On this episode of The Six Five Webcast, hosts Patrick Moorhead and Daniel Newman discuss the tech news stories that made headlines this week. The handpicked topics for this week are:
- NVIDIA Q4 FY2024 Earnings
- Lenovo Q3 FY2024 Earnings
- Synopsys Q1 FY2024 Earnings
- What to Expect at MWC 2024
- Third Generation Arm Neoverse
- Intel Foundry Direct Connect Event
For a deeper dive into each topic, please click on the links above. Be sure to subscribe to The Six Five Webcast so you never miss an episode.
Watch the episode here:
Listen to the episode on your favorite streaming platform:
Disclaimer: The Six Five Webcast is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors and we ask that you do not treat us as such.
Transcript:
Daniel Newman: Welcome back to another episode of The Six Five Podcast. It’s Friday. We’re on time, almost. It’s the best hour of the week for Patrick and I. For those of you tuning in, we hope it’s the best hour of the week for you as well. I mean, what else would be better? It’s a national holiday week. Yes, it’s NVIDIA day. Oh, I think there was actually a holiday this week too, by the way, on Monday. But I was talking about the one that everyone celebrated on Wednesday, and we’ll talk about that more today here on this show.
But Patrick Moorhead, it’s Friday. We’re getting on a plane tomorrow. We’re going across the pond, MWC. We’ll also talk about that later. But how are you doing, man? Are you ready? You relaxed? You feeling good? What’s your watch tell you about how you feel today? Because God knows, you wouldn’t actually answer it based on your own feelings.
Patrick Moorhead: Yeah, so for those who don’t know, I’ve instrumented my body. I haven’t implanted anything yet, but maybe in the future.
Daniel Newman: A bestie link.
Patrick Moorhead: I use an Apple Watch and I have an Oura ring, but I manage my stress. It’s not that I worry about stuff, but I’ve just gotten myself up into such a fit of excitement, working, crushing it for the last 35 years that I kind of overwork that adrenal system. So I literally have, on the face of my Apple Watch, a stress meter. So when I’m here with my bestie, it’s like in the teens, right? When I’m having a confrontation with somebody, it’s like 90. Then it bounces back and forth. But now it’s getting better, man. I saw my witch doctor yesterday and he says my physics-
Daniel Newman: Let’s tell that story, Pat. I mean, why don’t you give the background on your holistic medicine? We can do charts in a little while. We can get the charts up.
Patrick Moorhead: Possibly. Yeah, we can do that. Full blood test. But no, back in July I was really fat and unhealthy. Really wasn’t working out that much. Anyways, lost 34 pounds. I’ve kicked two out of three blood pressure medications. My vein health is up substantially. But no, I’m taking a lot of supplements, improving my gut health. Gosh, I’ve had like 250 different blood tests, a DNA cancer test.
Daniel Newman: Why don’t you tell people about the medicine you take? I think everyone would be super interested in hearing about that.
Patrick Moorhead: Oh, well, I think the biggest one is probably the fungus killer that I take. And that actually really is tough, right? But I’ve got a fungus overgrowth that’s going on down there. And bacteria is good, fungus is not. So I’m trying to get rid of that and what that does-
Daniel Newman: He’s talking about his gut, people, by the way.
Patrick Moorhead: I know.
Daniel Newman: He said he had fungus down there. He was talking about-
Patrick Moorhead: Fungus down there. Yeah, I think it’s yeast, to be specific. But anyways, my health is dramatically turning around. I feel better. My blood pressure is incredibly lower, to the point where I could probably get off the last one. But my doctor was like, “Give it another month. Let’s give you the reading. We’ll come back.” Inflammation is way down. Heart health is way up. So I don’t know, bestie, you might be stuck with me for 10 extra years than you were thinking. You’re probably like, oh my God, this guy’s going-
Daniel Newman: How do I get to just inherit all your business?
Patrick Moorhead: No, exactly.
Daniel Newman: I’m going to let it go under and then I was just going to just slowly slide there, slide into your customer DMs. Take them all away.
Patrick Moorhead: We should have-
Daniel Newman: I’m glad you’re healthy and I’m glad you’ve had the time to focus and dedicate to yourself. That’s good. And over the last few weeks, it’s been really good to get to know Pat a little bit more. Now we know you about your health. We know where your fungus lies. It’s been good to understand better how you’re using your Vision Pro personally. That’s good.
Patrick Moorhead: Yeah, the Vision Pro.
Daniel Newman: We getting inside the head of Patrick Moorhead. And by the way, everybody, when he flies to Europe, he’s planning to wear that thing. He’s going to be that guy. Now, I don’t think we’ve really redefined the glass hole category yet, Pat. But do you think there is one? Do you think people that wear their Vision Pros in public, I think you’re our very own Anshul Shah, was wearing it on a plane with a mask. I mean, he looked full on like part of a SWAT team.
Patrick Moorhead: I do think that looks cool. But I have to tell you, I am not going to be the one that’s bringing that. I mean, it’s huge. You know how I like to light pack. I’m going to Europe for a week and I’ve got basically a carry on and a backpack. So I’m not into that.
