Q&A: Nvidia chief explains his strategy for winning in mobile computing

Never say die: Jen-Hsun Huang has reinvented his graphics chip firm Nvidia three times now. And he just keeps going.

In 1993, he founded the company, now based in San Jose, with the aim of taking 3D graphics technology from Silicon Graphics supercomputers to ordinary PCs. The company beat out scores of rivals by pumping new versions of 3D graphics chips out every six months at a relentless cadence. Now it is the last major standalone graphics-chip maker, squaring off against microprocessor vendors Intel and Advanced Micro Devices.

Eighteen years later, Nvidia remains one of the innovation leaders of Silicon Valley, and recent events have shown that it never pays to second-guess the wisdom of Huang and his merry band of chipheads.

Analysts have long predicted that Nvidia would be squeezed out of the PC market as its rivals created combo chips that put graphics and microprocessor on a single piece of silicon. They criticized Nvidia for failing to create its own microprocessor. But Huang had a strategy in place. To prepare for this day, Nvidia added programmability to its graphics chips as far back as 2001, with the introduction of the GeForce 3. That enabled the company to start attacking Intel’s market dominance by getting graphics chips to do non-graphics computing tasks. Programmability has enabled Nvidia’s graphics chips to get into a wider range of devices, from video servers to supercomputers.

And now the company is in the midst of a new transition to mobile computing. This shift is perhaps the biggest of all, spearheaded by the company’s Tegra and Tegra 2 mobile processors. The Tegra 2 is in the Motorola Xoom tablet computer and the Motorola Atrix smartphone, both among a leading pack of Android devices that are challenging Apple’s dominance in mobile devices. And at the Consumer Electronics Show, Nvidia said it was working on a high-end ARM-based microprocessor called Project Denver. That microprocessor is expected to become one of a number of ARM chips that will be able to run an ARM-based version of Microsoft’s Windows operating system. Now that Huang has unveiled these big strategic moves, Nvidia doesn’t look so vulnerable anymore, and Huang has been able to spell out Nvidia’s strategy to Wall Street.

Huang recently held a dinner with a small group of reporters. He started the conversation with his own version of Nvidia’s history, and where the company is going next. Below is an edited transcript of the group conversation.

Q: Tell us about your strategy.

A: Mobile devices are a big topic and so we don’t spend much time talking about PC. Pretty crazy times. So we’ve had the, for many of you, I’ve had the privilege of spending time with you guys for some time and if I could just reframe the conversations that we’ve been having for some time and what we call inside the company Nvidia 3.0. For Dean, you’ve known Nvidia since practically our beginning and if you were to go back, Nvidia 1.0 was PC graphics. Nvidia 2.0 was the creation of this process that we call the GPU (graphics processing unit). It’s a programmable graphics processor and because of that it extended the reach of our market from PCs to all kinds of computing devices, anything that has visually rich expression.

And we call that Nvidia 2.0. That was the invention of programmable shader. The vast majority of today’s computer graphics that you see has elements of that invention in it. Modern computer graphics is really due to the invention of programmable shaders.

Well Nvidia 3.0 is about reinventing Nvidia so that we are able to address a much larger part of the computing market and there are several things that we wanted to achieve with Nvidia 3.0. We wanted to take the expertise that we’ve had for a long time and turn Nvidia into a parallel computing company.

And so Nvidia 3.0 inside the company is really a parallel processor company with several specializations. One of them is visual computing, anything that’s visually rich; we believe we can make a contribution to. And then the second thing is energy-efficient computing. Energy efficient computing has two elements to it and one side of it is mobile. We think that in the future the personal computer is a personal computer — not the PC — but a personal computer; the most desired personal computing device will be a mobile device. And the second aspect of it is the cloud. If you have millions and millions of processors in the cloud surely you would want it to be energy-efficient.

And so we invented this technology called CUDA which is at this point probably the world’s most pervasive parallel processing architecture. It powers the fastest supercomputer in Japan, the fastest supercomputer in China and many of the fastest supercomputers in the United States and around the world. Now inside the company we say the way we distilled Nvidia 3.0 down into actions is three arms. We say go parallel, go mobile and go ARM. Now a lot of you have asked me over the years what is our CPU (central processing unit, or computer’s brain) strategy and I’ve said over the years it was ARM. And I said it so matter of factly and it almost sounded like a joke. But it was the truth. It was the same thing I was telling our company inside that our CPU strategy is ARM. I believe that ARM will do for CPUs what Taiwan Semiconductor Manufacturing Co. did for foundries (the contract chip manufacturers). When I first came into the industry, there was a phrase and I forget who, I think it was either (chip executives) T.J. Rodgers or Jerry Sanders that said, do you guys remember it?

