Advanced Micro Devices, Inc. (AMD) Earnings Call Transcript & Summary

I was telling everybody that I spend 75% of my marketing talking about AI to start. Jean Hu, CFO. Happy to have you here. And I think maybe -- so a way to kind of frame some moving pieces. You were just at COMPUTEX, which is usually a PC show, but it had AI road maps from both companies.

So maybe kind of start with that. Two interesting areas, AI PC, but then obviously, data center AI. Maybe step through some of that, and then we'll get into some of the questions.

Yes. First, thank you. Thank you for having us, and thank you all for joining us this afternoon. You're right, we had really exciting COMPUTEX. There are a lot of announcements from AMD. Lisa Su, our CEO, actually did the keynote, and unveiled the leadership CPU, GPU and NPU architecture leadership. Actually AMD is the only company who have end-to-end solutions that cover CPU, GPU and NPU from a data center and then to PC and eventually our embedded business, which is Xilinx FPGA business, we do think Edge AI will happen in the future. So we have a broad portfolio to cover everything. So going back to, on the announcement side, the first is AI PC because it is a consumer show, we actually announced about our Ryzen AI PC 300 series. It's for premium, like actual thing notebook. It actually has the CPU and latest GPU and the NPU, which are all on one single chip. The TOPs, Microsoft one is 40 TOPs, we actually have 20% more TOPs, we are the only one who reached the 50 TOPs to really power the Copilot plus the AI PC. So it's one of the really amazing product. It's going to be available in July. We also announced our Ryzen 9000 desktop processors. It also has leadership performance in AI inferencing. So that's on the PC side. Then on the data center side, we actually previewed our next-generation Gen 5 EPYC CPU servers, it's code named Turin, Turin extends our leadership in both performance per watt and performance per dollar significantly. So that's going to be launched in second half of this year, very exciting. But most importantly is Lisa actually previewed our GPU AI accelerated road map on the annual cadence. So later this year, we're going to introduce MI325 which is going to have 288 gigabyte HBM3E memory and have significant memory capacity and bandwidth better than our competition. And then next year, we are introducing MI350, which is based on cDNA4 a new architecture, which is also going to have a 288 gigabyte HBM3E memory and also more importantly, if you look at the generation over generation, the performance improvement is 35x. That product will compete with literally Blackwell 200, and will have a similar generation over generation performance improvement. And then in 2026, we're going to introduce MI400, which along with NVIDIA's Rubin GPU that is going to continue to expand both the leadership inference performance and the training performance, so overall, very exciting road map, we feel pretty good about annual cadence because we have been the first one to adopt the chiplet architecture. That gives us tremendous flexibility to be able to have lot more memory and to be able to accelerate the road map.

Right. I think they're definitely -- I mean, obviously, NVIDIA has been vocal about a 1-year cadence, you kind of matched on the GPU. I would say the questions I get a lot is that they introduced GPU, they have a CPU, they have a NIC Card, they have NVLink Switch, Ethernet switch. So it's a whole portfolio that ultimately makes up a rack of AI servers. And I think the question that people have is how do you counter that? What partners have you talked about? And can you stay on a 1-year cadence for all of those other components as well?

Yes. Yes. That's a great question. On the GPU side, we believe we have a very competitive GPU. And on the networking side, if you think about the interconnect AMD has always been -- have the legacy to work on the Infinity Fabric, which link our GPUs today, it has been a very successful interconnect technology. And recently, there are 8 companies, which include Microsoft, Meta, Google, AWS, Broadcom, AMD, and Cisco. 8 companies form the UALink, and we are going to create an open standard, 1.0 we will come out literally in Q3. So the industry, the ecosystem will have an open standard for interconnect, which can link up to 1,000 GPUs, that's really important, right, is to really scale up the interconnect side. And on networking side, as you recall, we actually bought a company called Pensando, which is a -- it has really programmable architecture DPU, not only we have a tremendous customer interest today for stand-alone DPUs, but it's also very important networking technology for GPU road map going forward. In addition, there's Ethernet consulting with Cisco Broadcom, Arista, Marvell and other companies, which the ecosystem is doing -- is to promote the Ethernet networking. If you look at it today globally, Ethernet is the most prevalent networking technology across all data center. So we do think by working with the ecosystem partners, we're going to have the networking technology, the interconnect technology, the DPUs to really continue to drive our road map going forward.

