Thomson Reuters StreetEvents Event Transcript E D I T E D V E R S I O N Q4 2016 NVIDIA Corp Earnings Call FEBRUARY 17, 2016 / 10:00PM GMT ================================================================================ Corporate Participants ================================================================================ * Arnab Chanda NVIDIA Corporation - VP of IR * Jen-Hsun Huang NVIDIA Corporation - President & CEO * Colette Kress NVIDIA Corporation - EVP & CFO ================================================================================ Conference Call Participiants ================================================================================ * Deepon Nag Macquarie Capital - Analyst * Harlan Sur JPMorgan - Analyst * Hans Mosesmann Raymond James & Associates - Analyst * C.J. Muse Evercore ISI - Analyst * Steven Chin UBS - Analyst * Vivek Arya BofA Merrill Lynch - Analyst * Gabriel Ho BMO Capital Markets - Analyst * Matt Ramsay Canaccord Genuity - Analyst * Joe Moore Morgan Stanley - Analyst * Rajvindra Gill Needham & Company - Analyst * Mark Lipacis Jefferies LLC - Analyst * Sanjay Chaurasia Nomura Securities - Analyst * Ross Seymore Deutsche Bank - Analyst * Chris Rolland FBR & Company - Analyst * Chris Hemmelgarn Barclays Capital - Analyst * Brian Alger Roth Capital Partners - Analyst ================================================================================ Presentation ================================================================================ -------------------------------------------------------------------------------- Operator [1] -------------------------------------------------------------------------------- Good afternoon. My name is Ash, and I will be your conference operator today. I would like to welcome you to the NVIDIA financial results conference call. (Operator Instructions) I will now turn the call over to Arnab Chanda, Vice President of Investor Relations at NVIDIA. You may begin your conference. -------------------------------------------------------------------------------- Arnab Chanda, NVIDIA Corporation - VP of IR [2] -------------------------------------------------------------------------------- Thank you. Good afternoon, everyone, and welcome to NVIDIA's conference call for fourth quarter of 2016. With me on the call today from NVIDIA are Jen-Hsun Huang, President and Chief Executive Officer; and Colette Kress, Executive Vice President and Chief Financial Officer. I'd like to remind you that today's call is being webcast live on NVIDIA's Investor Relations website. It is also being recorded. You can hear a replay by telephone until the 24th of February 2016. The webcast will be available for replay up until next quarter's conference call to discuss Q1 financial results. The content of today's call is NVIDIA's property. It cannot be reproduced or transcribed without our prior written consent. During the course of this call, we may make forward-looking statements based on current expectations. These forward-looking statements are subject to a number of significant risks and uncertainties, and our actual results may differ materially. For a discussion of factors that could affect our future financial results and business, please refer to the disclosure in today's earnings release, our most recent Forms 10-K and 10-Q, and the reports that we may file on Form 8-K with the Securities and Exchange Commission. All our statements are made as of today, the 17th of February 2016, based on information currently available to us. Except as required by law, we assume no obligation to update any such statements. During this call, we will discuss non-GAAP financial measures. You can find a reconciliation of these non-GAAP financial measures to GAAP financial measures in our CFO Commentary, which is posted on our website. With that, let me turn the call over to Colette. -------------------------------------------------------------------------------- Colette Kress, NVIDIA Corporation - EVP & CFO [3] -------------------------------------------------------------------------------- Thanks, Arnab. Revenue reached a record in the fourth quarter, totaling $1.4 billion, up 12% from a year earlier, up 7% sequentially, and above our outlook of $1.3 billion. Our full-year revenue crossed above $5 billion to a record $5.01 billion, which was up 7% from the previous year. Quarterly growth was broad-based, with expansion across each of our four market platforms -- gaming, professional visualization, datacenter and automotive. Pacing gains were our GTX gaming platform, our datacenter platform, powered by deep learning, growing adoption and automotive. Viewed from a reporting segment perspective, Q4 GP revenue grew 10% to $1.18 billion from a year earlier. Tegra Processor revenue was up 40% to $157 million. NVIDIA's strategy remains sharply focused on creating platforms for our key markets. Our progress stems from our success in creating strong products that are targeted at growth markets. In Q4, our four market platforms contributed more than 85% of revenue, up from 78% a year earlier. Our growth platforms collectively increased 23% year over year. First let's start with our gaming platform. Gaming revenue grossed 25% year on year to $810 million, with good momentum carrying forward from the previous quarter. Maxwell-based GeForce GTX processors continue to lead our gaming growth, combined with growing anticipation for VR and the launch of holiday blockbuster games. The GeForce GTX 970 GPU stands as the world's most popular graphics card on the Steam gaming platform. And we continue to get strong traction with our GeForce GTX GPUs that power gaming notebooks. That includes the recently launched GTX 980 for notebooks, which has enabled a new category -- enthusiast-class VR-capable gaming notebooks. Excitement is growing around VR gaming, a key theme of last month's Consumer Electronics Show. We unveiled there our GeForce GTX VR Ready program to help gamers choose the best hardware for an immersive VR experience. And Oculus, which has now opened pre-orders for the Rift headsets, has exclusively certified GeForce GTX systems as being ready for VR. GeForce sales are driven by the launch of great gaming titles, and that again proved true this past holiday season. Fallout 4 was among the standouts, recording more than $750 million in sales in its first 24 hours. Other major hits were Star Wars Battlefront, Call of Duty: Black Ops 3 and Rainbow Six. We remain pleased with the continued success of GeForce Experience, our gaming platform that automatically optimizes your PC settings for each game and downloads the greatest game-ready drivers. At the end of January, just 2.5 years after its introduction, GeForce Experience subscribers stood at $76 million, up 37% on the year. Moving to professional visualization. Quadro revenue increased 7%, both sequentially and year over year, to $203 million. The refresh cycle of workstations continued to improve during the quarter, driven in part by new workstation configurations in the market. While VR is often portrayed as a consumer play, we are also excited by its potential in enterprise, particularly in areas such as medicine, architecture, education, and product design. Audi now has 20 virtual showrooms, with several hundred expected later this year, that lets customers experience new models, customize them in real-time, and take them for a virtual spin. In a very different application, a startup called Surgical Theater uses flight simulator technology and multiple GPUs to allow surgeons to use VR to fly through a patient's anatomy and rehearse complicated procedures before making the first cut. In datacenter, inclusive of Tesla and GRID, revenue rose 18% sequentially to a record $97 million, up 10% year on year. This reflects the extraordinary rise of deep learning, a field in which we are now engaged with nearly 3,500 companies and organizations, as well as growth in the number of high-performance computing applications that are GPU-accelerated. During the quarter, we launched key products for this market. And a number of the partners provided updates to their own work in this area that underscores the central role of the accelerated GPU platform. A key development came during November's Supercomputing 2015 Conference, with the release of the latest list of the worlds top 500 fastest supercomputers. It showed that more than 100 of these systems are now using accelerators. Two-thirds of these use NVIDIA accelerators, up 50% on the year. For hyper-scaled datacenters, we announced a platform that lets web services companies accelerate machine learning. It consists of both the NVIDIA Tesla M40, the most powerful accelerator designed for training deep [nolla] networks, and the NVIDIA Tesla M4, a low-power small form-factor accelerator for machine learning inference. Web services companies have enthusiastically embraced this trend. Shortly after our hyper-scale announcement, Facebook disclosed that it will use the Tesla M40s to power its next-generation computing system for machine learning applications. And earlier this month, AliCloud, Alibaba's cloud computing business, announced it will work with us to promote China's first GPU-accelerated, cloud-based, high-performance computing platform. They joined other web services giants embracing GPUs for machine learning. During the quarter, Google outsourced its TantraFlow deep learning framework, which can be accelerated on GPUs. Microsoft Computational Network Toolkit was integrated with Azure's GPU Lab, enabling neural nets for speech recognition that are up to 10 times faster than their predecessors. And IBM revealed that its Watson systems are now using GPUs. Progress continues to be made in our GRID virtualization