All right, good morning, everybody. Welcome to day one of the TD Cowen TMT Conference. My name is Matt Ramsay. My endeavor to cover the semiconductor market. Sometimes we do it well, sometimes not. But thank you so much for your attendance and your attention, and I'm thrilled to be here with John Forsyth from Cirrus Logic, CEO of the company, and we're gonna have a little chat about the company in general. I would just say the normal disclosure languages around Cirrus and around TD Cowen, and what we do in research and whatnot are in all your conference materials and on the website. So now that I got that over with, welcome, John.
Thanks a lot, Matt. Yep, delighted to be here, and that was very self-deprecating, but we think you do a great job.
I just wanted to kind of step back, John, and you guys are going through, I think, this year, refreshing some of the major content that you have with your large customer. It kind of occurred to me that some of us that followed the company for a long time might take some of that content for granted, right? It's been relatively stable for a long period of time. But maybe you could just, like, spend a couple of minutes talking about what the sort of core audio and amplifier products look like, and what kind of a generational jump we're gonna get this year.
Absolutely, yes. Yeah, thank you. So the products that you're talking about, the core content are codec, smart codec and boosted amplifiers. And these are some of the longest-running content we have. They've been through multiple generations in certain fruit-flavored devices that have been shipping for many years. We started shipping a codec, a custom codec, to our largest customer, about, you know, between 15 and 20 years ago, for that device. That's gone through many iterations. Over time, it gained boosted amplifier, the content gained boosted amplifiers as well to drive louder output from speakers than you could from a codec alone.
So those devices remain the core of our audio content in smartphones, and this year we have a refresh of both the smart codec and the boosted amplifier products coming. And so the smart codec is a 1-to-1 attach rate. The boosted amplifiers is a little bit different. They drive stereo speakers, so there are two of them for that, plus haptic feedback, so there's a third one, so it's a 3-to-1 attach rate there. And this has been a pretty significant R&D investment for us in both of those areas of development. Partly because in the case of the smart codec, we've migrated that from 55 nanometer down to 22 nanometer, which is a very big lift.
It's like you say, it's you know, because it's not a new socket to us, that could almost fly under the radar, but it's a huge investment. It's a really, really great investment from the point of view of the long-term revenue for the company, because we tend to see these products shipping for five years plus, and we can see that out into the future. And being on 22 nanometer allows us to deliver a lot more functionality, and get the customer to build more features around the product.
Anything that I guess this audience should expect hugely different from a content, ASPs, margins perspective as you go through that refresh?
Well, we'll certainly see some ASP accretion, yeah, and that, that's really reflective of the fact that there's a lot of innovation in those products, even though they're replacements for products that are there today. There's a lot of R&D investment that goes into that, and, you know, we need to get paid for that. We also feel that we're bringing a lot of additional value to the customer. So even though they have those products already, you can always deliver additional value, both through features and through integrating other components from around our chip on the board. That's one of the things we look progressively generation after generation to do. It saves board space and complexity for our customers. It grows our die size. It grows our the number of functions that we're integrating in our chip.
And so that, again, would be something that we, that will drive some ASP accretion in, in this cycle. Yeah.
Kind of working through the portfolio, I guess it was three years ago. I'm trying to remember where you guys introduced the first sort of camera processor in with the customer, and there's been, unlike what's going on in sort of the core codec and amplifier space, a pretty exciting content expansion and feature expansion path in the camera side. So if you could just maybe speak about that a little bit. I know at the time, when it first came out, it was very hush-hush as to what the features were going to be and the functionality, but I think we've all been able to see multiple generations of the phone now. So, if you could just spend a little bit of time on what that roadmap looks like.
Yeah, yeah, absolutely. Although I just wanna go back to your implication there that our codec and amplifiers were not exciting. We're very, very excited about that content. But I do get what you mean, which is that really in codec and boosted amplifiers, the rate of innovation had slowed somewhat. If you think about the early era of smartphones, to begin with, the audio was just very, very limited. Then boosted amplifiers allowed louder playback, then you got stereo, then you got haptics, and so on. So there was a much greater, much higher rate of innovation. That probably kind of got closer to the asymptote over the past 5-7 years.
