Vicor Corporation (VICR) Earnings Call Transcript & Summary

The Board of Directors of Vicor Corporation welcomes you to the 2024 Annual Meeting of Stockholders. I'm Jim Schmidt, Chief Financial Officer, Secretary and Treasurer of Vicor. I also am a member of the Board of Directors of the corporation. As provided for under Article 1 Section 8 of the corporation's bylaws, I will serve as Chairman of this meeting as well as Recording Secretary. Invitations to attend today's meeting were extended to stockholders of record as of April 26, 2024, and certain guests. Each attendee shall have registered in the lobby as you entered the building and received an agenda and our rules of conduct for today's meeting. As you may have seen upon entering the conference room, we have copies of our 2023 annual report on Form 10-K and our annual proxy statement available for attendees. On the screen is a comprehensive safe harbor statement comparable to the language we include in our other public statements regarding the company's financial or operational performance, such as our filings, press releases and our quarterly earnings calls. I will not recite the full text, but I remind you various remarks we make today may constitute forward-looking statements for purposes of the safe harbor provisions under the Private Securities Litigation Reform Act of 1995. Consistent with NASDAQ Listing Rule 5620, stockholders are afforded the opportunity to discuss company affairs with management at this meeting. However, as shown on the screen, Vicor is in its quiet period for the second quarter of 2024, and as such, we will not comment on specific elements of financial or operational performance for the second quarter or any forward period. Second quarter performance will be addressed during our upcoming earnings call. On the screen is the agenda for today's event. As shown, we'll conduct a formal business meeting, reviewing the proposal before stockholders as well as any business to be brought before the business meeting. As of now, I am unaware of any such additional business. After we take care of formal business, Phil Davies will present a brief strategy review. After that, Patrizio will take your questions regarding the company's business and products. Again, he cannot answer questions regarding our financial or operational performance for the second quarter of 2024 or for the full year. As has been the case in the past, we'll be posting a recording of today's meeting on the Investor Relations page of our website. I'll now introduce my 11 fellow members of the corporation's Board of Directors. Going forward, our Independent Directors are; in alphabetical order. Michael Ansour is a managing partner of March Partners LLC, an event-driven investment firm based in New York, New York. Michael is also a Director of the Hertz Foundation, a non-profit organization that awards fellowships to PhD students in the applied physical, biological and engineering sciences, for which he also serves as Chairman of the Foundation's Investment Committee. Michael previously served on Vicor's Board of Directors from 1993 to 2007. Jason Carlson, former CEO of Congatec AG, headquartered in Deggendorf, Germany, from 2015 to '21. Congatec is a leader in the industrial embedded computing field. He is the former President of Semtech Corporation, a leader in analog and mixed-signal semiconductors, as well as holding other positions of leadership in the electronics industry. Jason chairs both our Audit Committee and Compensation Committee. Estia Eichten is an original investor in the company and has been since 1989, a senior scientist at Fermi National Accelerator Laboratory in Batavia, Illinois, outside of Chicago, which he joined in 1981, the same year he joined our Board. Zmira Lavie has been a Partner of M&T Semiconductor, a consulting firm serving professionals and organizations throughout the semiconductor ecosystem, since January 2020. Previously, Ms. Lavie was with Tower Semiconductor, a manufacturer of integrated circuits using specialty process technologies, from 1988 to 2019. Zmira has over 30 years of experience within the semiconductor industry, including leading diverse business initiatives, R&D, complex projects and process development. John Shen has been a Professor and Director at the School of Mechatronic Systems Engineering of Simon Fraser University, in British Columbia, Canada, since January 2022. Previously, John had been the Grainger Endowed Chair Professor of the Illinois Institute of Technology from January 2013 to December 2021. John has 32 years of academic, industrial and entrepreneurial experience in power electronics, power semiconductor devices, renewable energy systems, microgrids, transportation electrification, sensors and actuators. Based on the assumed results of the shareholder vote to occur at this meeting, our employee directors for the forward term in alphabetical order will be; Phil Davies, who has served as our Corporate Vice President, Global Sales & Marketing since February 2011. He was appointed to the Board in 2019. Mike McNamara, who has served as our Corporate Vice President, General Manager, Operations since 2015. He also was appointed to the Board in 2019. Claudio Tuozzolo, who has been with Vicor since 2001, serves in a leadership role spanning R&D and sales and marketing. Claudio has been a Director since 2007. Patrizio Vinciarelli, Chairman of the Board, President and CEO, who founded the company in 1981. And myself. I joined Vicor in June of 2021 as Chief Financial Officer, Secretary and Treasurer, and was appointed as a Director at a meeting of the Board of Directors immediately following the 2021 Annual Meeting. Sam Anderson, CEO of IceMOS Technology Limited, a supplier of high-quality thick film bonded silicon on insulator wafers based in Belfast, Northern Ireland. Sam is the former CEO of Great Wall Semiconductor, an important component supplier to Vicor and which held an investment in non-voting convertible preferred stock. Great Wall was acquired in September 2015 by Intersil Corporation. Andrew D'Amico has been in the role of General Counsel for intellectual property matters for Vicor since 2006. Prior to his engagement with Vicor, Andrew had 18 years of private practice experience in the field of patent law, including patent litigation, patent licensing and patent prosecution as well as counseling in diverse technological areas, while with Fish & Richardson P.C. We're joined today by our outside counsel, Gabor Garai, who is remote, who chairs Foley & Lardner's Private Equity & Venture Capital practice, among other responsibilities with the firm. I also thank Foley & Lardner for providing the venue and support for today's event. Also in attendance is Stephen Briggs, our audit partner with KPMG LLP, our independent registered public accounting firm. Finally, I recognize Stephen Plefka, representing Computershare -- representing Computershare Trust Company, the corporation's Transfer Agent, who has been appointed to act as Inspector of Election, also known as the teller for the formal business portion of the meeting. Thanks to everyone for attending. Turning to the formal portion of the agenda. I call the 2024 Annual Meeting of Stockholders to order. On the screen is the agenda for the official business portion of today's meeting. Notice of the meeting was sent to all stockholders recorded as of the close of business on April 26, 2024, the record date. Only stockholders of record on that date are entitled to vote at this meeting, either by proxy or in person ballot. Similarly, only stockholders of record are entitled to present proposals during the business portion of this meeting. The affidavit of mailing the notice of meeting has been delivered by the teller to me and will be filed with the minutes and records of today's meeting. The proxy statement was filed with the SEC on April 29, 2024, and mailed to stockholders shortly thereafter and sets forth the proposal for stockholder vote here today. The Board recommends stockholders vote in favor of proposal 1. While Rule 14a-8 under the Exchange Act provides for the inclusion in our proxy statement of appropriately submitted proposals from eligible shareholders, I, in my capacity as secretary, confirm we did not receive any such proposals to be included in our proxy statement. This slide sets forth the number of shares of common stock and shares of Class B stock outstanding as of the prior 2 year ends and the record date. As shown, as of the record date, April 26, 2024, our outstanding shares consisted of 32,809,845 shares of common stock and 11,738,718 shares of Class B common stock. I remind you our outstanding Class B shares possess 10 votes per share, although they are exchangeable only on a one-for-one basis for shares of common stock. As shown, holders of our Class B shares, of which there are 11, possess as of the record date, 78.2% of the cumulative voting power of the 150,197,025 votes eligible as of the record date. Also as of the record date, Dr. Vinciarelli beneficially owned 29.6% of common stock shares outstanding and 93.9% of Class B shares outstanding, representing 79.7% of total voting power, giving him control over all governance matters. On a fully converted basis, Dr. Vinciarelli's holdings represented 47.6% of share ownership. Because of Dr. Vinciarelli's voting control, under the corporate governance requirements of NASDAQ Rule 5600, we are considered a controlled company, which is defined as a listed entity for which more than 50% of the voting power is held by an individual or an identified group. As such, we rely on certain exemptions under Rule 5600 and do not have a Board consisting of a majority of independent directors, nor is our nominating process the sole responsibility of independent directors. We do not utilize the available exemption from independent director oversight of executive compensation, but instead maintain a compensation committee comprised solely of independent directors. Of course, there is no exemption from maintaining an audit committee comprised solely of independent directors, which we do. Directors Ansour, Carlson and Eichten will be the sole members of our Board's Compensation and Audit Committees for the 2024 term. Returning to procedure. Our first order of business shall be to determine whether the shares represented at the meeting, either in person or by proxy, are sufficient to constitute a quorum for the purpose of transacting business. I introduced Stephen Plefka of Computershare earlier who has been appointed to serve as Inspector of Election, having pledged to perform those duties under Delaware law to support me in the conduct of this meeting. He pledged -- his pledge will be filed with the minutes and records of this meeting. Are there stockholders in attendance who have not submitted a proxy but wish to vote in person? If so, please raise your hand so that Stephen may confirm your eligibility as of the record date and give you a ballot. Stockholders present should be aware, for purposes of preparing the slides used today, we have assumed those who have signed and submitted a proxy, even though present, intend to vote by that proxy. However, are there any stockholders who have previously submitted a proxy who would like to change their vote on the matter before stockholders? If so, identify yourself to Stephen. Seeing none, I now ask the teller to inform us whether a quorum is present.

