This conference call transcript was computer generated and almost certianly contains errors. This transcript is provided for information purposes only.EarningsCall, LLC makes no representation about the accuracy of the aforementioned transcript, and you are cautioned not to place undue reliance on the information provided by the transcript.
spk01: Welcome everyone to today's webinar entitled Why Curve Earnings Results for the Third Quarter ended September 13th, 2023. My name is Jano and I'm your producer for today. During the presentation, all participants will remain on listen-only mode. If you require assistance at any time, please put the message in your chat box and send it to the host. I would like to advise all parties this conference is being recorded. And now I would like to hand it over to Julie Smith, CHEF Financial Officer. Please go ahead.
spk02: Thank you. Good afternoon and welcome to Vicor Corporation's earnings call for the third quarter ended September 30th, 2023. I'm Jim Schmidt, Chief Financial Officer, and I'm in Andover with Patricio Vinciarelli, Chief Executive Officer, and Phil Davies, Corporate Vice President, Global Sales and Marketing. After the markets closed today, we issued a press release summarizing our financial results for the three and nine months ended September 30th. This press release has been posted on the Investor Relations page of our website, www.vicorpower.com. We also filed a Form 8K today related to the issuance of this press release. I remind listeners this conference call is being recorded and is the copyrighted property of Vicor Corporation. I also remind you various remarks we make during this call may constitute forward-looking statements for purposes of the safe harbor provisions under the Private Securities Litigation Reform Act of 1995. Except for historical information contained in this call, the matters discussed on this call, including any statements regarding current and planned products, current and potential customers, potential market opportunities, expected events and announcements, and our capacity expansion, as well as management's expectations for sales growth, spending, and profitability, are forward-looking statements involving risks and uncertainties. In light of these risks and uncertainties, we can offer no assurance that any forward-looking statement will, in fact, prove to be correct. Actual results may differ materially from those explicitly set forth in or implied by any of our remarks today. The risk and uncertainties we face are discussed in item 1A of our 2022 Form 10-K, which we filed with the SEC on February 28, 2023. This document is available via the EDGAR system on the SEC's website. Please note the information provided during this conference call is accurate only as of today, Tuesday, October 24, 2023. BICOR undertakes no obligation to update any statements, including forward-looking statements, made during this call, and you should not rely upon such statements after the conclusion of this call. A webcast replay of today's call will be available shortly on the investor relations page of our website. I'll now turn to a review of our Q3 financial performance, after which Phil will review recent market developments, and Patricio, Phil, and I will take your questions. In my remarks, I will focus mostly on the sequential quarterly changes for P&L and balance sheet items and refer you to our press release or our upcoming Form 10-Q for additional information. As stated in today's press release, BICOR recorded total revenue for the third quarter of $107.8 million, up 1% sequentially from the second quarter of 2023, total of $106.7 million, and up 4.6% from the third quarter of 2022, total of 103.1 million. Advanced product revenue decreased 13.5% sequentially to 58.4 million, while brick product revenue increased 26% sequentially to 49.4 million. Shipments to stocking distributors increased 50.5% sequentially and 81.5% year over year. Exports for the third quarter decreased sequentially as a percentage of total revenue to approximately 62.8% from the prior quarter's 68.1%. For Q3, advanced product share of total revenue decreased to 54.2% compared to 63.2% for the second quarter of 2023, with brick product share correspondingly increasing to 45.8% of total revenue. Turning to Q3 gross margin, we recorded a consolidated gross profit margin of 51.8%, which is a 10 basis point increase from the prior quarter. I'll now turn to Q3 operating expenses. Total operating expense increased 7.7% sequentially from the second quarter of 2023 to $40.2 million. The sequential increase was primarily due to R&D spending and an increase in legal fees, which will remain at substantial levels through the completion of the investigation by the International Trade Commission of the unlawful importation into the United States of modules and systems that infringe the asserted VICOR patents. The amounts of total equity-based compensation expense for Q3 included in cost of goods, SG&A, and R&D was $693,000, $1,788,000, and $977,000 respectively, totaling approximately $3.5 million. For Q3, we recorded operating income of $15.7 million, representing an operating margin of 14.6%. Turning to income taxes, we recorded a tax provision for Q3 of approximately $1 million, representing an effective tax rate for the quarter of 5.9%. Net income for Q3 totaled $16.6 million. GAAP diluted earnings per share was 37 cents based on a fully diluted share count of 45,187,000 shares. Fully diluted EPS decreased approximately 3% sequentially compared to 38 cents in the second quarter of 2023 and increased