Q:  Jerry Sanders (former CEO of Advanced Micro Devices) said that real men have fabs (or chip factories).

A: Real men have fabs. The reason for that was during that time unless you owned a fab and had the ability to run a fab, develop the process technology to create the fab, you really couldn’t create a semiconductor company. Then TSMC came along. The idea of an open merchant foundry business where anybody can come and buy semiconductor manufacturing technology was really revolutionary. It created quite a few companies, including ours. Then I felt because of, if we believe in the mobile computing proposition, if we believed in that one of these days all of these mobile devices will be computers and they will be based on ARM processors and they will be reasonable computing devices.

Then the volume of mobile devices phones — one billion going to four — is so large it would attract developers. They would create applications that make these mobile devices more wonderful. This virtuous positive feedback system would happen for the mobile devices as it did for PCs, except it’s an order of magnitude larger. Instead of hundreds of millions of devices, it’s several billion. And so that would make the ARM processor the most valuable instruction set architecture (or chip processing architecture) in the world.

We’ve been talking to you about these three propositions for some time. We’ve been working towards this for sometime and at CES this year we were fortunate enough to package up a lot of these propositions and these ideas into a few announcements. The first announcement at CES was just simply announcing that we have real smartphones and tablets shipping with the Tegra Mobile processor.

We needed a really good operating system to expose the benefits of Tegra. And finally it came along with Android 3.0. It has all the virtues of a modern operating system, it has hardware acceleration, video hardware acceleration for graphics, has a rich applications programming interfaces (APIs) just like Direct X has for Windows. So it’s a wonderfully rich operating system, finally it came along and we were fortunate to partner with Google to bring that to the market. And so we were able to announce that at CES.

The second thing we announced was Project Denver. We’ve been working on a CPU internally for about three and half years or so. It takes about five years to build any full custom CPU. And Project Denver has a few hundred engineers working on it for this period of time and our strategy with Project Denver was to extend the reach of ARM beyond the mobile, the handheld computing space. To take the ARM processor, partner with them to develop a next-generation 64 bit processor to extend it so that all of computing can have the benefits of that instruction set architecture. It is backward-compatible with today’s ARM processors.

And so by partnering with ARM to build this processor, we felt that we could bring the ARM processor into the PC because the PC is still a very powerful device and it’s still able to do many things mobile devices cannot. We would go into the PC but retain the energy-efficient characteristic of ARM  and would enable a new class of personal computers that has many times less power consumption than today’s PCs but has the performance of today’s PCs if not more. And so that was our vision with Project Denver.

But we timed it with Microsoft so that 15 minutes later, key members of the press could walk across the hall, go into Microsoft’s press announcement and there Steve Sinofsky of Microsoft would announce that next generation Windows would be based on ARM. And so everyone now sees the picture that our CPU strategy really is ARM, that we intend to take the ARM for mobile devices all the way to supercomputers. ARM is now the only CPU in the world that will have deep penetration in the mobile devices, the PC, servers and supercomputers. And so I think at CES our strategy for Nvidia 3.0 was crystallized and people understood what we were trying to do — that we were trying to build a modern version of a computer technology company.

So that was kind of CES. People still thought a cloud over our heads was our big battle with Intel. People said that Nvidia’s Intel chip set (MCP) business is going away and of course we announced that our dispute with Intel has been resolved. We’ve extended our cross license with Intel and the licensing revenues that would come to our company would be approximately $1.5 billion over six years. That by and large replaces and some the business that we lost with MCP. If you think about the net income that it generates for our company and you multiply by whatever ratio you would like, six times for example for a 15 percent net income company that one and a half billion dollars translates to quite a large revenue contribution over six years. It more than compensated for the MCP business that we lost.

And so we were able to address part of the cloud over our heads with that settlement. And in the last part, and this is always the case when you demonstrate the first version of something, Tegra was our first success. All these smartphones and all these tablets that were announced at CES was really our first success. And so the question people had were maybe this Tegra was, is Nvidia really serious about mobile computing? Is this Nvidia and Tegra flash in the pan?

