Welcome to Broadcom's 2011 Analyst Day. Please welcome to the stage Chris Zegarelli, Broadcom's Director of Investor Relations.
Hey, everyone. It's really great to see so many people out with us today in our first ever New York-based Analyst Day, so I want to welcome you first to our 2011 event. I also want to thank the people that are joining us on the webcast as well. We know you all have a very busy schedule, so we definitely appreciate you joining us today.
Before we get started with today's event, I did want to remind you that we will be making some forward-looking statements today. These statements are based on our knowledge and understanding of the business as of today and are subject to the customary risks that you see on this cautionary statement here on the screen and you can find more detail on our SEC filings as well.
I did want to mention that we will be making some references to non-GAAP financial information. You can find reconciliations in the books that we provided to you at the reception, and you can also find the reconciliations on our website as well.
Just as in prior years, you will find an evaluation form in the Analyst Day books that you received when you checked in. Please fill those out by the end of the event and turn them in at reception, and you will receive a complimentary Broadcom-enabled device as a gift and a thank you for your feedback.
I did want to mention the agenda today. We'll have presentations from Scott McGregor, Bob Rango, Dan Marotta, Rajiv Ramaswami and Eric Brandt. We have a full agenda, so let's get right to it. Please join me in welcoming to the stage Scott McGregor, Broadcom's President and Chief Executive Officer.
Scott A. McGregor
Good afternoon. It's a great pleasure to be here in New York today, and we've been having these events now for a number of years, and we've been alternating generally between Silicon Valley and our Irvine headquarters. This is our first time to try it in New York, and I'm very gratified that we have twice the attendance of either Silicon Valley or Irvine, so it's great to have some additional folks here today and excellent.
New York is an impressive place and impressive city as you know, although I find that sometimes getting your mobile data to work in the middle of the day is not so good here, but we can help with that, and we'll talk a little bit about that. This is a group I think also that understands the importance of connectivity and bandwidth. You use that in your daily lives, and we see some very interesting trends happening in that space. We'll talk about that today, and how Broadcom, we believe, is uniquely positioned to really address the needs of an evolving world. You saw some of the eye candy coming in we had on the screen, and I think these are pretty interesting events that are going on. In the human species, we went hundreds of thousands of years before we had our first connected device. You can argue whether that was 1969 with the first ARPANET or maybe a little later with the first mobile phone or something like that. But next year, we will have the same number of connected devices as people on the planet, so quite a step forward. But we will make that next step to double that, okay, in less than 5 years. And so there is an exponential curve and an explosion of devices connecting to the Internet, connecting to each other, and that's going to continue to go on and go very quickly here. In fact, we believe by 2020, there are going to be 50 billion of these devices, so just an incredible growth.
So you have to ask yourself is, why is the incredible growth happening? What's driving this? And we believe it has to do with content. And if you look at the content consumptions and the traffic across the Internet, it started off fairly slow when we had devices like PCs, some of the original phones weren't very clever, SMS was this great invention that would allow you to do something not as good as an e-mail, sort of a tweet over your cell phone. It was miraculous when it happened. Kids still like it today. I think there's something really going on right now in the last decade where we've seen an explosion in that data traffic. And a great example is Netflix. Many of you have Netflix subscriptions, I imagine. There are statistics that show that if you take in the United States a typical evening, something like half of the entire Internet traffic is people watching movies over Netflix, okay, and I believe the delivery of content, movies moving to HD, moving to this rich content is going to increase dramatically over the next years, really pushing on the bandwidth capabilities of the Internet, really pushing on the devices. The richness of devices has really increased. Now you can carry with you a tablet or other device, very rich display, you expect to see high-quality media on that. And that's what's driving this curve up. And we think that the last 20 years is nothing compared to what the next 10 years are going to be. We believe that it's really going to take off here. And a couple of statistics to help you understand that, in the next few years we believe that something like 1 million minutes of video are going to be uploaded across the network every second, okay.
And it's also happening on the mobile side. And if you look at the growth rate of data traffic on the mobile side, it has a compound annual growth rate of something like 92%, okay, so that means 26x over the period from 2010 to 2015, so this explosion in content is really important. And I'll come back to this curve a little later and tell you why it's important to Broadcom and important to you as an investor.
So if we look at all of the different kinds of content, what's really enabling them is a dramatic multi-orders of magnitude improvement in communication standards. We put some of our favorite ones up here. You probably use many of these yourself, and we've seen anywhere from 10 to over 1,000 improvement in the speed that these technologies are capable of just in the last number of years. And the implications for this are not only does it enable that content you saw on the previous slides, but it allows you, for example, in your home to access very rich content. It allows you, we believe going forward, to eliminate the distinction between remote and local, okay. So in the past you always had to do the computing in front of you if you wanted to run the application there. Now you can run the application on the Web. The data can be on the Web. You've seen things like Siri in the iPhone, you pick up that, you call that. What it does is it transmits all the data to a Web application, runs it there, including interpreting your voice, and shifts the result back to the phone. Okay, so it doesn't actually have to run in the portable device. So this enablement of thin clients, okay, we believe is going to drive a lot of interesting changes going forward, again making devices less expensive and more powerful at the same time.
So let's talk about what does this mean to our customers. What are their challenges? Well I think their goal is basically to deliver on all of this bandwidth, okay, deliver more bandwidth and deliver functionality as much as they can. Now unfortunately they have some constraints. And so given that goal, some of their constraints are board space. Great example in a cell phone, you open up a cell phone, smartphone these days, there is no spare real estate in that device, it is absolutely crammed with electronics and other things and battery with everything that's left. So in order to add more functionality, it is a real challenge of how you do address the board space issue.
System cost is always a factor, that's always been a factor. There's usually some advantage in getting the price down, you'll see some elasticity in demand. People always want to get the system cost as low as possible.
And then power consumption. Power consumption is important in portable devices. I don't know about you guys, but if you have one of these new smartphones, usually by the middle of the day, you're looking for a power outlet, okay. That's not good. We need to fix that, so you have long time, long use on those devices. And in the data center, on the infrastructure side, the biggest cost component in a data center today is the electrical power, okay. It's not the cost of the hardware, it's not the people who run it, it's not the real estate, it's the cost of the power to run the devices and then to get all the heat out of the room when you're done. And so if you can reduce the power consumption, that's a tremendous gain there. So our customers want to solve all of these problems at the same time. And so this drives a couple of interesting things in our industry.
Let's look at a case study here of set-top boxes. And this is a set-top box. We took a picture of a circuit board of a set-top box from the year 2004. And you probably remember your set-top box from then. It sort of did a good job playing TV, that worked. If you look at the differences between 2004 and 2012, the functionality has absolutely exploded. The security has been upgraded dramatically because the content owners do not want you to rebuff their content. We've seen the advent of transcoders, and transcoders are going to be important going forward because they allow you to receive your video on the set-top box and then maybe sort of sling style watch it on your tablet, okay, and be able to have content anywhere you want it. We're seeing a dramatic increase in CPU speed. People want to run applications on these devices as well as have really nice user interfaces. And we see a lot of different ways these devices are going to connect, okay. HDMI just upgrading the video signal to your TV set but also connecting to the Internet, things with ethernet, Wi-Fi, Bluetooth and MoCA, different ways to connect across the home. And then power management because I think you've seen some studies that show that your set-top box may use as much power as your refrigerator, okay. It's on all the time, it uses a lot of power, okay. Well going forward, we'd like to reduce that to much lower power levels and so putting power management and integrating that into the box helps a lot.
So you would think if you added all of this content into the box, it would greatly increase the size of the box and the chips that go in it. Of course the answer is, no, it doesn't. You'll see in 2012, the goal of the industry has been to integrate a lot of these things, reduce the size, reduce the power. And in the device in 2012, we see a growth of something on the order of 20x the processor performance, at the same time, 5x more tuners, 6x more bandwidth and the capacity of the device itself and at the same time upgrading picture quality and other kinds of things. So these are really a challenge for all of the people in our industry as they go forward.
Let me use another example here. This is one where the complexity was the challenge and how can you use integration to reduce complexity. And in many networking devices -- high-speed networking devices, this is an example of a 10-gig switch. It had a large number of chips, okay, in it. And the goal is to reduce the size, at the same time increasing the performance of the device. And you see the 2011 version with one of our Trident chips here, dramatically smaller board space, 64 ports of 10 gigs, so an increase in the overall capability and at the same time lower power and lower cost. So this is our life in the semiconductor industry working with our customers to use integration to reduce complexity.
So let me go to what does it take for a semiconductor company to play in our world? Okay, if you want to do all those things that are going on in the last few slides I showed you, what do you have to have? We think there are 4 things that a semiconductor company has to have in order to be a solid player these days. The first one is R&D scale and execution. In other words, can you have the technology? Do you have the capabilities to do the technology? And can you execute on it, get it to market quickly? It's great if you can fund technology, but do you have the IP portfolio to do the kind of integration you had? Do you have all the pieces? Does your box of Legos, okay, have all the cool little parts that you need to make the desired object? Next is, do you have a complete solution? Okay, it's not always sufficient to just have the pieces you need to put in one box. Do you have the things that connect it elsewhere in the system, a complete end-to-end solution, perhaps a head end and a client end and a cable network, perhaps a base station and a smartphone in the wireless world? Do you have that complete solution? And then one thing we think that Broadcom is exceptionally good at, which is the ability to integrate. So let me look at each one of those in detail and explain where I think Broadcom comes out on these 4 essential things for an effective semiconductor competitor today.
First one is R&D scale, and when we look at Broadcom, we will spend about $2 billion on a GAAP basis this year in R&D. We're spending a large percentage. We have very high R&D intensity. We believe R&D is the source of innovation of Broadcom. That puts us #23 among all Fortune 500 companies in terms of absolute spend, #12 among all Fortune 500 technology companies and #3 in the Fortune 500 semiconductor companies. So the message is clear: Broadcom invests to win.
So the next question is IP portfolio. Do you have the breadth of IP? And what I've done here is I put up a list of companies, hopefully you're familiar with a few of these, and every year the IEEE organization, neutral third party, ranks all the technology companies in the world on their patent portfolios, and they don't just count the number of patents they have. They look at how powerful, and their words are, are they referenced by competitors, are they essential patents, are they important patents, are they really good patents, not just a body count number. And you can see here, Broadcom shows up #3 in this list, well ahead of many other people that are competitors of ours. And so this is an important area for us, an asset that we've created.
The next question is, who can put together a complete solution? And I've got a picture here illustrating somebody who is on their smartphone. Maybe they took a picture, and they're sending data. That data is going to go to a base station, a cellular base station. In that base station, it needs to get back to the Internet, that's called backhaul, it can be over microwave or fiber or Ethernet, different kinds of connections there. That will typically go into a wireless infrastructure, REN [ph] or RNG, going into a service provider, a packet network, to a data center where it might process an application. And then it goes to perhaps another user. If the person's at home, it could go over cable or DSL, or if they're in the office, it could go over Ethernet to their home. Typically, in their home, many people have Wi-Fi routers today, goes over Wi-Fi, and then to maybe their tablet where they would view this. It's interesting because Broadcom plays in every single one of these areas, okay. And in fact, we did an interesting calculation a few weeks ago that based on some conservative market share assumptions, we believe that 99.98% of all Internet traffic goes over at least one Broadcom chip. And in the case of wireless, which I'm showing in this picture, the answer is 100% because we have 100% market share in a lot of that wireless infrastructure that all the wireless data traffic goes over. And so the message here is: Broadcom can not only deliver the complete solution, but Broadcom is everywhere. And I'll say more about that a little later.
The next question is who can integrate things? And lots of companies talk about how they have pieces and whatnot, but who can actually put them together into a complete solution? We looked in the year 2007, this was an Ethernet switch back then. Year 2011, the number of things you had to put to have a competitive Ethernet switch went up dramatically. And the question we have for competitors, we believe no one else can build that chip on the right-hand side. Broadcom we believe is the only company that not only has the IP but has the ability to put it together and get it to work.
Let me give you another example. This is an example in connectivity. If we look at the year 2007 -- also Ethernet switch, sorry -- if we look at connectivity in the year 2008, we did our wireless LAN combo chip. It was wireless LAN, Bluetooth and FM radio. 4325 by the way is an incredible chip. I think we shipped close to 1 billion of those, okay, so very, very good chip in its time. However in the year 2011, we've improved that. We've created a enhanced die. It's dramatically smaller in size, and yet it has a lot more capabilities in wireless LAN, in Bluetooth and other features than its predecessor. And what's interesting is in this space, Broadcom has really taken the lead because of our ability to integrate. There is a long line of competitors who have said that they also have one of these chips. But you know what? They announce them, and they don't work. Okay, I think what's unique about Broadcom, we announce these chips, we put them in the market, they work, and they get rapid market acceptance. And we have yet to see a competitor able to do the level of integration we can in this space. Bob will talk later about how we plan to keep that advantage going forward.
So let me talk a little bit about our focus for 2012. One of the things that's very important for Broadcom is to expand our IP portfolio. I think we've done a great job creating probably the world's best communications IP portfolio. We're going to make that even bigger. We're going to continue to push new markets, new envelopes. We're going to invest across all our businesses, but we're going to put a special emphasis on investments in 2 areas. Let me highlight them for you. One is cellular. We're going to make an increased bet on winning in cellular in 2012. And we're going to invest more in wireless infrastructure because we believe these are some of the fastest-growing areas and areas that we, as a broad IP company with the ability to integrate, can deliver really good products in this space.
