Cirrus Logic, Inc. (CRUS) Earnings Call Transcript & Summary

Okay, we'll get started. Thanks to everyone for joining our discussion. I'm Rick Schafer, Oppenheimer's lead semiconductor analyst. And I'm joined today by John Forsyth. He's President of Cirrus Logic. John, I believe, has 20-plus years tech experience, joined Cirrus through the 2014, which I can't believe it was that long, John. 2014, Wolfson acquisition. So John, thanks for joining us. It's nice to have you here. Thank you. And before we get started, I thought maybe you could give us a brief history of your pre-Cirrus tech experience and the kinds of things you've been focused on since joining Cirrus, which is 6 years ago already.

Of course, yes. Yes. So thank you for making me feel pretty old there. I do have 20-plus years of tech experience, I -- pretty much all of that has been in embedded battery devices, so-called low powered environment. I started my first job in the industry, was with a company called Psion, P-S-I-O-N. Based -- headquartered in London, based out of the U.K., had an amazing alumni. I do not count myself as qualified to be amongst, but there's a lot of people that went from there to all kinds of organizations, including our largest customer, and were very influential in product developments there. From there, I worked in embedded operating systems and ultimately then into audio and Wolfson. So at Wolfson Microelectronics, prior to the acquisition, I ran the audio product portfolio, as I was responsible for product management and marketing and then continued running a big chunk of the portfolio once the Cirrus Logic acquisition took place. And since then, I've done a variety of jobs, most recently leading the strategic developments over the past couple of years with the executive team and then becoming President earlier this year. And a long way, moving to Austin to yet to enjoy the 110-degree real affable weather that we're experiencing at this time.

Yes. I [ here taught ] in Austin in August.

So maybe I'll start off just with the content question. I mean you guys have done a pretty fantastic job over the years, increasing your mobile content, not just at your lead customer, but sort of across the board. Sort of a classic land and expand, I think, sort of strategy. Content gains, we think, are becoming a lot more important as we look forward. And obviously, the markets mature, and we're not seeing as much growth apart from units. So I think a lot of your incremental content historically has been tied to audio, sort of your DNA. I'm just thinking, looking forward, where you might see some of your best content opportunities in the smartphone, whether that's audio, nonaudio, both like some of the adjacent markets that you started to talk about. But wherever you want to take that answer, but I'm just curious about the content gains basically.

Yes. We see broadly, we think of it in terms of broadly of 3 main avenues of growth. One of them is growing audio in the smartphone space. So that's our core market. It's been driving -- it's been the revenue engine for the company over the last decade. The second is can we drive audio into other new markets where we're particularly relevant? And the third is, can we leverage IP and expertise we have into adjacent areas, particularly in smartphones? So if you look at each of those in turn, I think our story in smartphone audio, as you say, has been very, very strong. In particular, in the past few years, we've seen real growth and momentum in boost amplifiers and the Android side of our business. And so -- but we also have to be realistic about how much runway is there for continued future growth. And the question we ask ourselves is, like how many more sockets that really matter to us? Do we need to win? And it's clear that in a market that units wise, I mean, maybe up a bit or down a bit, but broadly on a -- broadly at the top of that S-curve that the amount of runway for continued audio growth in the part of the market where we fit best is going to be limited. And by the bit that we fit best, what I mean is we generally deliver very high performance, very low power. And as a consequence, this is -- we are not going to thrive if we're having a street fight for $0.01 or $0.02 on ASP in the lower portion of the market. So we kind of think about -- if you take the mid-tier, the top half of that and up is where we can be most relevant. Then -- so there's still runway there, but it's not enough to continue and sustain the company's long-term growth. And at this point, I would say, honestly, I think like we're in most of the sockets that matter. So a lot of the goal there has to be to continue to win those, kind of run that marathon every year and continue to serve them, but also just concentrate fire on other new areas, which are going to drive growth. So then you look at the -- taking audio into the other markets and segments where we can be really, really relevant to our customers, and there are significant opportunities there, but there are also some challenges in certain segments where you might think, like, why aren't they all over soundbars or the smart home, given their audio expertise? Broadly, like the single biggest kind of outcomes razor we use there is, is it a battery-centric device? Because if it's a battery-centric device, then we will deliver a lot of value by doing whatever needs to be done on audio and voice at tremendously low power. If not, then we may still be able to provide a leading level of performance. But the likelihood is we've made all that investment to be doing what we're doing in a very advanced process node. We're not necessarily going to be the cheapest for a lot of that stuff. So if it's plugged into a permanent power supply, we can be relevant, but it's not going to be the big area of focus for us. So that's why when we look at expanding audio into -- and growing our share of audio in other segments, we talk about wearables, hearables, tablets and ultrabooks and things like where we think we can be meaningful because these devices tend to be really small, really thin, small batteries, a lot of constraints in terms of physical area for loudspeakers, things like that. So right up on really. And then the third area is an increasingly important one for us, which is leveraging IP into -- leveraging our IP and expertise into other areas with a particular emphasis on smartphones. And that doesn't mean exclusively smartphones. And I appreciate sometimes you look around other mixed-signal people, and they're all running away from the fire and we're the guy running towards the fire and say, hey, we're like we want more content and smartphones. The reason for that is when we want to get into new adjacent technology and product areas, like you can have a certain amount of truly new product, new technology, new market, new customer. But the best place for us to build it really is be in a new part of -- a new part of the system with a customer that we're very familiar with or a customer most that we're very familiar with in markets that we know because we understand the challenges that they face. We have good engineer to engineer relationships and so on. So that's what's driven an increasing focus of our investment on how do we take some of this key IP and apply it to these other segments. In particular, haptics was a good first example of that. And then most recently, we've talked about kind of vaguely, I admit the closed-loop controller, which will be coming to market later this year in a new part of the smartphone system. And then coming in the back end of next year calendar '21, we anticipate a power conduction product as well. So another new area, broadening our product portfolio, again, in the phone space. And again, leveraging IP that we're taking from the kind of audio-centric part of our business.