approximately 825% from 4 cents per share earned in the same quarter a year ago. Turning to our cash flow and balance sheet, Cash and cash equivalents totaled $227.8 million at Q3. Accounts receivable net of reserves totaled $62.6 million at quarter end, with DSOs for trade receivables at 42 days. Inventories net of reserves decreased 1.9% sequentially to $104.6 million. Annualized inventory turns were 2.1. Operating cash flow totaled $23.8 million for the quarter. Capital expenditures for Q3 totaled $7.7 million. We ended the quarter with a construction and progress balance primarily for manufacturing equipment of approximately $26 million, and with approximately $16.7 million remaining to be spent. I'll now address bookings and backlog. Q3 book to bill came in below one, and one year backlog decreased 19.6% from the prior quarter, closing at $174.7 million. Turning to the fourth quarter of 2023, with reduction in backlog, including overdue backlog, we are more dependent on terms orders, and that results in less visibility to our near-term outlook. While that is the case, our current expectation is that revenue, gross margin, and operating expenses will be approximately flat sequentially. With that, Philip will provide an overview of recent market developments, and then Patricio, Phil, and I will take your questions. I ask that you limit yourselves to one question and a related follow-up so that we can respond to as many of you as possible in the limited time available. If you have more than one topic to address, please get back in the queue. Phil?
spk03: Thank you, Jim. Let's begin with an update on our high-performance computing or HPC business, which will continue to be our major growth driver in the next few years. For the next few quarters, we will be focusing in three key areas. The first will be ramping production of our Gen 4 48-volt bus converter and factorized power point-of-load products in our new chip fab. The second will be completing development of our Gen 5 factorized power point-of-load solutions and delivering models and tools in Q1 of 2024. The third will be continuing to expand our customer base beyond the major accounts that dominated our revenues in the HPC market over the past five years. Regarding revenues in HPC, we expect customers using both lateral and lateral vertical Gen 4 products to be in production through 2025. before introducing new processors utilizing Gen 5 vertical power delivery or VPD solutions midway through 2025. In reference to the expansion of our customer base, we have continued to have substantial discussions with large data center, AI, and network processor companies on their challenges in powering next-generation high-current products. All of these companies recognize they need scalable access to more adept power system technology to effectively address the technical and operational challenges of generative AI at scale. The one major technology challenge that is foremost in everyone's minds and heard repeatedly was power density and power delivery. Power delivery to the processors, power delivery to the AI accelerator cards and their rack systems, and last but not least, power delivery to the data centers while driving towards a carbon-neutral objective. HPC customers are becoming aware that our current Gen 4 lateral vertical solutions can reduce power losses in an AI-enabled data center by 1 to 3 megawatts while enabling high processor performance Future processors will, however, require full vertical power delivery, or VPD, to continue power loss reduction. VICO's first-generation VPD solutions introduced in 2020 required complex stack packaging to incorporate bypass capacitors in a gearbox layer due to insufficient current multiplier density. Putting their customers at risk, competitors are going down the same route with even lower current density based on multi-cell, multi-phase solutions running into greater mechanical and thermal challenges as they try to deploy VPD using thick, heavy, and thermally inept stack packages. Our Gen5 MCM technology steps up current density by over three times. and reduces the multiplicity of bypass capacitors needed, eliminating a stacked capacitive layer, and enabling a second-generation VPD solution to the power system requirements of the AI card that is thinner, more thermally adept, reliable, and cost-effective. Customer engagements for our Gen5 VPD solutions are happening at an accelerating pace. and our objective in coming quarters will be to secure significant design wins. In view of the current density and performance gaps enabled by our 5G solutions and evolving AI power system demands, I am confident that within a few years we will gain a dominant share of the AI power system market. One of the major objectives in the design and development of our 5G product line was to have a scalable, low-cost, short cycle time and vertically integrated chip fab with a short timeframe for capacity expansion. We will need this capability to meet the supply chain demands of our customers, who are in two distinct groups at the moment, based on different priorities. The first group, where the priority is supply chain flexibility, is focused on a multi-source, multi-phase VR technology And the second group, where the priority is competitive advantage from product performance, is focused on the power system attributes needed to enable superior AI. Due to competitive market forces, our belief is that the first group will soon embrace the level of innovation and scalability enabled by 5G power system solutions from a multiplicity of chip fabs. I am very pleased with