And our mission also in 2012 is to continue to enable this whole next generation of products that go in the hand, the home and the infrastructure that plumbs all this together. By the way, Rajiv today is going to talk about infrastructure, and many of you have asked me and others, too, are there businesses at Broadcom that are ever overlooked? And I think people take it for granted, they see all the wonderful devices in our home that's Dan's going to talk about, the wonderful devices in your hand that Bob's going to talk about, and it's sort of just like magic that these things connect together and the plumbing is magic, okay. Plumbing takes a lot of work to get right, okay, and we believe that's an important area. And anybody who sees all these devices exploding knows that you need a way to tie them together. So that's an area I encourage you to pay a lot of attention to.
So let me talk a little bit about this graph again. I showed this at the beginning of my presentation, and this is network traffic over time, and I think any company that participates in a market that's growing exponentially like this, that's a good thing, okay. Lots of companies are going to make money, okay. Lots of companies are going to ship a lot more products. And the interesting question though is, for Broadcom we are gaining share, we are going into new markets, how do we shape up versus this curve? And so for fun, I plotted the number of units that Broadcom ships over the same time period. And you can see that Broadcom is ahead of the curve on an exponential curve in a fantastic market. I think this reflects our ability to execute, our ability to invest in R&D, the broad IP portfolio you saw and utterly important, the ability to integrate, which many of our competitors lack.
So as I close here, I want to leave you with the point that Broadcom is everywhere. There's a tremendous amount of stuff going on in the home, the hand and infrastructure. As you hear about new innovations in any of these spaces, okay, to the extent that there are new business models to deliver content in the home like I think you're going to see next year, the smartphone revolution and the democratization of smartphones, enabling smartphones to penetrate a huge population that hasn't known them before, and then the plumbing or the infrastructure to glue all this together, if you see those markets growing, that's good for Broadcom because Broadcom is everywhere, whether it's nearly 100% of all the traffic on the Internet, whether it's a lot of these multibillion devices for the hand or all of the new innovations in the home, we plan to be there and we see this as a very vibrant growth, which is why we believe the next 10 years are going to be much better than the 20 years so far.
So with that, let me stop, and I will take some questions at this point.
Scott A. McGregor
Let see, we have some folks with some microphones. Let's see, we'll start right there. Please say who you are, who you're with and then your question.
Shawn R. Webster - Macquarie Research
I'm Shawn Webster with Macquarie. So I'll ask a competitive question to start off with. So the last month, one of your competitors was talking about a new product where they integrate Wi-Fi into their baseband. Can you share with us your thoughts on what are the catalysts from your perspective on when that makes sense for you guys or for the industry in general to integrate the whole Wi-Fi into integrated basebands and application processors, and maybe you can share with us your view on market share next year and how this new entrant would fit into how you're viewing the landscape next year.
Scott A. McGregor
Let me answer your question just a little bit and I'm going to leave most of it for Bob Rango after his presentation because he's going to cover that in some detail. But we've had competitors in this space for a long time. We had a number of people claiming that they were going to integrate different things. I would observe a couple of things. One is beware of people announcing the ability to integrate before they've actually done it, okay, because we find that often it's a lot harder than they thought, okay, number one. Number two, there's a cadence issue, okay, between different technologies. So baseband technology moves at a certain pace, okay, and takes a certain amount of time to get certified. Wireless LAN technology moves at a different pace, okay. Bluetooth moves at a different pace. And we have learned over time that sometimes those cadences are incompatible. And so if you have the world's latest baseband, okay, and you had to freeze the wireless LAN too early, okay, and would you want to put the world's greatest baseband together with something that would be old-fashioned in wireless LAN? Is that an effective combination for a high-end smartphone? We don't think so. And so those are 2 sort of conceptual problems that we've seen in competitors, and I'll let Bob talk a little more, and you'll see some of the specifics on specific technologies in those spaces, and we'll come back to that later. A question right there.
Vivek Arya - BofA Merrill Lynch, Research Division
Vivek Arya from Bank of America Merrill Lynch. Scott, so just conceptually as far as you look at the market, I mean, there are 2 types of customers, right? One is the high-end customer, and there are fewer of them, and so there's a lot of buying power, there's a lot of concentration on what they want to do with specs. But then you also have this emerging market-type customers where the focus is more on cost. So the question to you is, how are you balancing your R&D in this right sort of bifurcated customer environment? And what are you noticing in terms of pricing trends as the market develops along these different dimensions? Are the needs very different among these 2 types of customers? I guess how you are balancing this in your R&D.
Scott A. McGregor
So the answer, how we balance, very carefully. Let me say a little more. There's some examples, for example in the infrastructure space, where some of our high-end customers have very much more demanding specs than some of the lower-end customers or in some of the emerging countries, and so we're able to have different products for different customers. And this is the advantage of our breadth and our scale, that we can have products that are tailored to that. Another example, in our set-top box business, we targeted the Indian market, which is very rapidly growing, and we didn't have the right chip initially, and we used a U.S. set-top box chip, okay, which is total overkill for what the market needed in India. And one of the reasons we saw some softness in our margins a year ago in that business was because we had to price that chip to be competitive in that market. It was over-featured. We since come out with specialized chips for each of the markets so they can be feature-appropriate and therefore, margin reasonable in both of those markets, so examples where because of the amount of money we spend on R&D, we're able to tailor that market. We also find some examples, though, where for example some of our combo chips, because of the extreme volumes we get on combo chips for the smartphone space, we're able to create incredibly price-competitive combo chips for use in PCs and games and automotive and other things, and where the customers are just amazed how little the chip costs, okay, and that's because you get the huge-scale benefits doing it in the other space, even though that had higher-end feature requirements. So you do get the complete mix of those things. And again that's why a company like Broadcom, with our breadth of products, can really address all the different pieces in the market. Question back there.
Ruben Roy - Mizuho Securities USA Inc., Research Division
Scott, Ruben Roy from Mizuho. As you talk about focusing your investments on cellular, I was wondering if you could maybe talk about your investments in baseband to date, and how would you grade how well you've done with bases? How do you think you need to change as you look ahead? And where and how are you going to invest to perhaps become more successful in baseband?
Scott A. McGregor
That's a fantastic question. I'm going to let Bob tell that story, so let's come back. If we don't properly answer your question in Bob's presentation, please come back and we'll fill in the details. Any questions on this side of the room? There's one right here.
Harlan Sur - JP Morgan Chase & Co, Research Division
Harlan Sur with JP Morgan. Scott, you talked about Broadcom's ability to take advantage of your integration expertise. I think Moore's law has been a big enabler to that. And so along those lines, maybe you can give us an update on the company's transition to the 28-nanometer node. When should we expect first tapeouts from Broadcom, and when do you expect to see volume production of 28 nanometer?
Scott A. McGregor
So 28 nanometer is the logical follow-on from the 40 nanometer we're doing today. We have already done test tapeouts in 28 nanometer. We will tape out products in next year in 28 nanometer. I think that time before volume will be a while though, so I think we're still shipping more 65 than 40, so it takes a little while for that to catch up. So I think you'll start seeing just the first volume in 28 towards the end of next year but more 2013. And in 2014 I think is when you're going to start to see that hit stride and really start picking up on the volume. And then of course there's 20 nanometers after that, and we're also working on 20 nanometers. Question back there.
Romit J. Shah - Nomura Securities Co. Ltd., Research Division
I'm Romit Shah, Nomura Securities. You mentioned that you spent about $2 billion on R&D this year. Curious how much cellular is as a percentage, and considering that you said that cellular would be a focus area of investment for next year, how does that -- how does cellular as a percentage of R&D spending change in 2012?
Scott A. McGregor
So cellular, we don't break out cellular, but we look at wireless as a total. And the reason that makes sense is that I think the distinction between baseband and application processor and video processor and all that's boring. We believe that in the future you're not going to talk about baseband chips anymore; they won't exist. There'll be cellular SoCs, okay, and we can have an interesting discussion of exactly what is in each cellular SoC and then you put the connectivity with the baseband or not. Those are all interesting conversations. But we don't they're going to be standalone baseband chips or standalone application processors or standalone Wi-Fi. It just doesn't make sense going forward given the integration model. And so when we look at the overall wireless spend, it's probably about half of our total R&D, which is corresponding roughly almost to where our revenue is flowing on that business. So we're probably a little more concentrated on R&D than revenue in that space, but that's a rough ballpark for you. Question over here, right there.
Edward Charles Long - Gillespie, Robinson & Grimm, Inc.
Ed Long from Gillespie, Robinson & Grimm, Inc. With regard to the investments, what you are you seeing in terms of improving returns on capital, and are you seeing a less bag for the buck in terms of your future investment?
Scott A. McGregor
I don't think we're seeing a less bang for the buck in terms of future investments. If anything, I think we're laying the groundwork for some pretty significant return going forward. I would say that we probably see a bit of a transition more towards investing in things that aren't going to generate return immediately, okay, so more of an investment cycle as we look at cellular wireless infrastructure. Wireless infrastructure in particular and things like automotive, which we've also talked about, have a long time between when you make the investment and when you see the result, so I'd say we slightly skewed our R&D investments more towards those kind of things, but we believe the bang for the buck is absolutely there. And I think in the case of automotive and some of the things we're doing in wireless, we are absolutely the pioneers in that space, and we believe we've got a great opportunity to lead. All right, with that -- one more question.
Sort of from a very high level, if I look at Broadcom stock performance, you basically came on board or even a little bit before that, it's basically flat with NASDAQ. And I look at the amazing things you guys make, I look at the growth in your markets, I look at the value-added you bring to the market, and I feel like it's sort of a transfer of wealth or something to the consumer because I as a consumer benefit, but I as an investor, not benefit because your stock is basically flat with NASDAQ. And unfortunately most of us in this room have to invest in stocks, not companies, the difference being we have to make money. So what changes going forward that makes this different, this model that you have done and is not too, too different in terms of vision from maybe last year or the year or the year before, but what changes in terms of value added to the investor that we could expect to see over the next year or 18 months or pick any timeframe?
Scott A. McGregor
That's a fair question, and let me answer it fairly. I believe, as you point out, Broadcom has single-handedly done wonderful things for the consumers, okay, and that's a good thing. We've established the Broadcom foundation to do even more good things for society, and that's a good thing. But we are not an altruistic company. We are a company that is -- we intend to deliver value to our shareholders. Let me tell you what we've done to this point and what we see going forward. We have a goal in this company to run this company with financial controls such that we will outgrow our peers on the top line, and we will deliver a model profitability between 20% and 22%. That's our model. We have reduced stock-based compensation to within a fair shake of our competition in that space, still need to do some for hiring, okay, but we have taken that down, and we have increased the amount that we are returning to shareholders, both in terms of share buybacks, and we've initiated a dividend. Now -- right now, we're particularly challenged because the industry compression is pretty severe and P/E multiples are way down. We believe that we will continue to outperform our peers. We will continue to deliver more to shareholders. Eric's going to talk about this a little bit in his presentation and show sort of where we've been. I believe over time we can increase our dividend, our plan is to create value for shareholders going forward. I think if we deliver on some of the investments we're making here, we will dramatically grow our market share and dramatically grow our top line, and I think we'll be able to hold our bottom line such that we'll be able to increase that faster than our top line. So that's our plan to deliver value to the shareholders. It does require the markets cooperate a little bit if the P/Es are compressing from that same period of time, it's a little hard to show as much value as we could, but we believe on normalized market conditions, we will generate a good value for our shareholders and be a good stock to own.
Okay, with that, I'm going to stop here and turn it over to our next presenter. We'll come back later and answer additional questions.
Please welcome Bob Rango, Executive Vice President and General Manager of Broadcom's Mobile and Wireless Group.
Robert Americo Rango
Good afternoon. I'm going to talk about Mobile and Wireless, but first I want to address this picture up here, the Brooklyn Bridge. Most of you probably know I'm from New York. I was born in Queens, then I was brought up in Long Island, and I never really got a chance to see the Brooklyn Bridge because I took the 59th Street Bridge, and no one would ever put a picture of the 59th Street Bridge up on a slide. Those of you who know it, every time you drive over it, you think it's going to collapse.
Well our story today is really based on really history of great results, great markets that we're in and great products and technologies that are coming. Where other competitors are saying they're going to diversify into these spaces, Broadcom's already there, and this slide kind of shows all the different places that we touch your life. When you wake up in the morning, your cell phone, possibly your Samsung Android phone with a Broadcom chip inside, you go into the living room, perhaps your kids are playing on the Nintendo Wii with Broadcom Bluetooth and Wi-Fi, your spouse might be Christmas shopping on an Apple -- a MacBook or an iMAC, and then you probably even drive to work and the Ford SYNC has Broadcom chips inside. So we have a broad footprint of technology across all these bases, and I keep hearing competitors of ours talking about diversifying into these spaces, but we're already there. And one of the things I want to make a point of already here is that we have enough breadth to launch a new technology without having our competitors join in. We could have -- we could create our own ecosystem because our chips are already found in all the different places, whether it be gaming, whether it's cell phones, whether it's PCs, whether it's tablets, Broadcom chips are there. So we have the propensity to launch a new technology or a new standard without the help of any of our competitors.
So let me tell you just very briefly what I'm going to talk about today. I want to talk just briefly about 2011 highlights. I do want to touch on market dynamics, technology update, but I'll be spending most of my time on the mobile platform. Eric has made it clear that you're here today to hear about the mobile platform and our cellular baseband business, and based on the questions that just came up, I'm sure we'll have some more interesting questions coming up.
But first let me start with the 2011, it was a record year for Broadcom's Mobile and Wireless. This is the team that has the cellular baseband, our Wi-Fi business, our Bluetooth business, GPS, NFC, application processor. Record Mobile and Wireless revenues and earnings, we contributed more to Broadcom's bottom line than any other group in 2011 on average. And then we extended our wireless connectivity leadership. Record number of shipments, I believe about 1.4 billion combo chips have been shipped to date. And we've launched new technologies and new products in 2011.