And so at Mobile World Congress, we demonstrated our next generation mobile processor working almost perfectly. You can already use this thing right now; it’s a perfectly working processor and was the world’s first quad core mobile processor. It is the world’s first mobile processor that can support a video resolution that is the highest of any computer in the world today. Whatever resolution you want it to support, we support, basically extreme HD. And as we demonstrated the world’s first consumer product of any kind to decode 1440p at 30 frames a second. I don’t think anything with the exception of large custom-made projector players have the ability to do that. There is no consumer appliance device in the world that can do that today.

And we also demonstrated a performance that was five times that of our last generation and so we introduced Kal-El. In the process of introducing Kal-El we introduced the idea that we would bring to the marketplace three other processors over the course of the next two years. Just like we do with GPUs, there will be a new generation of architecture every year.

Every single one of this project are fully funded and the expectation is within the next three or four years we’re going to bring to the mobile market performance that is nearly a hundred times higher than today’s’ PC. And that’s the roll out if you will of our Nvidia 3.0 strategy. Any questions?

Q:  At what point did you guys sort of decide that x86 (Intel-compatible microprocessors) had just many roadblocks for you?

A: I think we thought about x86 for a very long time. There are two reasons why we decided not to do x86. Aside from, well the second reason is what I said earlier that in fact it’s the wrong instruction set architecture. The first reason is simply very large of course. The world’s not waiting for us to build yet another x86 and we’re not going to go hire a bunch of the world’s best engineers so that we can wake up in the morning to go do something that somebody else has already done 25 years ago. It’s not logical.

And so it’s another way of saying it’s a commodity. Intel has got every single price point covered from $10 all the way up to $1000. There is not one nook and cranny we can cover by ourselves. AMD has covered everything else. And if you think about it for a second, how is it possible that AMD has a business model at all? And so that’s kind of the first reason. There is also a positive way of thinking about it. Why don’t we go find something where we can add a real contribution?

The negative way of thinking about it is there is some crazy person out there trying the same thing as Intel and it’s not working for them. At AMD, they actually make perfectly good CPUs. I’ve never met a CPU at AMD that I didn’t like. They are all fine. They just can’t win.

Q:  But if you go into the ARM world, there are more competitors.

A: Yeah but that’s the beautiful thing though. It’s like we’re all using TSMC for manufacturing. It’s an equal playing field for everyone. The fact that you have a CPU is equal. You start with ARM. But then building a computer out of that is very different. You can build a race car or a minivan. And we have to go choose the different types of markets where we could add a lot value because we’re passionate about it and we’re very good at it. Media computing is something that we’re really good at it.

Q:  You didn’t say it but you implied it. So does that mean that Intel is not an equal playing field?

A: At no time is there an equal playing field with Intel. At no time. And the reason for that is because 90 percent of every one of your customers’ livelihood depends on them. How is that an equal playing field? The way to think about it is to look at it from AMD’s view. If AMD’s customers were to give them 100 percent of their business, AMD would not have the capacity to serve them. If HP says I would like to give you a 100 percent of my business, AMD would just say I can’t do it. There is not enough capacity in the world to do it. That basically says your competitor is in a position that is formidable. It is impossible to beat them. That says something about Intel’s position.

Q:  Then they have the intellectual property position too. This is what your court battle was over. If somebody wanted to get into the x86 market right now, they’d have to pay Intel some ridiculous price?

A: You could but don’t forget they’re paying us a pretty significant royalty now.

Q:  If you would just assume it’s about processors and not graphics?

A: What is not a solvable problem is even if they give me rights to make an x86 chip, I will be building a commodity that at every price point they have an alternative to. And if they have an alternative to everything that I make, and it’s easier to buy from Intel, it’s just really not possible to distinguish yourself in an x86 world. And so that’s sort of the reason why, that’s one of the negative reasons why you don’t do it. But the positive reason is we all want to go make a contribution to something and make a difference in the world. I mean you’re going to go spend $1 billion in r&d, you go spend $1 billion building something that matters.

So that’s sort of the more first principle and logical thing to go to. And I think that when I think about investment, I tend to think about the time of our people. I don’t want to waste the time of our people. I mean you get all this great talent, you convince them to come to Nvidia, and guess what you go get them to build: something that someone else has done. It’s no different from writing an article that’s exactly been written already. You want to tell a story in a new way, you want to add a new perspective. And so the perspective that we thought that we can really do is that in this area of energy efficient computing. The way you go about computing is radically different.

Q: Now that you’re getting $57 million a quarter from Intel, you might start saying nice things about them, right?

