Q4 2023 Advanced Micro Devices Inc Earnings Call

January 30, 2024

Operator: Time, http://TheBusinessProfessor.com Thank you, John, and welcome to AMD's fourth quarter and full year 2023 Financial Results Conference call. By now, you should have had the opportunity to review a copy of our earnings press release and the accompanying slides. If you have not had the chance to review these materials, they can be found on the investor relations page of AMD.com. We will refer primarily to non-GAAP financial measures during today's call. The full non-GAAP to GAAP reconciliations are available in today's press release and the slides posted on our website.

Operator: Participants on today's call are Dr. Lisa Su, our Chair and Chief Executive Officer, and Jean Hu, our Executive Vice President, Chief Financial Officer, and Treasurer. This is a live call that will be replayed via webcast on our website. Before we begin, I would like to note that Mark Papermaster, Executive Vice President and Chief Technology Officer, will attend the Bernstein Tech, Media, Telecom, and Consumer One-on-One Forum on Tuesday, February 28th, and Jean Hu, Executive Vice President, Chief Financial Officer, and Treasurer, will attend the Wolf Research Semiconductor Conference on Tuesday, February 15th, and the Morgan Stanley Technology, Media, and Telecom Conference on March 5th. Today Speak only as of today and as such involve risks and uncertainties that could cause actual results to differ materially from our current expectations. Please refer to the cautionary statement in our press release for more information on factors that could cause actual results to differ materially. With that, I'll hand the call over to Lisa.

Lisa T. Su: Thank you, Mitch, and good afternoon to all those listening in today. We finished 2023 strong as data center sales accelerated significantly throughout the year despite the mixed demand environment. As a result, we delivered record data center segment annual revenue and strong top-line and bottom-line growth in the fourth quarter, driven by the ramp of instinct AI accelerators and robust demand for Epic Server CPUs across cloud, enterprise, and AI customers. Looking at our financial results, fourth quarter revenue increased 10% year over year to 6.2 billion, driven by significant double-digit percentage growth in our data center and client segments. On a full year basis, annual revenue declined 4% to $22.7 billion as record data center and embedded segment annual revenue was offset by lower client and gaming segment revenue.

Lisa T. Su: Importantly, data center and embedded segment annual revenue grew by $1.2 billion and accounted for more than 50% of revenue in 2023 as we gained server share, launched our next generation Instinct AI accelerators, and maintained our position as the industry's largest provider of adaptive computing solutions. Turning to the fourth quarter business results, data center segment revenue grew 38% year-over-year and 43% sequentially to a record $2.3 billion. Server CPU and data center GPU sales both set quarterly and annual revenue records as sales of our data center products accelerated throughout the year. We gained server CPU revenue share in the quarter driven by significant double-digit percentage growth in fourth-gen EPYC processor revenue and demand for our third-gen EPYC processor portfolios. In the cloud, while the overall demand environment remains soft, server CPU revenue increased year over year and sequentially as North American hyperscalers expanded fourth-gen EPYC processor deployments to power their internal workloads and public instances.

Lisa T. Su: Amazon, Alibaba, Google, Microsoft, and Oracle brought more than 55 AMD-powered AI, HPC, and general-purpose cloud instances into preview or general availability in the fourth quarter. By 2023, there will be more than 800 Epic CPU-based public cloud instances available. We expect this number to grow in 2024 based on the leadership performance, efficiency, and features of our EPIC CPU portfolios. In enterprise, sales accelerated by a significant double-digit percentage in the quarter as we built momentum with Forbes 2000 customers. We closed multiple wins with large financial, energy, automotive, retail, technology, and pharmaceutical companies, positioning us well for continued growth based on expanded production deployments planned for 2024. A growing number of customers are adopting EPYC CPUs for inferencing workloads, where our leadership throughput performance delivers significant advantages on smaller models like LLAMA7B, as well as to power head nodes in large training and inference clusters. Looking ahead, customer excitement for our upcoming TURN family of EPYC processors is very strong. Turin is a drop-in replacement for existing fourth-gen EPYC platforms that extends our performance, efficiency, and TCO leadership with the addition of our next-gen Zen5 Core, new memory expansion capabilities, and higher core counts.

Lisa T. Su: Internal and end customer validation work is progressing to plan, with Turin on track to deliver overall performance leadership, as well as leadership on a per core or per watt basis across a wide range of workloads when it launches later this year. Turning to our broader data center portfolio, our data center GPU business accelerated significantly in the quarter, with revenue exceeding our $400 million expectation, driven by a faster ramp for MI300X with AI customers. We launched our MI300 Accelerator family in December with strong partner and ecosystem support from multiple large cloud providers, all the major OEMs, and many leading AI developers. MI300X GPUs deliver leadership-generative AI performance by combining our high-performance cDNA3 architecture with industry-leading memory bandwidth and capacity.

Lisa T. Su: Customer response to MI300 has been overwhelmingly positive, and we are aggressively ramping production to support the dozens of cloud, enterprise, and supercomputing customers deploying Instinct Accelerator. For cloud, we are working closely with Microsoft, Oracle, Meta, and other large cloud customers on Instinct GPU deployments, powering both their internal AI workloads and external offerings. For enterprise customers, HPE, Dell, Lenovo, Supermicro, and other server vendors are on track to launch differentiated MI300 platforms later this quarter with strong demand from multiple enterprise customers. For HPC supercomputing, we shipped the majority of AMD Instinct MI300A accelerators for the El Capitan supercomputer in the fourth quarter and expect to complete shipments this quarter for what is expected to be the world's fastest supercomputer when it comes online later this

Lisa T. Su: We also closed new Instinct GPU wins in the quarter, including the flagship system at the German High Performance Computing Center, HLRS, as well as what is expected to be one of the world's most powerful enterprise supercomputers for energy company, ENI. In AI software development, we made significant progress expanding the ecosystem of AI developers working on AMD platforms with the release of our RockM6 software suite. The RockM6 stack significantly increases performance in key generative AI workloads, adds expanded support and optimizations for additional frameworks and libraries, and simplifies the overall developer experience. The additional functionality and optimizations of Rackham 6 and the growing volume of contributions from the open source AI software community are enabling multiple large hyperscale and enterprise customers to rapidly bring up their most advanced large language models on AMD Instinct Accelerator.

Lisa T. Su: For example, we are very pleased to see how quickly Microsoft was able to bring up GPT-4 on MI300X in their production environment and roll out Azure private previews of new MI300 instances aligned with the MI300X launch. At the same time, our partnership with HuggingFace, the leading open platform for the AI community, now enables hundreds of thousands of AI models to run out of the box on AMD GPUs, and we are extending that collaboration to our other platforms. Looking ahead, our prior guidance was for data center GPU revenue to be flattish from Q4 to Q1 and exceed $2 billion for 2024. Based on the strong customer pull and expanded engagements, we now expect data center GPU revenue to grow sequentially in the first quarter and exceed $3.5 billion in 2024.

Lisa T. Su: We have also made significant progress with our supply chain partners and have secured additional capacity to support upside demand. Turning to our client segment, revenue was $1.5 billion, an increase of 62% year over year and flat sequentially. We launched our latest generation Ryzen 8000 series notebooks and desktop processors in January, including our Ryzen 8040 mobile series that combine leadership compute performance and energy efficiency with an updated MPU that delivers up to 60% more AI performance compared to our prior generation that was already industry leading. Acer, Asus, HP, Lenovo, MSI, and other large PC OEMs will all offer notebooks powered by our Ryzen 8000 series processors, with the first systems expected to go on sale in February.

Lisa T. Su: To further our leadership in AI PCs, we launched our Ryzen 8000G series processors earlier this month, which are the industry's first desktop CPUs with an integrated AI engine. Millions of AI PCs powered by Ryzen processors have shipped to date, and Ryzen CPUs power more than 90% of AI-enabled PCs currently in the market. Our work with Microsoft and our PC ecosystem partners to enable the next generation of AI PCs expanded significantly in the quarter. We are aggressively driving our Ryzen AI CPU roadmap to extend our AI leadership, including our next-gen Strix processors that are expected to deliver more than 3x the AI performance of our Ryzen 7040 series processors. Strix combines our next-gen Zen5 core with enhanced RDNA graphics and an updated Ryzen AI engine to significantly increase the performance, energy efficiency, and AI capabilities of PCs.

Lisa T. Su: Customer momentum for Strix is strong, with the first notebooks on track to launch later this year. Looking at 2024, we are planning for the PC-TAM to grow modestly year on year, weighted towards the second half as AIPCs ramp up. We continue to see strong growth opportunities for our client business as we ramp up our current products, extend our AIPC leadership, and launch our next wave of Zen5 CPUs. Now turning to our gaming segment, revenue declined 17% year-over-year and 9% sequentially to $1.4 billion as lower semi-custom revenue was partially offset by increased sales of Radeon GPUs. Semi-custom SOC sales declined in line with our projections in the quarter.

Lisa T. Su: Going forward, we now expect annual revenue to decline by a significant double-digit percentage year-over-year, as supply catches up with demand in 2023, and we enter the fifth year of what has been a very strong console cycle. In gaming graphics, revenue grew both year over year and sequentially, driven by strong demand in the channel for both our Radeon 6000 and Radeon 7000 series GPUs. We expanded our Radeon 7000 GPU series with the launch of new RX 7600 XT series enthusiast desktop GPUs earlier this month that offer leadership price performance for 1080p gaming. We also launched new open source FidelityFX Super Resolution 3 software that can deliver significantly higher gaming frame rates on both GPUs and APUs.

Lisa T. Su: Turning to our embedded segment, revenue decreased 24% year-over-year and 15% sequentially to $1.1 billion as customers focused on reducing their inventory levels. We expanded our embedded portfolio in the quarter with new leadership solutions for key markets. We launched new Versal Prime Adaptive SoCs for the aerospace, test and measurement, healthcare, and communications markets that deliver industry-first support for DDR5 memory and increased DSP capability compared to our prior generation. In automotive, we launched new Versal SoC solutions that bring industry-leading AI compute capabilities and advanced safety and security features to next-generation vehicles.

Lisa T. Su: We also launched Ryzen Embedded Processors with unmatched performance and features for industrial automation, machine vision, robotics, and edge server applications. Looking ahead to 2024, we expect overall embedded demand will remain soft through the first half of the year, as customers continue to focus on normalizing their inventory levels. Longer term, we're very confident in the growth trajectory of our embedded business as our expanded product portfolio drove more than $10 billion of design wins in 2023, an increase of more than 25% compared to 2022. In summary, I'm very pleased with our fourth quarter and full year results.

Lisa T. Su: For 2024, we expect the demand environment to remain mixed, with strong growth in our data center and client segments offset by declines in our embedded and gaming segments. Against this backdrop, we believe we will deliver strong annual revenue growth and expand gross margins driven by the strength of our Instinct, Epic, and Ryzen product portfolios. Taking a step back, we believe AI is a once-in-a-generation transition that will reshape virtually every portion of the computing market, starting in the data center and then expanding into PCs and across multiple embedded markets.

Lisa T. Su: We have built excellent customer traction based on the strength of our multi-year AI hardware and software roadmaps, and we see clear opportunities to drive our next wave of growth as we deliver leading AI solutions across our portfolio. In the data center, we see 2024 as the start of a multi-year AI adoption cycle, with the market for data center AI accelerators growing to approximately $400 billion in 2027. Customer deployments of our Instinct GPUs continue to accelerate, with MI300 now tracking to be the fastest revenue ramp of any product in our history and positioning us well to capture significant share over the coming years based on the strength of our multi-generation Instinct GPU roadmap and open-source RockM software strategy. In PCs, we're focused on delivering our long-term roadmaps with leadership Ryzen AI NPU capabilities to enable differentiated experiences as Microsoft and our other software partners bring new AI capabilities to PCs starting later this year.

Jean Hu: At the same time, we are rapidly driving leadership AI compute capabilities across the full breadth of our embedded product portfolio. This is an incredibly exciting time for the industry, and an even more exciting time for AMD, as our leadership IP, broad product portfolio, and deep customer relationships position us well to deliver significant revenue growth and earnings expansion over the next several years. Now I'd like to turn the call over to Jean to provide some additional color on our fourth quarter and full year financial results. Thank you, Lisa, and good afternoon, everyone.

Jean Hu: I'll start with a review of our financial results and then provide our current outlook for the first quarter of fiscal 2024. AMD executed well in 2023 despite a mixed market demand environment, delivering revenue of $22.7 billion and earnings per share of $2.65. We drove year-over-year revenue growth in our embedded and data center segments. In addition, we successfully launched our AMD Instinct MI300 GPUs, positioning us for a strong ramp in 2024 in the AI market. For the fourth quarter of 2023, revenue was $6.2 billion, growing 10% year-over-year as revenue growth in the data center and the client segments was partially offset by lower revenue in our embedded and the gaming segments. Revenue was up 6% sequentially primarily driven by the ramp of AMD Instinct GPUs across several leading customers and higher revenue from EPYC server processing, partially offset by the decline in embedded and gaming segment revenue.

Jean Hu: Growth margin was 51% flat year-over-year, with higher revenue contribution from the data center and the client segment offset by lower embedded segment revenue. Operating expenses were $1.7 billion, an increase of 8% year-over-year, as we invest in R&D and marketing activities to support our significant AI growth opportunities. Operating income was $1.4 billion, representing a 23% operating margin. Taxes, Interest Expense, and others were $163 million.

Jean Hu: For the fourth quarter of 2023, diluted earnings per share was $0.77, an increase of 12% year-over-year. Now, turning to our reportable segment. Starting with the data center segment, revenue was $2.3 billion, up 38% year-over-year and 43% sequentially, driven by strong growth of both AMD Instinct GPU and 4th generation AMD IPEX CPU sales. Data center segment operating income was $666 million, or 29% of revenue compared to $444 million or 27% a year ago. Higher operating income was primarily due to operating leverage driven by higher revenue. Client segment revenue was $1.5 billion, up 62% year-over-year, driven by Ryzen 7000 series CPU sales. Kalyan segment operating income was $55 million, or 4% of revenue, compared to an operating loss of $152 million a year ago, driven by higher revenue.

Jean Hu: Gaming segment revenue was $1.4 billion, down 17% year-over-year and 9% sequentially due to a decrease in semi-customer revenue partially offset by an increase in Radeon GPU sales. Gaming segment operating income was $224 million, or 16% of revenue compared to $266 million or 16% a year ago. Embedded segment revenue was $1.1 billion, down 24% year-over-year and 15% sequentially as customers continue to work down their inventory levels. Embedded segment operating income was $461 million, or 44% of revenue compared to $699 million or 50% a year ago. Turning to the balance sheet and cash flow, during the quarter, we generated $381 million in cash from operations, and free cash flow was $242 million. Inventory decreased sequentially by 94 million to 4.4 billion.

Jean Hu: At the end of the quarter, cash, cash equivalents, and short-term investments were strong at $5.8 billion. In the fourth quarter, we repurchased 2 million shares and returned $233 million to shareholders. For the year, we repurchased 10 million shares and returned $985 million to shareholders.

Jean Hu: We have $5.6 billion in remunerative repurchase authorization. Now turning to our first quarter of 2024 Outlook, we expect revenue to be approximately $5.4 billion, plus or minus $300 million. Sequentially, we expect data center segment revenue to be flat, with a seasonal decline in server sales offset by strong data center GPU RAM. Embedded revenue will decline as customers continue to work down their inventory levels.

Jean Hu: Client segment revenue to decline seasonally, and in the gaming segment, as we enter the fifth year of what has been a very strong gaming cycle, and given current customer inventory levels, we expect revenue to decline by a significant double-digit percentage. However, year over year, we expect data center and client segment revenues to increase by a strong double-digit percentage given the strength of our product portfolio and share gain opportunities. Invest segment to decline and the gaming segment revenue to decline by a significant double-digit percentage. In addition, we expect the first quarter non-GAAP growth margin to be approximately 52%. Non-GAAP operating expenses are expected to be approximately $1.73 billion. The non-GAAP effective tax rate is expected to be 13%, and the diluted share count is expected to be approximately 1.63 billion shares.

Jean Hu: While we are not providing specific full-year guidance for 2024, let me provide some color. Directionally for the year, we expect 2024 data center and client segment revenue to increase, driven by the strength of our product portfolio and share gain opportunities. Embedded Segment Revenue to Decline and Gaming Segment Revenue to Decline by a Significant Double-Digit Percentage. We expect to expand the growth margin in 2024 and continue to invest to address the large AI opportunities while driving operating model leverage to deliver earnings per share growth faster than top-line revenue growth. In closing, we delivered solid financial results in 2023, further strengthening our product portfolio and establishing ourselves as a leading provider of data-centered We are very well positioned to build on this momentum and deliver strong financial performance in 2024 and beyond. With that, I'll turn it back to Mitch for the Q&A session. Thank you, Jean. John?

Operator: We're happy to poll the audience for questions. The next question comes from the line of Aaron Rakers from, Yeah, thanks for taking the question. I'm just kind of framing the outlook and the guidance for this calendar first quarter. I guess the first question is, you know, can you help us on a relative basis with the 400 million dollars of data center GP revenue that you expected in Q4, what that ultimately kind of fell out to be? And then on the guidance into 1Q, can you help us appreciate what seasonality, you know, is defined as as we think about the server business in the 1Q guide? Sure, Aaron, let me start and then see if Jean has something to add.

Lisa T. Su: So, you know, relative to the data center GPU business, look, we were very pleased with the performance that we saw in the fourth quarter. You know, it was always going to be a very sort of back-end quarter weighted as we were, you know, ramping the product, and we saw MI 300A, our HPC product, actually ramped very well. And then we saw MI 300X, the AI product, actually exceed our expectations based on strong customer demand, the way the qualifications went, and then the manufacturing ramp. So we were over 400 million for that business in the fourth quarter.

Lisa T. Su: And then going into the first quarter, you know, as we look at the business, you know, server seasonality, call it something around, let's call it high single digit, low double digit. There are also some other pieces of the data center business. I think the key piece of it is, We had originally expected the ramp to be a little bit more shallow for our MI300X, and what we're seeing now is, you know, the supply chain is operating really well, and the customer demand is strong, and so we will see MI300X increases as we go into the first quarter, and, you know, things are going relatively well. Yeah, Aaron, I'll give you some color.

Jean Hu: Aaron, I'll give you some color about client seasonality and others. So client is very similar to server. Typically, Q1 is high single digit to low double digit.

Jean Hu: That's consistent with the past. On the embedded side, it's very consistent with what we said in the past and consistent with what you see in the industry. The embedded business is going through a bottoming process, and we think Q1 will have a low double-digit sequential decline. That's embedded.

Jean Hu: On the gaming side, Lisa mentioned during her prepared remarks, we have the late stage of the product cycle in the year five of gaming consoles. But at the same time, we also have inventory at customers. So the combination of those impacts, we expect the Q1 gaming sequential decline to be probably more than 30%. So hopefully, that'll help you a little bit. Yeah, very helpful, Jean.

Lisa T. Su: And as a quick follow-up, I'm just curious about the traditional server demand that you see. When we look at server CPU shipments, you know, shipments are down, you know, north of 20% year over year. Are you seeing any signals? Or how are you thinking about a recovery in that traditional, call it non-AI general purpose server market as we move through 20? Sure, Aaron. So, look, I think, you know, I agree with your characterization of 2023 demand, although we did see, you know, some strong progress in the second half of the year, especially as customers in cloud and enterprise adopted our Genoa and our Zen4 family. So going into 2024, I would say the traditional server market is probably still mixed, especially in the first half of the year.

Lisa T. Su: There's still some cloud optimization going on, as well as, you know, sort of enterprises being a little bit cautious. That being the case, though, we also see opportunities for us to continue to grow our share in the traditional server business. I think our portfolio is extremely strong. The adoption of, you know, Genoa and Bergamo, as well as our new Sienna product lines, are getting bigger, And we also see Turin, our Zen5 product, coming in the second half of the year. So even in a mixed demand environment, I think we're bullish on what traditional server CPUs can do in 2024 and the next few years. Thanks a lot.

Lisa T. Su: Um, Lisa, I'm wondering if you can give us a little bit of sense in terms of the milepost that you're kind of, you know, marching toward on this $400 billion tan that you have for, you know, 2027. For example, do you think you can gain share at a rate that's kind of similar to the rate that you gain share for server CPUs, or maybe asked a different way?

Lisa T. Su: Is it reasonable to kind of look at your consumer GPU share of 20 plus percent? Is that a reasonable number? Or do you have aspirations higher than that? Yeah, thanks Tim for the question. You know, I would say a couple things. First of all, we're really pleased with the progress that we've made in our data center GPU business. I think the ramp that we've seen, the customer traction that we've seen, even in the last few months, has been great, and that gives us a lot of confidence in the growth of this business. I think the beauty of the AI market here is that it's growing so quickly that I think we have both the market dynamics as well as our ability to gain share in that framework.

Lisa T. Su: The point I would make is, you know, our customer engagements right now are all quite strategic, dozens of customers with multi-generational conversations. So, as excited as we are about the ramp of MI300, and frankly, there's a lot to do in 2024, we're also very excited about the opportunities to extend that into the next couple of years, you know, out into that, you know, 2025, 2026, 2027 timeframe So I think we're seeing a lot of growth. I think it's a little early to make market share projections, but I would say it's a significant growth driver given market demand as well as, you know, our own product capabilities. Thanks a lot.

Jean Hu: Jean, I guess as a follow-up, I know that you don't want to guide the full year, but I'm wondering if I can pin you down just to touch on maybe a, maybe a, you know, milepost that you're kind of marching to for 2024 growth is up 20% for the whole company. Is that a reasonable target? And then I guess within the data center, if you just add the incremental data center GPU revenue and you assume that the server business grows a little bit, it seems like that should maybe double year over year. But I'm kind of wondering if you can give us any ranges on those. Hi Tim. Thank you for the question. Yeah, we're not guiding for a year. It's the very early of the year, literally, you know, January.

Jean Hu: I think the way to think about it is, you know, Lisa mentioned during her prepared remarks that we feel pretty good about both our data center and client businesses growing in 2024. Of course, the largest incremental revenue opportunities are going to come from data center, between both the server side, getting more share, and the data center GPU side with a significant ramp of our MI300. I think that's how we should think about it.

Jean Hu: We do have a headwind from the gaming segment. We do think, year over year, we'll see a very significant double-digit decline in the gaming segment. And at the same time, embedded is going through the bottoming process. We do think in the second half, we'll see a recovery.

Operator: So those are the puts and the takes I can talk about. The next question comes from the line of Matt. Thank you very much. I wanted to ask, I mean, there's been so much.

Operator: Thank you all for tuning in. I'll see you next time, http://TheBusinessProfessor.com has progressed over the last six months. Software, and that you're rampant. What I'm interested in hearing a little bit more about, and you guys have been open about, some of the forward programs. Matt's perspective, how you're thinking about the roadmap. M.I.

Lisa T. Su: Accelerator. Parts that are, https://www.youtube.com.au The Bulletproof Executive 2013, www.tasktel.com Forward Roadmap, Yeah, sure, Matt. So I appreciate the comments. I think the, you know, the traction that we're getting with the MI 300 family is really strong. I think what's benefited us here is our use of chiplet technologies, which has given us the ability to have, you know, sort of both the APU version and the GPU version. And, you know, we continue to use that to differentiate ourselves.

Lisa T. Su: And that's how we get our memory bandwidth and memory capacity advantages. As we go forward, you can imagine, like we did in the EPIC timeframe, we plan multiple generations in sequence. That's the way we're planning the roadmap. You know, one of the things I will note about the AI accelerator market is the demand for compute is so high that we are seeing, you know, sort of an acceleration of the roadmap, you know, generations here, and we are similarly planning for an acceleration of our roadmap. I would say that, you know, we'll talk more about the overall roadmap beyond MI 300 as we get into later this year, but you can be assured that we're working very closely with our customers to have a very competitive roadmap for both training and inference that will come out over the next couple of years for that one. This is a follow-up. That's one of the questions that I've...

Lisa T. Su: Thanks for watching! The Bulletproof Executive, 2013. Check it out. If you could give us a little bit of your time today, we would love to hear from you, look under the hood as I guess. I've got a hundred versions of the same question, which is how the heck did you come up with this, Um, so if you could give us a little bit more, what are we talking about, and Accelerator Cards are we talking about? Silicon, Servers, and what kind of Unit Assumptions, any kind of thing that you can give us on market size, https://www.youtube.com.uk/watch?v=9ZZWjk. Amazing! Precise, that would, Sure.

Lisa T. Su: Well, Matt, I don't know how precise it is, but I think we said approximately $400 billion. But I think what we need to look at is the growth rate and how we get to those growth rates. I think we expect, you know, units to grow, you know, sort of substantial double-digit percentage, but you should also expect that content is going to grow. So if you think about, you know, how important memory and memory capacity are as we go forward, you can imagine that, you know, we'll see, you know, an acceleration there in just the overall content as we go to, you know, more advanced technology nodes. So there's some ASP uplift in there.

Lisa T. Su: And then what we also do is, you know, we're planning longer-term roadmaps with our customers in terms of how they're thinking about, you know, sort of the size of training clusters, the number of training clusters, and then the fact that, you know, we believe inference is actually going to exceed training as we go into the next couple of years, just given, you know, as more enterprises adopt it. So, as we look at all those pieces, I think we feel good that the growth rate is going to be, you know, significant and sustained over the next few years. In terms of what's in that, Sam? It really is accelerator time.