platform, which enables companies to deliver graphs-rich applications to employees on any device, anywhere. More than 100 companies are participating in an accelerated deployment program. Turning to automotive. Automotive revenue was a record $93 million, up 18% sequentially, and up 68% year over year. One of the biggest stories at CES was the introduction of our NVIDIA DRIVE PX 2 self-driving car platform, which utilizes artificial intelligence to address the profoundly complex challenge of autonomous driving. As many of you saw, DRIVE PX 2 is a supercomputing platform the size of a lunchbox that processes 24 trillion deep learning operations a second, and delivers 8 teraflops of processing power, equivalent to that of 150 [metso pros]. It is a flexible platform that automotive developers can scale from one to four processors. And it can utilize passive cooling or integrate seamlessly with the water-cooling systems of self-driving EVs. Capable of fusing data from cameras, lidar, radar and ultrasonic sensors, it creates a full 360-degree understanding of what is happening around the vehicle. It localizes the vehicle on an HD map, and it determines a safe path forward for using deep learning techniques. Volvo, well-known for its safety and reliability, will be the first to develop DRIVE PX 2, using it as the brain force fleet of 100 self-driving cars to be publicly available next year in its hometown of Gothenburg, Sweden. Just a couple weeks ago, the first autonomous shuttle, the WEpod, incorporating our deep learning platform, took its inaugural trip on public roads in the Netherlands, where it can be summoned with a smart phone app. The DRIVE PX 2 launch generated enormous interest around the world from car makers, Tier 1 suppliers, and others. We are now collaborating with more than 70 companies that are developing self-driving car technologies. Finally, our OEM and IP business was $198 million, up 3% sequentially, driven by the seasonal demand for notebooks. Now turning to the rest of the income statement. GAAP quarterly gross margin was 56.5%. Non-GAAP gross margin was a record 57.2%, slightly above our outlook. GAAP and non-GAAP gross margins increased from a year ago. GAAP operating expenses for the fourth quarter were $539 million, inclusive of $34 million of restructuring and other charges. Non-GAAP operating expenses, including litigation charges, were $445 million, in line with our outlook. For full FY16, our non-GAAP operating expenses were $1.72 billion, including litigation costs. Our focus on rigorous execution and enhancing efficiencies enabled our core operating expenses to remain flat from FY15, as we focused on expanding operating margins. GAAP operating income for the fourth quarter was $252 million. Non-GAAP operating income was $356 million, up 26% from $283 million a year earlier. GAAP net income was $207 million. GAAP earnings per diluted share were $0.35, including $0.04 of restructuring and other charges. Non-GAAP net income was $297 million. Non-GAAP earnings per diluted share were $0.52, an increase of 21% year over year. Now turning to some key balance sheet items. At the end of Q4, our cash and marketable securities balance was just over $5 billion. During the quarter, we paid $62 million in cash dividends, and we closed our accelerated repurchasing agreement with an additional 4.3 million shares returned. As a result, we have turned to shareholders an aggregate of $800 million in FY16, meeting our intention we communicated at the start of the fiscal year. Over the past four years, we've returned more than $3 billion to shareholders representing 98% of our free cash flow. As part of our ongoing commitment to deliver shareholder value through capital return, our intention is to return $1 billion in FY17 through quarterly cash dividends and share repurchases. For FY16, revenue reached a record $5 billion, up 7%. Our growth platforms increased 26% year on year. Non-GAAP gross margin was a record 56.8%, up 100 basis points on the year. Non-GAAP operating income grew 18% to $1.13 billion, with operating margin expansion up more than 200 basis points to 22.5%. Non-GAAP EPS grew 18%. Now turning to the outlook for the first quarter of FY17. We remain excited about our business prospects. Gaming remains strong and eSports, VR, and new exciting games will lift it further. GP-accelerated datacenters are expanding in both HPC and cloud, driven by the growth of deep learning. And autonomous driving continues to move forward. We have excellent positions in each of these growth markets. We expect revenue for the first quarter of 2017 to be $1.26 billion, plus or minus 2%. Our GAAP and non-GAAP gross margins are expected to be 57.2% and 57.5%, respectively, plus or minus 50 basis points. GAAP operating expenses are expected be approximately $500 million. Non-GAAP operating expenses are expected to be approximately $445 million. GAAP and non-GAAP tax rates for the first quarter of FY17 are both expected to be 19%, plus or minus 1%. Further financial details, including the CFO Commentary and revenue by market platforms are available on our IR website. We will now open the call for questions. We ask you that you limit your questions to just two. Ash, would you please poll for questions at this time? ================================================================================ Questions and Answers ================================================================================ -------------------------------------------------------------------------------- Operator [1] -------------------------------------------------------------------------------- Certainly. Our first question comes from the line of Vivek Arya with Bank of America Merrill Lynch. -------------------------------------------------------------------------------- Vivek Arya, BofA Merrill Lynch - Analyst [2] -------------------------------------------------------------------------------- Thanks for taking my question, and congrats on the very good growth and execution. My first question, Jen-Hsun, is on the gaming segment. For the last two years, it has grown at over 30% a year. I don't know many other multi-billion-dollar businesses and semis that are growing at this pace. The question really is, how sustainable is this growth? I understand the drivers. But could you help us ballpark that going forward over the next two or three years? Do you think of this as a 10%, 15%, 5% growth opportunity? Any guidance there would be extremely helpful. -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [3] -------------------------------------------------------------------------------- Yes, thanks a lot, Vivek. First of all, I think you captured the essence of it in your question. GeForce is really not a chip business anymore; it is really a gaming platform business. When you think about it from a gaming platform business, it has to be thought of in the context of the low gaming ecosystem and the gaming industry. It is $100 billion large. When you think about it that way and you drive the business that way and you create value that way, I think the prospects for our growth there is still quite significant. There are several different ways we can grow the market. First of all, when we introduce new game platforms -- and this last couple of years, we've introduced Maxwell, it's the most successful gaming platform we have ever introduced. The installed base of about 100 million GeForce gamers in the world has an opportunity to upgrade to a new platform. Another reason why we can grow is because the production value continues to increase, the graphics richness continues to increase. And we do that by inspiring the industry, providing a technology that helps it include our technology in a much easier way. And the way that we do that is called GameWorks. All the physical simulations, all of the visual simulations, all the lighting simulations and all of the things that makes games beautiful today are easy to include by just supporting GameWorks. It has been an enormous success for us. And of course, developing companies are still doing incredibly well. There are many countries around the world that are just starting to get into PC gaming. Southeast Asia is growing incredibly. And then not to mention that gaming is no longer just gaming. Gaming is all about sports now. But we're starting to see a new culture and a new dynamic in gaming even beyond that. It's really becoming a platform by which people could share, and a platform by which they could artistically express themselves. If you look at some of these games today, it's something you enjoy well-beyond just playing the game. You use the game as an editor to tell stories. And so these games like GTA 5, it's just fantastic for telling stories. And so you can see now that the gaming platform is going beyond games, it's going beyond sports, and now it is a creative platform. So there are just all these wonderful ways that the game industry has continued to be vibrant. And my sense is that we're going to continue to grow with it. -------------------------------------------------------------------------------- Vivek Arya, BofA Merrill Lynch - Analyst [4] -------------------------------------------------------------------------------- Got it. And as a follow-up, Jen-Hsun, on your automotive business, as you move from entertainment systems that are graphics-intensive to more advanced computing systems, do you think it changes the competitive environment? And what I am referring to is, we have seen Qualcomm and others enter the segment, and they are making the case they can integrate a lot of different [piece] parts and tie back to their processor. And I'm trying to draw a parallel with what you had on the smartphone side, where you were specializing in one part, but others could integrate other parts and become more successful. Is there going to be a similar situation in orders? Or do you think we should read it in a different way as to how your competitive situation can evolve in orders as computing becomes a bigger part of the application? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [5] -------------------------------------------------------------------------------- Well, we have to be mindful of competition. And there's a lot of ways to compete, there's a lot of different ways to bring value. And surely what you describe is one way to compete. Those aren't really the segments in a market that we'll address. The way we think about infotainment is, there are segments of the infotainment market -- surely the parts of the market that we serve incredibly well -- with the richness of the displays, the number of the displays. And how the displays are going to be used in the coming years are going to continue to expand. You know that display costs are coming down, and outlet displays are becoming cheaper. There are so many different ways to bring visualization into the car, to enhance the driving experience. You can also imagine how artificial intelligence technology can change the way infotainment systems are even used. One of our strategies of course, is to bring artificial intelligence technology to enhance how the driver communicates with the car. So there is all kinds of new technologies that we're going to deploy into the infotainment system, leveraging our expertise in deep learning. The part of the market that you are starting to talk about is the segment of the market that we really introduced into the marketplace, which is the autonomous driving computer. We started talking about it several years ago -- I think it was like three years ago, at CES, when I introduced the DRIVE PX. Where we imagined that in the future, a car would also have a supercomputer inside that is powered by deep learning, that is powered by artificial intelligence. And that takes in the sensor input continuously from the car, what's surrounding the car. And infer from it the appropriate thing to do. That vision three years ago seemed a little bit, if you will, outer space. But I think that it is very clear now that the technology that we are bringing to bear, deep learning, is really the best approach for helping car companies go beyond ADAS, which is going to be a commodity in the coming years, as you can imagine. Go beyond ADAS, and move towards assisted driving to full autonomous driving. So I think that we can add a lot of value there. PX2 was really invented to allow OEMs to scale that entire range, from assisted driving all the way to fully assisted driving. And that is one of the reasons why we can support one chip all the way up to four chips, from passive cooling all the way up to integrating directly into the self-driving EV water-cooling system, that is quite available for most EV cars with liquid cooling. So I think that our strategy there is going to work out quite well. We add a lot of value. It is very algorithm-rich, it's very software-rich. And I think our DRIVE PX platforms is really quite state-of-the-art. -------------------------------------------------------------------------------- Operator [6] -------------------------------------------------------------------------------- Our next question comes from the line of Mark Lipacis with Jefferies. -------------------------------------------------------------------------------- Mark Lipacis, Jefferies LLC - Analyst [7] -------------------------------------------------------------------------------- Thank you for taking my questions. The first question, on the TDS business, how -- I'm sorry, the Tegra development services, you noted that, that was an important driver of growth. Can you give us a little bit more color on that business? How big is that? What end markets are you working with? What are you helping customers do? Can you hear me? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [8] -------------------------------------------------------------------------------- Actually, I'm just trying to figure out what your question was. Let me see. I guess I'd be reluctant to announce anything today, but there are semi-custom businesses that people need our help on. And we are open for business to help select partners develop proprietary systems that leverages the wealth of technologies that we have, whether it is in visual computing or deep learning or supercomputing. So we can create systems and products and services that the world has never had before. That's an area that I think is of interest to us. It's an area that we will likely see a lot more success in the future. But there is not much to really announce today. -------------------------------------------------------------------------------- Mark Lipacis, Jefferies LLC - Analyst [9] -------------------------------------------------------------------------------- Okay, fair enough. And second question on the gross margins. They've gone over the last three years from the low-50%s, pushing through 57% now. At what point do these [asset tow out]? How should we be thinking about that? Thank you. -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [10] -------------------------------------------------------------------------------- I appreciate that question. There are a lot of moving parts in our gross margins. Of course, our enterprise business is richer in gross margins. Our consumer business tends to be lower in gross margins. But at the highest level, the way I would think of our Company's gross margins, it's the nature of our business model is changing. If you think about our business model, a long time ago, it used to be a chip business, but today we are a differentiated specialty platform business. What I mean by platform business is, it's of course the chips, the systems, but it's largely about the differentiated software that's on top of it. So increasingly, you're going to find that our business is software-rich, it's services-rich. And if that is the case, one would think that our business model would become increasingly of that nature. And I think you're just seeing the reflection of that. As our Company continues to move towards our differentiated platforms -- which was, call it, 50% just a few years ago, and is now reaching some 80% now -- as we move into these specialty differentiated platforms, the software content is just much, much higher. And our customers who work with us are not buying chips for their systems, for their commodity systems, but they are looking for a platform to solve a particular problem. And the problems that we would help solve, the solution that we bring to bear, is so high-valued, that I think that increasingly you should expect that -- well, you should hope, and I hope myself, that our gross margins continue to move along with the change in our business model. -------------------------------------------------------------------------------- Operator [11] -------------------------------------------------------------------------------- Our next question comes from the line of Hans Mosesmann with Raymond James. -------------------------------------------------------------------------------- Hans Mosesmann, Raymond James & Associates - Analyst [12] -------------------------------------------------------------------------------- Great, thanks, and congratulations, guys. Jen-Hsun, can you give us an update on the high-performance compute side of the business? How will Pascal compare to the upcoming other solutions in the market, specifically Intel's Knights Landing? And I have a follow-up. Thanks. -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [13] -------------------------------------------------------------------------------- Yes, Hans, thank you. Our high-performance computing business uses an architecture we call accelerated computing. Accelerated computing is a model of computing that we invented almost 10 years ago. It's a very unique way of computing, and it takes advantage of the strength of the CPU, as well as the advantage of the world's most parallel processor, called the GPU. It is very software-intensive, it is very mathematics-intensive, it's very algorithm-intensive. It's a problem that, when applied to some verticals, can accelerate computing dramatically. We see accelerations of 5x, 10x, 20x, quite normally, quite regularly. And the way that you translate this benefit to a customer is that it reduces their costs. Instead of building a supercomputer that may cost as much as $500 million, this supercomputer would be the world's best at $100 million. That's a pretty substantial reduction in expenses. The power-build that they would spend on a regular basis would be dramatically reduced. So datacenters and supercomputing centers save an enormous amount of money. On the other hand, the researchers see a substantial boost in their application throughput. So that is one of the reasons why accelerated computing is doing really well. We are seeing a couple of different drivers for accelerated computing and high-performance computing. Our datacenter business -- accelerated computing itself for supercomputing applications, whether it is weather simulation or molecular dynamic simulation -- continues to grow. The killer app that we are starting to see -- and we have been cultivating for several years now, and it's now really turning the corner and going into turbo-charge and growth -- is deep learning. Almost in every field of science, as well as for web services companies, artificial intelligence helps them wade through, comb through just massive mounds and oceans of data