EP 189: Analyzing ADI Earnings, Custom Silicon Shifts, and Semiconductor Debt

0:05Ben Bajarin: Hello, everyone. Welcome to another episode of The Circuit. I am Ben Bajarin.

0:11Jay Goldberg: Greetings, programs. I'm Jay Goldberg. All right.

0:15Ben Bajarin: We've got a range of things to talk today. Let's start with analog devices. are sort of, what do we say, quarterly check-in with the analog industry. Sometimes there's pockets of optimism.

0:34Jay Goldberg: Sometimes, yeah, people freak out.

0:37Ben Bajarin: But anyway, they had a good quarter. Obviously, as we've talked about, there will be a pretty consistent analog cycle pull-in, particularly around data center power and others. but we've noted it's been kind of lumpy, I guess. But I'll let you start on the signals from earnings.

1:00Jay Goldberg: So the big sort of event that took place at Analog Devices, they reported earnings this week, and industrial and consumer and all that stuff they usually do is doing well, which is important. But the most important stuff is they laid out a clear case for their opportunity in AI. And the analog companies have not participated the same degree as memory and digital logic companies have. But analog is starting to become more important in the data center. We're using so much electricity. You start to need a whole new category of electrical systems. And that's we're finally we've we've seen this for a while now because people have been talking about the 800 volt data center.

1:46Jay Goldberg: companies like TI and On and Navitas and Wolfspeed. And ADI has been kind of quiet about it. And this quarter, they came out and they said, they laid out and said, hey, this is what we think our addressable market is. And we're seeing data center is about 10 % of their revenue now. It doubled in the quarter, and they think it's going to double this year and double again next year. And so they're joining the party So interesting just that ADI is sort of the last to sort of talk about it They're probably the most conservative management team in the space But also interesting was they sort of scoped out the market for analog Which I don't think anyone else has really done yet It was interesting, they shared a lot of information it was interesting most go ahead go ahead no i was gonna say it was interesting the talk on margins where they do they i don't know if they're they are pretty high i mean others have slightly

2:57Ben Bajarin: lower margins which i heard and i was trying to listen to some of the investor questions also um i think which which was poking at the idea of they others having a little bit more pricing leverage to use if they're going to hold at lower gross margins since I think they're signaling 74 % and that can hold. But yeah, the margins I think was interesting, which again is just an absurd conversation when you look historically at analog margins, but so be it. So, all right.

3:33Jay Goldberg: So on the margin side, I think analog devices and TI tend to have higher margins than analog overall. Yeah. Right. Because they tend to sell directly to customers. Much of the rest of the analog space sells through the channel where there's less. You're sort of competing on price much more heavily. I think there is this weird thing, though, where you have a lot of these traditional longstanding companies that are now joining AI. and sometimes the questions they get on the calls show people just trying to figure that out and just trying to use the old framework for analyzing these companies in this very new way, in this very new market.

4:18Jay Goldberg: And so there's a little bit of disconnect. I think their margins there are... I think that's a secondary story. I think what was... They're going to do fine. The whole 800-volt data center story for analog is you're going to replace a bunch of lower value parts, not quite commodity, but almost commodity parts, with some very, very high spec parts. That's what 800-volt data center is, your big transition, in addition to just being large quantities of these chips now. Right. And so the long-term story for analog parts in the data center is this idea of we're selling a lot of high-priced parts, which is good news for them.

5:09Yeah.

5:10Jay Goldberg: I think to me, the really standout feature is they said, we think, they quantified how much analog content goes into a data center. And they said it's a billion to$1.5 billion of analog content per gigawatt. Right. And so we have NVIDIA saying 50 going to 100 billion per gigawatt to build a data center. And, you know, what's that 2 % of that goes to analog, which is small compared to NVIDIA, but is very meaningful for companies like AI and TI. It could be ADI and TI. I also, I really appreciated this. They didn't talk about TAM, total addressable market. They kept referring to their SAM, their serviceable addressable market.

5:59Right.

6:00Jay Goldberg: And the difference being TAM is like everything that goes into the space. SAM is everything that they can sell into. And so you may not have a part you can sell into the TAM, but SAM is what you actually can sell into. And I think the knock on ADI up until now has been their SAM is relatively small in the data center. And I think they thought that too. what they said on the call was over the last few months, they've sort of did a deep analysis of the market and they realized that their SAM was double what they thought it was. They also have an acquisition in there that brings them deeper into it.