Mr. Chairman, I can report, as of the record date, April 26, 2024, 32,809,845 shares of common stock and 11,738,718 shares of Class B stock were outstanding, representing an eligible total of 150,197,025 possible votes as shown on your slide. Given our receipt of Dr. Vinciarelli's proxy, representing 79.7% of eligible votes, far more than the simple majority of eligible votes required for a quorum are present here today. As you have just seen, I have received no new ballots nor any instructions to withdraw or change an existing proxy. So the numbers you previously reported and shown on your slide are accurate. As of the record date, Vicor had 96 holders of common stock, defined as held of record under Rule 12g5-1 of the Securities Exchange Act of 1934, representing the sum of stockholders identified by Computershare as distinct holders in possession of their shares or stockholders whose shares are held by a custodian for a single account. A complete list of stockholders as of the record date, compiled and certified by Computershare is open for examination at this meeting by any stockholder.

Thank you. I, therefore, declare a quorum. The polls are now open and will remain open until the matters before the meeting have been presented and the teller confirms no votes have been cast other than those received by prior proxy. We will now proceed with the discussion and voting on the proposals set forth in the proxy statement. Since the proposal was addressed in detail in the proxy statement and since 91.1% of shares eligible to vote are represented by proxy at this meeting, we will dispose with the formal motion and balloting for the proposal. Further, as mentioned, Dr. Vinciarelli's proxy, representing 79.7% of eligible votes, has been received and voted in favor of the proposal. Therefore, the proposal received prior to today's meeting has more than enough votes for approval. Each stockholder was entitled to vote for a maximum of 12 nominees. Cumulative voting was not permitted. Is there any discussion regarding the size of the Board and the election of the nominees? There being no further discussion regarding the proposal. I ask now whether there are any other matters to be properly brought by an eligible stockholder before the meeting? Hearing none, I ask once more if there is a stockholder who has not cast a ballot and now wishes to do so? Or is there a stockholder who has not yet had his or her ballot collected? There are appearing to be none. I now formally close the polls. Based on the vote totals previously provided by Computershare and seeing no additional votes reported by the teller, the slide shown sets for the following results. All 12 directors standing for reelection received an affirmative vote of at least 94.9% of the total amount of votes cast, representing at least 86.4% of shares entitled to vote. Therefore, all 12 nominees are hereby re-elected. The final results of today's ballot will be reported in detail on or before Thursday, June 27 via filing of a Form 8-K with the SEC. There being no other business to come before the meeting, I will entertain a motion to adjourn.

I so move.

Thank you. All in favor adjournment, please say aye.

Aye.

All opposed, say no. Vote is in favor of adjournment. This concludes the formal portion of our meeting. And I will turn it over to Phil.