Daniel Newman: One pair of jeans, one pair of pants, one sport coat, and four or five shirts, and you should be able to make this happen, right?
Patrick Moorhead: No, totally. I mean, you can turn your underwear and socks inside out.
Daniel Newman: But what about the fungus? You got to be careful with this. We have derailed the show. But we’ve got a good one on us today. We’ve got a mix. We’ve got some earnings, we’ve got some big industry updates. We’ve got a little preview of MWC. Pat’s going to talk chips, chips and chips. And no guacamole with these, no dips. This Super Bowl is behind us. We’re in the silicon era. We’re talking semiconductors. We’ve got a little conversation about just how far NVIDIA can go. We’re going to then hammer on Lenovo, one of NVIDIA’s biggest resellers, and their earnings. We’ve got Synopsys, we’ve got our Neoverse updates, MWC, and we’ve got Intel’s big Foundry Day. Sam Altman was there, kind of, Gina Raimondo. We had Satya Nadella, some big news out of Microsoft. I wrote a speculative piece on this one on MarketWatch. It was a banger, but it wasn’t nearly the banger of my next piece that’s coming out. I can’t tell you what it is because it ain’t out yet.
Patrick was on TV for like a hundred hours this week. I think he’s a full-on contributor now. He did a great job over on CB, Closing Bell, with our friend, John Fortt. And Pat, this was a heck of a week. Real quick though, because we are going to be talking about markets. Disclaimer. This show is for information and entertainment purposes only. And so while we’ll be talking about publicly traded companies, don’t take anything we say as investment advice.
Patrick, let’s kick right off. I mean, I started the show. I said, happy NVIDIA week. I started this week on Twitter, “Happy NVIDIA week.” It was actually more ridiculous than I thought it was going to be. The stock’s now trading at what, over $800 this morning. So they gained an AMD. They literally had a little bit of a pullback before and they grew by an AMD, $270 billion of market cap gain in the wake of its huge growth. What happened, Pat?
Patrick Moorhead: So first of all, it was definitely the circus. There literally were countdowns on the earnings coming up on CNBC, right? Kind of like days till Christmas or something like that that you would like as a kid. And I literally have not seen this exuberance since .com 1.0 when Cisco was essentially the NVIDIA. It’s miraculous. But hey, what did they do? I mean, they beat on revenue by 7.5%. They beat on EPS by about 11.5%, which by the way was the lowest beat percentage over the last four quarters. But it was the exuberance that the gravy train continues on.
And it looks to me like NVIDIA is becoming very much a retail stock. And the impact that has, it’s pluses and minuses. But when you grow by an AMD, after you beat by 8%, this is what happens. And it was no mystery when you look at the size of the different businesses that NVIDIA is in as they blew away data center, right? 18 billion. By the way, seven quarters before, data center was 3.8 billion. I mean, it’s just-
Daniel Newman: Same size as gaming, right?
Patrick Moorhead: Yeah, absolutely. Gaming, back in the first quarter of fiscal year ’23, it was 3.6 billion. So they were basically the same size. And since then, gaming has shrunk. They did have that quarter over quarter was the same, which isn’t great particularly because that was the holiday time period. But yeah, it was all about LLM, multimodal LLM inference and training. It’s so funny, CNBC was kind of wrecking their brain. I could tell like, “Hey, what do we talk about on this?” And they asked me about automotive. And absolutely no disrespect to the folks at NVIDIA automotive, but I just said, “Hey, in proportionality, you’re looking at a $280 million quarter versus an $18 billion quarter. And overall, a $22 billion quarter. It’s 1% of revenue. So inconsequential. And even Provis, at 463, is larger than automotive.”
And I think the big question, Dan, you and I got hit with this. And just weigh in on this. How long does this continue? And listen, current course and speed in competition that’s out there, on a percentage basis, Intel is going to lose percentage market share to AMD and to Intel and maybe even some of the ASIC startups. But when you have a rocket ship that’s growing like this, all companies can grow immensely in dollar proportionality, particularly NVIDIA. And Dan, that was the big basis of your piece that I don’t know if you published it yet on MarketWatch, but you and I had discussed this last night. What’s your take?
Daniel Newman: Listen, this is a perfect moment and a setup for a victory lap. When you get it right, you got to pat yourself on the back.
Patrick Moorhead: No, you got to do that. You have to.
Daniel Newman: I have an extensive op ed that’s going to come out. And I don’t want to spoil it for everyone out there that’s probably literally sitting at the edge of their chair being like, I wonder what Dan’s going to write next about NVIDIA. But let me tease it out for a minute. So in 2020, I said NVIDIA would be the next trillion dollar market cap company. The conditions were slightly different, but it was more based on two factors. It was the tie-up with Arm and the impending AI growth. In ’22 when the stock had fallen like $150, and by the way, it’s churning out 800 now. I went on CNBC and I got a lot of pushback from the hosts on the show. I was on Squawk Box, about why I said Microsoft and NVIDIA would be absolutely some of the best bets that people could make at that point merely because AI was going to come so fast, so furious. And nobody saw it.