to discover insight. So deep learning and using artificial intelligence technology across all fields of science -- I'm super-excited about the work that's going to be done in medicine. It's really going to see some great adoption. I think we mentioned that in just a couple of years ago, we had 100 companies working with us in the area of deep learning, and now it has ballooned to 3,500. That is quite a large-scale growth. It is in industries all the way from life sciences to supercomputing of course, to web services of course, to even industrial. And the application for industrial would be Internet of Things. All of these sensors all around the world collecting data needs artificial intelligence software, deep learning software, to review insights. In terms of our positioning relative to the competition, this is an area that we have a real advantage, and we have a real advantage for several reasons. The incremental cost to our Company, the incremental cost of engineering accelerated computing into our normal course of running our GPU business, GeForce business, is incremental. So the system, entire system, the 10,000 people in our Company, can quite easily, if you will, continue this rhythm. And quite a high-velocity rhythm, of bringing out new GPUs that are great for accelerated computing and great for gaming of course, and great for workstations. And it's a natural course of doing work. This is not an adjunct business to us. This is our core business. At the core, that is our fundamental advantage, that it's a singular motion, singular execution, singular investment, singular architecture, incredibly leveraged, and the execution, as a result, is just absolutely flawless. So I think in the end that, that's our greatest advantage. -------------------------------------------------------------------------------- Hans Mosesmann, Raymond James & Associates - Analyst [14] -------------------------------------------------------------------------------- Very helpful, thanks, Jen-Shun. -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [15] -------------------------------------------------------------------------------- Absolutely. -------------------------------------------------------------------------------- Operator [16] -------------------------------------------------------------------------------- Our next question comes from the line of Ambrish Srivastava with BMO Capital Markets. -------------------------------------------------------------------------------- Gabriel Ho, BMO Capital Markets - Analyst [17] -------------------------------------------------------------------------------- Hi, this is Gabriel Ho calling in for Ambrish. Thanks for taking my question. I want to follow up on your Tesla business. I think in a recent earnings call, Cray actually reiterated its expectation of over 50% of this $25 million revenue in the fourth quarter of this calendar year. And I think they cited DRIVE as one of three supercomputers that actually use Pascal. So how should we think about the benefit to your Tesla business? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [18] -------------------------------------------------------------------------------- Well, Cray is a very important partner of ours. The thing that is really exciting for me, is to see them transition their business -- not transition, but transform their business, from one that is really focused on supercomputing centers to one that is also working on big data. This is an area where we can add a lot of value to them. We have a lot of expertise in this area. And as they continue to evolve their market footprint beyond supercomputing centers and now into large enterprises, I think they can find a lot of success. They are seeing a lot of this success in this area. Big data analytics is square in their bull's eye, and I'm quite excited for them in the work that they are doing there. They are good customer for us, a great partner of ours and I'm excited to see their ongoing success. The thing we are all seeing is that big-data analytics, the most powerful weapon for big-data analytics has recently been discovered. Deep learning is just a fantastic new computing model. It is able to discover insight that is provably now superhuman. Its dimensionality in thinking through data is unrivaled in any approach that we have learned in the past. And that's one of the reasons why industries all over the world, from life sciences to industry to manufacturing to supercomputing, are jumping on the deep learning bandwagon. I think their adoption of Tesla, the NVIDIA GPU, is going to be quite a success for them. -------------------------------------------------------------------------------- Gabriel Ho, BMO Capital Markets - Analyst [19] -------------------------------------------------------------------------------- Thanks. As a follow-up, [your spend] seems to spend about $90 million in legal expense. How should we think about in FY17? And also, and can you give an update where you are in the case of [Qualcomm] and Samsung? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [20] -------------------------------------------------------------------------------- Just as a backdrop, we litigated against Samsung last year. The expenses was what you are referring to. At the core of it, fundamentally, philosophically, we believe that it is inappropriate and it's wrong for Samsung to use NVIDIA's technology -- technology that has cost us billions of dollars to invent and use it without compensating us. At the core, I just think that is just wrong. And we think it's wrong, and that's the reason why we decided to litigate to sue Samsung. The ITC has passed it's early decisions, and we disagree with them, we are disappointed by them. It's unfortunate that the business courts couldn't see through the obviously complex data associated with the technology. But we are disappointed by it. We have appealed for a review, and hopefully, in the near term, we will discover what the ITC will do. But I still believe that it was the wrong thing to do, for Samsung to use technology that companies who are specialized in these fields invent, and to use it without compensation. And I'm disappointed with the decision from the ITC, but so be it. Next year, we have plenty of things to go invest in, and we have plenty of growth drivers. You know that we have four powerful growth drivers in our Company. Gaming is one, VR is another, artificial intelligence and self-driving cars. And we have plenty of growth drivers to go focus our Company on. -------------------------------------------------------------------------------- Operator [21] -------------------------------------------------------------------------------- Our next question comes from the line of C.J. Muse with Evercore ISI. -------------------------------------------------------------------------------- C.J. Muse, Evercore ISI - Analyst [22] -------------------------------------------------------------------------------- Good afternoon, thank you for taking my first question. I guess first question, on the auto side. I'm trying to get my arms around how we should think about growth year in calendar 2016 off this 75% growth in 2015. If you could parse between your backlog for infotainment and your outlook there, as well as what kind of ramp you see with product development contracts on the ADAS side? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [23] -------------------------------------------------------------------------------- Two questions there. First of all, our pipeline. We've talked about our pipeline several times. We've shipped probably 5 million, 6 million cars. We have another 20 million, 25 million cars to ship in our pipeline. So these are design wins that took quite a few years to have won, and quite a few years of engineering to ramp into production. So we have a pretty good visibility of the pipeline and the opportunities that are ahead of us. Probably there is some market dynamics that's helpful to some of the design wins, the segments that we serve. Of course, at the time, a long time ago, it's hard to tell. But it's very clear now that the computerization of cars is a highly desirable end-user feature. The partners that we worked with, the car companies we worked with, to computerize their cars, whether it is Audi or Tesla, whose cars are heavily computerized. Their growth prospects in the coming years are quite good. So I think that's one. We have a clear view of the pipeline, and I think the mega-trends of the computerization of cars is in our favor. Now you mention -- secondarily, we introduced this platform called DRIVE PX. It is our economist-driving car computer platform. And the recent success of ADAS has really inspired just about every car company in the world to look beyond ADAS and what's beyond ADAS is self-driving cars. It could be partially assisted, it could be mostly assisted, and it could be completely assisted. And in each one of those levels of autonomy, a different amount of computation would have to be deployed. We've created a scalable architecture that allows car companies to develop cars that are partially assisted, all the way to completely assisted. We are working with quite a large number of customers now. Car companies, startup companies, companies that are largely cloud-based and have an enormous amount of data that they could transform into an automotive service, transportation-as-a-service. And so we are working with a whole lot of different types of companies, and I think this is going to be an area of quite a significant industrial revolution. And arguably quite a gigantic society good in the long term. So anyways, we are working on a lot of projects there. -------------------------------------------------------------------------------- Operator [24] -------------------------------------------------------------------------------- Our next question comes from the line of Steven Chin with UBS. -------------------------------------------------------------------------------- Steven Chin, UBS - Analyst [25] -------------------------------------------------------------------------------- Hi, thanks for taking my questions. Jen-Hsun, the first one for you, if I could. In terms of the deep learning, machine learning opportunity, I was wondering if you could help quantify sort of the longer-term silicon TAM for the opportunity, both in datacenter and automotive? And I guess more near term, any thoughts on what kind of development revenues could be generated for these machine learning-type platforms in the near term? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [26] -------------------------------------------------------------------------------- Sure, thanks. I think part of the answer is, I'm not sure. Part of the answer is, I'm not sure. So with that as a disclaimer, let me tell you why I'm so enthusiastic about it. There are many problems that computer science has been trying to solve which, algorithmically, is just impossible to solve. There is no known way of a human-described algorithm that completely captures the noisy and long tail of society. And it could be almost any problem. It can be weather-related-type problems. It could be market-related-type problems. It could be all kinds of purchasing-related challenges, and all kinds of data. It could be life sciences, as we know that the human body is not in a perfect condition all the time, that randomness that plays a role in understanding molecular science. There are so many different types of areas where there is no simple Newtonian physics equation that can describe the nature. So in that particular case, using an enormous amount of data to train the neural net, to train software, to rewrite the software, if you will, using an enormous amount of computation, is a pretty exciting computation model. I think this is a brand-new computing model, one that is going to augment the traditional model of symbolics and computer programming. This is going to be a data-driven type of computing model. In this particular case, GPU accelerated computing is really quite ideal and the computing model that we invented some ten years ago is really quite ideal. How big is it? I think that it could be quite significant. And we are starting to see, of course, the type of companies that are jumping on top of the deep learning bandwagon. They are great companies, from Google to Facebook to Vidu to IBM to Alibaba, to just about every hyperscale web services company in the world, is jumping on this. Because they have enormous amounts of data and it has very, very long tails. And traditional segmentation is too contrived of an approach to find great insight. Now that the companies with a great deal of web-based data, cloud-based data, have already starting to engage in this area, they are starting to implement artificial intelligence into one application after another. I think we already heard them announce that it's very likely they will put artificial intelligence into every single application they have. We are starting to see this sweep across industries. The automotive industry, of course, has the longest tail, as the world is a very noisy place. And in order to create a car that can navigate through it, the long tail of a very complicated world has to be handled somehow. Writing software programs is just not going to do it. So using an ongoing learning artificial intelligence network could be exactly the solution for it. Life sciences, industries, manufacturing, supercomputing, financial services -- the list goes on and on. And we are seeing a lot of enthusiasm. Before everybody can use deep learning, they have to train a network. And this is an area where we have a great deal of expertise. This year, as you know, we also announced our first hyperscale inferencing engine. It's our first end-to-end training-to-inferencing -- inferencing is predictions, the application of the network. So from training all the way to inferencing, we now have a complete architecture that is architecturally compatible. The Tesla M40 is for training, and the Tesla M4 is for inferencing. The M4 is a little, tiny credit card-sized GPU, and very low-power, incredibly energy-efficient. And you can connect it into just about any hyperscale datacenter in the world. And we are sampling customers now. The results are quite exciting. Customers are very enthusiastic about it. I think we could dramatically reduce the cost of datacenters all over the world, as they start to ramp up artificial intelligence in their everyday workload. -------------------------------------------------------------------------------- Steven Chin, UBS - Analyst [27] -------------------------------------------------------------------------------- Okay, I appreciate that color, Jen-Hsun. And as a quick follow-up, I have one for Colette on the inventory levels. Colette, just given where inventories ended for the quarter, it's -- on a days and a dollars basis, it's roughly comparable to a number of quarters back, when revenues were about 20% to 30% lower than where it is today. Is this a new level that the Company can continue operating at, just based on supply chain efficiencies? Or is there some -- a seasonal volatility in December or any other kind of average ASP of the [prox] that you are carrying? Thanks. -------------------------------------------------------------------------------- Colette Kress, NVIDIA Corporation - EVP & CFO [28] -------------------------------------------------------------------------------- Yes, Steven, thanks so much for the question. Our inventory levels that we are holding here, they are definitely going to swing a bit in terms of the mix, in terms of our platforms. But what we have right now, we do have a very healthy level of inventory. And we have a great team of people managing all of those different pieces, both for the channel, for our partners, and definitely for what we need to ship going forward. So I don't think we look at a number to exactly optimize in any single one quarter, as we do make sure that we are prepared for the platforms coming down the pipeline, as well as what customers need. But you are correct, it is probably at a fairly healthy low level at this time. -------------------------------------------------------------------------------- Operator [29] -------------------------------------------------------------------------------- Our next question comes from the line of Sanjay Chaurasia with Nomura Securities. -------------------------------------------------------------------------------- Sanjay Chaurasia, Nomura Securities - Analyst [30] -------------------------------------------------------------------------------- Hi, Jen-Hsun. One question on deep learning. I was wondering if you could talk about relative opportunity sizes in training and inference part of the deep learning? Clearly you have a strong position on the training side. I would love to hear your thoughts of the inference side. My understanding is, a lot of custom chips are being built in the industry on the inferencing side. So I would love to get your color on what competition you're foreseeing on that front? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [31] -------------------------------------------------------------------------------- Sure. Here is my guess. I think long term, training will be half of the overall market. And the reason for that is because training is so heavyweight. And in the long term -- well, not long term now -- you're training your network constantly. You create a network. You want to improve this network as fast as you can, because you have so much valuable data and so much insight that you can go after. And you deploy the network for inferencing, which collects brand-new data. And the world looks completely different to you. You now collect that data, and you use that data to train your network. I think that network training is going to be continuous basis, and we're seeing that, absolutely. Also there are more types of networks. The type of networks that are being created, the rate of revolution, the rate of innovation of networks -- network styles, network types, network configurations, network depths. It is happening every single week. I'm actually not exactly sure how you would design a custom chip for it. Which explains why there are only two chips today that are successful in inferencing. One of them is the Intel Xeon, and the other is the Tesla M40 and M4. So I think the ability to adapt to new algorithms quickly, it's really quite vital as we go through the next several years of this artificial intelligence revolution. There is just so much algorithms being developed, and I think you guys are reading about it constantly -- new breakthroughs in AI, new breakthroughs in network design. At the moment, I just really don't know how someone would settle down and design a custom chip for it. I happen to believe that long term, artificial intelligence is not a chip. Artificial intelligence is a computing model. And a computing model needs processors. And processors are programmable. And these programmable processors need to have rich software development environments around it, and these platforms need to be available all over the world. Today the NVIDIA accelerated computing platform is available in a PC, in a work station, in a laptop, in the cloud, in a car, in robots, in embedded environments, and it's all exactly the same architecture. I think that, that's really one