One of the other impediments for adoption has been in the software side. And I just wanted -- maybe you could talk about the progress that's being made in terms of open sort of software and having, obviously, if it's done in CUDA, it's a lot of manual labor to port, it's hard for the customer, and that's been an impediment. Where are we in that progression where a customer can use your silicon? And how long is that process at this point?

Yes. The ROCm 6.1, that's our most recent release of our software stack. It has library, compiler tools, we have been working with the ecosystem closely. If the model is writing based on PyTorch, Triton, and JAX, you actually can use MI300 out-of-box, that's why when you look at the Hugging Face, they probably have more than 700,000 models. All those models actually can run just on MI300, so that's the first one. Second one is, to your point, is if customers have been using CUDA and especially the kernel node level for CUDA we actually provide library tools for them to plot really efficiently. It's at a level right now for some customers, the porting work can be a day, can be a week, some may take longer if it's a more complex model, but it is so efficient, so it's largely porting is not a barrier anymore, and we'll continue to mature the ROCm software stack, library, tools and even more those to really help customers to convert very easily.

I want to ask you, NVIDIA mentioned on the last call that 40% of their business is inference. I know it might be tough to understand because they might do both, but in terms of your makeup of your business inference versus training, I don't know if you want to throw out a percentage or maybe I believe it's more inference weighted than kind of curious to your perspective as to why is your chip better for inference than -- and is there -- is it because you need to accomplish some things to get more of the training business? Or is it just because your chip is better at inference?

Yes. Yes. I think you're right. Today, if you look at our revenue, we are more indexed to the inference, I think there are several reasons. First, we are the second player to enter the market. Training absolutely has been going on for a while, right? People need to train the model first, then inference is where they make money. So initially, it's about the training of the model, and then inference is taking off. So when we enter the market, we actually see the inference applications demand continue to increase. Secondly, if you look at the MI300X, we do have 192 gigabyte HBM3 memory. So from that perspective, it is significantly higher memory capacity and bandwidth versus competition, for inferencing, that's really important, right, because you have a really large memory capacity, you can do the inference much faster. That's why the total cost of ownership for inference by using AMD's MI300 is much more, the TCO is really what the customer is focusing on. That's why we see customers like to use the MI300x for inference applications.

Wanted to ask you about supply. Last year, I think everybody figured out what CoWoS was and how their own models, and that was a big people following that. I'm curious, are you hearing more noise about HBM. So in just terms of I think, foundry capacity, I'm assuming you can get your chips done, but in terms of other components to ship, where is the tightness today? And is there any concern with HBM vendor supply.

Yes. I would say the industry continued to increase capacity, both on HBM side and on the CoWoS side. So it's absolutely the case. Our team has done a great job. But for the first half, we continue to see the tightness for both HBM and CoWoS. So capacity continue to be limited, but the team continued to work with the supply chain ecosystem, will continue to improve supply in the second half. I do think from a memory side, HBM side, we are working with all 3 memory suppliers and the capacity will continue to expand. That's the good news of the capacity side.

I did want to talk about the other 2 businesses, and we can go back to AI if we have time, but I wanted to ask on traditional servers because this year was thought of as a rebound year, when last year, it was thought of as like capacity plus wallet squeeze. I think to date, maybe the market is kind of only up modestly. So just your perspective on the traditional server market.

