Great, thanks. Thanks, everybody, for joining us. I'm Rick Schafer, Oppenheimer Semiconductor analyst. I'm joined today by MaxLinear CFO, Steve Litchfield. Steve's got a pretty long tech resume. I think most on the call probably know Steve. I think I've known him for 20, almost 25 years-
Long time
So listen, it's great to see you, as always, and thank you for joining us. And so if it's okay with you, I figure we'll just go ahead and kick it off.
Yeah, let's do it.
And then for folks in the audience, I'll try to save maybe five minutes or so at the end if there are any questions that come in. I'll be watching a board here, so I'll be able to see if you just throw your questions in the queue. So anyway, hey, Steve, maybe just start high level, if that's okay, like-
Mm-hmm
... macro cycle, kind of related stuff. I mean, you guys reported, I guess, 2-3 weeks ago, print guide were a little soft, but you certainly weren't alone in that. You know, a lot of folks have talked about just sort of a, kind of a slow snapback in demand. And, so I'm just curious kind of where you're seeing things at, at sort of that high, that sort of macro level, overall demand trends. And if in your answer, if you could sort of highlight any pockets of relative strength or weakness.
Sure. Sure. Yeah, yeah, no, it's a good question. We're, I think, like many others, we've kind of, I think you kind of said it right, I mean, a little bit of a false start. I think we've been working through inventory levels. I mean, that's clearly been the biggest headwind, you know, big cyclical upturn, and now we're seeing the ramifications of that. Felt like we had been seeing real good progress on burning down inventory. Frankly, we're still seeing good progress on that front. Demand has been, you know, reasonably good, in my opinion. Sell-through has been, you know, much higher than sell-in, and I think that gave us the confidence that, you know, that we'd see this burn down and growth kind of resume. It's just, you know, frankly, taken a lot longer than what was originally expected.
But, you know, there, there are different pockets of strength. Clearly, AI data center has been doing extremely well. Infrastructure, in general, has been doing okay. Telco, you know, telco spending has been soft, right? I mean, that's a big area, and it's, it's one that we have exposure to. Actually, most of that falls in our infrastructure area, but broadband connectivity, you know, has been tough. I mean, they, they typically take longer to recover, and I think some of the inventory that was out there was even higher than what was originally anticipated. I think a little bit behind. Well, you talk about some of this false start, I mean, ourselves, I mean, we thought that we would kind of get through this in the first half and start to see a recovery in the second.
You know, good demand, good sell-through, but just not enough. I mean, bookings have improved, but not enough. I mean, those are the type of things that, you know, I think that we hear internally. I mean, I look at a lot of CapEx trends overall, and CapEx seems a little softer, but not crazy soft. So I think you have to kind of dig deeper and see where those dollars are going. And clearly, you know, there's folks spending on data centers or on AI, no matter who it is, and so that's where a lot of those dollars are going. Whereas maybe they've, you know, been spending on infrastructure over the last 2-3 years, and so they've dialed back some of that, and so it's taken a little bit longer for it to recover.
But, you know, there's still some encouraging signs out there. I mean, we've lived in the semiconductor industry for a long time, so you know, these things are cycles. You get through them. I don't feel... You know, I've had a few questions, and I'll address it here. I mean, a lot of questions around, Well, has there been a structural shift on the broadband side? I don't think so. It's kind of the same dynamic. All the service providers are looking to have multiple, you know, suppliers. They don't wanna be beholden to one guy, so I don't think that's changed at all.
So, you know, I don't see any dramatic changes other than just kind of a typical, or I shouldn't say typical. It's rougher than usual, but it doesn't feel like there's any big structural shifts in the industry, though.
Okay. And I'm kind of jumping around a bit, but just since you-
No problem
... since you just mentioned broadband, I, I'm always curious. I mean, you guys, I think, have got a pretty hefty market share on the cable side-
Yep
... in that space. I think it's about half the market, but again, correct me, please correct me if I'm wrong. And I know there are certainly content or bill of material gains there, but, you know, when we kind of view that market, it seems like sort of a mature market. You know, you're still seeing cord-cutting and all that kind of stuff, and so how do you view that market for MaxLinear long term? I mean, you know, do you see that as, are there share gain opportunities for you there? Is it more like steady state share, and you're gonna grow through content gains over time? Or, you know, or do you look at it like a cash cow business? You know what I mean?
