Good morning. Good morning. Thank you for joining us for the first day of Needham & Company's 28th Annual Growth Conference. My name is Neil Young, and I'm a Semiconductor Analyst for Needham & Company. It is my pleasure to host this fireside chat with Silicon Motion. The company is the global leader in supplying NAND flash controllers for solid-state storage devices, including SSDs and embedded storage. Silicon Motion supplies more SSD controllers than any other company in the world for servers, PCs, and other client devices, and is the leading merchant supplier of eMMC and UFS embedded storage controllers used in smartphones and IoT devices. Joining me from the company today is Jason Tsai, CFO. Jason, thank you for joining us at the Needham & Company Growth Conference.
Thanks, Neil. Always appreciate coming here. It's a great way to kick off the year, so thanks for having us.
Of course. So, Jason, to kick things off, could you maybe give us a brief introduction to Silicon Motion for those in the audience that are less familiar? What are the main applications and customers for your products today? How did the company evolve to reach $1 billion annual revenue run rate? Excuse me.
Yeah. So, you know, we've been around for about 30 years now, but we started developing NAND flash controllers probably about 25 years ago. We've been doing this longer than really anybody else out there from a merchant supplier standpoint. We manage more NAND than anybody else out there, even the flash makers included. We've been able to steadily grow and expand our footprint from cards and flash drives early on to eMMC and UFS, and then now, more recently, over the last many years, PC SSDs. And then now, more recently, we're expanding into the enterprise space. What we've seen over the course of the last 20-plus years is that the complexity around managing NAND is getting greater and greater. NAND itself, while getting cheaper, is also getting more difficult to manage, and with that requires more robust controller technology.
We've been able to, again, because we manage more NAND than anybody else, we have a better understanding of NAND than anybody else. We're able to apply the learnings we have from managing one vendor's NAND to another vendor's NAND and develop controllers that are better, faster, cheaper than anything else in the market. And as a result of that, and as a result of higher cost of development, shifting priorities in terms of operating expenses and investment priorities, we're seeing more and more outsourcing opportunities today than we ever have. And so that's been certainly what we've been excited about.
Not only do we anticipate continuing to scale beyond the billion-dollar run rate that we're guiding to for the fourth quarter here, but certainly expecting us to continue to further scale as we increase share across all of our businesses and as we expand into new opportunities in automotive and enterprise, industrial, commercial, et cetera, and so these are what we're excited about that we believe puts us in the driver's seat in terms of driving our own growth longer term, expanding in spite of whatever the end markets are doing, and being able to deliver alpha to investors and drive growth across a variety of end markets, including several AI-related opportunities.
That's great. Thank you for that. So I'd like to discuss your win with NVIDIA's BlueField- 3 DPU and the content opportunity on BlueField- 4. So first, could you explain what the win on BlueField- 3 entails? If possible, could you share any details that might help us put our arms around the TAM or revenue opportunity for SIMO?
Yeah. So I'm not going to talk so much about the TAM because it's really not my place to really speak about kind of how big this business is for our customer. But for us, we started talking about this opportunity here a couple of quarters ago. Initially, we started off as a boot drive. It's still a boot drive. It's typically a lower density boot drive. The DPU, the data processing unit, is a server component. And so with it, there is a boot drive that stores all the boot-up code, et cetera. And so that's what we're providing. Typically, it's a lower density solution. Let's call it anywhere from 128 GB to 256 GB. It's a good solution. We're building the full drive, so we're buying NAND. We're pairing with our controller, and we're selling that to our customer that integrates that with the DPU.
We're looking at ASPs here roughly kind of in the $20-$30 range because it's a lower density. One of the advantages of being part of this ecosystem is that it enables us to understand how the high-capacity drives that we're developing for MonTitan and how those drives interact with the DPU. And so being part of the DPU gives us a better handle on how do we develop optimized controller technology for those high-capacity drives, which in our mind is a bigger opportunity longer term. Now, what's happened over the last six months, because of the success of BlueField- 3, we're now being pulled into additional opportunities. So in BlueField- 4, in additional switch products that significantly scale up in terms of both volume and density, and hence ASP. So we're moving from 128 GB and 256 GB solutions to 512 GB and 1 TB solutions.