6:38And so I just like,

6:40Jay Goldberg: everybody throws out these big TAM numbers and the ADI, again, conservative management team, a lot like AMD, but the ADI is management team throughout the SAM. because they're implicitly saying like, hey, we can throw out these big crazy numbers too, but we're just going to talk about what we can actually address. And it's bigger than I think, it's bigger than they thought. So it was good news for them.

7:06Ben Bajarin: Yep. SAMs are the way to go for those companies who are listening. SAM is a better number than TAM. So we always get this question in market models and data, which is like, what's their real market opportunity? Not how much is, you know, it's nice to say CapEx is going to be$10 trillion, but, you know, what are you actually selling into? It's a much smaller pie. Let's not mislead people, but I get it. Yeah, and I think, again, right, we did a power report a couple months ago just looking at the sheer pull of attach to power as all these data center racks get bigger and pull more power. But also just, like you said, 800 VDC, the transition that's going on in just serving power from grid to facility.

8:05Ben Bajarin: There's a lot of analog in there. You know, and it was interesting too, I don't remember if I said this, so sorry if I'm repeating everybody, But there's a real interesting way that kind of these analog play in 800 BDC for safety. So like when somebody is detaching a rash so that circuits can actually be managed and shut off accordingly. And so there's a whole new sort of path on what it's looking like for these very specific applications and serviceability and safety also as the entirety of a cluster and whatnot starts to pull far more energy. So, yeah, I mean, I think the content opportunity is big.

8:52Ben Bajarin: It's good. Good ADI. Put a little bit of numbers on that. But yeah, analog is going to be a beneficiary.

9:02Jay Goldberg: Yeah, right. And I think for ADI, the story, my assumption had been that they really served from the edge of the data center where the utility connected with power comes in, if you're doing behind the meter or whatever, where the electricity comes in to the wall of the data center itself. Right. Right. And now and then it turns out they actually had content inside the data center. And now what they're saying is they bought this company called Empower that actually brings them actually to the package level. You get the subvolt conversion stuff. And right. And they talked about optical, too, because everybody loves optical.

9:46Jay Goldberg: And they seem pretty excited about their optical opportunity. And again, they're not sitting next to the transceivers and the lasers, which is, I think, where the biggest volume is. they're on the control path so timers and power management and telemetry kind of stuff which is again not not the massive opportunity but still meaningful meaningful for them yeah and leaves the door open to do another acquisition

10:10Ben Bajarin: yeah i think what's interesting about this like

10:16Ben Bajarin: there's been they aren't the only example there's a handful of others uh kind of in this space but But I would also, again, say in things like FPGAs, like my report last week around FPGAs and talking with Lattice Management, even just stuff, conversations with Infineon and TI. It's interesting that they are sometimes hearing about new opportunities they didn't know about as customers are going deep in the weeds trying to solve engineering problems for compute racks or for power or for liquid cooling. And so I just think it's interesting that they'll be like, oh, we just heard of this new application that we didn't know we solved for X, Y, whatever.

11:05Ben Bajarin: And I say all of that to say we're in such this moment where there's a lot of engineering challenges kind of across the board. And it's creating interesting opportunities that are somewhat sometimes unpredictable by some of these vendors. Because they're being used in a unique way that perhaps they didn't slot for or design for. because an engineer is like, oh, I could use a PMIC for this, or I can use an analog sensor for this. So I think that's just interesting in, you know, call it this kind of very technical engineer's dream to solve some very hairy problems in building complex data centers and compute and all of that stuff.

11:52Jay Goldberg: Yeah. I mean, for me, the really interesting thing is you're seeing the analog companies data center revenue really running like both TI and ADI doubling their data center revenue and that's before we get to 800 volt data center right nobody's really building an 800 volt data center today it's all prototyping right it's just the design decisions are getting made today to solve those technical problems the real build up for 800 volt data center isn't until 2028 and we're already seeing these companies data center numbers run

12:25Jay Goldberg: I worry to your point about the linearity of this, does it sort of slow down next year before the 800 volt data center really takes off and we see a re-acceleration again in 28 it's going to be it's going to be a job to keep track of all this

12:47Ben Bajarin: this is why I go to all these conferences and I stick my head in these compute racks so I can look at when I send you all these pictures of microcontrollers and PMICs and analog sensors and capacitors because I'm like, look how many there are now. But you're absolutely right. It is. Anyway, I just love the engineering moment that we're in that that people are trying to solve new problems with existing technology, being super creative about it. And I just, you know, again, it's interesting that it's very, it's a little bit more unpredictable. Like some of it's predictable, but then some of it's like, well, how are you going to solve that problem?

