Vicor Corp (VICR) 2016 Q2 法說會逐字稿

Good day ladies and gentlemen, and welcome to the Vicor earnings results for the second quarter ended June 30, 2016 conference call. My name is Tawana and I will be your operator for today. At this time, all participants are in a listen-only mode. Later we will conduct a question-and-answer session. (Operator Instructions). I would now like to turn the conference over to your host for today, Mr. Jamie Simms, CFO. Please proceed.

Thank you. Good afternoon, everyone, and welcome to Vicor's conference call for the second quarter ended June 30, 2016. I'm Jamie Simms, CFO, and with me here in Andover are Patrizio Vinciarelli, Chief Executive Officer, and Dick Nagel, Chief Accounting Officer.

Today, we issued a press release summarizing our financial results for the quarter ended June 30. The press release is available on the Investor Relations page of our website, Vicorpower.com. We also filed a Form 8-K with the Securities and Exchange Commission in association with issuing this press release.

I'll remind listeners this conference call is being recorded and is the copyrighted property of Vicor Corporation. I also remind you various remarks we may make during this call may constitute forward-looking statements for purposes of the Safe Harbor provision under the Private Securities Litigation Reform Act of 1995. Our forward-looking statements are subject to risks and uncertainties that may cause actual results to differ materially from those explicitly set forth or implied in our statements. Such risks and uncertainties are discussed in the Form 10-K we filed with the SEC on March 8, 2016.

Please note the information provided during this conference call is accurate only as of today, Tuesday, July 26. Vicor undertakes no obligation to update any statements made during this call, and you should not rely upon such statements after the conclusion of the call.

A replay of today's call will be available beginning at midnight tonight through August 10, 2016. The replay dial-in number is 888-286-8010 followed by the passcode 36834933. In addition, a webcast replay of today's call will be available shortly on the Investor Relations page of our website.

I'll start this afternoon's discussion with a review of our financial performance for the second quarter and the first half of the year, and Patrizio will follow with his comments, after which we will take your questions regarding our business.

For the second quarter, as stated in this afternoon's press release, Vicor reported a net loss of $544,000, representing a net loss per share rounded to $0.01. For the prior quarter, we reported a net loss attributable to Vicor Corporation of $5.35 million, representing a net loss per share rounded to $0.14. Sequentially, a revenue increase of $6.9 million reduced our net loss by $4.8 million, a reflection of the production related leverage in Vicor's operating model.

For the first half of 2016, we recorded revenue of $99 million and a net loss of $5.9 million, representing a net loss per share of $0.15. In contrast, for the first half of 2015, we reported net income of $4.2 million, representing net income per share of $0.11. Recall results for the first quarter of 2015 in which we earned $0.09 per share, reflected the relatively high volume contribution of VR12.5 shipments for data center applications early that year.

As addressed in this afternoon's press release, second-quarter results reflected the recurring circumstances we have encountered for the last five quarters in terms of the negative influence on our revenue and profitability of the combination of delays in new programs, and the general weakness of demand for our legacy products due to broad macroeconomic conditions. However, volume did improve sequentially as consolidated product revenue increased 15% with improvement across all business units. Domestic revenue increased 11.2% sequentially and international revenue, consisting of exports and the revenue of our Japanese subsidiary, rose 17.7%, reflecting a rise in shipments by VI Chip and Picor of VR12.5 solutions to Asian contract manufacturers.

Of note, consolidated turns volumes, that is orders received and shipped within the quarter, returned to the level we had seen for the pronounced macroeconomic headwinds of Q4 2015 and Q1 2016. As we've discussed before, we consider turns to be only a coincident indicator of market health. However, we do know that sustained weakness in turns volume below a certain level is a red flag, so we are pleased to see this metric improve. Historically, turns volumes have been almost entirely related to BBU modules, but I should point out that calculation of turns for recent quarters has included in more meaningful contribution from VI Chip due to shortened leadtimes for certain higher volume programs. Of course, shortened leadtimes are associated with higher efficiencies, which contribute to higher product gross margins.