We've significantly expanded our breadth of IT. You're going to hear a lot today about how important it is to have all of the intellectual property in one place, so it can be integrated. Through acquisitions and organic hiring, we've added to our capability to do 4G LTE. We've added to our capability to do application processors, mobile application processors, NFC, 802.11ac and a lot more things I'm going to talk about today.
And then we've accelerated execution on the complete mobile platforms. We've invested in cellular RF. We've invested in power management. We're able now to deliver a complete platform to a customer with an operating system like Android. The importance of that can't be overstated because once we have that complete platform, our customers can quickly adopt it. And I'm going to show examples of how quickly a complete platform in the mobile space can kind of change the game in cellular against our competition.
Now I have an intensive competitive quality, and I always look at competition very intensively, so I have to show you this that Broadcom outgrew the top 7 competitors in the mobile and wireless space from 2008 to 2011. You can see here that in compounded average growth rate, Broadcom grew 31%. Competitor A, which I'll call out because I have a lot of respect for competitor A, 14%. But then competitors B through G, competitors 2 through 6 are all shrink if you put them together. So from 2010 to 2011, Broadcom revenues for Mobile and Wireless, up 20% versus 1% for the other 7 competitors combined, okay. We're looking at an 8-competitor picture and into -- from 2010 to 2011, you put all the other 7 competitors together, they grew 1%, Broadcom grew 20%. And we're one of only 2 significant players to grow in 2011, okay. So now if you look at the table on the right, you see the relative share as measured by dollars. You get Mobile and Wireless's share measured by dollars, like any big market, my challenge is to go get market share from weaker competitors, and I plan to do that. But in addition to getting market share from the weaker competitors, some -- many of which are either shrinking or losing money, the pie is growing. So I'm optimistic about our possibilities to go get share in the upcoming years and continue to grow Mobile and Wireless. My challenge is to keep making the Mobile and Wireless Group the fastest-growing group within Broadcom.
Now our formula for success, you're going to hear a lot about this. I think Scott mentioned it already. Breadth of IP, having all the technologies in one place, it allows us to integrate them together, it allows them -- it allows us to put all these technologies into a mobile platform and deliver it to a customer very quickly. But it takes a lot more than that to win. It takes the ability to integrate them, it takes this complete system approach so you have to work out all of the issues with coexistence, and you have a single monolithic die with multiple receiver transmitters on it, you have to work all that out, you have to put Android or Microsoft Windows on top of your platform, you need to have a whole system approach. And then you have to be able to execute on time, on price point in order for a customer to really rely on you. We're planning projects right now 2, 3 years out. We're working with our customers on projects that are going to go to production in 2013 and 2014. They have to be able to trust you. So all of these aspects make up our formula for success. And those competitors, the same ones that I showed that were struggling in the last couple of years, they fall down. They just miss on one of these 4 elements.
Okay so let's switch gears a little bit to market dynamics. I think everyone knows that the wireless space, smartphones are growing, but let's see how fast they're growing. It's skyrocketing actually. I mean, the market for these smartphones, like the one I have in my hand, are growing at a 23% compounded average growth rate. It's growing 4x faster than the 2 billion unit cell phone market; if you look at all handsets perhaps, 2 billion units. Industry analysts are forecasting 2 billion units. So it's growing 4x as fast as that. It's providing Broadcom with a $25 billion opportunity. And who can say enough about a smartphone or what they've done to change the world. Clearly, the handheld computers has done a lot to change -- I'll talk more about this, but I'm very excited about what we can do. In fact, what I have in my hand is a Broadcom prototype, and I'll show a demo of it later in the presentation.
And then tablets, whether you consider a tablet a big smartphone with a bigger screen or whether you consider it a more portable PC, it's carrying a lot of Broadcom mobile and wireless technology, okay. In fact, Wi-Fi turns out to be the preferred way to get data in and out of a tablet, and you could argue the tablet is the ultimate media consumption device. Wi-Fi is the preferred way to get data in and out. We're #1 in Wi-Fi. And industry analysts say that within tablets, we have 90% market share, whether it's the Apple iPad, whether it's the Samsung Tab, whether it's the Motorola Zoom, a Broadcom chip inside. Industry analysts are saying cumulative, a little less than 500 million units over the next 4 years, so a good opportunity for us to grow.
Then if you look at all the other devices, connecting everything, the Internet of things, whatever jargon you'd like to call, I mean, all these other devices, PCs, gamings, personal media players, set-top boxes, cable modems, these are all opportunities for Broadcom's connectivity technologies, and you can see that it's growing at a 27% compounded average growth rate, so 3 billion radio opportunity. And I took out mobile, I took out cellular, I took out tablets. If you put them back in here, this more than doubles. I just didn't want it to double count. So not only are we a leader but a huge growth opportunity for us going forward.
Now last year, I presented this same slide, and I put forth this thesis that consumer electronic devices were on an increasing basis going to be adopting multiple forms of wireless connectivity. They would adopt Wi-Fi, Bluetooth, GPS, NFC. Of course, some of these devices are also adopting cellular modems like tablets. But I wanted to just focus on connectivity for a moment, and you could see the area under the curve from 2008 to 2014 significantly increased because all these devices, like cell phones, since the advent of the iPhone, Wi-Fi has become a standard feature. We don't even argue Wi-Fi into smartphone anymore. Same thing with tablets, 100% attach rate for Wi-Fi in tablets, and the list goes on. So I think it's actually accelerating, and proof of the fact that it's now even accelerating faster than what I predicted last year is Broadcom has launched over the last 12 months over 800 different consumer electronics products with our chips in partnership with our customer base over 800.
Okay, and our customers, I won't go into this in a lot of detail but certainly, all the big customers in the wireless space are buying chips from us. It's hard to find a smartphone without one or more Broadcom chips inside. The leaders in the PC space, leaders in tablets, as I mentioned, we have great relationships with our customers, and those relationships are based on respect, fairness and our ability to execute. We prove our worth to our customers. Our customers want to partner with us because we're able to execute, again, on time.
Okay, so switching gears again to the technology update. I just want to briefly touch at a high level before I dive into the mobile platforms. Okay, this slide, which has a lot of arrows on it, it's not intended for you to look at and try to figure it all out. The X axis is not time. It represents all of the technology vectors that my team is working on and is focused on and is going to deliver to the market over the -- either has delivered or will deliver to the market. The point of this slide is to say that there's an art and science to picking the right things to integrate. Scott touched on this, but if you look at all the things that are happening in cellular SoC, for example, you have 4G coming. You have a lot more 3G coming. 3G 21-megabit modem, you have new application processor technology, you have all of the 4G LTE story. On Wi-Fi, you have a whole new standard coming. And every new standard in Wi-Fi in the past has been a big event for the PC, for the cell phone market. And there's a lot of things happening within Wi-Fi, things like Wi-Fi Direct, just a lot of things changing. Same thing with Bluetooth, Bluetooth 4.0 just came out. Bluetooth low energy, these low-energy sensors that allow you to track your dogs or whatever you'd like to track in your home and feed information into your smartphone, okay.
GPS. I want to talk a little bit about GPS because there's been a big inflection point in GPS recently. It's been the addition of the satellite constellation cold GLONASS. The Russians also have a satellite constellation up there and just recently, they're allowing it for commercial purposes. It turns out that one of our competitors tried to out-integrate us by putting GPS into a wireless connectivity chip, but they forgot to put in GLONASS. GLONASS significantly improves urban area navigation, places like New York, where you're in these urban canyons. When you look up now, you can see you either U.S. satellites or Russian satellites. And now you can triangulate off of more satellites, and that's a significant improvement. So if you make a mistake like that, you can strand the function in the chip. You take a margin hit, okay. And that was a mistake. It turns out that probably every company in the world now wants GLONASS as part of their GPS subsystem.
And finally NFC, a new wireless connectivity technology, that has some very interesting killer applications that you're going to want when you get your 4G phone. Again all these things are changing, and I'm going to anticipate a question at the end of the presentation you're going to say, what's the next thing we integrate? What are we going to put into the next set of chips? And I'm probably not going to answer because I'm going to tell you it's a trade secret because it is very difficult to pick the right things to integrate. It's both an art and a science of understanding the cadence of technology as well as when is the right -- when can customers deal with that integration.
And I wanted to say that no company has out-integrated us. Broadcom is at the forefront of integration at this point. We've integrated our connectivity technologies into combo chips -- in connectivity combo chips, and we've integrated our app processors, cellular modems and graphic subsystems into cellular SoCs, and we'll spend a lot of time talking about cellular SoCs, but Broadcom is certainly at the forefront of IC integration.
Okay, so let's start with the mobile platform. Again this is the topic that I think that we aim to listen to. I'm certainly very excited about what happened here. If you look at the opportunity, it's tremendous, right? We have already penetrated smartphones with our connectivity chips, various forms and flavors, Bluetooth, Wi-Fi, NFC, GPS. That's worth about $6 to $8 content depending on the customer and depending on the volume, okay, but once you go to the complete platform and you add a cellular SoC to it, and I'll talk a lot about the cellular SoC, and the PMU, the Power Management Unit, and the cellular RF, along with the connectivity, the ASP significantly increases from $6 to $8 to $12 to $30. But how do you get to that? Let's look inside of the cell phone. How do you get to that ASP? Well if you look -- slightly simplified but bear with me, there's a couple of major subsystems. This is what constitutes the complete mobile smartphone platform: A combo chip, which is one or more connectivity technologies, Wi-Fi, Bluetooth, GPS, NFC; a cellular SoC, which is the app processor, the cellular modem, the graphic subsystems; a Power Management Unit; and a cellular RF. The goal for me is to make sure that no -- we're not dependent upon any other supplier to supply this. This is essentially the entire cell phone, okay, and I'm going to talk in the presentation about each of these.
Let's take a look. I mean, I just got done saying it's a huge market. Can you -- well you can't address that market. You need multiple chips to address the entire market because it's such a huge market. Just the smartphone market alone, as I mentioned, is going to grow to be 1 billion unit market. So we see the market in tiers. The first tier being the super phones, the premium super phones as operators are currently calling it. These are things like the iPhone or the Samsung GALAXY S II. In this class of phone, the handset company is going to pick the best in class app processor, the best in class modem. In a lot of cases they're not integrated in this tier. And they're going to pick the best connectivity, what I call most cases is Broadcom's premium wireless connectivity, okay, and this will be the place where we launch our 4G LTE solution that I'll talk about later in the presentation.
Okay, the next tier, which I think is just as exciting as the top tier, perhaps more exciting, is the mass market smartphones, okay. In places -- in these BRIC nations, where operators are not subsidizing the phones, the cost of the phone really matters, okay. In places like the U.S., cost of the phone doesn't matter because AT&T and Verizon are going to subsidize the phone, and you can get it for $199, but all these other countries, phone price matters. I'm going to talk a lot about this mass market smartphone today because we see it as a huge opportunity for Broadcom. This is the platform that will enable us to expand our customer base quite significantly. This is based -- this is our 3G HSPA plus, this is 14-megabit, 21-megabit modems that connect to the operators, now what I call Broadcom's value connectivity, okay. I want to make it clear, this is where all of our investments are going. I mean, I think that questions about where we're investing, this answers it. It's going into these 2 segments: 4G, 3G, connectivity technologies that I'm going to talk about, new technology, new standards.
They're certainly -- the feature phone business is still out there. We certainly are still selling into it. In places like China, there's still a big feature phone. We're selling into it with our 2.5G EDGE SoC solutions and our entry wireless connectivity.
So with these 3 tiers, we're able to address both the mature markets and emerging markets. Okay, the good news for me is that's a diversification strategy. If the European economy deteriorates further, I'm exposed to the developing country markets like China and India and later on, I'm going to show you how quickly things can change in a place like India in the smartphone space.
Okay, so back to that complete platform, all the IP that you need to compete in this place, there's very few companies that have all the IP that you need to compete in wireless connectivity, that's Wi-Fi, Bluetooth, GPS, NFC, the cellular RF, the power management and then the cellular SoC, perhaps one of the most complex chips in the cell phone. So if Broadcom has all of this technology, and I'm going to talk a little bit about each one. I'll skip PMU and RF. I won’t bore you with that. I think the key elements are connectivity and cellular SoC.
Let's start with the wireless connectivity. So Broadcom is clearly #1 in combo chips. I mean, just about every smartphone out there has adopted our combo chips. How do we win? We win because we have the broadest line, broadest product lineup: we have premium connectivity, we have value connectivity, so we have entry connectivity. We have the industry's most comprehensive chip lineup. We add to that operating systems. Independence, we can port our software very quickly to IOS or Android or Microsoft, and we've done that. I mean every -- whether you're on the Apple iPhone or any of the new Microsoft Windows 7 phones or many of the Android phones, probably 70% of the Android phones, there's a Broadcom combo chip inside. And then we go even further, we add more software, more value add, things like security features for Wi-Fi, Hotspot 2.0., I won't go into every detail, but Hotspot 2.0 is very interesting. AT&T just announced that they put Wi-Fi into these -- the parks here in New York. And with Hotspot 2.0, you could walk out of the Starbucks on Wi-Fi in your cell phone, walk into the park, walk in the Battery Park or walk in the Central Park, and you still are connected on your Wi-Fi network. I'll talk more about how much work the operators are doing to offload their licensed spectrum to Wi-Fi -- unlicensed spectrum later in the presentation -- InConcert, which does the Bluetooth Wi-Fi selection for you, and then it goes on. I mean, now Wi-Fi is being used for peer-to-peer. When you go peer-to-peer, you need a software application layer to facilitate that, so Wi-Fi Direct, Wi-Fi Display. We, Broadcom, provide that to our customers. And essentially what we do is make it very easy for customers to adopt our combo chips, and you see that. I mean, whether it's -- the latest example of it is China. If you look at the Chinese handset manufacturers out there today, what you'll see is Broadcom combo chips have almost, I think, taken over the smartphone's business in China.