Lisa T. Su: So within accelerators, there are, you know, certainly GPUs, and there will also be some ASICs that are or other accelerators that are, at that time, as we think about sort of the different types of models from, you know, smaller models to fine tuning of models to, you know, very, the largest, large language models, I think you're going to need different silicon for those, you know, different use cases. But from our standpoint, GPUs are still going to be the, you know, sort of the compute element of choice when you're talking about training and inferencing on the largest language models. Thanks for watching! See you next time.

Lisa T. Su: Great, thank you. I think you talked about the mi 300 cloud workloads being kind of split between the more customer-facing workloads and versus internal. Can you talk about how you see the breakdown of that? And how, you know, how is your ecosystem progressing? This is a brand new chip. It seems impressive that you're able to support kind of a broad range of customer-facing workloads in the cloud. Yeah, sure, Joe.

Lisa T. Su: So yes, look, we are really happy with how the MI300 has come out. And, you know, we've now deployed and, you know, working with a number of customers, what we have seen is certainly Rokam 6 has been very important, as well as the direct optimization with our top cloud customers. We always said that, you know, the best way of optimizing the software is to work directly on the most important workloads.

Lisa T. Su: And, you know, we've seen performance come up nicely, which is what we expect, frankly, with, you know, the GPU capabilities that we would have to do some level of optimization, but that optimization has gone well. I think to your broader question, you know, the way I look at this is there are lots of opportunities for us to work directly with large customers, both on the cloud side, as well as on the enterprise side, who have specific training and inferencing workloads. You know, our job is to make it as easy as possible for them. And so, our entire tool chain, you know, all of, you know, our, you know, sort of the overall, you know, Rokam suite has really gone through, you know, significant progress. Over the last six to nine months.

Lisa T. Su: And then we're also getting some nice support from the open source community. So the work that Hugging Face is doing is tremendous in terms of just real-time optimization on our hardware, our partnership with OpenAI on Triton, and our work across a number of these open source models has helped us actually make very rapid progress.

Lisa T. Su: And for my follow-up, I guess, you know, a lot of the forecasting of your business that I'm hearing is coming from the supply chain. And we're sort of hearing AMD is building X in Asia. I guess, what would you ask us to think about that?

Lisa T. Su: You know, are you looking at being kind of sold out for the year? And so the supply chain would be close to revenue? Are you building for the best case scenario? Just, you know, I worry about sometimes expectations when people hear these supply chain numbers, and I'm just curious how you bridge the gap. Yeah, so, Joe, I think we updated our revenue expectations, you know, this quarter from, you know, our original number of two billion to three and a half billion to try to give, you know, some boundaries on some of the discussion out there. The way to think about the three and a half billion is that these are customers that we're working with who have, you know, given us firm commitments on what they need. As you know, the lead times on these products are quite long, so it's important, you know, to have those forecasts in early, and we have a strong order book.

Lisa T. Su: So that gives us, you know, good confidence to exceed the three and a half billion mark. From a supply chain standpoint, our goal is always to build more supply. From that standpoint, you know, we have also worked with our supply chain partners and secured significant capacity. You know, think about it as first half capacity is tight, and more capacity comes on in the second half of the year. But, you know, we've certainly made more progress there.

Lisa T. Su: So, you know, we do have more supply, and we're going to continue to work with our customers on their deployments, and we'll update that number as we go through the year. Next question comes from the line: Toshiya Hari with Goldman Sachs. Hi, thank you for taking the question. I had one on the MI300 as well, Lisa.

Lisa T. Su: I guess, first of all, how should we think about the quarterly trajectory beyond Q1? You talked about Q1 being up sequentially. Is it fair to assume kind of a straight line as we progress through the balance of the year, or is it more second-half skewed? How should we think about that? And I guess more importantly, some of the potential cloud customers that have yet to officially sign up for or sign off on the MI300. I guess what's the sticking point? Is it just a function of time, and you just need a little bit more time to go back and forth and tweak things? Or is there a software kind of concern?

Lisa T. Su: I guess what's holding them back at this point? Yeah, Toshiya, thanks for the question. So first, on the MI300 trajectory, I think you would expect that revenue should increase every quarter from now through sort of the end of the year. But it will be a bit more second-half weighted, and part of that is just customers as they're finishing up their qualifications on their lines, as well as sort of how our supply chain is ramping up. So yes, it should increase each quarter but be a bit more second-half weighted. And then to your comment about customers, look, we're engaged with all of sort of the large customers. These are all folks that know us really well, given our deep relationships in Epic.

Lisa T. Su: I think people just have different adoption cycles as they consider what they're trying to do in their roadmaps. But I view this as still being in the very, very early innings for us in this space. And I think a question was asked earlier.

Lisa T. Su: I think the key is this is not just about an MI300 conversation, but it really is about sort of our long-term multigenerational roadmap. And so that's the context in which we're working with our largest customers, as well as, as you know, there's a lot of demand coming from folks that are more AI-centric and not necessarily typical cloud customers but more enterprise or, let's call it, AI-specific companies that we're also very well engaged with. I got it. That's super helpful.

Jean Hu: And then as my follow-up, maybe one for Jean on the gross margin side, you're guiding Q1 to 52%. I'm curious, again, I'm sure you're not going to give, you know, quantitative guidance beyond Q1, but how to think about the trajectory for Q2 and beyond. I'm pretty sure you're working through some kinks as it pertains to the instinct ramp.

Jean Hu: Hopefully, that will improve over time. So that should be a tailwind. FPGAs, perhaps, the same. You know, it's turning for the better. And you've got server CPU volume growth throughout the year. So it feels like you've got multiple tailwinds as we think about gross margin progression on a sequential basis. But what are the potential headwinds as we move throughout 2024? Thank you. Yeah, Toshiya, thank you for the question. Yeah, you're absolutely right.

Jean Hu: We have some puts and takes that impact our gross margin. We guided the Q1 120 basis points higher than Q4 sequentially, primarily because the higher data center contribution actually more than offset the decline of embedded business in Q1. Going forward, the way to think about it is, as you said, the major driver is going to be data center businesses which are going to grow much faster than other segments. That mixed change will help us to expand the gross margin nicely. I think you also are spot on with embedded coming back in the second half, which will be a tailwind with the data center GPU. We are at a very early stage of ramp. We are improving testing time yield and continuing to expand gross margin, and we expect to be creative compared to corporate averages. So those are all the tailwinds coming in the second half.

Jean Hu: I would say the headwinds inside continue to be in the first half, where we see embedded business, not only in Q1, we see sequential decline. Q2 probably is going to be sequentially flattish versus Q1. That is a headwind for us because it does have a very nice gross margin. But overall, we feel pretty good about the trajectory of the gross margin improvement, especially in the second half. This next question comes from the line of... Thanks for letting me ask you a question. I wanted to get into the competitive environment first on the instinct side of things, how that's going, it doesn't seem to be slowing down your ramp whatsoever. But then also on the streets.

Lisa T. Su: On the CPU side of things. Lisa, you said you're gaining share in that area. But as we think about future roadmaps, pricing incentives, those sorts of things, any meaningful change in the competitive environment that you're seeing throughout 2024? Well, look, I think the environment for us is always competitive, so I think that has not changed.

Lisa T. Su: If I look at the instinct side, I think we've shown that MI300 and our roadmap are actually very competitive. There are some places where, let's call it, it's more even, like in the training environment, but as we look at the inferencing environment, we think we have significant advantages, and a lot of that shows through in some of our customer work. So we think for both training and inference, we will continue to be very competitive. And then, as you go into the CPU side, again, from our point of view, with each generation of Epic, we've continued to gain share. I think we exited the fourth quarter at a record share for AMD, and we're still quite underrepresented in enterprise. So I think there's an opportunity for us to continue to gain share as we go through 2024.

Lisa T. Su: From a competitive standpoint, what we see is Zen 4 is extremely competitive right now with Genoa, Bergamo, and Sienna. And as we go into Turin, we're deep in the design cycle for Zen 5 in Turin, and we feel very good about how we're positioned. Thanks for that. I guess as my follow-up, on the data center side, another theme that's been pervasive throughout 2023, at least on the GPU side, crowding out the CPU side, you mentioned that there was still a little bit of cloud digestion going on within your Epic business. But where do you see that standing?

Lisa T. Su: I know you're going to gain share, etc., but you guys clearly benefit from the Instinct side of the data center GPU side. But what about on the CPU side of things? Is that headwind now behind us? Or is it not?

Lisa T. Su: I think we expect the CPU business, from a market standpoint, to grow, Ross, as we go into 2024. I think the rate and pace of growth will depend a little bit on the macro and just, you know, overall CapEx, you know, trends. But, you know, from our standpoint, we are starting to see some of our larger customers plan their refresh cycles; there's a lot of what's called older equipment that has yet to be, you know, refreshed. And, you know, the value proposition for refresh is so strong, because, you know, the energy efficiency and sort of the footprint of the newer generations are so much better than sort of the four or five-year-old infrastructure, that we do I think the exact timing we'll have to, you know, understand more as the market evolves as we go through the year. Vivek Arya, Thanks for taking my questions.

Lisa T. Su: So first one, Lisa, you gave us a $2 billion plus number for MI before; now you've raised it to over $3.5 billion. And I'm curious, what drove the change? Was it incremental demand signals? Was it supply? And, you know, can you supply more if, let's say, demand is $4 or $5 or $6 billion, right? What is the limitation?

Lisa T. Su: And sort of related to that, on the competition side, your competitor will launch their B100 later in the year. Do you think that will change the competitive landscape? Yes, sure, Vivek.

Lisa T. Su: So I think what we said is, as we went from two to three and a half billion, it really is mostly customer demand signals. So as orders have come in on books, and as we've seen programs move from what's called pilot programs into full manufacturing programs, we have, you know, updated the revenue forecast. As I said earlier, from a supply standpoint, we are planning for success. And so we worked closely with our supply chain partners to ensure that, you know, we can ship more than three and a half billion units, substantially more, depending on what customer demand is, as we go into the second half of the year. And then, you know, in terms of, again, you know, roadmaps, as I said, we're very focused on a competitive roadmap, this, you know, that, you know, sort of what the next generations are beyond MI 300.

Lisa T. Su: So I do believe that, you know, we have a strong roadmap in place and continue to work with our customers to, you know, sort of adopt our roadmap as quickly as possible. And a longer-term question, Lisa: if I look at the success that AMD has enjoyed, there are many factors, but a few of them include your early adoption of chiplets and then the strong partnerships we have had with TSMC. But now, you know, we are seeing your x86 competitor, Intel, also adopt chiplet or tile technology, as they call it. And then, I think recently, the manufacturing update they gave, they said they were two years ahead, you know, in terms of incorporating gate all around and backside power delivery. So let's assume they are right, and they have either caught up to TSMC, or, you know, maybe they are ahead. What impact does that have on AMD in the medium? Yeah, sure, Vivek, look, you know, we're always looking at what's next, right?

Lisa T. Su: So on the chiplet technology, I mean, we're sort of on the fourth generation of chiplet technologies. I think we've learned a lot about how to optimize performance there. We are, you know, very aggressive with our adoption of leading-edge technology as it's needed. But I think those are only, you know, a few of the pieces.

Lisa T. Su: We're also focused on continuing to innovate in architecture and design. So I think the longer-term question that you ask is, you know, I think we're expecting that the competition is going to be using a similar process technology. And even in that case, you know, I think we feel like we have a very strong roadmap going forward, and we'll continue to drive both the CPU and the GPU roadmap very aggressively. Harsh Kumar: Yeah.

Lisa T. Su: Hey, thanks for letting me ask a question, guys. I had two questions. Let me start off with the accelerator side.

Lisa T. Su: The question we get a lot from our customers is they want to understand the value proposition of the MI300. So Lisa, I was hoping you could give us some understanding of the price versus power comparison or compute power. And then today, are you seeing your customers that are buying the MI300, are they primarily buying it for inferencing today, or are they using it primarily for tooling? And maybe for Jean, Jean, do you think it is possible for MI300 to finish the year at a run rate of about $1.5 billion? OK, Harsh, so let me start.

Lisa T. Su: We have a lot of memory capacity on MI300X compared to the competition, and what that means is for large language models that have many tens of billions of parameters, you could potentially do the workload in fewer GPUs. So it's a substantial system savings and allows you to do much more work within the same system. In terms of what customers are using MI300 for today, I would say there are a number of customers using it for large language model inferencing, and there are also customers that are using it for training. So I think the whole point is being a strong partner. When you put these AI systems in place, they are sometimes mixed-use systems, so they are used for both training and inferencing.

Operator: John, we have time for two more questions. Yeah, Harsh, let me answer your question about the MI300. By exiting Q4 2024, is it possible to get to 1.5? It is possible, right?

Jean Hu: Because Lisa mentioned earlier, we'll see sequential increases each quarter and more back and loaded in the second half. And we do have supplies of more than $3.5 billion. And of course, we'll continue to make progress with our customers. So the math, yeah, it's possible. But right now, we are really looking at focusing on the execution of the current $3.5 billion plus. This question comes from the line of Stacy Rasgon.

Stacy Aaron Rasgon: Hi guys, thanks for taking my questions. For the first one, you talked about that you had expected a more shallow ramp for the MI-300, and it's clearly doing better than that. So some of the upside, I guess, in the near term, is that being pulled forward from the second half, or is this like a step-up? Or is it more of a step-up in demand both in the first half and the second half relative to what you were? How do I interpret that like a shallow comment? Sure, Stacy.

Lisa T. Su: I don't think it's a pull forward of demand. I think what it is is, you know, we wanted to see how long it would take for customers to fully qualify and get their workloads performing. So, you know, that depends a lot on the actual engineering work that's done. And now that we're, you know, let's call it a quarter later, we've seen that it's gone really well. So it's actually gone, you know, a bit better than our original forecast. And as a result, we've seen stronger demand signals, and customers are gaining confidence in their ability to deploy a significant number of MI300s this year. Thank you. I want to ask Matt's question a little more directly, he didn't quite answer it, that the $400 billion number that you've got out there, is that just silicon and chips, or is there hardware and servers and stuff like that in that number as well? Like what's in that number? Yeah, I thought I had answered it.

Stacy Aaron Rasgon: But yes, I'll answer it again. It is, it is accelerator chips. It is not systems.

Lisa T. Su: So you know, think of it as, you know, GPUs, you know, ASICs that will, you know, be there. So those types of things, but obviously, it includes, you know, memory and other things that are packaged together with the GPU. Yeah, memory will be quite significant, right? So memory is a big portion. The final question comes from the line...

Operator: Thanks for squeezing in. Question for Lisa, as MI300 revenue... Thank you. Thank you.

Operator: double digits, uh... or half the revenue. I don't think it'll be one or two that are half the revenue, Chris. I think we're building this as a, you know, really happy to see, you know, sort of the broad adoption, as always with, you know, sort of the large cloud computing partners, we might see, you know, sort of one or two that are higher than others, but I don't think you see the type of concentration that you mentioned, and then just to follow up. You know, I feel very good about our partnership with TSMC; they You know, we'll see what happens over the next few years.

Operator: But, you know, I'd like to kind of reemphasize what I said earlier, even in the case of, you know, process parity, we feel very good about our architectural roadmap. And, you know, all of the other things that we add as we look at our entire portfolio of CPUs, GPUs, DPUs, and adaptive SOCs and kind of put them together to solve problems. I think we feel really good about what we can do with our customers. So, you know, we're always going to be paying attention to, you know, sort of the process race. But I think we feel very good about, you know, sort of our strategy and how we continue to sort of push the envelope on computing. Great, John. That concludes today's call. Thank you to everyone for joining us today, teleconference I'm, BF-WATCH TV 2021 www.globalonenessproject.org Downloaded From www.AllSubs.org Thank you for watching. I'll see you next time.

Operator: Bye-bye. ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? Time, for http://TheBusinessProfessor.com, Thank you, John, and welcome to AMD's fourth quarter and full year 2023 Financial Results Conference call. By now you should have had the opportunity to review a copy of our earnings press release and the accompanying slides. If you have not had the chance to review these materials, they can be found on the investor relations page of AMD.com. We will refer primarily to non-GAAP financial measures during today's call. The full non-GAAP to GAAP reconciliations are available in today's press release and the slides posted on our website.

Operator: Participants on today's call are Dr. Lisa Su, our Chair and Chief Executive Officer, and Jean Hu, our Executive Vice President, Chief Financial Officer, and Treasurer. This is a live call that will be replayed via webcast on our website. Before we begin, I would like to note that Mark Papermaster, Executive Vice President and Chief Technology Officer, will attend the Bernstein Tech Media, Telecom, and Consumer One-on-One Forum on Tuesday, February 28th, and Jean Hu, Executive Vice President, Chief Financial Officer, and Treasurer, will attend the Wolf Research Semiconductor Conference on Tuesday, February 15th, and the Morgan Stanley Technology, Media, and Telecom Conference on March 5th. Today' Please refer to the cautionary statement in our press release for more information on factors that could cause actual results to differ materially. With that, I'll hand the call over to Lisa.

Lisa T. Su: Thank you, Mitch, and good afternoon to all those listening in today. We finished 2023 strong as data center sales accelerated significantly throughout the year despite the mixed demand environment. As a result, we delivered record data center segment annual revenue and strong top line and bottom line growth in the fourth quarter, driven by the ramp of instinct AI accelerators and robust demand for epic server CPUs across cloud, enterprise, and AI customers. Looking at our financial results, fourth quarter revenue increased 10% year-over-year to $6.2 billion, driven by significant double-digit percentage growth in our data center and client segments. On a full year basis, annual revenue declined 4% to $22.7 billion as record data center and embedded segment annual revenue was offset by lower client and gaming segment revenue.

Lisa T. Su: Importantly, Data Center and Embedded Segment annual revenue grew by $1.2 billion and accounted for more than 50% of revenue in 2023 as we gained server share, launched our next generation Instinct AI accelerators, and maintained our position as the industry's largest provider of adaptive computing solutions. Turning to the fourth quarter business results, data center segment revenue grew 38% year-over-year and 43% sequentially to a record $2.3 billion. Server CPU and data center GPU sales both set quarterly and annual revenue records as sales of our data center products accelerated throughout the year. We gained server CPU revenue share in the quarter, driven by significant double-digit percentage growth in fourth-gen EPYC processor revenue and demand for our third-gen EPYC processor portfolios. In the cloud, while the overall demand environment remains soft, server CPU revenue increased year over year and sequentially as North American hyperscalers expanded fourth-gen EPYC processor deployments to power their internal workloads and public instances.

Lisa T. Su: Amazon, Alibaba, Google, Microsoft, and Oracle brought more than 55 AMD-powered AI, HPC, and general-purpose cloud instances into preview or general availability in the fourth quarter. By 2023, there will be more than 800 Epic CPU-based public cloud instances available. We expect this number to grow in 2024 based on the leadership performance, efficiency, and features of our EPIC CPU portfolios. In enterprise, sales accelerated by a significant double-digit percentage in the quarter as we built momentum with Forbes 2000 customers. We closed multiple wins with large financial, energy, automotive, retail, technology, and pharmaceutical companies, positioning us well for continued growth based on expanded production deployments planned for 2024. A growing number of customers are adopting EPYC CPUs for inferencing workloads, where our leadership throughput performance delivers significant advantages on smaller models like LLAMA 7B, as well as to power head nodes in large training and inference clusters. Looking ahead, customer excitement for our upcoming Turin family of EPYC processors is very strong. Turin is a drop-in replacement for existing 4th Gen EPYC platforms that extends our performance, efficiency, and TCO leadership with the addition of our next-gen Zen5 Core, new memory expansion capabilities, and higher core counts.

Lisa T. Su: In HPC supercomputing, we shipped the majority of AMD Instinct MI300A accelerators for the El Capitan supercomputer in the fourth quarter and expect to complete shipments this quarter for what is expected to be the world's fastest supercomputer when it comes online later this year. We also closed new Instinct GPU wins in the quarter, including the flagship system at the German High Performance Computing Center, HLRS, as well as what is expected to be one of the world's most powerful enterprise supercomputers for energy company, ENI. On AI software development, we made significant progress expanding the ecosystem of AI developers working on AMD platforms with the release of our RockM6 software suite. The RockM6 stack significantly increases performance in key generative AI workloads, adds expanded support and optimizations for additional frameworks and libraries, and simplifies the overall developer experience.

Lisa T. Su: The additional functionality and optimizations of Rackham 6 and the growing volume of contributions from the open-source AI software community are enabling multiple large hyperscale and enterprise customers to rapidly bring up their most advanced large-language models on AMD Instinct accelerators. For example, we are very pleased to see how quickly Microsoft was able to bring up GPT-4 on MI300X in their production environment and roll out Azure private previews of new MI300 instances aligned with the MI300X launch. At the same time, our partnership with HuggingFace, the leading open platform for the AI community, now enables hundreds of thousands of AI models to run out of the box on AMD GPUs, and we are extending that collaboration to our other platforms. Looking ahead, our prior guidance was for data center GPU revenue to be flattish from Q4 to Q1 and exceed $2 billion for 2024.

Lisa T. Su: Based on the strong customer pull and expanded engagements, we now expect data center GPU revenue to grow sequentially in the first quarter and exceed $3.5 billion in 2024. We have also made significant progress with our supply chain partners and have secured additional capacity to support upside demand. Turning to our client segment, revenue was $1.5 billion, an increase of 62% year over year and flat sequentially. We launched our latest generation Ryzen 8000 series notebooks and desktop processors in January, including our Ryzen 8040 mobile series that combine leadership compute performance and energy efficiency with an updated MPU that delivers up to 60% more AI performance compared to our prior generation that was already industry leading. Acer, Asus, HP, Lenovo, MSI, and other large PC OEMs will all offer notebooks powered by our Ryzen 8000 series processors, with the first systems expected to go on sale in February.

Lisa T. Su: To further our leadership in AI PCs, we launched our Ryzen 8000G series processors earlier this month, which are the industry's first desktop CPUs with an integrated AI engine. Millions of AI PCs powered by Ryzen processors have shipped to date, and Ryzen CPUs power more than 90% of AI-enabled PCs currently in the market. Our work with Microsoft and our PC ecosystem partners to enable the next generation of AI PCs expanded significantly in the quarter. We are aggressively driving our Ryzen AI CPU roadmap to extend our AI leadership, including our next-gen Strix processors that are expected to deliver more than 3x the AI performance of our Ryzen 7040 series processors. Strix combines our next-gen Zen 5 core with enhanced RDNA graphics and an updated Ryzen AI engine to significantly increase the performance, energy efficiency, and AI capabilities of PCs.

Lisa T. Su: Going forward, we now expect annual revenue to decline by a significant double-digit percentage year over year, as supply catches up with demand in 2023, and we enter the fifth year of what has been a very strong console cycle. In gaming graphics, revenue grew both year over year and sequentially, driven by strong demand in the channel for both our Radeon 6000 and Radeon 7000 series GPUs. We expanded our Radeon 7000 GPU series with the launch of new RX 7600 XT Series Enthusiast Desktop GPUs earlier this month that offer leadership price performance for 1080p gaming. We also launched new open source FidelityFX Super Resolution 3 software that can deliver significantly higher gaming frame rates on both GPUs and APUs.

Lisa T. Su: We also launched Ryzen embedded processors with unmatched performance and features for industrial automation, machine vision, robotics, and edge server applications. Looking ahead to 2024, we expect overall embedded demand will remain soft through the first half of the year as customers continue to focus on normalizing their inventory levels. Longer term, we're very confident in the growth trajectory of our embedded business as our expanded product portfolio drove more than $10 billion of design wins in 2023, an increase of more than 25% compared to 2022. In summary, I'm very pleased with our fourth quarter and full year results. For 2024, we expect the demand environment to remain mixed, with strong growth in our data center and client segments offset by declines in our embedded and gaming segments.

Lisa T. Su: Against this backdrop, we believe we will deliver strong annual revenue growth and expand gross margins driven by the strength of our Instinct, Epic, and Ryzen product portfolios. Taking a step back, we believe AI is a once-in-a-generation transition that will reshape virtually every portion of the computing market, starting in the data center and then expanding into PCs and across multiple embedded markets.

Lisa T. Su: At the same time, we are rapidly driving leadership AI compute capabilities across the full breadth of our embedded product portfolio. This is an incredibly exciting time for the industry and an even more exciting time for AMD as our leadership IP, broad product portfolio, and deep customer relationships position us well to deliver significant revenue growth and earnings expansion over the next several years. Now I'd like to turn the call over to Jean to provide some additional color on our fourth quarter and full year financial results. Thank you, Lisa, and good afternoon, everyone.

Jean Hu: I'll start with a review of our financial results and then provide our current outlook for the first quarter of fiscal 2024. AMD executed well in 2023 despite a mixed market demand environment, delivering revenue of $22.7 billion and earnings per share of $2.65. We drove year-over-year revenue growth in our embedded and data center segments. In addition, we successfully launched our AMD Instinct MI300 GPUs, positioning us for a strong ramp in 2024 in the AI market. For the fourth quarter of 2023, revenue was $6.2 billion, growing 10% year-over-year as revenue growth in the data center and the client segments was partially offset by lower revenue in our embedded and the gaming segments. Revenue was up 6% sequentially, primarily driven by the ramp of AMD Instinct GPUs across several leading customers and higher revenue from EPYC server processing, partially offset by the decline in embedded and gaming segment revenue.

Jean Hu: Growth margin was 51% flat year-over-year, with higher revenue contribution from the data center and the client segments offset by lower embedded segment revenue. Operating expenses were $1.7 billion, an increase of 8% year-over-year, as we invest in R&D and marketing activities to support our significant AI growth opportunities. Operating income was $1.4 billion, representing a 23% operating margin. Texas interest expense and other was $163 million.