Concluding on consolidated revenue, recognized distribution revenue for the second quarter rose over 24% with each of our stocking distribution partners recording improved results with our new products representing a larger share of activity.

Turning to product level profitability, the increase in production volume and shipments contributed to a greater than 4 point increase in consolidated gross profit margin as a percentage of sales, which rose from 42% for Q1 to 46.2% for Q2. BBU module and configurable revenue increased by 8.3% sequentially, driving Andover's gross profit margin higher, primarily due to a rebound in shipments of bricks for industrial and transportation applications.

The operational challenges associated with the consolidation of our six custom units into three wholly-owned subsidiaries are behind us, and our customer revenue increased 36.5%, on track with forecast. Combined gross margin for the three custom subsidiaries rose over 4 points, reflecting higher volumes.

VJCL, our Japanese subsidiary, recorded a second consecutive quarter of higher dollar revenue resulting from the 15% year-to-date appreciation of the yen to the dollar through June 30. However, trading conditions in Japan remain challenging as yen revenue declined 2.3% sequentially, while gross margin declined over 3 points despite lower yen costs to import our modules used in VJCL's product portfolio.

VI Chip revenue increased 37% sequentially, driving the unit's gross profit margin higher by over 8 points, primarily due to the aforementioned increases in shipments of VR12.5 solutions and sustained volumes of older VI Chip models for industrial and test instrumentation applications. This margin improvement occurred although we continue to operate well below capacity, supporting our expectation of further margin improvement for VI Chip with the volumes we are forecasting for 2017.

Picor, our fabless IC subsidiary, experienced a 26% increase in revenue, reflecting its contribution of SIP regulator components to the increased volume of VR12.5 deliveries. Vicor's new products also gained traction within our stocking distribution channel. Picor's gross margins were largely unchanged for the quarter.

Consolidated operating expenses only rose 1.3% quarter-to-quarter, and for the first half of 2016 were at 2.3% or $1.2 million lower than the total recorded for the first half of 2015. Consolidated SG&A expenses rose 2.1% sequentially while consolidated R&D expenses were flat, rising only 0.3 of a percentage point.

Within SG&A for the second quarter, the increase in revenue drove related variable expenses, notably commissions, which rose 10.7%. The sizable seasonal decline in audit costs and related accounting fees largely offset a seasonal rise in shareholder reporting costs tied to our annual shareholder meeting and the publication of our annual reported proxy as well as anticipated increases in travel, advertising and related sales and marketing costs. R&D costs saw a 4.6% sequential increase in prototype spending, but this was offset by reductions in spending across other categories.

Our consolidated operating loss of $601,000 compared favorably to the $5.4 million operating loss reported for the first quarter. As was the case in the first quarter, our second-quarter tax calculation did not reflect unusual or nonrecurring activity.

Turning to cash flow and our cash position, our second-quarter net loss combined with a $4.5 million swing in working capital to cause a decline in our cash and equivalents of $5.6 million to $54.2 million. Capital expenditures for the second quarter rose to $2.8 million from the prior quarter's $1.9 million, higher than we had forecast, as the timing of in-service dates for certain projects shifted. Recall that we completed the buildout of our new VI manufacturing line during the fourth quarter and substantial construction retrofit projects continue. We continue to expect capital expenditures to remain in the recent range of plus or minus $2 million for the foreseeable future.

The quality of our receivables portfolio remains excellent with days sales steady at 46 days. Similarly, our aging schedule remains in very good shape with 97% of accounts under 60 days.

We were in touch with our Turkish distributors last week and are comfortable with our modest exposure there. No other markets present a challenge or represent a concern for us at this time.

Annualized turnover of consolidated inventories declined slightly from 4.7 to 4.5 times, reflecting an increase in raw materials held by VI Chip. No unusual or notable activity occurred in our inventory reserve accounts.

Employee headcount as of June 30 was 1,013, of which 975 were full-time employees. This quarter, I will provide greater detail on the composition of changes to our headcount.