Now we do have the best technology but, again, are we in the right phones? Are we in phones that consumers care about? Well I've put on this chart the Editors' Choice top smartphones. I think this is indicative of what consumers want in smartphones. And if you look at these 3 top Editors' Choice, Gizmodo, Broadcom is in 4 of the top 4 Gizmodo phones; CNET, 4 of the top 5; and PCWorld, 9 of the top 10. Now I want to make a mention that these are phones that operate on different carriers, there's different operating systems, and there's different cellular basebands inside. Okay, so to kind of exemplify the nimbleness of our connectivity business, I think this slide talks to that.
So just a quick update on NFC, a new technology that's coming. There's a lot of news in the press recently about NFCs, whether it's the sort of the debate between Verizon and Google on which wallet is going to get loaded on to Android phones, but that's all good news if you ask me. Broadcom recognized how important NFC was going to be a couple of years ago, and we started investing, we acquired companies, we put more people into it. And I'll give you this quick status report. There's some very killer applications that NFC enables: e-wallets, taking your -- turning your cell phone -- turning your smartphone into your wallet, getting rid of your wallet, so your wallet becomes a thing of the past, and it becomes even better than your wallet, that's called mobile payments; and simplified connectivity, okay, peer-to-peer connectivity, when you want to pair a smartphone, where you have your pictures and your video, to your TV, you do it simply by tapping them together, okay. So these are some interesting killer apps, and they're motivating third-party app companies to go develop NFC apps. That's good.
The ecosystem though is coming in place. Let me focus on the ecosystem for mobile payments because as I mentioned earlier, Broadcom has enough presence in all the various segments to make the simplified connectivity model work. In fact, you already see it working. You see companies like Nokia launching NFC with just simple pairing as the current used models, but the ecosystem for mobile payments is very interesting because what you see is Google Wallet. A lot more advertising, I just recently saw Citibank advertised the Google Wallet at compatibility with their online banking; Isis, which is the Verizon AT&T version; and then you have Vodafone and Orange launching their own wallets. So a lot of things are happening. I think the pieces are coming together between the carriers, the banks, the merchants and the OEMs.
In fact, I think in 2012, we'll see a big increase. I think industry analysts are already forecasting this, almost the tripling of the number of units on Wi-Fi radios in a number of different consumer electronics including smartphones. And I can announce today that the chip that we talked about last year, Broadcom's NFC-tagged reader chip already has established multiple design wins in the marketplace. I won't talk about those yet because I'd like to see those customers go to production, but multiple design wins I can tell you with our NFC chips.
Okay, now something that's really important, a whole new standard for Wi-Fi. I mean, I think most people in the audience remember the transition that I think Apple started with 802.11g, and then there was a transition to 802.11n about 3, 4 years ago. Well there's a new standard now called 802.11ac, and we think it has a tremendous amount of value. And this slide kind of talk to the value because if you look at the smartphones, what you see is there's 2 ways to get information in and out of a smartphone, right? There is your 3G channel, there's the operator channel, and there is the 802.11n channel, the unlicensed band. What's happening, as you see, is carriers like AT&T have very precious licensed spectrum and they want to take information off of -- you want to take video downloads, music downloads off of their 3G and put it on the Wi-Fi whenever they can. But one of the challenges is, there's more devices in your home, and all these devices want to share information, right?
You want to move information from your smartphone to your TV or from your smartphone to your PC. And in fact, over-the-top services are becoming even more prevalent so the amount of video that will be coming off the Internet over the next couple of years is going to increase very dramatically. So what's the point? The point is that 3G is going to run out of bandwidth, and we all know that 3G will get upgraded to 4G. 4G LTE is already being launched in the, what I'd call the mature nations like the U.S. You see AT&T and Verizon talking a lot about 4G, but that spectrum is precious spectrum and the carriers will still be very motivated to move information downloads from off of that licensed spectrum to Wi-Fi.
So our answer is to upgrade the Wi-Fi to 11ac. And in fact, 11ac goes very nicely with NFC, which is -- that's the universal symbol for NFC up there, you'll see that printed on phones coming in the future. You use NFC to pair devices and you use ac, which is 3x to 4x faster than 11n. It's lower powered, it's higher-power efficiency than 11n. So a new standard, and I'll talk about it, but we really believe that 802.11ac and NFC complement 4G LTE. When you get a phone with 4G, you want the user experiences that NFC enables: You want the e-wallet, you want the simple pairing, and you want ac so that you can share your videos at a faster rate than you could before because there's going to be more traffic on both the licensed and the unlicensed band over the next couple of years.
So we will be, I predict we will be first to market with 802.11ac. And as usual, as I just mentioned before, we're going to do more than just provide a chip that runs 11ac. The standard gives you this ability to run over this unlicensed spectrum on the 5 gigahertz band, the standard gives you the capability to run at a faster rate. In fact, 802.11ac will allow you to transmit data at gigabit rates.
So for those of you who don't follow 3G, 4G: On a good day, 4G, assuming you're not in downtown New York and you're not everybody contending for the same spectrum, 100 to 150 megabit is the 4G rate you'll get if you're the only one on the cell tower. 802.11ac will run at gigabit rates, okay, and better coverage because 802.11ac versus 11n will give you a better rate at range, as we call it. It will also be better power efficiency, okay? So, I mean, for those of you who want to hear about the power efficiency, I can talk about it later, but since we can use different coding schemes, we can be more efficient.
And then all the other things that are needed to have customers adopt it very quickly; everything from NFC to Wi-Fi Direct software, the software that sits under DLNA, the standard that allows you to move information in a living room. In the future, you won't have to worry about this. The app provider will -- you'll download an app, the app provider will kick the middleware and it will be transparent, but you will be able to move information very quickly from the Internet and from device to device. Okay, so 802.11ac is something I'm very excited about. I think it pairs very well with 4G.
Okay, so finally, I looked around the home, looked at some devices that were not Wi-Fi connected yet and I found a few. There's not many but there's a few. And if you look at that, appliances are not connected yet. Thermostats, for example, are not connected yet; health and fitness appliance. These things could all be made smarter and better if they can connect to the Internet. In fact, I'm giving an example here: If you can connect your appliances, there's an opportunity for retailers to make money. I mean, how much would you guys pay to know when your refrigerator is going to fail? If you put a Wi-Fi chip in it and you connect it to the Internet, perhaps you could predict when it was going to fail based on how long the motor has run, total number of hours that it's run. That's an opportunity for retailers to make money, so they're motivated.
When you add to the mix smartphones and tablets as the sort of the change agents with applications from third parties that allow you to control things like thermostats or home security systems, you really complete the picture, right? Because these wired third-party application vendors, they are also motivated to provide this ecosystem.
So just looking at this little mini-ecosystem here. It adds another $6 billion of TAM for our Wi-Fi business. So I've told our Wi-Fi guys, perhaps they need to raise their numbers because new TAM is coming online. And we do see it already. A lot of these appliances are going to go connected. And with Wi-Fi, you don't need any sort of intermediary, you don't need something like ZigBee where you have to convert into TCP/IP. Wi-Fi allows you to go directly to the Internet.
Okay, so covered connectivity. Back to this mobile platform and all the devices, I want to talk about the cellular SoC next. Cellular SoC has the modem in it, the cellular modem; applications processor; graphics and video. But before I get into that, I just want to dig a little bit deeper. I want to look at the smartphone market with a little bit finer granularity.
If you look underneath the covers, underneath the chart I showed you earlier, what you see is that the growth in smartphones is going to come from 3G and 4G, and Broadcom is investing in both 3G and 4G, okay? But what you see, here is ABI Research, their analysis shows that 3G, cumulative over this period, over the next 4 years from 2011 the 2015, 3 billion handsets, 3 billion; and 4G, 310 million, but a significant growth rate. Both of these markets are very attractive for us and we're investing in both markets.
But let's go one level deeper. Where is the growth? It turns out that -- this chart is showing mature markets like the mature countries versus emerging, the BRIC countries, and what you see is the opportunity for smartphones in this BRIC countries is almost twice as it is in the mature venues. And then if you scratch a little deeper under the surface and you see what industry analysts are saying about the price requirement of the smartphone: In order to address the emerging markets, the price has to be less than $300, and we're not talking about a subsidized phone now. We're saying $300 or less is where most of the volume is, okay, 2.4 billion units for devices that are less than $300, okay? So keep that in mind as we walk through this. Certainly, my goal is to address both of the markets. As I said earlier, the tiers for super phones we're addressing, mass-market smartphones we're addressing, that's where our investment is going.
So back to the SoC now, with all those things in mind about where the market is growing, where the money is to be made over the next 2 to 3 years. I figured you guys would want to hear about what happens in the next 12 to 24 months so I focused on the SoC roadmap for the mass-market smartphone.
Let's start with the application processor, mobile application processors, obviously something that's very important. Android, Microsoft Windows runs on the application processor, and Broadcom has invested a significant amount over the years to improve our performance. This chart shows our improvement over the last 4 generations of our cellular SoC chip. And we've made, from our first generation to our fourth, a 40x improvement in application processor performance.
And what we're doing now is we're applying all the things that we know how to do best, including latest process node technology. We have a great relationship with TSMC, we have access to the latest process node from them. And they just won't partner with anyone, they want to partner with a company that can add value. Obviously, to achieve gigahertz speeds on application processors, you need the latest process node technology.
Low-power design, most of our chips have multiple power islands. When you're not using some part of the chip, you have to turn it off to conserve battery life. And advanced packaging, because in order to deal with the power dissipation of some of these high-performance application processors, you need advanced packaging.
All of these things will go into our next-generation app processor, which will have a 120x improvement over the gen 1, but 3x compute improvement over the last the gen. So each generation, we're improving compute performance by 3x to 4x, okay?
And for the first time, we've been acknowledged by an industry analyst, Strategy Analytics. Just recently this month, Broadcom broke into the top 5 smartphone application processor chip supplier rankings, okay? So, and that's basically due to the chips that I'm going to talk about very shortly. So we're in the top 5 when it comes to application processors for smartphones.
Cellular modems, another very key component. We've been investing in this area for many years, and you can -- I won't bore you with a lot of the technical data but just a couple of the very key data points. Our phone's based on Broadcom cellular SoCs or operating on an increasing number of the world's important operators, and we're currently in an LTE carrier trial with our emerging LTE solution. When you look at power consumption from generation to generation, right, we have been able to reduce received power by 50%, going from one generation to the next on our cellular baseband, staying very competitive.
And we're widely considered to have the best RF team in the industry. And we even started investing in cellular RF technology a couple of years ago, actually, 4 or 5 years ago. Our 2G went into production, it was capable of supporting 4 bands. Our 3G is already in production, it can support up to 9 bands. And our 4G is in development and we plan to support up to 36 bands. So you can see the complexity increases going up significantly in order to support LTE, but we're up to the challenge.
Okay, so finally, the graphics. What you see here is an image out of one of the games that runs on Android, it's Modern Warfare 3. You can see how complex this image is. And what's important here is the amount of gradients on the color, on the shading, the resolution of the screen, how many frames a second you can render. Those are all things that are important. People are demanding a console gaming experience on their cell phone.
And Broadcom has been investing for many years, actually, since 2004 when we did an acquisition, in graphics. In fact, we call it VideoCore, and that, it's maybe a misnomer, it should be called MediaCore because this dedicated IP block does graphics, it does image signal processing. When your image comes off the camera, you need to post-process it, that's called image signal processing, okay? And it does video. So you can't do those functions well with standard application processors. You need to do that with dedicated hardware, dedicated customized hardware, and that's called VideoCore.
Now let's see how we do versus the industry's competition. One of the most recognized benchmarks that's out there is called Taiji. It's the OpenGL ES 2.0 benchmark most people will recognize as benchmark, as a very important benchmark. And what you see here is Broadcom versus Qualcomm versus TI. In fact, this TI chip, I think, is running the latest version of some of Ice Cream Sandwich phones that are out there. And you can see that Broadcom's VideoCore is able to render over 50 frames a second while some of the competition can barely get to 30. And in fact, just another data point comparing Broadcom VideoCore 4, all this -- again, this is a fair comparison because it's comparing what's in production to what's in production. Our VideoCore 4 is in production in many different Nokia phones, smartphones. And Nokia's multimedia experience is widely considered to be one of the best. Now comparing VideoCore 4, which again is in production, to one of Imagination's latest IP cores, we're 1/2 the power and 2x the performance.
So some of our competitors don't have this IP. They go often license it from a company like Imagination. It sounds good on paper until you have a problem. And a customer calls you up and says, "Hey, this game, this Modern Warfare 3 won't run," and that company has to go call Imagination. Okay, Broadcom doesn't have to do that. We're a one-stop shop. All this IP that I'm talking about is owned and within Broadcom so I can walk down the hall, knock on the engineer's door and say, "What were you thinking when you designed this?" and I usually get an answer very quickly. And I think that's the respect we have with our customers, okay? We have the IP in-house. Okay, so the industry's best graphics performance and power consumption.
Okay, so all that technology that I just mentioned goes into this roadmap that I call the mass-market smartphone roadmap. This addresses that 3 billion cumulative units from 2011 to 2015. It starts with this device called the 21552 and 21553. It has a gigahertz app processor, 7.2-megabit modem, it's complemented by 2 other devices that are sampling. The 21552 is in production.