So significantly higher density, significantly higher ASPs. And that's certainly becoming a lot more interesting in itself. So the boot drive opportunity, while certainly initially it's great to be in the ecosystem, what's evolved is that it's allowed us to grow meaningfully the TAM within that customer, within that opportunity as new products ramp, new additional opportunities scale over the course of the next, let's call it a couple of years. We're very optimistic that this is going to be a very meaningful new revenue opportunity for us.
Okay. So it sounds like you hinted at it a little bit, but I just want to maybe dig a little deeper. So now that we know about your content, BlueField- 3, it sounds like there might be additional content opportunities on the next platform. So I know maybe you spoke about that a little bit, but maybe if you could dive into, I know you don't want to speak about the TAM, but maybe as much detail as you could on the next generation at that customer.
Yeah. So with BlueField- 4, with the additional networking switch products that we're being brought in on for the boot drive, we are seeing, again, higher volume opportunities as we go from one boot drive to multiple drives across these different products. BlueField- 4 is expected to be certainly a higher volume solution relative to 3. As Vera Rubin begins to ramp in the back half of this year, we're excited. I think from a density standpoint, as you go from a 128 to 256 configuration to 512 and 1 TB, that significantly increases the ASPs for us. And so we're certainly very excited as that begins to scale here later this year. And then going into certainly 2027, that opportunity further expands.
Great. Thanks for that. So NAND flash prices rose sharply in 2025 due to supply shortages driven by AI demand and continue to rise so far in 2026. I know on the 3Q call, management talked about how module maker customers prepared 8 to 12 months of NAND inventory ahead of price increases. Does this insulation mean you can continue to grow even if NAND prices stay high or supply stays tight? And if so, why?
Yeah. So look, for us, what we're focused on is things that we can control. And so what that means is how do we get more share, right? Independent of whatever the end markets are doing, higher NAND flash prices, higher memory costs, certainly there is going to be some element of demand destruction at the end of the day. But for us, it is more about how do we continue to gain share in client SSDs. Our new products are commanding north of 50% share. And eMMC and UFS, we're confident that we can continue to further increase our share over the next few years. So those are the things that we can control. As I touched upon earlier, we're now in an environment where memory makers are trying to balance between a wide range of investment priorities.
The cost of development of these controllers, even for consumer applications like PC SSDs, like eMMC and UFS, has only gotten more expensive. Between balancing the right investment portfolio at our customers and us being able to be part of their strategy longer term, we're seeing significant opportunities here to scale our share independent of what the end markets are doing. To your point, we do certainly a lot of our larger module makers have amassed a fair bit of inventory coming into this cycle. The flash makers were very clear over the last year that prices aren't coming up. The module makers have taken steps to ensure that they have adequate supply to last them for a fair bit of time. We're well positioned with those folks. We're well positioned with our flash maker customers.
Ultimately, at the end of the day, more than 70% of our customers' revenue is coming from OEMs, so PC OEMs, smartphone OEMs, server OEMs, these types of customers where they have better access to products, where they have stronger balance sheets, stronger financial health that can still weather a lot of these higher price solutions, and so we're excited that, again, based upon kind of our customer profile and their ability to weather this, that we'll continue to do well.
Okay. So I mean, are you seeing any changes in order and patterns yet, such as customers accelerating purchases or drawing down inventory due to the pricing environment?
Module makers have done this countless times, right? I mean, we've seen up and down cycles within the memory market for every two or three years, right? So it's not something that's really new. Certainly, the level of increase in NAND flash pricing is certainly a lot more aggressive than we've seen in past cycles. But I think they've all done a good job in managing that. I wouldn't say that we've seen any changes in the demand profile yet. We have seen some third-party research that would imply that PCs and smartphone markets, the growth rates may temper a bit because of the higher memory costs. However, again, back to kind of what my point was earlier is that we have no control over kind of how the end markets behave. What we do have control over is how can we gain more share?