Year-to-date, total headcount has increased by net 28 with the net addition of 11 full-time employees across the organization, a net reduction of nine part-time employees, and a net increase of 26 co-op students typically involved in programming and IT support. The year-end cycling of these co-ops skewed the full-time and part-time figures I provided last quarter.

Now I'll turn to bookings and our outlook for the third quarter. Bookings for the second quarter rose 7.2% sequentially to $52.5 million, the highest level in five quarters. Total BBU bookings rose 9% overall with contributions from the Andover module business, our custom business, and VJCL both in dollars and yen. VI Chip and Picor bookings reflected the continued order flow, which had resumed in the first quarter for BTM and PRM solutions for powering processors using the Intel VR12.5 standard. Reflecting these orders, VI Chip bookings sequentially increased 7.1%. As I described before, contract manufacturers frequently do not match BTM and PRM order volumes, and second-quarter bookings for Picor were again out of sync with bookings declining 23% after rising 87% for the prior quarter. With this mix of bookings, our consolidated book-to-bill ratio for the second quarter stood at just under 1 to 1. Backlog as we entered the third quarter totaled $30.3 million, a slight decline of 3% from the beginning backlog for the second quarter. We are expecting the improvement of turns volumes to continue with their growth more than offsetting the slight decline in backlog scheduled to ship during the quarter. We also are expecting additional VR12.5 orders through the first half of 2017 before the transition to Intel's VR13 standard finally occurs. We have been successful in expanding our 48 Volt to Load solutions to additional server manufacturers or the new VR13 standard and anticipate increasing orders and shipments to additional computing and networking customers to occur starting in 2017.

As Patrizio will address, our momentum in the 48 Volt to Point of Load segment continues. Also, customer interest in our new chips and chips in VIA packages for a range of applications beyond computing is rapidly expanding. This is all very encouraging, but sales cycles for new, disruptive products are unpredictable and can be long. Similarly, we cannot predict when macroeconomic conditions will generate confidence rather than uncertainty.

As discussed before, our breakeven quarterly revenue level is in the range of $53 million depending upon product mix. As such, we expect to operate for the next two quarters close to breakeven, given our backlog and the aforementioned turns expectations. However, major variables outside of our control include the timing of large new programs beginning to ramp.

I'll now turn the call over to Patrizio.

Thank you Jamie. As Jamie noted, second-quarter results reflected the rebound from the weak first quarter, which we believe the percent is a turning point in our long-term performance strength. He also spoke to the expanding level of market enthusiasm associated with the differentiated 48 Volt Direct to Point of Load solution. Coming off a high profile event earlier this year, notably the Open Compute Project Summit in March at which Google affirmed its commitment to the 48 Volt bus architecture for data centers.

Momentum in other markets continues to build with the expanding offering of the high-performance bar system building blocks, specifically chip modules and chips within VIA platforms, for use in applications for which their matched power density make them compelling, cost-effective alternatives to competing solutions in larger traditional form factors.

As mentioned last quarter, we are actively engaged with our new products with customers in a range of promising markets, including electric vehicles, and within that field, autonomous vehicles in particular, LED lighting and wireless technical communications infrastructure. We also are seeing an acceleration in interest in our new products from customers in markets served by BBU, as these customers undertake new designs of products in which our legacy bricks have been de facto standards. Our global teams are busy responding to interest ranging from initial engineering inquiries to a sense of design engagements.

Introduction of new products continues, contributing to the diversification of our customer base. Notably, just last month, we expanded our family of VIA packaged DCMs. DCMs are isolated regulated DC to DC converters offering a matched power density. When deployed in our VIA package for either board or chassis mounting, these converters offer best-in-class performance and a high level of design flexibility. With our June introductions, we now offer DCM solutions for challenging defense electronics applications, complementing commercial offerings for communications, industrial, instrumentation, robotics and transportation applications.

The broadened DCM family is fully integrated into Vicor's power component design methodology, the basis of our strategy for the growth markets we serve. Our focus on an end-to-end modular power system design methodology is unique in the industry as it provides system designers with predictable components and system functionality.