Where maybe it doesn't grow so fast, but it's gonna generate a lot of cash for you guys.
Yeah. It's a little bit of both. So at a high level, broadband, you know, it's always grown low single digits, right? It's never been a big growth business, to your point. But it's also been consolidating to the point where, you know, there's a couple of players, and so you don't have to invest heavily. There are long product life cycles, so you're only having to upgrade, call it, you know, every 5 or 6 years. And so it ends up being ultimately a good cash cow business. For us, there are certain pockets of it that we weren't investing or, I'd say we didn't have a very broad portfolio. We didn't have Wi-Fi, we didn't have Ethernet. And those are 2 areas that I would say are growing more than others.
I mean, there's certainly ASP growth on the processor side as well, that we're taking advantage of and will continue to. We, MaxLinear, have been a little bit more exposed to the cable side, but we've, you know, over the last, call it, 2.5 years, been very focused on PON and just fiber in general. That's where most of the investments go, and that market is twice as big as the cable market. We haven't exited the cable side by any means. And you've got an upgrade cycle there that's happening there. I mean, you're familiar, Rick. I mean, we always kind of think of this as we'd like to be that arms dealer. These service providers, if they're gonna go cable, then we've got solutions for that.
If they go fixed wireless access, we've got solutions for that. If they go PON, we've got solutions for that, right? So we don't really care, but clearly there's been more of a rollout on the PON, fiber side, and that's where, you know, we've been getting good traction with a couple of the big North America guys, but Europe is starting to ramp up as well. So that business three years ago was less than $10 million for us. It's this year it'll be a little shy of $50 million, but then next year, kind of going up $60-$80 million, as we have some new customers that are ramping. And so that's exciting, and as the industry returns, we'll certainly see more beyond that.
So some of the areas that we're underexposed to, we've got a lot of potential to grow, you know, much larger than the overall market. But, you know, but I wouldn't say we're necessarily pounding the table that broadband is the end-all, be-all. We've been actually investing heavily on the infrastructure front, in data center, in wireless infrastructure, as more, you know, growth potential anyway, and that's where a lot of the focus has been.
So, clearly you're looking for growth in PON. I guess you might as well finish the broadband segment-
Yeah, sure, sure-
Since we're talking about it.
Yeah, no problem.
So you see growth there, and then, you know, I guess that I'm assuming most of that is 10G PON, and I don't know what your split is, if you share that, sort of 10G, and then sort of maybe some of the slower speed PON. And, again, I know there's a lot of questions in here, but the last one really for me is, some other folks in the high-speed PON space are kind of overexposed to China. I think you guys are kind of unique. I believe you don't have as much China exposure-
Mm-hmm
... on the PON side. But again, correct me if I'm wrong, I'm just-
Sure
...curious, sort of, where you're seeing the growth.
Yeah, so the PON – so you're exactly right. I mean, the PON market, much larger... I mean, so for the products that we're selling, which I think you're actually referencing some guys that we don't compete against. So the PON side, I mean, the competitive landscape is primarily Broadcom, to some degree MediaTek, and North America and Europe are really much – I mean, they're the bigger guys there, and that's where we're capturing share. It's a market we've been underexposed to. It has similar dynamics, though. I mean, like a PON gateway is gonna have, you know, $50-$60 worth of content, you know, as we go forward, and which is very similar to the cable market, right?
So and then we're able to capture, you know, not only the processor, but you also get Wi-Fi, you get Ethernet, you get power. So all that goes into the gateway, and that's where we've been successful with a lot of those same service providers that rely on us, you know, much, much like they do in the cable markets, to have the, you know, the software, firmware capabilities and the support that's required as they, you know, do these larger ramps. So, we do sell into Asia, but it's, it's not as big of a market, and I would say it's a little lower end as well. But that's where you get into, you know, bigger SoC guys like, like Huawei or, you know, to some degree, MediaTek, you know, really pursue that market on a bigger basis.
But North America and Europe are the ones that are getting more traction for us.
Well, thanks. Thanks for that color. And maybe I'll circle to China now just quickly, 'cause I know, you know, U.S.-China, the tensions there have it certainly you guys are not unique in that it's weighing on your, your, your guys' business as well. And I think you said, and I was- I'm just looking at the number, but it looks like, you know, you kind of sized the, the impact at $5 million-$8 million in 2Q, and, and, I think about twice that for the whole second half. So really, my question is, are we at a new baseline for your guys' China business? Is... You know, would you consider that these levels de-risked? Is this the new normal, or, or do you expect to kind of mean revert to kind of where you were with your, with your China business?