And maybe, John, if I could. Just I know you're limited in what you can talk about with that power part that you just mentioned there last. But I know a lot of people are -- their ears pricked up on your call when you started talking about that. You sort of opened the Kimono a little bit on that. I don't know if you can provide just a little color on what maybe the functionality or what existing IP you're leveraging there. I mean, I don't know -- if you could talk about a sense of ASP. I'm throwing stuff out of so maybe you can hit on -- I don't know what story you want to talk about...

Can you tell us anything. Yes. Yes. No, no. I totally get it. I appreciate the difficulty -- I mean, obviously, it's difficult for us to talk about a lot of stuff that we do, but I correspondingly appreciate the difficulty from your side when it comes to that. So it -- we're limited in what we can or we want to say about it. Regardless of the customer, we never want to be in a position of revealing something about our customers' products. I can probably add a little bit of color about like what is the expertise that's relevant, that makes us relevant to a space there that you probably associate -- when we say -- when we mentioned some power conversion, there's probably a bunch of other names that come to mind before Cirrus. So how has that come about? And what's the IP backlog that we're building on there? Partly, it's not IP, of course, partly, the big important part of it, I think, is about just being a really good collaborator with our customers. It's something that our most strategic customers value a lot. So the willingness to invest and work with them on something new is -- like that can open a lot of doors in the mixed-signal space in general. But -- and we have said that this is a custom part. But I would also highlight the kind of path that we've been on with our boosted amplifiers has taken us into it -- more and more into the space that -- around power and elements of power and battery stuff in the system. The reason for that -- in fact, that's a lot of what lies behind -- the boosted amplifier investments are a lot of what high behind the closed-loop controller stuff as well. So if you want to think about like a very, very crude simplistic, I would literally probably get crucified for this by some of our engineering staff and its lack of accuracy, but a very simplistic view of some of the evolution of thinking around the boosted amplifiers. It would be that, okay, the boosted -- the amplifier started off as a relatively dumb -- a simple part. You fed a signal in and it sent power to the transducer which moved and moved down and make noise. As we put more and more voltage into the transducer, the issues that we started encountering, not just us, but as the industry started encountering what to do with the state of that transducer and the likelihood that you damage it. So you're trying to make more a move in a very constrained space. And out of that came the whole field of speaker protection, which is trying to monitor the state of the transducer, the diaphragm and the voice coil, monitor how it's moving and the temperature of it to make sure you don't drive it too hard, you don't damage it. So you went from a dumb part of this loopback and processing around making sure that you didn't damage the transducer. And that was the kind of second big phase of amplifier evolution. Much more recently, as again, we've been driving more and more power through the amplifiers. We've been driven to look at a different aspect of the system, but kind of in the same vein, as we realized that the amount of power we're putting through the amplifier into the transducer is such that you're really responsible for a lot of the peak current events that occur in the life cycle of the phone. And the peak is very high relative to the kind of baseline of normal phone operation. And as a consequence, you risk damaging things in the system. So you risk damaging the user experience by browning out the system. It may not only -- you maybe have 1 amplifier doing it, you may have stereo amplifiers doing it, you may have other categories of device, which have 4 amplifiers, driving very hard at the same time, making huge demands on the battery. So driving out the system or damaging the battery start to become a serious risk. And therefore, having smarts around that, understand at the amplifier level -- understanding the state of the battery, the state of the power consumption in the system and being able to modulate what you're doing in order to preserve the integrity of the user experience and protect the battery and preserve the health of the overall system starts to become very important. So we've gone along that path where we put a lot of investment and thought into how we look upstream from the amplifier at those system effects. And we've developed technology in that space. Some of that having come to market. But that provided us with an excellent platform for the opportunity that we've talked about. So at least in terms of -- while I'm not able to talk about what we're doing, at least in terms of the lineage of it and some of the background that we have that makes us at least kind of appropriate and incredible in that space. That's how we've kind of arrived into this adjacent area and believe we have some real innovation to offer.