I kept talking. Pat, remember how I talked about tech being inflationary? Well, AI is the most inflationary aspect of technology. Anything that enables you to do more with less people. Remember Altman and the billion dollar one person company that he keeps talking about. Well, now it’s almost the opposite, Pat. It’s like you get to the other side of this thing and it’s like once everybody’s getting on the bandwagon, it’s kind of time to step back and go, hmm. Now when everybody’s like, “Yes, buy, buy, buy.” This is like that inverse Kramer thing that everyone jokes about. By the way, someone showed a video yesterday of Kramer talking about shorting NVIDIA when it bottomed at $130. So this stuff is fun.
Patrick Moorhead: I mean, by the way, it’s fun to pile on Kramer, but-
Daniel Newman: No, no, no. I’m not piling on him, I just mean-
Patrick Moorhead: No, I know that.
Daniel Newman: It’s part of history. We pile on at all moments, Pat. What were you saying? Go ahead. Sorry.
Patrick Moorhead: No, no, no, I was just saying, rational people would be like, that totally makes sense.
Daniel Newman: Well, when things are going bad, the assumption is it’ll never get better. And when things are going well, the assumption is that it’ll never get worse. And both of those assessments are emphatically wrong. And so the real question, to your point, is how long does it continue? Well, you made a good, astute assessment that the growth is actually already slowing. Now, this is a law of large number of things. You could align it to the cloud numbers and their growth. Again, none had growth that was as astronomical as what we’re seeing here with NVIDIA. But NVIDIA has done this with effectively no competition. They’ve effectively had no meaningful competition. And so now over the next few quarters, for the first time, meaningful competition is popping up.
And the second kind of factor about competition is there seems to be a meaningful consensus among the competition that they know they need NVIDIA, but they also know NVIDIA is too strong. And so whether that was Satya Nadella talking about networking chips and basically developing chips that would enable data flow in Azure, that would not require as much dependence on NVIDIA. AWS already did this with its adoption of DGX Cloud. It did not use NVIDIA’s networking. You have Google GPUs, Meta’s developing its own chips, even though it’s the biggest buyer of NVIDIA chips. You have Oracle building with Ampere. You just kind of go up and down the stack. Everybody wants to build their own. And then you had AMD, of course, Lisa Su’s got this big forecast. And this is because there are companies that want to have a competitive choice.
And also, Pat, what does it mean with what Intel is doing, with what Sam Altman is suggesting? These are all very speculative, but clearly Sam Altman thinks we need an alternative. I mean, he’s not saying pour it all in and help NVIDIA get bigger. He is saying, let’s create an alternative. So you’ve got the different architectures, Pat. You’ve talked a little bit about FPGA, but definitely the ASIC architecture, very specific workloads on recommendations and filtering. Those are just a couple. For instance, you’ve got power requirement issues that are to be considered. And GPUs are good at a lot of things, but they’re not great at any one thing. That’s kind of how they’re designed, a lot like general purpose CPUs. So all these things are kind of just a series of conditions that say NVIDIA will keep growing and it’ll keep being successful, but not at the same rate.
And then when the rate slows down, investors, especially retail investors are fickle. Like, oh, they’re only growing. So that next quarter will be the first time that the company is growing on a year over year against the generative AI boom quarters. So what if next quarter after they grew 250% revenue a year ago, they only grow 25% over the 250%? People are going to be like, “Oh my God, it’s slowing.” It’s like, well, it’s actually growing really fast. And so these are all the conditions that I’m looking at saying, “Look, they’re doing great. It’s just different when competition rises. And these are real competitors. These are formidable companies.
And Pat, what about Apple? I mean, we haven’t even heard about what Apple’s going to do, but I seriously doubt they’re going to want to be, in their vertical integration strategy, as dependent on a single company as most of these cloud providers have been on NVIDIA.
Patrick Moorhead: Yeah, Apple recognizes they can’t replicate that in the cloud. But I got to tell you, their cloud spend is big. Apple used to do servers back in the day, but they don’t do it anymore. Hey, Dan, what would you think about, is there a potential acquisition play here?
Daniel Newman: Acquisition for who?
Patrick Moorhead: For NVIDIA. You got to keep the growth going and you don’t know how to.
Daniel Newman: I don’t think they can acquire anybody, Pat. I don’t see a circumstance. I think what they are doing that’s really smart is some of the seed investing that they’re doing in some of these very cool startups. They’ve got all this excess cash flow, Pat. They’re putting bets all over the market. It looks like a roulette table. There’s a hundred interesting startups. They’re probably putting money in 90 of them right now. And there’s going to create a loyalty and a commitment and access to technology and IP. I don’t see any circumstance in which they could make a big deal right now. Do you? Do you think they could do it?
Patrick Moorhead: It’s such a hard thing to answer, because you’re trying to get in the head of the regulators who are crazy right now, particularly over in Western Europe. So I got to say no in the current environment. Even if it has nothing to do with their current market, like buying a Qualcomm or something.