Jean Hu: For the fourth quarter of 2023, diluted earnings per share was $0.77, an increase of 12% year-over-year. Now, turning to our reportable segment. Starting with the data center segment, revenue was $2.3 billion, up 38% year-over-year and 43% sequentially, driven by strong growth of both AMD Instinct GPU and 4th generation AMD IPEX CPU sales. Data center segment operating income was $666 million, or 29% of revenue compared to $444 million or 27% a year ago. Higher operating income was primarily due to operating leverage driven by higher revenue. Client segment revenue was $1.5 billion, up 62% year-over-year driven by Ryzen 7000 series CPU sales. Kalyan segment operating income was $55 million, or 4% of revenue, compared to an operating loss of $152 million a year ago, driven by higher revenue.

Jean Hu: Gaming segment revenue was $1.4 billion, down 17% year-over-year and 9% sequentially due to a decrease in semi-customer revenue partially offset by an increase in Radeon GPU sales. Gaming segment operating income was $224 million, or 16% of revenue compared to $266 million or 16% a year ago. Embedded segment revenue was $1.1 billion, down 24% year-over-year and 15% sequentially as customers continue to work down their inventory levels. Embedded segment operating income was $461 million, or 44% of revenue compared to $699 million or 50% a year ago. Turning to the balance sheet and the cash flow. During the quarter, we generated $381 million in cash from operations, and our free cash flow was $242 million. Inventory decreased sequentially by $94 million to $4.4 billion.

Jean Hu: Client segment revenue is expected to decline seasonally and in the gaming segment. As we enter the fifth year of what has been a very strong gaming cycle and given current customer inventory levels, we expect revenue to decline by a significant double-digit percentage. Year-over-year, we expect data center and client segment revenues to increase by a strong double-digit percentage, given the strength of our product portfolio and share gain opportunities. We expect invest segment revenues to decline, and the gaming segment revenue to decline by a significant double-digit percentage. In addition, we expect the first quarter non-GAAP growth margin to be approximately 52%, and non-GAAP operating expenses to be approximately $1.73 billion. The non-GAAP effective tax rate is expected to be 13%, and the diluted share count is expected to be approximately 1.63 billion shares.

Jean Hu: We expect to expand the growth margin in 2024 and continue to invest to address the large AI opportunities while driving operating model leverage to deliver earnings per share growth faster than top-line revenue growth. In closing, we delivered solid financial results in 2023, further strengthening our product portfolio and establishing ourselves as a leading provider of data-centered GPUs for AI. We are very well positioned to build on this momentum and deliver strong financial performance in 2024 and beyond. With that, I'll turn it back to Mitch for the Q&A session. Thank you, Jean. John?

Operator: We're happy to poll the audience for questions and for more information. Thank you for watching. I'm Bob Lepofsky.

Operator: Have a great week. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. Bye. By

Operator: Next question comes from Aaron Rakers of, Yeah, thanks for taking the question. I just kind of framing the outlook and the guidance for this calendar first quarter. I guess the first question is, you know, can you help us on a relative basis with the 400 million dollars of data center GP revenue that you expected in Q4, what that ultimately kind of fell out to be? And then on the guidance into 1Q, can you help us appreciate what seasonality, you know, is defined as as we think about the server business in the 1Q guide? Sure, Aaron, let me start and then see if Jean has something to add.

Lisa T. Su: So, you know, relative to the data center GPU business, look, we were very pleased with the performance that we saw in the fourth quarter. You know, it was always going to be a very sort of back-end quarter weighted as we were, you know, ramping the product, and we saw MI300A, our HPC product, actually ramped very well. And then we saw MI300X, the AI product, actually exceed our expectations based on strong customer demand, the way the qualifications went, and then the manufacturing ramp. So we were over 400 million for that business in the fourth quarter.

Jean Hu: Aaron, I'll give you some color about client seasonality and others. So client is very similar to server. Typically, Q1 is high single digit to low double digit.

Jean Hu: On the gaming side, Lisa mentioned during her prepared remarks that we are in the late stage of the product cycle for the year five of gaming consoles. But at the same time, we also have inventory at customers. So the combination of those impacts, we expect the Q1 gaming sequential decline to be probably more than 30%. So hopefully, that'll help you a little bit. Yeah, very helpful, Gene.

Lisa T. Su: And as a quick follow-up, I'm just curious about the traditional server demand that you see, I know, when we look at server CPU shipments are down, you know, north of 20% year over year. Are you seeing any signals? Or how are you thinking about a recovery in that traditional, call it non-AI general purpose server market as we move through 20? Sure, Aaron.

Lisa T. Su: So look, I think, you know, I agree with your characterization of 2023 demand, although we did see, you know, some strong progress in the second half of the year, especially as customers in cloud and enterprise adopted our Genoa and our Zen4 family. So going into 2024, I would say the traditional server market is probably still mixed, especially in the first half of the year. There's still some cloud optimization going on, as well as, you know, sort of enterprises being a little bit cautious. That being the case, though, we also see opportunities for us to continue to grow our share in the traditional server business. I think our portfolio is extremely strong. The adoption of, you know, Genoa and Bergamo, as well as our new Sienna product lines, are getting bigger, And we also see Turin, our Zen5 product, coming in the second half of the year.

Lisa T. Su: So even in a mixed demand environment, I think we're bullish on what traditional server CPUs can do in 2024, until the next from the line. Thanks a lot. Um, Lisa, I'm wondering if you can give us a little bit of sense in terms of the milepost that you're kind of, you know, marching toward on this $400 billion tan that you have for, you know, 2027. For example, do you think you can gain share at a rate that's kind of similar to the rate that you gain share for server CPUs? Or, I guess, maybe asked in a different way?

Lisa T. Su: Is it reasonable to kind of look at your consumer GPU share of 20 plus percent? Is that a reasonable bogeyman? Or do you have aspirations higher than that? Yeah, thanks, Tim, for the question. I would say a couple things.

Lisa T. Su: First of all, we're really pleased with the progress that we've made in our data center GPU business. I think the ramp that we've seen, the customer traction that we've seen, even in the last few months, has been great. And that gives us a lot of confidence in the growth of this business. I think the beauty of the AI market here is that it's growing so quickly that I think we have both the market dynamics as well as our ability to gain share in that framework.

Lisa T. Su: The point I will make is, you know, our customer engagements right now are all quite strategic, dozens of customers with multi-generational conversations. So, as excited as we are about the ramp of MI300, and frankly, there's a lot to do in 2024, we're also very excited about the opportunities to extend that into the next couple of years, you know, out into that, you know, 25, 26, 27 timeframe. So I think we're seeing a lot of growth. I think it's a little early to make market share projections, but I would say it's a significant growth driver given market demand as well as, you know, our own product capabilities. Thanks a lot.

Jean Hu: I think the way to think about it is, you know, Lisa mentioned during her prepared remarks that we feel pretty good about both our data center and client business growing in 2024. Of course, the largest incremental revenue opportunities are going to come from data centers between both the server side, the Ganymode share, and the data center GPU side with a significant ramp of our MI300. I think that's how we should think about it.

Jean Hu: We do have a headwind from the gaming segment. We do think, year over year, we'll see a very significant double-digit decline in the gaming segment, and at the same time, embedded is going through the bottoming process. We do think in the second half, we'll see a recovery.

Jean Hu: So those are the puts and the takes I can talk about. The next question comes from the line of Matt. Thank you very much. Good afternoon.

Operator: I wanted to ask, I mean, there's been so much progress made over the last six months, doubtless. Software, product. The Bulletproof Executive 2013, What I'm interested in hearing a little bit more about, and you guys have been open about... Some of the forward programs, Matt's perspective on how you're thinking about the roadmap. M.I.

Operator: Accelerator, and I can. Parts that are. QC....

Lisa T. Su: The Bulletproof Executive 2013, Yeah, sure, Matt. So I appreciate the comments. I think the, you know, the traction that we're getting with the MI 300 family is really strong. I think what's benefited us here is our use of chiplet technologies, which has given us the ability to have, you know, sort of both the APU version as well as the GPU version. And, you know, we continue to use that to differentiate ourselves.

Lisa T. Su: And that's how we get our memory bandwidth and memory capacity advantages. As we go forward, you can imagine, like we did in the EPIC timeframe, we plan multiple generations in sequence. That's the way we're planning the roadmap. You know, one of the things I will note about the AI accelerator market is the demand for compute is so high that we are seeing, you know, sort of an acceleration of the roadmap, you know, generations here, and we are similarly planning for an acceleration of our roadmap. I would say that, you know, we'll talk more about the overall roadmap beyond MI 300 as we get into later this year. But you can be assured that we're working very closely with our customers to have a very competitive roadmap for both training and inference. That will come out over the next couple of years. This is a follow-up. That's one of the questions that I've... Thanks for watching!

Operator: Thanks for tuning in. The Bulletproof Executive 2013, If you could give us a little bit of your time, we would love to hear from you. Thank you. Look under the hood as I guess.

Lisa T. Su: I've got a hundred versions of the same question, which is how the heck did you come up with, Um, so if you could give us a little bit more. Are we talking about AcceleratorCard, about silicon servers, and what kind of Unit Assumptions, any kind of thing that you can give us on market size? Thank you. Thank you.

Lisa T. Su: ,,, Precise, that would. Sure. Well, Matt, I don't know how precise it is, but I think we said approximately $400 billion. But I think what we need to look at is the growth rate and how we get to those growth rates. I think we expect, you know, units to grow, you know, sort of substantial double-digit percentage, but you should also expect that content is going to grow. So if you think about, you know, how important memory and memory capacity are as we go forward, you can imagine that, you know, we'll see, you know, an acceleration there in just the overall content as we go to, you know, more advanced technology nodes. So there's some ASP uplift in there.

Lisa T. Su: And then what we also do is, you know, we're planning longer-term roadmaps with our customers in terms of how they're thinking about, you know, sort of the size of training clusters, the number of training clusters, and then the fact that, you know, we believe inference is actually going to exceed training as we go into the next couple of years, just given, you know, as more enterprises adopt it. So, as we look at all of those pieces, I think we feel good that the growth rate is going to be, you know, significant and sustained over the next few years. In terms of what's in that, Tam, it really is accelerator TAM. So within accelerators, there are, you know, certainly GPUs, and there will also be some ASICs that are or other accelerators that are in that TAM.

Lisa T. Su: As we think about sort of the different types of models from, you know, smaller models to fine tuning of models to, you know, very, the largest, large language models, I think you're going to need different silicon for those, you know, different use cases. But from our standpoint, you know, GPUs are still going to be the compute element of choice when you're talking about training and inferencing on the largest language models. Great, thank you.

Lisa T. Su: I think you talked about the mi 300 cloud workloads being kind of split between the more customer-facing workloads and versus internal. Can you talk about how you see the breakdown of that? And how, you know, how is your ecosystem progressing? This is a brand new chip. It seems impressive that you're able to support kind of a broad range of customer-facing workloads in the cloud. Yeah, sure, Joe.

Lisa T. Su: And, you know, we've seen performance come up nicely, which is what we expect, frankly, with the GPU capabilities. We would have to do some level of optimization, but that optimization has gone well. I think to your broader question, you know, the way I look at this is that there are lots of opportunities for us to work directly with large customers, both on the cloud side, as well as on the enterprise side, who have specific training and inferencing workloads. You know, our job is to make it as easy as possible for them.

Lisa T. Su: And so, our entire tool chain, you know, all of, you know, our, you know, sort of the overall Rokam suite has really gone through, you know, significant progress over the last six to nine months. And then we're also getting some nice support from the open source community. So the work that Hugging Face is doing is tremendous in terms of just real-time optimization on our hardware, our partnership with OpenAI on Triton, and our work across a number of these open source models has helped us actually make very rapid progress.

Jean Hu: And for my follow-up, I guess, you know, a lot of the forecasting of your business that I'm hearing is coming from the supply chain. And we're sort of hearing AMD is building X in Asia. I guess, how would you ask us to think about that? You know, are you looking at being kind of sold out for the year?

Lisa T. Su: And so the supply chain would be close to revenue? Are you building for the best case scenario? Just, you know, I worry about expectations sometimes when people hear these supply chain numbers, and I'm just curious how you bridge the gap. Yeah, so, Joe, I think we updated our revenue expectations for this quarter from, you know, our original number of two billion to three and a half billion to try to give some boundaries on some of the discussion out there. The way to think about, you know, the three and a half billion is that these are customers that we're working with who have, you know, given us firm commitments on what they need. As you know, the lead times on these products are quite long. So it's important, you know, to have those forecasts in early, and we have a strong order book.

Lisa T. Su: So that gives us, you know, good confidence to exceed the three and a half billion. From a supply chain standpoint, our goal is always to build more supply. We, and so from that standpoint, we have also worked with our supply chain partners and secured significant capacity. You know, think about it as first half capacity is tight, and more comes on in the second half of the year.

Lisa T. Su: But, you know, we've certainly made more progress there. So, you know, we do have more supply, and we're going to continue to work with our customers on their deployments, and we'll update that number as we go through the year. Next question comes from the line: Toshiya Hari with Goldman Sachs. Hi, thank you for taking the question. I had one on the MI300 as well, Lisa.

Lisa T. Su: I guess first of all, how should we think about the quarterly trajectory beyond Q1? You talked about Q1 being up sequentially. Is it fair to assume kind of a straight line as we progress through the balance of the year, or is it more second half skewed?

Lisa T. Su: How should we think about that? And, I guess more importantly, some of the potential cloud customers that have yet to officially sign up for or sign off on the MI300, I guess what's the sticking point? Is it just a function of time, and you just need a little bit more time to go back and forth and tweak things? Or is there some kind of software-related concern?

Lisa T. Su: I guess what's holding them back at this point? Yeah, Toshiya, thanks for the question. So first, on the MI300 trajectory, I think you would expect that revenue should increase every quarter from now through sort of the end of the year, but it will be a bit more second half weighted. And part of that is just customers as they're finishing up their qualifications on their lines, as well as sort of how our supply chain is ramping up. So yes, it should increase each quarter but be a bit more second half weighted. And then to your comment about customers, look, we're engaged with all of sort of the large customers. These are all folks that know us really well, given our deep relationships in Epic.

Lisa T. Su: I think the key is this is not just about an MI300 conversation, but it really is about sort of our long-term multi-generational roadmap. And so that's the context in which we're working with our largest customers, as well as, as you know, there's a lot of demand coming from folks that are more AI-centric and not necessarily typical cloud customers but more enterprise or, you know, let's call it AI-specific companies that we're also very well engaged with. I got it. That's super helpful.

Toshiya Hari: And then as my follow-up question, maybe one for Jean on the gross margin side, you're guiding Q1 to 52%. I'm curious, again, I'm sure you're not going to give, you know, quantitative guidance beyond Q1, but how to think about the trajectory for Q2 and beyond. I'm pretty sure you're working through some kinks as it pertains to the instinct ramp.

Jean Hu: Hopefully, that improves over time. So that should be a tailwind. FPGAs, perhaps, the second half, you know, turns for the better. And you've got server CPU volume growth throughout the year. So it feels like you've got multiple tailwinds as we think about gross margin progression on a sequential basis. But what are the potential headwinds as we move throughout 2024? Thank you.

Jean Hu: Yeah, Toshiya, thank you for the question. Yeah, you're absolutely right. We have some puts and takes that impact our gross margin. We guided the Q1 120 basis points higher than Q4 sequentially, primarily because the higher data center contribution actually more than offset the decline of embedded business in Q1. Going forward, the way to think about it is, as you said, the major driver is going to be the data center business, which is going to grow much faster than the other segments. That makes change will help us to expand the gross margin nicely. I think you also are spot on with embedded coming back in the second half, which will be a tailwind with the data center GPU. We are at a very early stage of ramp. We are improving testing time yield and continuing to expand the gross margin, and we expect it to be a creative to the corporate average. So those are all the tailwinds coming in the second half.

Jean Hu: I would say the headwinds inside continue to be, in the first half where we see embedded business, not only in Q1 do we see sequential decline, but Q2 probably is going to be sequentially flat versus Q1. That is a headwind for us because it does have a very nice gross margin. But overall, we feel pretty good about the trajectory of gross margin improvement, especially in the second half. The next question comes from the line of... Hi, thanks for letting me ask you a question. I wanted to get into the competitive environment. First, on the instinct side of things, how that's going. It doesn't seem to be slowing down your ramp whatsoever. But then also on the street.

Lisa T. Su: From a competitive standpoint, what we see is Zen 4 is extremely competitive right now with Genoa, Bergamo, Sienna, and as we go into Turin, we're deep in the design cycle for Zen 5. And we feel very good about how we're positioned. Thanks for that. I guess as my follow-up question, on the data center side, another theme that's been pervasive throughout 2023, at least on the GPU side, crowding out the CPU side, you mentioned that there was still a little bit of cloud digestion going on within your Epic business. But where do you see that standing?

Lisa T. Su: Or is it? I think we expect the CPU business from a market standpoint to grow, Ross, as we go into 2024. I think the rate and pace of growth will depend a little bit on the macro and just, you know, overall CapEx, you know, trends. But, you know, from our standpoint, we are starting to see some of our larger customers plan their refresh cycles; there's a lot of what's called older equipment that has yet to be, you know, refreshed. And, you know, the value proposition for refresh is so strong, because, you know, the energy efficiency and sort of the footprint of the newer generations are so much better than sort of the four or five-year-old infrastructure, that we do I think the exact timing we'll have to, you know, understand more as the market evolves as we go through the year. Vivek Arya, Thanks for taking my questions.

Vivek Arya: So first one, Lisa, you gave us a $2 billion plus number for MI before; now you have raised it to over $3.5 billion. And I'm curious, what drove the change? Was it incremental demand signals? Was it supply? And, you know, can you supply more if, let's say, demand is $4 or $5 or $6 billion, right? What is the limitation?

Lisa T. Su: So I think what we said is, as we went from two to three and a half billion, it really is mostly customer demand signals. So as orders have come in, and as we've seen programs move from what's called pilot programs into full manufacturing programs, we have updated the revenue forecast. As I said earlier, from a supply standpoint, we are planning for success. And so we worked closely with our supply chain partners to ensure that we can ship more than three and a half billion units, substantially more, depending on what customer demand is as we go into the second half of the year.

Lisa T. Su: And then in terms of, again, roadmaps, as I said, we're very focused on a competitive roadmap for what the next generations are beyond MI 300. So I do believe that we have a strong roadmap in place and continue to work with our customers to sort of adopt our roadmap as quickly as possible. And a longer-term question, Lisa; if I look at the success that AMD has enjoyed, there are many factors, but a few of them include your early adoption of chiplets and then the strong partnerships we have had with TSM.

Lisa T. Su: But now, you know, we are seeing your x86 competitor, Intel, also adopt chiplet or tile technology, as they call it. And then, I think recently, the manufacturing update they gave, they said they were two years ahead, you know, in terms of incorporating gate all around and backside power delivery. So let's assume they are right, and they have either caught up to TSMC, or, you know, maybe they are ahead. What impact does that have on AMD in the medium? Yeah, sure, Vivek, look, you know, we're always looking at what's next, right?

Lisa T. Su: So on the chiplet technology, I mean, we're sort of on the fourth generation of chiplet technologies. I think we've learned a lot about how to optimize performance there. We are, you know, very aggressive with our adoption of leading-edge technology, as it's needed. But I think those are only, you know, a few of the pieces.

Harsh V. Kumar: Hey, thanks for letting me ask a question, guys. I had two questions. Let me start off with the accelerator side.

Lisa T. Su: So your question about the value proposition for MI300, again, customers are using it for different reasons, but presume that there is a performance per dollar benefit to using AMD. So that's one piece of it. The other piece of it, though, is that we intrinsically have more. So you can see the high capacity on MI300X compared to the competition. And what that means is for large language models that have many tens of billions of parameters, you could potentially do the workload on fewer GPUs. So it's a substantial system savings and allows you to do much more work within the same system. In terms of what customers are using MI300 for today, I would say there are a number of customers using it for large language model inferencing, and there are also customers that are using it for training. So I think the whole point is being a strong partner. When you put these AI systems in place, they are sometimes mixed-use systems, so they are used for both training and inferencing.

Operator: John, we have time for two more questions. Yeah, Harsh, let me answer your question about the MI300. By exiting Q4 2024, is it possible to get to 1.5? It is possible, right?

Stacy Aaron Rasgon: Hi guys, thanks for taking my questions. For the first one, you talked about that you had expected a more shallow ramp for the MI-300, and it's clearly doing better than that. So some of the upside, I guess, in the near term, is that being pulled forward from the second half, or is this like a step-up, or is it more of a step-up in demand both in the first half and the second half relative to what you were? How do I interpret that like a shallow comment? Sure, Stacy.

Lisa T. Su: And as a result, we've seen stronger demand signals, and, you know, customers are gaining confidence in their ability to deploy a significant number of MI300s this year. Thank you. I want to ask Matt's question a little more directly, because he didn't quite answer it. The $400 billion number that you've got out there, is that just silicon and chips, or is there hardware and servers and stuff like that in that number as well? Like, what's in that? Yeah, I thought I had already answered that.

Stacy Aaron Rasgon: But yes, I'll answer it again. It is, it is accelerator chips. It is not systems.

Lisa T. Su: So you know, think of it as, you know, GPUs, you know, a six that will be there. So those types of things, but obviously, it includes memory and other things that are packaged together with the GPU. Yeah, memory will be quite significant, right? So memory is a big portion. The final question comes from the line of...

Operator: Thanks for squeezing in. Question for Lisa, as MI300 revenue, Thank you, customers, double digits, uh... or half the revenue. I don't think it'll be one or two that are half the revenue, Chris.

Lisa T. Su: I think we're building this as, you know, really happy to see, you know, sort of broad adoption, as always with, you know, sort of the large cloud computing partners, we might see, you know, sort of one or two that are higher than others, but I don't think you see the type of concentration that you mentioned, and then just to follow. I feel very good about our partnership with TSMC, they continue to execute extremely well, you But, you know, I'd like to kind of reemphasize what I said earlier. Even in the case of, you know, process parity, we feel very good about our architectural roadmap.

Lisa T. Su: And, you know, all of the other things that we add. As we look at our entire portfolio of CPUs, GPUs, DPUs, and adaptive SOCs and kind of put them together to solve problems, I think we feel really good about what we can do with our customers. So, you know, we're always going to be paying attention to, you know, sort of the process race, but I think, I think we feel very good about, you know, sort of our strategy and how we continue to sort of push the envelope on the computing roadmap. And that is Great job, John. That concludes today's call. Thank you to everyone for joining us today. Telecom

Tower, operator assistance during the conference. Please press Star Zero on your telephone keypad and as a reminder, this conference is being recorded it.

It is now my pleasure to introduce to you Mitch Haws, Vice President Investor Relations. Thank you Mitch you may begin.

Thank you John and welcome to Amd's fourth quarter, and full year 2023 financial results Conference call.

By now you should have had the opportunity to review a copy of our earnings press release and the accompanying slides if.

If you have not had the chance to review these materials can be found on the Investor Relations page of AMD Dot com.

We will refer primarily to non-GAAP financial measures during today's call to full non-GAAP to GAAP reconciliations are available in today's press release and the slides posted on our website.

Participants on today's call are Dr. Lisa Su, our chair and Chief Executive Officer, and Jean Hu, Our executive Vice President Chief Financial Officer and Treasurer.

Greetings and welcome to the a M D fourth quarter and full year 2023 conference call. At this time all participants are in a listen only mode. A brief question and answer session will follow the formal presentation. If anyone should require operator assistance. During the conference. Please press star zero on your telephone keypad.

This is a live call and will be replayed via webcast on our website.

Before we begin I would like to note that Mark Papermaster Executive Vice President and Chief Technology Officer will attend the Bernstein Tech media Telecom and consumer one on one forum on Tuesday February 28, and Jean Hu Executive Vice President Chief Financial Officer, and Treasurer will attend the Wolfe Research semiconductor conference on Tuesday February.

As a reminder, this conference is being recorded.

It is now my pleasure to introduce to you Mitch Haws, Vice President Investor Relations. Thank you Mitch you may begin.

Mitch Steves: Thank you John and welcome to Amd's fourth quarter, and full year 2023 financial results Conference call.

15th and the Morgan Stanley Technology Media and Telecom conference on March 5th.

Mitch Haws: By now you should have had the opportunity to review a copy of our earnings press release and the accompanying slides if.

Today's discussion contains forward looking statements based on current beliefs assumptions and expectations speak only as of today and as such involve risks and uncertainties that could cause actual results to differ materially from our current expectations.

Mitch Steves: If you have not had the chance to review these materials can be found on the Investor Relations page of AMD Dot com.

Mitch Steves: We will refer primarily to non-GAAP financial measures during today's call. The full non-GAAP to GAAP reconciliations are available in today's press release and the slides posted on our website.

Please refer to the cautionary statement in our press release for more information on factors that could cause actual results to differ materially with that I'll hand, the call over to Lisa.

Speaker Change: Participants on today's call are Dr. Lisa Su, our chair and Chief Executive Officer, and Jean Hu, Our executive Vice President Chief Financial Officer and Treasurer.

Thank you Mitch and good afternoon to all those listening in today.

Finished 2023 strong as data center sales accelerated significantly throughout the year, despite the mixed demand environment.

Speaker Change: This is a live call and will be replayed via webcast on our website.

Speaker Change: Before we begin I would like to note that Mark Papermaster Executive Vice President and Chief Technology Officer will attend the Bernstein Tech media Telecom and consumer one on one forum on Tuesday February 28, and Jean Hu Executive Vice President Chief Financial Officer, and Treasurer will attend the Wolfe Research semiconductor conference on Tuesday February <unk>.