Yeah. So two things here. I'll talk about the impact in the quarter and some of the export issues that we had in 2Q that you referenced. So we said $5 million-$8 million was the impact in 2Q. It came late in the quarter. And then, yeah, about twice that in the back half of the year. The way I explain this is that, like, tensions have gotten, you know, tough. It's at the high end as well as at the low end. I mean, you're seeing, you know, lots of regulatory challenges around processors and, you know, number of cores, things like that, that they're very much focused on. But they're also looking at low-end capabilities as well, and making sure that they're limiting that...
You know, to some degree, on what lands in China's hands. But the overarching theme is that they continue to be, you know, they kind of just keep tightening the screws around some of the entities, especially entities that are on the Entity List, right? And so that's kind of, to some degree, we've not gone into all the excruciating detail, but that was some of the dynamics about what happened in the quarter. BIS is also getting a little tougher to deal with, whereas historically, you could work through problems quickly. You know, you can't, and they would just rather just deny everything at this point.
So the good in all this, if there is such thing, is that this was a legacy piece of our business that wasn't. It wasn't a huge growth driver for us. It probably $15-$20 million a year. And so, but to your point on the baseline, this is probably the last piece of business that's of any size, and although I would say even at $20 million, it's probably not that big for us, but I think it now, with it out, I don't expect it to come back. If it does, great, but we're not counting on that, and so that'll limit some of the exposure to it. So that's the export side. Let me talk briefly about China, because that's not the only business we do in China, and we did reference kind of softness in general.
Where we have more exposure in China is probably a little bit on the industrial side. Domestic, I don't think it's a surprise here, many other peers, TI, ADI, and Microchip, some of these folks have seen softness, and globally, but then also domestically in China. So, so we have been impacted by that. I expect, you know, there continue to be some softness. I don't feel like it's... there's a ton of inventory or anything out there, because it's been soft for a while. So I think as we go into next year, I think our industrial end market, China and otherwise, kind of starts to get back to low double-digit growth rates, which is what we would expect more on a long-term basis.
And a question I probably should have asked earlier when we were talking about inventory and stuff, that you said the word and it jarred a memory. Where is the channel inventory now versus sort of where you'd like to, like, say, target it or where you'd like to see it?
Well, honestly, I feel like we're getting pretty close to, you know, normalized inventory levels. I mean, I think it was slightly elevated in Q2, but I would've expected us to be at normal levels in the back half of the year. But, you know, I think this is where the demand has kind of come into the picture, and if demand softens a little bit, well, now I got a little bit more inventory than I thought I did, right?
Right. Right.
That, to me, is kind of where we're at, so it feels like we're really close. You know, it's a big mix issue, right? So I mentioned earlier, a lot of customers are really counting on us having inventory, either us or a distributor or an ODM or something, and I think that's actually been depleted. We're getting to that level where it's a little bit, you know, challenging for them, and I think they're gonna get surprised in some cases. But that has to... That does mean that demand has to continue to be, you know, maybe not robust, but it has to be normal anyway. And if that does continue, I think you're gonna see some shortages out there in certain pockets. And then we'll get back to some normalcy, right?
I mean, we see customers ordering inside of lead times, and they're counting on us having product, and, you know, several customers even last quarter were burned by that. And I think more and more you'll see people saying, "Okay, I understand. Yeah, I got 20-week lead times, and I'll honor that, and I'll put the orders in further in advance.
So I know this is a weird question to ask right now, when everything's, you know, we're still kind of on the softer side of demand, but, I mean, you and I have been through multiple cycles in our career, right?
Sure.
The hockey stick. We never see it coming, right?
Yeah.
But it always comes.
Sure.
So how do you protect or prepare MaxLinear for that day when everybody's hair's on fire and you're getting those phone calls, you know? I mean, 'cause it always seems like that's kind of, you know, not only the sweet spot of the cycle, but-
Yeah
... supplies start chasing demand, but the great time for companies like MaxLinear to take share, usually, you know, historically speaking.