Yes, traditional server market is still quite mixed. Last year, we all know traditional server market actually declined. And if you look at this year, in the cloud environment, it's continued to see customers, some of them must continue to optimize, right? Is the AI investment optimization is still going on. But we have gained tremendous momentum with our Gen4 server CPU platforms. Both Genoa and Bergamo have been ramping quite significantly, adoption and the market share gain has been really, really impressive. If you look at the Q1, we actually got to 33% revenue market share. So when you look at the enterprise market, we actually started to see early signs for refreshing cycle, the way to look at it is when you look at the CIOs today, they are facing a lot of challenges. They are limited by power, by space. And also, they are trying to figure out how to adopt AI. So with all those things, TCO will become really important, when you look at the AMD solution, we actually can provide the same amount of compute with 40% less servers with our Gen4 family. What that means is you can cut CapEx by half at the very beginning and the operation cost to operate those servers will be also 40% less, so when you look at the whole TCO, we do think that will help the refresh cycle, right? Because then they will have more space, more power, they can utilize, adopt AI, do better planning within their data center.

It's a great lead, and I was going to ask you in terms of -- I mean, if you listen to NVIDIA, you would think that a GPU is going to do every workload ever over time. But clearly, there's a lot of workloads that a CPU is going to handle. And in many cases AI might even create more workloads for CPU. I think can you talk about just how old is this installed base of CPUs? And can they continue to just kind of ignore spending. And then you mentioned a better performance, is it looking at this market wrong on a unit perspective, you're getting much bigger super high core counts like Bergamo, is it potentially not a growth unit market, but then ASP is really the way to look at it.

Yes. Yes. I really appreciate the question. So first is, when you look at different workload, really fundamentally, there are so many different workloads. The data explosion continues, different workloads really need different compute engine. When you look at the traditional foundational applications, your ERP system, your database and your shopping website, your Meta, Facebook, Instagram, all those things, you don't need the GPU, the CPU has the best TCO for those kind of foundational traditional applications. And those things continue to increase. Generative AI is incremental. It's in addition to that foundational data and the foundational workload. So your question is spot on, when we look at the server CPU market, we actually continue to increase core counts per unit. So unit actually is not a good way to look at this market at all, because when you have 192 core counts with our next generation Turin, literally, you are addressing a lot of problems in the general compute side and the core count has been increasing double digit, both our competition and ourselves. So we are pushing the core counts, continue to be higher and higher. So the right way to look at it is actually core counts, unit is actually declining, but core count has been increasing. So from ASP perspective, we do think ASP will increase because core counts increase. In general, our view is, hey, this is a mature market. It's going to continue to grow. It may not be as high as generative AI, but it's a very healthy market, and we'll continue to gain share in this market.

I wanted to ask you the other announcement at COMPUTEX, AI PC, the first wave came out, they're ARM-based, you came out with a product that has maybe 10 more TOPs. And I think for a customer, it's going to be seamless from a software perspective. So just kind of your perspective, there have been some high numbers thrown out there for ARM-based PCs. I mean, kind of what's the selling point for those, is there a battery life argument they can make or I mean, because I think on the other side, they're not priced any cheaper. They don't run all the software. It seems like if AI PC is going to happen, it should be a good thing for you as well as Intel, but it's kind of painted somewhat as a negative.

It's interesting, right? ARM PC has been around, you and I, we have been in this industry for a long time. And the ARM PC, probably this is the second round, first round it did not go anywhere. We all paid a lot of attention. This is second round. I think fundamentally, if you look at AMD solution on the AI PC side, not only we have our latest generation CPU, GPU and also NPU performance with 50 TOPs, we actually know the whole PC ecosystem much better compared to ARM PC when you think about the ecosystem, especially commercial side, right, all the applications, everything has been really enterprise based on the X86 generation over generation. So that backward compatibility that the ecosystem is really important, and the performance is also very important. So I think from our perspective, we do think AI PC is going to really help with the PC refreshing cycle and our leadership product, the AI, we are going to be on shelf literally July, we do think that's going to have the leadership features and the capabilities that will help customers to adopt AI PC.