As a result, we delivered record datacenter segment annual revenue and strong topline and Bottomline growth in the fourth quarter driven by the ramp of instinct, AI accelerators and robust demand for epic server Cpus across cloud enterprise and AI customers.

Looking at our financial results fourth quarter revenue increased 10% year over year to $6 2 billion driven by significant double digit percentage growth in our data center and client segments.

Speaker Change: 15th and the Morgan Stanley Technology Media and Telecom conference on March 5th.

Speaker Change: Today's discussion contains forward looking statements based on current beliefs assumptions and expectations speak only as of today and as such involve risks and uncertainties that could cause actual results to differ materially from our current expectations.

On a full year basis annual revenue declined 4% to $22 7 billion as record datacenter and embedded segment annual revenue was offset by lower client and gaming segment revenue.

Speaker Change: Please refer to the cautionary statement in our press release for more information on factors that could cause actual results to differ materially with that I'll hand, the call over to Lisa.

Importantly, datacenter and embedded segment annual revenue grew by $1 2 billion and accounted for more than 50% of revenue in 2023, as we gain server share launched our next generation instinct, AI accelerators and maintained our position as the industry's largest provider of adaptive computing solutions.

Lisa T. Su: Thank you Mitch and good afternoon to all those listening in today.

Lisa T. Su: We finished 2023 strong as data center sales accelerated significantly throughout the year. Despite the mixed demand environment. As a result, we delivered record datacenter segment annual revenue and strong topline and bottomline growth in the fourth quarter driven by the ramp of instinct, AI accelerators and robust demand for epic server CPU.

Turning to the fourth quarter business results datacenter segment revenue grew 38% year over year, and 43% sequentially to a record $2 3 billion.

Server CPU and datacenter GPU sales, both set quarterly and annual revenue records as sales of our data center products accelerated throughout the year.

Lisa T. Su: Use across cloud enterprise and AI customers.

Lisa T. Su: Looking at our financial results fourth quarter revenue increased 10% year over year to $6 2 billion driven by significant double digit percentage growth in our data center and client segments.

We gained server CPU revenue share in the quarter driven by significant double digit percentage growth in fourth Gen epic processor revenue and demand for our third Gen epic processor portfolio.

Lisa T. Su: On a full year basis annual revenue declined 4% to $22 7 billion as record datacenter and embedded segment annual revenue was offset by lower client and gaming segment revenue.

And cloud while the overall demand environment remains soft server CPU revenue increased year over year and sequentially as North American Hyperscale has expanded <unk> epic processor deployments to power their internal workloads in public instances.

Lisa T. Su: Importantly, datacenter and embedded segment annual revenue grew by $1 2 billion and accounted for more than 50% of revenue in 2023, as we gain server share launched our next generation instinct, AI accelerators and maintained our position as the industry's largest provider of adaptive computing solutions.

<unk> Alibaba, Google, Microsoft and Oracle brought more than 55, AMD powered AI HBC and general purpose cloud instances into preview or general availability in the fourth quarter.

Exiting 2023, there were more than 800 epic CPU based public cloud instances available.

Turning to the fourth quarter business results datacenter segment revenue grew 38% year over year, and 43% sequentially to a record $2 3 billion.

We expect this number to grow in 2024 based on the leadership performance efficiency and features of our epic CPU portfolio.

Lisa T. Su: Server CPU and datacenter GPU sales, both set quarterly and annual revenue records as sales of our data center products accelerated throughout the year.

In enterprise sales accelerated by a significant double digit percentage in the quarter as we built momentum with Forbes 2000 customers.

We gained server CPU revenue share in the quarter driven by significant double digit percentage growth in fourth Gen epic processor revenue and demand for our third Gen epic processor portfolio.

We closed multiple wins with large financial energy automotive retail technology, and pharmaceutical companies positioning us well for continued growth based on expanded production deployments planned for 2024.

Lisa T. Su: And cloud while the overall demand environment remains soft server CPU revenue increased year over year and sequentially as North American Hyperscale or expanded forged an epic processor deployments to power their internal workloads in public instances.

A growing number of customers are adopting epic Cpus for inferencing workloads, where our leadership throughput performance deliver significant advantages on smaller models like Lama <unk> as well as the power head nodes and large training and inference clusters.

Lisa T. Su: Amazon Alibaba, Google, Microsoft and Oracle brought more than 55, AMD powered AI HBC and general purpose cloud instances into preview or general availability in the fourth quarter.

Looking ahead customer excitement for our upcoming turn family of Epic processors is very strong.

Turn is a drop in replacement for existing fourth Gen epic platforms that extends our performance efficiency and Tcl leadership with the addition of our next Gen 10, five core new memory expansion capabilities and higher core counts.

Lisa T. Su: Exiting 2023, there were more than 800 epic CPU based public cloud instances available.

We expect this number to grow in 2024 based on the leadership performance efficiency and features of our epic CPU portfolio.

Internal and end customer validation work is progressing to plan with turn on track to deliver overall performance leadership as well as leadership on a per core or per watt basis across a wide range of workloads when it launches later this year.

In enterprise sales accelerated by a significant double digit percentage in the quarter as we built momentum with Forbes 2000 customers.

Lisa T. Su: We closed multiple wins with large financial energy automotive retail technology, and pharmaceutical companies positioning us well for continued growth based on expanded production deployments planned for 2024.

Turning to our broader datacenter portfolio, our data center GPU business accelerated significantly in the quarter with revenue exceeding our $400 million expectation driven by a faster ramp for <unk> 300, <unk> with AI customers.

Lisa T. Su: A growing number of customers are adopting epic Cpus for inferencing workloads, where our leadership throughput performance deliver significant advantages on smaller models like Lama <unk> as well as the power head nodes and large training and inference clusters.

We launched our <unk> 300 accelerator family in December with strong partner and ecosystem support from multiple large cloud providers, all the major Oems and many leading AI developers.

Lisa T. Su: Looking ahead customer excitement for our upcoming turn family of Epic processors is very strong.

<unk> 300, ex Gpus deliver leadership generative AI performance by combining our high performance cdna, three architecture with industry, leading memory bandwidth and capacity.

Lisa T. Su: <unk> is a drop in replacement for existing fourth Gen epic platforms that extends our performance efficiency and Tcl leadership with the addition of our Nextgen Zen five core new memory expansion capabilities and higher core counts.

Customer response to <unk> 300 has been overwhelmingly positive and we are aggressively ramping production to support the dozens of cloud enterprise and supercomputing customers deploying instinct accelerators.

Lisa T. Su: Internal and end customer validation work is progressing to plan with turn on track to deliver overall performance leadership as well as leadership on a per core or per watt basis across a wide range of workloads when it launches later this year.

In cloud, we are working closely with Microsoft Oracle meta and other large cloud customers on instinct Gpus deployments powering both their internal AI workloads and external offerings.

Lisa T. Su: Turning to our broader datacenter portfolio, our data center GPU business accelerated significantly in the quarter with revenue exceeding our 400 million expectation driven by a faster ramp for <unk> 300, <unk> with AI customers.

For enterprise customers HPE, Dell Lenovo Supermicro and other server vendors are on track to launch differentiated 300 platforms. Later this quarter with strong demand from multiple enterprise customers.

Lisa T. Su: We launched our <unk> 300 accelerator family in December with strong partner and ecosystem support from multiple large cloud providers all of the major Oems and many leading AI developers.

In HBC supercomputing, we shipped the majority of AMD instinct mid 300, <unk> accelerators for the El Capitan supercomputer in the fourth quarter and expect to complete shipments this quarter for what is expected to be the world's fastest supercomputer when it comes online later this year.

Lisa T. Su: <unk> 300, ex Gpus deliver leadership generative AI performance by combining our high performance cdna, three architecture with industry, leading memory bandwidth and capacity.

We also closed new instinct GPU wins in the quarter, including the flagship system at the German high performance Computing Center HRS as well as what is expected to be one of the world's most powerful enterprise supercomputers for energy company Eni.

Lisa T. Su: Customer response to <unk> 300 has been overwhelmingly positive and we are aggressively ramping production to support the dozens of cloud enterprise and supercomputing customers deploying instinct accelerators.

On AI software development, we made significant progress expanding the ecosystem of AI developers working on AMD platforms with the release of our <unk> <unk> software suite.

Lisa T. Su: In cloud, we are working closely with Microsoft Oracle meta and other large cloud customers on instinct Gpus deployment powering both their internal AI workloads and external offerings.

The <unk> stacks significantly increases performance in key generative AI workloads adds expanded support and optimizations for additional frameworks in libraries and simplifies the overall developer experience.

Lisa T. Su: For enterprise customers HPE, Dell Lenovo Supermicro and other server vendors are on track to launch differentiated 300 platforms. Later this quarter with strong demand from multiple enterprise customers.

The additional functionality and optimizations of <unk> six and the growing volume of contributions from the open source AI software community are enabling multiple large hyperscale and enterprise customers to rapidly bring up their most advanced large language models on AMD instinct accelerators.

Lisa T. Su: In HBC supercomputing, we shipped the majority of AMD instinct 300 accelerators for the El Capitan supercomputer in the fourth quarter and expect to complete shipments this quarter for what is expected to be the world's fastest supercomputer when it comes online later this year.

For example, we are very pleased to see how quickly Microsoft was able to bring up GPT for on mid 300, <unk> in their production environment and rollout Azure private previews of new MRI 300 instances aligned with the 300 X launch.

Lisa T. Su: We also closed new instinct GPU wins in the quarter, including the flagship system at the German high performance Computing Center HRS as well as what is expected to be one of the world's most powerful enterprise supercomputers for energy company Eni.

At the same time, our partnership with hugging face the leading open platform for the AI community now enabled hundreds of thousands of AI models to run out of the box on AMD Gpus, and we're extending that collaboration to our other platforms.

On AI software development, we made significant progress expanding the ecosystem of AI developers working on AMD platforms with the release of our <unk> <unk> software suite.

Looking ahead, our prior guidance was for datacenter GPU revenue to be flattish from Q4 to Q1 and exceed $2 billion for 2024.

Lisa T. Su: The Broadcom six stack significantly increases performance in key generative AI workloads adds expanded support and optimizations for additional frameworks in libraries and simplifies the overall developer experience.

Based on the strong customer pull and expanded engagements. We now expect datacenter GPU revenue to grow sequentially in the first quarter and exceed $3 5 billion in 2024.

Lisa T. Su: The additional functionality and optimizations of <unk> six and the growing volume of contributions from the open source AI software community are enabling multiple large hyperscale and enterprise customers to rapidly bring up their most advanced large language models on AMD instinct accelerators for.

We have also made significant progress with our supply chain partners and have secured additional capacity to support upside demand.

Turning to our client segment revenue was $1 5 billion, an increase of 62% year over year and flat sequentially.

Lisa T. Su: For example, we are very pleased to see how quickly Microsoft was able to bring up GPT for on <unk> 300, <unk> in their production environment and rollout Azure private previews of new 300 instances aligned with the 300 X launch.

We launched our latest generation Ryzen 8000 series notebooks and desktop processors in January including our Ryzen $80 40 mobile series that combined leadership compute performance and energy efficiency with an updated MCU that delivers up to 60% more AI performance compared to our prior generation that was already industry.

Lisa T. Su: At the same time, our partnership with hugging face the leading open platform for the AI community now enabled hundreds of thousands of AI models to run out of the box on AMD Gpus, and we're extending that collaboration to our other platforms.

<unk>.

ASER, a sous HP Lenovo MSI and other large PC Oems will offer notebooks powered by our Ryzen 8000 series processors with the first system is expected to go on sale in February.

Lisa T. Su: Looking ahead, our prior guidance was for datacenter GPU revenue to be flattish from Q4 to Q1 and exceed $2 billion for 2024.

Lisa T. Su: Based on the strong customer pull and expanded engagements. We now expect datacenter GPU revenue to grow sequentially in the first quarter and exceed $3 5 billion in 2024.

To further our leadership in AI Pcs, we launched our Ryzen 8000 G series processors earlier this month, which is the industry's first desktop Cpus with an integrated AI engine.

Lisa T. Su: We have also made significant progress with our supply chain partners and have secured additional capacity to support upside demand.

Millions of AIP CES powered by Ryzen processors have shipped to date and ryzen Cpus power more than 90% of AI enabled Pcs currently in market.

Lisa T. Su: Turning to our client segment revenue was $1 5 billion, an increase of 62% year over year and flat sequentially.

Our work with Microsoft and our PC ecosystem partners to enable the next generation of <unk> expanded significantly in the quarter.

Lisa T. Su: We launched our latest generation Ryzen 8000 series notebooks and desktop processors in January including our rising $80 40 mobile series that combined leadership compute performance and energy efficiency with an updated MCU that delivers up to 60% more AI performance compared to our prior generation that was already industry.

We are aggressively driving our ryzen AICPA roadmap to extend our AI leadership, including our next Gen. Strict processors that are expected to deliver more than three X. The AI performance of our ryzen $7 40 series processors.

<unk> combines our next Gen Z five core with enhanced rdna graphics, and an updated ryzen AI engine to significantly increase the performance energy efficiency and AI capabilities of Pcs.

Lisa T. Su: <unk>.

Lisa T. Su: ASER, a sous HP Lenovo MSI and other large PC Oems will offer notebooks powered by our Ryzen 8000 series processors with the first system is expected to go on sale in February.

Customer momentum for <unk> is strong with the first notebooks on track to launch later this year.

Lisa T. Su: To further our leadership in AI Pcs, we launched our Ryzen 8000 G series processors earlier this month, which is the industry's first desktop Cpus within integrated AI engine.

Looking at 2024, we are planning for the PC Tam to grow modestly year on year weighted towards the second half as AI Pc's ramp.

We continue to see strong growth opportunities for our client business as we ramp our current products extend our AIP see leadership and launch our next wave of <unk> Cpus.

Lisa T. Su: Millions of AIP CES powered by Ryzen processors have shipped to date and ryzen Cpus power more than 90% of AI enabled Pcs currently in market.

Now turning to our gaming segment revenue declined 17% year over year, and 9% sequentially to $1 4 billion as lower semi custom revenue was partially offset by increased sales of Radeon Gpus.

Lisa T. Su: Our work with Microsoft and our PC ecosystem partners to enable the next generation of AI Pcs expanded significantly in the quarter.

Lisa T. Su: We are aggressively driving our ryzen AICPA roadmap to extend our AI leadership, including our next Gen. Strict processors that are expected to deliver more than three X. The AI performance of our ryzen $7 40 series processors.

Semi custom Soc sales declined in line with our projections in the quarter.

Going forward, we now expect annual revenue to decline by a significant double digit percentage year over year as supply caught up with demand in 2023, and we enter the fifth year of what has been a very strong console cycle.

Lisa T. Su: <unk> combines our Nextgen Zen five core with enhanced rdna graphics, and an updated ryzen AI engine to significantly increase the performance energy efficiency and AI capabilities of Pcs.

And gaming graphics revenue grew both year over year and sequentially driven by strong demand in the channel for both our Radeon 6000, and Radeon 7000 series Gpus.

Lisa T. Su: Customer momentum restricts as strong with the first notebooks on track to launch later this year.

Lisa T. Su: Looking at 2024, we are planning for the PC Tam to grow modestly year on year weighted towards the second half as AI Pc's ramp.

We expanded our Radeon 7000, GP Gpus series with the launch of New Rx 7600, XT series enthusiast desktop Gpus earlier. This month, then offer leadership price performance for 10 ADP gaming.

We continue to see strong growth opportunities for our client business as we ramp our current products extend our AIP see leadership and launch our next wave of <unk> Cpus.

We also launched new open source fidelity FX Super resolution three software that can deliver significantly higher gaming frame rates on both Gpus and apu's.

Lisa T. Su: Now turning to our gaming segment revenue declined 17% year over year, and 9% sequentially to $1 4 billion as lower semi custom revenue was partially offset by increased sales of Radeon Gpus.

Turning to our embedded segment revenue decreased 24% year over year, and 15% sequentially to $1 1 billion as customers focused on reducing their inventory levels.

Semi custom Soc sales declined in line with our projections in the quarter.

We expanded our embedded portfolio in the quarter with new leadership solutions for key markets.

Lisa T. Su: Going forward, we now expect annual revenue to decline by a significant double digit percentage year over year as supply caught up with demand in 2023, and we enter the fifth year of what has been a very strong console cycle.

We launched new Versal Prime adaptive ssds for the aerospace test and measurement healthcare and communications markets that deliver industry first support for DDR, five memory and increase DSP capability compared to our prior generation.

Lisa T. Su: And gaming graphics revenue grew both year over year and sequentially driven by strong demand in the channel for both our Radeon 6000, and Radeon 7000 series Gpus.

In automotive, we launched new vessel Soc solutions that bring industry, leading AI compute capabilities and advanced safety and security features to next generation vehicles.

Lisa T. Su: We expanded our Radeon 7000, Gp's GPU series with the launch of New Rx 7600, XT series enthusiast desktop Gpus earlier this month that offer leadership price performance for 10 ADP gaming.

We also launched ryzen embedded processors with unmatched performance and features for industrial automation machine vision robotics and edge server applications.

Lisa T. Su: We also launched new open source fidelity FX Super resolution three software that can deliver significantly higher gaming frame rates on both Gpus in AP use.

Looking at 2024, we expect overall embedded demand will remain soft through the first half of the year as customers continue to focus on normalizing their inventory levels.

Longer term, we're very confident in the growth trajectory of our embedded business are as our expanded product portfolio drove more than $10 billion of design wins in 2023, an increase of more than 25% compared to 2022.

Turning to our embedded segment revenue decreased 24% year over year, and 15% sequentially to $1 1 billion as customers focused on reducing their inventory levels.

We expanded our embedded portfolio in the quarter with new leadership solutions for key markets.

In summary, I am very pleased with our fourth quarter and full year results for 2024, we expect the demand environment to remain mixed with strong growth in our data center and client segments offset by declines in our embedded and gaming segments.

Lisa T. Su: We launched new Versal Prime adaptive ssds for the aerospace test and measurement healthcare and communications markets that deliver industry first support for DDR, five memory and increased DSP capability compared to our prior generation.

Against this backdrop, we believe we will deliver strong annual revenue growth and expand gross margin driven by the strength of our instinct epic and ryzen product portfolios.

Lisa T. Su: In automotive, we launched new vessel Soc solutions that bring industry, leading AI compute capabilities and advanced safety and security features to next generation vehicles.

Taking a step back we believe AI as a once in a generation transition that will reshape virtually every portion of the computing market starting in the data center, and then expanding into Pcs and across multiple embedded markets.

Lisa T. Su: We also launched ryzen embedded processors with unmatched performance and features for industrial automation machine vision robotics and edge server applications.

Lisa T. Su: Looking at 2024, we expect overall embedded demand will remain soft through the first half of the year as customers continue to focus on normalizing their inventory levels.

We have built excellent customer traction based on the strength of our multi year AI hardware and software Roadmaps and we see clear opportunities to drive our next wave of growth as we deliver leadership AI solutions across our portfolio.

Lisa T. Su: Longer term, we're very confident in the growth trajectory of our embedded business are as our expanded product portfolio drove more than $10 billion of design wins in 2023, an increase of more than 25% compared to 2022.

In the data center, we see 2024 as a start of a multiyear AI adoption cycle with the market for data center AI accelerators growing to approximately $400 billion in 2027.

Lisa T. Su: In summary, I'm very pleased with our fourth quarter and full year results.

Customer deployments of our instinct Gpus continues accelerating with <unk> 300, now tracking to be the fastest revenue ramp of any product in our history and positioning us well to capture significant share over the coming years based on the strength of our multi generation instinct GPU roadmap and open source rock on software.

Lisa T. Su: For 2024, we expect the demand environment to remain mixed with strong growth in our data center and client segments offset by declines in our embedded and gaming segments.

Lisa T. Su: Against this backdrop, we believe we will deliver strong annual revenue growth and expand gross margin driven by the strength of our instinct epic and ryzen product portfolios.

<unk>.

In <unk>, we're focused on delivering our long term roadmaps with leadership ryzen AI <unk> capabilities to enable differentiated experiences as Microsoft and our other software partners bring new AI capabilities to Pcs starting later this year.

Lisa T. Su: Taking a step back we believe AI as a once in a generation transition that will reshape virtually every portion of the computing market starting in the data center, and then expanding into Pcs and across multiple embedded markets.

At the same time, we are rapidly driving leadership AI compute capabilities across the full breadth of our embedded product portfolio.

Lisa T. Su: We have built excellent customer traction based on the strength of our multi year AI hardware and software Roadmaps and we see clear opportunities to drive our next wave of growth as we deliver leadership AI solutions across our portfolio.

This is an incredibly exciting time for the industry and even more exciting time for AMD as our leadership IP broad product portfolio and deep customer relationships position us well to deliver significant revenue growth and earnings expansion over the next several years now I'd like to turn the call over to Jim to provide some additional color on our fourth quarter and full year.

Lisa T. Su: In the data center, we see 2024 as a start of a multiyear AI adoption cycle with the market for data center AI accelerators growing to approximately $400 billion in 2027.

Our financial results.

<unk>.

Thank you Lisa and good afternoon, everyone I will start with the reveal file financial results and then provide our current outlook for the first quarter of fiscal 2024.

<unk> executed well in 2023, despite mixed market demand environment Delevering in revenue of $22 7 billion and the earnings per share of $2 65.

Lisa T. Su: <unk>.

Lisa T. Su: In Pcs, we're focused on delivering our long term roadmaps with leadership ryzen AI <unk> capabilities to enable differentiated experiences as Microsoft and our other software partners bring new AI capabilities to PC. Starting later this year.

We drove year over year revenue growth in our embedded and data center segments. In addition, we successfully launched our AMD <unk> I might be Henry the Gpus.

Lisa T. Su: At the same time, we are rapidly driving leadership AI compute capabilities across the full breadth of our embedded product portfolio.

<unk> asked if our strong ramp in 2024 in the AI market.

Lisa T. Su: This is an incredibly exciting time for the industry and even more exciting time for AMD as our leadership IP broad product portfolio and deep customer relationships position us well to deliver significant revenue growth and earnings expansion over the next several years.

For the fourth quarter of 2023 revenue was $6 2 billion growing 10% year over year as the revenue growth in the HSN chair and the client segments was partially offset by the lower revenue in our embedded in the gaming segment.

Lisa T. Su: Now I'd like to turn the call over to Jim to provide some additional color on our fourth quarter and full year financial results.

New was up 6% sequentially, primarily driven by the ramp of AMD <unk> team with Gpus. It crosses several leading customers and higher revenue from <unk> Silva, our processors, partially offset by the decline in embedded in the gaming segment of revenues.

Lisa T. Su: Gene.

Jim: Thank you Lisa and good afternoon, everyone I will start with the review of our financial results and then provide our current outlook for the first quarter of fiscal 2024.

Jim: <unk> executed well in 2023, despite the mixed market demand environment Delevering in revenue of $22 7 billion and the earnings per share of $2 65.

Gross margin was 51% flat year over year with higher revenue contribution from that data center and client segments offset by lower embedded segment revenue.

Jim: We drove year over year revenue growth in our embedded and data center segments. In addition, we successfully launched our AMD <unk> I might be hungry to Gpus positioning us for a strong ramp in 2024 in the AI market.

Operating expenses were $1 7 billion, an increase of 8% year over year as we invest in R&D and marketing activities to support our significant the AI growth opportunity.

Operating income was $1 4 billion, representing a 23% operating margin.

Jim: For the fourth quarter of 2023 revenue was $6 2 billion growing 10% year over year as the revenue growth in the HSN chair and the client segments was partially offset by the lower revenue in our embedded in the gaming segment.

<unk> interest expense and other was the one country at the end of $63 million.

For the fourth quarter of 2023 diluted earning per share was <unk> 77, an increase of 12% year over year.

Jim: Revenue was up 6% sequentially, primarily driven by the ramp of AMD <unk> team to Gpus crosses several leading customers and higher revenue from <unk> Silva, our processors, partially offset by the decline in embedded in the gaming segment revenues.

Now turning to our reportable segments.

Starting with the datacenter segment revenue was $2 3 billion up 38% year over year, and 43% sequentially driven by strong growth of both AMD in steam to GPU and fourth generation AMD <unk> Cpus sales.

Jim: Gross margin was 51% flat year over year with higher revenue contribution from that data center and client segments offset by lower embedded segment revenue.

Data Center segment operating income was $666 million or 29% of revenue compared to 444 million or 27% a year ago.

Jim: Operating expenses were $1 7 billion, an increase of 8% year over year as we invest in R&D and marketing activities to support our significant the AI growth opportunities.

Higher operating income was primarily due to operating leverage driven by higher revenue.

Client segmented revenue with $1 5 billion up 62% year over year, driven by Ryzen 7000. This series of CPE sales.

Jim: Operating income was $1 4 billion, representing a 23% operating margin.

Jim: Taxes interest expense and other was one country at the end of 63 million.

<unk> segment operating income was $55 million of 4% of revenue compared to an operating loss of 152 million a year ago driven by higher revenue.

Jim: For the fourth quarter of 2023 diluted earning per share was <unk> 77, an increase of 12% year over year.

Jim: Now turning to our reportable segments.

Gaming segment revenue was $1 4 billion down 17% year over year, and 9% sequentially due to a decrease in semi customer revenue, partially offset by an increase in Raytheon GPU sales.

Jim: Starting with the datacenter segment revenue was $2 3 billion up 38% year over year, and 43% sequentially driven by strong growth of both AMD in steam to GPU and fourth generation AMD <unk> Cpus sales.