Daniel Newman: Too powerful.
Patrick Moorhead: Like that.
Daniel Newman: Too powerful.
Patrick Moorhead: Totally different markets. I got to tell you though, five years ago, that would’ve been a shoo-in when you look at the definition of markets. But again, we’re in this wacky time where big companies be bad just because they’re big.
Daniel Newman: Listen, you got to admire the success of NVIDIA. This is one of those where I’m not criticizing for the sake of criticizing. I’m just saying it’s gotten to the point where everybody’s on the bandwagon. And that always scares me. When everyone says it’s like an endless upward and to the right, you’re like, uh-oh. Same thing with the doubt. And everyone’s like, “This is never coming back.” You know, Pat, you like to devour history. You talk about that.
And when you read history, you do realize that there’s a reason that people make those statements about cycles and history repeating itself. And there’s kind of a cast of different reasons that happens. But one of them is because we’re stupid. As a society, we’re just stupid. Things happen. We see it happen and we let it happen again. And by the way, it’s kind of like that. You kind of wonder, are we actually that stupid or are we just being led on as a society? But, I don’t know.
Patrick Moorhead: Yeah, Dan, I know. Listen, this is an important topic, and I know we’re talking a long time about this. But what is unprecedented historically is that the big companies keep getting bigger. Every company had cycles where they would be dominant for about a decade. Probably the longest run of domination was IBM, and they’re still a very important company out there. But the dominance, I’ve never seen before. We had the run-up in Cisco. And yeah, they’re still dominant in networking, have a great business, but they’re not dominant overall. And Apple is the one that stymies me the most. Consumer electronics. Every 10 years a new company comes along, whether it’s Sony or-
Daniel Newman: LG, Samsung.
Patrick Moorhead: Toshiba what was dominant for a time as well. So out of one side of my mouth, the big company bad is odd to me. But then again, if you look at the concentrated value, isn’t NVIDIA worth more than all equities in China now, the entire China stock market?
Daniel Newman: Something like that. I don’t have that number in front of me, but we saw that it’s bigger than the entire S&P 500 energy sector, all the companies in it put together on market cap. So remember when energy companies were the big, bad companies?
Patrick Moorhead: Yeah.
Daniel Newman: Now we have a tech company that’s worth more than the entire index, all the companies in it.
Patrick Moorhead: Did you know we’re pumping more oil in the US than the Saudis now?
Daniel Newman: We have to, because we can’t make the transition as fast as everybody thinks we can. Again, I don’t know, there’s probably a theme here, but maybe people are stupid. Could be the new name of the podcast. We got to read stuff. And don’t just read the headlines, actually read the text, read books, read history, and you’ll learn a little bit here. And Pat, by the way, it was pretty cool that you were in the Steve Jobs bio, not by name, but you had a role to play. It was fun. It was fun reading through that one.
Patrick Moorhead: Yeah, maybe we’ll talk about that later. I was feeling good once after 25 years. I realized that my product as a product manager, Jobs tore apart on stage at the iMac announcement. Yeah, that was pretty cool. But anyways, you know what, we should probably move on.
Daniel Newman: Yeah. All right. That was the one I was really excited to talk about though. Let’s talk about one of NVIDIA’s biggest partners in the ecosystem. Let’s talk about Lenovo, Pat. So you and I had the chance to sit down with both Ken Wong, the president of the SSG business and Kirk Skaugen, the head of the infrastructure business, and talk a little bit about the company. Now it’s interesting with these large technology OEMs. They’re so much more than that now, and I know they always get frustrated. I don’t know the exact way they’ve been titled because they’re device makers, they’re infrastructure, they’re service providers. We need a new category for these kinds of companies that do so much. They’ve got on-prem cloud consumption services.
And the company did return to growth, which was a really, really good thing, Pat. It’s been a tough couple of years for the likes of the Lenovos, the HPEs, the Dells. And even though a lot of them have managed through very well, you can manage very well when you’re doing $50 billion or $100 billion a year of revenue. It hasn’t been on a growth-oriented. Remember, NVIDIA grew 265% or something like that. Lenovo grew 3%. So 15.3 last year, 15.7 billion this quarter. It has however continued to see, on a quarter to quarter basis, strong sequential growth of income.
And probably another thing, Pat, that I’ve been paying a lot of attention to with Lenovo has been the growth of the company’s non-PC revenue. And the company got a few more bits there on an annual basis. Now 42% of its revenue, non-PC. And why is this so important? Well, you and I both know the PC AI, AI PC boom is coming. Signal 65 will be looking very closely at this. That’s our testing arm by the way. We’re going to be looking at testing what AI PC can do. We are also looking very, very closely at how companies that are dependent on these kind of cyclical technologies that tend to be also under incredible margin pressure are able to perform.
And by being able to move into the non-PC strong growth and SSG, they had a $2 billion service billing. Some strong alignments with companies like ServiceNow. Their AI strategy continues to grow. They’re seeing year-on-year growth in their PC portfolio for the first time in a long time, Pat. And this answers a question about a cycle that everyone’s been worried about, is the PC cycle coming back? So they were able to diversify away from PC. And now that the PC business is returning, they’re able to start to see some growth.
