Advanced Micro Devices has had a few tough years as it struggles to stay afloat financially and compete with Intel, the world’s biggest chip maker.
Yet Lisa Su, chief executive of the Sunnyvale, California-based chip design company, remains an optimist. She told us in an interview that 2016 is going to be a good year for AMD, with products that are in the pipeline finally seeing the light of day. Among those is AMD’s new ARM-based 64-bit processors for data centers, which are being launched in volume sales today.
AMD is also working on new chips based on the FinFET manufacturing technology, which will result in a whole new line of microprocessors, accelerated processing units (APUs, which combine graphics and processor on a single chip), and graphics chips from the company’s Radeon Technologies Group.
But like we said, it’s been a struggle. AMD sold off part of its test and manufacturing business in October for $436 million. Su’s predecessor, Rory Read, left the top post earlier than expected in October 2014, in part because of AMD’s continuing financial struggles. Su didn’t talk much about financials, as the company is reporting its earnings on January 19 for the fourth quarter of 2015. AMD is expecting revenues to drop 10 percent in the fourth quarter, compared to the previous third quarter.
One of the bright spots is AMD’s semi-custom business, which includes the accelerated processing units that it builds for all three major video game console makers — Sony, Microsoft, and Nintendo. And AMD has more semi-custom chip designs in the works.
We participated in a small roundtable of journalists in a Q&A session with Su during the 2016 International CES, the big tech trade show in Las Vegas last week.
Here’s an edited transcript of our interview.
Question: How are things going?
Lisa Su: 2016 is going to be a good year. The notion of what we’re trying to do at AMD is about bringing out great technology, great products. 2015, we level-set a bunch of things in terms of what we wanted to do with the product road map, particularly the key transitions over the next 18 to 24 months. There’s a lot of payoff that should be coming through in 2016 and the first half of 2017. We’re very excited about that.
I know that there was a lot of conversation about our Polaris launch here at CES. Hopefully you’ve all seen the demos and the numbers. When I step back and look at what’s important to AMD, it’s about graphics leadership — visual computing leadership — as well as a strong computing experience. We have the capability to integrate those two together.
On the graphics side, this Polaris beginning for us is one of a number of things that you’ll see us do in the graphics business to elevate the importance of strong hardware coupled with software in an open ecosystem, to bring together the best that visualization can do. We’re very excited about it. I know many of you have met Raja Koduri. He’s a visionary in the industry. We see giving him the keys to our Radeon technologies is going to be very exciting for the industry and for us over the next 12 to 18 months.
On the computing side, there’s a lot of anticipation for Zen. We’re very much committed to deliver Zen on the time scale that we have previously stated, which is: 2016 you’ll see samples, hardware, some early volume, but 2017 is the first full year of volume in servers and desktops. We’re excited. You have a road map that’s very compelling. It’s on us to execute. That’s where I’d like to be, starting in 2016.
Question: What would you like people to be saying this time next year about AMD?
Su: I’d like them to say AMD is back.
Question: Do you feel like you have enough coming to create that?
Su: I do. It’s never one thing that makes the difference. It’s the combination of years of work on the CPU road map, getting to a competitive CPU standpoint. That’s critical for us. On graphics, the overall industry is excited about what’s possible with virtual reality, immersive computing, augmented reality. That makes graphics overall very exciting, but particularly relative to AMD.
We know what it’s like to be number one. We’ve been number one before. We have every opportunity to get back there. We’re very focused on these core markets, where some of our competitors are focused on many other things.
Question: If you had to look at, say, three years, what sort of company is AMD at this point in time? Is it the culmination of the trends you’re talking about, or do you plan to expand in other areas?
Su: We have three core focuses. Graphics leadership across the market segments — if you look at consumer, professional, and integrated, that’s a key focus area for us. Strong computing leadership relative to both servers and PCs. We think computing will continue to be important as mobile has gone up and saturated.
The thing that differentiates us a bit more is our ability to integrate those for custom applications. I view us as a strong partner for those companies that want to vertically integrate. The best example is game consoles, but there are many other examples of companies that want to vertically integrate, that know you have to optimize hardware, software, and systems together. AMD is the best partner in the world for that. That’s how we see expanding into different applications. Whether that’s cloud and enterprise, or consumer focus, there are lots of opportunities for us.