Gaming segment operating income was 224 million or 16% of revenue compared to $200 to 66 million or 16% a year ago.

Jim: Data Center segment operating income was 666 million or 29% of revenue compared to 444 million or 27% at year ago.

<unk> segment revenue was $1 1 billion down 24% year over year, and 15% sequentially as customers continue to work down their inventory levels.

Jim: Higher operating income was primarily due to operating leverage driven by higher revenue.

Embedded segment operating income was 461 million or 44% of revenue compared to 699 million or 50% a year ago.

Jim: Client segment revenue was $1 5 billion up 62% year over year, driven by ryzen seven and the series of CPE sales.

Jim: Client segment operating income was $55 million of 4% of revenue compared to an operating loss of 152 million a year ago driven by higher revenue.

Turning to the balance sheet and the cash flow during the quarter, we generated 381 million in cash from operations and our free cash flow was $242 million.

Jim: Gaming segment revenue was $1 4 billion down 17% year over year, and 9% sequentially due to a decrease in semi customer revenue, partially offset by increase in radeon GPU sales.

Inventory decreased sequentially by 94 million in Q4 4 billion.

At the end of the quality of our cash cash equivalents and short term investments was strong at $5 8 billion.

In the fourth quarter, we repurchased two 2 million shares and returned 233 million to shareholders for.

Gaming segment operating income was 224 million or 16% of revenue compared to 266 million or 16% a year ago.

For the year, we repurchased 10 million shares and returned $985 million to shareholders.

Jim: <unk> segment revenue was $1 1 billion down 24% year over year, and 15% sequentially as customers continue to work down their inventory levels.

We have $5 6 billion in remaining share repurchase authorization.

Now turning to our first quarter of 2020 for outlook.

Embedded segment operating income was 461 million or 44% of revenue compared to 699 million or 50% a year ago.

We expect revenue to be approximately $5 4 billion, plus or minus $300 million sequentially.

Sequentially, we expect datacenter segment revenue to be flat.

The seasonal decline in silver sales offset by strong data center GPU ramp.

Jim: Turning to the balance sheet and cash flow during the quarter, we generated 381 million in cash from operations and free cash flow was $242 million.

Embedded revenue to decline as customers to continue to work down their inventory levels.

Client segment revenue to decline seasonally in the in the gaming segment and Tim.

Jim: Inventory decreased sequentially by $94 million to $4 4 billion.

We entered the fifth year of what has been a very strong gaming cycle and given current customer inventory levels, we expect revenue to decline by significant double digit percentage.

Jim: At the end of the quality of our cash cash equivalents and short term investment was strong at $5 8 billion.

Jim: In the fourth quarter, we repurchased two 2 million shares and returned 233 million to shareholders for.

Year over year, we expect datacenter and the cloud and the segment revenues to increase by strong double digit percentage given the strengths of our product portfolio and the share gain opportunities.

Jim: For the year, we repurchased 10 million shares and returned $985 million to shareholders.

Jim: We have $5 6 billion in remaining share repurchase authorization.

In <unk> segment to decline in the gaming segment revenue to decline by significant double digit percentage.

Speaker Change: Now turning to our first quarter of 2020 for outlook.

In addition, we expect our first quarter non-GAAP gross margin to be approximately 52% non.

Speaker Change: We expect revenue to be approximately $5 4 billion plus or minus $300 million.

Speaker Change: Sequentially, we expect data center segment revenue to be flat.

non-GAAP operating expenses to be approximately 173 billion.

Speaker Change: The seasonal decline in silver sales offset by strong data center GPU ramp.

non-GAAP effective tax rate to be 13% and a diluted share count is expected to be approximately $1 six 3 billion shares.

Speaker Change: Embedded revenue to decline as customers continue to work down their inventory levels.

Sure.

Speaker Change: <unk> segment revenue to decline seasonally and the in the game in Sacramento as we enter the fifth year of what has been a very strong gaming cycle and given current customer inventory levels, we expect revenue to decline by significant double digit percentage.

While we're not providing specific full year guidance for 2024, let me provide some color directionally for the year. We expect 2020 for data center and client segment revenue to increase driven by the strength of our product portfolio and the share gain opportunity in.

Speaker Change: Year over year, we expect datacenter and the client the segment revenues to increase by strong double digit percentage given the strengths of our product portfolio and the share gain opportunities.

<unk> segment revenue to decline in the gaming segment revenue to decline by significant double digit percentage.

We expect to expand gross margin in 2024 and to continue to invest to address the large AI opportunities, while driving operating model leverage to delever earnings per share growth faster than top line revenue growth.

Speaker Change: <unk> segment to decline in the gaming segment revenue to decline by significant double digit percentages.

Speaker Change: In addition, we expect our first quarter non-GAAP gross margin to be approximately 52% non.

In closing, we delivered a solid financial results in 2023 further strengthening our product portfolio and establishing ourselves as the leading provider of a datacenter gpus for AI.

Speaker Change: non-GAAP operating expenses to be approximately 173 billion.

Speaker Change: non-GAAP effective tax rate to be 13% and a diluted share count is expect to be approximately 163 billion shares.

Very well positioned to build on this momentum and deliver strong financial performance in 2024 and beyond.

Speaker Change: Yeah.

Speaker Change: While we're not providing specific full year guidance for 2024, let me provide some color directionally for the year, we expected 2020 for data center and the client segment revenue to increase driven by the strength of our product portfolio and the share gain opportunities in.

With that I'll turn it back to Mitch for the Q&A session.

Thank you Jane John we're happy to poll the audience for questions.

Thank you Mitch we will now be <unk>.

A question and answer session.

To ask a question. Please press star one on your telephone keypad, a confirmation tone will indicate that your line is in the queue.

Speaker Change: <unk> segment revenue to decline in the gaming segment revenue to decline by significant double digit percentage.

First start to if you'd like to remove a question from the queue.

For participants using speaker equipment, it may be necessary to pick up your handset before pressing the star keys.

Speaker Change: We expect to expand gross margin in 2024 and continue to invest to address the large AI opportunities, while driving operating model leverage to delever earnings per share growth faster than top line revenue growth.

Ask that you please limit yourself to one question and one follow up thank you.

One moment, please while we poll for questions.

And the first question comes from the line of Aaron Rakers from Wells Fargo. Please proceed with your question.

Speaker Change: In closing, we delivered a solid financial results in 2023 further strengthening our product portfolio and establishing ourselves as the leading provider of a datacenter gpus for AI.

Yes, thanks for taking the question.

Just kind of framing the outlook and the guidance for this calendar first quarter I guess the first question is can you help us on a relative basis, the $400 million of datacenter GPU revenue that you expected in Q4.

Speaker Change: We're very well positioned to build on this momentum and deliver strong financial performance in 2024 and beyond.

That ultimately kind of fill out to be and then on the guidance into <unk> can you help us appreciate what seasonal.

Speaker Change: With that I'll turn it back to Mitch for the Q&A session.

Mitch: Thank you Jane John we're happy to pull the audience for questions.

Is defined as as we think about the server business.

Into the <unk> guide.

Jane: Thank you Mitch we will now be <unk>.

Sure. Let me start and then <unk> has something to add so relative to the datacenter GPU business look we were very pleased with performance that we saw in the fourth quarter.

Jane: A question and answer session, if you'd like to ask a question. Please press star one on your telephone keypad, a confirmation tone will indicate that your line is in the queue. You May press star two if you'd like to remove a question from the queue.

There's always going to be a very.

Jane: For participants using speaker equipment, it may be necessary to pick up your handset before pressing the star keys.

Sort of back end quarter, our weighted as we were ramping the product and we saw 300 days.

Jane: Ask that you please limit yourself to one question and one follow up thank you.

Our HPV product actually ramp very well and then we saw 300 X.

Jane: One moment, please where we pull for questions.

The AI product actually exceed our expectations based on strong customer demand the way the qualifications wins and then the ramp manufacturer. So we were over $400 million.

Jane: And the first question comes from the line of Aaron Rakers from Wells Fargo. Please proceed with your question.

Aaron Rakers: Yes, thanks for taking the question.

Aaron Rakers: Just kind of framing the outlook and the guidance for this calendar first quarter I guess the first question is can you help us on a relative basis to $400 million of datacenter GPU revenue that you expected in Q4.

For that business in the fourth quarter, and then going into the first quarter as we look at the business server seasonality.

Call it something around let's call it high single digit low double digit there also some other pieces of the data center business I think the key piece of it is.

Aaron Rakers: That ultimately kind of fell out to be and then on the guidance into <unk> can you help us appreciate what seasonal.

We had originally expected.

<unk> to be a little bit more shallow of Rmi 300 X and what we're seeing now is the supply chain is operating really well and the customer demand is strong and so we will see.

Aaron Rakers: Is defined as as we think about the server business.

Aaron Rakers: Into the <unk> guide.

Speaker Change: Sure Erinn, let me start and then <unk> has something to add so relative to the datacenter GPU business look we were very pleased with performance that we saw in the fourth quarter.

<unk> 300 ex increase as we go into the first quarter and.

Things are going relatively well so yes.

Speaker Change: There's always going to be a very.

Sort of back end quarter, our weighted as we were ramping the product and we saw.

Give you some color Eric I'll give you some color about.

Clay on the seasonality and others. So clay on these very similar to silver typically Q1 is high single digits to low double digit.

Speaker Change: 300 days.

Speaker Change: Our HPV product actually ramp very well and then we saw 300 X.

Speaker Change: The AI product actually exceed our expectations based on strong customer demand the way the qualifications wins and then the ramp manufacturing ramps. So we were over $400 million.

That's the consistent davita posture on the embedded aside it's very consistent with what we said in the pasture and consistent with what you see in the industry.

Speaker Change: For that business in the fourth quarter, and then going into the first quarter as we look at the business server seasonality.

Embedded business is going through a bottoming process and we think of Q1, it will have a low double digit sequential decline.

Speaker Change: Call it something around let's call it high single digit low double digit there are also some other pieces of the data center business I think the key piece of it is.

That's impacted.

The gaming side Lisa mentioned during her.

Speaker Change: We had originally expected the ramp to be a little bit more shallow of Rmi 300 X and what we're seeing now is the supply chain is operating really well and the customer demand is strong and so we will see.

Prepared remarks.

We have the late stage of a product cycle in the year five.

Gaming consoles, but at the same time, we also have inventory at our customers. So the combination of those the impact we expect the Q1 gaming sequential declines probably more than 30%. So hopefully that help you a little bit.

Speaker Change: 300, <unk> increase as we go into the first quarter and.

Speaker Change: Things are going relatively well so yes.

Speaker Change: Give you some color Eric I'll give you some color about.

Yes, very helpful gene and as a quick follow up I'm just curious.

Speaker Change: Clay on the seasonality and others. So client on these very similar to silver typically Q1 is high single digit to low double digit.

Traditional server demand that you see I know when we look at server CPU.

Rents are down north of 20% year over year.

Speaker Change: That's the consistent Davita posture.

Are you seeing any signals or how are you thinking about a recovery in that traditional call. It non AI general purpose server market as we move through 2004.

The embedded aside it's very consistent with what we said in the pasture and consistent with what you see in the industry.

Sure Aaron So look I think.

Speaker Change: Embedded business is going through a bottoming process and we think of Q1, it will have a low double digit sequential decline.

I agree with your.

Characterization of the 2023 on demand, although we did see.

Some strong progress in the second half of the year, especially as customers in cloud and enterprise adopted our general and our Zen four family so going into 2020 for I would say the traditional server market is probably still mixed especially into the first half of the year Theres still some cloud optimization going on as well as sort of.

Speaker Change: That's impacted.

Speaker Change: On the gaming side, Lisa mentioned during her.

Speaker Change: Prepared remarks.

Speaker Change: We have the late stage of a product cycle in the year five.

Speaker Change: Gaming console, but at the same time, we also have an inventory at our customers. So the combination of those the impact we expected that Q1 gaming sequential declines probably more than 30%. So hopefully that helps you a little bit.

Enterprise.

Being a little bit cautious that being the case, though we also see opportunities for us to continue to grow share.

In the traditional server business I think our portfolio is extremely strong.

Speaker Change: Yes, very helpful gene and as a quick follow up I'm just curious.

Adoption of general and Bergamo, as well as our new <unk> product lines.

Speaker Change: Traditional server demand that you see I know when we look at server CPU.

We're getting a lot of <unk>.

And then we also see turn our Gen five product coming in the second half of the year. So even in a mixed demand environment I think we're bullish on what traditional server Cpus can do in 2024.

Speaker Change: Shipments were down north of 20% year over year.

Speaker Change: Are you seeing any signals or how are you thinking about a recovery in that traditional call. It non AI general purpose server market as we move through 2000 and for.

Speaker Change: Sure Aaron So look I think.

And the next question comes from the line of Timothy Arcuri with UBS. Please proceed with your question.

I agree with your.

Aaron Rakers: The characterization of the 2023 on demand, although we did see.

Thanks, a lot. So I'm wondering if you can give us a little bit of sense in terms of the mileposts that you are kind of marching toward on this $400 billion Tam that you have for 2027.

For example, do you think you can gain share at a rate that's kind of similar to the rate that you gained share for server CPU.

Or I guess, maybe asked a different way is it reasonable to kind of look at your consumer GPU share of 20 plus percent is that a reasonable bogey or do you have aspirations higher than that perhaps.

Aaron Rakers: Enterprise.

Aaron Rakers: Being a little bit cautious that being the case, though we also see opportunities for us to continue to grow share.

Yes, Thanks, Tim for the question I would say a couple of things first of all we're <unk>.

Aaron Rakers: In the traditional server business I think our portfolio is extremely strong.

Really pleased with the progress that we've made in our data center GPU business.

Adoption of general and Bergamo, as well as our new <unk> product lines.

Think the ramp that we've seen the customer traction that we've seen even in the last few months I think has has been great.

Aaron Rakers: We're getting.

And then we also see turn in our Gen five product coming in the second half of the year. So even in a mixed demand environment I think we're bullish on what traditional server Cpus can do in 2024.

That gives us a lot of confidence in the ramp of this business I think the beauty of the AI market here is growing so quickly that I think we have both the market dynamic as well as our ability to gain share.

Aaron Rakers: And the next question comes from the line of Timothy Arcuri with UBS. Please proceed with your question.

In that framework the <unk>.

I will make is our customer engagements right now.

Timothy Michael Arcuri: Thanks, a lot. So I'm wondering if you can give us a little bit of sense in terms of the mileposts that you are kind of marching toward on this $400 billion Tam that you have for 2027.

All quite strategic.

Dozens of customers with multi generational conversations so as excited as we are about the ramp of <unk> 300, and frankly, there is a lot to do.

Timothy Michael Arcuri: For example, do you think you can gain share at a rate that's kind of similar to the rate that you gained share for server CPU.

In 2024, we're also very excited about the opportunities to extend that into the next couple of years out into that 'twenty five 'twenty six 'twenty seven timeframe. So I think we see a lot of growth I think it's a little early to make a market share projections, but I would say, it's a significant growth driver given the market demand.

Timothy Michael Arcuri: Or I guess, maybe asked a different way is it reasonable to kind of look at your consumer GPU share of 20 plus percent is that a reasonable bogey or do you have aspirations higher than that perhaps.

Speaker Change: Yes, Thanks, Tim for the question I would say a couple of things first of all we're really pleased with the progress that we've made in our data center GPU business.

As well as our own product capabilities.

Thanks, a lot Jean I guess.

Speaker Change: Think the ramp that we've seen the customer traction that we've seen even in the last few months I think has has been great and that gives us a lot of confidence in the ramp of this business I think the beauty of the AI market here is growing so quickly that I think we have both the market dynamic as well as our ability to gain share.

Follow up I know that you don't want to guide.

Full year, but I'm wondering if I can pin you down just to touch on maybe.

Mileposts you are kind of.

In March and two for 2020 for growth is up 20% for the whole company is that a reasonable target and then I guess within data center. If you just add the Incrementals datacenter GPU revenue and you assume that the server business grows a little bit it seems like that should maybe double year over year, but I'm kind of wondering if you could give us any ranges on those numbers. Thanks.

Speaker Change: In that framework the point I'll make is our customer engagements right now are all quite strategic dozens of customers with multi generational conversations. So as excited as we are about the ramp of <unk> 300, and frankly, there's a lot to do.

Hi, Tim Thank you for the question yes.

Yes, we're not guiding a year.

Very early of the year literally January.

Speaker Change: In 2024, we're also very excited about the opportunities to extend that into the next couple of years out into that 'twenty five 'twenty six 'twenty seven timeframe. So I think we see a lot of growth I think it's a little early to make a market share projections, but I would say, it's a significant growth driver given the market demand.

I think the way to think about it Lisa mentioned that during her prepared remarks, we feel pretty good about both our data center and client businesses to grow in 2020 for of course, the largest incremental revenue opportunities that are going to come from datacenter.

Speaker Change: As well as our own product capabilities.

And both the silver side again, your mall share and datacenter GPU side.

Thanks, a lot Jean I guess.

Speaker Change: Follow up I know that you don't want to guide the full year, but I'm wondering if I can pin you down just to touch on maybe.

Significantly ramp up of our <unk> hundred I think.

That's how we think about it we do have a headwind from <unk>.

Speaker Change: Maybe.

Speaker Change: Mileposts that you are kind of marching to for 2020 for growth is up 20% for the whole company is that a reasonable target and then I guess within data center. If you just add the Incrementals datacenter GPU revenue and you assume that the server business grows a little bit it seems like that should maybe double year over year, but I'm kind of wondering if you could give us any ranges on those.

Gaming segment, we do think of year over year, we will see very significant.

Double digit decline on the gaming segment and at the same time embedded is going through the bottoming process. We do think the second half where we'll see the recovery. So those are the puts and takes.

Speaker Change: Thanks.

I can talk about it.

Speaker Change: Hi, Tim and thank you for the question yes.

Yes, we're not.

And the next question comes from the line of Matt Ramsay with TD Cowen. Please proceed with your question.

Speaker Change: <unk> a year.

Tim: Very early of the year literally January.

Thank you very much good afternoon.

Lisa I wanted to ask I mean, there's been so much.

Focus on scrutiny as there should be on the really exciting progression with that money.

300.

I mean, we've progressed over the last six months from.

I think some doubts and the investment community.

Tim: And both the silver side at <unk> and datacenter GPU side.

The software and your ability to ramp the product now.

Tim: Significantly ramp up of our <unk> 300, I think.

Proving that youre ramping it with I think you said thousands of customers across different end markets.

Tim: That's how we think about it we do have a headwind from <unk>.

What I'm interested in hearing a little bit more about and you guys have been open about what.

Tim: Gaming segment, we do think of year over year, we will see very significant double digit decline on the gaming segment and at the same time embedded is going through the bottoming process. We do think the second half where we'll see the recovery. So those are the puts and takes.

Some of the forward programs in your traditional server business look like from a roadmap perspective I'd be interested to hear how you're thinking about the roadmap and your MRI.

To accelerate our family is that correct.

They're going to continue to be parts that our CPU and GPU together is it a primarily a GPU only roadmap what kind of cadence are you thinking about.

Speaker Change: I can talk about it.

Speaker Change: And the next question comes from the line of Matt Ramsay with TD Cowen. Please proceed with your question.

Any kind of color you can give us on some of the forward roadmap trajectory for that program would be really helpful. Thanks.

Matthew D. Ramsay: Thank you very much good afternoon.

Sure Matt So I appreciate the comments I think the.

Matthew D. Ramsay: Lisa I wanted to ask I mean, there's been so much.

The traction that we're getting with the <unk> 300 family is is really strong I think what has benefited us here is our use of triplets technologies, which has given us the ability to have sort of both the the Apu version.

Matthew D. Ramsay: Focus on scrutiny as there should be on us really.

Matthew D. Ramsay: The exciting progression with MRI three.

Matt Ramsay: 300.

Matt Ramsay: I mean, we've progressed over the last six months from.

Matt Ramsay: I think some doubts and the investment community on that.

As well as the GPU version and we continue to use that to differentiate ourselves and that's how we get our memory bandwidth in memory capacity advantages as we go forward you can imagine like we did in the epic timeframe. We planned multiple generations in sequence. That's the way we're planning the roadmap one of the things I.

Matt Ramsay: Software and your ability to ramp the product now.

Matt Ramsay: Proving that youre ramping it with I think you said thousands of customers across different end markets.

Matt Ramsay: What I'm interested in hearing a little bit more about and you guys have been open about what.

Matt Ramsay: Some of the forward programs in your traditional server business look like from a roadmap perspective I'd be interested to hear how you're thinking about the roadmap and your M. I.

You'll note about the AI accelerator market is the demand for compute is so high that we are seeing.

Accelerator family.

And acceleration of the roadmap.

Matt Ramsay: They're going to continue to be parts that our CPU and GPU together or is it a primarily a GPU only roadmap what kind of cadence are you thinking about.

Generation here and we are similarly planning.

Acceleration of our roadmap I would say that we will talk more about the overall roadmap beyond them by 300 as we get into later this year, but you can be assured that we're working very closely with our customers to have a very competitive roadmap for both training and inference.

Matt Ramsay: Just any kind of color you can give us on some of the forward roadmap trajectory for that program would be really helpful. Thanks.

Speaker Change: Yeah sure Matt so.

Speaker Change: I appreciate the comments I think the the <unk>.

Speaker Change: Traction that we're getting with the <unk> 300 family is is really strong I think what's benefited us here is our use of triplets technologies, which has given us the ability to have sort of both the the Apu version.

That.

We will come out over the next couple of years.

Thank you for that Lisa just as a.

A follow up.

I guess one of the questions that I have.

And getting a lot in different forums.

Speaker Change: As well as the GPU version and we continue to use that to differentiate ourselves and that's how we get our memory bandwidth in memory capacity.

With respect to the $400 billion Tam that you guys have laid out for 2027.

Maybe you could give us.

Speaker Change: <unk> as we go forward you can imagine like we did in the epic timeframe. We planned multiple generations in sequence. That's the way we're planning the roadmap one of the things I will note about the AI accelerator market is the demand for compute is so high that we are seeing.

Look under the Hood.

I guess.

We've got 100 versions of the same question, which is how that did you come up with that number.

So if you could give us a little bit more in terms of are we talking about system and accelerator cards are we talking about just the silicon or are we talking about full servers.

Speaker Change: Sort of an acceleration of the roadmap generations here and we are similarly planning.

And what kind of.

Unit assumptions any kind of thing that you can give us on market sizing or what gives you the visibility so early into this general trend.

Speaker Change: Acceleration of our roadmap I would say that we will talk more about the overall roadmap beyond them by 300 as we get into later this year, but you can be assured that we're working very closely with our customers to have a very competitive roadmap for both training and inference.

To give a precise number of three years out.

That would be really really helpful. Thank you.

Well, Matt I don't know how precise it is but I think we said approximately $400 billion, but I think what we need to look at is growth rate and how do we get to those growth rates.

Speaker Change: That.

Speaker Change: We will come out over the next couple of years.

I think we expect units to grow sort of substantial double digit percentage, but you should also expect that content is going to grow. So if you think about how important memory and memory capacity is.

Speaker Change: Thank you for that Lisa just as a follow up.

Speaker Change: I guess one of the questions that I have.

Speaker Change: <unk> been getting a lot in different forums.

With respect to the 400 billion dollar Tam that you guys have laid out for 2027.

As we as we go forward you can imagine that we'll see.

<unk>, there and just the overall content as we go to more advanced technology nodes. So there's some ASP uplift in there and then what we also do is we are planning longer term roadmaps with our customers in terms of how they're thinking about.

Speaker Change: Maybe you could give us.

Speaker Change: A little look under the Hood.

Speaker Change: I guess.

Speaker Change: We got 100 versus at the same question, which is how that did you come up with that number.

Speaker Change: So if you could give us a little bit more in terms of are we talking about system and accelerator cards are we talking about just the silicon or are we talking about full servers.

Sort of the size of training clusters, the number of training clusters.

Speaker Change: And what kind of sort of unit assumptions any kind of thing that you can give us on market sizing or what gives you the visibility so early into this general trend.

Then the fact that we believe influence is actually going to exceed training as we go into the next couple of years just given.

As more enterprises adopt so I think as we look at all of those pieces.

Speaker Change: To give a precise number of three years out.

Speaker Change: That would be really really helpful. Thank you.

I think we feel good that the growth rate is going to be significant and sustained over the next few years in terms of what's in that Tam.

Speaker Change: Sure, Matt I don't know how precise it is but I think we said approximately $400 billion, but I think what we need to look at is growth rate and how do we get to those growth rates.

It really is accelerator town so within accelerators.

Matthew D. Ramsay: I think we expect units to grow sort of substantial double digit percentage, but you should also expect that content is going to grow. So if you think about how important memory and memory capacity is.

There are certainly Gpus and there will also be some asics.

That our other accelerators that are in that Tim as we think about sort of the different types of models from smaller models to fine tuning of models to the largest large language models I think youre going to need different silicon for those different use cases, but from our.

Matthew D. Ramsay: As we as we go forward you can imagine that we'll see.

Matthew D. Ramsay: Acceleration, there and just the overall content as we go to more advanced technology nodes. So there's some ASP uplift in there and then what we also do is we are planning longer term roadmaps with our customers in terms of how they're thinking about.

Endpoint Gpus are still going to be the sort.

Sort of the compute element of choice when Youre talking about training and inferencing on the largest language models.

Matthew D. Ramsay: The size of training clusters, the number of training clusters, and then the fact that we believe influence is actually going to exceed training.

And the next question comes from the line of Joe Moore with Morgan Stanley. Please proceed with your question.