So this is a company that is going to be very instrumental in providing infrastructure to cloud providers. They do that around the world. They’re providing infrastructure to the enterprises. They’re going to enable AI for the enterprises through services. And of course, they will be one of the largest unit movers in the AI PC era, Pat. And so it’s definitely one to watch. It won’t get the number of headlines that NVIDIA’s getting, and that’s why we’re talking about it here. But it’s returning good earnings. It’s creating good income on revenue. It’s growing, albeit steadily, Pat. Overall solid. It’s hard to follow up NVIDIA. It’s just hard because I think they had 467% earnings growth. But yes, good quarter for Lenovo.
Patrick Moorhead: Listen, by the way, I like the overall assessment. It is staggering to me that NVIDIA can go up by that much. And companies that have servers like that won’t go up that much. It’s kind of mind-boggling. Now, we did see a company, Supermicro. By the way, Supermicro’s stock, and it’s interesting, I get hit on this on CNBC on occasion, but their stock is up 215% in the last six months. It’s absolutely mind-boggling to me. $50 billion market cap. Basically, they do ODM servers.
And I think that they are getting a lot of benefit from this lift. Some really positive stuff from Lenovo. Essentially they resumed their year-on-year growth, which again, very positive. They did make an interesting confirmation that, like I said on stage at CES, AI PC stimulates the refresh cycle. 50% of PCs will be AI capable by ’26. And by the way, when you disconnect desktop from notebooks, it probably means that 80% of notebook PCs will be AI capable. And the reason for why it’s not larger overall is essentially that Intel and AMD haven’t decided to put the big NPU in those desktop parts or not.
In the SSG side, Ken, great conversation with him. Huge milestone. That division’s only been in existence for three years, but they’re up $2 billion. And I think that’s a pretty strong milestone. And sure, they didn’t start at zero because they did have a lot of the services that pulled through from the hardware. But when you look at the percentage of it, it’s less than 50%. Actually 55% of SSG’s revenue comes from non-hardware type business. And if you would’ve told me three years ago Lenovo in services would be strong, I would’ve said, “You got to be kidding me. That’s going to be a stretch.” But to their credit, they’ve come out swinging here and really focused on the areas that matter, like hybrid cloud, remote desktop, PC as a service. They’ve got this new AI fast track professional services out there. So hats off to the service group.
ISG, we just got off a video with Kirk Skaugen, looking good sequentially. And by the way, all time record for storage, all time record for edge revenue. It’s that core server that’s been the challenge. And at least, in my channel checks, a lot of the decisions that were made for LLM servers three years ago, it wasn’t supposed to be that big, ended up getting huge. And a lot of times these hyperscalers will do a mixed vendor bid. And there’s the LLM ones and then there’s the standard non LLMs, and that’s where all the cloud guys are putting all their investment.
So IDG, it’s interesting, they came out swinging, right? They claim to be increasing their market share. I mean, they’ve been number one forever. And they also said they got 27% of commercial, which is really a Dell stronghold. And probably the biggest surprise to me is that their operating margin, they said industry leading operating margin. I’m going to check that against Dell, but I’d be very surprised if IDG has a higher operating margin than Dell because Dell does less consumer, which is lower profit. And Dell does more commercial, which is higher profit. Just to summarize, looking forward to meeting with the Lenovo ISG team at Mobile World Congress. We’re going to shoot a half a day of videos there and hope you can tune in.
Daniel Newman: Yeah, it’s going to be a lot of fun. Looking forward to doing that, Pat. That’s what we do. That’s what-
Patrick Moorhead: It’s part of what we do.
Daniel Newman: It took us 30 minutes to get through two topics.
Patrick Moorhead: Yeah, we got to rock and roll.
Daniel Newman: I want to be clear that I thought the eight minutes we spent talking about your fungus down below-
Patrick Moorhead: It’s important stuff, dude.
Daniel Newman: A healthy Patrick is a good analyst Patrick.
Patrick Moorhead: The happy Patrick. Yep.
Daniel Newman: Happy Patrick.
Patrick Moorhead: We’ve got 17 minutes.
Daniel Newman: We got to get moving, then. I just kind of want to keep going back to this fungus. I don’t know, it just does something for me.
Patrick Moorhead: We can do a full segment on it next time.
Daniel Newman: Why don’t we talk about Synopsys because, talk about a company that’s got some work that it’s going to have to do to support all this AI explosion, Pat.
Patrick Moorhead: Yeah. So Synopsys is tip of the spear for all this chip chippery, right? Not only do they do EDA or SOCs and chips and IP, but they also do system simulation and they’ve got this big potential Ansys acquisition that should hit in ’25. But listen, they beat on revenue or they met on revenue, but they beat on EPS by 4%. They got a great lift on the stock market. They had record quarterly earnings revenues of 21%, a big OpInc increase at 3.5%. But no, I liked the way that they came out and talked about the industry first, right? First GAA, that’s gate all around, Arm Cortex X tape out, leading mobile supplier, by the way.