A: I didn’t say anything bad about them. That’s the reality.

Q: So did you swap that Intel for Qualcomm and Samsung there (for your rival)?

A: I don’t know whether we swapped for anybody. We’re still competing as Intel; they’re still competing against ARM. I don’t think anybody was swapped. We have simply added rivals. We have more competitors today than we ever had.

Q: I remember back in late 1990s there were 60 competitors in graphics.

A: That’s right.

Q: How will you win?

A: Well anything worthwhile doing you’re going to have a lot of competition. You can’t choose something that’s worthwhile to do that you’d be so so arrogant to think that you’re the only person in the world that figured it out. It won’t be for long. Well the question is can we add a unique perspective? That’s mobile computing on the whole account if we all haven’t figured it out that’s the future. If everyone hasn’t figured it out that’s because they’ve got their head in the sand. If graphics is important, if visual computing is important, we can add a perspective. If building a high performance computing device is important we can add a perspective.

I think that at least mobile devices are still in ancient times. I mean frankly this is a whole bit like Windows 2.0. We’re going to look back at these phones and they’re just going to be gross. They’re going to be disgusting. I mean yucky. And you’re going to look at it and go; I can’t believe I used that.

Q: Just that I’d be interested to hear your views of kind of where you see the consumer in the computing industry in say three years? I ask the question in the perspective of what we should expect to see when we go to Best Buy say in two years. Is the smartphone and tablet section going to be much bigger or the PC sector? Are consumers going to know the difference between what they’re buying if Windows works on all of them?

A: Excellent question. The old personal computer industry was, you bought a PC at Best Buy it’s based on Wintel (Windows and Intel). That’s the old PC industry. But we all know that the new personal computer industry is including that but it’s a mobile device and will be increasingly a mobile device. And it’s a foregone conclusion our children are much more enamored with iPhones and Android devices and tablets and now with PCs. But you know you don’t buy mobile devices at Best Buy. You buy them at AT&T, Verizon stores. So Best Buy is really under a lot of pressure. If ARM is the most important chip architecture, then the most important processor in your computer is going to be an SOC (system on chip, or a processor with many other parts built into it).

Q: Do you expect consumers to care about what’s happening at the chip level?

A: That’s the amazing thing. That’s why it’s even more over than we think. The consumers aren’t going to worry about that at all. Nobody ever picked up an iPhone goes, I wonder if this has an x86 or an ARM inside. I just know it’s an iPhone. I know, that’s how sad it is. That says everything.

Q:  Where does Qualcomm fit in this world?

A: We compete with Qualcomm. I think about Qualcomm as a mobile company that is growing into computing. They grew up on the mobile side. Our strength is that we know computing really well. The more sophisticated the computers or tablets or smartphones, the more we have an advantage.

Q:  How fast does Denver have to get into the market? Can yo wait until Windows on ARM arrives?

A: Denver can come out ahead. It does everything. Denver can’t just be another CPU port. We have to bring a perspective about energy efficiency that the world has never seen and so that’s something that we are quite excited about.

Q: What do you think of the OS wars?

A: Operating systems are really an ecosystem with all the companies around them. So other software companies, hardware companies, OEMs (original equipment manufacturers). You learn how to work with that operating system in a way that really hard to put into words. Thousands of people are working together on a daily basis and you can’t put that down on a piece of paper, or you just can’t answer how do you work with Microsoft? I don’t know where to start. It’s been 18 years; you know most people aren’t married that long.

And so when we started working on mobile devices with them, it’s all the same people here. Instead of having Microsoft work with the ARM side of our company, we moved the ARM processor work into the PC set of our company because it is a lot easier to do that than to reconnect 4,000 relationships. We have that relationship with others as well. When you think about a new operating system, you think about what it is capable of doing as well as the ecosystem around it that makes it sticky or not sticky.

There is a Windows ecosystem and an iPhone ecosystem that Apple has created. There is a Blackberry enterprise ecosystem. The question is how much more can you build like that? I think the answer is there are not many more out there.

Q:  Where do you come down on WebOS?

A: It’s an OS for HP because they can always use it for the printers and things like that. The question is what is everybody else doing? If you have an HP operating system, you disconnect a very large part of the computer industry from it. Dell and Acer don’t get involved in that. The question is do we get involved with them and that’s a very tough question because supporting an operating system is an enormous undertaking as operating system gets more and more complicated.We support Windows, Macintosh, Linux because of our supercomputing work, RIM, and Android. Basically, that’s a huge investment for us already.