These 2 other devices are sampling with single and dual gigahertz app processors, enhanced graphics and 7- and 21-megabit modems, okay? So these are the chips we announced earlier this year: 21654, 28150. One of our competitors recently said that, "Hey, in order to run Ice Cream Sandwich, you need a quad application processor." Well, a month after we got Ice Cream Sandwich, we ported it to our dual gigahertz application processor. It's actually in my hand and I wanted to show you a quick demo. Yes, it certainly is a prototype, but I do want to show you a demo.
Okay, so this is the phone based on Broadcom's dual gigahertz application processors. And I'm, you actually see it on the camera here, so I apologize, but it's certainly -- I'm going to hold it and you can see the Ice Cream Sandwich logo, there is the actual home screen. And one of the indicative things of Ice Cream Sandwich is the blending that you see on the icons and you can see that blending effect, that's a new feature. And then I just want to show you a video. This is a 1080p video from the movie Hugo.
Robert Americo Rango
Again, this is an HD video preloaded on the phone because it's the best way to do it here.
Okay, well, that's coming off through the phone, and I think you'll agree that's, it's good enough to watch. Long-term, it's just like a big screen. The resolution is 1080p, high definition.
And just for completeness, let me also show you: Angry Birds is already running on this prototype. This is one of my favorite games.
Robert Americo Rango
And I don't know if you guys play Angry Birds, but you can see the snappiness of the interface. And for those of you who don't play Angry Birds, the reason the birds are angry is the pigs stole their eggs. This is the beginning of the game. And I'll take one shot and you guys can bet on whether or not I'll be able to hit this with one shot, but here we go. Yay! I don't have too much time to play Angry Birds, I've only gotten past the first stage, okay?
So I think the proof is in the pudding that Broadcom has invested significant amount on app processors. Again, that was Ice Cream Sandwich, that's Android 4.0. The other place you can find Android 4.0 right now is on the Samsung Nexus and this is a very high-end phone, a very specialized phone.
So Ice Cream Sandwich is very relevant for the next 12 to 24 months. It's already running on these chips that I just showed you, and those are the chips that I think are going to really have a major impact on the marketplace. And I might have bored you a little bit with some of the technical specs, but let me show you how these chips do against real-life comparisons. Again, this is a slide that's indicative of how we're changing the economics of the smartphone space.
That 21654 that's based on a single gigahertz app processor, compare that to the HTC EVO 4G, which was on the editor's choice for 2011, okay? The graphics compares to a Motorola Droid 2, the modem compares to the Samsung GALAXY S. That Samsung GALAXY S was also on the editor's choice. But here's the kicker: The average bill of materials to build these 3 phones, based on TechInsights -- look at these phones, it's $158. Because of the 21654 level of integration and all of the RBOM, what we call the bill of materials surrounding the phone that we have integrated into the SoC, we can reduce the price of that phone down to 50%, down to $70 to $90, so a 50% reduction in what it cost the handset company to build a phone that has a single gigahertz app processor.
And if you want even more processor performance, let's look at the 28150, this is the dual gigahertz app processors. If you look at that, compare a Samsung GALAXY S II: 3D performance compares to the HTC EVO 3D; and the modem, the 21-megabit modem, compares to the Motorola ATRIX 2.
Now most of you may not even have these phones. Probably, most of you have iPhones, iPhone 4S, I would guess. So iPhone 4S, 14-megabit modem, unless you're on a CDMA2000 network, if you're on a CDMA2000 network, then your modem speed is 3.2 megabits per second. So my point is these chips are very competitive and they changed the economics of that mass-market smartphone. They allow us to go into the mass market and address it with these 2 chips. And it's not just 1 chip, it's 3 different chips: It's the 21552, the 21654 and the 28150. All at different price performance points, all targeted at the mass-market smartphone, okay?
Now maybe that's not enough, but here's another data point for you. Recently, that 21552, that first chip on the roadmap, was designed into a Samsung phone called the Samsung GALAXY Y. And because it was a complete platform, it was Broadcom RF, Broadcom Wi-Fi, Broadcom GPS and Broadcom Bluetooth, we put the whole platform together, we delivered it to Samsung, they were quickly able to adopt it, they were quickly able to deliver it. And in India, it's one of the hottest smartphones in India. It's got to be 1 of the 2 biggest markets in the world, I think it's got 1.3 billion people. Flipkart, which is the Amazon of India, it's the top-selling smartphone on the Flipkart website now. And The Mobile Indian, which is like the CNET of India, for 2 weeks in a row, has rated it numero uno for smartphones INR 10,000 or less and that would be $150 or less. So this is based on Broadcom's complete mobile platform, okay?
So now, earlier in the presentation, I said one of our goals is more customers. Certainly, Samsung continues to be a fantastic customer of ours, great partnership; Nokia, as well. We've talked a lot about Nokia launching 3G phones in 2012, that's an event that's still going to happen. But I want to talk about our new customers. We've made a point of expanding into China and we already have a new -- some new customers to talk about, including ZTE, ZTE was recently rated #5 in total handset sales in the world; other customers like TCL-Alcatel and YUHUA and G'FIVE.
As you get into China, what you'll realize is -- you might not recognize these brand names, but G'FIVE is the third-biggest cell phone provider in India. So these are very notable companies, and we have more models and more customers coming over the next year that we'll announce, okay? So more customers are already there with these phones have already been launched in operators in places like Europe.
And a by the way, this represents 2G, 3G and 4G footprint, okay? And what's important about 4G is we didn't start from scratch. Through the acquisition of a company working on OFDM technology already -- in fact, that company we acquired last year had already gone through 3 generations of OFDM, and OFDM is the same technology that's used in 4G LTE. They've shipped 20 million units into real operators, and this 21750 is one of their latest chips, okay? It turns out that we are working on a 4G LTE solution with a lot of characteristics that are built on the technology that we were able to acquire. Now we've added quite a few people to this effort already and we are working on a 4G chip with the following attributes.
It'll be multimode, for sure. 4G without multimode doesn't get you very far, as some of our competitors have learned. It'll be FDD and TDD, it will be cat 4 running at 150 megabits per second. And it is in carrier trials now, okay? So we are investing in the future. We're investing in application processors for the future, we're investing in 4G for the future.
Okay, so where does that leave us? Again, back to that intensive competitive analysis, I always like to look at how we're doing against our competitors. And what does it take to win in this space? Again, I showed you that phone, Samsung GALAXY Y, what other companies could deliver that phone in the time that we delivered it? Well, let's just start with the Android.
What do you need to build it? You need Android, and everybody can get Android. You need an app processor. Well, you can get a standard app processor from ARM, we all know that. So no competitors fall off the list yet, but the minute you add a competitive graphics subsystem, like I showed earlier, the VideoCore, one of our top 8 competitors -- 7 competitors falls off the list. If you add a competitive 2G, 3G modem, other competitors fall off the list; and a 4G modem, more competitors fall off the list. If you add state-of-the-art wireless connectivity like combo chips with 802.11n and 802.11ac, Bluetooth, GPS and NFC, other competitors fall off the list.
Notice you're left with, really, Broadcom and 1 other competitor that we think is viable to survive in this market long term. But you need more. You need the multimode RF I talked about, you need that multiband RF. You need smart PMUs. You need software drivers to stitch it all together, you need platform integration. You need to IOT-test it, GCF-test it, debug everything, be able to deliver a complete platform. In fact, some of the competitors fall off more than once, but I didn't want to do too much injustice by putting too many Xs through them, okay? So just Broadcom and really only 1 other competitor, we think, have the wherewithal to last in this market long term.
So you've probably heard that story from some other of our competitors, and again, back to that intensive competitive analysis that I do, I think our formula for success is paying off. If you look at our total revenue ranking in 2007 against those 7 same competitors, the top guys in mobile and wireless semiconductors in the world, in 2007, Broadcom was #6. In 2008, we moved up to #5. In 2010, we moved up to #3. The formula must be working because I'm proud to announce, in 2011, roughly, if everything stays on track between now and the end of the year, Broadcom will become #2. Okay, so by the end of 2011, which is December, we are the #2 guy right now and we think we're a very strong #2. Only a handful of these companies are actually making money in mobile and wireless, which is why I think it'll be a struggle for them to keep up with the investment, going forward.
So in summary, I think we have the right products for the right markets. I hope this, I really did convince you this mass-market cell phone, mass-market smartphone opportunity will be huge. And we are investing in 4G. So we're in the right markets and we have the right products for the right markets.
Our wireless connectivity continues to distance itself from the nearest competitors with 11ac and NFC, and these complete smartphone platforms are a huge opportunity. And we have everything it takes to go in so I feel very confident about our prospects, going forward.
Okay, so with that, I want to open it up to Q&A.
Robert Americo Rango
Yes? Oh, you need to get a speaker -- a microphone.
Shawn R. Webster - Macquarie Research
Shawn Webster, again, sorry. Two questions. One, my first question is, I was wondering if you could add to maybe Scott's response to my earlier question in terms of Wi-Fi integration your view on at what point is it a process node decision or more just when the technology matures, maybe expand on that. And then my second question, on NFC, I was wondering if you could wrap some kind of financials around your view on NFC in terms of how you're going to monetize that, make money on it, what do you think the market size is for the kind of products that you're going to be shipping and how that's going to evolve integration-wise, as well.
Robert Americo Rango
Well, I can't add much to what Scott said on Wi-Fi integration. I think the time for Wi-Fi integration will be when we know that including in a big SoC matches the cadence of that SoC and matches the value point of that SoC. I mean, I don't know about you, but when I buy my 4G phone, I want to make sure the Wi-Fi is as current as the cellular modem and I want those killer applications that NFC is going to bring. I mean, I already have a smartphone with 3G and 11n. My next phone is going to be 4G and it's going to have 802.11ac and it's going to have NFC. So we think the right complement for a 4G solution and a 4G phone is 802.11ac and NFC together. Now regarding your question about revenue for 2012 on NFC, I mean, we don't break out that individual piece, but industry analysts are predicting a big growth. And we see a tremendous amount of customer traction out there with NFC. I mean, that's as much as I can say, but I do think that our customers, our handset customers and even carriers have recognized the value of NFC. And the exciting part for us is -- again, to build the smartphone of the future, you need to have NFC, Wi-Fi, Bluetooth, GPS, cellular modem, app processor and graphics. I mean, all of that stuff will come together in one shape or the other. The question is: What's the right shape? And we think we have the right formula for that. Yes?
James Schneider - Goldman Sachs Group Inc., Research Division
Jim Schneider, Goldman Sachs. Bob, I was wondering if you can talk about expectations in the cellular business. I think that business for you on the SoC side has ramped slower than the market, as expected, broadly speaking. And could you talk about whether you think that's because it's mainly a technology issue in terms of baseband modem or applications processor or more of a marketing customer issue in terms of the initial target customers you picked or the level of high-end versus low-end smartphones you picked and what kind of milestones you would look to over the next year to convince us that you're making more progress there.
Robert Americo Rango
Okay, well, certainly, we started out in the cellular baseband business with 2 customers, Samsung and Nokia, as all of you know. And Nokia, unfortunately, has lost some market share. We know that they've lost their way and they're reorganizing, so that certainly has been a downside for us. But going forward, my goal is to take this complete platform and proliferate it. And the only way you can get that quick proliferation is with a complete platform. Again, if you -- I would like to compare what, years ago to today. Years ago, it took a handset company almost 3 years to develop if they piece-parted things together. With a complete mobile solution, it's a fraction of the time. I mean, and I think the Samsung GALAXY Y is indicative of how quickly the things can change. So I'm very optimistic. I think, with a complete platform and our ability to deliver things like Android on top of those chips I just talked to you about, including Ice Cream Sandwich that I demoed up here, I think we have a really promising future in cellular basebands and complementing, making sure that our cellular basebands are unified with our connectivity. Aligning those roadmaps together is one of my goals.
James Schneider - Goldman Sachs Group Inc., Research Division
In terms of milestones?
Robert Americo Rango
I'll be back here next year and I'll come back and tell you about more customers. And I think what you'll see is that those chips, like the 28150, going to production in 2012. And that's, I mean, back to Scott's cadence argument, these things do take some time to get to production. But once they're there, they're tremendously valuable and tremendously impactful in terms of how a customer can take them to market. And you see that with our customers already. I mean, China customers are taking our complete platform and proliferating it. Good questions. And now we're behind you, in terms of the microphone.
Ross Seymore - Deutsche Bank AG, Research Division
Bob, it's Ross Seymore from Deutsche Bank. Just a similar question on the cellular side. We've been for many years about the plan for Broadcom to get traction there whether for customer reasons or end market reasons. The connectivity side has done great. The cellular side has been falling behind a little bit more. You took that business over, and coming up on a year ago, what sort of changes have you made to emulate the success that you've had in the connectivity side on the cellular side?
Robert Americo Rango
Three words: smartphone, smartphone, smartphone.
Ross Seymore - Deutsche Bank AG, Research Division
So Broadcom didn't like smartphones before?
Robert Americo Rango
Well, I mean, before, we were focused on feature phones, and now we're focused on smartphones. Okay, I mean, that's really the short answer. And our connectivity was focused on smartphones and that's where we were enjoying such a great growth on Wi-Fi and Bluetooth. We were able to intersect the ramps. But certainly, well, we've changed other things to. We've changed the organization, but I won't bore you with that. Really, the focus is on focusing on smartphones for those 2 segments that we talked about before: these super-premium super phones, as operators are calling them now, and the mass-market smartphone. Yes?