This is a company you have to listen to the earnings call to actually get it. But they had some really great highlights on AI, showing how it’s getting DSO.AI. Drove a 20% revenue uplift, winning multi dive package designs, record revenue and software integrity. But no, very solid thing. And I’m really interested to see. I think that Synopsys is going to be one of these power brokers, like big power brokers in the next five years. I don’t know if they’ll be the next NVIDIA. But I got to tell you, this space is so hot. It’s democratizing the way that companies can do design. It would’ve been nuts to think that an AWS or a Microsoft could develop their own SOC. But with tools from companies like Synopsys and Cadence, it’s the real deal.
Daniel Newman: Yeah, I had the chance to talk to their CFO, Shelagh Glaser, and it was good to get the kind of run down. I think you said something that was worth double clicking on, Pat. And that’s, this is not a company that you can just sort of tip tap in for a moment and get. What they’re doing is so important. Between what the second-largest portfolio of IP for chip makers as well as EDA, and now the Ansys deal is going to make them an even bigger power broker. By the way, the size of that deal was not immaterial. I mean, you look at that, remember how big you thought the AMD, Xilinx deal was? Similar in size. Why is this? Because this era of building chips, this multi-trillion dollar sort of economic expectation, and you can talk about whether that’s chips, infrastructure, electricity, design, whether it’s packaging, all the different things that go into that.
But there’s a lot of complexity and there’s also a lot of companies and players that want to enter the market. They want to speed execution. They want to speed design. They want to continue to advance with the most powerful designs as well as, again, trying to address the sustainability issues. Which, Pat, this is a practical version of that, not a greenwashed version of that. We’re going to be deploying massive amounts of scale to compute, and that’s going to require a ton of consideration. So whether it’s the generative AI tools that they’re using to streamline and shorten the design cycles or it’s the IP that their customers are able to license. This company is going to be more and more critical, more in the forefront of the discussion around AI.
And as long as AI continues to be the boon, by the way, AI will fuel more regular compute chips. AI will fuel more networking chips. AI will fuel more lagging edge chips. All this stuff kind of happens and it all coincides. And I think we always want this sort of zero-sum, like, oh, it’s all about the leading edge, newest node AI chip. But look at a vehicle. It went from a handful to dozens, to hundreds, to thousands of chips. This is a variety of silicon that’s going to be in these intelligent vehicles. And Pat, it’s companies like Synopsys that are going to be critical to the contributions required to put these things together.
So you kind of hit on the results. They’re good results. The leadership of the company seems to be doing the right things operationally on margin. They moved the midpoint up on guidance. They seem to be stable on profitability. It was just an overall really good quarter, but it’s a company you’re going to need to spend a little more time to understand. It’s not GPUs or CPUs that people kind of all get. There’s a lot of very specific technical prowess that is underneath this company, but they are what, Pat? They’re the critical partner to just about every one of these major fabless and design and manufacturers of silicon in the world.
Patrick Moorhead: Yeah, statistically, Synopsys and Cadence, you have to buy both of them because they have some different strengths and weaknesses. But good talk.
Daniel Newman: Yeah, well, I mean we’re playing the game out there for everybody that wants to know. By the way, when people kind of go, what are the other picks for AI? Because the natural gravity is always about to pick one of two things. It’s either the chip or the software that everybody knows. This is the software and IP that nobody knows is out there. And when you’re kind of trying to say what are the kind of smart companies to maybe invest in, this could be one. Again, we’re not giving investment advice.
All right, pal, let’s talk about MWC. So we’re on our way out there. And listen, I’ve been on a few pre-briefings, you’ve been on a few pre-briefings. This is my 10th or 11th MWC now, I think every year except for the ones that got canceled. So what are going to be the themes, Pat? Top of mind, 5G, I know it’s crazy. In 2018 or ’17, I was doing 5G readiness assessments going into Mobile World Congress. We’re still, in some ways, doing 5G readiness. By the way, we didn’t even talk about the fact that AT&T was down for a lot of people all day yesterday. How crazy is that?
Patrick Moorhead: So dumb.
Daniel Newman: Not only is it dumb, but our lives are completely at a stop when you’re not connected. I don’t know when this happened, but all of a sudden I don’t know how to function if I’m not connected. That’s going to be a big theme here at MWC. You’re going to see the connecting of everything. So we kind of have heard about IOT and industry and all these things, but the bottom line is the importance of networking, the importance of connectivity, the importance of the service provider networks and all the things that we’ve grown to become dependent on is what makes and enables us to use this technology every day. So we’ve got 5G, we’ve got connectedness. Something that we spend some time on always is, of course, they always spend some time on the manufacturing side of things. Smart manufacturing industry, I expect that to be a continuation.