Question: You mentioned that this past year has been about setting some plans for the future. It’s been a relatively quiet year in terms of people talking about AMD. Does that mean 2016 is a really important year for getting people to talk about AMD again and be excited about the future for AMD?
Su: Absolutely. The way I think about it is, what excites people is great products and great technology. When you think about the transitions we’re going into in 2016, bringing out a whole new set of FinFET products across all of our product lines starting in 2016, that affords a step function improvement in performance and performance per watt. Bringing out some of the new architectural features, both hardware and software, that we’re talking about.
The foundation that we were setting in 2015 is: We’re going to focus on our products. As a result, we took some time. We weren’t out there saying that we’re going to take over the world. Not until we have something to show you. But the plans were, we wanted to be able to show you FinFET product at CES. I’m proud of the team for doing that. We’ll ship that in the middle of 2016. We see everything on schedule. You’ll hear more about Zen progress as we go through 2016, but it’s very much — we want to show the world what AMD is capable of in 2016.
Question: What do you think you’ve learned from the game console cycle? You guys dominated that last generation, made all these chips, and moved on to more semi-custom businesses. What are you learning from game consoles that you’d take to that larger business?
Su: A lot of the initial game consoles were — let’s call it consolidating some of the AMD IP that was there. Going forward we’re seeing that the innovation may very well come from the custom applications. The drive that these leading OEMs have in terms of where to take the business gives us a unique view into road maps five years out. That comes back and is very synergistic with our standard product business.
I see the two as hand in hand. We have a standard product business that’s very much based on industry, market trends, and then we have a custom business that’s very much based on divisions of some of the leading OEMs in the industry. You bring those two together and you have a lot of visibility into what’s important.
Question: Would it be fair to say that semi-custom business for the console market is your most stable business at this point in time?
Su: Because of the game console cycle being a five- to seven-year cycle, it affords a bit of stability once you get into it. But I wouldn’t take anything for granted. It’s extraordinarily competitive in this market. The importance of having strong IP, strong relationships, and setting a five-year vision for where we’re going is critical to these customers.
Even today, you still see PlayStation 3 and Xbox 360 selling. There will be a lifetime for these consoles going forward.
Question: One thing that’s happened in previous generations of the console business, there have been changes in the architecture of the chips mid-cycle in some consoles. Will that happen in this cycle?
Su: Without talking about any particular console road map, because our console customers are very sensitive, there will be opportunities to cost reduce. As we know, the consoles tend to be very sensitive to price point. As the price comes down, the console volume goes up. There will be opportunities to do that in this cycle as well.
Question: Do you think you’re building up enough new IP to justify new generations of consoles?
Su: It’s less about our IP and more about the user. Are there user benefits to more graphics horsepower? I think the answer is yes. For that reason, if there’s an application that can use it — I don’t know if you guys visited our booth, but I spent 15 minutes there with some of the guys and they showed me the latest Oculus and HTC demos. They’re good, but there’s a lot of room. Those things are still very expensive relative to the consumer price point.
Both console and PC gaming have a long way to go yet to make those $199 price points. When you think about the real mainstream market, when you get to $199 a lot of people have access. There’s a lot of opportunity yet.
Question: Is VR going to be a core focus for you, to aid these companies that are bringing VR to the mainstream?
Su: Absolutely. Both from a hardware and a software standpoint. VR is a great showcase for technology. The focus has been very much around gaming, but there are many other interesting applications outside of gaming when you look at medical, education, other interactive opportunities. We can do a lot on hardware to get the price point into the right range. We can do a lot on software to enable developers to work more efficiently.
Question: When can we expect to see enough content to drive mainstream users to VR?
Su: That’s the question. It’s going to evolve over time. We’ll have some early stuff in 2016, but as we go through the next couple of years we’ll need to really get the content going where it needs to be.
Question: Whenever you mention VR to many consumers, they think of those ‘90s headsets and the collapse of VR. Do you feel like this generation will be the one to hit the mainstream?
Su: I believe this generation is very credible. My perspective is, over the last 12 to 18 months, we’ve been watching the progress. It’s gone so fast. Every three months the content is better. The user experience is better. The feeling of it is there. The way people react to it — you guys have all seen it. It’s very easily accessible. You don’t have to imagine what the VR experience is.
So yes, I absolutely think this generation is capable of becoming mainstream. We have to continue to work hard on all the pieces of the ecosystem. You guys have seen it. There’s groundswell in all parts of the ecosystem.