Stacy A. Rasgon - Sanford C. Bernstein & Co., LLC., Research Division
Stacy Rasgon, Sanford Bernstein, a couple of questions. So first you were talking about increasing cellular investment in 2012, can you give us some feeling for how much more incremental investment on the revenue base you're doing now versus what you were doing previously? Secondly, how long does it actually take these incremental investments to actually start driving meaningful revenues? Should we expect revenues in 2012, or is it a 2013 or later issue? Third question: Do you have any -- can you give us any sort of, I guess, view of what the scales is required in the business to actually realize, at least at a minimum breakeven, profitability? How big does it need to be before it covers all your costs? And fourthly, given some of the headwinds that you have in one of your, at least in one of your customers, is this enough, do you see enough upside in the new opportunities to make up for that?
Robert Americo Rango
Okay, well, those are 4 questions. What I can tell you is barring any major economic event in places like Europe, which I know we're all watching, and probably, people in the audience have a much better understanding of it than I do, barring any of that, I plan on growing our business year over year. And we don't really break out a lot of those details in terms of investments, but I absolutely think that we're making the right investments. We've looked across all of the portfolio in our cellular business, like you would expect us to do after a change in focus from -- to these smartphones. I think you can already see the results on this roadmap. I mean, these chips are all smartphone chips. They all have gigahertz-plus application processors. They're already running state-of-the-art operating systems with ecosystem support, Android. We're going to build all of our chips to make sure they can support Microsoft, going forward, Microsoft Windows Phone. So we don't break out some of those specifics in your questions 2, 3 and 4, but I guess that I'm optimistic about our, the potential, going forward.
Sanjay Devgan - Morgan Stanley, Research Division
Bob, Sanjay Devgan from Morgan Stanley. Just to follow up, you talked about mobile and wireless being an investment-focus area for you next year, and you've already also talked about some of these new customers that you have coming on. I was wondering, as you look at the investment, is this more IP related, or is this kind of to support the new customers engagements? And then as a little bit of a follow-on, as you look at the opportunity to garner additional Tier 1s, how do you feel about your ability to support Tier 1s as you bring them on?
Robert Americo Rango
The investment is going into both supporting more customers and IP development. And I think our ability to bring on additional Tier 1s is good because we're putting that capability in place now. And again, I keep going back to this same thing, which is the complete platform allows us to kind of provide a standard product at a, with a higher level of functionality with, hopefully, less customer-specific support, okay? So I mean, as we know, Android, at least from a semiconductor standpoint, is the game changer. I mean, Google Android has kind of leveled the playing field in terms of semiconductor vendors. We all have to provide Android and Microsoft at this point. Perhaps, there's other operating systems coming, but those are the ones that we have to provide, and we think we have a great ability to do that. We've done that on the Samsung Y -- GALAXY Y. You'll see other Android phones coming. And I showed you Ice Cream Sandwich demo up on the stage. So investments going into both. I think I have time for a few more questions. People with a microphone can pick. I can see the hands, but I can't see the people with the microphones, so...
Ambrish Srivastava - BMO Capital Markets U.S.
There's one here, Bob. Ambrish from BMO. I just had 2 quick clarifications. You mentioned that VideoCore is in production with Nokia, and I was just wondering, is that with the Windows-based phones or the existing phones that you are in already? And then, I have a follow-up.
Robert Americo Rango
Okay, good question. On VideoCore, we're in production with Nokia Symbian phones, currently. On the Windows Phones that they're launching, we have connectivity in the Windows Phones: Wi-Fi, Bluetooth.
Ambrish Srivastava - BMO Capital Markets U.S.
Okay. And then my quick -- on timing, ac, what kind of timing should we expect for ac? And also, what ASP bump do we get?
Robert Americo Rango
Yes. So the timing on ac, we will launch 802.11ac in calendar year 2012. I'm trying to save some of the thunder for some of the guys who are slaving over all the details on launching 11ac and all the software enhancements that we do. I definitely will see an ASP boost from it. Just like 11n, I think it'll start at a premium. And then over time as optimizations happen and fab yields improve, I see all the prices coming down on 11ac. And there's other standards on Wi-Fi coming, by the way. I didn't talk about them, but if you guys look closely at the chart, there's 2 other standards that are coming, and we'll talk about that next year. Okay, one other question, I think?
Arnab K. Chanda - WJB Capital Group, Inc., Research Division
Yes, you can't see me, I have to stand up. It's Arnab Chanda from WJB. If I take your presentation, the tenor of it is you have a great market share in connectivity. You don't have a great market share in baseband, according to Broadcom standards, so you can improve that and it'll help your ASP. Now why would that be more likely than the opposite side where you have 90% market share in connectivity? It's a standards-based market. In other standards-based market, people don't maintain 90% market share. Baseband, on the other hand, where application processors is a system differentiator, it's easier to maintain a higher market share there. Why wouldn't the opposite happen? The guys that are not successful in connectivity today, why wouldn't they gain share? For example Atheros, outside of these handsets, used to have maybe 25% of the market, 30%, while you were a little bit higher. It wasn't such an amazing difference. Could you talk about that?
Robert Americo Rango
Yes. I mean, first of all, connectivity is not simpler than the cellular SoCs. All of our combo chips have a little embedded processor on them and there's a lot of software that runs on those processors and up on the host. And we have customers who are writing customized codes for them. So certainly, it's, there's a lot of challenges on connectivity. And number two is, if you talk about cellular SoCs, you're right, it is a challenging market in cellular SoCs. I wouldn't be confident if we hadn't started our investment many years ago and have been building over time all the capabilities we have within Broadcom. And I think that the success we have in Samsung GALAXY Y can be expanded. I think it can be -- I think that, once you prove that you can deliver a complex subsystem like this, a complete smartphone platform, customers really take notice. And you might ask, why? Why would they take notice? And that's because, really, there's only one company out there that's been able to do it reasonably well to this point so there is a huge need in the marketplace for a competitor to that company. We think we can fill it. So I really do think that we will gain share in cellular SoCs over the course of 12, 24 months.
Okay, so I am a couple of minutes over. I think that's enough. I think we're going to have a Q&A later and you can ask Scott some more questions on the mobile platform. With that, I want to thank you for coming and asking all those great questions. Thank you.
Please welcome Dan Marotta, Executive Vice President and General Manager of Broadcom's Broadband Communications Group.
Daniel A. Marotta
Okay, good afternoon, everybody, and thank you for attending our Analyst Day here in New York City. My name is Dan Marotta and I am responsible for the broadband business at Broadcom.
What I want to cover today are really 2 topics. One is, what are the growth drivers in the broadband business? Again, very different from the business Bob talked about, more of an industrial product business but obviously a lot of consumer trends there. What are the drivers? How do we see the broadband business growing? How do we see ourselves competing.
And then on the technology side, why are we able to win? What are we doing that has allowed us to really maintain a #1 position in these markets for since their inception, since the start of the very first QAM demodulator to the point today where we're the undisputed leader in these 2 product segments.
A couple of things on the growth drivers, if I could leave you with some key messages for the broadband business. Obviously, a lot of data today that you'll hear even more from Rajiv and from Eric and then further questions.
What's driving the broadband business today? The North American and European markets are very technology and feature rich. And although they aren't growing too quickly, the importance of these markets for the structure of the broadband business is paramount. If you look at many of our smaller competitors, we've driven them out of these markets, and without them, you can't maintain a healthy business, a model business within Broadcom. These are extremely important. And then, this technology shifts throughout the world, as I'll show you in the subsequent charts.
The second item is, a lot of our growth is coming from these BRIC markets that you're familiar with today. Obviously, a population and, in many cases, deregulation and digitization of cable and satellite networks that are providing a home for our products that we have today.
And then lastly, we're just becoming more and more competitive, and I'll show you many, many examples of that. A lot of it is also leveraging technology that Rajiv, and Bob have talked about.
So who is the broadband group? What do we do today? I mentioned that we're the leader in these markets. We have a long history, 20 years of participating in, initially, cable digitization, cable modems and then added many product lines today. But if you look at the set-top and broadband access, we're really covering all the product segments that relate to service providers and with this particular focus on the digital home. So we're covering all those product segments and they are quite complex and, in many cases, very fragmented, which for a leader like us, is very good.
Of course we're not satisfied with that. We've added some technologies since last year. One of them is the acquisition of a powerline networking company, Gigle, which I'll talk about shortly. That's given us powerline technology that we can use not only as a standalone product but in conjunction with our set-top box products particularly in Europe where powerline is a method to connect up set-top boxes to broadband access networks.
The second area of expansion is the femtocell market. You heard Scott and even Bob talk about the data that is being demanded by smartphone users. Filling those gaps in, filling the coverage in is an area where femtocell is excelling.
And then, other intrinsic areas within our business focused more on retail and set-top box, there's little pockets out there that are quite interesting. And then, of course, continuing to expand into adjacent markets that I'll talk about shortly.
Why is the broadband group winning? And this chart is just a example of many of the operators that we engage in today. And I think, from 20 years ago, and maybe we focused on a Comcast and a couple of operator, cable operators in North America. Today, we are engaged with virtually every operator in the world that is focused on broadband access and pay TV. And that is something that many of our competitors just do not have the scale, they don't have the talent, the sales reach and the technology to do. This isn't running around with a PowerPoint chart and showing it to an operator and getting on a plane and coming home. This is working on their middleware, engaging on their roadmap and tapping into their product development that they want to see 2 or 3 years from now and keeping that pipeline full. And I'll talk more about this chart and the implications and why Broadcom is winning and is able to basically tap into the trends. And many of these markets and customer, operator segments are highly fragmented, which for a new entrant, is quite daunting. For a company like us, they can take advantage of quite a defensible barrier.
One of the reasons we're winning at operators is we offer end-to-end solutions on many of the access products that we sell to the service provider community. Many of those come from Rajiv's business but also from broadband. And why is this important? Why is this just like any other product category? Because many of the standards that operators deploy all start with the infrastructure side. All the new link standards and features and functions start way into the infrastructure and are worked on many, many years in advance with very specialized scientists that many of our competitors just simply can't attract those type of personnel to work in the company.
The other benefit for a company like us is it gives us a catbird seat into the transition in the high-volume CPE segments. And it gives us a chance to have a more defensible entry strategy into the CPE markets where you typically would get more competitors.
How is broadband going to continue to grow our business? I know many of you believe that the broadband markets are growing slowly. And I -- if you look at this chart, the idea and the concept here is we have a two-pronged assault on growing the broadband business. The bottom is really the technology-rich developed countries Europe and America, which are adding lots of functions. I'm going to go through in some detail about the kind of trends in the market. Some of the user trends that even Bob talked about are relating back to the infrastructure or actually taking advantage of those as we speak today. And those, of course, are translating to increases in our selling prices of our products.
The other vector is additional share gains, adjacent markets that we're prosecuting and, in some cases, acquisitions that are getting us key technology that we can enter new market spaces.
One of the areas that is always a little confusing is the client-server transition that many operators are undertaking today, planning some ahead, some behind, mostly U.S. and European initiative. And one thing I want to do is debunk the myth that this will devalue the broadband set-top box business, and eventually, when there's an IP transition, all will just go away somewhere.
And I'm going to start out by just breaking the set-top box into the legacy set-top box today, which is a very highly functional, secured device. Although it's an island of functionality, that's not networking. And what does that mean? In my house, I have more TVs than set-top boxes. Operators would like to have a set-top box in all your nodes. That's how over-the-top is attacking them today. If you have a TV without a set-top, you're more than likely going to turn to a Netflix, a VUDU to get VoD content. That's not good for them, and of course, it's not good for us. We would like to have an expanded number of set-tops per home.
The other area is, because it's a standalone device, it has a lot of redundant technology when you look at your whole home, particularly hard disks, putting one per box, not necessary; and cable card, which is a security card that's quite costly that has to be put in each device, again, another $50 item that, if centralized, could be removed. And then lastly, because there's no network, you're missing a lot of features that I'm pointing out here.
Fast-forward to a client-server model, and all of a sudden, you're able to, because of the cost reduction and the elimination of redundant elements, you're able to add a set-top to each node. The operator can then garner VoD revenues on every TV outlet in your home. That expands our TAM. That's a good thing for us. They're eliminating the hard disks. As we know, the, on the recent news, they don't float. That's, the operators would like to get rid of the host. Cable cards are expensive. You just need one, really, in your home network today. That's going to reduce their SAC costs. That is going to allow them to shift capital towards outfitting your entire house, every node, with a set-top that they can garner VoD revenues. And this is even more important for a satellite operator where there's no satellite-ready television. If you don't have a receiver, you simply can't get the signal and that leads to some dissatisfaction and churn.
And then lastly, just the addition of a networking component. And for Broadcom, it doesn't matter what it is. It could be MoCA, very popular in North America; it could be Wi-Fi; it could be ethernet; it could be powerline. Allows a number of applications to be deployed. One of them is mobile device viewing, and I'll talk about that shortly and what that means to the content of our chips and the capability with just a simple adder.
So think about client-server. Increasing our TAM and removing and modernize the basic standalone set-top box architecture is a good thing for our broadband business.
Let's take a look at a typical home. You have maybe a couple of set-top boxes, like I have. You have a bunch of TVs that don't have those today. Operators would like to outfit your entire home with set-tops. That is good, but removing those redundant elements, and client-server concentrates all those elements in one set-top, thereby saving money and allows Broadcom to increase its available market because, now, operators have the ability, because of reduced subscriber cost, to drive this to every node, and I'll show you many examples of how this happens. The network that loops around these TVs, it could be anything today, we don't care, enables a lot of different applications.
The other thing that server allows you to do and some technology that we have developed, that we're going in production with today, is increase the number of tuners. Operators today are talking 6 to 8 tuners. In the future, we're building chips that will have up to 32 tuners, which is quite a transition away from a couple of tuning elements that are on a typical set-top today. Again, huge increases on functionality that we have developed, the breakthrough technology that allows you to put that kind of functionality in set-top box.