Thanks, John, and I appreciate it. I do really appreciate it the limitations. That was good color. So I appreciate that. And another one, which, again, seems to be sort of almost like one outside your comfort zone, but certainly away from your core DNA is CLC, the closed-loop controller is, as you say, ramping this year. And some have speculated that the initial design is for handset camera module. I think anybody's looked online about it, seen your patent there and stuff. Is there anything you can tell us about CLC uses today and where that tech could proliferate? I mean camera is obviously one, but are there other logical places that, that could go?

Again, I mean, I don't want to comment on speculation. People look at our patent filings and sometimes define things we're doing and sometimes put 2 together -- 2 and 2 together and get 98. So no comment on that. But the closed-loop controller technology, as you say, we're ramping that in the back end of this year. The -- we're ramping it now, hopefully, for launch in the back end of this year. And yes, to pick up on that earlier point is kind of -- is leveraging the aspect of amplifier technologies well in the driver side. So making -- taking an input from the outside world, doing the data conversion into the digital domain and then figuring out what to do with it and then driving back out in the world, moving something around in response to that stimulus. So does it have a role to play in multiple areas? I mean, I think it's pretty -- closed-loop controller is such a generic term that you could argue that either way. I mean you could easily get closed-loop controllers all over the system or we have multiple products with them. And I think specifically, though, what you're getting at is like, how scalable is this opportunity for you? And I would be cautious about that in that we think we've designed something really, really good to tackle a particular problem in a subsystem. It's -- in terms of its broad market applicability, in general, I think we try to be cautious about that and under-promise, I wouldn't want to raise any expectations about that now. But the path of opportunity for us there is much more, I think, 1 of evolving the feature set and continuing to grow it and add to that over time and probably deliver multiple skews of that kind of device so that it can populate a whole range of categories of products. But that's likely the path of growth there rather than doing something similar for a whole bunch of different parts of other systems.

Okay. And I think you talked about haptic plus the closed-loop controller being $1 billion opportunity. I mean, other than the -- I think they're both based on amplifier tech. And I think very similar type ASP. Is there any other reason to lump those 2 together? Are they twins for any particular reason? Or is that -- you just wanted to give a sample size of, hey, this is a $1 billion TAM that we're going after.

Yes. There's some shared lineage there. I think as we get -- yes, I think as we go forward, we'll probably break them apart in a more meaningful way. Winding back probably 2 or 3 quarters ago when we did that. The -- we wanted to give an order of magnitude size for that combined opportunity without starting to become overly specific about the new stuff, but we didn't want to -- as I said, we don't want to be in the business of talking too much about customer products. But I'm sure we'll break that out. But all the same, that's a -- yes, that's a reasonable kind of ballpark figure for what we think the opportunity is. And yes, and at a fundamental level, that there's a lot of similarity in how those devices came to be and how they function.

Is it wrong to think that it's kind of an equally balanced opportunity between the 2? Or is that giving too much color?

No, that's about right. Yes. It's probably 50-50 there. Yes.

Okay. And so switch to haptics, again, just real quickly. You've outlined multiple benefits of your haptic technology, improved water resistance. I mean all the things that kind of came with that in prescreen space. As you talk with customers, how are they prioritizing haptics in future designs. I mean, I think you've talked about that market doubling or -- yes, will double by 2024. I didn't want to misspeak. That implies a solid 15% CAGR. I mean, is that -- do you expect to grow with the market, outgrow the market? Any color there would be...