our progress in other key markets, which are critical to developing a robust and flexible business portfolio of our own. We have reached a very important milestone in our automotive business development, achieving PPAP qualification for three of our platform power modules for 800 volts to 48 volt power conversion, which enables production for these products in the second half of 2024. As in previous quarters, we continue to develop new collaborations with OEMs and Tier 1 suppliers who value lightweight, high-density power system solutions. These collaborations will enable design wins for 2026 production and beyond. Towards the end of Q3 and early into Q4, we have seen demand strengthen in our broad industrial, aerospace, and defense markets for both large OEMs and smaller customers who purchase through our channel partners. Demand in China remains weak for both legacy brick and advanced products. And we have been shifting our focus in recent quarters towards the Korean and Asia Pacific markets, where we see new opportunities for our advanced power modules across a broad range of industries. Our new sales and marketing team in Japan has been making excellent progress, developing new pipelines of opportunity with large industrial automation accounts, and are on track to add significant revenues in the coming years. Momentum with our operational excellence initiatives continues with teams working on specific performance improvements with our top 100 customers. And with new products set to launch in Q4 and Q1 of 2024, we are on track to meet our OGSM goals. Customer visits from our top 100 accounts to view or audit our new chip fab have been averaging one per week in recent months. This pace will pick up in 2024 as we ramp production and have the ability to host more customers. Visitors have left impressed with our new fab and its capacity and scalability. They understand their need to access our FAB capacity for power systems with power density and performance attributes that cannot be supported by multi-source, multi-phase solutions that cannot keep up with our current density and PDN flexibility. They also understand our commitment to operational excellence as clearly reflected in the equipment, process, and systems of our chip FAB. Thank you, and with that, we will now take your questions.
spk02: Okay, operator, we're ready for questions now.
spk01: Sure. Everyone, if you wish to ask a question, please click on the rise hand icon at the bottom of your screen. If you would like to withdraw your question, please click on the hand icon again. Thank you. And the first question is coming from John Langton. You are unmuted.
spk08: Please go ahead. Thank you for taking my questions, guys. First one, I was just wondering how qualifications have gone at the new facility. Can you talk about when you'll start shipping volume through the vertical plating capacity and kind of the initial demand that you're seeing and the response that you're seeing from customers?
spk04: I think we had trouble hearing part of your question. Could you repeat it, please?
spk08: Sure. I was wondering how the qualification process went for the initial customers for the new vertical vertically integrated plating capacity and when you expect to start shipping volume on that new equipment.
spk04: Okay. So, as Phil pointed out in his prepared remarks, we've had visits at an accelerating pace, frequent visits, weekly visits. As he mentioned, these visits have generally gone quite well. Visitors are impressed with the equipment, the processes, the systems we put in place to achieve much shorter cycle time in scalable capacity. This is very evident to every visitor that has come through the FAB. The FAB at this point, in terms of its vertical integration, is what I would say essentially complete. That's not to say that all of the equipment has been delivered and installed. The strategy that the operations team has pursued in terms of enabling initial capacity has been to get the core equipment with respect to our two-dimensional integrated packaging capabilities in place while deferring some of the other measures in order to facilitate some of the challenges by some of the vendors. That other measure is coming in and is getting installed over the next few months. But while we are, in effect, not all done, we are done to a very large extent. We can, at this point, make complete units, including all of the three-dimensional package integration steps that are essential to our 4G products in terms of vertically integrating processes that have been outsourced. as well as enabling a 5G platform capability, which has some very exciting new process steps.
spk08: Okay. To be clear, are you already shipping, or are you still in the qualifications and maybe finishing out phases of the facility? Are you producing out of that piece that you've been working so hard to insource over the last year or two?
spk04: We are using it for some of our in-house needs, I would say most of our in-house needs at this point, and beginning to selectively use it for certain customers who are seeking short cycle time and rapid turnaround to service their needs, but not on a mass scale yet.
spk08: Got it. Okay. Just to follow up, Patricio, when do you think the book-to-bill will start to creep back over one and you see your backlog increasing again?
spk04: Well, that could happen relatively soon. It could take a little longer. We're not going to make any specific prediction. I think it's a complex landscape and there's a number of variables at play. So I think we need to be noncommittal with respect to that.