Curious if you're seeing what the particular drivers are? I mean I think on the enterprise side, there's Copilot. It seems like though no one is really making a distinction that it has to be run locally, right? So to the consumer or even the enterprise like they want -- they're going to have maybe 6, 7 years' worth of different models. Intel, AMD, there's a whole mix, and they may -- so to draw a hard line and say, no, it won't run, you can't run this certain Copilot version might be impractical. So if that's the case, I mean when we talk about an upgrade cycle, like one, what applications are you seeing that are interesting that might drive it? And then do you see eventually someone drawing a hard line saying, you need to have a 1-year-old CPU or newer, otherwise you can't do it.

Yes, great question. So it's very interesting if you think about AMD, we actually introduced the AI PC first in the market. So the Ryzen 7000 actually is an AI PC, but there were not many applications. So even though we have an AI PC, there are no applications from a customer perspective, either enterprise or consumer, the most important things are applications. The key thing is really all those applications can help enterprises to improve productivity and help consumers for content creation or for running their whatever family photos, videos to be helpful. That's why our view is we'll see second half of this year when this AI application come out. Then next year potentially is where you can see the AI PC adoption because only when we enterprise customers, we all if there are AI application, we can use offload on NPUs, which can improve our productivity, we're going to adopt it, right? So I do think it's important to align the applications and the ecosystem to make sure we can really pay premium to get AI PC, but get more productivity from that.

Curious if you think of just list of the business in terms of share, the PC share gains kind of moderated out. I think you're still gaining share in servers. I think part of the part of the PC market is the penetration on the enterprise side. I think Intel just has a very dominant franchise. It's hard to kind of crack that. So I'm kind of curious, are people thinking about it wrong if I think that your PC share gains are going to remain at whatever 20%. And I think it fits into the kind of like monetizing AI too. Can you get enterprise share, that will be a benefit.

Yes. Maybe if we take a step back, if you look at the enterprise side, enterprise really requires different go-to-market. The commercial, each enterprise, their CIO, how they buy PC, how they buy servers, it's very different from a consumer, from a hyperscale data center. So AMD has made a tremendous effort during the last 2 years, investing in go-to-market side. We hired our new Chief Sales Officer from IBM. One of the objective is absolutely focused on enterprise go-to-market approach, not only have more feet on the street, but also understand how to approach the enterprise customer. So the success we have been seeing is on the server side. First actually is we are able to show CIOs the total cost of ownership benefit, so they can convert to AMD servers. And the same thing on the PC side, you literally have to convince enterprise CIOs to change in order to expand your market share. So that takes a longer time. But with the go-to-market approach we have and the capabilities and the leadership of our product portfolio, we do think we'll continue to make progress, just like how we are gaining share on the server side, we do think that on the PC side, the commercial side, we'll continue to gain share, make the progress. Go-to-market is very, very important there.

So I've asked you all these strategy questions. You did a fantastic job. I'll ask you a CFO question. I actually want to know if you think about next year, there are some big moving pieces. Who knows what the AI number is going to be. Obviously, AI PC might be additive to gross margin, data center stuff you tell me might be impressive to gross margin. Kind of how do you think about those moving pieces? Obviously, we don't have the magnitudes of each, but how do we think about your gross margin where it is today and where it could be depending on those swings.

Yes. So if you look at our gross margin, we have been making progress. In 2023, the company's gross margin is 50% and in Q1, we actually ended up improving gross margin to 52.3%, and we actually guided Q2 to 53%. And so one of the key drivers of the gross margin increase actually is because of data center mix. Data center has been growing much faster than our other business despite a headwind on the embedded Xilinx business side, we're able to improve gross margin. Second half, we think will have the same dynamics starting embedded business are going to recover, but gradually. And the major driver for gross margin improvement continue to be data center. I think next year, it will be the similar picture. It's because the mix change is going to be more favorable. Data center in general has higher than corporate average gross margin. And then hopefully, when the embedded business stabilized and come back, that will be the tailwind to continue to help us to improve gross margin. I think one thing, as you know, is our gaming business has lower than corporate average gross margin and the gaming cycle is at the fifth year going to 6, 7 year. So gaming business is going to be more muted, which -- it's not good, but it actually helps the gross margin mix.

Okay. With that, we're out of time. Thank you, Jean.

Okay. Thank you so much, everyone. Thank you.