I'd be disappointed if we didn't grow with the market. I think we're leaders in that field of haptics driver. I think it will proliferate into segments where we're maybe less relevant. But the big picture view of it is, yes, like we see that continued emphasis and transition towards physical button replacement with virtual buttons depending on relatively sophisticated haptics drivers. I would say, though, that just to maybe coloring the picture a little relative to what we share on the time slide and so on in the investor deck. The pace of evolution and change there in the Android side of the smartphone market has been pretty uneven. I would say it probably falls into that category of things that is a little slower than we anticipated. And for sure, it's -- you can see plenty of signs that it's happening, but it's taken a little while. And I think that more than anything comes down to the fact that until relatively recently, there wasn't a lot of support in the Android code base in terms of a rich haptics framework, which has been addressed now. And typically, that's something that a year or so after Android implements a new framework, you start to see new OEMs adopt and leverage those technologies. So I think it's sometimes easy to overlook. And certainly, for us, it's easy to overlook how important having that core support is for a lot of Android OEMs because they are in a race to market with comparatively short development cycles relative to, for example, our largest customer. And so if something is like a load of engineering effort and complexity to figure it out, like it might just not make it on the list. So it's taken a little while, but we see more and more devices going down that kind of route away from physical buttons. I would also say that certainly, within Cirrus looking at the haptics market and talking with our experts on haptics, they continue to be pretty optimistic about how far this can reach into mid-tier and lower devices, largely because if you look at -- if you look at like mid-tier phones do almost everything they can today. If you walk into a store, they do almost everything they can to look like a flagship phone. So they -- so which requires the removal of bezels, the removal of home buttons, the kind of streamlining of the physical design. And so they may find ways of optimizing the cost of doing that. But that route takes you down the path of needing to use haptics driver to kind of square the circle on the removal of those buttons.

Got it. That makes sense. And maybe just pivot a little bit to another market opportunity for you. I was hoping we could talk for a second about digital headsets. And the world is sort of increasingly moving to wireless and ANC, and you guys called that, I think, half a dozen years ago, this was where we were headed. I mean, as this process plays out, do we just -- do you think we do see Inbox wired headsets go away over the next few years? And maybe more specific to Cirrus, but could you discuss your content today and like truly wireless headsets and how that evolves over the next few years? Because I do feel like ANC is definitely penetrating the wireless headset space, becoming a bigger piece.

Yes, right, right. Yes, I think it is, which has got to be a good thing. And hopefully, if you think for Cirrus in the process. Yes. I mean, I would say, even -- I mean, you referenced 6 years ago, I would say, even 6 years ago, when Cirrus bought Wolfson, I think at the time, I can remember having conversations about, okay, at some point, wireless disappear, right? Like they've got to disappear. This doesn't make any sense. So yes, for sure. Like -- I mean you would expect at some point, it goes away. As usual, I would not want to be in the position of saying anything about any of our customers' products and their plans. But you would expect that. So for us, it'd be highly desirable to transition to more content in the truly wireless space and not just as a kind of, well, we need to make up some content there. But really, it's very obviously intangibly a place with a lot of innovation and a lot of opportunity for features, given that there's real embedded processing in those devices. So yes. Right now, we have a range of products in the truly wireless space that go from amplifier level in some of the pro devices that are out there, which everybody can see in teardowns. And if you want to think about the value of those, each of those does not equal 1 boosted amplifier because we would -- blow your eardrums. Think about them both combined together as roughly having the value of regular single boosted amplifier. But then we can -- there's every reason to believe that we can beat a path north from there, add in other categories -- or other OEM devices in the truly wireless category. We have smart codecs sitting in each year. And today, what those are doing is providing -- driving the transducer, providing the voice wave -- voice system kind of functionality and a few other features. We have a new product that's sampling to customers right now, which will be our first -- actually, the first product that's focused on truly wireless from us that introduces ANC. And so that's with some of our customers now. And that also includes DSP processing capability and software that brings a whole bunch of other smart features as well, which -- a number of which, which I find really interesting and to do with like hearing health stuff, what's called the PSAP, personal and sound amplification product, that kind of use where people are using it. Particularly people who don't, for whatever reason, want or quite need a prescription hearing aid, this is something -- it makes something which can enhance you're hearing in certain situations, totally tunable by you and it greatly improves your quality of life. That kind of hearing augmentation functionality looks like being 1 of several areas of innovation around truly wireless that we're excited about. So yes, the ASP for those kind of devices, the codecs, the smart codecs that sit in wireless earbuds, we believe that those will range from $0.60 up to $1.50, depending on how rich the functionality in there. But we are, for sure, excited about what that market can represent. It's still relatively small in units, but seems like one where the consumer response is really unambiguous that people like the devices a lot.