Yeah, no, look, you're absolutely right. These two things, I, you know, the sales teams kind of never project it right in neither direction, frankly, and they'll be late once again. I think that's fair. Look, I think we prepare in the areas where we see that it could take off quicker, or we may see tightness, right? So, I mean, you, you've already heard this in the industry, I mean, where you've got, you know, some of the lower geometries or, you know, 3, 4, 5 nanometer nodes where things are tight. It's also where data center demand is, so you've got to be prepared for that one.
But then there's other pockets as well, even in, you know, call it 60 nanometer, for example, that you've got to be prepared for, that could pick up because they just weren't, you know, anticipating it. I think right now there's sufficient demand out there. I mean, I'm sorry, sufficient supply out there at some of those older nodes, so I think that'll be okay, and I think we can be a little more patient. I think you've also got to, you know, you gotta work during these times, it's just incredibly important to stay close to the customer and make sure that those relationships are secure because they also don't know in a lot of cases, right, as to where that demand and how they're gonna to play it.
But where we can take advantage with a relationship and, you know, kind of get ahead of this, that's what we want to do. We want to get closer with them so that we know how to take advantage of the competition and have something when the- when the competitor doesn't. And there's definitely as a small guy, want to kind of take advantage of a situation where, you know, maybe that customer is being ignored by one of the bigger semiconductor guys 'cause they can't get supply or they're not focused on the right things. That, those are great chances for us to step in and become a much bigger supplier to some of these customers.
... Thanks. So maybe to back to some of your statements, and I'm curious on optics. I think you guys have said you're tracking to, like, and correct me if I'm wrong, but $30 million or so this year in optics revenue, and I'm curious, you know, if you were to break that out, how much of that is PAM4 versus other? Like, what are some of the other p- you know, major pieces-
Sure
... in there besides PAM4?
Yeah. So yeah, that's exactly right, Rick. In May, we upped our guidance. Originally, coming into the year, we thought it'd be between 10 and 30. We said in May that it would be above the $30 million mark. We didn't give an exact number, but it's going well, and we're in various phases of qual. The short answer to your question, it's primarily PAM4 DSPs. That is the majority of the product portfolio. That's expanding into several broader products, but also different applications, right? We also are mostly focused around the module itself, right? We're selling into a PAM4 module that's going into a data center, hyperscaler.
That's where our biggest focus has been, but that broadens out into AECs, AOCs, all kinds of derivatives there, as I'm sure you're quite familiar with. So that's where we've been working with customers. We're in various stages of qual. You also know well, I mean, while we focus on the hyperscalers and have great relationships there, we also have to be really intimate with the module providers as well, and that's a broad set of customers, right? Each one of these guys has three or four guys that they like to use. We need to have, you know, a relationship as well. We need to have qual slots with each of these module vendors as well, to really, you know, to drive growth there, you know, over the long term, so both relationships are important.
Is it, it sounds like it, it's sort of a back half loaded year. Sounds like you're building some really good momentum in that business. You know, can we... In terms of thinking of 2025, and I'm sure you're not talking a ton about that yet, but, I mean, in terms of looking at run rates exiting 4Q, would, would that be sort of a good baseline assumption for sort of how we—like, a trajectory, what we think of for-
Sure
... for-
Yeah, I mean-
... for 2025?
... I kind of think of it, if you're doing north of $30 million, I mean, that business ought to double next year. So I don't know, if you're up in the $60-$80 million range, I mean, you're more, you know... Most of this ramp is happening in the second half of this year, so you're probably exiting this year in the, you know, $15-$18 million, which, you know, makes you a lot more comfortable that next year can be north of $60 million. I mean, can it get up to $80-$100 million? Potentially, yes. I mean, the market next year - I think it depends a little bit on what these rollouts look like.
We'll move into production in the back half of this year, but then we have several other customers that will start to go to production in kinda Q2, Q3 of next year. And that's where some of the bigger guys, some of the bigger volumes can really take place. So, can we get to triple digits? I mean, that's exciting for us, but I think we can. But you know, a couple of things have to fall our way, and assuming the customers ramp in the timeframe that they've shared with us, I mean, that's definitely a possibility.
Is that primarily, you know, 800G, and by the way, I should have asked that question earlier?
No, it's okay.
Is 800G driving-
Yeah
... most of the ramp this year, or is it more like next year we see 800G really kicking in?