Reliability, fast design cycles. [DCCs] configurability and scalability, recognition of these unique power system capabilities is catching on across a broadening base of customers.

I will conclude my remarks. Despite appropriate caution and conservatism regarding near-term financial results by once again affirming my belief that Vicor is close to fulfilling its long-term vision of transforming our competitive landscape. Yes, microeconomic uncertainty remains a challenge for our legacy products, and this ramp of innovations evidently uncertain sales cycles. However, with customers recognizing we have the right products, our momentum is finally building.

I'm sure listeners have many questions, so I will open the call. Operator?

(Operator Instructions). Jim Bartlett.

Last quarter, you said you expected to see bookings take positive steps this quarter, the next quarter and the quarter after that as far as I can see, visibility, the same progression of increasing bookings and shipments. I would take it from your more cautionary note this time that perhaps some of these just aren't -- projects aren't developing quite as rapidly, and/or more macro uncertainty. Could you help me understand a little more about what's happening here?

Yes. So, by the way, we continue to expect an increasing level of bookings and shipments quarter by quarter as far as the eye can see. One thing, though, that has changed since last quarter is that we've been informed of further delays with the actual start date of the next generation VR13, which, as many of us are painfully aware of, has kept pushing to the right with a high degree of unpredictability.

Now, the silver lining is that delays in the ramp of VR13 are giving rise to new orders for VR12.5. However, the offsetting element is that, I think as we've discussed in the past, with VR13 we have a considerably broader customer base, so the makeup of VR12.5 sockets doesn't fully compensate for the delays with new projects, which are only going to be with us as of the time when VR13 finally takes off.

So, to recap, we still believe that we saw a turning point in the revenue performance and online performance by far of the Company last quarter. We are still looking forward to increase bookings this quarter, the quarter after that, and as far as the eye can see, but the delay, the further delay with VR13 has mitigated the steepness of the ramp that is associated with those particular programs, which, as of last quarter, were anticipated to ramp at the end of the year and beginning of 2017, and they now appear to have been set back by another one or two quarters.

So now it's second quarter 2017?

Well, as far as we can tell. And obviously, we've had -- we are not alone in this. This is not a predicament that's unique to us, but the industry at large. We've been surprised before, so I can't rule out the possibility that we could be surprised again.

Now, again, to look at this in the grander context, while VR13 has been delayed again, it has been made up, at least in part, by continued shipments of VR12.5 products. And we are making progress in getting production orders with other programs that are in a different space or spaces unrelated to the vagaries of subsequent generations of processors.

Needless to say, our strategy of diversification, both within the data center space in terms of the types of processors we power, and outside of the center space in terms of the types of solutions that we offer in other markets, is aimed at limiting our dependency, or significant dependency, on any one single product generation introduction schedule, which, unfortunately, has been very unpredictable.

So that would mean that, if potential new data center customers that were awaiting -- they were (technical difficulty) VR13, so getting some of those new customers would have to wait (technical difficulty)

Yes. We have a broadened spectrum of applications at the Point of Load associated with VR13 sockets, which, unfortunately, awaits the actual availability of these new processors.

Thanks for explaining.

[John Dillon].

Hi Patrizio. I just want to go back to Jim's question a little bit more. So, in your prepared remarks, you were talking about the unpredictable delays in the ramp of certain major new programs. So is that the VR13 customers, is that really what the delays are?

That's been the major factor.

And I appreciate your answer to Jim. It makes a lot of sense. It's very hard to follow Intel because they have so many different iterations of the Skylake processor, and Skylake is really what's got the VR13. But you are really looking for the server version of Skylake, and there's all kinds of different versions of the server. So, I mean is it really early -- is that what you are looking for, the early platform when that -- because I think that's scheduled to come out the second quarter of 2017.

I'm not going to provide you reasons to comment on the specific type of processor in question. It may not be one only a play, given the diversity of applications that we have been pursuing.

Okay, so (multiple speakers)

I think with your general knowledge of delays that have occurred with respect to at least some, you can infer that we have been dependent and will continue to depend on when this processor will actually become available in volume.

