The Six Five team discusses AMD’s latest earnings.

If you are interested in watching the full episode you can check it out here.

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: AMD, Pat – a company we are both bullish on, but how did they perform this quarter?

Patrick Moorhead: Yeah, they performed pretty much exactly as they said and versus expectation. They beat on EPS by six and a half points, and they beat on beat/met by revenue… and by the way, that beat/meet is within 1% for me. They beat on revenue by 0.84%. And going into that, I think that’s pretty good, particularly when you look at the market, right? What they’re walking into is an extremely light PC market, and overall, the server market is difficult. On the plus side, they really crushed it and embedded and really showed the street, again, one of the reasons that they acquired Xilinx was, bringing in Xilinx makes the highs less high, but makes the lows less low. And that’s exactly what we’re hitting here. The PC and the server market are in the sh**s, and Xilinx really, really helped them. The number, they absolutely crushed it for growth and embedded.

Yes, I get it. It still has some acquisition numbers in it, but 163%, even though the server market was down, they still had double-digit epic sales. Lisa made a comment on the call, that she said that, “Yeah, we think the conditions are right that we could continue to gain more server share this year.” That’s in alignment with what I’ve been saying about about 2023. Intel has a much stronger roadmap in ’24 and ’25. One of the things that Lisa did do – and we’re going to talk about this even when it comes to Qualcomm – is Lisa gave a little bit of a hint about the future of AI. She called it on the call “The company’s number one strategic priority”, and Lisa doesn’t mince words, she doesn’t get ahead of things. She pretty much… what Lisa says happens. They formed an AI group under ex-Xilinx CEO, Victor Pang, which by the way is an ex-AMD guy, ex-ETI guy. And he has been working for a while cross-collaborating with the team of how to move this forward.

And AMD is not new to AI. Actually, most AI inference is done on a CPU, although I know it’s not cool to talk about that. But when it comes to discreet cards, whether that be training or inference, NVIDIA owns the day. If they don’t have 90%, 95% market share, I know you’ve pointed this out in your market watch articles, it’s still a lot, but AMD has had a lot of success with National Labs where it’s sure it’s about flops, but also as they move to AI, it’s about AI. And this MI-300 accelerator that’s coming out. I’m really interested to hear about.

The number one thing I want to hear about, is the software strategy. I 100% believe, and I told a Forbes columnist this, the print, not the web folks like me, that I fully believe that AMD can feel the right hardware. The question I have is on the software. We don’t do rumors on this show, but Bloomberg did bring out a rumor, interestingly enough when I was giving this interview, that said Microsoft and AMD are going to be collaborating an AI accelerator. So I’m going to leave it there. I’ve had more to say. I might come back and loop around, but that’s my overall thoughts.

Daniel Newman: Why don’t we do rumors?

Patrick Moorhead: I just did it. I just did it.

Daniel Newman: Why not?

Patrick Moorhead: But we can dive in. Dan, why don’t you… why would Microsoft ever want to collaborate with a company on, and not just have it be an all NVIDIA shop?

Daniel Newman: I don’t know. I mean, price, performance, flexibility, open source, scale, ESG, lots of different reasons. Right now the pot of gold is very, very large, and everybody’s trying to figure out how to wrap their arms around it with AI. We’re going to see massive growth, mid double-digit CAGR for AI over the next several years according to numerous different surveys out there. And with the speed and trajectory, I am actually, between you and me, not sure that those surveys are fully caught up to the opportunity that actually is AI. I think we’re going to see bigger CAGR growth, and again, AI is going to go from being sort of a vertical or a technology adjacency to an overarching theme, like digital transformation, which by the way, I know a guy that knows a little bit about this.

You know, quickly… you hit a lot of the key points to me. This was probably business… I kind of think I quoted it as business as usual for AMD. They put forth a number, they beat it, they guided, a little bit soft. I think Lisa’s being conservative right now. There’s still a lot of uncertainty in the macro, and it’s a responsible CEO that’s going to look at that situation and say, “I’m going to put numbers forth that I’m very confident that we can achieve and if the market gets a little bit better, and we’ll beat it.” And I prefer that to overly optimistic. There’s still too many uncertainties out there. I mean, didn’t another major bank fail this week? It was FRC. I think there’s three or four more teetering. We raised interest rates again.

And let’s be clear people, all these things are intertwined. This is not like, “Oh, the bank issue’s over here.” The banks are where the money is being held for companies that are paying and operating and investing. Not to mention, banks themselves are buyers of all, net buyers, of a lot of technology. And right now you’ve got an entire subset other than the big three or four banks that are all teetering. They’re all on stilts right now. Our economy’s a little bit on stilts because I don’t think the market is fully caught up to all the macro things going on.

So back to a AMD though. AMD’s performing well, it’s performing fine. This quarter was, like I said, business as usual. Everything that’s going on in the macro, especially the PC numbers, all took place and that’s to be expected. The overall consensus from Lisa and other CEOs, has been that the back half of this year will be better for semis. And by the way, I believe semis, because they sell ahead of when the OEMs actually ship units. We’ll start to see some improvement in semis probably before we’ll start to see it in devices sans Apple. Apple’s numbers are always just mind-boggling to me.

The biggest thing for AMD though, and what I put out there in MarketWatch, is that NVIDIA’s in a really unique juxtaposition right now, because NVIDIA has close to 90% of market share. And NVIDIA has enjoyed this wild lead and competitive advantages across the board, but everything within NVIDIA has been built around CUDA. And CUDA is its proprietary language. It requires that companies basically commit and lock into this environment, but you’ve got the rise of new competition. We talked about the Microsoft rumor, we talked about what’s going on with AWS and Inferencia, Google TPUs. There’s an advancement there. And, of course, you’ve got Intel and AMD that are saying, we’re not going to go away and not play in the space. You’ve got startups like Groq who we are investor in. We’ve got SambaNova and others that are trying to build lower cost accelerators that can do large language model workloads at a much lower cost, and are much more efficient from a power perspective.

So even if the market gets bigger and NVIDIA keeps growing at a really high growth rate, I would not be surprised if they start to cede share. This is the interesting inflection point, and I think AMD could be the company that capitalizes first. It has the popularity, it has the woo, it has the investor sort of backing, and people really seem to like it, and it seems to have a plan. Now whether this comes to fruition, we shall see.