Advanced Micro Devices' (AMD) CEO Lisa Su Presents at Credit Suisse Technology, Media & Telecom Conference (Transcript)

| About: Advanced Micro (AMD)

Advanced Micro Devices, Inc. (NYSE:AMD)

Credit Suisse 20th Annual Technology, Media and Telecom Conference

November 30, 2016 11:30 AM ET

Executives

Lisa Su - President and CEO

Analysts

John Pitzer - Credit Suisse

John Pitzer

I would like to welcome everyone to the session this morning. My name is John Pitzer, I am the semiconductor analyst here at Credit Suisse. It's my distinct pleasure to able to introduce Lisa Su, who is the President and CEO of Advanced Micro Devices. We’re going to go through a fireside-chat conversation here over the next 20 to 25 minutes. We do have mics kind of roaming the room, so if you’ve got a question, please don’t be shy, raise your hand.

And with that, Lisa, I’ll kick things off, and I think it's always a good way to start off these fireside-chat sessions, especially if there is anyone in the audience that’s not intimately familiar with the AMD story. Just spend a few minutes positioning the Company. What's your core IP? What end markets you try to tackle that core IP with? And I guess especially given your perspective of having been in the seat for what about two plus years now, just over two years kind of the challenges you saw coming-in and kind of what you have addressed over those two year period?

Lisa Su

Yes, absolutely. So, John, it's great to be here. Thanks for having us. AMD is really a high performance computing company. So if you think about what our core IP is, it's really high performance microprocessors, and high performance graphics processors. And I think the exciting part of the market today is they are just huge growth markets. If we look at the TAM that we’re addressing, we think it's about a $50 billion TAM across gaming, data center and immersive platform, so we like the markets a lot.

I think what I have been trying to do at AMD is really focus us on our technology. We are a product company, so we live or die on great products. And the focus for the last two years has really been on execution of that roadmap. And so we have a multi-year roadmap. It's a multi-year journey. But I think we’ve made some good in-roads into getting our high performance graphics and processor roadmaps in a very strong position.

John Pitzer

How should we think about those roadmaps over the coming two to four quarters, because you do have some interesting new products that are coming out, Polaris near-term on the GPU side, next year Zen first as a high-end desktop product, and then there is the server market. So help us understand how you are thinking about the rollout out of these -- the roadmap over the next several quarters?

Lisa Su

The key for us is really execution for our customer set. Our customer sets want to see long-term roadmaps. 2016, this year, was really the year of the GPU. So our Polaris launch, we’re gaining share in the graphics market, as well as our custom business with game consoles is a very strong franchise. Sony, our customer just launched PlayStation 4 Pro. We’re very pleased, and proud to be in that system. Apple just launched their MacBook Pro, very pleased and honoured to be in that system.

And as we go forward, 2017 we really roll-out sort of our high performance roadmap. A few years ago, we decided that CPU performance was absolutely critical, and we needed to start from scratch. And so Zen is our high performance CPU that is really a significant, significant enhancement over what we've had in the past. That’s unscheduled. We’re launching that in the first half of 2017, first in desktop market and then going into the high performance server market. But we also have our Vega architecture and graphics that’s coming out for high performance graphics in the first half of '17. So, it’s a very busy time for us over the next number of quarters.

John Pitzer

Just sticking on the discrete graphics market, how should we think about your share aspirations with Polaris now ramping, Vega next year. Where do you think you can get your share back to within that market?

Lisa Su

Yes, so we’ve made very good progress, I would say, in the first couple of quarters of this year in terms of graphics. I think it’s nice, when you look at graphics, it’s really both in the consumer side on the channel, as well as in the OEM business. And then on the professional graphics market, there is a large opportunity. I think we’ve gained a good amount of share over the last few quarters. We’re going to continue consistently drive. We believe that there is no reason we can’t be 50-50 share, over-time, but it will certainly take some time to get there. The key thing is really enhancing our relationships with customers, because we believe that this is also important to have a very sticky business going forward.

John Pitzer

One of the interesting announcements within your GPU business was a couple of weeks ago at the Super Computer Conference where Google came out and said that they would be making available your GPUs in the cloud for acceleration applications. And most of the conversations around GPU and AMD are on the gaming side. Now, we’re starting to see within the data center and the acceleration side. Can you give us a little bit more detail on that relationship with Google? And, I guess, more importantly, how do you think about exploiting your GPU IP within the data center and acceleration market?