Okay. What about memory sockets or other sockets? Are you getting any new memory or other sockets before the release of the VR13?

So, generally speaking, when it comes to other sockets that are related to the deployment of the new processors, as you can imagine, they are related to the deployment of the new processor and they are subject to two corresponding unfortunate delays.

Okay. That makes sense, that they are all in the same design. So, in light of that, when we are talking about design wins, have you lost any design wins that you've previously had because of some of the new competitors that are coming out with 48 Volt, or do you still have all the design wins that you've had?

We have not lost any design wins. It is uncertain whether there's any real competitive product that is truly available. And with respect to that, which has been projected as becoming available, the performance attributes of these projected products is well below the performance attributes of our existing products, nevermind new products that we would be introducing relatively soon.

So to be clear, our visibility is limited to what has been advertised as becoming available. We've seen no evidence of real-world availability, and even if that were to become real, that which is being projected in terms of density is far below the density of our products, and in other performance attributes is also considerably the lesser performer.

So, we don't feel in any way threatened by the competition. We believe that our competition is good, particularly competition of this kind where we have a very significant performance advantage, and we have a very good cost card. So we believe that the onset of a broader participation in the 48 Volt to Point of Load applications and broadening requirements by multiple customers for these kinds of solutions is a development that will accelerate our revenue growth and profitability as we are positioned to capture a very good share of that market.

That all makes -- that actually makes perfect sense to me. But what I think I'm hearing is that, because of these delays, you've got a number of customers who are just lining up waiting for the VR13 to hit, and then they are going to be putting in their orders and you will start shipping product. But like right now, you've got one basically data center that's taking the 12.5; you've got a multiple number who are going to be taking the 13. You've got additional lots that you are going to be taking over when the new 13 comes out, and you've got other customers that are coming on. I almost feel like there's a dam here, and all your customers are backing up, and then all of a sudden the VR13 is going to be here and you're going to see a very, very significant increase in bookings and revenue. Am I on track for that, or can you elaborate on that a little bit?

I think the comment I would make is that VR13 and the general evolution powers 48 Volt-based solutions is without question a great development for us. It's a perfect match for the strength of our technology.

Again, we have, particularly in terms of density, efficiency, low noise performance, which is more and more mission-critical and the other respects that I will not be specific about. Because of this variety of performance advantages and a very cost-effective card, this is a development that very much favors our opportunity.

But I think it would be a mistake to put so much emphasis on just that end market. We made tremendous progress over the last couple of years with the expansion in our broad portfolio in a number of different directions with other complementary products that make it more and more real for customers at large, and not just in the data center space, to architect their power system based on a combination of modular building blocks that Vicor uniquely makes. And these are front-end blocks. These are different kinds of Point of Load devices, including in particular a whole family, expanding family of ZVS Point of Load regulators we've developed within Picor, and other kinds of products, again, in the front end space. DCMs, DFMs, these are building blocks which we've developed. These are families that have been expanding with, again, very differentiated performance attributes which meet the demands of our customers in a variety of end markets. So we are finding ourselves in the catbird seat in a multiplicity of end markets, and we are applying all our resources in the front end of the business, in the back end of the business, to pursue in a balanced way this multiplicity of opportunities. And we can do it without spreading ourselves thin because of the unique methodology that we've developed in terms of being able to reuse the same power conversion engines, being able to reuse the same packaging technologies, the same manufacturing infrastructure. So, we have the wherewithal to pursue a large multiplicity of opportunities with common denominator capabilities. And for that reason, it would be a mistake to in effect put all the focus in the data center space in server power. There's a lot more to the opportunity than just that.

And again, that kind of goes to my question. The heart of my question is it sounds like you're going to have increasing bookings and revenue for the next couple of quarters, but then it sounds like it's a backlog of customers just waiting to get these new products and you're going to see a significant step up in revenue, bookings and revenue, six months -- assuming there's no additional delays. Am I right in assuming that, that you will see a very significant step up in bookings and revenue in six months if you don't see any additional delays?