Yes. I would agree. I've got some. So pivoting again to process, you just recently announced on your call, the beginnings of the move, the qualification efforts starting on 22 nanometer. And I was curious, which products you expect to migrate there first? Maybe a sense first of just production timing. A lot of people get excited when they hear about the move to '22 and is that something that's going to start in 2 quarters? Or is it more like 2 years, kind of a loose time frame. And John, just maybe as part of your answer because I don't want to interrupt you when you start speaking, the gross margin impact. I mean, because it's obviously going to help with your cost of manufacturing, and it would seem like a richer mix going forward for you, plus lower cost could lead to some gross margin expansion. And I don't know if that's the wrong way to look at things in your model.

Yes. Maybe I'll start with that and just say, like, I think the gross margin story there is broadly as steady as she goes. I wouldn't expect the big swing one way or the other. Obviously, when we're doing anything like that, we look at it from the point of view of making sure it's supportive of our gross margin target. But I think given the various trade-offs there, like I wouldn't necessarily expect it to skew enormously northwards either. I think the -- but I'll circle back to that point and cost and so on in a second. I mean, I think broadly, yes, just to summarize, we're on that path to 22 nanometer. We got silicon back for a test chip in the back half of the last year or towards the end of last year, and we've been really happy with the way all of that's looking. That is really focused on the IP. To your question about like which products, that's really focused on the IP associated with our codec products. So I wouldn't say there's necessarily a ton of -- ton of logic right now to doing something like a boosted amplifier. I'm trying to take that amps to 22. But the mix between digital and analog on our codecs is such that if you were to grow the digital, -- 55 is a really great place for what we've been doing to date. If you were to take what we have today and put that in directly on 22, you gain on the digital side, it shrinks a lot. You lose a bit on the analog side. It stays broadly the same, but the wafers cost more. It's probably not incentive enough in itself to do it. Like I think the ASP effect of that, assuming constant margins would be broadly a wash. So the logic to doing it is to drive more content on the digital side. So more processing that somehow needs to be close to that analog portion of the system. So we're engaged in doing that. It will be focused on the codec. And the driver -- well, I should qualify that, the drivers are really twofold. One is increased digital content. And the second driver is, okay, maybe you don't increase digital content, but you're designing for systems where physical area is absolutely paramount because we will be able to make for a given function assuming it's some mix of analog to digital, it will likely end up, in total, being smaller. So those are the 2 main driving forces and the -- to circle back to the margin question. If you think about the overall equation of that, the cost of our development and design increased with 22. The tapeout costs increased with 22. The wafer cost increased with 22 per square millimeter, you're doing something that's more expensive. So the reason we would be doing that is we believe we've got a very strong case that we'll be able to grow functionality and therefore, deliver more value and get the ASP benefit that will offset those increased costs and give us upside. In terms of timing, yes, there's -- you've got to be patient in semiconductors in general. I mentioned that we had silicon back at the end of last year. We have a design team working now on the back of that, on what the product looks like. We don't have a firm committed target yet, but I would expect in calendar '22 or 23, that first 22 nanometer product will land.

Okay. Thanks for that. I feel like there's one -- another big elephant in the room, I just wanted to talk about is VoiceBio, it's just been out there. I know it's near and dear to you and in particular. And I think it's obviously proven to be a tough problem to solve even for audio experts. So I was hoping maybe you could just update us on where we are in terms of development, maybe what the tall tent-pole is in terms of hurdles still yet to overcome. And any sense of like revenue timing? Or is it just too soon to talk about?