of our advantages, that we have the ability to be adaptable, programmable, and yet we are available in literally every single computing platform form-factor you can imagine. And the accessibility of NVIDIA's architecture is literally global, worldwide and within reach for anybody. -------------------------------------------------------------------------------- Operator [32] -------------------------------------------------------------------------------- Our next question comes from the line of Ross Seymore with Deutsche Bank. -------------------------------------------------------------------------------- Ross Seymore, Deutsche Bank - Analyst [33] -------------------------------------------------------------------------------- Thanks for letting me ask a question. One for either Jen-Hsun or Colette. In your first-quarter guidance, it looks like the down-10% has some seasonality to it. But you also have a lot of businesses that have secular trends behind them. So I was hoping that you could provide a little color on seasonality versus secular? Or which of the drivers would be better or worse than that 10%? Acknowledging also that you lose a week of business guiding into that April quarter. -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [34] -------------------------------------------------------------------------------- Yes, first of all, I appreciate your question. I think, first of all, it's just the guidance, and it's our best view today, it's our most prudent view today. And as you know, although there are many things we know, the world is a very uncertain place. There's a few things that we do know. The gaming market is quite vibrant. It continues to be quite vibrant. We monitor it literally every day, every week. And we monitor it all over the world, and it remains quite vibrant. In the coming months, there are some really wonderful games that are coming out, that we think are going to be spectacularly successful, whether it is The Division or Tomb Raider or -- the list goes on. So I think the gaming market appears to be quite vibrant. Our automotive business is vibrant, and the work that we do in self-driving is really gaining traction and capturing the imagination of just about every car company around the world. Our deep learning work and supercomputing work, our high-performance computing work is accelerating. In a lot of ways, I understand where you're coming from. But we don't want to ignore seasonality. Q1 is Q1, and we recognize that the market is uncertain. We will see how it plays out. At the end of the quarter, we'll come and report it again. -------------------------------------------------------------------------------- Operator [35] -------------------------------------------------------------------------------- (Operator Instructions) Our next question comes from the line of Harlan Sur with JPMorgan. -------------------------------------------------------------------------------- Harlan Sur, JPMorgan - Analyst [36] -------------------------------------------------------------------------------- Hi, good afternoon, and congratulations on another solid quarter. Jen-Hsun, you talked about the installed base and the upgrade opportunity in gaming, I think, last call. You quantified it as around three-quarters of the installed base that really needs a more updated GTX processor. Given the 37% growth in the gaming business last year and the success of Maxwell, it actually does appear that you did drive some meaningful upgrade in the installed base. So the question for you is, is there any updates to your views on where the installed base sits at from that upgrade opportunity? And then the second question is, do you get a sense that the cadence of these upgrades will be accelerating, given the advancements you and your gaming engine partners are bringing to the market every year? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [37] -------------------------------------------------------------------------------- Those are really good questions. We monitor our installed base pretty carefully. Currently in the installed base, we basically have three architectures still in operation. We have the Maxwell architecture, and we have the Kepler architecture, and the Tesla architecture. All of those -- oh, excuse me, the Fermi architecture. Those architectures are all running in the installed base at the moment. We have managed upgrade about one-third of the installed base. Meanwhile, it is the case that ASPs of our chief use are going up, because the graphics richness and the graphics production value is going up. The quality of games -- because the market for games is so high, game developers can really create much more beautiful games, and take the risk to do that. The developing markets are growing. The number of genres, like eSports, of games are growing. So there's a lot of different growing vectors. Meanwhile, all of that is on top of our desire to upgrade our installed base, so they can enjoy games the way that it ought to be enjoyed. I think there is still a fair amount of growth opportunity ahead of us, and we will monitor it carefully and report it once a quarter. -------------------------------------------------------------------------------- Operator [38] -------------------------------------------------------------------------------- Our next question comes from the line of Deepon Nag with Macquarie Capital. -------------------------------------------------------------------------------- Deepon Nag, Macquarie Capital - Analyst [39] -------------------------------------------------------------------------------- Yes, thanks a lot, guys. So Quadro grew for the first time, I think, in several quarters. Can you describe what drove that growth, and how we should think about the growth profile for the rest of this year? Is it possible that it could grow in the mid-single-digits for all of 2017? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [40] -------------------------------------------------------------------------------- Yes, Deepon, thanks for the question. I was delighted myself. I will just put that out there. We worked really hard to improve our Quadro business. The team works incredibly hard. We invented a new technology for rendering called Iray. It's the world's first physically modeled photo-realistic renderer that is accelerated by GPU. The results of it is really quite remarkable, and they continue to add new capabilities to it. We, this last quarter, also benefited from the enthusiasm and excitement around VR. And we have VR SVKs and collaborations with just about every ISB in the world that is working on VR. So I think there's a lot of good reasons to be enthusiastic about Quadro. We don't believe for a moment that the design quality and design production value of movies or games or architecture or manufacturing will continue. We believe absolutely that it's going to continue to improve. And visual realism and the productivity of the engineers that are involved, the artists that are involved, needs to continue to increase. So we think this is going to be a vibrant growth area ahead. I think that what drove recently the uptick on the OEMs refreshing workstation cycle. And I think we should enjoy some of that for the coming quarters. But I still think long term, the real opportunity, surely the market is there. We know that more and more of design and creativity is done digitally. So at the core, the market is there. The opportunity for us is to bring new forms of rendering, new forms of design. And as you can see, Iray is for rendering, and VR is for interacting with the design. These type of capabilities require just an enormous amount of GPU capability. And I think at the core, that is going to be our long-term growth drivers. -------------------------------------------------------------------------------- Operator [41] -------------------------------------------------------------------------------- Our next question comes from the line of Joe Moore with Morgan Stanley. -------------------------------------------------------------------------------- Joe Moore, Morgan Stanley - Analyst [42] -------------------------------------------------------------------------------- Great, thank you. Can you give us some color on what you are seeing in emerging markets, with a lot of macro concern about some of your end markets, notably China? And you guys keep putting out very good numbers. Can you just sort of talk about that demand by geography, and how you weigh the economic and currency risks over the course of this year? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [43] -------------------------------------------------------------------------------- Sure. For some reason, I still tend to believe that what drives the gaming market is great games. And I think that there is some evidence that the continued release of great games and great production value games, the vibrancy of eSports, the fact that eSports is really not just for competition, but drives the dynamic of sharing and social. Those kind of factors continue to drive our gaming business. I'm quite enthusiastic about the developing markets. Southeast Asia, for one, is really starting to adopt PC gaming quite rapidly. It is a market that is extremely underserved. India is a market that is extremely underserved. They are underserved because broadband Internet hasn't been available to those marketplaces until just recently. And there are surely demographics in these markets that would love to jump onto gaming. Who doesn't need a PC? So almost every market develops around PCs quite rapidly. I think the way of enjoying games is so affordable by adding a GeForce GTX to a PC that you already own, it's the most affordable form of entertainment, if you think about it that way. Most of eSports are free to play anyhow. So much of it is, anyhow. So it is a wonderfully affordable way to enjoy entertainment. So I think the Southeast Asia, India, are really quite exciting developing markets. -------------------------------------------------------------------------------- Operator [44] -------------------------------------------------------------------------------- Our next question comes from the line of Matt Ramsay with Canaccord Genuity. -------------------------------------------------------------------------------- Matt Ramsay, Canaccord Genuity - Analyst [45] -------------------------------------------------------------------------------- Thank you very much. Jen-Hsun, there's been a lot of speculation about how the emergence of VR headsets, in particular for gaming, would drive your GPU business. I guess I would like to hear your perspective on a couple of things. One, obviously it requires high-end and high GPU use to support some of these applications. So of the early adopters, maybe a sense of which of those folks might already own those type of GPUs, and which ones might have to upgrade in the near term? And second, how VR might penetrate the [East Forts] phenomenon over time and drive more of those upgrades into the higher-volume mainstream parts of the gaming market? Thanks. -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [46] -------------------------------------------------------------------------------- Yes. There are really two questions in your question; they are both good. One question is, how do I feel about VR and its impact on gaming? The second part is, how will VR impact our business? Let me tell you the second part first. We are not forecasting and not assuming any upside in VR. However, there's no way but good that VR will bring to our business. And we will take it a day at a time. We believe that it's going to be an exciting growth driver. We believe that is going to be helpful to our high-end GPU business. But when the time comes, it will be a nice bonus. So we're going to run our business as if we don't count on it. However, obviously we care very deeply about it, because we think that the experience of VR is quite amazing. Anybody who has tried it is surprised how immersive it is, how it takes you away from where you are and into another world. You are really suspended in disbelief, and it's as close to a holodeck as we have ever experienced. So we believe strongly that VR is going to be fantastic for entertainment, it's going to be fantastic for games. We also believe that it's going to be fantastic for all of our pro-business. I wouldn't be surprised if the segment of our business that it helps the earliest may very well be our professional business. The reason for that is because there are many applications that are mission-critical. Even though the headsets are not free, it's quite affordable. And for people who have power walls and use large displays, VR is actually an incredible cost reduction. Almost anybody can now have a virtual reality cave, which cost tens of thousands of dollars. Anyone can now have a power wall, which costs tens of thousands of dollars. And now for just a few hundred dollars, have all the benefits of that. So I think that you could tell that I'm very enthusiastic about it. We are developing a lot of fresh technology and new enabling technology to make it possible. We are working with all the market leaders to develop the market and cultivate the market. And then from a financial perspective, we will just see how it plays out. My sense is that it's going to be a really nice bonus. -------------------------------------------------------------------------------- Operator [47] -------------------------------------------------------------------------------- Our next question comes from the line of Chris Hemmelgarn with Barclays. -------------------------------------------------------------------------------- Chris Hemmelgarn, Barclays Capital - Analyst [48] -------------------------------------------------------------------------------- Thanks very much for taking my question, and congrats on the good quarter. -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [49] -------------------------------------------------------------------------------- Thanks a lot, Chris. -------------------------------------------------------------------------------- Chris Hemmelgarn, Barclays Capital - Analyst [50] -------------------------------------------------------------------------------- First of all, we've talked a lot about some of the really exciting businesses that have a lot of great growth prospects for you. I thought we would look at, as well, some that have been a little more stagnant. And I was curious to your take on, for example, the OEM business. It looks like it may finally be stabilizing after declining. How do you look at that business in the coming fiscal year? Do you think it can even get back to growth, or do you expect continued declines? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [51] -------------------------------------------------------------------------------- Well, my sense is that it will stabilize, and let me tell you why. Our OEM business is not about gaming, because it is our gaming business. And the OEMs are not about design, because that's our design business. But the thing that some OEMs do is, they include our technology to differentiate a mainstream platform from a premium platform. So by adding NVIDIA's technology, you turn a commodity PC into a premium PC. And the experience is better, the performance is better, and everything just works. So there is a real benefit in using it as a premium multimedia PC, if you will. So we'll probably see continued interest in doing that. And we are delighted by that. We don't count on it, but we're delighted by it. So my sense is that it will likely stabilize. -------------------------------------------------------------------------------- Operator [52] -------------------------------------------------------------------------------- Our next question comes from the line of Chris Rolland with FBR & Company. -------------------------------------------------------------------------------- Chris Rolland, FBR & Company - Analyst [53] -------------------------------------------------------------------------------- Hey, guys, thanks for the question, and great quarter. Speaking about your installed base briefly, can you guys share with us your view on GPU replacement rates? How fast are they or how long are they now, and whether they are shrinking or speeding up here? And also perhaps how they might differ by geography? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [54] -------------------------------------------------------------------------------- I don't know if I have the precise granularity by geography, except for just a few countries. But let me address it overall. The installed base takes a couple two or three years to upgrade. And on the lower end, three to four years. On the higher end, one to two years. And so overall I guess, because the lower-end products are higher in volume, it would probably weigh the overall average to call it three-plus years. But the ASP, of course, if you look at it from an ASP perspective, it's a little bit different -- it would probably drift up. The rate of upgrade appears to be increasing. I might also explain that the ASP of our overall portfolio increasing. I think the reason why it's increasing is because the size of the gaming market has now grown to a level that developers can take a fair amount of risk to add a rich content, rich production value in graphics, which they didn't used to be able to do. They now have the benefit of a large installed base, a PlayStation an Xbox, a Nintendo and PCs, that they could actually create content that is really, really quite beautiful and technologically rich. Which drives up the adoption of higher-end GPUs, because you need higher-end GPUs to process it. I think that is quite a significant factor. I think the other factor is that the game consoles, although nowhere near the performance level of our high-end GPUs, on average, is higher in terms of capability than our average installed base. That is actually terrific news for us, because a PC gamer who wants to enjoy games that are adapted from game consoles, which all of them are, now would have to upgrade their GPUs to enjoy at least a game console experience. So I think that bottom half of our installed base has a real opportunity to enjoy at least a game console experience for just $150. For a $150 graphics card, you can get experience that is superior to a game console. That is quite an amazing value. That, I think, is also another reason to spur adoption. -------------------------------------------------------------------------------- Operator [55] -------------------------------------------------------------------------------- Our next question comes from the line of Rajvindra Gill with Needham & Company. -------------------------------------------------------------------------------- Rajvindra Gill, Needham & Company - Analyst [56] -------------------------------------------------------------------------------- Thanks, and congrats as well. Just a follow-up question on the virtual reality market and how you are looking at that. Can you talk a little bit about some of the PC requirements that are going to be necessary to use in Facebook, Oculus Rift headset when it comes out this quarter? And the cost that is going to be needed for the individual user? And contrast that from the business model that Sony is employing, which is basically, based on my knowledge, a bundling strategy with their PlayStation 4, which wouldn't require an upgrade to the graphics card or buying a new desktop? Just wondering how you think about those two different business models? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [57] -------------------------------------------------------------------------------- There are two different questions in there. One question is, what is the minimum requirement for VR today? Using today's graphics card as an example, the GTX 970, which is the most popular graphics card in the world, is the min spec for the Oculus Rift. And the reason why, of course, is because Oculus and their PC focus wants to have the best possible experience for the early adopters of VR, and I think that is a really prudent strategy. You want to delight all of your early adopters with the best possible experience. But the way to think about it long term is that, as the market continues to grow and more content comes and VR moves into the mainstream, there is no question in my mind that our $100 and $150 GeForce GTX cards will, in the future, be able to play VR just perfectly. So this is not a question about the availability of technology or the cost of technology. As we know, technology continues to advance, and whatever experience today will continue to get more affordable long term. Whatever the Sony PlayStation does, I think is just fantastic either way. What we would like to do is get people excited about VR. And in the final analysis, there are TV gamers and then there are -- there are console gamers and there are PC gamers. They are different genres, and they are different applications and different styles, and very largely different customers. So I think we are just enthusiastic about VR, period. And over time, the technology will more and more affordable. -------------------------------------------------------------------------------- Operator [58] -------------------------------------------------------------------------------- Our final question comes from the line of Brian Alger with Roth Capital Partners. -------------------------------------------------------------------------------- Brian Alger, Roth Capital Partners - Analyst [59] -------------------------------------------------------------------------------- Thanks for squeezing me here at the end. I'll be very brief. The opportunity in automotive obviously continues to grow. And your unveiling at CES was impressive by anyone's standards. However, there seems to be two approaches to the automotive market as we move forward. There is obviously the closed system approach that Mobile [Life] seems to be pursuing, and what appears to be more of an open-architecture approach that you seem to have. Can you maybe describe how things shape up for you as you look out over the horizon with sensor fusion and the various OEMs, and getting things right from a safety standard space, et cetera? Do we need the control of a closed system, or can we get it done with collaboration with others? -------------------------------------------------------------------------------- Jen-Hsun Huang, NVIDIA Corporation - President & CEO [60] -------------------------------------------------------------------------------- Sure. I appreciate the question. It's really a good question. This is really a matter of philosophy. And philosophically, this is how we see the world. We believe the self-driving car is not a solved problem. I say that as a statement of fact. I don't think anybody would dispute it. I also believe that self-driving cars is a field that's going to require the technological muscle of a very, very large industry, and that no one company with a few hundred employees is going to solve it all by themselves. The idea that a unsolved problem of such incredible daunting levels, that an entire computer industry is in the process of trying to solve, could possibly be a closed system tied around a chip, seems illogical to me. That's number one. Number two, I believe that long term, our car company, the soul of the car company is the driving experience of that car. The soul of the company is the safety record of that car. The soul of the company is the functionality of that car. In the future, the functionality, the safety, the driving experience of the car is going to be largely software-defined. It's going to be artificial intelligence network-defined. I just can't imagine great companies like BMW and Mercedes and Audi and all of the world -- Toyota, and the list goes on -- and many great companies that are emerging into this marketplace. I just can't imagine that these companies would somehow outsource the soul of their car to a chip company. That is a second philosophical belief. So what we have decided to do is to create an automotive autonomous car computing platform and all of the rich software that is necessary to enable this incredibly high-throughput computer to behave in a really energy-efficient way and cost-effective way. And to be able to apply our deep learning expertise, so that these cars can benefit from artificial intelligence to solve these really complicated world problems. And that by partnering with every single car company in the world, that together we might be able to solve this incredibly daunting challenge, and hopefully bring some society good. So that is our approach as the open platform. And it starts really from a philosophical approach. Now, that philosophical approach results in a very substantial technological difference. Notice that our platform is completely programmable. We have rich tools. We know that developers all over the world can easily buy themselves a GeForce TITAN and write CUDA applications. And those CUDA applications will tomorrow run on a DRIVE PX just seamlessly. So I think that is a wonderful way for designers all over the world to be able to develop software -- which is really hard to do right now -- and then quickly deploy it into the car. So our strategy is just very, very different, and that is our approach. And my sense is, at the moment, it appears to be quite a good approach. Okay, I really appreciate that question. NVIDIA is the world leader in visual and accelerated computing, which is helping to create exciting growth markets like VR, AI, and self-driving cars. Which will transform many industries and positively impact the future of society. Our strategy is to leverage one core investment into four growth markets -- gaming, professional visualization, datacenter and auto. And it's delivering results and gaining momentum. Our goal is to balance investments to capture the enormous opportunity ahead, while maintaining a keen focus on improving near-term financial performance. I also want to remind everyone that our annual GPU Technology Conference will take place April 4th through the 7th in San Jose. We will be focusing on VR, artificial intelligence and autonomous driving. We will also be holding Analyst Day there, and I look forward to seeing all of you. Thank you. -------------------------------------------------------------------------------- Operator [61] -------------------------------------------------------------------------------- Ladies and gentlemen, that does conclude the call for today. We thank you for your participation, and ask that you please disconnect your lines. -------------------------------------------------------------------------------- Definitions -------------------------------------------------------------------------------- PRELIMINARY TRANSCRIPT: "Preliminary Transcript" indicates that the Transcript has been published in near real-time by an experienced professional transcriber. While the Preliminary Transcript is highly accurate, it has not been edited to ensure the entire transcription represents a verbatim report of the call. EDITED TRANSCRIPT: "Edited Transcript" indicates that a team of professional editors have listened to the event a second time to confirm that the content of the call has been transcribed accurately and in full. -------------------------------------------------------------------------------- Disclaimer -------------------------------------------------------------------------------- Thomson Reuters reserves the right to make changes to documents, content, or other information on this web site without obligation to notify any person of such changes. In the conference calls upon which Event Transcripts are based, companies may make projections or other forward-looking statements regarding a variety of items. Such forward-looking statements are based upon current expectations and involve risks and uncertainties. Actual results may differ materially from those stated in any forward-looking statement based on a number of important factors and risks, which are more specifically identified in the companies' most recent SEC filings. Although the companies may indicate and believe that the assumptions underlying the forward-looking statements are reasonable, any of the assumptions could prove inaccurate or incorrect and, therefore, there can be no assurance that the results contemplated in the forward-looking statements will be realized. THE INFORMATION CONTAINED IN EVENT TRANSCRIPTS IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL 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 CONFERENCE CALLS. IN NO WAY DOES THOMSON REUTERS OR THE APPLICABLE COMPANY ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY EVENT TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S CONFERENCE CALL ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS. -------------------------------------------------------------------------------- Copyright 2019 Thomson Reuters. All Rights Reserved. --------------------------------------------------------------------------------