You’ll see things like network slicing and different opportunities to create secure connectivity at the edge. And Pat, AI, I waited till last just because I felt like it was so obvious to me, but AI is in the headlines. Whether we’re mobile or in the data center, an on premises, AI will continue to be a massive topic. It will be a topic from the keynote stage for sure at Mobile World Congress. And I’m looking forward, by the way, to spending time with the usual suspects. We’ll spend time with Qualcomm and Intel, Cisco and HPE, Dell. But then there’s also a lot of cool new companies that are there. There’s like Snowflake, are very involved. Microsoft, very involved. IBM, very involved. And so we’ll see a good diversity. And Pat, we’ll be running quickly and fast. The Six Five will be on the ground. And of course, you and I will be there analyzing and advising the markets. So there’s my take. Quick and to the point.
Patrick Moorhead: Yeah, I’m just going to end with, it’s AI, AI, AI in a very similar way that we saw at CES, just in a different form. And it takes really two forms. The first form is how can AI be used in products and services to improve the outcome? So for instance, the way that traffic is handled either at the base station or at the core network or even on the edge. And then there’s, how are these B-to-B companies helping other companies, let’s say, accelerate AI traffic that goes from the edge device to the edge, to the core switching network or the base station? And those are literally going to be the two themes. That’s going to cross the industrial IoT. It’s going to be talking automotive. We’re going to see consumer markets. But that is undoubtedly going to be the push.
Now the interesting part is, a lot of the algorithms that really make these types of workloads better are more machine learning than generative AI. At the end point, totally generative AI. But I have yet to see what generative AI can do, let’s say, in the network as opposed to machine learning. But I’m excited to get there. I leave on Saturday, I think like you do. And we’ll be there for most of the week.
Daniel Newman: Right on. So two topics. And how are you doing on time?
Patrick Moorhead: Oh, I’ve got a hard stop.
Daniel Newman: You do?
Patrick Moorhead: Yep.
Daniel Newman: All right, we got to go fast then, Pat. So we’re hitting Arm Neoverse.
Patrick Moorhead: We are hitting Arm Neoverse, baby. So it’s been interesting. And the numbers are hard to pin down, I think quite frankly because of China. But Arm went from close to 0% market share in servers, to between, I’ll say 15% and 20% in about a decade. Now, the first eight years were the hardest, but now that we see the proliferation of Arm in the data center and custom design from the top CSP, and yes, I’m expecting Google to come out with an Arm based server SOC, there’s even numbers that say Arm has 40% server share in China, which is just staggering. So they’ve gone from no respect to very respected.
And most of the growth, if you followed Arm’s earnings, was because of the data center. So essentially what they came out with was a couple of things. First of all, two new Neoverse CSS products. And I stand corrected, I was corrected that CSS is not a white glove service. It is a product. Now, that product is not just off the shelf. Every customer takes in different types of capability. In fact, Microsoft Cobalt, Azure’s new, or actually overall Microsoft Arm based SOC was cited as a CSS design, which I think is pretty impressive.
The company also flashed some gen AI Neoverse performance. Again, the whole CPU. Doing it on the CPU. The industry’s dirty little secret is that holistically, more AI is done on the CPU than a GPU or an ASIC or an FPGA. Nobody really wants to talk about that because I think people view it as a weakness. And they talked about the new CSS roadmap. Didn’t go through details on it. It was basically, yeah, we’ve got a next generation for V3 that’s performance optimized, and N, which is a performance per watt optimized. They put up some benchmarks that, no, we did not run at Signal 65. But they seemed very provocative, not only on generative AI but also versus Xeon.
Daniel Newman: Yeah, it seems like this would be a good one to dial into. And definitely, it’d be good to get some testing metrics, Pat. I am going to let you have the floor on this one because we only have three minutes, and I want to make sure we do get to the last topic. We’ll come back, we should talk more about this. So Arm continues to make great advancements, and their stock has doubled in less than a month. So doubled, everybody.
All right, so Intel had a Foundry Day. This was a big one, Pat. You and I had a lot of preamble on this one. A lot of run up, there’s a lot of excitement. Pat Gelsinger did his victory lap on five and four. Had the chance to bring Gina Raimondo on stage to talk about the US, I think she called Intel a national treasure or something of that sort. And basically talked about a vote of confidence from the administration that they’re going to continue to support the company. But he also got some huge backing from Satya Nadella, which announced that it would be partnering with Intel. It was able to declare. Which, Pat, to some dissatisfaction, I think about the $15 billion lifetime value of its pipeline. A big number. But again, remember, NVIDIA grew 276 billion in a day in market cap. So I think everyone was kind of puzzled about that one. But Satya’s support and Microsoft buying in was important.
But he had to cast to everybody there. They had Synopsys there, they had Cadence there, they had Broadcom there, they had Microsoft there. I mean, there was a huge support structure. Arm, right? Rene Haas was there. And you had Sam Altman there. Now unfortunately, I wasn’t able to be in the room, so I did not get to hear what was said there and I was devastated about it. But Pat, this was a big moment for Intel in terms of a coming out around its Foundry.