13:33Ben Bajarin: Well, I'm not sure. And engineers are just being creative and some company can all of a sudden benefit from that they didn't see coming. So I think that's interesting. That was all, that was the only point I wanted to make on that. All right. So apparently AI has solved cancer or cancer has been solved. There's now a vaccine, so I don't have to keep going down to the guy down by the river who keeps offering me the cancer vaccine for 20 bucks a shot, right? I'm just kidding. It's always my joke. I'm like, yeah, I just got the cancer vaccine from this guy down by the river.

14:07Jay Goldberg: People are like, what?

14:08Ben Bajarin: I'm just kidding. All right. You had a point to make on this, so I'll let you go.

14:15Jay Goldberg: So there was big news in pharma this week. Moderna, a company which created the COVID vaccine seemingly overnight, has announced it came out with a vaccine for melanoma for skin cancer. And it is novel in that there is a degree of personalization to their treatment, where every patient gets a slightly different combination of chemicals. And there is a heavy amount of compute involved in sort of calculating all that for each patient. And so the headline was that Moderna used AI to cure cancer. And if you go back and actually read their press releases, and I think they did this with Merck, they go out of their way to not call it AI.

15:13Jay Goldberg: They call it bioinformatics, which is an older term. And I think there was a misperception that they actually used some big AI training model to do this. And that's not the case. Right. And I mentioned this because I think there are a lot of people who assume that AI is going to is going to come up with novel drugs. And I actually think that is going to happen, but it hasn't happened yet. And I think that's important. Like and you can you can read that as glass, a 50 percent glass. Is it half full or empty? We haven't solved it, but maybe we still can.

15:53Jay Goldberg: and it makes me think of this company I met in when I was in China earlier this year I went to see this company called InSilico in Shanghai and InSilico is interesting it's only about five or six years old and they they make a big deal about their use of AI for developing new drugs And what they've basically done is they've taken all of their data, all the data they can, and fine-tuned some LLMs to do sort of every part of the drug processing pipeline. But even they are not using AI to come up with new drugs. Yeah.

16:40Ben Bajarin: Right.

16:41Jay Goldberg: They're enhancing the process of refining it and sort of getting it to approval stages.

16:51Jay Goldberg: But it's so it's interesting. Like this, again, this is a company that is using AI very, very heavily. And they're just using it to enhance the process. Right. Which is not sort of novel. They just process existed. They're not coming up with new ways to fight a drug. and and they've done incredible things like they're a five-year-old company they have 30 drugs in their pipeline right now in development which is a big number for a small company like that and so one I think Intelco is very interesting but I think it also shows the limits of what AI is able to do today right we're not at the point where AI is curing cancer.

17:33Jay Goldberg: Yeah.

17:34Ben Bajarin: I thought it was interesting that everybody that I follow who I thought was a technology semiconductor investor all of a sudden was a biopharma expert. Yeah. They're putting out reports. I was like, wait a minute, didn't you just write about GPUs? No, but actually, your point is valid. Obviously, people are working on how AI can be used to help cure diseases and cancer. There's optimism that that happens someday. Obviously, that will play a role. But yes, I think, again, we have people getting ahead of their skis, to use the term lightly, in some framing. But I think that will become more of a story in pockets over the next few years about how compute simulation and running that work, that research they do can maybe make time to trial faster or something.

18:47Ben Bajarin: But yes, AI is not solving all of the world's problems yet. Instead, people are still arguing about data centers showing up in their hometowns and fighting them like it's nobody's business. So sentiment on data center builds is not good.

19:07Yeah, and I think that's part of the dialogue is that AI was so heavily hyped up in its early years in 2023.

19:16Jay Goldberg: and people use chat GPT to make silly memes and then kind of thought that was it. And I have to think if we start to get more meaningful advances because of AI, that will change people's tunes. But we haven't had that yet. And I'm optimistic, right? I'm optimistic. Again, these companies are barely scratching the surface of it. They're using Ampere and Hopper. what happens when they get orders of magnitude more compute in some of these big systems, right? And I think if you look at some of the people who are really, really leading and very, very advanced and they're thinking about AI, you know, like some of the guys from Google who just left,

20:03I think they think that's all possible, right?