Q: What are your priorities?

A: Our priorities are Windows first, Android second.

Q:  So what is your relationship with the Nokia? If you want to have, if you’ve got a relationship with Windows, you may have a relationship with Nokia.

A: The exciting thing for us is that Nokia used to be a non-opportunity for us because we never supported Symbian. Now that they are moving to Microsoft, we have a chance.

Q:  Does Apple have to lose market share for your market opportunity to get bigger?

A: No. Apple has to get to a point where the number of projects that they have in the company exceeds their capacity as a company to build internally.

Q:  Is ARM on the Mac OS possible?

A: I don’t know their plans but if you look at it from 10,000 feet, it seems to make sense, right? Because if they go Mac on ARM, they could address some of their concerns with their own SOC. So instead of paying $150, they can pay $15.

Q:  Just to follow, who do you have to displace to get into Nokia? Who is there now?

A: Mostly Nokia, when I say Nokia I mean I shouldn’t say it that way, I would say ST Ericsson. ST Ericsson is almost 100 percent of Nokia. Used to be TI. I will say even maybe some TI still.

Q:  Do you think it is possible that Apple could have external CPU suppliers and internal going at the same time.

A: Yeah. They use some of their own chip sets, they buy some of ours but it’s only when the business opportunity get so large that they don’t know how to address it.

Q:  Where is the Project Denver opportunity? Is is mobile, PCs, or servers?

A: We are all about energy-efficient computing.

Q:  Does that mean Windows and ARM can address the entire market?

A: I don’t know. Nobody actually knows. My question is this: do you guys actually believe in five years time, whether you are in a desktop computer, or a mobile computer, is there some way that energy-efficient processing is not going to be enough for you? Can you actually imagine it? I think we are having a hard time finding markets it doesn’t serve. It does HD, it plays games. Multiply the performance by 50. What can a computer do with that kind of performance? Do you need more?

So it turns out there are still a ton of content creation opportunities for those supercomputing workstations. But for most of our normal computing average computing needs, I think a mobile device with a couple of watts of power consumption, which is 50 times less than the power consumed by your desktop today, is going to be enough. So my imagination tells me that the mobile computing market is going to become unquestionably the most important computer in the world. Mobile processors will address a very large part of the market.

We are going to get to a point where you can enjoy the same game on any device. If sitting in front of your living room and you can play game on your big screen TV but if you want to pick up where you left off, you could just get back into the game on your mobile device.

Q: Is there still a need for game consoles?

A: Well I still think that there is a market for a dedicated device. It’s just simply wont be as large as it used to be, don’t you think?

Q:  I am not sure.

A: There are enough people out there who want a wonderfully designed game device. There is a market for people like that.

Q:  Well you could just have a little Nintendo chip that you put in your Android phones?

A: Yeah or a Nintendo store on that phone. When Nintendo realizes that what they are really all about is helping people enjoy games, then they could do that. Why can’t you enjoy their games on any mobile device? And so maybe they realize that at some point and they’ll say if you have a mobile device, we’d like you to play Mario on it and you can buy from this store. But if you want a dedicated device to play Mario, you can have it this way. If you want to play on TV, you play this way.

Q:  Do you think there will be another round of consoles coming?

A: Oh, no question about it.

Q:  And can you predict when it will be in terms of how many years from now?

A: We will build one of them, right. And the reason for that is because the world doesn’t have enough engineering talent for anybody to build three of them at one time. It takes the entire livelihood of a computer graphics company to build one of them. And every single time they build one, my life is in danger. You build it once every five or seven years, but you have to build it all in a very short time. That’s because they wait and wait and then they say, ‘Can I have it next week?’

Q:  So how much longer a life do you think Xbox 360 and the PS 3 have?

A: Oh 10 years. But the next new console will happen before that. They are still selling PlayStation 2s.

Q:  So Project Denver raises the question as to whether there is use for more than four cores on a client-side device. So my hunch about Denver was maybe it’s like a GPU with four cores attached to it.

A: You are trying to trick me into telling you everything about Project Denver. Okay I fess up, and here is how it works.

Q:  Philosophically, we could keep adding more cores into the chip, right? But if you don’t need more than eight threads of execution, or maybe 16 threads, then what else do you do for the CPU? Whereas if you can just add more GPU hardware, and more cache, then you can see the benefit.