Now, when I made a bull case about a week and a half ago, I wrote a bull case on MarketWatch. And one of my three items, biggest one is Foundry. Yes, the company has improved its process. It does commit itself to technology leadership over the next couple of years. That is a big challenge. Having said that, the desire to have a US-based or west based Foundry doing the leading edge with differentiation and packaging, which Intel has proven already that it has, and the packaging being perhaps more important in the long term than wafer, I know. I said it. I’ll let you weigh in on that one, Pat.
But in this overall AI boom, it seems like Intel is in a very interesting place, that it could really see some acceleration of its Foundry business. This has been the fastest growth part of its business. It’s been a bit of an encouragement point. It’s got good support from the policy makers, government around the world. The company’s expanding in Europe, it’s expanding in Israel, it’s expanding here in the US. And this was a moment for the company to celebrate that. Having said that, they also did it during a different national holiday, NVIDIA earnings day. Joke, I can’t help myself. But it was a big day, Pat, and it went all day long. So I know you got to run. I wanted to talk fast, hopefully give you at least 60 seconds to a minute, two minutes, whatever you need to weigh in on your side. And we can come back to this later, talk more about it if it makes sense.
Patrick Moorhead: Yeah, we had Intel Foundry, I guess he’s the CEO now, Stu Pann, on The Six Five. And what’s interesting is, you go back and read that interview and you read between the tea leaves, there was a lot of confidence and a, hey, you really should join us for Direct Connect Day. And they delivered. I didn’t expect a major customer announcement like Microsoft. And yeah, I got caught up in the $15 billion contract. But it ended up that that’s $15 billion so far overall lifetime value for the entire IFS business, which, by the way, is up 50% from Intel earnings day. So I think that was a pretty big deal.
Listen, three years ago I caught up with Pat Gelsinger right as he made a reentry back into Intel. And I wrote a piece that essentially said if they execute on this, this and the other, they will be back. I can’t tell you the attacks that I got on that. But here’s the deal, I was Intel’s biggest customer. I was at Compaq. I was Intel’s biggest competitor when I was at AMD. And I know this company very well and what they are capable of. I didn’t Babe Ruth it. I didn’t say they would. I said this is what they would have to do that they do it. And then that turned into five nodes in four years. And everybody said, there is no way. And by the way, I totally get it. After Intel stumbling on seven nanometer, I mean it’s like, why would you ever think that this would be possible?
Really good progress so far. That number needs to go from 15 to 50 in short order. But if I look at it, I do believe that they will do NVIDIA packaging. And not just because TSMC can’t do enough CoWoS. It’s because Intel’s packaging is really good, with things like Foveros and EMIB, and you’ll see a lot of designs that will use both. And you have MediaTek who, by the way, is taking low-end market share away from Qualcomm, which is interesting. So I do think once Intel shows up with the goods, maybe it’s after 18A, I do think Qualcomm will do something with Intel.
And what a lot of people don’t understand from Qualcomm is they have the most diversified manufacturing of any SOC provider out there. I mean, they do GF, they do Tower, they do TSMC, they do Samsung. I mean they’re all over in the support that they do. And I got to tell you, I am going to Babe Ruth Intel Foundry and say that it will be a very successful ongoing concern that ultimately could be split as a tracking stock out there. Now, people are talking about splitting the company up into two. That would be stupid because until you have full design autonomy and using all industry standard tools, which Intel doesn’t currently do, it’s about a seven-year process to do that. I know that because I saw AMD go through it when we were going from proprietary tools to Cadence and Synopsys.
Daniel Newman: Yeah, and I’ll pull together here really quick and we’ll close this baby down. But yeah, Pat, it also would be just stupid from their business model standpoint. You could see the outsider view on why they would maybe want that to happen. And of course, could it drive more dollars and support across fabless companies if there was, but that’s like SoftBank and Arm. You own 90% of it. It’s like, is it really public? It’s kind of.
All right man, listen, this has been a great show, another great week, great way to start my Friday, Pat. We’ll be coming to you a lot and frequently and live from MWC. Thanks for tuning in so much. Hit that subscribe button. Join us for all of our shows. We appreciate the community here, all of us, you, me and Pat’s fungus. We all enjoy you being part of The Six Five community. But for this week, for this episode, time to say goodbye. Pat, get to your call. It’s breakfast time. We’re out of here.
Author Information
Daniel is the CEO of The Futurum Group. Living his life at the intersection of people and technology, Daniel works with the world’s largest technology brands exploring Digital Transformation and how it is influencing the enterprise.
From the leading edge of AI to global technology policy, Daniel makes the connections between business, people and tech that are required for companies to benefit most from their technology investments. Daniel is a top 5 globally ranked industry analyst and his ideas are regularly cited or shared in television appearances by CNBC, Bloomberg, Wall Street Journal and hundreds of other sites around the world.
A 7x Best-Selling Author including his most recent book “Human/Machine.” Daniel is also a Forbes and MarketWatch (Dow Jones) contributor.
An MBA and Former Graduate Adjunct Faculty, Daniel is an Austin Texas transplant after 40 years in Chicago. His speaking takes him around the world each year as he shares his vision of the role technology will play in our future.