How do we take more of this market? And that's what we've been honestly very successful in doing and laying the groundwork for the last few years that we'll really capitalize on here going forward.
Okay. So staying on a subject, I know in the past, management has commented that rising prices could actually push OEMs to adopt QLC NAND faster. Could you first elaborate why the shift to QLC NAND would be a benefit to SIMO? And then are you starting to see any of that shift?
Yeah. So QLC, 4-bit per cell QLC NAND, the challenge with 4-bit per cell is that the endurance and reliability tends to be a little worse than TLC NAND. However, cost per bit becomes also lower. So we are seeing certainly a lot more interest across all of our businesses. We're seeing it in PC SSDs, which we've been supporting for quite some time. We're starting to see it in enterprise for our MonTitan products. We're starting to see it in eMMC and UFS for our smartphone customers. And so QLC is certainly something that's been more top of mind for a lot of folks. QLC, especially in the enterprise applications, is very, very new. Historically, this is a market that's relied on TLC and MLC NAND for high-performance compute SSDs.
Our MonTitan product basically takes these lower-cost dollar-per-bit NAND, QLC NAND, and builds our customers can build high-capacity SSDs around that that can help basically drive high-capacity, high-performance, low-cost SSDs into warm storage applications that are ideal for AI inference and training, and so you get the best of both worlds, right? You get the high capacity, you get the low cost, but you also get the high performance of SSDs, and this is a new market that historically hasn't existed before. Either you pay a lot for high-performance SSDs that are low density, or you pay a little bit for high-capacity hard disk drives that are lower performance, and so now we're starting to bridge that gap with these high-capacity SSDs, and we're really excited about the opportunity here. In smartphones on QLC, historically, again, this is a market that's been driven by TLC NAND.
Just given some of the performance requirements around smartphones, TLC historically has been better suited for this. But due to improvements in controller technology, improvements in the quality of QLC coming out of the fabs, we're starting to see flash makers, excuse me, we're starting to see handset OEMs really seriously consider and starting to put in QLC, especially in more mainstream handsets that are more cost-sensitive. So over the course of the next few years, we do expect QLC NAND to also penetrate the smartphone market and expand further there. So our advantage here is that we've managed more QLC NAND than anybody, even the flash makers. We've managed more QLC across the board, everybody's QLC, whereas even some of the largest flash makers are only managing their own.
And so that gives us a really unique opportunity to understand the dynamics around what the strengths and weaknesses around QLC are. How do you manage that better? How do you accentuate the strengths? How do you try to mitigate or reduce kind of where the vulnerabilities are to drive with better error correction, with better wear leveling, with better performance characteristics than what you would kind of traditionally get out of QLC? And that's been extremely successful in us taking share and winning new opportunities.
That's great. So I know you guys have talked about also on the last call, NAND makers increasingly leaning on SIMO to provide controllers as they redirect their own R&D towards other technologies. I guess I think we know the answer, but maybe speak a little bit about that dynamic where NAND makers are focusing their in-house development efforts and sort of how that's allowing you guys to take share across multiple products.
Sure. I mean, you've got basically two types of memory makers out there, right? Folks that are also developing DRAM and HBM and folks that are just centered around NAND. If you're also developing DRAM and HBM, obviously there are significant resource allocation issues in order to make sure that you've got enough supply for the AI infrastructure folks that require increasingly more DRAM and HBM. And so the priorities have shifted significantly. If you're just a NAND-focused provider, the end markets for NAND are also shifting quite a bit, right? You're going from just PCs and smartphones to now more enterprise, more automotive, more industrial, commercial, all these other applications. One of the opportunities around NAND for us and for the industry is that NAND changes every, let's call it, two to three years, right?