Matthew D. Ramsay: As we go into the next couple of years just given.

Great. Thank you I think you talked about the 300 cloud workloads being kind of split between the more customer facing workloads in versus your internal can you talk about how you see the breakdown of that and how.

Matthew D. Ramsay: As more enterprises adopt so I think as we look at all of those pieces.

Matthew D. Ramsay: I think we feel good that the growth rate is going to be significant and sustained over the next few years in terms of what's in that Tam.

How is your ecosystem progressing this is a brand new chip.

Matthew D. Ramsay: It really is accelerator town so within accelerators.

It seems impressive that you're able to support kind of a broad range of of customer facing workloads and cloud.

Matthew D. Ramsay: There are certainly Gpus and there will also be some asics.

Yes sure Joe So yes look we are really happy with how the mid 300 has come up and we've now deployed.

Matthew D. Ramsay: That our other accelerators that are.

Matthew D. Ramsay: In that time.

Matthew D. Ramsay: As we think about sort of the different types of models from smaller models to fine tuning of models too.

And working with a number of customers. What we have seen is certainly <unk> has been a very important.

Matthew D. Ramsay: The largest large language models, I think youre going to need different silicon for those.

As well as the direct optimization with our top cloud customers. We always said that the best way of optimizing the software is working directly on the most important workloads and we've seen performance come up nicely, which is what we expect frankly with.

Matthew D. Ramsay: Use cases, but from our standpoint Gpus are still going to be the.

Matthew D. Ramsay: Sort of the compute element of choice when Youre talking about training and inferencing on the largest language models.

Matthew D. Ramsay: And the next question comes from the line of Joe Moore with Morgan Stanley. Please proceed with your question.

The GPU capabilities that we would have to do some level of optimization, but the optimization has gone well I think to your broader question.

Great. Thank you I think you talked about the 300 cloud workloads being kind of split between the more customer facing workloads and versus your internal can you talk about how you see the breakdown of that and how.

I look at this is there are lots of opportunities for us to work directly with large customers. Both on the cloud side as well as on the enterprise side, who have specific training and inferencing workloads, our job is to make it as easy as possible for them and so our entire tool.

Joe L. Moore: How is your ecosystem progressing this is a brand new chip.

It seems impressive that you're able to support kind of a broad range of of customer facing workloads and cloud.

Chain.

Speaker Change: Yes sure Joe So yes look we are really happy with how the 300 has come up and we've now deployed.

All of our sort of the overall Rockin suite has really gone through.

Significant progress.

Over the last six to nine months and then we're also getting some nice support from the open source community. So the work that hugging faces doing.

And working with a number of customers. What we have seen is certainly Rocco six has been a very important.

Is tremendous in terms of just real time optimization.

Joe L. Moore: As well as the direct optimization with our top cloud customers. We always said that the best way of optimizing the software is working directly on the most important workloads and we've seen performance come up nicely, which is what we expect frankly with.

On our hardware.

Our partnership with open AI on Triton.

And our work across.

A number of these.

These open source model has helped us actually make very rapid progress.

Joe L. Moore: The GPU capabilities that we would have to do some level of optimization, but the optimization has gone well I think to your broader question.

Great for me.

A follow up I guess, a lot of the forecasting of your business that I'm hearing is coming from supply chain and we're sort of hearing Amd's building action in Asia.

Joe L. Moore: The way I look at this is there are lots of opportunities for us to work directly with large customers. Both on the cloud side as well as on the enterprise side, who have specific training and inferencing workloads, our job is to make it as easy as possible for them and so our entire tool.

Yes.

How would you ask us to think about that are you looking at being kind of sold out for the year and so the supply chain would be closer to revenue are you building for the best case scenario.

I worry about sometimes expectations when people hear the supply chain numbers and I'm just curious how do you bridge the gap.

Joe L. Moore: Chain.

Yes, so Joe I think we updated our revenue expectations.

Joe L. Moore: All of our sort of the overall rosin suite has really gone through.

This quarter from our original number of 2 billion to $3 5 billion to try to give.

Joe L. Moore: Significant progress.

Over the last six to nine months and then we're also getting some nice support from the open source community. So the work that hugging faces doing.

Some some.

Bounding on some of the discussion out there the way to think about the $3 5 billion is these are customers that were working with.

Is tremendous in terms of just real time optimization.

Joe L. Moore: On our hardware.

Who have given us firm commitments on what they need as you know the lead times on these products are quite long so it's important.

Joe L. Moore: Our partnership with open AI on Triton.

Joe L. Moore: And our work across.

A number of.

Joe L. Moore: These.

To have those forecasts in early and we have a strong order book so that gives us.

Joe L. Moore: These open source model has helped us actually make very rapid progress.

Good confidence to exceed the $3 5 billion.

Great: Great and for my follow up I guess, a lot of the forecasting of your business that I'm hearing is coming from supply chain and we're sort of hearing Amd's building action in Asia.

From a supply chain standpoint, our goal is always to build more supply.

And so from that standpoint, we have also worked with our supply chain partners and secured significant capacity I think about it as first half capacity is tight.

Great: I guess.

Speaker Change: How would you ask us to think about that.

Speaker Change: You <unk>.

Speaker Change: Looking at being kind of sold out for the year and so the supply chain would be close to revenue are you building for the best case scenario.

And more comes on in the second half of the year, but we've certainly made more progress there so.

Speaker Change: Worry about sometimes expectations when people hear the supply chain numbers and I'm just curious how you bridge the gap.

We do have more supply and we're going to continue to work with our customers on their deployments and we will update that number as we go through the year.

Speaker Change: Yes, so Joe I think we updated our revenue expectations.

And the next question comes from the line of <unk> Hari with Goldman Sachs. Please proceed with your question.

This quarter from our original number of 2 billion to $3 5 billion to try to give.

Speaker Change: Some.

Hi, Thank you for taking the question I had one on the 300 as well Lisa.

Joe L. Moore: Some bounding on some of the discussion out there.

The way to think about the $3 5 billion is these are customers that were working with.

I guess first of all how should we think about the quarterly trajectory beyond Q1, you talked about Q1 being up sequentially.

Joe L. Moore: You have given us firm commitments on what they need as you know the lead times on these products are quite long so it's important to.

Is it fair to assume kind of a straight line as we progress through the balance of the year or is it more second half skewed how should we think about that and I guess more importantly, some of the cloud potential customers that have yet to fulfill.

To have those forecasts in early and we have a strong order book so that gives us.

Joe L. Moore: Good confidence to exceed the $3 5 billion.

Joe L. Moore: I'm a supply chain standpoint, our goal is always to build more supply.

Sign up for sign off on the <unk> 300, I guess, what's the sticking point is just a function of time and you just need a little bit more time to go back and forth and tweak things are.

Joe L. Moore: And so from that standpoint.

Joe L. Moore: We have also worked with our supply chain partners and secured a significant capacity I think about it as first half capacity is tight.

Is there a software kind of concern.

I guess, what's holding them back at this point, yes, yes sure. Thanks for the question So first on the.

Joe L. Moore: And more comes on in the second half of the year, but we've certainly made more progress there so.

Joe L. Moore: We do have more supply and we're going to continue to work with our customers on their deployments and we'll update that number as we go through the year.

300 trajectory I think you would expect that revenue should increase every quarter from now through sort of the end of the year, but it will be a bit more second half weighted and part of that is just.

Joe L. Moore: And the next question comes from the line of Toshi Hari with Goldman Sachs. Please proceed with your question.

Customers as they're finishing up their qualifications in their lines as well as sort of how our supply.

Toshiya Hari: Hi, Thank you for taking the question I had one on the 300 as well Lisa.

Toshiya Hari: I guess first of all how should we think about the quarterly trajectory beyond Q1, you talked about Q1 being up sequentially.

Chain is ramping so yes, it should increase each quarter, but be a bit more second half weighted and then to your comment about customers look we're engaged with.

Toshiya Hari: Is it fair to assume kind of a straight line as we progress through the balance of the year or is it more second half skewed how should we think about that and I guess more importantly, some of the cloud potential customers that have yet to fulfill.

All of sort of the large customers.

These are all folks that know us really well given our deep relationships in epic.

I think people just have different adoption cycles as they consider what they're trying to do in their roadmap, but I view this as still very very early innings for us.

Toshiya Hari: <unk> signed up for sign off on the <unk> 300, I guess, what's the sticking point is just a function of time and you just need a little bit more time to go back and forth and tweak things are.

In this space and I think a question was asked earlier I think the key is this is not just about <unk> 300 conversation, but it really is about sort of our long term multi generational roadmap and so that's the context on which we're working with our largest customers as well as.

Toshiya Hari: Is there a software kind of concern.

Toshiya Hari: I guess, what's holding them back at this point yeah, yes. So.

Speaker Change: Sure. Thanks for the question so first on the.

Speaker Change: 300 trajectory I think you would expect that revenue should increase every quarter from now through sort of the end of the year, but it will be a bit more second half weighted and part of that is just.

As you know Theres a lot of demand coming from.

Folks that are more AI centric and not necessarily typical cloud customers, but more enterprise or let's call. It AI specific.

Speaker Change: Customers as they're finishing up their qualifications in their lines as well as sort of how our supply.

Companies that were also very well engaged with.

Got it that's Super helpful. And then as my follow up maybe one for gene on the gross margin side, you're guiding Q1 to 52% I'm curious.

Speaker Change: Chain is ramping so yes, it should increase each quarter, but be a bit more second half weighted and then to your comment about customers look we're engaged with.

Im sure Youre not going to give quantitative guidance beyond Q1, but how to think about.

Speaker Change: All of sort of the large customers.

Speaker Change: These are all folks that know us really well given our deep relationships in epic.

The trajectory for Q2 and beyond.

Pretty sure Youre working through some kinks as it pertains to the.

Speaker Change: I think people just have different adoption cycles as they consider what they're trying to do in their roadmap, but I view this as still very very early innings for us.

The instinct ramp hopefully that improves over time, so that should be a tailwind.

<unk>, perhaps the second half.

Speaker Change: In this space and I think a question was asked earlier I think the key is this is not just about <unk> 300 conversation, but it really is about sort of our long term multi generational roadmap and so that's the context in which we're working with our largest customers as well as.

Turns for the better.

<unk> got some Super server CPU volume growth throughout the year. So it feels like you've got multiple tailwind as we think about gross margin progression on a sequential basis, but what are the potential headwinds as we move throughout 2024. Thank you.

Speaker Change: As you know Theres a lot of demand coming from.

Yes. Thank you for the question, yes, you're absolutely right. We have some puts and takes that impact our gross margin.

Speaker Change: Folks that are more AI centric and not necessarily typical cloud customers, but more enterprise or let's call. It AI specific.

Kind of the Q1 120 basis points higher than Q4 sequentially.

Speaker Change: Companies that were also very well engaged with.

Primarily because of the higher datacenter contribution or actually more than offset the decline of impacted the business in Q1 going forward. The way to think about it is as you said that you see the major driver or is it going to be datacenter business is going to grow much faster than other segment that makes.

Speaker Change: Got it that's Super helpful. And then as my follow up maybe one for gene on the gross margin side, you're guiding Q1 to 52% I'm curious.

Speaker Change: Curious im sure Youre not going to give quantitative guidance beyond Q1, but how to think about.

Gene: The trajectory for Q2 and beyond.

Change it will help us to expand the gross margin nicely I think you also.

Gene: I'm pretty sure you're working through some kinks as it pertains to.

Gene: The instinct ramp hopefully that improves over time, so that should be a tailwind.

Our spot.

On the embedded coming back in second half, which will be a tailwind.

Gene: FPGA is perhaps the second half.

The datacenter GPU.

Gene: Turns for the better.

Gene: Gotcha Super server CPU volume growth throughout the year. So it feels like you've got multiple tailwind as we think about gross margin progression on a sequential basis, but what are the potential headwinds as we move throughout 2024. Thank you.

At the very early stage of ramp we are improving testing timing yield and continue to expand gross margin and we expect that to be accretive to corporate averages. So those are all the tailwind to come in in the second half.

Gene: Yes.

I would say.

Thank you for the question, yes, you're absolutely right. We have some puts and takes that impact our gross margin. We guided the Q1 120 basis points higher than Q4 sequentially.

The hydro Windsor side to continue to be in the first half where we see embedded in the business.

Not only Q1, we see sequential decline Q2, probably are going to be sequentially flattish Q versus Q1 <unk>.

Primarily because of the higher datacenter contribution actually more than offset the decline of embedded business in Q1 going forward. The way to think about it is as you said that you see.

Our hydro wind for us.

Because it does have a very nice gross margin, but overall, we feel pretty good about.

The trajectory over the gross margin improvement.

Gene: The major driver or is it going to be datacenter business is going to grow much faster than other segment that makes change it will help us to expand the gross margin nicely I think you also.

<unk> second half.

And the next question comes from the line of Ross Seymore with Deutsche Bank. Please proceed with your question.

Hi, Thanks for let me ask a question I wanted to get into the competitive environment first on me.

Gene: Spot on on the embedded coming back in second half of which will be a tailwind with the datacenter GPU.

Instinct side of things.

How thats going it doesn't seem to be slowing down your ramp whatsoever. But then also on the street server CPU side of things you said youre gaining share in that area, but as we think about future roadmaps pricing incentives and those sorts of things any meaningful change in the competitive environment that youre seeing through the 2024.

Gene: We are at a very early stage of ramp we are improving testing time yield and continue to expand gross margin and we expect them to be accretive to corporate average. So those are all the tailwind coming in the second half I would say.

Sure.

Gene: Headwinds aside to continue to be in the first half where we see embedded the business.

The environment for US is always competitive so I think that that has not changed.

If I look at the instinct side I think we have I think we've shown that.

Gene: Yes, not only Q1, we see sequential decline Q2, probably are going to be sequentially flattish Q versus Q1 that are.

300, and our roadmap are actually very competitive.

Hi, the window for us.

There are some places where let's call it.

Gene: Because it does have a very nice gross margin, but overall, we feel pretty good about.

It's more even like in the training environment, but as we look at the inferencing environment.

Gene: The trajectory of the gross margin improvement, especially.

We have significant advantages and that's that's showing through in some of our our customer work. So we think for both training and inference, we will continue to be very competitive.

Gene: Especially second half.

Gene: And the next question comes from the line of Ross Seymore with Deutsche Bank. Please proceed with your question.

Then as you go into the CPU side again.

Ross Seymore: Hi, Thanks for let me ask the question I wanted to get into the competitive environment first on me.

From our view with.

Ross Seymore: Instinct side of things.

With each generation of epic we've continued to gain share I think we exited the fourth quarter a record share for.

Ross Seymore: Thats going it doesn't seem to be slowing down your ramp whatsoever, but then also on the street server.

Ross Seymore: <unk> side of things, you said youre gaining share in that area, but as we think about future roadmaps pricing incentives those sorts of things any meaningful change in the competitive environment that you're seeing through 2024.

For AMD.

We are still quite underrepresented in an enterprise. So I think there is an opportunity for us to continue to gain share as we go through 2024 from a competitive standpoint.

What we see is Zen four is extremely competitive right now.

Ross Seymore: Sure.

Hello.

Speaker Change: The environment for US is always competitive so I think that that has not changed.

With Genoa Bergamo Sienna.

And as we go into <unk>.

Speaker Change: If I look at the instinct side I think we have I think we've shown that.

Keep in the design in cycle four zone, five and turn and we feel very good about how we're positioned.

Speaker Change: 300, and our roadmap are actually very competitive.

Speaker Change: There are some places where let's call it.

Thanks for that and I guess as my follow up on the data center side. Another thing Thats been pervasive throughout 2023 at least the GPU side crawling up the CPU side, you mentioned that there is still a little bit.

Speaker Change: It's more even like in the training environment, but as we look at the inferencing environment. We think we have significant advantages and that's that's showing through in some of our our customer work. So we think for both training and inference.

Digestion going on within your epic business, but where do you see that standing I know youre going to gain share et cetera.

Speaker Change: We will continue to be very competitive.

Speaker Change: And then as you go into the CPU side again.

You guys clearly benefit from the <unk> side of the datacenter GPU side, but what about on the CPU side of things is that Edwin now behind us or is that still an issue.

Speaker Change: From our view.

Speaker Change: With each generation of epic, we've continued to gain share I think.

I think we expect the CPU business from a market standpoint to grow Ross as we go into 2024, I think the rate and pace of growth will depend a little bit on the macro and just overall capex trends, but from our standpoint, we are starting to see some of our larger customers plan there.

Speaker Change: The fourth quarter a record share.

Speaker Change: For AMD.

Speaker Change: We're still quite underrepresented in an enterprise. So I think there is an opportunity for us to continue to gain share as we go through 2024 from a competitive standpoint.

Speaker Change: What we see is Zen four is extremely competitive right now.

Refresh cycles, there is a lot of let's call. It older equipment that has yet to be.

Speaker Change: With Genoa Bergamo Sienna.

Speaker Change: And as we go into turn we're deep in the design in cycle four and five in turn and we feel very good about how we're positioned.

Refreshed and the value proposition for refresh is so strong because the energy efficiency and sort of the footprint.

Of the newer generations are so much better than sort of the four or five year old infrastructure that we do see that refresh cycle happening as we get into 2024 I think the exact timing we will have to understand more.

Speaker Change: Thanks for that I guess as my follow up on the data center side. Another theme that's been pervasive throughout 2023 at least the GPU side crowding out the CPU side, you mentioned that there is still a little bit of digestion going on within your epic business, but where do you see that standing I know youre going to gain share et cetera.

As the market evolves as we go through the year.

And the next question comes from the line of Vivek Arya with Bank of America Securities. Please proceed with your question.

Speaker Change: You guys clearly benefit from the <unk> side of the datacenter GPU side, but what about on the CPU side of things is that Edwin now behind us or is that still an issue.

Thanks for taking my questions. So first one.

Visa from you gave us a $2 billion plus number for them I before now you have related to over $3 5 billion and I'm curious what drove the change was it incremental demand signals both the supply end.

Speaker Change: I think we expect the CPU business from a market standpoint to grow Ross as we go into 2024, I think the rate and pace of growth will depend a little bit on the macro and just overall capex.

Can you supply more of let's say demand is four or five or 6 billion what is the.

Speaker Change: Trends, but from our standpoint.

Speaker Change: We are starting to see some.

Limitation and sort of related to that.

Some of our larger customers plan their refresh cycles. There is a lot of let's call. It older equipment that has yet to be.

On the competition side your competitor will launch there will be 100.

Later in the year do you think that has changed the competitive landscape in anyway.

Speaker Change: Refreshed and the value proposition for refresh is so strong because the energy efficiency and sort of the footprint.

Yes, sure Vivek. So I think what we said is as we went from two to $3 5 billion. It really is mostly customer demand signals. So.

Speaker Change: Of the newer generations are so much better than sort of the four or five year old infrastructure that we do see that refresh cycle happening as we get into 2024 I think the exact timing we will have to understand more.

<unk> orders have come on books and as we've seen programs move from let's call. It pilot programs into full manufacturing programs. We have updated the revenue forecast as I said earlier from a supply standpoint, we are planning for success and so we worked closely with our supply chain partners to ensure.

Speaker Change: As the market evolves as we go through the year.

Speaker Change: And the next question comes from the line of Vivek Arya with Bank of America Securities. Please proceed with your question.

Vivek Arya: Thanks for taking my questions. So first one.

Or that we can ship more.

Then $3 5 billion substantially more depending on what customer demand is as we go into the second half of the year and then in terms of again Roadmaps as I said, we're very focused on a competitive roadmap.

Visa from you gave US a 2 billion plus number for them I before now you have related to over $3 5 billion and I'm curious what drove the change was it incremental demand signals both the supply end.

Vivek Arya: Ken can you supply more of let's say demand is four or five or 6 billion what is the.

That sort of what the next generations are beyond my 300, So I do believe that we have a strong roadmap in place and continue to work with our customers too.

Vivek Arya: Limitation and sort of related to that.

Vivek Arya: On the competition side your competitor will launch there'll be 100.

Sort of adopt our roadmap as quickly as possible.

Vivek Arya: Later in the year do you think that has changed the competitive landscape in anyway.

Got it.

Ken: Yes, sure Vivek. So I think what we said is as we went from two to $3 5 billion of it really is mostly customer demand signals. So.

Longer term question, Lisa if I look at the success that <unk> enjoyed its many factors, but a few of them included your early adoption of AAA and the strong partnerships we have had.

Ken: <unk> orders have come on books and as we've seen programs move from let's call. It pilot programs into full manufacturing programs. We have updated the revenue forecast as I said earlier from a supply standpoint, we are planning for success and so we worked closely with our supply chain partners to ensure.

The SMC, but now Youre seeing your X 86 competitor Intel also adopt chipset or biotechnology as they call. It and then I think recently the manufacturing update they gave they said that two years ahead.

In terms of incorporating gate, all around and backside our delivery.

Ken: Or that we can ship more.

So, let's assume there right and they have caught up to TSMC or maybe they got ahead, what impact does that have on AMD and kind of the medium to long ago.

Ken: Then $3 5 billion substantially more depending on what customer demand is as we go into the second half of the year and then in terms of again Roadmaps as I said, we're very focused on a competitive roadmap.

Yeah sure Vivek look we're always looking at what's next right. So on the triplet technology.

On the fourth generation of the triplet technologies I think we've learned a lot about how to optimize our performance. There we are very aggressive with our adoption of leading edge technology.

Ken: That sort of what the next generations are beyond my 300, So I do believe that we have a strong roadmap in place and continue to work with our customers too.

Ken: Sort of adopt our roadmap as quickly as possible.

It's needed, but I think those are only a.

Speaker Change: Got it and.

A few of the pieces. We're also focused on continuing to innovate on architecture and design. So I think the longer term question that you ask is I think we're expecting that.

Speaker Change: Longer term question, Lisa if I look at the success that <unk> enjoyed its many factors, but a few of them included your early adoption of AAA and the strong partnerships we have had.

The competition is going to be on a similar process technology and even in that case.

Speaker Change: The SMC, but now Youre seeing your X 86 competitor Intel also adopt chipset or biotechnology as they call. It and then I think recently the manufacturing update they gave they said that two years ahead.

I think we feel like we have a very strong roadmap going forward and we will continue to.

To drive both the CPU and the GPU roadmap very aggressively.

Speaker Change: In terms of incorporating gate, all around and backside our delivery.

And the next question comes from the line of harsh Kumar with Piper Sandler. Please proceed with your question.

Speaker Change: So, let's assume there right and they have caught up to TSMC or maybe they got ahead, what impact does that have on AMD and kind of the medium to long ago.

Yes.

Thanks for letting me ask a question guys I have two questions, let me start off with.

Speaker Change: Yeah sure Vivek look we're always looking at what's next right. So on the triplet technology I mean, we're sort of on the fourth generation of the triplet technologies I think we've learned a lot about how to optimize our performance. There we are very aggressive with our adoption of leading edge technology.

Accelerated aside.

Can we get a lot from our customers is they want to understand the value proposition of BMI three hundreds. So Lisa I was hoping you could give us some understanding of price versus power comparison, our compute power and then today are you seeing your customers that are buying PMI 300 are they primarily buying it for inferencing.

Vivek Arya: It's needed, but I think those are only a.

Vivek Arya: A few of the pieces. We're also focused on continuing to innovate on architecture and design. So I think the longer term question that you ask is I think we're expecting that.

Or are they using it primarily for tooling and maybe for Jean Jean do you think is it possible for mm 300 finished the year at a run rate of about $1 5 billion.

Vivek Arya: The competition is going to be on a similar process technology and even in that case.

Yeah.

Okay harsh so let me start to your question about the value proposition for <unk> 300.

Vivek Arya: I think we feel like we have a very strong roadmap going forward and we will continue to.

Again customers are using it for different reasons, but presume that there is a performance per dollar.

Vivek Arya: To drive both the CPU and the GPU roadmap very aggressively.

Benefit to using AMD. So that's one piece of it.

Vivek Arya: And the next question comes from the line of harsh Kumar with Piper Sandler. Please proceed with your question.

The other piece of it though is we intrinsically have.

Harsh V. Kumar: Yeah, Hey, Thanks for letting me ask a question guys I have two questions, let me start off with.

More bandwidth in memory capacity on 300 X compared to the competition and what that means is for large language models that are many tens of billions of parameters.

Harsh V. Kumar: Accelerated side. The question, we get a lot from our customers is they want to understand the value proposition of BMI of three hundreds. So Lisa I was hoping you could give us some understanding of price versus power in comparison, our compute power.

You could potentially do the workload and fewer gpus, so it's a substantial.

Our system savings and allows you to do much more work.

Harsh V. Kumar: And then today are you seeing your customers that are buying PMI 300 are they primarily buying it for inferencing today or are they using it primarily for tooling and maybe for Jean Jean do you think is it possible for 300 to finish the year at a run rate of about $1 5 billion.

Within the same system.

In terms of what customers are using <unk> 300 for today I would say there are a number of customers using it for large language model inferencing and there are also customers that are using it for for training. So I think the the whole point is being a strong partner when you put these AI systems in place.

Speaker Change: Okay harsh so let me start to your question about the value proposition for <unk> 300.

Sometimes mixed used systems. So they would be used for both training and inference.

Speaker Change: Again customers are using it for different reasons, but presume that there is a performance per dollar.

John we have time for two more questions.

Yes, let me answer your question about the 300.

Speaker Change: Benefit to using AMD. So that's one piece of it.

Speaker Change: The other piece of it though is we intrinsically have.

<unk> Q4, 2024 is it possible to get to one five.