So it's really, really rich and incredibly hard problems. I'm definitely not disagreeing with you on that. I would say, though, that, that is not the barrier to adoption at this point. I think market acceptance and the appetite to go take that to market has been more of a limiting factor at this point from the technical progress on the underlying technology because we certainly reached a point where I've mentioned elsewhere, we've passed this FIDO certification. FIDO got baked into Android. That is like, okay, this is the way biometric authenticators will be certified as being secure. And FIDO, on the back of that, introduced the certification program for voice authentication. We sailed through that on the first attempt. To date, there's no other voice authentication solution that's been through that and we've been really happy with the performance and the customer response and dialogue around that has been very, very positive. There are -- nonetheless, there are challenges when it comes to, okay, how much risk is associated with introducing any biometric. How intertwined is this thing with my voice assistant technology, which all of our customers have, which kind of lives in the cloud, but you want to sell a chip that sits over here. And so -- and there's just -- the overall movement of the industry towards voice biometrics has just been slow. In spite of the fact that still every conversation -- I think to date, every conversation I've ever had with an OEM about it has started off with the premise where they tell you, yes, you don't need to do the pitch. We totally believe this is a problem that we want to solve. So that side of it has been disappointing. On the plus side, where that has taken us is in having a number of really good conversations with our customers around building blocks in the technology. So yes, we've made a chip -- part of the reason we made a stand-alone chip for it was security, but part of it was just so that we didn't pile that risk onto any of our programs, right? So we always envisaged that ultimately, the technology could have a number of paths to market, but we needed a vehicle for it to get proven out. Where we -- where that's taken us is of having a number of really good conversations with customers about how we integrate aspects of that technology into their system. So into other products that we make that go into their system. And I think that is likely to be the route by which it gets to market. So in terms of revenue, it's probably -- around ASP, it's going to be hard to -- harder to break out because of that. I think it's going to be wrapped up in other stuff, but it's probably going to be a really nice, sticky technology that also drives a reasonable amount of processing and probably, probably looks much more attractive on 22 and so on and on higher nodes. So I think over the next 3 years, say, we're likely to see that start to come to fruition. It's obviously going to look a little bit different to the way we thought maybe 3 years ago.

Okay. Thanks. And just for those in the audience, I don't see any questions in queue, but we've got about 5 minutes left. So if you would put a question through the queue, I'm happy to read it and have John answer it. And until something pumps up in queue, I might ask an Android question, if I could, John. And I'd like to discuss, not only your largest Android customer, but just in general, you've obviously done a good job, increasing your content and growing, as you said earlier, into the China market. I guess I'm just curious if you could discuss like sort of where your average content is today across your customers in China, in particular. I know it's well below, let's say, flagship phones. And what the trend has been the last few years for content growth? And sort of where you see that headed? I mean, are we going to close the gap as we move into the 5G world? Or between flagship and sort of the Chinese branded phones? Or...

I mean, I think that's a great question. Just to give you like some general guidance and rules of thumb for how to think about our content across all of those. In the vast majority, so boosted amplifiers are leading the charge in Android for us. There are pockets of other stuff from codecs. But the main story is around boosted amplifiers. If -- they're in the $0.40, $0.45 range pair -- in investor deck. Anywhere we're winning sockets, it's going to be a stereo device, so you get a couple of those. That's that like -- as a baseline, that's how to think about our content in any new design win. Now when we get more advanced and have a deeper relationship with customers or just get a little further down the road, that has the potential to grow, in particular, with the addition of haptics driver, which isn't broadly the same ballpark. We -- I think we say $0.50 to $0.75 as ASP guidance for those. But for the current generation of haptics drivers most of them being at the lower end of that. They still didn't incorporate for the analog front end and so on. And then there's the additional opportunity for accessory stuff that sits around the phone. So when you think about our largest Android customer, we have stereo, boosted amplifiers, haptics driver and then these truly wireless earbuds sitting alongside, which have a smart codec in each ear. And that's how to think about the overall kind of content picture there. Those -- and when we look at the other China-based OEMs, I think, okay, there may be some opportunities in accessories there. We're not falling over ourselves to chase low volume accessory opportunities. It tends to be a -- just because of very challenging acoustics those things tends to be a ton of work. So if it's a low volume thing, we'll look at that very carefully. But there's every reason to shoot for 2 boosted amplifiers and haptics driver as being the core of the business there. And then, of course, with customers, going back to your first point about land and expand, I mean with customers that are -- where we have a really strategic relationship, we always want to then look at their system and try to understand what we can integrate, how we can solve their problems better and grow the value over time.

Great. And John, that was a perfect ending. You stuck the landing. That's our time. So really appreciate your time today, and it was good seeing you.

Yes, you too, Rick. All right. Well, thank you very much. Thanks for the invitation, and stay well.

Yes, you, too. Thanks, John from Cirrus. That's our call.

Thank you.