And Bob talked about all these mobile viewing options. Operators, of course, are very concerned and want to make sure they can get their content on these mobile devices so they're not disintermediated by maybe a younger generation that's very comfortable watching movies on an iPad. And by adding those into your home network, operators are adding dual Wi-Fi, not just Wi-Fi for your data but a separate channel for video because the video distribution has to be wireless in order to communicate with these new devices.
And then, the networking. Today, we're agnostic. We, with our recent acquisition and the Wi-Fi business we have today, we can go into an operator and, based on their demographics and their network, offer them any solution that they feel they can go to market with and serve their customers well. You add on top of that, with the network, the ability to use DLNA to basically view all your media that you have in your home. Of course, that's growing by leaps and bounds with all the digital photography and music distribution.
You might say, "Hey, this is going to be something far in the future. We'll all be retired when this happens." Here's a study of a service provider that many of you have today, DIRECTV. DIRECTV has always been a very innovative service provider, and the satellite guys typically are. They're under the most pressure in that they're selling a single service. I think DIRECTV, with their content distribution and some of the technology they deployed, is basically doing the client-server today. They have a server that's got 6 tuners, it's got MoCA built into it, it's being deployed in homes as we speak. They are introducing wireless clients. They're introducing digital outdoor unit that eventually is going to be an IP device that can deliver IP into a home network that Broadcom is participating in, and I'm going to talk about that shortly. And there is the link to mobile for control and also viewing content that, obviously, with our mobile business, there is a way to take advantage of that, along with viewing content within your home. It's quite advantageous for Broadcom.
So here is an example today, and I think many of you have seen the ads where DIRECTV is watching TV in any room, those are the start of this client-server model for DIRECTV. And other operators are moving very rapidly to transition their network, for obvious reasons.
Now turning to the gateway side of the equation. For most of you, you have a cable modem, a DSL modem, a fiber modem in your home. And it's a simple device, it takes in from the WAN network and gives you a ethernet jack and you connect it up to your stuff in your house, it all works. That's changing. And this picture I have here basically embodies a specification called universal services gateway that a major MSO is out there building. And we have built a reference design that basically incorporates all those functions. And you can see, if you look down your printed page, it is quite a Broadcom sideshow. Not only does it have a cable modem which I talked about, this breakthrough technology that has 8 tuners in it, but it has 2 Wi-Fi channels: 1 for video and 1 for data. It has a digital cordless phone built into it. It has a MoCA network built into it. And now you can see the huge value increase and the transition from your simple modem to a very complex gateway and which basically sets the stage for Broadcom to eventually integrate this and develop a very significant, defensible position in this important market. And basically, at a high level, it basically taken 3 functions: your router, your cordless phone and your modem, and combined them into a single platform.
And again, why is the operator focused on this? This mobile viewing has put great demands on the router and the Wi-Fi network such that the operator wants to control that channel to get a better user experience for mobile video. They want to have their own wired network, in this case it's MoCA, and they want to greatly increase the number of channels. And that's why they're shifting away from this traditional simple modem. Obviously, broadband being in a leadership position in this market, it's no surprise that we're in the forefront of this technology transition. All very good for us.
Now turning to some secular drivers in our business both on the shared growth side and also digitization and population growth that's driving the business today. I mean, everyone's heard of the growth in the BRIC markets, I don't think this is anything too surprising. But just to kind of put a point on 2 areas for us today, one is China. And a few years ago, the Chinese government put together an initiative called NGB or Next Generation Broadband where they basically were going to allow cable operators to, one, sell data service and, number two, drive rapidly towards digitizing their networks, essentially mimicking what the U.S. had done 20 years ago.
And I think the results are very significant for our company in that they are taking a lot of technology that was developed in the United States and rapidly transitioning in HD networking, multichannel receivers, video recording. And in the case of China, there is a couple of hundred million subscribers for their cable plant and those are expected to all eventually transition to digital broadcasting. And so it represents a huge market for us today and it's been growing very briskly, particularly the HD portion of it.
The second is India. And within the past, I don't know, couple of months, the Indian government, basically working with the cable operators, which is quite a fractured organization for those of you that are from India, to basically mandate that the Indian cable networks now digitize.
And why is that good for us? Because, now, they have to shift towards a traditional pay TV model and the expectation is there'll be a need of about 100 million set-top boxes to fuel this transition to a hybrid fiber coax network within the India cable network. I think one of the reasons they're doing it is, for those of you that are familiar, there's a flourishing direct-to-home business in India with 6 different satellite operators, pretty amazing that there are 6 operators actually broadcasting throughout India. Broadband group is shipping the 5 or 6 these today with our satellite products, and we look forward to working closely with the many, many hundreds of Indian cable operators to drive our cable products in the same direction.
I talked about a new breakthrough technology that we announced, it's called Full-Band Capture. And it's allowed us to both enhance our current products but also get into some new markets. And one of them is that outdoor unit, which is obviously synergistic with our leadership satellite business today. And what does this allow us to do? Today, everyone who has a satellite receiver has to have an outdoor unit and it's typically an analog device, and so you get a channel and you have a wire that goes to your set-top and you have another wire that goes to another set-top. This Full-Band Capture technology essentially eliminates all the RF circuitry in favor of digitizing this entire 2150 gigahertz signal with a digitizer, allowing us to essentially replace dozens and dozens of tuners with a monolithic device, setting this up to be eventually an IT device. So having a digital approach to this product allows us to enter this market which is quite synergistic with our existing satellite business. So very attractive and leveraged from this breakthrough technology that we're applying to many, many different markets within broadband.
I talked about the growth of digital and HD. And one of the expenses, capital expenses that operators look at today is reclamation. HD channels take up more space and there's been a lot of competition around HD channels between satellite operators and cable operators. And one thing Comcast looked at was how do I reclaim that analog spectrum in a way that keeps my customers satisfied but doesn't cost too much? And after a lot of study, they determined that reclaiming this with a new CTE device they call a Digital Transport Adapter, which I have picture here at the bottom of the screen, was highly capital-efficient, and their estimates are 10% of what it would cost to go and actually change their infrastructure they can -- for those of you that have a Comcast service, you probably got a couple of these in the mail. And you went to a TV that I talked about that didn't have a set-top box, and you hook this up and you've got standard definition digital service. And we've shipped a lot of these units and we had a very high market share for this product, again because of our close relationship and our ability to basically work closely with them to drive our roadmap. Time Warner is going to be introducing a similar reclamation device to do the same thing next year, and both Comcast and Time Warner are going to have an HD version so that you can get premium HD content on all these nodes that today don't have a set-top box that broadband will be selling you HD DTA very shortly.
This same product category, which is a purposeful product for this application, is also used to go address this huge analog cut-off the digital terrestrial market of which there's may be 1/3 of the countries have really shut off analog, the U.S. being one of them. So this is also a big growth and boom market which we're using similar technology with different front ends to be able to go attack that growth market, very exciting.
Not to forget the modem business. Digital music really drove the need to have a broadband modem. If you remember, downloading digital music was quite challenging if you didn't have a good broadband connection of some type. And now, video, whether it's IP video so that operates can deliver more data, user-created video that we've all talked about, Netflix, video within the homes, going from the 10-megabit modem that you may have today to a modem that goes into hundreds of megabits. That's going to outfit your house with a massive amount of capability to deliver video throughout your home, it's a key. And operators have laid the track and the infrastructures so that these can be deployed throughout the world. It's very capital-efficient. to change the end. You don't change the actual infrastructure, and they can briskly deploying DOCSIS 3.0, fiber here in New York City with Verizon and VDSL throughout the world. And this -- obviously, this business is something that Broadcom is the leader in today, and it's no surprise that one of the only company that's able to offer all the different access technology, and of course all the synergy and efficiencies that we can garner by being able to scale this business up.
I talked about powerline networking, home networking and what it means to us. It's an adder today. And there's actually a lot of synergy with powerline network across the company, and not just in connecting set-top boxes together for European operators who are maybe MoCA is not an option or Wi-Fi, but also synergies in backhauling Wi-Fi where you might want your -- access point is not in a good position. You need to backhaul that data to your cable modem, backhaul your Femtocell or maybe you need to have it in the right position in your home to get maximum coverage.
And then, next year, we'll be adding a new product category based on a new standard called G.hn, which is what AT&T and China Telecom are contemplating transitioning to for their next-generation home networking standard. So again, acquisition last year and being able to quickly build upon that with some new product category.
Femtocell product is something that we acquired from Percello also about a year ago. And today, we have a number of very key design wins for this 3G product. It's very low power, deploying the product today and obviously used to fill the gaps, whether it be in voice coverage or to fill the gaps in data coverage with different operators, both in Europe, the U.S. and in Asia. And we're actively working on our mixed mode product. Next year we'll be introducing a 3G LT mixed mode product that we can capitalize on all the technology and all the progress we've made on this product this year.
You know, Scott, I think even Bob, had a chart like this. And it's quite staggering, I know it's a very simplistic chart, to look at the IP that actually goes into our chips today. And certainly since we've amassed this over a 20-year period, it kind of slowly creeps up on you. But I want to touch on just one of these areas and just show you the actual implications from a selling price, and let's go back to mobile viewing, right?
Operators want to make sure that their content can be viewed mobile-y on tablets and phones so that that next-generation isn't watching Netflix, is watching their content. They are able to send the content directly to this iPad via a specialized wireless connection, a 5-gig connection. And that obviously is good for us. We're able to incorporate that to our set-top box product.
And number two, the picture, the picture that's actually on the screen here is a 1080p picture, not suitable for your iPad. iPad uses 720p. And so there is immediately a demand for a very complex device called the transcoder that's going to convert to this picture to the picture that's appropriate for your phone or your iPad and do it in real-time because the operator can't do that at the head end. There's a lot of rights issues with doing that. You need to do it locally.
And so a product that we're going to production today has 2 transcoders. Operators wanted to support 2 tablets. And as we completed that product, a guy like Verizon comes up and says, "Hey, I want 4 of these. They want their whole nuclear family sitting around with 4. And of course we said, "Absolutely. We will add 4 of these devices into the next-generation chip that we're at, that we'll be working on taping out."
So here's a new application that sprung up, and immediately, a direct impact with quite a complex function which there's actually startups out there that's been building stand-alone chips. And here we are integrating that function very cost effectively, which is going to be a key enabler for adding that across the whole set-top box ecosystem.
One of the reasons that broadband has been successful, and if you refer back to the worldwide chart there of all the operators, since the start of the group 20 years ago, we have done extremely well with being able to go out and work with operators, embrace what they're doing and internalize that with our own roadmap and quickly transition along this curve that I've outlined here and not react to it but drive it. And as an example, many of our competitors and why the U.S. market is so important, in the MPEG-4 transition, we actually lost a lot of competition in North America. They had made a huge mistake in transitioning to this technology. Through an acquisition, a lot of hard work, we were able to capture significant share during that technology transition, very disruptive.
Another example is MoCA, the home networking. We were able to incorporate that in our product and essentially have a differentiated product against the whole competitive base where they had to use a discreet solution and we had an integrated and very cost-effective solution. So these transitions, if done in a reactive way, can be very devastating to your business.
Let me give you a more concrete example quickly. Here's that chip I talked about, the 7425, a chip we've already announced, has those 2 transcoders I talked about. You can see not a trivial amount of area, and then the cost of the chip is proportional to the area, minus the package. This chip has a dual core 1.3 gigahertz custom CPU that my team developed, so very high performance. Bob talked about this video core processor. Well, guess what is in this product. There's a video core processor that we scaled up, we've boost up, and essentially the same software and firmware we take from Bob [ph] in Cambridge, England. And we're able to leverage, and it's extremely high performance. And we can keep scaling that up to basically hit different performance levels based on whether a customer wants Android or whatever it may be in order to scale up with their middle ware.
And it's not really who can just build this chip. Let's just say a lot of our competition could gain access to this IT. I think the secondary factors are also very interesting. What companies have the methodology? What companies have the manufacturing capability to build it? Very nontrivial, just throw this and send a file to TSMC, and it comes back and it's a great chip. That's how some people think but that's not what happens. Our manufacturing capability is very significant to do this. So it's not just the IT, but who can make a device like this to make money out of it is a trick in itself. And of course, it's developed with a couple of hundred people thrown all over the world. The management to do that is also a trade secret as well.
Going back to the competition, and I think Eric will touch on this, our competition has actually begun to shift the focus players with this whole worldwide fragmented onslaught that you have to go and prosecute in order to become a real player, has weighed very heavily on their share and of course their profit. And I think you guys have seen some of that in the press. Some of the competition I've listed here has suffered, and they are losing money. And things have begun to shift towards large-scale players, not necessarily that they'll do very well, but at least they have the money and they can as they go out of business. And I think that's a tribute to the just the strength of our team, the IP and our ability just to not only maintain a #1 position but actually continue to increase that.
Finally, I hope I've left you with a good prospect for our business. And that although we have a very significant business today in this space, I believe we actually can continue to grow share, and we will grow share both in North America and in the BRIC countries. Today, our market is over $4 billion. We have more than 1/2 that market, but we will be growing share next year in North America and of course in the BRIC countries with population growth and share capture as well.
These functionality increases. And I think it's very clear that a lot of these new applications, which are real applications that operators and consumers are demanding today, is adding significantly to our selling prices. And we're able to capitalize on that now because of the strength of our team, and in some cases, being able to take those acquisitions and quickly internalize that and utilize their technology.
And then of course, you wrap Broadcom around the broadband business and it's just an amazing story over the past 20 years to where we are today and the future is very bright and very exciting.
Thank you. I guess we'll take some questions. Right up here in front.
