.
As everyone seems to be chasing the data center market and it becomes more.
Got it. Yes, thank you. So, on the second part of your question. Yes, this should overtime make integration possible, to get ColorZ to market. We took a more step-by-step approach where the parts are not integrated today. But you're right, the fact that all the parts are in CMOS would make it, possible to integrate in the future and that potentially give us more power and cost advantages absolutely.
Okay, great. Nice job. Thanks guys.
Thank you. And our next question comes from Dave Kang from B. Riley. Your line is now open.
First question regarding ColorZ and Microsoft. Some say that Microsoft could hold back on ColorZ until they find the second source, your comment on that?
We haven't heard that from them, you should ask them directly, but that's not the plan as we understand it. You go ahead, Dave. You had another question.
Yes, sure and secondly, sticking with ColorZ and Microsoft. Some clients [ph] also said that ColorZ is only suitable for Microsoft type of architecture, what's your comment on that one?
Well, if you're implying that Microsoft is only data center, worried about power and cost and space. It could be, but last time I checked every cloud data center is extremely worried about power cost and space. So again, ColorZ is a very optimized solution for 80 kilometer. It's very complementary to a coherent solution. What ColorZ really announced the cloud data center to do is to connect data center to our 80 kilometers apart, as if they're one giant virtual data center.
And if you were to be able to, you couldn't actually achieve the advantages we're talking about the 3.6 terabyte front panel switch density. The low power QSFP28. The 80 kilometer reach, the cost effectiveness with other solution. So this is a very optimized solution focused on upgrading what today is 10 gigabit, DWDM to 100 gigabit DWDM which is what ColorZ provide.
It's also focused on being on complementary coherent and we believe, and online it will actually create more bandwidth that will eventually help coherent volumes grow. So very complementary solution, very focused on 80 kilometers and for that 80 kilometer distance it's the most power efficient, most space efficient, most cost effective solution on the market.
Has anybody done some kind of study on cost differential or TCO differential between PAM4 or ColorZ versus coherent DWDM because I've heard from other industry executives seems like 40 kilometers to 80 kilometers, your ColorZ versus Coherent could really run into each other?
Again, I mean we do not believe they'll run into each other. We believe that the complementary. Obviously, large data center like Microsoft would not have gone forward with that extensive research and you could probably get that research from them, if you ask this there, to share not ours. Others are currently doing that same research and we believe that there will be strong interest and we should be able to be on track to announce another partner towards end of the year, early next year.
Got it, thank you. And then, just wondering can you break out your mix between 40 gig versus 100 gig?
Dave, we don't ship very much 40 gig at all. We do ship probably 10%, 15% of our communications revenue is going into 10 gig that's using multiples of 10 gig for 40 gig applications, but that's really the 10 gig series technology that we acquired from Cortina. That's about all we do in the 10 or 40 gig space.
Yes, makes sense. And lastly, actually a couple more. One of your competitors talked about having strong backlog from number of ACO customers. I think you may have touched upon it, but can you just remind me whether you have a similar type of strong backlog from ACO customers and how many ACO customers are we talking about here?
So, Dave we do have a fairly healthy backlog. We typically don't quote specific backlog. But we do have a good backlog in that area, in particular because we're in virtually all of the merchant designs for ACO, CFP2 for our amplifier technology. Go ahead.
Oh and then, I think before you talked about DCO for China, is that happening now or is it more of a second half event?
DCO is happening.
It's happening now, got it. All right, that was it from me. Thank you.
Thank you. And our next question comes from Joe Moore from Morgan Stanley. Your line is now open.
I wonder if you could just talk a little bit about the general service provider carrier spending environment, if you guys are still seeing strength in China and what you're seeing globally there.
Thanks, Joe. This is Ford. We - no changes till, on the same trends that we've seen in the past. We indicated on the past quarterly call, still seeing strengths for now.
Okay, great. And then with regards to the legacy business kind of being flat lining the rest of the year, I guess, if I look back historically it seems like it's always been down a little bit. Is there a reason why it kind of maybe stepped on little bit more in Q1 and then flattens out? Or is there any risk as we go through the year, with that volume further?
Yes, historically this business is down in the first quarter for a couple of reasons. One is, just seasonality because if you look at equipment vendors who want to make their year-end at the end of March, they have to have their chip technology in-house in the fourth quarter. So the fourth quarter has usually got some momentum built into it because of that and then. You do have some seasonality in the first quarter in part because of some of the Asian customers and Chinese New Year in the first quarter.
So that in conjunction with price renegotiation for certain customers that also effect this industry for in the first quarter and to bring their revenues down a little harder in the first quarter and then we'll have, as we had, as we suggested relative flatness through the balance of the year.
Got it, thank you very much.
Thank you. And our next question comes from Richard Shannon from Craig-Hallum. Your line is now open.
Probably just a few from me. I guess first on your 45 gigabit products the amps and drivers. You talked about this, I think last quarter as well. Just to be clear, is that product specifically intended for 200 gig and above or is that also, used for 100 gig products for your customers?
It's intended for mostly for higher rates.
Higher rates, okay. I know there's some systems shipping out there with 200 gig today. Is that revenue stream from the amps and drivers together for 45 gig, is that anywhere near the size of your 28 gigabit products or is that still quite a difference there?
Not booking it up. We don't break up by which gigabit, we haven't broken these in the past.
I think you can assume in general, since we just recently introduced the 45 gigabit that we're still shipping much more 32 gigabit than we would 45 so far.