Whereas, to your point, in the camera content, we've seen a very aggressive kind of innovation and a very aggressive schedule of updating camera components for us. So we introduced our first camera component a few years ago, so three or four years ago, and we're already on the third generation of that. And what our camera controllers do, principally control elements within the camera module, relating to stabilization and auto focus or moving. The way to think about it is we're very, very good at motion-sensing movement and moving small elements around with incredibly high levels of precision, and incredibly rapidly. And that translates into a number of things, but one of them, for example, is very high quality stabilization.
So you can take great night shots, or if you have a telephoto lens or a longer focal length, which typically shows up camera shake much, much more, you can stabilize that and take great shots from a distance. So we've been through this iteration, series of iterations, where to begin with, we were just doing optical stabilization, then also the sensor shift stabilization and autofocus, and then most recently, some of the most innovative features in the most recent generation of our largest customers' devices. In the Pro Max series of cameras, there's a very, very innovative prismatic lens there, which also relies on our camera controllers to drive some of the elements within that assembly.
So we've seen over time, there are multiple kind of dynamics or multiple contributing factors to our value accretion in the camera space. One is adding more features at a higher ASP. Another is, as those features cascade, we get a greater total amount of content across all the camera SKUs over time. And that's been reflected in us just seeing the value, the total value of our camera content increase year- on- year, each year we've been shipping that. As we look out to the future, we don't see that letting up. We see a really exciting roadmap.
Mm.
There's no doubt in my mind the camera will continue to be a kind of a marquee feature of these devices, a key differentiator for our biggest customers, and we will be kind of central to driving a lot of the features involved in delivering that experience.
Got it. No, that's really helpful. First of all, everybody, if there's questions in the audience, please just wave your arms around or yell at me or something, and we'll get those addressed. I wanted to kind of think about the camera evolution. One of the things that you and I have been talking about for years is to... There was a big push, 5, 6, 7 years ago, of trying to take the core audio codec business and amplifier business into the Android market. We had lots of fits and starts there. Are the dynamics different in camera?
Like, independent of some of the audio stuff, I mean, if the Cupertino continues to innovate on camera and the Android camp has responses at the high end, are there opportunities there for your company just in that space, or maybe not?
It's, like, it's not inconceivable, but I would say, when we stack rank all of our opportunities-
Yeah
... that doesn't make the cut.
Got it.
And it's not that we couldn't bring great features there, but we can't do everything. And one of the things that I think Cirrus does really well is focus. We talk about focusing on our biggest bets, resourcing them to win, and making sure that we, you know, we drive long-term success and value with key customers. I think with Android, long-term value has been sometimes a challenge as well-
Yeah
... because every generation needs to be won again. You get swapped out for somebody who's three cents cheaper. You know, and that is particular kind of market, which is actually very, very different from our, the smartphones of our largest customer, where tends to be a huge kind of front-end discussion around what you're gonna build, but then once it's in, it tends to ship for quite a long time. And that really helps us invest pretty, pretty heavyweight R&D in being an innovation partner for the customer. And I think that works really well in that environment. I don't think that lends itself so well to Android. There'll be opportunistic kind of stuff we do in Android. As you know, we have plenty of business in the audio space there.
The camera space, it just hasn't made the cut today.
Got it. That totally makes sense. Just popped in my head, you had mentioned the, the exciting stuff around, codec and amplifier, and, maybe I underplayed that a little bit, but I, I have a feature request.
Sure.
If you could somehow make it to where when my mother talks on speakerphone in a public place, that it would shut the phone off-
Yeah
... that'd be, that'd be pretty swell.
All right. I'll do my best.
It's endemic. Anyway, I wanted to really dive into the— you mentioned placing big bets-
Yeah
... and putting all the energy of the company behind the big bets. You have a couple of things that are, let's say, relatively newer into the portfolio. One is to go after the PC market, one is to go after some of the power conversion spaces. Whichever order you like, maybe we can kind of walk through some of the new bigger bets you're making.
Yeah, absolutely. And they are quite different, but very complementary. Maybe just to back up a second, when we talk about our strategy, we typically talk about three kind of prongs to that. First of them being, maintain and build our leadership in smartphone audio.
Yeah.
The core new products that I talked about earlier on, the smart codec and the boosted amplifiers, are really, really kind of enshrine that. The second prong is about expanding our content within smartphones beyond audio, in what we call the high-performance mixed signal space. And then, the third prong is leveraging a lot of that IP across audio and high-performance mixed signal into new markets. So in the case of the PC market... just to take that first, yeah, we're very excited about our momentum there. But one of the key attributes is that we're able to leverage a lot of the IP that we've developed for our smartphone products into PC products, which may need to be repackaged, may need to be...