A: No, I mean there are many things that you said that are quite true. For example, for any application about six to eight threads is about all it could support. I mean you’ve got an audio thread, you got an artificial intelligence thread, you got a physics thread or a few of them, you got some basic collision detection threads. You don’t have that many threads, but you could spawn off probably six to eight threads and in fact that’s the reason why Xbox 360 has six cores and it keeps them pretty busy. I don’t disagree with you about the eight threads.  The thing that I would also say though is that world has become highly multitasking, especially on these mobile devices.

And when you’re reading a book, listening to music or you’re accessing a website, you want it to pop up real fast. When your web site is coming in, especially once you get Wi-Fi or 4G, you want it to come it fast. You get all these Java scripts all loading at the same time. That’s a whole lot of threads you could process at the same time. So interactivity, the snappiness of an experience, the fact that you’re doing a whole of things at the same time — you could be downloading an application, listening to music and browsing the web — you don’t want that to slow down.

Silicon cores are basically free. You really want never to give up the opportunity to delight somebody with a snappy experience. I think over time our expectation is that stuff ultimately will work and deliver a great experience. The silicon is almost free. The cores are free. I’m saying just buy them from me.

Q:  Do you thing that TSMC will lag behind in updating their process technology at a certain point?

A: No in fact the exact opposite is true. There was a time when only Nvidia and AMD pushed the leading edge capacity of TSMC. TSMC really relied on Nvidia and AMD to drive their early ramps. But now that Qualcomm and others are now in the mobile computing business, there are more and more semiconductor companies that are pushing the high performance nodes and that’s a good thing. I mean Qualcomm is now pushing the high performance nodes because they want more performance, right? And TI wants more performance and we want more performance, Marvell wants more performance.

Q:  Why did Intel miss mobile computing?

A: I think when you are in a different position, you look at the world differently. It’s hard to see the world clearly when you are in the midst of enormous success and it’s not their fault. You almost can’t see it; you know what I am saying? I mean you can’t see the other side because what you are doing is working so well. I mean could you imagine, I started a Chinese food restaurant and I got lines piled up to the other state. I would think that everybody in the world wants to be Chinese and so I will just keep building it out. The reason why we are certain that the future was going to be mobile computing devices was by extrapolation. We looked at what was inside mobile devices and extrapolated. When the ARM processors first showed up in a TI phone, do you guys remember that? I mean that was a wake up call for the world. That was 10 years ago, but that was a wake up call.

Q:  Where do you think the price of the tablets is going to settle at this year?

A: It depends on what kind of tablets. The problem is touch has a lot of different types. There is the person who doesn’t mind resistive screens. There are 10-finger capacitive touch screens, or four-finger screens. There are more screens coming. If you want perfect touch and handwriting, that could still be an $800 tablet.  You can go down to resistive touch screen and then pound the living daylights out of the screen before it knows you hit the screen, then that could easily be $199.

Q:  I think it depends on what Apple does.

A: No. I believe that Apple does what Apple does and nobody else is going to compete with them.

Q:   So you don’t think the Apple has a sort of particular pricing advantage?

A: I don’t think Apple products are going to be cheap and let’s think about that for a second, I think first of all, they have one advantage that everybody else doesn’t have which is the Apple Stores. Best Buy commands a 30-point margin. When you sell at Best Buy, you lose 30 points. I mean that is huge. Apple has a cost advantage.

Q: Are you shipping Tegra 2 chips for the Motorola Xoom?

A: Oh yeah we are shipping the daylights out of it.

Q: Back to Windows on ARM again: did you know about it before you greenlit Project Denver on ARM?

A: We had to commit on Project Denver first.

Q:  So what made you think though that Microsoft would follow through with Windows ARM?

A: We would have done Denver with ARM without Windows on ARM.

Q:  For whom?

A: For mobile devices that don’t use Windows.

Q:  Was there a sigh of relief when Microsoft made the announcement?

A: Oh yeah. When Windows decided to use ARM, our market opportunity went from a billion units to 1.4 billion and that’s the way I think about it. There are a billion smart phones and tablets and whatever and then now you add to it, 400 million notebooks and PCs. So the increase of our total available market by 40 percent is not a bad thing. So I was delighted by that. Okay, I moonwalked a little. Dan (Vivoli, senior vice president of marketing at Nvidia) and I hugged. Dan wept. I was happy, he wept, he wept, it was noticeable, he wept. He was that happy, he wept.