Lisa Su

Yes, so I think GPUs, right now, the market is on fire. I mean it’s a great market. Our strategy in GPUs is really to play across the entire market. We think we have the IP, the capability, the talent to do that. We started with gaining share in consumer, and that’s why Polaris was so important for us this year. As we go into more of the higher end markets with our next generation Vega architecture what you'll find is the hardware is very, very competitive.

Working with the cloud guys there is a lot of software work to be done, and so we have been investing in that area, and we’ll continue to invest in that area. I think GPU and the data center is certainly going to grow over-time. It’s still relatively small compared to the CPU market. But what’s unique for AMD is we have access to both high performance CPUs and high performance GPUs, which we believe over-time will be a critical advantage for us.

John Pitzer

And Lisa maybe you could help me understand, people the audience, because it’s clear that from a Silicon perspective you’re catching up to your closest competitor. How are we going to [cap] [ph] the software? Because if you look NVIDIA has put a lot of effort on the software ecosystem and one of the questions I get, I have a hard time answering it or quantifying, is where is AMD on that slope or in that process?

Lisa Su

Well, I think when you look at it, the hardware foundation is extremely important, and so there's certainly significant investments from us in that area. But on the software side, you actually see that you can make very rapid progress, particularly our focus is on providing an open compute environment. I mean we really believe in an open ecosystem, working with the entire ecosystem, trying to accelerate progress. If you look just on the gaming side, what we’ve been able to do with our -- with all the games developers, we’ve dramatically improved that ecosystem with device drivers and everything that’s necessary. We’re doing the same on the work-station and GPU server side. So, I view it as a great opportunity for growth over the next couple of years.

John Pitzer

That’s helpful. Switching gears on to the CPU side with Zen. Part of the survival mechanism that I use in tech is to try to block-out a lot of the noise, because we get noise every day. But there is a lot of chatter about the timing of Zen, and is it late, is it not on track. Wondering if you could just level-set us. When were your expectations for Zen on the desktop, Zen on the server? And have those targets moved over the last couple of quarters?

Lisa Su

So, I think it’s very fair to say, when you do a brand new high performance microprocessor, it’s going to take three or four years. And that’s the level-set. We’re actually very pleased with Zen. It's right on track. Our expectation is that we will launch in desktop in the first quarter of 2017. We’ve given little bit of information in terms of where we are relative to the performance aspects and there’ll be more coming shortly.

Customers have been actively testing and working with Zen over the past number of months, and it’s looking quite good. So, you should expect Zen in desktops in first quarter of 2017, Zen in servers in the second quarter of 2017, and Zen in notebook form factor second half of 2017. So, it’s a very -- actually it's quite a fast roll-out of Zen throughout our product roadmap.

John Pitzer

And before your time, but the historic knock on AMD has been they were always going to gain in unit market share, but not necessarily the profit market share. When you think about Zen, what part of the stack are you going after? And what are the implications as you gain unit share in Zen around your own gross margin profile?

Lisa Su

So, it was very important for us with Zen to be competitive really throughout the price bands. So if you look at Zen at the very high end in desktops, we believe we’ll be very competitive with the Core i7 type product line. When you look at Zen in servers, we believe we’ll attack good 80% of the TAM. So there is a lot of opportunity there. And from our standpoint, these are areas where it’s a significant ASP uplift from where we are today. So, what we want to do is provide both performance, but as well as price performance in these market segments.

John Pitzer

And what do you think a realistic share expectation of Zen in the server market. And I’ll preference it by saying, when I go back and look at when AMD had peak market share in servers, it was sort of 2005-2006 time period, and it was really based upon the Opteron platform. And if I remember back then, Opteron had a clear distinct advantage over what Intel could offer there around Memory I/O and bandwidth. Then you got up to about 25%-26% market share. So my question here is, help me understand is there a memory bandwidth with analog for Zen where you think you’re going to have significant better performance than Intel’s product offering, and kind of what share expectation do you think is reasonable?