spk08: Okay, fair enough. I'll jump back in queue. Thank you.
spk04: Thank you.
spk01: The next question is coming from John Dillon. You are unmuted. Please go ahead.
spk05: Hey, Phil. I'm wondering if you can give us a little more color on the bookings. For example, how do the bookings look for the next sequestration? couple of quarters, are we gonna see sequential increases in the bookings, or do you expect them to be flatter down?
spk04: We may see substantial increases. Again, we don't want to be very specific because of uncertainties that are outside of our control, but there is a range of scenarios that include pretty steep climb, but then again, it might take a while longer.
spk05: Can you just give us a little more color on that? What's going on? It sounds like you could possibly see a nice increase in bookings, and I understand bookings. Trust me, I really do. But can you just give us a little more color on that? That would be helpful.
spk03: So, John, this is Phil. So, as Patricio said, it's a complex landscape of bookings older programs ramping down, newer programs coming up. There's a whole host of things out there, complex-wise, in terms of the deployment of next-generation solutions. So, as Patricio said, it's better to just take a little bit of a wait-and-see approach to that at the moment, given, as I said, the complexities that we're looking at.
spk05: I guess I understand that the problem is there's a bunch of people who are short on the stock, and The shorts are running around saying, you know, you guys aren't going to see any bookings increase and you're out of the GPU business. Maybe you can just answer that question a little bit better because they're pretty vocal about the demise of Litecoin. They're really betting against you and the company. So I'm just wondering if you can help us out a little bit here, give us a little color on, you know, what you see as far as GPU land is concerned.
spk04: John, as you know, I think you've been an investor for a long time. We don't run a company based on what the shorts might do or any near-term type of consideration. We take the long view. We are very committed. We believe we are in a very unique position, the only position that can support critical market needs. So that's how we run the company. That's how we operate. We're not going to get into any debate regarding that which to us is quite obvious with respect to the tenure of our technology, the need for it. The customers that are literally flocking to us in terms of their next-generation AI needs, it confirms what we believe and is subjectively supported by numbers such as current density and PDN flexibility and vertical power delivery capabilities that are without equal in the industry. That's what I think investors should be relying upon. Obviously, they have a choice to believe us or not believe us. We believe ourselves, and time will tell with respect to that.
spk05: I completely understand. I was involved in programs where there was voltage drop, and I understand that all too well. I don't know if all the investors do, but I completely understand. I understand how your technology is.
spk04: So if some don't, that's going to be too bad for them. But I think we need to move on to the next topic.
spk05: I'll be back in a few. Thank you very much.
spk04: Thank you.
spk01: And the next question is coming from Quinn Bolton. You are unmuted. Please go ahead.
spk07: Hey, guys. Can you hear me? Yes. Yes. I guess first question, you know, last quarter you guys seemed pretty excited about this new AI platform that you expected to ramp in the fourth quarter of 23 and throughout 2024. Can you just give us any updates on that program? How are you feeling about it? Is it still on track? You started to see more bookings associated for that. And a related question, there are now two platforms On the market, one by NVIDIA, one by AMD, that I think both are rated at 1,000-watt TDP. It seems like 1,000-watt is kind of the key power level where you guys bring some real advantages, especially with lateral vertical. Can you confirm whether this AI platform is indeed 1,000-watt skew?
spk03: So, Quinn, this is Phil. So the first part of your question, I think, was what we talked about the last call. And my prepared remarks, I basically said that the first area of focus was to ramp in Q4 our 48-volt bus converters and factorized power solution. So I hope that answers that question pretty clearly. The second, in terms of We're seeing 1,000-watt GPUs and network processors and all sorts of ASICs and mixes of CPU-GPU combos out there. As Patricio talked about, we have the technology to address those that offers, as I mentioned in my prepared remarks as well, significant power savings with lateral vertical solutions. of the scale of megawatts in data centers, which is really critical to any data center company. and we with our gen 5 technology coming along i mean the level of engagement now is very very high as i said we're having substantial conversations now with customers that will diversify us away from the two big guys that we've been doing business with for the next for the last few years so so as i said i'm very confident in the future and where we stand and uh it's a bit of a complex landscape as i said because you've got new programs Starting up, old programs ending. There's a bunch of other stuff going on out there with technology and product introduction. So, you know, as I say, we're confident in our position in the market.