20:06Jay Goldberg: And, you know, those people were pretty deep into this. So I think it's possible. I just, I don't know when. Yeah.

20:13Ben Bajarin: Yeah, I know I went to years ago when Amazon used to do this event called Remars, which was sort of just a meeting of the minds of everything to automation, robotics, AI, et cetera. Someone from big pharma, I can't remember who, but came in and basically showed like what they do to go through drug discovery today. And it was expensive. It was cumbersome. Yes, they threw a lot of compute at it, et cetera. But the whole point was, you know, what happens when we can speed this up via, you know, compute and AI and intelligent simulations grounded in real world science, you know, etc. So that was years ago.

20:55Ben Bajarin: So like it's always been talked about. But yeah. But to your point, like I – it's just we're in a weird moment with AI sentiment. You know, I don't know how long it takes to get over that. but it's you know you hear about these tensions it's getting talked about a lot data center builds being stalled and whatnot because of counties getting angry and i'm witnessing this in our backyard as amazon's building three multiple hundred megawatt data centers and so i'm on these like facebook community boards and i see them sharing like notes for meetings they're all angry we're gonna run out of water you're gonna pollute our ground you're right next to a pepper farm like anyway it's it's interesting so the joys the joys of this moment of of ai in a build out okay um yeah let's talk about uh custom asics there was some uh some news around custom asics uh marvell and google have a enhanced relationship that was good for the stock but also So people, again, misinterpreted this news as a little bit different, I guess.

22:13Ben Bajarin: There was some knock on Broadcom, which was not the case, which we'll talk about. But if you read the news very clearly, it was definitely this deal is more for IP and custom work around the periphery, not necessarily the TPU itself which is what Broadcom does and then does with MediaTek to agree in that design and other vendors but that's not bad being in the periphery, being in the optical I noted very quickly I think there has a lot to do with CXL around this and disaggregated memory in kind of the report that we wrote a couple weeks ago about why DDR4 reuse is kind of an immediate TCO unlock Google's got a lot of DDR4.

22:59Ben Bajarin: Marvell's the best position in the full stack for CXL. But yeah, people were like, this is a terrible Broadcom. It's like, it's not terrible Broadcom. This is also a huge growing market in dollars for custom silicon. But Marvell's playing an important role, but it is more on XPU attach, if you will, to use a technical term.

23:24Jay Goldberg: yeah it was i mean we talked about this three four four weeks ago where or maybe even longer it's clear it's clear that uh you know broadcom has been with google that signed tpus from the beginning and it's just gotten so it's google's tpus ambitions have just grown and the complexity of chips that they're designing has grown. I think it's to the point now where they want to diversify. It's well known that Google is uncomfortable paying Broadcom's margins, but they are stuck with them for a variety of reasons, mostly around Surge's IP. But as they expand, it's only natural that they're going to want to take on additional suppliers to help them bring all this to market.

24:20And I think, you know, there's obviously Google.

24:23Jay Goldberg: Google has a lot of its own tools itself, right? They've designed their own EDA tools in a lot of cases. So they have a lot of chips they want to build.

24:36Jay Goldberg: So it is, I mean, you know, I understand why it's bad for Broadcom. It is they're losing a little bit of share inside Google. But also, to your point, it's growing so big, they were never going to hold on to 100 % of it, not with the size that Google's chip ambitions are. And I think you're right. I think there's an argument that the things that Marvell will do for Google are not necessarily businesses that Broadcom wants that badly. It's not that big a loss, maybe. Maybe you could argue this is something that they couldn't do. I don't know where their state of their CXL is.

25:20Jay Goldberg: So not great for Broadcom, but not the end of the world. I think they're still likely to hold on to a lot of the core TPU business.

25:28Ben Bajarin: Yeah, but I think you hit on the tension that I think people have pointed out kind of regularly, which is both these companies actually want diversity. I don't think either one wants to be holding to each other. you know yet at the same time broadcom is going to get a lot of wafers right they're going to get in some cases a lot more wafers than other people because of their scale and so you know you look at the sure you can do some of this with other vendors but you still need them for a good portion of your volume but you know you you had this even early on right broadcom this goes back a couple years but we've sensed the tension where people are like hey you got too much customer concentration.