Okay. Thank you, Jim. Good morning, everyone. One of my favorite times is actually meeting people outside of the formal part of this annual meeting. So, nice conversations. And I brought some samples along and some of you got a chance to see some of the Gen 5 technology. And if we didn't get to do that, I'm going to be around after the meeting. And if you want to see that in person, I can give you a little demo of what we have there. Okay, let's get into the market update and strategy session. Here we go. Okay. So, in terms of preparing for the meeting today, I spend a lot of time thinking about what to really talk about. And I think these words from Patrizio, which came from the Q1 statement that we put out really capture, I think, the essence of what we want to talk about today, because 2024 is the year of transition. And so, I wanted to talk about the product strategy, the licensing strategy, the customers, the markets. But I want to start, first of all, with that clarity of purpose that Patrizio mentioned here in the words in the statement. So let's begin with that. So the good news is the clarity of purpose for our corporate strategy has not changed since we put it in place almost 3 years ago now. You can see here that we talk -- and I talk about this quite regularly on the quarterly update meetings, the top 100 customers that we are now focused like a laser beam on in terms of the 4 markets that we're also focused like a laser beam on. We're all about operational excellence around those top 100 customers using our new chip fab that you just saw. And we will never ever take the focus off, having the proprietary technology leadership in the highest density, both on the power front and on the current density front for our power modules. And then, we also have revenue from the products and also now a new revenue stream coming from IP licensing. So that is our clarity of purpose for our corporate strategy, and that hasn't changed. So that's very, very important, I believe. In terms of our business, we have a little -- few changes around the edges here. But again, clarity of purpose for our business is to really focus on 2 very big market segments, the AI market or high-performance computing area, with a whole host of technologies, and I'll talk about each of these markets in a little bit more detail. Automotive power modules, which is a new market segment that we entered in at the late 2018, I think it was time frame. And couple of things about our industrial and aerospace market. We've actually doubled our revenue in the industrial market the last 5 years, and I'll talk about what we're planning to do for the future. The aerospace and defense, it's a little bit of a different market, but we've doubled our revenue there in 7 years. And we view that actually as a growth market, and it's got very nice gross margin. So it's a very important market for Vicor. So, we'll spend a little bit of time looking into that as well, okay? So, opportunities and challenges, let's talk about those for a moment. The opportunities, I'll spend more time on. The challenges for the company, and I'll take this right up front here is really the reduction in revenues that we really enjoyed from the high-performance computing market. I think, as everybody knows in this room, we were sole-sourced at NVIDIA for a very long time, and their product strategy, their technology strategy has changed. And our objective now, not just at NVIDIA, but at a whole host of other big AI processor chip and hyperscaler companies, is to regain share of that market that we once enjoyed. And I believe we have a strategy to do that, and we'll talk about that as part of the presentation here today. So, in terms of our product strategy, I've shown this slide before. Again, it hasn't changed. We've got very, very foundational IP on 48-volt technology, developed almost 20 years ago, if not more than that. Our focus of that foundational technology on high-performance compute is really to focus on current multiplication at the point of load. That's a big differentiator for us. Nobody does current multiplication at the point of load and that is really coming to the fold with the advent of 2,000 AMP AI processors and network processors, requiring vertical power delivery. In the automotive world, there is far more emphasis now placed on weight, on power density, as pure electric, mild-hybrid and even plug-in hybrid vehicles come to market. And the power levels are increasing, not decreasing. So our technology advantages there are giving us great opportunities as well. And then, on our top 100 customers in the industrial and aerospace markets, we are really focused on those customers that have very tough power delivery network challenges. That's how we selected that top 100, okay? We needed to work with customers that had that power delivery challenge. And so that's the reason that we focus on them. In terms of the product strategy, one of the things that Vicor has always done is to anticipate the needs of the market. And we've done that in high-performance compute and in automotive. And so those 2 markets really drive the innovation and the product strategy within the corporation, okay? You can see the different power modules that come out of that. And again, the curve on the right-hand side showing the improvement in power density over the years. Significant advances every 2, 2.5 years that we bring to market. In terms of derivative products, we have our platform breakthrough products. We now have derivative products. We are going to be introducing 20-plus new products this year, derivative products off those products developed for HPC and automotive. And we're bringing those forward into targeted segments and targeted customers in the industrial and aerospace and defense market. In terms of IP licensing, we have 4 pillars of intellectual, if you like, of innovation at the company. They're shown here; topologies, control systems, semiconductor components, and last but not least, that beautiful factory that makes those amazing packages that we put that technology into, okay. You need all 4 to get those current density and power density performance levels that I showed you. So that is what is the cornerstone of the IP licensing. And there are 2 pieces of that. There are licenses that give access to the power modules themselves to basically second source, if you like, the Vicor products, or to even develop unique products. But we're really focused on IP licensing that really drives supply chain activity at this point in time, and that is our OEM licensing strategy that we've talked about on the quarterly calls recently, okay. We have opportunities across 4 markets. So those top 100 customers, they live within these 4 markets that we focus on. And as I've mentioned, the enormous opportunity in front of us is about $5 billion a couple of years ago, and it's growing to about $11 billion in the coming years. And that's with those top 100 customers. All of our other customers that go through the channel is additive to that. So, we are still focused on the channel and the wider customer base, but we want to execute on the top 100 in these 4 segments and drive a share of that eventual $11 billion market opportunity for Vicor, okay. Let's talk about high-performance computing for a moment, because this is a very important, if you like, regaining our footing in that particular marketplace. So let's jump into that and talk about vertical power delivery. So, vertical power delivery, we've already had a Generation 1, and we're moving on to Generation 2. Generation 1 is shown here. If you look at an AI ASIC or a networking ASIC, right underneath it, you will have a whole field of hundreds, if not thousands of bypass capacitors. And those bypass capacitors are needed for noise cancellation and they're needed for transient energy surges that the processor needs. They live there. And so what we did to solve the vertical power delivery problem was to integrate that in a stacked package that you can see on the top right. So, we have our current multiplier, and on top of that, we had what is called a gearbox. And that is that capacitive interdigitated layer to connect that product to the ASIC above. That has to be custom. The footprints have to match. So it's not a big advantage to have custom products all the time, okay. So, that's Gen 1. Very low current density requiring that stack package technology. And our competition is now copying that approach to vertical power delivery. They have heavy, very low current density products, around 1.25 amps per millimeter squared. That has significant challenges to do the vertical power delivery that ASIC companies are looking for today, okay? But Vicor has moved on. And this is our Gen 5 technology, the second-generation VPD, where the current density is so great, it's over 3 amps per millimeter squared, and those current multipliers can now sit and nestle in to those bypass capacitor areas on -- underneath the ASIC or the GPU, really simplifying the current delivery to the processor, okay. No stack packages, no thermal problems, low weight, very easy to implement. And so this Generation 5 technology is being brought to market, now. That's the samples I have in the box, and I'll show you later. In terms of the evaluation board we're providing to the customer, this is it. This shows 2 current multipliers that will sit underneath the board to deliver 1,200 amps of DC current to the product I'll show you in a moment. That's at the heart of the current multiplier, vertical power delivery for this marketplace of Gen 2 VPD. On the top, we have the load cell that simulates the GPU ASIC. And then we have 2 other products. We have our PRM that regulates the 48-volt bus that comes in, and then we have a driver, the MCX, that drives the current multipliers. So that's a 3-chip set or a 4 chip set in this case, for the customer, and this is the evaluation system that we will be building in the next month. And in August, we will be demonstrating this to lead accounts one-on-one, in meetings that are being set up now. And then these boards will also be delivered to customers as we go through the year. And from that, we are hoping to regain our leadership position back in the HPC market with this new technology that really advances the state of the art in vertical power delivery. So, our current density is so great that the other opportunity for us is in the advanced packaging world, where vertical power delivery is great, but you can further reduce power losses by actually putting the current multipliers inside these advanced packages that are being developed. Now these advanced packages are being developed for chiplets on a single GPU scale or a 2 GPU scale, but they're also being developed for wafer scale technologies similar to what Cerebras has, but not exactly like Cerebras, because that's a radical system, but similar to that. And so the current multipliers that we are developing can go inside these packages, further reducing PDN, further improving performance. It's that dense in terms of the current density. So, this is an area that we are now going to be working on with a couple of lead customers over the coming 6 months to a year. Let's talk about our Industrial business. Focused on 4 main areas where we call these the megatrends within industrial or the new industrial. So, robotics, e-mobility, ATE and battery cell formation and test. These are the markets, again, that we're focused on in industrial. Our strategy here, and I'm going to do a little bit of a build to explain it. It starts with identifying, again, those segments within those 4 areas that really value the density and the performance of Vicor's technology. We've identified 36 accounts that we're going to be focused on as part of our top 100 strategy. And then the products that are derivative products that have been developed for HPC and automotive, we map them to those segments and to those customers. Then we bring in our channel partners to then create more customers that are also within those vertical segments. So, again, we are laser-focusing the channel resources that we have, DigiKey, Mouser, Arrow and Avnet across the world to build out what we are calling our top channel accounts, okay. This all leverages the technology for the top 100, and it's additive to our $11 billion SAM number. So we're going for 2x growth again. So the last 5 years, we grew this business 2x. We're going to grow it again 2x over the next 5 years, that's our objective. Aerospace and defense, we see this as a growth market. So we've grown this 2x in the last 10 years. We're viewed as a strategic supplier by the North American and European defense companies. We have incredibly close relationships with them. They look to Vicor to deliver new technologies of density and performance that they need for their new systems, both offense and defense. The share that we have of that market at the moment, actually, is 26%. We have a very nice share of that market, but we can grow that significantly with new products. So, we're bringing out new technologies this year, as I mentioned, 20-plus derivative products aimed -- 10 or 12 of those aimed at this market segment, both on the DC-to-DC front and on the AC-to-DC front, as you can see here. These are the markets that we're focused on and we're going to double this business again, but this time in about a 7-year period, little bit different than industrial because of the gestation period that happens in defense and aerospace. It's a little longer. The AC to DC market is a very interesting market here. We've taken a product that we've developed for the 50 hertz, industrial and HPC and automotive markets, and we've turned it into a 400 hertz product for the aerospace chassis systems. It's a big market. It's got a TAM -- a SAM of $800 million. So we've got this product that's launching now. We're sampling this now to lead customers, and we expect this to be a very big market and part of that doubling of this business over the next 7 years. It's a very important strategy for AC to DC for us. Automotive; so we've made great progress since 2018. We were talking outside earlier about a little bit of a delay and there was a bit of a delay in this market due to the whole COVID situation that we all went through where customers basically put on hold any new product development for a couple of years. But that's in full steam now. So that's really moving ahead very fast. So we're fully engaged in this market. We've secured Stage 1 of the IATF standard quality standard certification. We've got 2 OEMs going into production in '25. Early production this year actually. We're shipping 3 automotive qualified products today, and we'll talk about that. Our pipeline is over $1 billion, and we can continue to grow that. We've got design wins additive to the ones above there at 4 additional customers and our production ramps, as I mentioned, will begin in '25. Our strategy here is about high-density, scalable power modules. So we have about 4 or 5 products that scale to meet the power delivery network needs, both for onboard charging for, if you like, 400 to 800 volt charging station compatibility with 400 to 800 volt batteries in vehicles that are on the road today. And we also have 48-volt zonal architecture, both getting from high voltage at 800, 400 to 48, and then from 48 down to -- the 48-volt loads down to 12 volt loads. So this is our strategy. And we can do this with about 5 or 6 power modules, very low weight, very high density, everything that the OEMs are looking for. 48 is the new 12, we said this about the HPC market about 10, 12 years ago. At the time, I don't think anybody believed us. But Elon Musk came out with a cyber truck, and that's all 48 volts zonal. And since that time, we've had so many conversations with OEMs and Tier 1s about how to do 48 volts and Vicor being the leader in 48 volts. So that's really helped us in the market with our positioning and with the conversations that we're having with OEMs and Tier 1s. So this is a very important message. We talked about BEV, pure electric vehicle. But we're not just in pure electric vehicle. We have collaborations ongoing with mild hybrid car manufacturers, with plug-in hybrid platforms, and also with 48 volts coming to the ICE market as well. So we're not just focusing on electric vehicles. We have design wins and collaborations ongoing across the whole automotive platform front, okay, and that's a very important thing. Now, dollar content, of course, changes per vehicle, per segment, but there's still very good dollars in a very big market for us to get here, okay. So, in summary, I haven't put any numbers here, please forgive me, but it shows you what we're looking at in terms of growth. OEM business starts in '25 and goes from there. We've got great collaborations and design wins in this place -- in this space, and we are going to, I believe, make a very big business out of this that will be additive to our new HPC business that we will rebuild with Gen 5. On the top of that, we have our industrial business, and we have our aerospace and defense business, both doubling in the next 5 to 7 years. So I believe the future for Vicor is very bright. We have the technology. We have the customer engagements, and I'm very, very positive, as I always have been, about the future of this company. Okay. So, with that, we'll now take your questions. Thank you very much.