As I said earlier last quarter and reiterated at the beginning of this call, we see increasing bookings and increasing revenues as far as the eye can see. It's not been the case as far as I can remember that we would, as we recently have had to do, in effect turn away opportunities, because we have more customers and more requirements coming to us that in spite of all the efficiencies that are brought about by our unique methodology that we can address. So we've become more and more selective with respect to the threshold of business opportunity and commitment to pursue new kinds of customers and applications.

Great. I'll get back in the queue. Thank you so much. Sounds good.

[Don McKenna].

Hey Patrizio. I don't want to beat the horse again, and I do appreciate the fact that you are broadening the base of your customers, but can you explain to me why there is that delay that is now 15, if not 18, months? Is it because they are not able to produce a reliable, functional product, or have they lost market share to other competing products from other chip design people?

Well, it's really not up to me to comment on what stands behind these delays. I think that question should be posed elsewhere. We are, as our customers are, dependent on the availability of next-generation solutions. And I think, to your point, as time goes on, more and more solutions are becoming available in the special building for alternatives.

Have they not indicated why it is they have the continued delay?

Again, it's really not up to me to comment. There could be perfectly -- I'm sure there are perfectly legitimate and valid reasons for these delays. I think -- I'm fond of the expression that people in glass houses should never throw stones, right? Everybody, every company that makes disruptive technology a core element of its strategy will from time to time suffer delays. You can't have your cake and eat it too if you want a high degree of credibility. But it's an issue. You are taking baby steps that are highly predicable, but they come at the expense of not enough innovation, which ultimately is key to long-term success.

Vicor has, in the past, as you know, as we all passively remember, had its delays. We've had numerous delays, certainly when it comes to the chip products, as I go back a few years. Because of the challenges inherently in the knowledge of the technology, we saw developments slip by a year or two in some cases. The good news is that we've got a fast interface and I think, as we've all seen over the last year and a half, we've been able to introduce products at an accelerating rate, and that's the trend that is continuing because we transitioned from a certain phase in development of brand-new technology to a different phase in which we can apply it with a higher degree of predictability while still leveraging the unique attributes of very disruptive technology. It's a different pace. It's really not within my competency to be able to opine on what makes sense behind the VR13 delays. I'm sure you can find explanations from other sources.

Okay. Thank you.

Dick Feldman.

Hello? Hello?

(Operator Instructions). [Alan Hicks].

Good afternoon. I had a couple of questions on some of your other markets, specifically supercomputing, automotive and wireless telecom. What are your opportunities there?

So, supercomputing, we are getting more and more involved. By the way, unfortunately, one of the supercomputing applications is VR13 based, so that is also suffering from the same delays, so that's an example of a Point of Load supercomputing application.

We are also involved in supercomputing applications from a front end perspective. There are two major initiatives on that front that come to mind, in answer to your question, with some of our front end products. So on the supercomputing front, we are making inroads for the Point of Load and in terms of front end solutions.

Was there another market that (multiple speakers)

Automotive and wireless telecom.

Okay. Automotive. So in the automotive space, we are selling DCMs, and this program is ramping in China with our 46.3 DCM chip, (inaudible) chips per system. We are also involved in other programs, and in some cases I can't comment about the specifics because of neutrality constraints. But I can tell you that they are exciting programs, and ones that have great potential in years to come, particularly when it comes to autonomous driving systems.

And wireless telecom?

So, in wireless telecom, we have some important -- one particularly important design win with our PFM in Asia. So, that's, again, an example of a different kind of product with a different end market, a different type of opportunity, but still leveraging those common denominator capabilities in terms of engines and packaging technology that I was referencing earlier.

Okay. And finally, your BBU business, do you think that's back on the growth path, or is that still going to be kind of flat?