26:13Ben Bajarin: So, okay, well, great. Yes, Google's our biggest customer, but fine. But they, they also want that diversity. So you kind of have those two dynamics that play into each other. That makes sense, right. In terms of each one's strategy, but you know, it's like people complained for so long, particularly on Broadcom that now that they are focusing on diversity and they have X number of, of bigger customers, still an obscene amount of wafers at TSMC, people think it's bad.

26:46Jay Goldberg: So, you know, in terms of, you know, relative, relative, uh, impact, it's a small, it's a small negative for Broadcom. It's a massive, massive win for Marvell because the dock on them has very much been, they're entirely dependent on Amazon for their custom basic business. And they've done a little bit to diversify. And now this is a fairly sizable step for them and de-risks them to some degree. They have a long way to go though. So it's, but this is, it's good. It's good now that we can, you know, because I mean, we've talked about this for like two years. The whole debate about Marvell was, are they, are they losing share at Amazon?

27:28Jay Goldberg: How much are they losing at Amazon? now it's like okay we have they have more than one one customer to hang their hat up

27:37Ben Bajarin: yeah no i agree and i think you know again i keep coming back to it you know we've we've tried to slot marvel cleanly into the role they play in xpus we have constantly sort of come back to they keep saying it's uh it's xpu attached right what does that mean on the periphery io tiles networking, et cetera. And I just think it's becoming clearer that is the role that they play. They can, again, get into other parts of kind of off XPU, but still a part of the XPU versus on package. But thinking of them as a periphery XPU attached beneficiary is actually, I think, a good positioning for them as they continue to get more of those deals with other vendors versus the CPU tile or the ASIC tile, which is not what they can do, but it's not what they are doing.

28:34Jay Goldberg: Yeah, it's weird because they have a pretty solid processor team, but it's pretty solid, right? But they have really, really sort of leading class networking. Remember, they acquired Cavium, they acquired Enfy. This is one of the leading networking chip companies out there. and that sort of got lost in the we stopped talking about it for years and I think they've realized they need to talk about it more and they have a lot of opportunity there so something new for my belt

29:07Ben Bajarin: alright so then now Broadcom's getting some more news because they are and I don't know I think this is a rumor but still reported but looking to raise capital for some of these TPU deals. So this got some speculation, a lot of question, why would you need to do this when you've got very big customers who also have a lot of cash? And anyway, I'll let you go into it. We can kind of talk about what that means and why more people are doing this types of stuff, Intel included. But yeah, you start. So I think Bloomberg broke the story. legitimate media outlets are saying that Broadcom is raising, I think,$60 billion for a special

29:53Jay Goldberg: purpose vehicle, an SPV, to do something in the data center, probably with Anthropic. And I think this is very much in line with what we saw last week with NVIDIA and their $500 billion, their own SPV to fund data centers. And I think what's clear now is the chip companies are starting to use their balance sheet, just like the hyperscalers are using their balance sheet to compete. And it's important, right? I think if you think back, what was it, last year, roughly, six, nine months ago, NVIDIA sort of changed the nature of competition. They said, we're not just competing on chips. We're not competing on servers.

30:42Jay Goldberg: We're doing rack scale system level design. And that sort of forced their competitors to follow suit. We had AMD buy ZT systems so that they could do rack scale design. Now, NVIDIA is coming along and saying, we're also competing with Balance Sheet. We're going to help fund the data center. They've been doing it for a while. They've really amped that up. And their competitors have been following suit, too. It started with backstops on NeoCloud capacity, and it's expanded into something probably much bigger than that now. NVIDIA is using every tool it can. They've used a lot of the technical tools.

31:23Jay Goldberg: Now they're using balance sheet as a tool to compete. And that has put pressure on AMD and now Broadcom to follow suit. If you want to win business here, you can't just have the best chips. You can't just have the best systems. You need to have the best. You need to bring the balance sheet into play as well, which is very, very different. It's something very new for the semiconductor industry. it is uh it's a little it's a little concerning right these are companies that historically have had very very little debt for good reasons and um there's but now you know they they have to they have to start using these tactics in order to to compete against nvidia

32:15Jay Goldberg: so that's you know as i'm talking about it i'm getting more and more worried about it right you think about it for for for 50 years semiconductor industry i mean it's like one of the things you've just known like oh chip companies don't have debt right because the problem with debt is that it forces you onto a schedule do you always have to repay things on time. It's hard to do. Chips are cyclical. Technology cycles come and go. You don't want to go bankrupt because you've missed a product. That's what debt can put you into very difficult positions because you have to sacrifice cash flow from, say, R &D in order to make your debt payments.