Thanks, Phil. So, as Phil said, at this time, we'll take your questions. Patrizio primarily will lead the questions for us. Once again I remind everyone, we are in our Q2 quiet period, and accordingly, Patrizio will not be commenting on current operational or financial performance. I also remind you, as Chairman of the meeting, I will be following the rules of conduct distributed to attendees. So, please limit yourselves to one question with a permitted follow-up. With that, I turn the microphone over to Patrizio, and I would just like to ask anyone who has a question to please step to a microphone. We're recording this and it's very helpful to be able to hear you if you have a question. Please go ahead.

Bob Johnston with Herald Investment Management. Could we continue on the discussion about auto? And you talked about that ramping up in 2025. Can you give us an idea how that might play out in terms of sampling and starting to see revenues? And is it across different markets, North America, Asia, Europe or should we look for 1 specific market or 1 specific OEM?

Okay. So, I'll take the last part of your question first. So it's across all markets. So we have engagements in China, Korea, Japan, Europe and North America. I would say North America is really behind, lagging behind the other countries. I think GM, Ford, Stellantis are still really trying to figure out their electric vehicle strategy, but it's full steam ahead in China and Korea with what we've seen out there. In terms of product strategy, it's right across those power modules that I showed you. So it's not just 1 application. It's right across BEV, MHEV, PHEV and ICE. So it's very, very broad with those power modules that I talked about, both at 800 and 400-volt. And we've got design wins, we've got design-ins, we've got collaborations probably across about 15 OEMs and Tier-1s.

So, in a single model as a sole source or dual source or how do you have would that play out potentially?

In a single model sole sourced at the moment.

So, to be clear, to begin with, these are what you might call very high-end automobiles. They're not manufactured by the OEMs in millions or tens of millions of units per year. They're manufactured in thousands, possibly going to the 10,000 level. The content per car is significant, and if we gauge it in terms of revenue opportunity for Vicor over the next 5 years, it will ramp starting in '25 to, based on current design wins, the $15 million, $20 million level.

For the wins, yes.

In one notable customer application. So what is designed in and beginning to go into production, to be clear, is not mass market automobiles. They automobiles that value the attributes of power system technology that we can offer, density, lightweight, which contribute to range, particularly with the BEV.

I'm Andrew Fones, I'm with Brown Capital Management. I had a, I guess a question about your history. Congratulations on getting the chip fab open and up and running. But it obviously took many, many years longer than you had initially expected. So, my first question was like, what lessons did you learn from that experience and those delays? Obviously, it was more than just COVID.

Yes. So it's -- if we put the building our first chip fab into context, a proper context, I think we can be generally proud with what we accomplished, without question, there is a dose of optimism, right, that permeates many of us feel in particular, right? So -- but in the times that optimism, which is pervasive, leads us to anticipate something might happen faster than it can be made to happen, given the realities of the challenge. In a discussion before the meeting, I drew an analogy between our first chip fab and the presence in the semiconductor industry. And there's an audible presence in the semiconductor industry with wafer fabs. I don't remember how long it took to build the first wafer fab. Claudio might know, but it wasn't built or developed or brought to fruition in a year time scale because of the complexity of the processes, the equipment that needs to be developed. So, to be clear, as I was mentioning earlier, the way power converters are still manufactured today by the industry at large is using assembly techniques that are common denominator to the electronics industry. But building, let's say, GPU and building a power converter, particularly a high-density power converter where thermal management, being able to carry hundreds, if not thousands of amperes out of the package, these are very different challenges. And so the industry has been building and continues to build converters as if there were electronic assemblies for the most part. Yes, there is some specialty process that comes in, in the latter stages of manufacturing. Heat sink is attached some structural adhesive, thermally conductive interfaces. But for the most part, the methodology and the thought process that goes into the conventional assembly of competitive converters is rooted in the general practices of the electronic industry. And Vicor is the only company that realized a long time ago that when it comes to a density power converters with their unique attributes, we would need a very different kind of methodology, which you can see, to some degree, reflected in the video that was shown at the beginning. And the very first part of the video with electronic assembly, that was not really different and unique, but you saw in the latter part where large panel was going through chemical process steps that to build our fab, we had to bring about a high degree of novelty and one that gives us a level of scalability that would be highly leveraged both in terms of continue to advance the performance, but most importantly, achieve levels of cost effectiveness that come about with the scaling attributes of manufacturing power components in panels as opposed to making them without those kinds of techniques. So, no doubt, in answer to your question, we learned a few lessons, we always do. Things never go according to expectation, particularly if there is a healthy dose of optimism. But we got it done and we now have capacity that can support, we think, upwards of $1 billion worth of revenues.