Lisa Su

So, I think all of that is good context. It is true back in the Opteron days that our peak we were at 25-ish percent market share. And I think what we’ve always -- our aspiration is to provide significant value, it’s not just a price statement but it’s a technology and innovation statement. I think what we’ll have with Zen is an opportunity with certain workloads to really do very well. I think the differentiation for us is going to be not just in the CPU itself, but all of the I/O surrounding the CPU and then how we connect to the outside world.

Our view is that having both high performance CPUs and high performance GPUs will also give us some advantage in terms of how we interconnect those systems. So what's a reasonable share aspiration, look, I think today it's fair to say we’re less than 1%, so it's all upside. From here, it's a $15 billion market. It's very, very strong margin contribution. If I weren’t aspiring for double-digit share then I don’t think that would be aspirational enough. I think it will take us some time. We’re very cognizant of the fact that competing at the high end means it will take time to both win customer's confidence and trust for these applications but we’re well on our way with the design wins, the engagements, and the idea that we can provide something special with our CPU and GPU combination.

John Pitzer

And then you were one of the first to embrace and endorse the idea of heterogeneous compute, and in an interesting sort of way it's playing out a little bit differently than I think a lot of us thought, because we thought it was going to be an architectural battle between x86 and ARM. I would like to get your update on kind of where you are on that product offering? But the interesting heterogeneous dynamic that’s in play now is CPU versus GPU versus FPGA. Love to get your perspective on that architectural battle as well.

Lisa Su

Yes, it's a very, very good point. I mean we’ve been thinking about heterogeneous computing really for the last seven, eight years in terms of how it comes about. And I think what's true is we’re a firm believer that you’re going to need CPUs, you’re going to GPUs, you probably will need some FPGAs, and there will be some ASIC applications. So there is no one size fits all. However, what you will have is the ability to customize or mix and match the different flavors in a fairly cohesive way is very important.

And it's because the workloads are so different. There are lots of different workloads, whether you’re talking about in the cloud or in specific enterprise applications, and the mix between CPU and GPU, or CPU and FPGA, will change. And so we’re a big believer in the open ecosystem required around that, the open interconnect that’s required around that, and having good foundational elements. It's clear to us that heterogeneous computing is going to be more important, but there is not going to be a -- it's a one to four factor or one to 16 factor, or whatever it is. It's going to vary depending on the workloads.

John Pitzer

And then update us on your ARM roadmap?

Lisa Su

So, from an ARM standpoint, we believe the ARM architecture is certainly important in certain market segments. From our view, we offer ARM as part of our custom roadmap. From a standard product roadmap, though, all of our eggs are in the x86 basket, and the point is there is just so much TAM for us to address with x86 and GPUs that’s where we’re focusing the business.

John Pitzer

And you mentioned customization, which is a good segue into my next set of questions, just around your custom and semi-custom business. Maybe you can just remind us the strategy there, where the success stories have been. But more importantly, where do you see taking that semi, semi-custom business over the next three plus years?

Lisa Su

Yes, so the semi-custom business has been a real nice way for us to differentiate ourselves. And it's the idea that if you really want to optimize the system, you want to optimize hardware, software, and system attributes together. The best examples are what we’ve done in the game consoles with Sony and Microsoft we have built really a very nice franchise. I would say those systems, both the PlayStation 4 and the Xbox One, have been extremely successful by all accounts. And now we’re doing mid-cycle refreshes there.

We view customization as one of our differentiators. And so our job is to help our customer's products become better. And so there is a co-engineering aspect of it that’s important. And it’s not just custom silicon, but it’s also custom software and custom system and thermal envelopes that we can get to. And at the end of the day, sometimes I get asked, what does this mean. It's just a lot of extra cost. I view it as it’s extra value add for the customer, and it also gives us a stickiness and a long-term potential. If you look at what we’ve been trying to do with AMD over the last few years, it’s really to become a much higher player in the value chain of our customers solutions.

John Pitzer

Can you talk a little bit about the longer term profitability aspects of the semi, the custom, semi-custom business, because when you look at the CPU market and the GPU market, the road to higher profitability is pretty tangible as long as you execute. It’s unclear I think to lot of investors the profitability roadmap for that semi, semi-custom business because historically it's been a good business but not necessarily the most profitable business. I think what people trying to figure out is that something inherent about semi semi-custom, or custom semi-custom, or is that inherent more in the gaming market which was the first area of success, and as you branch out from gaming into other areas, should we expect higher levels of profitability?