Additional layers of 3D NAND create more complexity around the controllers needed to manage those products. So newest generation 3D NAND requires more robust controller technology. And that gives us an opportunity to step in. Even though the interface speeds are not changing, the interface standards are not changing every two or three years, the NAND is. And so that opens up opportunities for us to come in, to take share, to offload some of that internal development that the flash makers historically have taken on their own and bring that in-house to us and enable that outsourcing to be part of that longer-term strategy. I've been with the company for, on and off, for 17 plus years now. And I tell you, the amount of opportunity we're seeing today is significantly more than anything we've seen in the past.
We're not talking about one or two projects at a flash maker. We're talking about entire segments of the business that they want to outsource to us. So this isn't something that we're a band-aid solution or to get them to the next level. This is something where they're telling us, "Look, our priorities have shifted." While it's important for us to be in this market still, it's not important enough for us to do this on our own. So please come in and help us. And so the challenge that we've had isn't about being able to generate new business. The challenge we have right now is to be able to scale up resources enough to manage and to partner with our customers and take on more of these opportunities.
So, you've heard us speak to this on our earnings calls for the last many quarters. Is that we're resource constrained and we'll continue to hire in order to support that inbound interest across all of our businesses.
Great. I want to shift a little bit to the PC market. So SIMO made a big push into the high-end PC segment. Your eight-channel Gen 5 controller seems to be ramping strongly. Revenue for this controller grew 45% quarter over quarter in three Q and now accounts for over 15% of your client SSD sales. Additionally, you've talked about securing design wins with four of the six NAND makers and nearly all module makers for this product. So I guess, how are early design wins faring? And more broadly, management's talked about the high-end PCs being 10%-15% of the market. Do you expect to secure a large portion of that slice?
Yeah. So PCIe 5, the latest interface standard within the form factor for PC SSDs, this is an opportunity that we've been working on for a while. We've developed really innovative controller technologies first at the eight-channel level for high performance and then also a four-channel solution for more mainstream products that'll scale later this year. But these are developed and manufactured at 6 nm , which is leading edge for us. I know it's not leading edge compared to a lot of other folks in the industry, but it's leading edge for us. But development costs are getting more expensive, right? For us, 6 nm tape out is anywhere from $5 million-$10 million. So whether we're bearing that cost or a flash maker is bearing that cost, somebody's bearing the cost of these innovative controller solutions. We have four flash makers with the eight-channel.
We believe we'll capture in excess of 50% share on the high-end PCIe 5, certainly as well as the mainstream PCIe 5 that's upcoming later this year. We're seeing, again, tremendous interest to take over both TLC and QLC SKUs within the flash makers and expanding our roles here. And so it's a huge opportunity here that we're starting to scale. While we have already grown the eight-channel business, our PCIe 5 business pretty meaningfully throughout this year, that'll scale throughout last year, that'll scale more meaningfully into 2026 as you now start seeing PC OEMs bring PCs to market with PCIe 5 installed. And so we're now going to see that beginning to scale here in 2026, and they'll continue to ramp.
Second half, the later part of this year, we'll have the four-channel beginning to ramp that'll drive PCIe 5 from the high-end market 10%-15% to more mainstream segments of the market that will further expand our share longer term. Through our wins in PCIe 5, we anticipate being able to increase our market share on PC SSDs from 30% historically to 40% over the next few years as PCIe 5 ramps.
Great. So I know you hinted at it a little bit, but I wanted to ask on the mainstream. You've talked about some design wins in the product, sort of how you talk about the four out of six NAND makers in the eight-channel. In the four-channel, could you walk us through some of the design wins you currently have? And I know you talked a little bit about that ramp, but maybe how do you envision the ramp of that controller in 2026?
Yeah. So we'll start seeing early unit volumes late this year and then ramping more meaningfully next year. We also have four flash makers as a different set of four flash makers for four-channel versus the eight-channel. But these additional customers that we're working with and virtually every module maker as well will scale, and that should drive, again, that higher share gain that we've talked about over the next few years for client SSDs.