You know, they are dedicated products for PCs, but they're leveraging a lot of IP investment we've made over the years. So in the PC space, you know, we're still in relatively early innings, but we have a great pipeline of opportunity, and we think we're building really good momentum. We see a lot of customer activity driven by the reference design, Intel's Lunar Lake reference design that we're where we have multiple pieces of silicon on there. That means we're gonna see, you know, we see Cirrus silicon coming to market this year in products. We've already seen that, but we're gonna see a lot more as we get into the next calendar year.
I'll just park the PC stuff there for a moment. I'm sure we can get into it in more detail. But to go back to your point about power, yeah, when I say high-performance mixed-signal and expansion within the smartphone, camera represents a significant component of that, haptics, too. But one of the biggest kind of new unconquered or only partially conquered bits of territory for us is around power. So, we did a few years ago ship our first power conversion and control chip. Since then, we've been making a lot of IP investment in developing additional technologies around the battery, both upstream and downstream of the battery. We see that as an area where we have something unique and innovative to bring to the market.
So that represents a pretty significant chunk of what drives our SAM expansion over the coming years.
Maybe just to pick up on that last point, we within investor circles in my conversations, we have a lot of conjecture about what is AI gonna be in a smartphone, what is AI gonna be in a PC? And I kind of come at it, I don't claim to know, but I come at it from a little bit of a technical angle, which is, in these devices, you can pretty much tell that to bring on board on-device inference, the processing and the memory parts of the device are gonna have to take more power.
Yeah.
Then, I would imagine that the people that make these devices with limited battery life are trying to figure out how the heck to get power from everything else.
Right.
Is that short summary the driver behind what you guys are investing in?
I think that's a very fair analysis. I would throw into that, board space as well.
Okay.
So I think the requirement for additional onboard processing, if you're gonna, like, yeah, to take a simple example, you're gonna have some kind of generative AI capability on the device. You're gonna be running a model-
Mm.
... that has, that's gonna require substantial memory and, most likely, additional processing cores relative to what, what are there today. All of that's gonna require more space as well as more power. And so we think of, power and board space as being at a premium, probably that premium increasing relative to-
Yeah
... the way it's been over the past few years. That creates a huge opportunity for us. For a start, everything we do is incredibly power efficient. So if you compare Cirrus mixed signal devices to our competitors in various categories, we're typically significantly ahead of them in terms of the geometry that we work at. That's expensive and it's hard, but the way it benefits you is greatly increased power efficiency, higher processing density, so the ability to do more processing within a given space, and the ability to shrink the chip further. So we are really, really well positioned to deliver some of the key benefits that are gonna be required, these key attributes that are gonna be required as a consequence of on-device AI.
Specifically in the power space, I think what we have done in the power conversion and control chip is actually a really good kind of signpost to some future opportunities. Because, one of the things that that chip is doing is managing the state of the battery and how the system draws charge from the battery during peak power events.
Mm.
So there are various things which drive kinda peak demand on the battery. One of them is in fact the use of kinda neural cores. You know, when you fire all those up, the lights everywhere else in the system kinda go dim. And that can damage the battery, but it can also really harm the user experience elsewhere. If you're trying to use the camera at the same time and it starts to you know, drop frames or something, that's a terrible user experience. So what our chip does is manage the power across those different demands during these peak events in a way that preserves the battery health and preserves the user experience. I can only see that getting more important-
Yeah
... over time.
Yeah. No, totally makes sense. Please.
Just as a follow on the kind of AI question, if you look at some of the recent demos by some of the, you know, tech companies, it seems like they're thinking the way in which you interact with your phone is probably going to change and be more back and forth from a voice perspective. Can you just talk about any future opportunities or anything you're working on there with respect to how the, you know, interaction with the phone or the PC changes?
Yeah, I certainly think anything that drives the voice interaction side has the potential to benefit Cirrus. We are in the voice path, so ironically, you can hear that my own voice is recovering right now from a childborne virus over the weekend. But yeah, we're in the voice path, and actually we added significant value to the system during that first era of the voice-driven interface. You know, when the first voice assistants came, there was voice wake, a kind of novel feature and so on. All of that centered around the codec and the kind of processing that we were able to do on the codec.