And so my follow -up, you obviously have lost share, you talked about that on some of the leading AI boards. And this was during a time where you had like a design lead. You talk about still having a design lead, but what gives you confidence? What's different now that would enable you to regain that share? And should we be concerned that you mentioned some of the complexities and novelty of that manufacturing process that it's the concerns about those types of issues that have resulted in your lost share and potentially questions about whether you could regain it? Like how are you convincing your clients that you've overcome those issues?

So the share that we lost at NVIDIA, let's talk a moment about it. One needs to understand the cause of that -- in order to again frame the opportunity, the challenges in proper context. It wasn't that we didn't have the best technology. It wasn't that we couldn't enable a higher-performance GPU. It wasn't that in enabling a high-performance GPU at the same power level, we couldn't save 120, 150 watts worth of heat. It was a business decision predicated on making sure that there could be no capacity gates from single-source suppliers at the time when the demand for those kinds of products exceeds the ability to make them. So it didn't have to do with apparent density not being well ahead of the competition. I was having dinner with John Shen last evening who occasionally gets presentations from competitors or a variety of experts in the industry, and he pointed to the fact that he saw one slide with a variety of companies in the field, and Vicor all the way to the right with an entry in terms of power density stood well above the shoulders of the competition. So, we've had that all along. It's typically been not a marginal differentiator, but a very major differentiator. We don't like to just be 20%, 30% better the competition. We like to be at least 2x, 3x, 4x. And because we think ultimately, business competition is enabled by technology capability that when it comes to the power system, cannot be avoided in terms of it becoming enabling, meaning that if any company at any one point in time as a market lead and can, in effect, approach that opportunity by commoditizing in their product assembly strategy, the market forces will quickly bring about a number of competitors that see an opportunity, in effect, leveraging the best power system technology, among other things, in order to be able to offer products that are superior. So, those market forces, coupled with the strong technology advantage that we've had, and we are expanding with our 5G technology, I think the pillars of our expectations with respect to future success in the artificial intelligence field as well as the other markets that Phil talked about.

Hi, John Dillon. Can you talk a little bit -- give us kind of a deep dive into your NRE, the non-recurring engineering that you're receiving? Who's paying the NRE? What kind of opportunities are you seeing? And what are the future opportunities that we can expect from the NRE that you're getting right now?

Yes. So, we, I think, report NRE together with royalties.

It's separated, but I can describe -- I mean, we get -- it's a variety, and Phil can comment on this as well. So it's -- it can be aerospace and defense, John. Automotive would be a source of NRE and other customers. I don't know if you want to elaborate, Phil.

No, we've got a high-performance compute customer that also contributes NRE for special product development. So, yes, it goes across those 3 markets.

I think Vicor has a long history of being able to get customers to fund NRE.

I understand that. I'm particularly interested because the numbers seem to be growing. And NRE for me is really important because it really shows your technology advantage because companies are paying you because they need you and you're modifying the product. Am I wrong in this? Or is this how you see the NRE?

Well, yes. The NRE funds at times it's customizing, but it doesn't have to be. And some of the recent examples include space products, development for space applications. But you're right, though, there is a desire and understanding that Vicor has very, very skilled engineering talent and the development is worth funding oftentimes.

Phil, can you comment on the HPC customer and what they're using the NRE for and how this can translate into future opportunities?

So that's a VPD type of opportunity, which has been evolving. It's now in production and escalating production based on what Phil described as Gen-1 type of VPD. And now, as part of the development project plan to migrate to the Gen 2 5G-enabled type of VPD. And to set things in perspective, this will take the current density from less than 1 amp per square millimeter up to the 3 amp per square millimeter capability that the 5G current multiplier will support.

Hi, Jon Tanwanteng with CJS. I was wondering if you could discuss the NBM ITC case. Where does that stand today? And assuming you do get an injunction, how can you quantify what kind of benefit might follow that, if that gets a favorable ruling for you?

So, the ITC case on the NBM, to your point, is our first ITC case. We brought it in July of last year. We held a trial in, I think, end of April, beginning of May. There was a 1 week trial in Washington DC. All post-trial briefing is now complete, and the case rests with the administrative law judge, and the current expectation is that it's going to rule by the beginning of October. In parallel with that activity, we had to deal with a number of related matters. So, the defendants, which in the case of the ITC are also called respondents, scrambled to, in effect create diversion to some degree and present risk by launching lawsuits against Vicor in a variety of jurisdictions. They also brought a request for Inter Partes Review or IPRs at the patent trial and appeal board looking to get the so-called PTAB to find some of the claims or all the claims that are asserted in the ITC case to be invalid. We scored a major win a few weeks ago when the PTAB, in a matter of couple of weeks, ruled on each of the 3 requests or IPRs denying all 3 of them. I was mentioning earlier in the discussion before the meeting to so gauge our degree of success in these matters that statistically, when a party brings a request for IPR, that party has a 70% probability, based on PTAB statistics, of getting the case instituted; i.e. there is a 30% probability that the case will not be instituted. So with 3 attempts at 3 separate patents, our probability at priori of not having any one of them instituted would have been 30% times 30% times 30%, which is very, very small percentage. And remarkably, the [ PAM ] board found that there wasn't even a question with respect to any one of the claims and the institution was denied. So, that affirms, generally speaking, the validity of the patents and the risk that infringers face of being subjected to exclusion order by the International Trade Commission. And following that, damages, potentially willful damages in the case we brought in DC or Texas against the named infringers. And that's not a complete list. There is more to be added. So this is a first major case for us. Again, it's got 2 prongs. One is the ITC prong which seeks to get an exclusion order that will keep not just those companies, but their customers, including very large OEMs that manufacture or have manufactured and bring into the country literally billions of dollars' worth of server artificial intelligence hardware. There could be a risk with respect to being able to import that hardware. So that's one prong. And the complementary prong, which is the monetary damages prong. That's currently stayed pending the outcome of the ITC case in the district court in this district of Texas, in Marshall, Texas. So it's a 2-pronged approach. We've done very, very well so far. We feel good about the team effort, both in our offensive actions in asserting the patents, and all the defensive actions we've taken with respect to some preposterous initiatives that our opponents took. As an example, one of the respondents argued at the International Trade Commission that they should be dismissed because they had bought NBMs from Vicor and their purchase orders had some fine print to the effect that by which you're buying some Vicor product, they would get a royalty free global perpetual license to our IP, which really doesn't pass the common sense test. But they had the [indiscernible] to argue that at the ITC, the judge scolded them. They went to China, looking for an observing panel to -- in China, to, in effect, agree with them that somehow they acquired a license to our IP just by buying some products. And we had to go to District court of Massachusetts to get a restraining order, which we secured where the district court judge ordered them to withdraw their initiative in China. So in a number of venues, we've done very well. I'll conclude the commentary on this front by pointing back to the presentation with respect to technology and product strategy, and what I regard as a diversified business model that is leveraging the track record, the innovation, and intellectual property in a variety of ways, by having the best products, giving customers the greatest opportunity, but also of late, pursuing opportunities to monetize our IP portfolio with royalty income that has become substantial and likely will become even more substantial. Sorry for the lengthy answer to your question.