Speaker Change: More bandwidth in memory capacity on 300 X compared to the competition and what that means is for large language models that are many tens of billions of parameters.

It is possible right because Lisa mentioned earlier, we will see sequential increase each quarter and the mall back end loaded in second half and we do have our supplies more than $3 5 billion and.

Speaker Change: You could potentially do the workload and fewer gpus, so it's a substantial.

Of course that will continue to make progress with our customers. So that the math, yes, it's possible, but right now we are really looking at.

Speaker Change: System savings and allows you to do much more work.

Speaker Change: Within the same system.

Speaker Change: In terms of what customers are using <unk> 300 for today I would say there are a number of customers using it for large language model inferencing and there are also customers that are using it for for training. So I think the the whole point is being a strong partner when you quit these AI systems in place.

Focus on the execution of the current $3 5 billion plus.

And the next question comes from the line of Stacy <unk> with Bernstein Research. Please proceed with your question.

Hi, guys. Thanks for taking my questions.

The first one you talked about you expected a more shallow ram.

Sometimes mixed used systems. So they would be used for both training and inference.

<unk> 300, and it's clearly doing better than that.

Some of the upside I guess in the near term is that being pulled forward from the second half or is this like a step up or is it more of a step up in demand both in the first step in the second half relative to what you were seeing before.

Speaker Change: John we have time for two more questions.

John: Yes, let me answer your question about the 300.

John: Exiting Q4, 2024 is it possible to get to one five.

How do I, how do I interpret that but shallow comment that you made sure Stacy I don't think it's a it's a pull forward of demand I think what it is is we want to.

John: It is possible right because Lisa mentioned earlier, we will see sequential increase each quarter and the mall back end loaded in second half and we do have our supplies more than $3 5 billion and.

We wanted to see how long it would take for customers to fully qualify and get their workloads performance. So that depends a lot on the actual engineering work that's done and now that were let's call. It a quarter later.

John: Of course that will continue to make progress with our customers.

Speaker Change: The mass, yes, it's possible, but right now we are really looking at.

Seen that it has gone really well so it has actually gone.

Speaker Change: Focus on the execution of the car into $3 5 billion plus.

A bit better than our original forecast and as a result, we've seen stronger demand signals and customers are gaining confidence.

Speaker Change: And the next question comes from the line of Stacy <unk> with Bernstein Research. Please proceed with your question.

And their ability to deploy a significant number of EMI three hundreds this year.

Hi, guys. Thanks for taking my questions.

Stacy: For the first one you talked about you expected a more shallow ram <unk> 300, and it's clearly doing better than that.

Got it thank you for my follow up.

Want to ask Matts question, a little more directly you didn't quite answer it the $400 billion number you've got out there is that.

Stacy: And some of the upside I guess in the near term is that being pulled forward from the second half or is this like a step up or is it more of a step up in demand both in the first half in the second half relative to what you were seeing before like how do I, how do I interpret that like shallow comment that you made sure Stacy I don't think it's a it's.

Just silicon and chips or is there hardware and servers and stuff like that in that number as well like what's in that number.

I thought I had answered it but yes I'll answer it again it is.

It is accelerator chips is not systems, so think of it as.

Speaker Change: Pull forward of demand I think what it is is we we want to.

Gpus.

Speaker Change: Wanted to see how long it would take for customers to fully qualify and get their workloads performance so that.

Six that will be there. So are those those types of things, but it includes obviously it includes memory and other things that are packaged together with the.

That depends a lot on the actual engineering work, that's done and now that were let's call. It a quarter later.

But with the Gpus.

Yes memory will be quite significant right. So memory's a big portion of it.

Speaker Change: We've seen that it's gone really well so it's actually gone a.

Speaker Change: A bit better than our original forecast and as a result, we've seen stronger demand signals and customers are gaining confidence.

And the final question comes from the line of Chris Danley with Citi. Please proceed with your question.

Hey, Thanks for squeezing me in team.

Speaker Change: And their ability to deploy a significant number of M. I three hundreds this year.

I guess question for Lisa.

I 300 revenue ramps, how do you see the customer concentration, let's say a year or two from now do you think youll have one or two customers that are double digits or one or two that are half the revenue or do you think it will be totally fragmented.

Speaker Change: Got it thank you for my follow up.

Speaker Change: Want to ask Matts question, a little more directly you didn't quite answer it the $400 billion number you've got out there is that just silicon and chips or is there hardware and servers and stuff like that in that number as well.

I don't think it will be one or two that are half the revenue Chris I think we're building. This as a really we're happy to see sort of the broad adoption as always with sort of the large cloud.

Speaker Change: In that number yes.

Speaker Change: I thought I had answered it but yes I'll answer it again it is.

Speaker Change: It is accelerator chips is not systems, so think of it as.

Gpus.

Partners, we might see.

Speaker Change: Six that will be there. So are those those types of things, but it includes obviously it includes memory and other things that are packaged together with the.

One or two that are higher than others, but I don't think youll see the type of concentration that that you mentioned.

Great and then just a follow up on somebody else's question.

Speaker Change:

Speaker Change: But with the Gpus.

Intel's roadmap versus TSMC, So I'm sure you're sure you're intimately familiar with Tsmc's manufacturing roadmap and we've all seen them until opened up the kimono on on what they expect to happen over the next couple of years.

Speaker Change: Yes memory will be quite significant right. So memory's a big portion of it.

Speaker Change: And the final question comes from the line of Chris Danley with Citi. Please proceed with your question.

Do you think intel's going to close the gap somewhat with.

Christopher Brett Danely: Hey, Thanks for squeezing me in team.

Your foundry over the next couple of years do you think they'll be able to maintain the lead.

Christopher Brett Danely: I guess question for Lisa.

Christopher Brett Danely: I 300 revenue ramps, how do you see the customer concentration, let's say a year or two from now do you think youll have one or two customers that are double digits or one or two that are half the revenue or do you think it'll be totally fragmented.

I feel very good about our partnership with TSMC, they continue to execute extremely well.

We will see what happens over the next.

A few years, but.

I'd like to kind of reemphasize, what I said earlier, even in the case of process parity.

Lisa T. Su: I don't think it'll be one or two that are half the revenue Chris I think we are building. This as a really we're happy to see sort of the broad adoption as always with sort of the large cloud.

We feel very good about our architectural roadmap and all of the other things that we add as we look at our entire portfolio of Cpus Gpus, Gpus adaptive <unk> and kind of put them together to solve problems.

Lisa T. Su: Partners, we might see.

I think we feel really good about what we can do with our customers. So we're always going to be paying attention to sort of the process race, but I think I.

Lisa T. Su: One or two that are higher than others, but I don't think youll see the type of concentration that that you mentioned.

Speaker Change: Great and then just a follow up on somebody else's question.

I think we feel very good about sort of our strategy and how do we continue to.

Speaker Change: Intel's roadmap versus TSMC, So I'm sure you're sure you're intimately familiar with Tsmc's manufacturing roadmap and we've all seen them until opened up the kimono on on what they expect to happen over the next couple of years.

Sort of push the envelope on the computing Roadmaps.

Okay and that is the end of the question and answer session I would like to turn the floor back over to the AMD team for any closing comments.

Do you think Intel is going to close the gap somewhat with.

Great John that concludes today's call. Thank you to everyone for joining us today.

Speaker Change: With your foundry over the next couple of years do you think you'll be able to maintain the lead.

Speaker Change: I feel very good about our partnership with TSMC, they continue to execute extremely well.

And this concludes today's teleconference. You may disconnect. Your lines at this time. Thank you for your participation.

Speaker Change: We will see what happens over the next few years, but.

Speaker Change: I'd like to kind of reemphasize, what I said earlier, even in the case of process parity.

Speaker Change: We feel very good about our architectural roadmap and all of the other things that we add as we look at our entire portfolio of Cpus Gpus, Gpus adaptive <unk> and kind of put them together to solve problems.

Speaker Change: I think we feel really good about what we can do with our customers. So we're always going to be paying attention to sort of the process race, but I think.

Speaker Change: I think we feel very good about sort of our strategy and how do we continue to.

Speaker Change: Sort of push the envelope on the computing Roadmaps.

Speaker Change: Okay and that is the end of the question and answer session I would like to turn the floor back over to the AMD team for any closing comments.

Speaker Change: Great John that concludes today's call. Thank you to everyone for joining us today.

Speaker Change: And this concludes today's teleconference. You may disconnect. Your lines at this time. Thank you for your participation.

Speaker Change: Hum.

[music].

Speaker Change:

Speaker Change: Okay.

Speaker Change: Hum.

Speaker Change: Hello.

Speaker Change: Oh.

Speaker Change: Mhm.

Speaker Change: [music].

Speaker Change: No.

Speaker Change: Oh.

Speaker Change: [music].

Speaker Change: Okay.

Speaker Change: Yeah.

Speaker Change: Sure.

Speaker Change: Mhm.

Speaker Change: Hum.

Speaker Change: [music].

Speaker Change: Greetings and welcome to the a M D fourth quarter and full year 2023 conference call. At this time all participants are in a listen only mode. A brief question and answer session will follow the formal presentation.

Speaker Change: Anyone should require operator assistance during the conference. Please press star zero on your telephone keypad as a reminder, this conference is being recorded.

Speaker Change: It is now my pleasure to introduce to you Mitch Haws, Vice President Investor Relations. Thank you Mitch you may begin.

Mitch Steves: Thank you John and welcome to Amd's fourth quarter, and full year 2023 financial results Conference call.

Mitch Steves: But by now you should have had the opportunity to review a copy of our earnings press release and the accompanying slides if.

If you have not had the chance to review these materials can be found on the Investor Relations page of AMD Dot com.

Speaker Change: Participants on today's call are Dr. Lisa Su, our chair and Chief Executive Officer, and Jean Hu, Our executive Vice President Chief Financial Officer and Treasurer.

Speaker Change: This is a live call and will be replayed via webcast on our website.

Speaker Change: 15th and the Morgan Stanley Technology Media and Telecom conference on March 5th.

Lisa T. Su: Thank you Mitch and good afternoon to all those listening in today.

Use across cloud enterprise and AI customers.

Lisa T. Su: On a full year basis annual revenue declined 4% to $22 7 billion as record datacenter and embedded segment annual revenue was offset by lower client and gaming segment revenue.

Lisa T. Su: Turning to the fourth quarter business results datacenter segment revenue grew 38% year over year, and 43% sequentially to a record $2 3 billion.

Lisa T. Su: Server CPU and datacenter GPU sales, both set quarterly and annual revenue records as sales of our data center products accelerated throughout the year.

Lisa T. Su: We gained server CPU revenue share in the quarter driven by significant double digit percentage growth in fourth Gen epic processor revenue and demand for our third Gen epic processor portfolio.

Lisa T. Su: And cloud while the overall demand environment remains soft server CPU revenue increased year over year and sequentially as North American Hyperscale or expanded fourth Gen epic processor deployments to power their internal workloads in public instances.

Lisa T. Su: Amazon Alibaba, Google, Microsoft and Oracle brought more than 55, AMD powered AI HBC and general purpose cloud instances into preview or general availability in the fourth quarter.

Lisa T. Su: Exiting 2023, there were more than 800 epic CPU based public cloud instances available.

Lisa T. Su: We expect this number to grow in 2024 based on the leadership performance efficiency and features of our epic CPU portfolio.

Lisa T. Su: In enterprise sales accelerated by a significant double digit percentage in the quarter as we built momentum with Forbes 2000 customers.

Lisa T. Su: Looking ahead customer excitement for our upcoming turn family of Epic processors is very strong.

Lisa T. Su: <unk> is a drop in replacement for existing fourth Gen epic platforms that extends our performance efficiency and Tcl leadership with the addition of our next Gen 10, five core new memory expansion capabilities and higher core counts.

We launched our <unk> 300 accelerator family in December with strong partner and ecosystem support from multiple large cloud providers, all the major Oems and many leading AI developers.

Lisa T. Su: <unk> 300, ex Gpus deliver leadership generative AI performance by combining our high performance cdna, three architecture with industry, leading memory bandwidth and capacity.

Lisa T. Su: In cloud, we are working closely with Microsoft Oracle meta and other large cloud customers on instinct Gpus deployments powering both their internal AI workloads and external offerings.

Lisa T. Su: In HBC supercomputing, we shipped the majority of AMD instinct mid 300, <unk> accelerators for the El Capitan supercomputer in the fourth quarter and expect to complete shipments this quarter for what is expected to be the world's fastest supercomputer when it comes online later this year.

Lisa T. Su: On AI software development, we made significant progress expanding the ecosystem of AI developers working on AMD platforms with the release of our <unk> <unk> software suite.

Lisa T. Su: The <unk> stacks significantly increases performance in key generative AI workloads adds expanded support and optimizations for additional frameworks in libraries and simplifies the overall developer experience.

For example, we are very pleased to see how quickly Microsoft was able to bring up GPT for mid.

Lisa T. Su: 300, <unk> in their production environment and rollout Azure private previews of new MRI 300 instances aligned with the 300 X launch.

Lisa T. Su: Looking ahead, our prior guidance was for datacenter GPU revenue to be flattish from Q4 to Q1 and exceed $2 billion for 2024.

Lisa T. Su: Based on the strong customer pull and expanded engagements. We now expect datacenter GPU revenue to grow sequentially in the first quarter and exceed $3 5 billion in 2024.

Lisa T. Su: We have also made significant progress with our supply chain partners and have secured additional capacity to support upside demand.

Lisa T. Su: Turning to our client segment revenue was one 5 billion, an increase of 62% year over year and flat sequentially.

Lisa T. Su: We launched our latest generation Ryzen 8000 series notebooks and desktop processors in January including our Ryzen $80 40 mobile series that combined leadership compute performance and energy efficiency with an updated MCU that delivers up to 60% more AI performance compared to our prior generation that was already industry.

Lisa T. Su: <unk>.

Lisa T. Su: ASER, a sous HP Lenovo MSI and other large PC Oems will offer notebooks powered by our rising 8000 series processors with the first system is expected to go on sale in February.

Lisa T. Su: To further our leadership in AI Pcs, we launched our Ryzen 8000 G series processors earlier this month, which are the industry's first desktop Cpus with an integrated AI engine.

Lisa T. Su: Millions of AIP CES powered by Ryzen processors have shipped to date and ryzen Cpus power more than 90% of AI enabled Pcs currently in market.

Our work with Microsoft and our PC ecosystem partners to enable the next generation of <unk> expanded significantly in the quarter.

Lisa T. Su: Strikes combines our Nextgen Zen five core with enhanced rdna graphics, and an updated ryzen AI engine to significantly increase the performance energy efficiency and AI capabilities of Pcs cut.

Lisa T. Su: Customer momentum restricts as strong with the first notebooks on track to launch later this year.

Lisa T. Su: Looking at 2024, we are planning for the PC Tam to grow modestly year on year weighted towards the second half as AIP cease ramp.

Lisa T. Su: We continue to see strong growth opportunities for our client business as we ramp our current products extend our AIP see leadership and launch our next wave of <unk> Cpus.

Now turning to our gaming segment revenue declined 17% year over year, and 9% sequentially to $1 4 billion as lower semi custom revenue was partially offset by increased sales of Radeon Gpus.

Lisa T. Su: Semi custom Soc sales declined in line with our projections in the quarter.

Lisa T. Su: And gaming graphics revenue grew both year over year and sequentially driven by strong demand in the channel for both our Radeon 6000, and Radeon 7000 series Gpus.

We also launched new open source fidelity FX Super resolution three software that can deliver significantly higher gaming frame rates on both Gpus and apu's.

Lisa T. Su: Turning to our embedded segment revenue decreased 24% year over year, and 15% sequentially to $1 1 billion as customers focused on reducing their inventory levels.

Lisa T. Su: We expanded our embedded portfolio in the quarter with new leadership solutions for key markets.

Lisa T. Su: We launched new virtual prime adaptive ssds for the aerospace test and measurement healthcare and communications markets that deliver industry <unk> support for DDR, five memory and increased DSP capability compared to our prior generation.

Lisa T. Su: In automotive, we launched new virtual Soc solutions that bring industry, leading AI compute capabilities and advanced safety and security features to next generation vehicles.

Lisa T. Su: We also launched ryzen embedded processors with unmatched performance and features for industrial automation machine vision robotics and edge server applications.

Lisa T. Su: Looking at 2024, we expect overall embedded demand will remain soft through the first half of the year as customers continue to focus on normalizing their inventory levels.

Lisa T. Su: In summary, I am very pleased with our fourth quarter and full year results for 2024, we expect the demand environment to remain mixed with strong growth in our data center and client segments offset by declines in our embedded and gaming segments.

Lisa T. Su: Against this backdrop, we believe we will deliver strong annual revenue growth and expand gross margin driven by the strength of our instinct epic and ryzen product portfolios.

Lisa T. Su: In the data center, we see 2024 as a start of a multiyear AI adoption cycle with the market for data center AI accelerators growing to approximately 400 billion in 2027.

Lisa T. Su: Customer deployments of our instinct Gpus continues accelerating with <unk> 300, now tracking to be the fastest revenue ramp of any product in our history and positioning us well to capture significant share over the coming years based on the strength of our multi generation instinct GPU roadmap and open source rock them software.

Lisa T. Su: <unk>.

Lisa T. Su: In <unk>, we're focused on delivering our long term roadmaps with leadership ryzen AI <unk> capabilities to enable differentiated experiences as Microsoft and our other software partners bring new AI capabilities to Pcs starting later this year.

Lisa T. Su: At the same time, we are rapidly driving leadership AI compute capabilities across the full breadth of our embedded product portfolio.

Lisa T. Su: Now I'd like to turn the call over to Jim to provide some additional color on our fourth quarter and full year financial results.

Lisa T. Su: Gene.

Jim: Thank you Lisa and good afternoon, everyone.

Jim: Starting with the reveal file financial results and then provide our current outlook for the first quarter of fiscal 2024.

Jim: <unk> executed well in 2023, despite the mixed market demand environment Delevering in revenue of $22 7 billion and the earnings per share of $2 65.

Jim: We drove year over year revenue growth in our embedded and data center segments. In addition, we successfully launched our AMD <unk> I might be hungry the Gpus positioning us for a strong ramp in 2024 in the AI market.

Jim: Revenue was up 6% sequentially, primarily driven by the ramp of AMD <unk> Gpus crosses several leading customers and higher revenue from <unk> silver our processors, partially offset by the decline in embedded in the gaming segment of revenues.

Jim: Gross margin was 51% flat year over year with a higher revenue contribution from the data center and client segments offset by lower embedded segment revenue.

Jim: Operating expenses were $1 7 billion, an increase of 8% year over year as we invest in R&D and marketing activities to support our significant AI growth opportunity.

Jim: Operating income was $1 4 billion, representing a 23% operating margin.

Jim: Taxes interest expense and other was the one country at the end of $63 million.

Jim: For the fourth quarter of 2023 diluted earning per share was <unk> 77, an increase of 12% year over year.

Jim: Now turning to our reportable segments.

Jim: Data Center segment operating income was $666 million or 29% of revenue compared to 444 million or 27% a year ago.

Jim: Higher operating income was primarily due to operating leverage driven by higher revenue.

Jim: Client segment revenue was $1 5 billion up 62% year over year, driven by Ryzen 7000. This series of CPE sales.

Jim: <unk> segment operating income was $55 million of 4% of revenue compared to an operating loss of 152 million a year ago driven by higher revenue.

Jim: Gaming segment revenue was $1 4 billion down 17% year over year, and 9% sequentially due to a decrease in semi customer revenue, partially offset by increase in Raytheon GPU sales.

Gaming segment operating income was 224 million or 16% of revenue compared to $200 to 66 million or 16% a year ago.

Jim: <unk> segment revenue was $1 1 billion down 24% year over year, and 15% sequentially as customers continue to work down their inventory levels.

Jim: Embedded segment operating income was 461 million or 44% of revenue compared to 699 million or 50% a year ago.

Turning to the balance sheet and cash flow during the quarter, we generated 381 million in cash from operations and free cash flow was $242 million.

Jim: Inventory decreased sequentially by 94 million to $4 4 billion.

Jim: At the end of the quality of our cash cash equivalents and short term investments was strong at $5 8 billion.

Jim: In the fourth quarter, we repurchased 2 million shares and returned $233 million to shareholders for.

Jim: For the year, we repurchased 10 million shares and returned $985 million to shareholders.

Jim: We have $5 6 billion in remaining share repurchase authorization.

Speaker Change: Now turning to our first quarter of 2020 for outlook.

Speaker Change: We expect revenue to be approximately $5 4 billion, plus or minus $300 million sequentially.

Sequentially, we expect data center segment revenue to be flat.

Speaker Change: The seasonal decline in silver sales offset by strong data center GPU ramp.

Speaker Change: Embedded revenue to decline as customers to continue to work down their inventory levels.

Speaker Change: Client segment revenue to decline seasonally in the in the gaming segment and Tim as we enter the fifth year of what has been a very strong gaming cycle and given current customer inventory levels, we expect revenue to decline by significant double digit percentage.

Speaker Change: Year over year, we expect datacenter and the cloud on the segment revenues to increase by strong double digit percentage given the strengths of our product portfolio and the share gain opportunities.

Speaker Change: <unk> segment to decline in the gaming segment revenue to decline by significant double digit percentage.

Speaker Change: In addition, we expect our first quarter non-GAAP gross margin to be approximately 52% non.

Speaker Change: non-GAAP operating expenses to be approximately 173 billion.

Speaker Change: non-GAAP effective tax rate to be 13% and a diluted share count is expect to be approximately $1 six 3 billion shares.

Speaker Change: Sure.

Speaker Change: While we're not providing specific full year guidance for 2024, let me provide some color directionally for the year, we expect 2020 for data center and the client segment revenue to increase driven by the strength of our product portfolio and the share gain opportunities in.

Speaker Change: <unk> segment revenue to decline in the gaming segment revenue to decline by significant double digit percentage.

Speaker Change: We expect to expand gross margin in 2024 and to continue to invest to address the large AI opportunities, while driving operating model leverage to delever earnings per share growth faster than top line revenue growth.

Speaker Change: Very well positioned to build on this momentum and deliver strong financial performance in 2024 and beyond.

Speaker Change: With that I'll turn it back to Mitch for the Q&A session.

Mitch: Thank you Jane John we're happy to pull the audience for questions.

Thank you Mitch we will now be <unk>.

Mitch: A question and answer session, if you'd like to ask a question. Please press star one on your telephone keypad, a confirmation tone will indicate that your line is in the queue. You May press star two if you'd like to remove a question from the queue.

Mitch: For participants using speaker equipment and may be necessary to pick up your handset before pressing the star keys.

Mitch: Ask that you please limit yourself to one question and one follow up thank you.

Mitch: One moment, please while we poll for questions.

Mitch: And the first question comes from the line of Iron Rackers from Wells Fargo. Please proceed with your question.

Aaron Rakers: Yes, thanks for taking the question.

Aaron Rakers: Just kind of framing the outlook and the guidance for this calendar first quarter I guess the first question is can you help us on a relative basis, the $400 million of datacenter GPU revenue that you expected in Q4.

Aaron Rakers: That ultimately kind of fell out to be and then on the guidance into <unk> can you help us appreciate what seasonal.

Aaron Rakers: Is defined as as we think about the server business.

Aaron Rakers: Into the <unk> guide.

Speaker Change: Sure. Let me start and then <unk> has something to add so relative to the datacenter GPU business look we were very pleased with performance that we saw in the fourth quarter.

Speaker Change: There's always going to be a very.

Speaker Change: Sort of back end quarter, our weighted as we were ramping the product and we saw 300 days.

Our HPV product actually ramp very well and then we saw 300 X.

Speaker Change: The AI product actually exceed our expectations based on strong customer demand the way the qualifications wind and then the ramp manufacture them. So we were over $400 million.

Speaker Change: For that business in the fourth quarter, and then going into the first quarter as we look at the business server seasonality.

Speaker Change: Call it something around let's call it high single digit low double digit there also some other pieces of the data center business I think the key piece of it is.

Speaker Change: <unk> 300 ex increase as we go into the first quarter and.

Speaker Change: Things are going relatively well so yes.

Give you some color Eric I'll give you some color about.

Clay: Clay on the seasonality and others. So clay on these very similar to silver typically Q1 is high single digits to low double digit.

Clay: That's the consistent Davita posture.

Clay: The embedded aside it's very consistent with what we said in the pasture and consistent with what you see in the industry.

Clay: Embedded business is going through a bottoming process and we think of Q1, it will have a low double digit sequential decline.

That's the embedded.

Clay: On the gaming side, Lisa mentioned during her.

Clay: Prepared remarks.

Clay: We have the late stage of a product cycle in the year five.

Clay: Gaming console, but at the same time, we also have inventory at our customers. So the combination for those the impact we expect the Q1 gaming sequential declines probably more than 30%. So hopefully that help you a little bit.

Speaker Change: Yes, very helpful gene and as a quick follow up I'm just curious the traditional server demand that you see I know when we look at server CPU.

Speaker Change: Shipments were down north of 20% year over year.

Are you seeing any signals or how are you thinking about a recovery in that traditional call. It non AI general purpose server market as we move through 2000 and for.

Sure Aaron So look I think.

Speaker Change: I agree with your.

Aaron Rakers: The characterization of the 2023 on demand, although we did see.

Aaron Rakers: So strong progress in the second half of the year, especially as customers in cloud and enterprise adopted our general and our Zen four family so going into 2020 for I would say the traditional server market is probably still mixed especially into the first half of the year Theres still some cloud optimization going on as well as sort of.

Aaron Rakers: Enterprise.