32:57Jay Goldberg: That's not a good look. You miss the next product cycle. It puts you in a bad place. now things obviously change the chip companies are much much larger and have much stronger cash flows and nobody's sort of chipping into problem territory on their ratings but it's it's not there are risks here and we were just talking about Marvell a second ago, part of the deal with Google is that they gave them equity in warrants Right. Right. We saw AMD do that.

33:35Ben Bajarin: Tied to revenue. Yeah. Right. Exactly. Tied to revenue and payments. Yeah.

33:39Jay Goldberg: Right. And we saw AMD do something similar with OpenAI a few months ago.

33:44Ben Bajarin: Right.

33:44Jay Goldberg: Right. These companies are deploying their balance sheet, both equity and debt, in order to compete.

33:53Which is, you know, I guess it's a thing to do.

33:58Jay Goldberg: I guess it's better than, you know, giving them 10 % off the price of the chips. But it's getting, I mean, like I read all of this as the processor landscape is getting, is amping up the competition. Like the competitive intensity is amping up. Yeah. And so that's, you know, and, you know, NVIDIA is leading the way, but they're, you know, they're having, they have to do it. they're not unassailable because like it wouldn't have to raise 500 billion dollars in debt if if their technical lead was so big

34:38Ben Bajarin: yeah i don't know yeah and but i also think like you wouldn't do some of this and i get the risk portion but you wouldn't do some of this if you also didn't have fair visibility and confidence from your customers that there is still some coming ramps that are not necessarily guaranteed, but fairly high confidence in their capacity needs, their roadmap needs. Because again, like we talked about, I do find this interesting, the LGA waterfalling of how much the supply chain has to then just go do more deals with every bit of supply chain further down. And that's, you know, there's a lot of capital, right, that goes into that to make sure you can meet your commitments.

35:26Ben Bajarin: So it is, I mean, I think, again, right, your point is, you're right, this is a new, a new dynamic for the semiconductor industry. To some degree, this is kind of, I guess, what happens when a capital intensive industry has to go and build out in a longer timeline than they're used to. Um, but it's just, this whole cycle is causing difference, like stuff we've never seen before, uh, because of the build out. So, um, but yeah, I think people will continue to worry about it because obviously there is risk. Yeah.

36:02Jay Goldberg: And I think it's telling too, that every, everybody doing this is structuring these as SPVs rather than putting it on their own balance sheet. And that adds costs, right? If you start bringing in other players, an SPV will typically have other players involved, private equity or some other funding institution will be involved and they expect to get paid. And so, and we see the hyperscalers doing this too. They're all doing it in SPVs. And I think the idea is if you put it into a remote entity like an SPV, you can keep it off the company's books, sort of. Like it's not actually going to appear in their financial statements.

36:46Jay Goldberg: But buried deep beneath all these things, there are real obligations tied to them. And we probably won't see this until it's too late.

36:56Ben Bajarin: Yeah, I mean, honestly, like I said, there's a lot different. But at the same time, this is an industry that's, I guess, not had to do in the supply chain what they're being asked to do now to meet demand. Yeah.

37:13Jay Goldberg: I mean it's interesting to me. I'm getting calls from fixed income investors and fixed income journalists. I've never talked to these people in my life. I've never had – you cover tech companies. You never have to pay attention to the balance sheet. And now suddenly we've joined the fixed income party. We get to go visit the fixed income clubhouse now.

37:40Ben Bajarin: Awesome. Well, balance sheet competition. Let's keep that on our radar. And perhaps a report needs to be written about the balance sheet era of competing, not just who has supply chain scale, who's got the balance sheet and supply chain scale. Anyway, all right, everybody, a little shorter episode for you today, but looking forward to next week because I will be at Hot Chips. It is a great, great session. If anybody happens to be at Hot Chips, hit me up. But this is always one of my favorites to talk about and glean insights from. And then we have some earnings next week from some big names. Marvell, NVIDIA, I believe.

38:24Ben Bajarin: Who? Yeah, it's next week. Uh, yeah. So awesome times. We're looking forward to it. All right, everybody. Have a good one. We will talk to you next week. Thank you, everybody. Tell your friends.