No, no, that was great. My second question is, how do you plan on addressing the second source issue with your largest potential customers and prior customers? Do you need to get the design wins first before going out and securing a partner or a second manufacturing site to address that issue? Or is that something you can set up now in advance of potentially getting ahead of that if they need it?

Do you want to take that?

No.

Okay. I'll take it.

I'll be too excited about it.

Okay. So, I think we have -- we are approached frequently with respect to that question. I think some of the competitors that have assessed the value potential of our technology and the market opportunity would be keen to partner with us. The challenge when it comes to alternate sourcing is always one of finding a structure that works in terms of giving customers what they need, right, in terms of scalable capacity and more than one supplier without undermining the profitability opportunity. And we think we have ways of accomplishing those goals. The issues are different depending on the end market. So, take automotive as an example, and AI or HPC is another example, Quite possibly in HPC, if we were to succeed with our initiative to prevail with the NBM patents, potentially with other patents that are key to those markets, there could be an opportunity for an alternate source to be a supplier to OEMs that have licensed under Vicor OEM licensing practice, and we have a notable one in that category. There may be more. If there were to be enough, we could have a partner that, even without paying royalties to Vicor, could sell into a large enough HPC market, the one that is comprised of the Vicor OEM licensees. Automotive is a different market with different issues. There, there is perhaps more opportunity for an established automotive supplier to come in, and in effect bring more value by virtue of their scale or their history with those customers. So, this is a question that needs to be approached with a combination of elements, and -- but again, we're very focused on making the most out of the opportunity with a very diversified strategy. Diversification is key, obviously, to avoiding disruption should the kinds of events that have happened with us in AI with a customer buying over $100 million worth of products per year going down to nothing, right? If we didn't have a very diversified business strategy, we'd be suffering a lot more and working on making the most out of them.

If I could just follow up for more detail specifically on the point of load products. Is it possible to, or maybe is the answer simply that the customer may have different SKUs. One of them involves your product and the other one may involve a more commodity style product, and that might be the solution.

Yes, that's also. We've seen that where customers have had different boards, and the Vicor board is higher performance, but they also have a second source board different layout. I mean, that's what they do. I mean they talk about multiphase being a commodity. But the controller chip that goes with the multiphase is different from company to company. So, that actually is a single source for that controller chip. So even though they think they have multi sourcing, they have separate boards because the controller chip layouts are different. So each board is different. So again, having a Vicor board with a higher performance SKU is another possibility.

It's all different because obviously, when a multi-phase solution is multi sourced and all of the suppliers that multiphase have essentially equivalent performance, the board itself has equivalent performance. In our case, a valuable position is to deliver higher performance. So the issues are a little different. But to Phil's point, the notion that a board is multi source, that's actually not really the case. You still have the logistical complexities, right? You have the flexibility of being able to go to different suppliers, but you're actually building different assemblies.

I think -- could you go to the microphone or -- there's a few people in the -- sorry, There's a few people that queued up back there.

Hi, Quinn Bolton with Needham. I guess my first question, you've talked about 2024 and probably early '25 as sort of -- you have a number of opportunities and a number of challenges. I think a lot of the near-term opportunities seem to revolve around the outcome of the ITC case and potential licensing deals that may come about to -- certainly, if you're successful. My question is, you've got a $1 billion revenue fab. That's beautiful, but you need to fill it. And so, to the extent that you are able to secure licensing deals, what's the likelihood you might be able to secure a portion of that customer's products as part of that licensing transaction?

So, our OEM license is already structured to incentivize customer's OEMs to avail themselves of outsourcing, but procure a substantial enough fraction of their requirements from us to discount the royalty rate. So the royalty rate is discounted up to a very large fraction. I can't tell you how much, but it's a very substantial fraction. As long as the OEM within the past year has procured modules from Vicor above a certain level, the tens of millions of dollars. And so that structure is really architected to, in effect, incentivize OEMs to take advantage of Vicor as a supplier of modules for very high-performance applications as well as being the repository of the enabling technology so that with a license, they can, without concern of disruption in their supply, procure from other sources solutions, which typically will not be at the same level of performance, but they may be good enough for some of the requirements. So we have the scaffolding built into the OEM license agreement.

Perfect. And you sort of touched on my second question in the response to the last question. But it seems to me that every vertical power design would effectively be a custom solution, especially if competitors have to use gearboxes. How does Vicor convince the customer to sort of give you a portion of their business sole sourced? What do you need to do from a customer service support quality? Is there anything -- what do you think the biggest challenges are to securing the 5G vertical power delivery opportunity?

So, to your point, the form of VPD that is practiced by our competitors, which is, as Phil described earlier, a stacked VPD, and is to say, at the semiconductor level, stacking is something that the industry has been pursuing for a variety of good reasons. But, when it comes to the power system delivery, having to stack a gearbox with energy reservoir, a large array, literally thousands of semi-capacitors under other layers, which, as practiced by our competitors, involve layers of inductors and layers of power stages. That stacking is very challenged in many respects. It's challenged mechanically because it's quite thick, typically around 5 millimeter, 7 millimeter thick. It's challenged thermally, because the layers generate heat, particularly the inductor layer, the power stage layer. And one of the challenges is to be able to extract that heat, to get it out from the sack itself. The mechanical thermal assembly yield challenges are really taking their toll with the very few OEMs that have gone down that path, because they needed to take that step in order to enable their GPU's, CPU's, what have you. What we understand is that their assembly yields are quite poor. They have a cost issue, because these custom assemblies are, by the standards of the industry, very expensive. So, I'm answering your question by pointing to the motivation these OEMs have to have a much better solution, because the first generation of EPD, through gearboxes and stacking, which Vicor pioneer and patented, right, for which we've not yet asserted patents, but we could. That's got a lot of challenges that make these OEMs a very difficult time. There's actually been one large OEM, an automotive company that went down that path and had to abandon it. The key people that expiated that effort got fired, and the company is now buying solutions from other parties because they couldn't make it work. So you can infer from those developments in the industry that first generation VPD is fundamentally handicapped, as first generation technologies tend to be, because fundamentally, they were brought about by an early process that was not as mature as a process required to enable a truly scalable solution, which is what we're now embarked upon with the second gen VPD and 5G kind of multipliers with much higher current density. So we believe that OEMs are motivated. There is obviously a challenge of single source and a single fab to manufacture those products. We're going to have to work those issues over time. As I was suggesting earlier, there could be an alternate source that makes the investment in a fab. And within HPC sells into Vicor OEM licensees, there is a lot that remains to be done in order to make the most out of the opportunity. But we have fertile ground because the OEMs that are, in a sense, compelled to take these risks and deal with these challenges because of their GPU, TPU needs are or should be very motivated for better solutions.