So I don't want to talk about specifics in relation to that, but we do think of that as being a fundamentally interesting set of opportunities there, yeah. And the same would go in the PC space. I mean, I was kind of referring to smartphones there, but we're now shipping codecs into a lot of PCs. And again, all the microphones are connected to our codec. We're going to see all the signals, means we can do a lot of, a lot of useful, valuable processing.
Getting onto the PC conversation, I guess I have two questions and they're, I guess, different topics. But one is, the motivation in the PC industry to start to pull the high-performance stuff that you guys have into the PC market. Like, just at first principles, like, what, what's the motivating factor that—I mean, Intel's pulling you guys into reference designs, and there's pulls now. Is it, as simple as Apple has 10-hour battery life, and the Windows ecosystem has to figure out what the heck to do, and now they're trying to... I mean, at the first order? Or, are there consumer features that require more audio? I'm just trying to get to the root cause of—'cause I've been asking you, and asked Jason the same thing before, like, there's these op-adjacent market opportunities.
Like, it seems like that pull could have happened before, but it's now, it's happening now, so just trying to wonder what the difference is.
Yeah, both, both the pull, and we have, we have very deliberately set out to, to chase them as well.
Okay.
So that's been a very conscious ambition and choice on our part. I think for many years we were running just to stand still to keep up with our largest customer. I think we achieved a kind of scale of R&D where we could actually say: Okay, like, now can we leverage more of this IP into attacking other markets? So that's a big part of what's driven our engagement with the PC landscape. But there has been a pull as well, and that's twofold. In the audio space, that's very much driven by the experience, driven by a couple of things, but one above all, the experience of having meetings through the screen, which-
Okay
... obviously accelerated very significantly with COVID. Coming out of COVID was kind of when we started engaging with our laptop customers, so we're in design with all of the top five laptop vendors. And we heard pretty consistent stories from all of them about the fact that, you know, in the past, it just didn't matter. I mean, honestly, PC-based audio was not good, let's say. Whereas that had started to show up because nobody bothered with PC audio, so it didn't, so nobody was aware that it wasn't that good. When you're having meetings through the screen every day, both audio on the input side, so the quality of the processing of the voice signal, and on the output side, so can I clearly hear everybody that I'm talking to?
Is there adequate volume to have a meeting with a group of people? All of that started becoming significantly more important. At the same time, yeah, the competitive benchmark moved quite significantly with custom silicon from our largest customer. That meant that you needed to be able to deliver more in more processing or more capability for longer in a thinner chassis. These things like, you know, pick one.
Yeah.
It's very, very hard to achieve all of those, but our technologies do significantly contribute towards being able to achieve them. The reason for that is, one, on the audio side, a shift to a boosted amplifier architecture allows you to create much higher volume with a much smaller cavity. So if you wanna make something loud, you've got to move a lot of air. And typically, in you know, a relatively inexpensive laptop, there'll be like a big volume of air around the speaker, the transducer. But you can't afford that in something that needs to be really slim. So you've got a tiny back cavity for the transducers. It means you need to drive those speakers much harder, much higher voltage.
That means you need a boosted voltage, so higher voltage than you're getting from the battery. And this is really what happened in smartphones like 10 years ago. It's happening now in the laptop space. So we have quite significantly the best boosted amplifier architectures for smartphones. We leverage those into the PC space, then people can build really thin laptops that still sound great, where you've got large volume or loud volume playback at high quality. On the power side, our goal is to provide the most efficient power conversion. So you whether you're charging the battery or taking power from the battery and feeding it to various devices, there will be various stages of power conversion, so from one voltage to another, for example.
Every time you do that, the efficiency really, really matters, because if you're inefficient, you're generating heat. You are losing some of that energy through heat. And if you want to build the thinnest possible device, ideally fanless, that has a good user experience, then you're gonna want the most efficient power conversion you can possibly get, because you're really designing to a thermal envelope. And that's what we set out to do. A few years ago, we acquired a company called Lion Semiconductor, who had novel and a very high performance power conversion technology. We've taken that, and that's formed the basis of our first laptop focused power conversion chips. They were on the reference design that I referred to earlier, and they're in design with customers now.
Well, against our handy dandy shot clock here in the back, I think we might have to cut things here, John. But I just say that it's a testament to you and your company that you guys have been able to make happy the most demanding customer in the industry for so long.
Thank you.
Well done. Thank you for the partnership-
Yeah
... and have a good day.