Lisa Su

Yes, it’s fair to say. So, if you look at our margin profiles today, they are heavily weighted by our game console business. And our game console business tends to have gross margins that are below the corporate average. But frankly have operating margins that are actually very healthy, because it’s a model where customers actually paid for the engineering. And so the fall-through is very high. I think at a broader level though, at a broader level, we believe that margin expansion is extremely important in our business, certainly in the semiconductor business margin expansion is critical.

As we get into the higher end segments, such as high-end PCs, high-end servers, high-end GPUs. You will see our margin profile shift up with that product mix. As it relates to the custom business I think it’s more market specific than business model specific. So, if we were doing custom in the data center, I think you would see higher margins versus game consoles, which happen to be, let’s call it very high volume silicon TAM but tend to be more price sensitive.

John Pitzer

That’s a good segue into my next question. I got you’re going to pre-announce any new customers in the custom, semi-custom business. But as you think about moving up to those higher market segments within that business, what’s the timeline? How do we think about milestones over the next, call it, four to eight quarters?

Lisa Su

Well, I think -- if you look today at the semi-custom business, it's been primarily game console driven. And it’s been again very, very pleased with the franchise that we’ve built around that. We do have engagements outside of game consoles, and that will ramp, let’s call more in the 2018 timeframe. The other thing you should expect is I talked about our Zen roadmap for our products, so in terms of desktops servers and notebooks. But one should expect Zen in our semi-custom roadmap as well as we go -- as we look beyond 2017 into the '18, '19 timeframe. And so, we really do view this as we’re developing foundational IP that can go into a number of different markets. And we have good prospects in those areas.

John Pitzer

I had Intel on stage yesterday talking about the autonomous driving market, and it might be 10 to 15 years out. But if you look at an endpoint, you can make the argument that self driving car is effectively a server on wheels. And I think Intel is selling about 20 million units in their data center business today, and it is about 100 million cars they could get sold every year. So, you can make argument that market is five times larger than the data center opportunity at least today. And again the end point could be a decade plus out. I guess I was a little bit surprised with the Tesla announcement where they came out chose NVIDIA as their control vehicle for level three autonomous driving. But what surprised me was just -- but the good things they had to say about you. Maybe you can…

Lisa Su

Of course, I mean that surprise you, John...?

John Pitzer

Only because it’s not a very -- I mean, as you have been very focused on the end markets that you had talked about attacking, and that was not one that you guys have talked historically. So, I would love to get your thoughts on how you’re thinking about that market?

Lisa Su

Well, I think it goes to this view of where is high performance important, okay. And I think you said it very well, an autonomous car maybe a high performance server on wheels. The fact is that, there are very, very few players that have access to high performance CPUs and high performance GPUs. And I think we can name them unless in one hand. What you should think about with us is, we really believe that our hardware is competitive across the entire spectrum of these markets, but we’re going to be very targeted in how we go after the markets.

And they really -- it starts with a notion of, as I said in GPUs, we started with consumer, because it’s a place that we know well and that market moves very fast. And we’ve made good progress there. You will hear us talk more about our professional workstation roadmap, as well as our server GPU roadmap. And I think down the road we will as well be in the autonomous vehicle range. But we’re really looking at how do we go market-by-market in very specific fashion.

John Pitzer

That leads to another good question. I mean, to an extent over the last several years, your opportunity set, I wouldn’t say is limited but it’s been constrain by your financial resources, your profitability on your P&L and your balance sheet. You took a big step this year in I think helping to show up the balance sheet. How do we think about the growth in R&D from here, because there is clearly a lot of incremental opportunities you could be going after as your financial performance improves. How do we think about R&D growth relative to revenue growth?

Lisa Su

Yes, I think it’s fair to say, we’ve made good progress on the balance sheet and that was extremely important with our debt transaction that we did a couple of months ago. I think, going forward, the way you should think about it is, we are a growth company. We are aspiring to be a growth company, and with and that comes a needed investment in R&D. The fact that we have been able to retire some debt and refinancing debt actually gives us a more capacity to invest in R&D.