My name is Ron Feinstein. I'm a Private Investor from the 1990s. So my question is in the context of that, and I'm a 1% shareholder. So I have a vested interest in your progress and it's directed at you as CEO. It's a sales and marketing question. You're not a big company. The complexity of your diversification is overwhelming to me, as a non-technical person. Do you ever have strategic discussions about your satisfaction with the level of focus that you're giving to market niches? It's not a product question, it's a sales and marketing question, because of the limitations that occur. I've always found complexity as the enemy and focus is the value creation. When I see the slides and I see the range of opportunities, and it's not a question for Philip. I'm curious how you react to that.

That's a very good question. And what comes to my mind in answer to your question is that there is an underlying principle to our approach, which is reflected in the concept of component solutions for your power system. And you might have seen it through the slides that in an essential way simplifies the challenge. I'm not trying to contest your proposition of the complexity of the challenge. Without question, there's a good deal of complexity to the challenge and we are a small company with big aspirations that over the years, decades, have not been fulfilled. We fully appreciate that. But there is an underlying simplification, which has to do with the products, as power components for customers' power systems being largely common denominator, meaning the same type of power components, converter technology, control system technology underlying semiconductor components, fab methodology that is required in order to make cost-effective building blocks for fast charging or a 48-volt zone architecture in an automobile. That same methodology, but in terms of engineering development, product definition, fabrication in a fab is equally leveraged in HPC or the industrial markets or even aerospace, which gives us a level of flexibility and diversification with common denominator type of building blocks and methodologies. In one way, it adds to the complexity of the challenge, but it also has, I think, value in terms of simplifying it in many respects. Ultimately, if the industry and I have no doubt with respect to this outcome, comes around to treating power components like other components in the electronics industry from earlier decades, which had been standardized as, in effect, modular, leveraged building blocks. Ultimately, that will lead to the core answer to your question of addressing the large [ multiplicity ] of markets with all their niches and all the unique requirements with the common denominator strategy that leverages a number of benefits from all the common denominator attributes. But without question, this complexity is the challenge. I know the question was addressed to me. Maybe Phil wants to add something here.

Point of clarification. Sales and marketing is not going to come back and say that they don't have the capacity. It's just the nature of the beast. And the question deals with just how obsessively focused on the priorities. I hear the common denominator. I may even understand it. But do you feel that this sales force that you have and the marketing organization that you have isn't diluted with their talent and their focus?

So it's structured to be compartmentalized to a certain degree with respect to their respective end markets and opportunities. We just actually had the quarterly business review 2 days ago. And the presentations from the business leaders from each of these various end markets were focused on the respective challenges and opportunities, what they needed in terms of products in order to deliver to the targets of growth within the respective markets. And they are going about to the degree of understanding their customer base, their potential customers and their needs in a very focused way while communicating with their peers and the rest of us with respect to what the enterprise as a whole needs to do in order to enable their goals and their targets. Remarkably, there are common denominator opportunities where fundamentally the same products or very similar products are key to being able to achieve their respective opportunities. I understand that the underlying question, which I think all of us have in our minds, I think is, given the great attributes of the technology, the novelty, the ingenuity, how is it that we're still such a small company, right? You are a 1% shareholder. Vicor has got a small share typically of the market at large. We do have a larger share, in some cases, much larger share of certain end markets. We have had a large share in AI. We now have a very small share in AI. And there are explanations for all of these developments. So at the end of the day, the proof is in the pudding and this is all about turning all of these capabilities into revenues and profits, and we're very focused on that.

My name is Bill Dillon. John is going to say I'm the black sheep in the family.

No relation.

He's from Ireland. We're related. So I'm going to ask you a question. It might be tough, but I hope you rise to the occasion. Almost 6 years ago, in the Q3 earnings report, you said there's no competitor that's even close because we are, I could argue, 10 years ahead of the competition, and we have really 100 patents standing in the way of the competition, and we've been working diligently over a long time to planting a minefield for any unscrupulous competitor who would want to try and chase our tracks. Never mind the fact that there are fundamental technical challenges to doing that independent of the IP. A very strong statement.

That sounds like me.

It was you.

[indiscernible] who would have guessed?

As a shareholder, I can tell you, and I know others here really enjoy the conference calls because your optimism comes across and it's -- if you're along the stock, it's exactly what you want to hear. But I want to give you a point of reference, in that same October 2018, a competitor, Monolithic, had $389 million in revenues. You had $291 million in revenues. Today, monolithic has $1.83 billion in revenues. We have $391 million. You started to address the question that Jonathan asked and admitted some of the deficiencies. For shareholders, it hasn't been a lost decade, but it's been 6 years. Your stock at the time was $35 -- I'm sorry, it was $37 in October 2018. Today, it's $35. You did successfully sell, raised $110 million in 2020 at $65 a share. But everybody who bought into your vision then is sitting with a large loss. The stock, by the way, of Monolithic, is $852 today. It had been $113. And that isn't the only company that's done well. The Philadelphia semiconductor index has doubled in the last 5 years. So I get that you acknowledge we haven't had much success on Wall Street yet. And my question, which some of our friends from Wall Street will acknowledge, one of the most dangerous phrases is this time, it's different. And I think you want shareholders to believe that the Gen 5 and the focus of the company is going to be different. Can you tell us why you have confidence that you can deliver the Gen 5 product at a price point, I know you've outlined the advantages of it, and I'm sure the advantages are there. But how do you know you can deliver it at a price point that the market will accept and that will be profitable for Vicor shareholders?

Well, there's a lot of good points there and implicit questions or explicit questions. But let me start from the very last one. The fab, of which we saw glimpses, there was not a very comprehensive view of the entire fab. There's a lot to it that wasn't shown for a variety of reasons. That fab, we believe, has the cost structure, both in terms of the cost of processing panels for car multipliers or other building blocks, and the essential spread in terms of the [indiscernible] cost that goes into a panel to be cost effective for mainstream variable volume applications. In fact, I would argue, as I did in a discussion before the meeting, that Vicor is way ahead of the competition with respect to envisioning the need for a power component fab, a chip as in converter housing package fab and bringing it to fruition. So we're very focused on cost structure. We understand that ultimately, to succeed, we have to have a cost card. And we believe that we are uniquely equipped with our first chip fab to deliver that. Let me touch on another question, specific question, I think you raised regarding disappointment with respect to the performance of Vicor stock. And obviously, it's been a wild ride, right? Because back then, you pointed out it was in the mid-30s. We were, I think, at that point in time, on the eve of a major ramp with an AI customer. I think the stock got all the way up to about $160 and it's come back down to the mid-30s. What comes to my mind in that regard is that, if I go back to an earlier time, while a significant opportunity in terms of the technology and the products was being put in place, the stock was at $5 and sat there and nobody had any interest in it. There are times, as most of us know, where there are disconnects, sometimes major disconnects between value -- real value and stock price. I understand that obviously, the stock price is the best measure of value at any one point in time, but we also recognize that there are times of opportunity, and I believe this is one of them, where if we execute, and obviously, there are legitimate questions with respect to that, right? Given our technology capability and the general market opportunity, 5 years from now, we might be hearing about a similar statement, as you summarized from 2018, having been a predictor of the creation of great deal value. We have that opportunity. There is, without question, a lot of challenges, given the complexity of the task, there are risks. There are different business models, such as the one that you pointed to, which leverage copying, and achieving economies of scale and undercutting prices that have been remarkably successful in terms of delivering shareholders' value and I don't dispute that and I take my head off for the financial performance and the stock price performance, but that's a different business model, one that doesn't work for us. It's worked for those shareholders, and they may continue to work for those shareholders. Time will tell. We have planted a minefield. You can see some of the mines beginning to go off. So I think all the statements that I made back then were true, as I knew them at the time. And that's, I think, still true today. But, without question, there is complexity and unpredictability to everything we do.