So it's primarily 800 to both of your questions, but I mean, there's definitely some 400 gig sockets that are still ongoing, so some of that's mixed-
Okay
... into it. And then, you know, and so then as we get into next year, you start looking at 1.6T. I mean, we did an announcement earlier about our 1.6T that is, you know, in design phase right now. We'll have silicon... You know, the market, other than NVIDIA, I'll say it that way, looks to more of that starts to happen in the back half of next year. If you look at the market data here, your bigger hyperscalers are a little bit later next year for 1.6T. That's the part that I think is has been pushed a little bit, but clearly everyone is working on 200G per lane right now, as are we, and very excited about that market. Tough technologies, they'll take a little bit longer.
I mean, you're seeing some of these push-outs on, like, Blackwell, for example. I, I think, I don't know, it's a real simplistic view, but, I mean, these are really hard technologies, right? And it's not uncommon to see some of these things push a month or two or four, whatever it is, and I think that's what the industry is experiencing a little bit. I think we just have to continue to stay focused on supporting our customer base and, and getting... Because there's a massive rollout right now, just on 800 gig, and I think that's where you've got, whatever, 80% of your resources just focused on making sure we're hitting production windows for customers.
Right. So 1.6T will be a material contributor, a material piece of the mix for you guys next year, or is it more like 2026 when that comes?
Yeah, I think, I mean, I think if you look at the market data, more of the market is gonna grow in 2026. I think for us, while we'll have products out, and, you know, maybe we realize some revenues in the back half of next year, I think it'll be pretty modest, and I think the majority of that ramp will be in 2026.
I get off the PAM4 thing in a sec, but I-
Sure
... I'm very curious on 1.6, 'cause you're hearing all these things about whether it's NVIDIA maybe with some, you know, designer packaging issues maybe delaying some volume there. But you also hear on the PAM4 side, especially at 1.6, that getting to 200 gig lanes, as you say, is no mean feat. You know, so you hear about lasers a lot, and, like, 200 gig VCSELs, and it seems like that's maybe not ready for prime time yet, if ever. I'm just curious your opinions.
Yeah
since they are on
No, I mean-
How, how do we get to the 1.6? What, what are the hurdles, I guess?
Yeah, I mean, I think everything that you're highlighting is a challenge, right? I mean, for us, it's the DSP, but are there laser challenges? Of course, there are. I mean, there are system-level challenges. There's packaging issues that you're hearing about right now, and so, yeah, I mean, all of this is an issue. Do I think they move into production? Yeah, I mean, next year, I mean, I think they do. And, but, you know, how fast is the rollout? I think the other thing is, you know, if you think of some of these closed-in systems, they're gonna have, you know, more ability to get something out, even if it's at higher power levels, et cetera, so, so they can get something done.
But the broader, you know, kinda hyperscaler universe, you know, may take a little longer, as it normally does. And, you know, to get the right interop to make sure that, you know, the lasers to the optics, I mean, everything, packaging, et cetera, yeah, I mean, it certainly can push a little bit.
No, that's, that's really clear. Maybe one last one. Just, I believe, but again, correct me, but I think you've got sort of three big enterprise customers, I think, today in PAM4. I think that's sort of right. So I feel like your mix today, backward-looking, maybe skews a little more enterprise. I'm curious as we move as 800G and then 1.6T become bigger contributors-
Yeah
... does that sort of shift your mix more maybe to-
Yeah
the hyperscaler-
I think
... CSPs?
Yeah, I mean, look, enterprise. I mean, we've got several Tier 2s that are ramping that are not your, you know, top three hyperscale guys. That being said, I mean, those, those are also... You know, we're working with all, all the big four guys, if you will, and so, but they're at least three of them... I mean, you know, you could argue Google has a little more of a customized approach, but the other three guys, I mean, they, they really demand a lot of interop, and they're, you know, more, more than likely, as you well know, a little bit delayed as they get a more robust ecosystem defined and ramp, which I think has been, you know, clearly heard in the market. I don't think I'm sharing anything new. But, but those are...
So, to your point, yeah, I mean, that is what will drive that revenue. And that's why it kinda starts to happen in earnest kinda mid-next year.
Okay. And is it, is it more, is it pretty balanced AOC and AEC, I mean, and, and maybe you're agnostic-
So, look, don't forget, I mean, the majority of the PAM4 market, 80%+ is modules, right? It's optical modules.
Right.