Okay, so that's kind of what I thought. Thanks for that. Maybe just couple quick ones more from me. On the CDR product. Sounds like you're pretty excited about that ramping here in early part of the year. Any sense to what kind of share you might be able to achieve with that product by the end of the year or timeframe you'd like to discuss?
Again, we don't really break up the CDR versus the PAM, versus our Gearbox. I mean, we already an existing CDR that is shipping today, 100 gigabit. That's a new CDR the similar power CDR. If you look at that business, the networking business we got our first generation Gearbox that was focused on CFP module, then we took a second generation Gearbox that was for LAN cards. We get another first generation CDR that was for CFP. We get this lowered power CDR that's for CFP24 and CFP28.
And we got these Cortina 5's [ph] for 10 and 40 gigs. So I mean, would all these products comprise in addition to obviously PAM4, 40, 100, 50 to 100, 400. So all of these would comprise our networking business. And we don't break up the different components, Richard.
Okay that's fair enough. Maybe just one last question on the memory side of your business. What's your outlook on qualifying on the buffer side there? And you're working with all three of the guys, all three of the main memory module guys or you're going to give us any thought and clarity on how that's progressing?
So the buffer side is really a focus right now on Skylake. The buffer for Haswell. We're going to have some shift, but again most of our business for Broadwell would be register based, Richard. We've always been consistent that 2016 would be mostly register based, there's very little buffer in 2016. So same as we said in the past. We haven't changed that, so.
Okay, fair enough. That's all the questions from me. Thank you very much guys.
Thank you. And our next question comes from Quinn Bolton from Needham & Company. Your line is now open.
I apologize, I missed the prepared comments. But wanted to just touch base on couple of questions. First on the ColorZ, it looks like that it requires some purely engineered links with amplifiers and just version composition modules and it seems like Microsoft may own their fiber, which makes it easy to engineer that link. Kind of, did you look at other prospective customers. How important is it for those other customers to also own the fiber and if they don't own the fiber, is that something that they can, can they get those highly engineered links through their fiber provider?
Quinn, you're right. This is a solution that's for the more sophisticated customer. It's not going to be a solution that a direct enterprise customer would adopt. It's a solution that's probably focused on the mega cloud data centers or systems OEMs that would work with them to provide those engineered links, as you suggest.
Microsoft obviously had a very competent network engineering staff that has done all their work. The difference now that providers who are currently evaluating ColorZ have a similar type of capability and we expect them to be able to, to do what Microsoft has done and then, for the folks that do not have the capability in-house or working through a third-party or system OEMs, who would then come in with a solution and help those customers role this up.
Got it, great. Okay. Second question, just coming out of OFC. It sounds like many of the your peers in the optical and component side saw pretty good order trends and better visibility sort of going out of OFC and just wondering, if you look back or if you look today, your visibility relative to say coming out of last quarter. Will you say visibility is improved or is it generally the same as it was 90 days ago?
So if you recall Quinn, we had a bit of pause in like the Q2, Q3 of last year then Q4 - we went back to become positive and so I would say, we're the same as Q4, no different still strong and moving forward.
Thanks. For just your comment about the DCO in China, kind of wondering is that a CFP form factor. I guess I was surprised to hear that, China's already moved on to the DCO given that the CFP2 ACO's are just now starting to ship and almost sample volumes. I mean, we're not even in main production, had on ACO's. What kind of DCO form factor are we talking about in that china market?
Well the only DCO form factor available in the market today is in CFP form factor. So we're talking about CFP, DCO and that has been shipping about a year ahead of the CFP2 ACO. And you should think about the volume today on the CFP DCO being about 2X what the volume are today on CFP2 ACO.
Now time will tell, which one will when, it's going to be hard to say, how it will split longer term. We do not expect the DCO folks to stand still. We do expect CFP2 DCO to show up and you should expect us to support all the form factors. So we're not going to pick sides here. We're supporting CFP DCO. We're supporting CFP ACO. We expect to support CFP2 DCO.
Got it, great. And then just lastly on the buffer side. It sounds like, you're sort of really targeting the Skylake generation with the current generation of buffers. Given that you're sampling this probably 12 to perhaps 15 months ahead of the Skylake introduction. Do you think, you'll have to end up re-spinning that part before Skylake ramps? Or do you think, that this is sort of production worthy and you guys are just not far ahead of you know the next Skylake generation of servers?
So if you look at our business so far on the memory side of the business. We have had a solid register offering and it's probably fair to say that on the buffer, we have been playing catch up. We feel our 2016 from a register versus buffer is on track. 2017, it's in development right now, Quinn. So I'd rather not discuss it. I mean, we're taking at a quarter at a time, as John as said.
Okay, thanks Ford. Thanks, John.
Thank you and this does conclude our question-and-answer session portion of the call. I'd now like to turn the call back over to John Edmunds for any further remarks.
Thank you, Soniya. Inphi plans on attending the Jefferies Conference in Miami on May 11; the Deutsche Bank one-on-one conference in San Francisco on May 19; the B. Riley Conference in Los Angeles on May 25; the Craig-Hallum Conference in Minneapolis on June 1; the Bank of America Conference in San Francisco on June 2; and the Stifel TMT Conference in San Francisco on June 5 and 6. Ford and Debra, and I would like to thank you for joining us today and we look forward to speaking with you again in the future.
Ladies and gentlemen, thank you for participating in today's conference. This concludes today's program. You may all disconnect. Everyone 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: email@example.com. Thank you!