spk07: Thank you, Phil. I guess a question, and this is kind of highlighted in, I think, some of your legal proceedings and ITC complaints against Delta, but it looks like one of your big motherboard customers has moved It looks like in scale to a two-stage architecture and away from factorized power. And I think Delta has come in and won some of the sockets away from your MBM. I guess, can you give us any sense of what's going on in the motherboard business? How much of that business do you think goes two-stage? How much of it stays single-stage? Because it seems like, you know, a big customer has moved away from VICOR, you know, with that transition to two-stage architecture and, you know, Delta's come in to do the 48 to 12-volt module.
spk04: So, as you may know, VICOR pioneered the MVM bus converter. We have substantial IP. We've asserted three different patterns coming at it from three different directions, and they're not the only patterns we have that cover that technology. We have licensed it. There are OEMs that are paying royalties for those NBMs, and they're the ones that they have certainty or continuity of supply There are other ones that have been taking chances, and those are going to go lying down in the next year. That's what's going to happen.
spk03: And, Quinn, I would say that when our Gen 5 technology is introduced to the market next year, what have been, you know, the CPU requirements have been creeping up, you know, 500, 600 amps now for some of these motherboards. I think our Gen 5 solutions will cause a number of customers to reevaluate that two-stage approach, given the performance that they can get from our Gen 5. So I think, you know, the market has moved from 1 to 2, and it'll go back to 1, in my opinion. Yeah, I second that.
spk04: So, to be clear, in answer to your question, the NBM is not, you know, as much as we love it, we invent it, right? But you'll hear from me first. It's not a long-term solution to 3 amp per square millimeter, high-density, VPD, scalable, robust, cost-effective EPD for Next Generation AI. It is not. It's an interim step. It's a sideshow. It's an attempt to, in effect, drive a multi-source, multi-phase capability that is fundamentally handicapped. It's not really turning out to be when you get up to the kilowatt level or the thousand amp level. truly multi-source, it has brought about significant trade-offs in terms of performance, which are not sustainable in a competitive landscape.
spk07: Got it. Thank you. Thank you.
spk01: Just a reminder, everyone, if you join using the phone line and if you wish to ask a question, please press star and three. And the next question is coming from John Dillon. You are unmuted. Please go ahead.
spk05: Yeah. On the last call, you stated you had a lateral and a lateral vertical opportunity with a major customer.
spk04: I was wondering if that was for the same... John, sorry to interrupt you, but it's hard, at least for me, to hear what you're saying.
spk03: Phil, can you hear? It's coming across a little bit muted and mumbled, John. If you can do something with your speaker there, that would be great.
spk05: Is this a little bit better?
spk03: Oh, that's better, yeah.
spk05: Okay. On the last call, you stated you had a lateral and a lateral vertical opportunity with a major GPU customer. I was wondering, was that for the same GPU or was it for two separate GPUs?
spk03: So what we talked about there was that we have lateral and lateral vertical solutions for not just one customer. We are bringing that solution forward, and we have customers looking at both of those solutions. Certainly, the number of customers looking at lateral deployments is a little bit higher than the number with lateral vertical, but we have both.
spk04: That's also the way in which systems have evolved, but make no mistake. The future, and the future is coming next year, is no longer with lateral PDMs. In fact, all of the design activities that we're engaged in at this point is beyond lateral PDMs. I'm on a call with an important potential customer tomorrow, we're not even going to or entertain a lot of opinion. It's either a lot of vertigo or what we call second-generation VPD, which is a more advanced, more scalable, more robust form of VPD that falls on the heels of the first-generation VPD that VIGO pioneered and companies have tried to copy.
spk05: So with the one major customer, though, do they have two different designs with you? One for, you know, are they working with the one customer? Do they have two different GPUs that are going to be using one using lateral, the other is going to be using lateral vertical? Or is it the same GPU that's initially going to be used lateral, then going to lateral vertical?
spk04: John, we're not going to talk about any one customer. I'm sorry, but, you know, bear with us. That's not a level of specificity that we want to get involved with.
spk05: Okay. And then one last question then. When do you expect the GMs, the gross margins, to improve with the new factory?