It's not AEC, it's not AOC, and that, that's where our focus is. Are our DSPs being used in AEC and AOC applications? Do we have customers doing all of that? Yes, absolutely, we do. But, but the sweet spot of the market is kinda where our efforts are, are spent, and that's where the majority of the revenues will come from initially. Of course, we're gonna have customers that'll broaden that out. And, and remember, I mean, AECs and AOCs are just beginning as well, and to some degree, slightly delayed. I mean, you, you've got a couple of customers that are. I mean, well, there's one competitor that, as you well know, has a, has a slightly different solution, and they can go to market sooner. But, but the adoption in general, is slightly delayed.
But again, I mean, I don't know that... I think we're well-positioned to take advantage of that, but really, you know, the dollars are focused around the DSP mo- the module itself, right? That, that's where the money is.
Right. But you didn't say it, but the AEC, AOC for you guys is 100% R&D reuse, right? I mean, it's basically-
Right
... just a new application.
That's right. Yeah, yeah.
It's just the same expansion.
So it's the same, same DSP. That's exactly right, so same DSP. So that's kind of an easy derivation. And keep in mind, all your module guys, but hyperscalers, if you will, I mean, they're all gonna adopt this also. And so, yeah, they're working with the same customer, and that's where I kinda come back to the importance of support. And it's also why incumbency can be very powerful because they're in the labs already, et cetera, right? So that's. I would say, that's been a challenge for us over the last year and a half, and I think we've really gotten ourselves over that. We've got, you know, started to really establish that trust with a lot of these customers, and they're seeing the... Had an investor earlier asking about that differentiation and why they choose us.
But, I mean, the power levels is, you know, this is a massive challenge for these guys, and they're looking for alternatives, and they wanna see, you know, more capability, and I think we're delivering that. I think we're showing up from a support standpoint, firmware, software. I mean, that, that ends up being critical as well, right? Not just the hardware and not just the power, but you've also be being able to support the customer matters tremendously right now.
Yeah, no, fair enough. And I wanted to come back just quickly, just one question on-
Yeah
... carrier or maybe one or two. 'Cause you mentioned that, I mean, look, it's been soft for everybody, I feel like forever, like, what, two years and counting-
Feels like
... or something. Right? So I guess I'm curious if you have any thoughts on, like, when that turns. I mean, does... Do you sort of envision wireless, wireline sort of picking up... sometime in the first half next year? I'm not trying to put anything on you-
Yeah
'Cause I don't think anybody knows.
Sure. Sure, I look, I mean, if I think of telecom, like our wireless infrastructure, I mean, we're seeing signs it's improving, but I mean, it's just—it's not. I wouldn't call it robust by any means. But it also doesn't feel like there's just massive amounts of inventory. So, there's some decent signs there. I mean, is it the first half of next year? I mean, I don't know. I feel good about 2025. It—does it happen in Q1 or Q4 or Q2? I don't know. It's really hard to call right now, especially kind of given, you know, how they're spending money, what do interest rates do, and things like that globally, right? So but, I just—I do feel continued improvement in the channel, and frankly, even in demand as well.
So, are you starting to see it, or was it just too soon yet, even in that business, to start seeing it in your order book or your backlog?
I mean, we're seeing it-
At this stage.
... but it's small. So how do you see it? I mean, you see some of these orders that you haven't seen in some time come in, and it's like, "Ah, okay, that that's one that we knew would be coming." What else do you see? You see expedites. I mean, that's always a kind of a good leading indicator where they were expecting... You know, they at the last minute kind of come in and order, you know, "Hey, I need a product in eight weeks." That I mentioned this example earlier, but I mean, that's always a good signal as well that, you know, they were waiting to order, waiting to order, and then they waited, frankly, too long. But those are the little things that we're starting to see.
And so as that becomes a more of a growing trend, I think we'll be a lot more confident as to, you know, which quarter that improvement starts to take off.
Fair enough. And then a quick one, if I could, on Ethernet. You know, I think, I think you mentioned, that's a $100 million business sometime in the next year or two, and I know you're, or I believe you're ramping, with a pretty big U.S. Tier 1, like mid-next year. Is that the kind of key driver there, or is there more stuff at play, more-
Yeah, so-
More things driving that?