Great. And then last question on PCs, given the noise of NAND prices impacting PC pricing and ultimately demand, like we talked about earlier, what's your outlook on the PC market itself for 2026?
Look, we'll wait and see on kind of what some of the third-party folks out there talk about, the Gartner and the like. The initial numbers we've seen coming through is obviously that there is expected to be some demand destruction. But the reality is that most investors probably aren't investing in Silicon Motion for 1%-2% end market growth, right? That's not the value proposition that we're bringing to market. We talk a lot about share gains. They want to see that our SSD market share go from 30%-40%. They want to see our other businesses scale. They want to see new opportunities scale, right? That's what we're focused on. I have no control over memory pricing and how that's going to affect end market demand.
What I do have control over is making sure I've got enough resources to capture those additional opportunities that are coming my way.
Great. And then I know we talked about it earlier, but NAND flash makers walking away from the eMMC and UFS controller market. Can you talk about this dynamic, how you're capitalizing on this? Are NAND vendors basically outsourcing all of their mobile controller needs to you at this point? And then lastly, provide any quantitative sense of how much share you've gained in the past year or so in the eMMC and UFS controller market.
Yeah. I'll give you some examples of what we're doing here, right? I mean, we've taped out our new UFS 4 controller a year or a year and a half ago. We're working on our second-generation product right now for that. Samsung with their eMMC business for automotive, they've outsourced that completely to us, right? And this is going back to that dynamic I talked about earlier. As new generation NAND comes online, as older generation NANDs kind of go end of life, they need to bring on a new generation of controllers to manage NAND across all these different end markets. And that's certainly what we're seeing with Samsung, with other flash makers. We've seen module makers becoming a bigger portion of this market. I think historically, eMMC and UFS have been dominated by the flash makers.
We're now seeing module makers becoming more relevant, more germane to the supply chain here, especially within the Chinese handset OEM ecosystem. And so if you take a look out there, there really isn't anybody from a merchant controller standpoint that has really any experience with eMMC and UFS aside from us. And we've been doing this for, gosh, 10 plus years now. And so we started off with Hynix early on. We started off with Samsung early on. We went with Micron. We're working with additional flash makers now for UFS. So the opportunity for us continues to scale pretty meaningfully. But more importantly, we have this long history of experience of developing controller technology that there isn't any other merchant guy out there.
When module makers are now becoming more involved in the space, there really isn't anybody else out there that can step in and help them out. We've been able to capture quite a bit of share there. We're excited. Historically, we've been about a 20%-25% market share player in this market, which is virtually the entirety of the merchant market. We do expect to increase our share over the next few years to north of 30%. As we continue to scale new products, as we continue to win additional sockets, we're confident that we can continue to move that market share needle up.
And then looking beyond smartphones to the broader eMMC and UFS market, you've highlighted in the past several applications and markets outside of phones that are growing. Could you discuss how SIMO is tapping into these other markets and where you're seeing the strongest growth and any particular new use cases and embedded memory that you're excited about?
I wouldn't say necessarily there are new use cases, but certainly if you take a look at automotive, right? You're going from historically one storage device for infotainment to now upwards of 10 storage devices, many including eMMC and UFS. You're talking about new devices such as AR, VR, smart glasses that are doing well, that eMMC is an ideal solution for. So while eMMC standard is old, has been around for a very long time and from a smartphone basis have largely evolved into UFS, there's still a tremendous market, probably close to a billion units a year in eMMC. So while it may not necessarily be the newest, sexiest thing out there, it's a huge market opportunity that we are doing really well in here.
And so it's not necessarily so much about new products or new end markets, but just servicing the existing opportunities here within this market is already a massive TAM for us.
Great. And then I want to move to enterprise SSDs and your MonTitan platform. So you've been talking about MonTitan platform for enterprise SSD controllers. For those in the audience, maybe not so familiar, what exactly is it? What makes MonTitan important for your growth? How does it differ from traditional client controllers? And just talk about your competitive advantage.