spk04: Well, I think we are on a general upper trend. That's not to say that we're going to see a positive improvement each and every quarter, but we have been in the low 50% range, and I think the three of us here, Phil, Jim, and myself, have line of sight to considerably higher numbers, partly due to the unique FAB capabilities that we put in place, a considerable investment, but also significantly in terms of the 5G point of load technology that takes us to a much stronger position in terms of not just performance, but cost-effectiveness. And some of that is going to be passed along to customers in the form of more cost-effective solutions. But at the same time, those more cost-effective solutions at the lower price point in terms of cents per amp to a customer will result in substantially higher margins for Vigo.
spk05: Great. Thank you very much.
spk01: The next question is coming from the attendee who joined over the phone. So please introduce yourself before you take your question. You are on mute now.
spk06: Hi, Richard Shannon with Craig Hallam. Can you guys hear me?
spk00: Yes.
spk06: All right, excellent. Let's hear a couple questions. Maybe following on the topic of bookings here, I think last quarter you felt fairly confident that the backlogs would improve either in the ending third quarter or fourth quarter. Now it seems you're less certain of that. And I think if I understood one of the answers to a prior question, you're still expecting the design with the major customer that is still active here. So it seems like a couple simple explanations for that would be either that design is delayed or Your share of the size of the opportunity is more limited than what you initially had thought, or potentially there's other reasons. So can you help us understand those dynamics relative to your last conference call?
spk04: I think I made clear that we really don't want to go into that level of detail. And to be clear, while I appreciate the reason for the interest, the curiosity, It's really got very little to do with Viagra's opportunity in the medium and long term, and that's what we're really focused on.
spk06: Okay. I guess I'll jump to another question here, maybe looking a little bit longer term. And, Phil, I'm going back to your prepared remarks here about time frame for five-year technology to be ramping. I think you said in 25, if I caught your commentary correctly. Is this the expectation of one that would intersect with the first ramp of a three-nanometer accelerator you see in the market, or is there some dynamic you're expecting to intersect with it?
spk03: No, I think that 5G is, I believe, looking at what it's capable of doing to the market is going to be very ubiquitous. I think we'll see. design ins as we've talked about even down in the you know the low hundreds of amps because of the density and cost effectiveness and performance level of the technology so it's going to be ubiquitous across all sorts of different high performance computing markets but with regards to what we see from a deployment perspective 24 will be obviously a big design win year for us with increasing number of customers and then a ramp to production in the 25 timeframe some customers early in 25 other customers sort of midway through the year so it'll be a mix and a blend of of both you know domestic and international customers actually so so again that that's that's going to set the scene for us to take really significant share in the ever-expanding AI market That's what we're looking at right now, Richard.
spk06: Okay, thank you. I'll jump back in the queue. Thank you.
spk01: The next question is coming from . You are unmuted. Please go ahead.
spk07: There we go. I guess just a follow-up there, Richard's question. You know, if 2024 is sort of a transition year, design year for Gen 5 and 2025 is the real ramp of the Gen 5 technology, would you expect, you know, quarterly revenue to sort of stay flattish at about current levels until you get to that Gen 5 ramp? Or do you see the opportunity that as some of the Gen 4 lateral and lateral vertical solutions come to market, that you could actually see growth in quarterly revenue. Now, I'm not trying to get you to give us a quarter when it may grow. I'm just, how do you think about revenue over the next four or five quarters? You know, do we need Gen 5 to ramp before we see a significant increase in quarterly revenue, or can that happen sooner with Gen 4 lateral and lateral vertical?
spk03: Thanks. So, Quinn, as we said, it's a complex landscape right now. You know, we've got confidence in our Gen 4 lateral and lateral vertical solutions. We're going to see how that goes, but I believe that that technology, again, as we talked about, you can't not save megawatts for your end customer and not consider it seriously for deployment, right? I just go back to that. So it's a very important step for us towards Gen 5. But we've got great products that we're going to be ramping this quarter and into next quarter. So we have a bridge, if you like, to the Gen 5. Got it.
spk07: Thank you, Phil.
spk01: The next question is coming from John . Please go ahead. You are muted.
spk08: Thank you. I was wondering, Jim, if you could break out the actual legal expense in the quarter and kind of what you expect the run rate to be as the ITC does its investigation over the next couple of quarters.
spk02: Incrementally, John, I won't quote an exact number, but it's in the millions of dollars, and it's very substantial for us as a smaller company. But we believe it's well worth taking on because of the strength of our position.