So there, I guess there's kind of 2 or 3 areas. One is just, look, I mean, this business started out more focused on kind of the gateway, 'cause each one of these gateways has a 1 gig port. It's transitioning to 2.5 gig. So that, that's why the business started, you know, a couple of years ago, and that's happening. We'll see a recovery next year, you know, as that transition happens. But I think, to your point on where some of these other customers and how we're broadening this out, how do you get it to $100 million? We have said 18-24 months. So yeah, we do see that line of sight there. Enterprise switching guys, there's a big North America switching guy that has designed us in.
We do expect that to ramp next year, but there's a lot more enterprise switching guys that are looking-
Okay
... for alternatives, that are making that upgrade to 1 gig to 2.5 gig, and so that is exciting. It's in kind of a, I, I call it a niche part of the market. It's big for us, but it's also not something that our peers are really focused on. So I think we're able to service this market and, and really gain some market share here, and so, so that piece of it is exciting. You also have the industrial market that's going through the same thing. You know, you're seeing more and more networking out in these factory, you know, implementations, where they've had 1 gig ports, and they're all upgrading to 2.5 gig, as well as more and more data is necessarily for some of these applications.
So all three of those applications, I think, are growing, and that gives us the confidence that this can be a big grower for us going forward.
Fair enough. I'd be remiss if I didn't ask the CFO a couple questions on the model.
Sure.
Um-
Not, not a problem.
You know, you know, gross margin obviously down a little bit or guided down a little bit-
Yep
... in the quarter, I think, but still at a high number, right? 58.5%, I believe, so close to 60% still. You know, are we sort of... Would you call this trough gross margin? You know, and, or so we, have we kind of... And then basically, what, if any, changes have you made to your, to your target gross margin?
Yeah. So quickly, no change to the long-term target. It's mostly driven by the, just the lower revenue. I mean, I think our gross margin has kind of held in pretty well. We're seeing a little bit of a slip here as revenues kind of decline even further, and so the fixed costs start to weigh in. I mean, we're a fabless guy, so we don't have a ton of fixed costs, but there are some, and, you know, it starts to weigh a little bit. So we've seen it dip below 60, but I do think we start to see recovery next quarter, and for that matter, I mean, next year, you know, we're probably 59%-60%, I think, next year.
And then, you know, as infrastructure grows, as the revenue kind of recovers a little bit, we get back into the low sixties with the long-term goal of being in the mid-sixties.
Fair enough. Maybe my last one, just on OpEx. I know you talked about 25% cost reduction, I think, in fiscal 2025, but correct me, and I'm just curious, like, what areas are you focused on to sort of rationalize that spend? And it's always tough, right? When you're the size company you are-
Sure
... really a growth company, you know, the choices you have to make on R&D, like, I'm just curious how-
Yeah
... how you think through all that in terms of-
Yeah
-functions.
It is tough, but I mean, one of the benefits to being in the markets that we play in is that, you know, we're not on this, you know, this constant hamster wheel to some degree, like a consumer company or something, where you're having to turn chips all the time, and we've got a little longer product life cycles, so that's super helpful. We have had several areas like PON, like Wi-Fi 7, like the Keystone and Rushmore product for DSPs. So we've had some big investments, but, I mean, they were really expected to kind of moderate in 2025. So fortunate from that perspective, that's just kind of how the cycle goes from an R&D perspective. But getting to 25% cut is never easy.
Majority of that is in R&D. We've seen, you know, some good growth in R&D spend over the last couple of years in order to support some of the development products that we had. So, I'm comfortable. We're moving aggressively. We'll have those cuts done this quarter. There'll be some residual follow-through in Q4, but I think, you know, we should be on track to be at a $220 million run rate next year, down from, call it, $290-ish this year. So, I'm pleased with the progress we've made on the cost front. As you would guess, we're really focused on where are the growth drivers? I mean, data center, DSP, I mean, that's where we're gonna be investing. That's where those R&D dollars are gonna go.
Where are we slowing down or not spending as much? I mean, in some of our broadband markets, in some of our industrial markets. It just doesn't require that level of spend on a go-forward basis.
Got it. Fair enough. That's actually got us on time. We stuck the landing, as they say, so-
See, you're very punctual, as always.
Well, it's great to see you. Thanks for joining us.
Great
... and have a great rest of your day.
Yeah, we appreciate it. Thanks for having us, and, you know, happy to do any follow-ups. Thanks, Rick.
Great. Thanks, Steve. See you.
All right. Take care.