I very much appreciate your sincere answer. One short follow-up, you've said that you think there might be a need for a second foundry and you've talked about licensing. If you only have a certain capacity, at what point might we see customers placing orders if there's going to be a shortage of your ability to actually deliver product? I'm talking about the Gen 5?

Well, so, we got to first go through the designing process and that's happened with some customers, smaller customers. It is happening with some larger customers. It will happen, I believe, with other bigger customers, as events unfold and strategies get modified in light of competitive developments. So, I think that we should be satisfied with the accomplishment of bringing together first chip fab. Nowadays, that fab is grossly underutilized, right, given our existing run rate, our backlog, the fact that book-to-bill has been below 1 for quite some time. But I have no doubt that fab is going to be at full capacity. And to your point, customers that have their solutions dependent on components coming out of that fab would want to be securing their capacity going forward, because there's going to be competition for that capacity as there was a few years ago in the middle of an AI ramp. We frankly were stretched for capacity to the point where we were not able to service the customer community at large, and that was very painful for us and painful for some of the smaller customers. So, all these issues are in a way with us and we're looking to address them with customer satisfaction in mind. As Phil pointed out, that's an essential part of a long-term successful business strategy and that may involve, before too long, bringing an outer source with a fab. But we may need some more running with our existing fab before a party is ready to invest $300 million in setting up in Japan or some other place, a second facility with similar capabilities.

Jim Liberman. As a layperson, I think that I heard you mention that you thought that the interim power supply solutions were unsustainable. And I wondered if you could comment on that, if you still believe that and give as much color as possible.

So the solution that you are, I think, referencing works to a degree, but could only be made to work to that degree by tottering down compute power and in effect, falling short of targets with respect to compute capability and the level of advance in that relative to the power generation. In there it is, I think, a very significant fact that I think should be taken into account in assessing a number of things. The opportunity for Vicor, the logic of what OEMs competing for a certain space might apply going forward. The fact of the matter is that in that particular application, with that OEM having a dominant position in that market, being able to deliver a level of performance that at this point in time is still unique, ahead of the competition, at least in some respects, is enough. And the compromise in performance relative to what could have been had with a better power system is an acceptable trade-off. But again, as suggested in earlier remarks, competitive forces are not going to allow that to be sustained going forward. It's not, as you pointed out, a sustainable strategy, because in the world of technology, as we all know, performance at good cost ultimately wins. And the competitive marketplace always brings that about. So, it's a compromise that the CEO and the Board of Directors made and I think one that can't be criticized in terms of their business opportunity at this point in time, but not a sustainable strategy for the long term.

Okay. As it appears there. Oh, sorry there's -- sorry, I was -- go ahead.

There is 2 more.

Okay.

My name is Fletcher Cole. I'm an individual shareholder. And Patrizio, I believe in your -- you alluded to the fact in a previous question that there are times when the stock market price does not reflect the underlying value of the company. And if I heard you right, I thought that there might be an opportunity for investors at this point in time. I believe the company has a share repurchase authorization currently outstanding, and I'm a bit perplexed as why the company does not complete that share repurchase authorization at this point in time.

Yes. So, personally, just a few weeks ago, I purchased, I think it cost me over $8 million to acquire options or exercise options in Vicor. So, in terms of my personal belief, I think that's the best indication of my sense of the opportunity. With respect to the company, we have a substantial cash position. It may make sense to do a buyback. I can't tell you if we are ready or close to being ready to doing that now, but it's certainly a consideration. So, I believe in the value. I think that there is great leverage. I think there are many scenarios that could lead to great returns, and these scenarios involve, again, a diversified strategy where we have actually a multiplicity of paths to great success, ranging, first and foremost, with adoption of our 5G technology in AI, HPC and automotive as well as the other markets that are growing very nicely. But also in terms of the parallel effort with respect to monetizing on our intellectual property that is now getting broadly infringed.

As a non-employee of the company, I don't get -- I don't have the opportunity to get stock options. So I'd like to see the company putting its money where its mouth is. Actions speak stronger than words, and I don't want the company to be compromising its financial strength in any way, obviously. But I think by merely completing the authorization that's currently outstanding would go a long way to reassuring some of our shareholders. And once that's completed, I would certainly like to see the Board reauthorize a larger authorization.

Hi, Michelle Waller with Segall Bryant & Hamill. I wanted to go back to the royalties. So if you -- if I think about the A100, right, I think that you guys are -- that's the share that you have today with NVIDIA. And if I think about the fact that your royalties have an incentive for NVIDIA to allocate more share to you guys for the H100 and presumably the B100. As the A100 goes away, wouldn't you guys, like -- wouldn't it make sense if there was an incentive that they would allocate you guys some share on the H100 or no? Like, I mean, I'm just trying to make a sense of when you're talking about the incentive structure of the royalties, why they wouldn't give you guys share for other generations besides just the A100?

So I cannot discuss royalties or a license agreement in the context of a licensee because the license agreement is strict in terms of not allowing any disclosure. So, you made remarks regarding the A100 and it's in the public domain. Obviously, NVIDIA itself publicized the fact that our solution was single source in that generation of product. I think it's no mystery that NVIDIA, in anticipation of the market opportunity and volumes that they could achieve with the H100 made a strategic decision to multisource. And that led to the current state of affairs where Vicor was not part of that H100 power system solution. At various points in time, there were prospects for opportunities for derivative products or other things. We have retained a presence from a hardware perspective. It's not on the scale of what it was, but it's non-negligible, maybe, I would say, substantial. And again, where it goes from here, time will tell. But, again, the major, I think, observation with respect to this evolution is that a unique set of circumstances and business priorities and business strategies played the dominant role with respect to steering the power system solution in the action that wasn't one of achieving the best performance, either for the power system or the GPU. It was really predicated on a pragmatic choice based on a set of priorities that I think makes sense and disputably logical given the value proposition that NVIDIA has been able to achieve.

Okay. As it appears, there are no more questions. I will adjourn the 2024 Annual Meeting of Shareholders. Thank you for attending.

Thank you.