spk08: Got it. Do you expect it to increase from what you did in the past quarter here?
spk04: No. We're prepared to invest as much as necessary. Again, in one way of looking at it, the international property side of our business is profitable. The operating expense relating to a certain intellectual property is more than covered by intellectual property-related income. And we expect that to continue to move in the right direction, particularly as it becomes evident to OEMs who have been taking a wait-and-see attitude with respect to how this may all play out. that they're going to be left in the large by, you know, companies that have stolen our technology.
spk08: Got it. Okay. Phil, I was wondering if you could expand just a little bit on the enthusiasm you're seeing out there for the Gen 5 EPD. What's, like, the breadth of the customer that you're seeing today versus where you were, you know, maybe in the same time in the development cycle for the current gen of AI processors? Would you say that there's A lot more volume and opportunity out there today just because of all the, I guess, the amount of competitors that are out there, number one. Number two, the size of the opportunity increasing. Just help me size, you know, what that opportunity looks like, Gen 4 versus Gen 5, if you can.
spk03: Yeah, the current density improvement, well, first of all, you've got the drivers of the much higher currents coming along for all of the processes across networking, AI, CPU, right? That's the first thing. Secondly, the market's expanding at an incredible rate, particularly on the AI side. But as you add more AI, you also need to add more CPU. You also need to add more networking. So everything's sort of getting the big uplift. Now, then, you factor in that our Gen 5 has got three times the current density over the Gen 4 technology. And the reaction to that, and we've shown a lot of the customers, actually, when you go down the list of the majors in data center, AI, processors, or network processors, we've shown them examples of the mechanical examples of the packaging that we're going to have. I mean, the reaction is astounding. They're astonished that we can put as much current as we can into that small of a package. And if they're looking at VPD today, because that's where they're going to go, as Patricio said, all of the conversations now are about VPD. It's not about lateral. The current Gen 4 is lateral vertical, but Gen 5 is all about vertical. They look at that and they say, my God, I don't need to disturb my capacitive layer underneath my board. I can put one of these tiny little current multipliers there. I don't have to worry about mechanicals, thermals, you know, everything sort of fits. And so the reaction is incredible when compared to companies, similar Chinese companies that are trying to copy our MBMs, trying to do this thing with VPD, copying basically the wrong thing, which is a stacked package, really heavy, They're really terrible. You know, they're leading customers down the wrong path, and the customers can see that when they see our Gen 5 technology. So the excitement is really incredible, and I'm very confident, as I said, we're going to take major share here in this market.
spk04: First Generation VPP is an horror show from a manufacturability perspective and from an IP perspective as well. So, there too, unscrupulous foreign suppliers are putting, not just themselves, but major OEMs at substantial risk with respect to continuous supply, both in terms of, you know, the viability of the solution from a robustness perspective. and from the perspective of viability from an IT perspective or non-infringement perspective, I should say.
spk08: Got it. So fair to say that the pull is much stronger on Gen 5 than it was in the Gen 4, Gen 4.5 that you have out there?
spk03: Yes. I think that customers are going to get both performance, right, and power savings, and we talked about all the advantages of reliability, of cost, that add on to all of that. And with our scalable chip fab, we're going to be able to work with customers on flexible supply chain solutions. So we're sitting in a very good place, John.
spk08: Okay, great. Thank you.
spk01: Next question is coming from the attendee who joined over the phone. So please introduce yourself before you have your question. You are unmuted now.
spk06: Hi, Richard Shannon from Craig Allum again, guys. Can you hear me still? Yes. Okay, great. Thank you. Phil, I wanted to ask you about one of your statements in your prepared remarks here about 5G technology and the kind of targeting two different groups with different priorities here. The last part of that statement was you believe the first group will embrace level of innovation for 5G from a multiplicity of chip fabs. Since you only have one at least that you've talked about or in plan, can you parse the statement a little bit more and tell us what you mean by that and over what timeframe we could see a multiplicity of chip fabs, please?