So, the BBU business should be viewed in terms of legacy products, which, as Jamie pointed out in his remarks, do tend to be particularly dependent on the macroeconomic cycles. And they are also very much dependent on defense cycles. But this -- the new breed of opportunities that are enabled by DCMs and other chip and VIA products that we have been introducing within the last year, year and a half, (inaudible) of new applications with respect to these products are growing very rapidly. And they clearly point to a time frame in the next few years where the mix of business in terms of the center of gravity of what was historically legacy BBU business will have shifted progressively towards the new products, which, needless to say, offer much greater density, much greater efficiency, better cost-effectiveness in combination with a number of other improved performance attributes. So we are continuing to enjoy the steady revenue and profitability of the legacy bricks. Don't get me wrong, those are not going anywhere. But the good news is that we are also building on top of that a new set of applications with the existing traditional BBU customer base and with some new customers that enable to buy the new products.

Okay, sounds promising. Thank you.

John Dillon.

I wanted to talk a little bit more about some markets also, in particular the communications market. I think what I saw is the OCP actually supported the 48 Volts, or brought it into the OCP for communications. And as you know, I'm not sure if everyone else does, but communications has always been 48 Volt. So how is that market embracing you? How are you doing in that market?

So, while it is true that communications is very light on a 48 Volt bus, communications for the last 15, 20 years has used that bus through a so-called intermediary bus, so intermediate bus of 12 volts. And while 48 Volts has been the backbone of communication power, distribution of power infrastructure, visually all of the sockets in recent times have not been provided by way of 48 Volt to Point of Load converters. Virtually all the sockets have been supported by 12 Volt to Point of Load converters. And that is changing.

Remarkably, companies that historically have been very entrenched, very comfortable in their paradigm of 48 Volt, 12 Volt to the Point of Load by way of bus converter followed by an oscillated Point of Load regulator have recently made an agonizing appraisal of what it takes for them to be able to compete and be able to make systems that have the requisite density and efficiency. Fundamentally, need is a catalyst for innovation and I think a combination of desperate need for greater density and efficiency, coupled with some of the public pronouncements that took place earlier this year, have caused mountains to move that I would have never thought would be moving as fast as they are away from the 12 Volt intermediate bus and intermediate bus architecture.

Now, I am not suggesting that this is going to be a change that will take place overnight, but it is indicative of a broad change that is taking place now across different end markets and a change that will cause greater opportunity for 48 Volt solutions react to the Point of Load with better economies of scale and greater performance. It is self-reinforcing, and it is indicative of a trend that should bring about very significant changes in the power system architecture across these different markets.

So it sounds like this is a situation where success will beget more success because they are seeing your success in other markets with 48 Volt and they are starting to adopt it. Is that what I'm hearing?

I think that's an element, but there's also another catalyst which has to do with just need. You can't keep solving communications power systems challenges by bus converter bricks and Point of Load regulators. So, if we look at what has happened over the last 10 years, it used to be that the typical motherboard was powered by bus converter bricks with the capability of hundreds of watts. And these boards, over the years, have gone from consuming hundreds of watts to consuming -- so 100 watts to consuming 700 watts, to consuming over 1,000 watts and that trend keeps on going. And the requirements in terms of bus converter building blocks and Point of Load regulators keeps getting more and more challenging to the point that system designers can no longer solve their requirements with the same old solutions, even accounting for the evolution of those solutions, meaning the bus converter technology has gotten better over time, but it's at the point where it no longer works. And part of the reason has to do with the inefficiency of the intermediate bus architecture in terms of its power distribution losses and other factors. So, the reason why these changes are taking place more and more and across different end markets has to do in part with emulation and in part with need. And it's a combination of the two that provides the powerful catalyst.

Great. Are you seeing design wins then, are you seeing many design wins in the communications market?

We are seeing design activity that I think will take some time to turn into design wins. But I think it's fair to say that that's a land of opportunity for our solutions because, once again, we are very uniquely equipped in terms of power conversion engines, of functional building blocks, and our chip capabilities in particular to address these needs with the highest performance with at least a factor to greater density than anything else that's competitive and with better (inaudible) in terms of the key parameters such as low noise and other attributes, and other development that is worth referencing because it bears on all of this is that we are seeing more and more requirements for solutions in terms of the Point of Load devices, processors, other kinds of ethics that demand 0.7, 0.8 volt rails, 200, 300 amperes. That kind of requirement cannot be effectively addressed in a rational way with the same kind of solutions that have been deployed in the past. It calls for both a more advanced architecture and better engines. So, two things have to change to address these kinds of requirements, the architecture and the engine in terms of the capability to deliver in particular a current multiplication function at the Point of Load, which is what in one way of looking at this sets us uniquely apart from everybody else.