Being a little bit cautious that being the case, though we also see opportunities for us to continue to grow share.

Aaron Rakers: In the traditional server business I think our portfolio is extremely strong.

Aaron Rakers: Adoption of Genoa in Bergamo, as well as our new <unk> product lines.

Aaron Rakers: We're getting a lot of traction and then we also see turned in our Gen five product coming in the second half of the year. So even in a mixed demand environment I think we're bullish on what traditional server Cpus can do in 2024.

Aaron Rakers: And the next question comes from the line of Timothy Arcuri with UBS. Please proceed with your question.

Timothy Michael Arcuri: For example, do you think you can gain share at a rate that's kind of similar to the rate that you gained share for server CPU.

Speaker Change: Yes, Thanks, Tim for the question I would say a couple of things first of all we're really pleased with the progress that we've made in our data center GPU business.

Speaker Change: Thank the ramp that we've seen the customer traction that we've seen even in the last few months I think has has been great and that gives us a lot of confidence in the ramp of this business I think the beauty of the AI market here is growing so quickly that I think we have both the market dynamic as well as our ability to gain share.

In that framework the point I'll make is our customer engagements right now are all quite strategic dozens of customers with multi generational conversations. So as excited as we are about the ramp of <unk> 300, and frankly, there is a lot to do.

Speaker Change: <unk> growth driver given the market demand as well as our own product capabilities.

Speaker Change: Thanks, a lot Jean I guess.

Speaker Change: Follow up I know that you don't want to guide the full year, but I'm wondering if I can pin you down just to touch on maybe.

Speaker Change: Mileposts that you are kind of.

Speaker Change: Marching to for 2020 for growth is up 20% for the whole company is that a reasonable target and then I guess within data center. If you just add the Incrementals datacenter GPU revenue and you assume that the server business grows a little bit it seems like that should maybe double year over year, but I'm kind of wondering if you could give us any ranges on those numbers. Thanks.

Speaker Change: Hi, Tim Thank you for the question yes.

Speaker Change: Yes.

Speaker Change: <unk> a year.

Speaker Change: Very early.

Speaker Change: The year literally January.

Speaker Change: I think the way to think about it Lisa mentioned during her prepared remarks, we feel pretty good about both our data center and client businesses to grow in 2020 for of course, the largest incremental revenue opportunities are going to come from data center between both.

Speaker Change: The silver side again, your mall share and datacenter GPU side.

Significantly ramp up of our I might 300 I think.

Speaker Change: That's how we think about it we do have a headwind from <unk>.

Speaker Change: Gaming segment, we do think of year over year, we will see very significant.

Speaker Change: Double digit decline on the gaming segment and at the same time embedded is going through the bottoming process. We do think the second half we will see the recovery. So those are the puts and takes I can talk about it.

Speaker Change: And the next question comes from the line of Matt Ramsay with TD Cowen. Please proceed with your question.

Matthew D. Ramsay: Thank you very much good afternoon.

Matthew D. Ramsay: Lisa.

Matthew D. Ramsay: Wanted to ask I mean, there's been so much.

Matthew D. Ramsay: Focus on scrutiny as there should be on the really exciting progression with M. I.

300.

Matthew D. Ramsay: I mean, we've progressed over the last six months from.

Matthew D. Ramsay: I think some doubts and the investment community.

Software and your ability to ramp the product now.

Proving that youre ramping it with I think you said thousands of customers across different end markets. So.

Matthew D. Ramsay: What I'm interested in hearing a little bit more about and you guys have been open about what.

Matthew D. Ramsay: Some of the forward programs in your traditional server business look like from a roadmap perspective I'd be interested to hear how you're thinking about the roadmap and your MRI.

Matthew D. Ramsay: To accelerate our family is it going to.

Matthew D. Ramsay: They're going to continue to be parts that our CPU and GPU together is it a primarily a GPU only roadmap what kind of cadence are you thinking about.

Matthew D. Ramsay: Just any kind of color you can give us on some of the forward roadmap trajectory for that program would be really helpful. Thanks.

Speaker Change: Yeah sure Matt so.

Speaker Change: I appreciate the comments I think the <unk>.

Speaker Change: <unk> that we're getting with the <unk> 300 family is is really strong I think what has benefited us here is our use of triplets technologies, which has given us the ability to have sort of both the the Apu version.

Speaker Change: As well as the GPU version and we continue to use that to differentiate ourselves and that's how we get our memory bandwidth in memory capacity.

Speaker Change: Vintages as we go forward you can imagine like we did in the epic timeframe. We planned multiple generations in sequence. That's the way we're planning the roadmap one of the things I will note about the AI.

Speaker Change: <unk> market is the demand for compute is so high that we are seeing.

Speaker Change: Sort of an acceleration of the roadmap generations here and we are similarly planning.

Speaker Change: Acceleration of our roadmap I would say that we will talk more about the overall roadmap beyond my 300 as we get into later this year, but you can be assured that we're working very closely with our customers to have a very competitive roadmap for both training and inference.

Speaker Change: That.

Speaker Change: We will come out over the next couple of years.

Speaker Change #100: Thank you for that Lisa just as a follow up.

Speaker Change #101: I guess one of the questions that I have.

Speaker Change #102: <unk> been getting a lot in different forms.

Speaker Change #103: With respect to the 400 billion Tam that you guys have laid out for 2027.

Speaker Change #104: Maybe you could give us a little look under the Hood.

Speaker Change #104: I guess.

Speaker Change #104: 100 versions of the same question, which is how that did you come up with that number.

If you could give us a little bit more in terms of are we talking about system and accelerator cards are we talking about just the silicon or are we talking about for servers.

Speaker Change #104: And what kind of sort of unit assumptions any kind of thing that you can give us on market sizing or what gives you the visibility so early into this generative AIG trend.

Speaker Change #104: To give a precise number of three years out.

That would be really really helpful. Thank you.

Speaker Change #105: Well, Matt I don't know how precise it is but I think we said approximately $400 billion, but I think what we need to look at is growth rate and how do we get to those growth rates.

Speaker Change #105: I think we expect units to grow sort of.

Speaker Change #105: Double digit percentage, but you should also expect that content is going to grow. So if you think about how important memory and memory capacity is.

Speaker Change #105: As we as we go forward you can imagine that we'll see.

Acceleration, there and just the overall content as we go to more advanced technology nodes. So there's some ASP uplift in there and then what we also do is we're planning longer term roadmaps with our customers in terms of how they're thinking about.

Speaker Change #105: The size of training clusters, the number of training clusters, and then the fact that we believe influence is actually going to exceed training.

Speaker Change #105: As we go into the next couple of years just given.

Speaker Change #105: As more enterprises adopt so I think as we look at all of those pieces.

Speaker Change #105: I think we feel good that the growth rate is going to be significant and sustained over the next few years in terms of what's in that Tam.

Speaker Change #105: It really is accelerator town so within accelerators.

Speaker Change #105: There are certainly Gpus and there will also be some asics.

Speaker Change #105: That our other accelerators that are in that Tam as we think about sort of the different types of models from smaller models to fine tuning of models too.

The largest large language models, I think youre going to need different silicon for those different use cases, but from our standpoint Gpus are still going to be the sort.

Speaker Change #105: Sort of the compute element of choice when Youre talking about training and inferencing on the largest language models.

Speaker Change #105: And the next question comes from the line of Joe Moore with Morgan Stanley. Please proceed with your question.

Joe L. Moore: Great. Thank you I think you talked about the MRI 300, cloud workloads being kind of split between the more customer facing workloads and versus your internal can you talk about how you see the breakdown of that and how.

Joe L. Moore: How is your ecosystem progressing this is a brand new chip.

Joe L. Moore: It seems impressive that you are able to support kind of a broad range of of customer facing workloads and cloud.

Speaker Change #106: Yes sure Joe So yes look we are really happy with how the <unk> 300 has come up and we've now deployed.

Joe L. Moore: And working with a number of customers. What we have seen is certainly <unk> has been a very important.

Joe L. Moore: The GPU capabilities that we would have to do some level of optimization, but the optimization has gone well I think to your broader question.

Joe L. Moore: I look at this is there are lots of opportunities for us to work directly with large customers. Both on the cloud side as well as on the enterprise side, who have specific training and inferencing workloads, our job is to make it as easy as possible for them and so our entire tool.

Joe L. Moore: Chain.

Joe L. Moore: All of our sort of the overall rosin suite has really gone through.

Joe L. Moore: Significant progress.

Joe L. Moore: Over the last six to nine months and then we're also getting some nice support from the open source community. So the work that hugging faces doing.

Joe L. Moore: Is tremendous in terms of just real time optimization.

Joe L. Moore: On our hardware.

Joe L. Moore: Our partnership with open AI on Triton.

Across.

Joe L. Moore: Number of.

Joe L. Moore: These.

Joe L. Moore: These open source models has helped us actually make very rapid progress.

Speaker Change #107: Great and for my follow up I guess, a lot of the forecasting of your business that I'm hearing is coming from supply chain and we're sort of hearing Amd's building action in Asia.

Speaker Change #107: I guess.

Speaker Change #108: How would you ask us to think about that are you looking at being kind of sold out for the year and so the supply chain would be close to revenue are you building for the best case scenario.

Speaker Change #108: I worry about sometimes expectations when people hear the supply chain numbers and I'm just curious how do you bridge the gap.

Yeah, So I mean, Joe I think we updated our revenue expectations.

Speaker Change #109: This quarter from our original number of 2 billion to $3 5 billion to try to give.

Speaker Change #109: Some.

Speaker Change #110: Some bounding on some of the discussion out there the way to think about the $3 5 billion is these are customers that were working with.

Speaker Change #110: Who have given us firm commitments on what they need.

Speaker Change #110: As you know the lead times on these products are quite long so it's important.

To have those forecasts in early and we have a strong order book so that gives us.

Speaker Change #110: Good confidence to exceed the $3 5 billion.

From a supply chain standpoint, our goal is always to build more supply.

Speaker Change #110: And so from that standpoint.

Speaker Change #110: We have also worked with our supply chain partners and secured significant capacity I think about it as first half capacity is tight.

Speaker Change #110: And more comes on in the second half of the year, but we've certainly made more progress there so.

Speaker Change #110: We do have more supply and we're going to continue to work with our customers on their deployments and we'll update that number as we go through the year.

Speaker Change #110: And the next question comes from the line of Toshi Hari with Goldman Sachs. Please proceed with your question.

Hi, Thank you for taking the question I had one on the 300 as well Lisa.

Toshiya Hari: I guess first of all how should we think about the quarterly trajectory beyond Q1, you talked about Q1 being up sequentially.

Toshiya Hari: Is it fair to assume kind of a straight line as we progress through the balance of the year or is it more second half skewed how.

Toshiya Hari: How should we think about that and I guess more importantly, some of the cloud potential customers that have yet to.

Toshiya Hari: Officially sign up for sign off on the <unk> 300.

Toshiya Hari: What's the sticking point is just a function of time and you just need a little bit more time to go back and forth and tweak things are.

Toshiya Hari: Is there a software kind of concern.

Toshiya Hari: What's holding them back at this point, yes.

Speaker Change #111: Thanks for the question. So first on the 300 trajectory I think you would expect that revenue should increase every quarter from now through.

Speaker Change #112: Sort of the end of the year, but it will be a bit more second half weighted and part of that is just.

Speaker Change #112: Customers as they're finishing up their qualifications in their lines as well as sort of how our supply.

Speaker Change #112: Change is ramping so yes, it should increase each quarter, but be a bit more second half weighted and then to your comment about customers we're engaged with.

Speaker Change #112: All of the large customers.

Speaker Change #112: These are all folks that know us really well given our deep relationships in epic.

Speaker Change #112: I think people just have different adoption cycles as they consider what they're trying to do in their roadmap, but I view this as still very very early innings for us.

Speaker Change #112: In this space and I think a question was asked earlier I think the key is this is not just about <unk> 300 conversation, but it really is about sort of our long term multi generational roadmap and so that's the context on which we're working with our largest customers as well as.

Speaker Change #112: As you know Theres a lot of demand coming from.

Speaker Change #112: Folks that are more AI centric and not necessarily typical cloud customers, but more enterprise or let's call. It AI specific.

Speaker Change #112: Companies that were also very well engaged with.

Speaker Change #113: Got it that's Super helpful. And then as my follow up maybe one for gene on the gross margin side, you're guiding Q1 to 52% I'm curious.

Speaker Change #113: Im sure Youre not going to give quantitative guidance beyond Q1, but how to think about.

Gene: The trajectory for Q2 and beyond.

Speaker Change #114: Pretty sure Youre working through some kinks as it pertains to the.

Gene: The instinct ramp hopefully that improves over time, so that should be a tailwind.

Gene: <unk>, perhaps the second half.

Gene: Turns for the better.

Gene: <unk> got some Super server CPU volume growth throughout the year. So it feels like you've got multiple tailwind as we think about gross margin progression on a sequential basis, but what are the potential headwinds as we move throughout 2024. Thank you.

Speaker Change #115: Yes, Toshi I. Thank you for the question, yes, you're absolutely right. We have some puts and takes that impact our gross margin.

Speaker Change #116: The Q1 120 basis points higher than Q4 sequentially.

Toshiya Hari: Primarily because of the higher datacenter contribution actually more than offset the decline of embedded business in Q1 going forward. The way to think about it is as you said is the major driver is going to be datacenter business is going to grow much faster than other segment that makes.

Toshiya Hari: <unk> change it will help us to expand the gross margin nicely I think you also.

Toshiya Hari: Spot.

Toshiya Hari: On the embedded coming back in second half of which will be a tailwind.

Toshiya Hari: The datacenter GPU.

Toshiya Hari: At the very early stage of ramp we are improving testing time yield and continue to expand gross margin and we expect that to be accretive to corporate average. So those are all the tailwind to come in in the second half.

Speaker Change #117: I would just say.

Speaker Change #117: The hydro Windsor side to continue to be in the first half where we see embedded in the business.

Speaker Change #117: Not only Q1, we see a sequential decline Q2, probably are going to be sequentially flattish Q versus Q1 that.

Speaker Change #117: Our hydro wind for us.

Because it does have a very nice gross margin, but overall, we feel pretty good about.

The trajectory over the gross margin improvement.

Speaker Change #117: <unk> second half.

Speaker Change #117: And the next question comes from the line of Ross Seymore with Deutsche Bank. Please proceed with your question.

Ross Seymore: Hi, Thanks for let me ask a question I wanted to get into the competitive environment first on me.

Ross Seymore: Instinct side of things how.

Ross Seymore: How thats going it doesn't seem to be slowing down your ramp whatsoever. But then also on the street server CPU side of things you said youre gaining share in that area, but as we think about future roadmaps pricing incentives those sorts of things any meaningful change in the competitive environment that youre seeing through the 2024.

Sure.

Ross Seymore: Hi.

Ross Seymore: The environment for US is always competitive so I think that that has not changed.

If I look at the instinct side I think we have I think we've shown that.

Ross Seymore: 300, and our roadmap are actually very competitive.

Ross Seymore: There are some places where let's call it.

Ross Seymore: It's more even like in the training environment, but as we look at the influencing environment. We think we have significant advantages and that's showing through in some of our our customer work. So we think for both training and inference, we will continue to be very competitive.

Ross Seymore: And then as you go into the CPU side again.

Ross Seymore: From our view.

Ross Seymore: With each generation of epic we've continued to gain share I think we exited the fourth quarter a record share for.

Ross Seymore: For AMD.

Ross Seymore: We are still quite underrepresented in an enterprise. So I think there is an opportunity for us to continue to gain share as we go through 2024 from a competitive standpoint.

Ross Seymore: What we see is Zen four is extremely competitive right now.

Ross Seymore: With Genoa Bergamo Sienna.

Ross Seymore: And as we go into <unk>.

Keep in the design in cycle four zone, five and turn and we feel very good about how we're positioned.

Speaker Change #118: Thanks for that and I guess as my follow up on the data center side. Another thing Thats been pervasive throughout 2023 at least the GPU side crowding out the CPU side, you mentioned that there is still a little bit of color.

Speaker Change #118: Digestion going on within your epic business, but where do you see that standing I know youre going to gain share et cetera, but you guys, but clearly benefit from the <unk> side of the datacenter GPU side, but what about on the CPU side of things is that Edwin now behind us or is that still an issue.

Speaker Change #119: I think we expect the CPU business from a market standpoint to grow Ross as we go into 2024, I think the rate and pace of growth will depend a little bit on the macro and just overall capex trends, but from our standpoint, we are starting to see.

Some of our larger customers plan their refresh cycles. There is a lot of let's call. It older equipment that has yet to be.

Speaker Change #119: Refreshed and the value proposition for refresh is so strong because the energy efficiency and sort of the footprint.

Speaker Change #119: Of the newer generations are so much better than sort of the four or five year old infrastructure that we do see that refresh cycle happening as we get into 2024 I think the exact timing we will have to understand more.

Speaker Change #119: As the market evolves as we go through the year.

Speaker Change #119: And the next question comes from the line of Vivek Arya with Bank of America Securities. Please proceed with your question.

Vivek Arya: Thanks for taking my questions. So first one.

Vivek Arya: Lisa from you gave us a $2 billion plus number for that my before now you've raised it to over $3 5 billion and I'm curious what drove the change was it incremental demand signals both the supply end.

Vivek Arya: Can you supply more of let's say demand is four or five or 6 billion what is the.

Vivek Arya: Limitation and sort of related to that.

Vivek Arya: On the competition side your competitor will launch there will be 100.

Vivek Arya: Later in the year do you think that will change the competitive landscape in anyway.

Speaker Change #120: Yes, sure Vivek. So I think what we said is as we went from two to $3 5 billion. It really is mostly customer demand signals. So.

Speaker Change #120: <unk> orders have come on books and as we've seen programs move from let's call. It pilot programs into full manufacturing programs. We have updated the revenue forecast as I said earlier from a supply standpoint, we are planning for success and so we worked closely with our supply chain partners to ensure.

Or that we can ship more.

Speaker Change #120: And then $3 5 billion substantially more depending on what customer demand is as we go into the second half of the year and then in terms of again Roadmaps as I said, we're very focused on a competitive roadmap.

Speaker Change #120: That sort of what the next generations are beyond my 300, So I do believe that we have a strong roadmap in place and continue to work with our customers too.

Speaker Change #120: Sort of adopt our roadmap as quickly as possible.

Speaker Change #121: Got it.

A longer term question, Lisa if I look at the success that <unk> enjoyed its many factors, but a few of them included your early adoption of triplets and the strong partnerships we have had.

Dsm's.

Lisa T. Su: But now we are seeing your X 86 competitor Intel also adopt chip led or tied to technology as they call. It and then I think recently the manufacturing update they gave they said that two years ahead.

Lisa T. Su: In terms of incorporating gate, all around and backside our delivery.

Lisa T. Su: So let's assume they are right in there.

Lisa T. Su: Either caught up to TSMC or maybe they got ahead, what impact does that have on AMD and kind of the medium to long ago.

Yeah sure Vivek look we're all.

Lisa T. Su: Always looking at what's next right. So on the triplet technology I mean, we're sort of on the fourth generation of the triplet technologies I think we've learned a lot about how to optimize performance. There we are very aggressive with our adoption of leading edge technology.

Speaker Change #122: It's needed but.

Speaker Change #122: Those are only.

<unk> of the pieces. We're also focused on continuing to innovate on architecture and design. So I think the longer term question that you ask is I think we're expecting that.

Speaker Change #122: The competition is going to be on a similar process technology and even in that case.

Speaker Change #122: I think we feel like we have a very strong roadmap going forward and we will continue to.

Speaker Change #122: To drive both the CPU and the GPU roadmap very aggressively.

Speaker Change #122: And the next question comes from the line of harsh Kumar with Piper Sandler. Please proceed with your question.

Speaker Change #122: Yes.

Harsh V. Kumar: Thanks for letting me ask a question guys I have two questions. Let me start off with the accelerated aside.

Harsh V. Kumar: Can we get a lot from our customers is they want to understand the value proposition of BMI three hundreds. So Lisa I was hoping you could give us some understanding of price versus power comparison, our compute power and then today are you seeing your customers that are buying PMI 300 are they primarily buying it for inferencing.

Harsh V. Kumar: Or are they using it primarily for tooling and maybe for Jean Jean do you think is it possible for 300 to finish the year at a run rate of about $1 5 billion.

Lisa T. Su: Okay harsh so let me start to your question about the value proposition for <unk> 300.

Harsh V. Kumar: Again customers are using it for different reasons, but presume that there is a performance per dollar bench.

Harsh V. Kumar: Benefit to using AMD. So that's one piece of it.

Harsh V. Kumar: The other piece of it though is we intrinsically have.

Harsh V. Kumar: More bandwidth in memory capacity on 300 X compared to the competition and what that means is for large language models that are many tens of billions of parameters you make you could potentially do the workload and fewer gpus. So it's a substantial.

System savings and allows you to do much more work.

Harsh V. Kumar: Within the same system in terms of what customers are using <unk> 300 for today I would say there are a number of customers using it for large language model inferencing and there are also customers that are using it for for training. So I think the the whole point is being a strong partner when you put these.

Harsh V. Kumar: Systems in place.

Harsh V. Kumar: Sometimes mixed used systems. So they would be used for both training and inference.

Speaker Change #123: John we have time for two more questions.

Let me answer your question about the 300 <unk>.

Speaker Change #124: <unk> Q4, 2024 is it possible to get to one five.

Speaker Change #125: It is possible right because Lisa mentioned earlier, we will see sequential increase each quarter and the mall back end loaded in second half and we do have our supplies more than $3 5 billion and <unk>.

Speaker Change #126: So we'll continue to make progress with our customers so that the math, yes, it's possible, but right now we're really looking at.

Speaker Change #126: Focus on the execution of the car into $3 5 billion plus.

Speaker Change #126: And the next question comes from the line of Stacy <unk> with Bernstein Research. Please proceed with your question.

Stacy: Hi, guys. Thanks for taking my questions.

Stacy: The first one you talked about you expected a more shallow ram.

Stacy: <unk> 300, and it's clearly doing better than that.

Stacy: Some of the upside I guess in the near term is that being pulled forward from the second half or is this like a step up or is it more of a step up in demand both in the first half in the second half relative to what you were seeing before.

Speaker Change #127: How do I, how do I interpret that like shallow comment that you made sure Stacy I don't think it's a it's a pull forward of demand I think what it is is we.

Speaker Change #127: We want to.

Speaker Change #127: We wanted to see how long it would take for customers to fully qualify and get their workloads performance so that.

Speaker Change #127: That depends a lot on the actual engineering work, that's done and now that were let's call. It a quarter later.

Speaker Change #127: We've seen that it has gone really well so it's actually gone a.

Speaker Change #127: A bit better than our original forecast and as a result, we've seen stronger demand signals and customers are gaining confidence.

Speaker Change #127: Their ability to deploy a significant number of three hundreds this year.

Speaker Change #128: Got it thank you for my follow up.

Speaker Change #129: Want to ask Matts question, a little more directly you didn't quite answer it the $400 billion number you've got out there is that just silicon chips or is there hardware and servers and stuff like that in that number as well.

Speaker Change #129: In that number yes.

Speaker Change #130: I thought I had answered it but yes I'll answer it again it is.

Speaker Change #130: It is accelerator chips that is not systems, so think of it as.

Gpus.

Speaker Change #130: Six that will be there so those those types of things, but it includes obviously it includes memory and other things that are packaged together with the.

Speaker Change #130: But with the Gpus.

Speaker Change #130: Memory will be quite significant right, so memories big portion of it.

Speaker Change #130: And the final question comes from the line of Chris Danley with Citi. Please proceed with your question.

Christopher Brett Danely: Hey, Thanks for squeezing me in team.

Christopher Brett Danely: I guess question for Lisa.

Speaker Change #131: I don't think it will be one or two that are half the revenue Chris I think we're building. This as a really we're happy to see sort of the broad adoption as always with sort of the large cloud.

Partners, we might see.

Speaker Change #131: One or two that are higher than others, but I don't think you see the type of concentration that term that you mentioned.

Great and then just a follow up on somebody else's question.

Speaker Change #131: Sort of Intel's roadmap versus TSMC, So I'm sure you're sure you're intimately familiar with Tsmc's manufacturing roadmap and we've all seen them until opened up the kimono on on what they expect to happen over the next couple of years.

Speaker Change #131: Do you think in total it's going to close the gap somewhat with with your foundry over the next couple of years do you think there'll be able to maintain the lead.

Speaker Change #132: I feel very good about our partnership with TSMC, they continue to execute extremely well.

We'll see what happens over the next few years, but.

Speaker Change #132: I'd like to kind of reemphasize, what I said earlier.

Speaker Change #132: Even in the case of process parity, we feel very good about our architectural roadmap and all of the other things that we add as we look at our entire portfolio of Cpus Gpus, Gpus adaptive <unk> and kind of put them together to solve problems I.

Speaker Change #132: I think we feel really good about what we can do with our customers. So we're always going to be paying attention to sort of the process race, but I think.

Speaker Change #132: I think we feel very good about sort of our strategy and how do we continue to.

Speaker Change #132: Sort of push the envelope on the computing Roadmaps.

Speaker Change #133: Okay and that is the end of the question and answer session I would like to turn the floor back over to the AMD team for any closing comments.

Speaker Change #134: Great John that concludes today's call. Thank you to everyone for joining us today.

Speaker Change #135: And this concludes today's teleconference. You may disconnect. Your lines at this time. Thank you for your participation.

Q4 2023 Advanced Micro Devices Inc Earnings Call

Request a DemoDemo

AMD

Earnings