Yeah. Enterprise is obviously a very different beast compared to consumer, right? The requirements around performance, endurance, and reliability are much higher, much, much higher than anything on the consumer side. We touched a little bit upon this earlier in the conversation when we talked about QLC, but as AI is becoming, is now front of mind for everybody, storage infrastructure, storage data infrastructure has been a challenge, right? I mean, you need faster access to more data more quickly in order to drive inference and training better, get better results more quickly. The challenge in the past is that you've got really high performance SSDs, but they're really expensive and they're lower density. You've got hard disk drives that are really high capacity, but lower performance, and so there hasn't been anything that bridged that gap.
With the latest generation of 3D NAND QLC, you're able to get upwards of 2 Tb dies for QLC. Why that's important is that with the 2 Tb dies, you can build high capacity drives, 128 TB and 256 TB drives. And at those type of densities, because of the smaller footprint, because of the lower power consumption, lower heat generation, lower HVAC cost, your TCO for using SSDs has now become pretty comparable to hard disk drives. So now you no longer have to make a choice between, do I go for something that's more affordable or do I go for something that's more high performance? With QLC, high capacity QLC SSDs, you can get the best of both worlds. And that's really what our MonTitan platform is designed for.
We went to a lot of these data center folks, flash makers, enterprise users several years ago and asked them, how do we differentiate, right? When we were starting to build this MonTitan platform, we knew enterprise was going to be big, but we wanted to figure out how do we differentiate, and so we didn't want to build a me too solution because then you just ultimately compete on pricing and that's a no-win situation really across the board. What we heard from folks was that QLC was going to become more important over time. And as I said before, we have more experience managing QLC than anybody else in high performance environments and cost-effective environments across the board, and so with that experience, we're able to leverage those five, 10 years of experience managing QLC to the enterprise platform.
And so being able to develop those solutions on enterprise has been significant wins for us. We've got two tier one customers. We've got four additional customers that have a long history of supplying tier one, tier two enterprises and CSPs. We're well positioned both with U.S. customers as well as Asian customers. And we're waiting for the industry. The industry is waiting for the ramp of these 2 Tb dies that'll be upcoming over the next, let's call it 6-12 months. We're confident that as these products ramp, as this technology availability improves, that we'll be able to significantly scale the MonTitan business starting later this year and certainly into 2027 and beyond.
Yeah. So just to put some numbers around that scaling, I know you talked about MonTitan contributing 5%-10% of total revenue by late 2026 or late 2027, I guess. Could you just talk about what needs to go right to achieve that number? And it sounds like you guys are on track at this point.
Yeah, we are. I mean, a couple of things that are kind of holding it back at this point is that you've got certainly the need for availability. While there have been some flash makers that are in early production, we anticipate broader availability across all flash makers to happen over the next call it six months or so. And so as there is more broader availability, as more multi-sourcing opportunities come to light, we see QLC adoption with MonTitan expanding rapidly later this year and into next year. We're also waiting on some of our customers who want to develop their own firmware or are developing their own firmware to complete that process and scale as well, right? So there are a couple of things kind of that are tempering the near term in terms of how quickly can it ramp.
Once these elements have been resolved, we anticipate significant scale over the next few years.
Great. Looking at your R&D investments, you've been clear that the company will continue to invest heavily into new technology. You noted that OpEx should roughly grow in line with revenue as you fund new projects. In terms of specific development activity, you are planning for further tapeouts in 2026, as you mentioned. So I guess how many tapeouts are currently planned for 2026 and then also for 2027? Could you help provide any rough timing on the 2026 tapeouts?