Just one last question, and that's how about the graphics market. How are you doing in that?

Very well there too. Design-in activity going on on that front as well.

Great. Thank you so much. It sounds really exciting. Thank you.

Dick Feldman.

Patrizio, you mentioned that you have gotten orders and are addressing markets that are not tied to the 13 server. I wonder when those markets can start to be significant enough to start to move the needle if there were to be further delays with the new server chip.

Well, these other opportunities, and again, there are many of them and they are different stages, are the basis for our general confidence in the growth opportunity irrespective of the ultimate timing of VR13. I'm not suggesting in any way that VR13 isn't going to come around; it will. It's just a matter of exactly when, and obviously our passive detections were based on the information, the best information, that was available to us, and that information in hindsight turned out not to be accurate. There have been numerous resets, but I don't expect that it is going to go on forever.

I think, at this point in time, whether it's the first or second quarter, worst case the third quarter next year, VR13 as far as we are concerned with the procedure skews that we need to address and the bus system requirements that we will provide, they come around. And when that happens, we expect to see a tube effect on the growth that should otherwise be more and more based on a broader set of opportunities that are providing along in complementary end markets with complementary products.

So what you're saying is some of the new markets you want to enter are tied to the introduction of the new chip, or are there things that are independent of all of this that may provide some growth?

Sorry for interrupting, but there are many things that are totally independent of VR13 that are progressing along. So VR13 will be a significant boost as it ramps, but if let's say a (inaudible) for whatever reason -- I'm not suggesting that would be the case -- if it were never to happen, which again I would not expect to be the case, I think it's only a matter of when, not a matter of if. But even if it did not come about, we would still be looking at the growth story with respect to the complement of products that we've been developing and getting designed in into a variety of end markets and applications.

The design wins that you have gotten, when do you anticipate they would start to go into production in a meaningful way?

So, as an example, in answer to the earlier question regarding supercomputing applications, (technical difficulty) with a design win for front end components that will amount to millions of dollars worth -- or is projected to be millions of dollars worth of business next year and ramping further into 2018. So, it's a long list of different products, different customers, different applications, different end markets, and they're all at different phases of gestation.

One last question. You made some comments on your call earlier in the call that there are -- you are deluged with customers making inquiries about using these products that you feel you have to be perhaps more selective in who you can address. I wonder if you could give us further insight into that comment.

Yes. So, it is apparent to us that it has become apparent to many OEMs that their challenging our system requirements can in some cases only be addressed by Vicor power components. Even companies that historically we have found difficult to penetrate for a variety of reasons have recently come to us looking to engage us to have products designed to meet their needs.

So, I'll give you, without mentioning a name, an example in the communications space. A very substantial company, Asian company, has come to us with a very unique requirement, and they believe we are capable with our technology to address it in a way that cannot be addressed otherwise. So we've engaged with them to address this requirement. We have been approached by other companies. And what I am referencing here is these are all things that have happened within the last six months. A number of different engagements that we've taken on and other opportunities that we have selectively chosen not to pursue because they were deemed to be, even though interesting in many respects given other opportunities that we can pursue, not worthwhile.

So what I see into this is a growing realization by OEMs in a variety of end markets -- in some cases some of the competitors have also come to us for help -- that our technology, because of this very high density, high efficiency flexibility and general capability, can address needs, higher levers, in their own specific ways, and in fact they want us to be part of the solution. I think it is symptomatic of, again, a growing realization of our capabilities, which bodes well for our opportunities going forward. It gives us the privilege of being able to select and choose what we have interest in going after, and that's a good place to be.

Okay. Thank you and good luck.

Thank you. If there is one more question?

There are no more questions in the queue.

Very well. Thank you very much.

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