spk04: Well, I'll try to answer that question. Again, even the a lack of sharp visibility with respect to that. So bear with me as I answer your question in general terms. As you know, it took many years and a very substantial investment to bring together a first chief. In some respects, this initiative is similar to the kinds of fabs most of us are familiar with, semiconductor fabs. And as in that case, the technical complexity, the complexity from the perspective of equipment and processes is such that bringing to closure a first fab takes, again, considerable time, persistence, and investment. But the great news with respect to having done it is that replicating it is something that can be done on a much shorter cycle time with a high degree of profitability, and the investment involved while substantial. It's in the hundreds of millions. In the right perspective, from the perspective of OEMs wishing to have, you know, flexibility from a sourcing perspective, given their system capabilities and valuable positions, in that perspective, a relatively small investment. So we do anticipate more than one FAB coming to the market to bring about an ecosystem where large OEMs that cannot, for good reason, rely on capacity from just the first VICOR FAB will avail themselves of capacity from power facilities.
spk06: Okay. I guess we'll parse that statement and ask about it in the future. Patricio, thanks for that. I'm going to follow up on one of your answers to a prior question here talking about I think, and I'll probably not get your wording correct here, but you're only engaged with customers that are interested in vertical-only solutions, either first or second gen here. Is that essentially saying, are you saying that you're not seeing anyone with a sufficient level of power, and I'll use 1,000 amps as maybe that demarcation line, tell me if it's different, that are not looking, are you telling me they're not looking at lateral at all, or you're just talking with the guys that are only looking at verticals?
spk04: So they are typically looking at programs on different timescales, in some cases a few months, or the better part of a year, or a year and a half, with different kind of requirements. And, you know, our focus is to engage where we can contribute substantial value through, in particular, 5G solutions. with their unique set of attributes, and those are particularly differentiated when you get past lateral, right? It can be lateral-vertical. We demonstrated that with lateral-vertical relative to lateral at current levels of roughly 1,000 amps, you can achieve something of the order of nearly 150 watts PDN, both direct and indirect power settings. Those are very substantial. And those, when applied across data centers, result or can result in the megawatt scale settings that Phil referenced earlier. So there's a valuable position there. And that can be done with our 4G technology, even though the current density of that technology is, you know, a small fraction of current density with 5G. So it is a depth, a lot of vertigo. There's a valuable position there, and we can engage there to enable customers to achieve their objectives. But the much bigger opportunity is looking a little beyond the next few months or six, nine months. of systems that are going to go into production. As Phil said, late next year or early in 25, there were, with 5G MCMs, MCMs per square millimeter, there are all new world or capabilities that are enabled, and in turn enable solutions that are around, have got all the right attributes, right? They're scalable, they're robust from a mechanical perspective, they're robust from a terminal management perspective, and they're more cost-effective.
spk03: So, Richard, this is Phil, just to add on to that. A couple of weeks ago, I got invited to a big data center company's strategic supplier day where they had their executives presenting on the future roadmaps and the challenges of uh building ai out in their existing data centers as well as adding more data centers going forward and as you can imagine as patricio mentioned here we're talking about data centers of the 20 megawatt plus and getting power into those while also trying to get carbon neutral is a massive challenge. So if you look at the amount of compute that's going to be needed to support AI going forward, both on the CPU side and on the GPU or ASIC side, plus the network processing, even if you save 10 watts or 5 watts per the CPU, That's a massive savings when you multiply the amount of compute in a data center. It gets over megawatts very quickly. So even to save 10 watts or 15 watts in a 400-amp or 300-amp CPU motherboard, and you can do it vertically, why not do it vertically if you can offer that to your customer base? So that's the change that's going on in the market. To add to that, historically, the answer to
spk04: to fields is a very good question. Why not do it that way? Is that you couldn't do it that way. You need technology that has got both the high current density, the high bandwidth, that enables, you know, a significant fraction of the caps they started getting taken out the socket to be swept away.
spk03: So power savings is going to be it as this stuff gets deployed, because you can't power it. You're competing also with all the EVs out there. The same grid supports all of that, plus the data centers. So they're also trying to use renewables, but really significantly reducing the impact, you know, on a megawatt scale is massive value add for any of these data center hyperscalers. You know, it's really important.
spk02: Okay, Operator, I think we're going to have to take Maybe one more question, please, and then we'll have to wrap it up.
spk01: All right. So the last question is coming from John . Please go ahead. Your line is unmuted. John? I'm sorry. I didn't have my hand raised.
spk08: That wasn't me.
spk01: Oh. Sorry. In this case, we have no more questions.
spk02: Okay. Thank you. Thank you, everyone, for joining the call.
spk01: Thank you, everyone. That marks the end of your webinar. Thank you for joining, and have a nice day.
Disclaimer