Yeah, we'll certainly provide more details around kind of what 2026 OpEx and tapeout schedules and timeframes look like when we report our fourth quarter results here in early February. But with that said, we have talked about the need for additional controllers because of new generation of NAND that's coming online. Those new generation NAND tends to be, again, more complex in order to manage, but also higher performance, et cetera. So you need a new generation of controllers. So we're building a new generation of UFS controllers. We're building a new generation of PCIe 5 controllers to support these next generation NAND. We also anticipate building a PCIe 6, taping out a PCIe 6 controller this year for our enterprise business. That's going to be at 4 nm.
As many of you know, 4 nm is not exactly a cheap process geometry, but this is something that we will have more color on for folks during our earnings call in February. We'll talk more about kind of the rationale as to why right now, right? PCIe 6 is really not yet in the market yet, but it'll continue. It'll start growing over the next few years. We want it to be well positioned. But given the high development costs, we want to make sure we provide the transparency to investors about why it makes sense, who the customers are, how does this scale, what does ROI look like, why are we confident that this is the right product, why are we confident that this is a good time for us to be putting those investment dollars in and kind of what that long-term ramp looks like.
So stay tuned for that. But it's an opportunity that we're really excited about, just given some of the inbound interest we're seeing with a lot of the enterprise-centric customers that we have. So our goal certainly is to continue to drive operating margin leverage. We've been investing in a number of new products that have really not yet seen any ROI yet, right? Over the last few years, we've invested in enterprise. We've invested in PCIe 5. We've invested in UFS 4. We've invested in a lot of these expensive six-nanometer products that are now just beginning to come to market. And so as those begin to scale, we're going to see a lot more of that operating margin leverage kicking in in spite of these higher tapeout costs, in spite of the higher team, bigger teams that we're trying to build out at this point.
But certainly the goal here is that with our fourth quarter guidance we gave, where we talked about 19%-20% operating margin, we're getting to the point where kind of we're near our historical norms of 20%-25%. Longer term, our goal still continues to be greater than 25% operating margin. There's nothing fundamentally changed in our business that says we can't get there. And so as these investments that we've been making now start coming to market and starting to scale, we see significant leverage happening over the next few years as that begins to ramp.
Great. And then last topic, we have to ask about the legal situation, the MaxLinear merger termination and resulting arbitration. Could you provide an update on the process? And then as you do, what is at stake here for SIMO? I know originally you filed claims for the $160 million termination fee plus additional damages. Should we be thinking in that ballpark or something a little bit more?
Yeah. So we sued for the $160 million plus fees and interest as well as significant damages. We haven't publicly quantified what those significant damages are, but significant. The arbitration process began in October as expected. Closing arguments will happen in March and then a decision will be rendered by the tribunal sometime in the next three-to-six months after that. So you're looking at kind of second half of this year before this all gets resolved. Not much more I can say comment on that besides just stay tuned as there's more updates that we can provide, we will provide.
Okay. And then finally, how would a resolution, say a cash payment factor into your plans?
We'll cross that bridge when we get there. I mean, I think historically, if you just take a look at our capital allocation strategy, we've been fairly balanced around dividend payments and share repurchases, right? Dividend payments we've been making for the better part of the last 10 years went from started at $0.60 a share to now $2 per ADS, and that's increased gradually over the last 10 years. If you take a look at share repurchases, while we historically have not had a consistent share repurchase program in place, we have consistently had periodic share repurchases programs. If you take a look at over the course of the last three to five years, about half of our free cash flow has gone to share repurchases. So it's a meaningful part of our capital allocation strategy. The third leg is acquisitions.
We historically have not been very acquisitive, but it's certainly something that we continue to look at as controller technology continues to evolve, as performance of storage becomes more intertwined with the interconnects. Certainly, there's potential opportunities for us to try to advance some of those synergies, and so those are things that we will continue to look at. I think it'd be very unlikely for us to go something, do something that's completely out of left field, but certainly something that we feel can be additive to the business longer term, so those are the things that we think about. I'm not going to comment specifically on use of cash with any given scenario, but those are the things that we think about in terms of capital allocation.
Great. Well, those are all the questions I have. Thank you for your time.
Thank you so much. Thank you, everybody.