So, you should see us continue to ramp-up investments. In terms of areas, particularly around GPU and the software required to enable some of these new applications. But we’re going to be very diligent in the process, right, on building long-term roadmap for the company. I think, we’re excited about some of the progress we’ve made so far. But we also are extremely excited about the opportunities we have going forward. And the nice thing about it is, there are so many opportunities, it’s up to us to really prioritize which ones and in which order to go after them.

John Pitzer

Two other important milestones that happened for the company this year, first was the China IP licensing deal. Wondering if you can just give us an update on how that’s progressing the milestone payments? But more importantly, you’ve talked about potentially more opportunities of monetizing your IP. Maybe you can elaborate a little bit further on how those efforts are going?

Lisa Su

So, we did complete our China joint venture earlier this year. It’s a joint venture in servers building server chips, specifically for the China data center market. It’s going very well. Actually the joint venture has ramped-up, both in terms of engineering resources and just overall capability. I think we view it as a way to augment our own product sales. So we’ll continue to obviously sell AMD branded CPU chips into the China market. But this allows us to overlay something that’s designed more locally for the market.

And the way I view licensing for AMD is, again, we have over 10,000 patents, some of them are very seminal patents in high performance CPUs and GPUs we have a lot of technology capability. And so it's an overlay to augment profitability and value creation for the company. First and foremost, we’re a product company and we’re going to spend time on building great products. But if we can find these opportunities to augment the business with licensing, we will do so. And particularly as it opens up new markets, so the China market is number two data center market out there. It's critical to have significant offerings in that area. So we’re very pleased with that.

John Pitzer

So timing of IP deals are always extremely difficult to predict?

Lisa Su

Yes.

John Pitzer

Would you like to provide any milestones for calendar year '17, is there an expectation that we should see one or two more of these? Or is it just too difficult to tell?

Lisa Su

I would choose not to put that market out there, but I will say we have a good view that you should see additional IP deals from us. And they're all be slightly different, they’re all very custom but, yes, we’re making good progress.

John Pitzer

And then the other important milestone is that, was the renegotiation of the wafer supply agreement with the GlobalFoundries, and it's clear that GlobalFoundries remained a very important supplier and partner for AMD. But I'm kind of curious what is that renegotiation gives you as you go to market over the next couple of years?

Lisa Su

Yes, so as you said, GlobalFoundries is a very important partner for us. We have a long term wafer supply agreement. But what I was trying to do with this renegotiation is really give us an opportunity to have a very clear long-term access to high performance technology and more flexibility from a technology roadmap standpoint. So, one of the knocks on AMD perhaps has been we’ve had some execution issues in the past. I'm extremely focused on execution and how do we ensure that we have strong execution. And so the wafer supply agreement gives us some additional flexibility as it relates to future technologies.

And in particular, this year we ramped actually a large number of chips in both 16FF with TSMC and 14LPP with GlobalFoundries. There was a large undertaking. And having two suppliers actually was helpful, frankly, because we’ve turned over the majority of our product line into FinFET over the last seven or eight months. As we look forward, we believe that 7 nanometer technology generation is going to be very critical, it's going to be a big node, whereas 10 is probably going to be a smaller node. And so the wafer supply agreements, we will certainly work very closely with GlobalFoundries, but we’ll also work with other foundry suppliers for 7 nanometer.

Question-and-Answer Session

Q - John Pitzer

Any question in the audience?

Unidentified Analyst

[Technical difficulty] because of a certain game is coming out and also you both had new architectures. So I am curious about your view on sustainability of the unit growth and also ASPs continuing to rise?

Lisa Su

Yes, that’s a good question, and just for the audience around the GPU market expansion. Look, it was definitely a very good year for GPUs, particularly this past quarter with our expansion and then just the overall market expansion. I do think some of that is due to the fact that we both have new architectures out there. So, that’s some of the expansion. But the fact is PC gaming is one of the few areas that we are seeing overall growth, and the mix is towards the higher end ASPs. So, we like the market very much, I think it’s a good market for us and we’ll see how it continues.

John Pitzer

Great. With that I think we’ve run out of our time in this session. But I really want to appreciate Lisa for joining us this morning.

Lisa Su

Very good. Thank you, John.

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.

THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.

If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!