Muni Rivers [ph] with Trust Company of the West. The cable operator vision of the world is set-top box-centric. Is that how they look at clients? It seems that we have a whole new generation of people who just want a high-speed connection, and that's it. They don't want to deal with set-top boxes. They don't want to have wires hanging from TVs, get flat screen, beautiful sets, and you have all these stuff dangling around. So with that, I mean, maybe the set-top box is not going to be the center of the universe?
Daniel A. Marotta
Well, the question is, and I think there's all of the misconceptions. Operators, a set-top box is a necessary evil. It is not the service that they provide today. A set-top box is a security device, right? And operators, if they could snap their fingers and have secure content to guarantee ESPN and Disney that nobody would have left their service, and Comcast is a media delivery company and an Internet access company. And the set-top box is really the vehicle to get it there. And I think those devices are getting smaller and more mobile, and the interfaces are getting more modernized. So I don't view that as so much the issue. It's really, it's all about the content. If you look at a company like Comcast, with their NBC Universal and Time Warner, the content part of the equation is always the key. They have this reoccurring revenue and the set-top box is that necessary evil that it takes to distribute that in a secure way.
Right there, third row.
Harlan Sur - JP Morgan Chase & Co, Research Division
Dan, I'm Harlan Sur with JPMorgan. As your customers move towards this next-generation connected home architecture, media server and, I think, client architecture, I think there's a view out there that semiconductor content potentially goes down because obviously you're not putting a full-blown set-top box under every TV. Has Broadcom done any studies to ascertain whether or not silicon content or dollar opportunity actually increases or decreases for Broadcom as you move towards this next-generation architectures?
Daniel A. Marotta
Good question. It will increase. And I think a couple of things I tried to highlight here. One is the TAM expanded. There's just more units that maybe you, like me I've got 2 set-tops, I got 6 TVs, okay? What, that's not good. I'm watching Netflix with a TV without a set-top. Comcast doesn't like that, I'm telling you that right now. Number two is they're shifting a lot of functions around. Operators are going to have 24 to 32 tuners in a centralized server. There won't be 2 in every set-top, that's inefficient to do that. They're going throw away the discs and just have one disc. They're going to throw away the cable cards. That's where the savings is coming from. For broadband for Broadcom, increasing the TAM significantly because in the U.S., Europe, certainly the number of TVs is out of mix with where the VoD content. And I think there's a real competitive threat over the top. It's going to be what's viewed on those nodes and they're going to lose 1/2 of subscribers. They're going to lose that VoD subscriber unless they do something. I pointed out Digital Transport Adapters, that's just some of it, give some premium content capability. So I see this as extremely positive.
Question way in the back?
John Pitzer - Crédit Suisse AG, Research Division
John Pitzer with Credit Suisse. One near-term, one long-term question. On the near term, my understanding is about 1/2 of your set-top boxes are shipped with HDDs, so I'd like to get your perspective on the Thailand issue and I guess specifically, how to drive manufacturers go through the allocation process when things get tight. And then secondly, on a longer-term view, relative to all the growth drivers you put up there, where do you think that the 5-year growth of this business will be versus the last 5 years? And can you help sort of quantify, are we just kind of treading water here to have a similar growth rate? Or do you think some of the growth drivers you talked about are actually additive to what you had been growing over the last 5 years?
Daniel A. Marotta
Well, let me answer the second one first. I think they are additive. And I think if you look at our long-term growth rate, let's say, from the start of the decade, it's been pretty significant. Although a little bit stair step, I think we have every other year, we jump up and get a huge piece of share somewhere. It could be a drag, it could be an echo, and we jump up a lot. And I think this year has been one of those kind of years coming off of '10. And I think a lot of the technology and expansions throughout the world market is going to allow us to go take a big chunk out of the remaining $2.4 billion in share that we don't have today. And I think, when guys look at our narrow competition, and I think the results are there. If you look at their P&L, that we have taken significant share there. Going to the hard disk issue, I mean, we talked to all our customers for hard disk because it reflects back in our forecast. I think we've taken care of that in Q4 and in Q1. Everyone has a different approach to it. People are out there trying to change suppliers and change mix. You have operators actually buying units themselves and paying higher prices because obviously getting subscribers is more important in the short-term than set cost increases. So everyone has a different approach, and in some segments, it's worse than others. You've seen news from Pace [ph] but you haven't seen things from other customers, and I think we've taken these things into account at this point. Question back there?
Just in flavor with my previous question. If I look at your growth driver in developed markets and emerging markets, they were pretty much the same last year. And you claim that you gained market share this year, yet your broadband revenues were declining this year. So can you help me understand why it is that this market is not growing? And when you look at next year, do you expect to grow? And what is a reasonable growth expectation from this division?
Daniel A. Marotta
Sure. I think in terms of getting market share, I think one thing about our business is a lot of our products take a while to go to production. And so there is no question that we have gained market share. And as you break it down regionally, and I can talk with you offline about it, they -- things take a while to go to production. And some of the products that were designed in last year are really going into production in the back half of 2012, maybe even beyond. So it's a business that takes a 1.5 years to really get a product in production which snatch away some business from one of our competitors. And I think in the case of these BRIC markets, the same is true. Although they have the population growth, it takes time to digest that. And I think we're also coming off the back of 2010 that was very strong, and in some cases, some of our customers are getting a little carried away in the back half of the year.
Oh we got one in the very, very back.
Daniel A. Marotta
I believe so, yes. I'm not going to comment on the number here, but I could talk with you offline about it.
Brian Blair - Wedge Partners Corporation
Brian Blair from Wedge Partners. Given some of your comments about the move from legacy set-top box to this client/server model, can you talk a little bit about how you view Microsoft's recent move, the move last week, where the Xbox now can actually deliver cable content. I mean, they have -- files is on there as of yesterday. Comcast is about to deliver an app. So that now becomes, in essence, a smart cable box that has all these other features, social features, gaming features. Can you talk about how you view that, maybe looking out a year? And do you see it as a meaningful threat over time to cable set-top boxes?
Daniel A. Marotta
Well, I think it's actually interesting and it wouldn't be a surprise that we're talking with Microsoft and we have a relationship. They have a large set-top infrastructure supplier to AT&T and Swisscom and others. And so the units, as of today, is for clear channels. And I think they have some vision of incorporating secure features in this device, and it wouldn't be a surprise if we were talking to them. And it could be very interesting for sure, I think, given the success of their product. And this can easily be an interesting retail vehicle, kind of like a TiVo but with the attraction of the game box that we would be interested in, and it might be a surprise that we are talking with them. And it's something that we certainly want to follow a path towards a secure capability versus what's there today is clear channel like a television. And the question is, will that migrate towards a secure device? And I believe if they wanted to, that would be a good path for that.
Brian Blair - Wedge Partners Corporation
It must be clearly secured though right now, if cable guys have said, "Look through our content over," right? Let's have enough security now?
Daniel A. Marotta
Sure. And that's something that's interesting to us as well.
Let's see, the middle of the room there.
Barry Stewart with Omega Advisors. Early in your talk, I think you spoke of the some households will have a super cable box with Wi-Fi and it will transmit various streams to the other televisions in the household. But wouldn't it be feasible to instead of using a bunch of other TVs to each member of the family once they have a tablet, but maybe if they only have 1 or 2 TVs conducted and the other people are all looking at a tablet. If that's the case, it's hard to see that your TAM would be improved. So tell me is that practicable and...
Daniel A. Marotta
Well, that's in fact what the point of adding that function is. It's not to stream the data to the television, but what this operator is trying to achieve is that what they find is that using off-the-shelf routers connected to the cable modem, the experience of getting that video Wi-Fi-ed to the tablet as user content, that's the intention, is -- got a lot of issues. There's a lot of problems with doing it. There's a lot of satisfaction issues with just lighting up that service. So what they're trying to do is control that hardware and that software so that they can control that experience to make sure when that video goes through, it hits the tablet with some channel that it actually works properly. So it's not the deal with the TV, it's actually make the streaming to a tablet have a very good consumer experience. Why that's important is they're collecting all that into a much more advanced modem and setting the stage for a company like Broadcom to be able to integrate that and of course offer all those functions today, increasing our profit.
One more? One more question? So way over to the side there. I hope you’re the last one. First, we have a little reception here so we can chat further.
Alex Gauna - JMP Securities LLC, Research Division
Alex Gauna, JMP Securities. If I'm not mistaken, your trial alongside of an Intel-based solution for the next generation our view set-top boxes from DIRECTV, can you talk about what's different about your solution versus them. And is this a bake-off? Or do you expect that to be a split solution going forward? And then also, with DIRECTV, I know you're involved with them in terms of the set-top box less Wi-Fi solution. How do you participate in that from a TV side? And does it need .11 ac before it goes forward? Is it a 3x3? And what do you think we'll see from TVs?
Daniel A. Marotta
Yes, I think on the Wi-Fi, I think DIRECT is looking at today 3x3, but they will upgrade like everyone else. Obviously, they can't get their hands on that today. I think an operator probably a little bit behind a PC guy that typically are early adopters of the technology, Bob talked about this 802.11 ac. But there's no reason why an operator like DIRECT would not move to that. It's just the natural consequence of when the product is mature and pricing and so forth. But today, a guy like DIRECT is looking at 3x3. I think in the case of an our view client, which is a TV client, I think those -- it just depends on the TV manufacturer. I'm not sure the Intel thing makes sense. But on the TV client, it would be whoever makes the TV. Makes sense?
Okay, thank you. Thank you very much everyone.
We will now take a break. Please be back in 15 minutes.
Please welcome Rajiv Ramaswami, Executive Vice President and General Manager of Broadcom's Infrastructure & Networking Group.
Good afternoon, folks, and welcome back. I see some of you are still settling in. Come on in, take a seat. I'm here to talk to you about our Infrastructure & Networking business. When you guys think about Broadcom, a lot of you think about Bob's business and the mobile wireless, and you know a lot about Dan's business, the broadband group. And what I find as I've been hitting the road talking to you folks over the last of couple of years here is not much is known about the infrastructure business. So I'm going to start out by giving you a very quick bird's eye view on what this business is all about before we really dive into it, and take it from there.
So if you look at it at a very high level, this Infrastructure & Networking business really what it is, it's a set of products that a network could connect users or people to the content that they want to see. That's what we do. You can think of it as a black box in the middle, but we decided to put a red box in there, it's the holiday season after all, and just look at what's in that box. And you see here, people -- that's people at work at the bottom, people watching TV at home and people on the road. So whether -- wherever you're located and whatever form you are accessing the content, you're getting the content across the network. And that, in a nutshell, is what this business is all about.
Now let's open up that black box, or red box I should say, and see what's in there. So when you open this thing up, it actually gets a little complicated, but let me summarize the different components of this business. There is a service provider network, that's in there. The service provider network has a mobile portion, that's sitting in here, there's also a fixed access portion, and there is the service provider packet core. So really 3 components to that service provider network. You have your enterprise LAN, enterprise local area network, and then you have the data centers. You have your enterprise data centers and you have your cloud data centers. And as we all know, cloud data centers are getting bigger and bigger, more and more contents. In fact, people talk about 70% to 80% of contents out there now being unstructured data and generated by users other than operations out there.
So if you look at the 3 sides of different components, and there's networking in each of this, there's the service provider network enterprise network and the data center network. And if you look at the infrastructure business today, what we do is provide solutions for each of these places in the network.
Now this is not something that you guys start and peel and get to see. And to many people, it looks like it's plumbing. But plumbing is actually a very good honorable business to be in. Now it's about 25% of the business for Broadcom, and the telephone [ph] is about 40% of the overall profit, so it's a very profitable business to be in. And I'd be happy to be a plumber any day with this kind of performance here.
Now if you -- the reason I just -- at that way is because this is a highly complex business. It's not easy to get in the significant barriers to entry. There's a vast diversity of products and solutions required to tap into each place in that market. There is significant software hurdles to get over. And the net-net of this is there's a lot of room for differentiated solutions in the market and stickiness with our customer set.
So that's the reason we are where we are today, and we are pretty much across all of these markets. Just to give you a rough breakdown of the revenue in the infrastructure group, about 40% of the revenue is coming from the service provider portion, about 30% is coming from the enterprise LAN, about 20% comes from the data center. We also have about 10% of our business coming from small and medium businesses that's not shown on this chart. And we also started making a foray into the automotive networking wall, with still many years ahead, and really not a factor from a revenue perspective in the short term.
So that, in a nutshell, is what this business is about. And what's happening in this network, in this business here is that there's massive transformation, massive demand growth happening. So let's talk about what is going to be driving growth in this business. So for the next 20 minutes or so, we'll talk to you about what's happening from a demand perspective on the infrastructure, going exponentially as you can imagine.
In response to all those demands, these networks that you saw in the picture there are transforming. They're changing in very fundamental ways. Broadcom today, we are already a leader in this market. And with the transformation happening, we're not sitting quietly, we're executing through those transformations, anticipating those transformations. And we are in a unique position to continue to lead in this market. That's what we're going to cover in the next 25 minutes. So let's start with the demand picture.
So I'm not going to go through every one of these, and it's shown over there, all these demands are something that you see. You see social networking being an integral part of your life, you see smartphone products operating, plus more tablets, driving more and more consumer broadband growth, Verizon deploying FiOS for example. And then you see the advent of cloud computing in both the enterprise markets, enterprise data center and as well as in public cloud data centers like Amazon or Google or Facebook or you name it.
So what all of this is doing to the network is driving massive bandwidth growth. And you can quantify that in terms of petabytes per month and it just continues to grow on this exponential ramp. And since all of this stuff is data traffic. This is all data. This is not your old voice stuff. So the more users coming to the net, more content being delivered across the net, all of this requires more bandwidth. And guess what? It's all delivered over Ethernet, which is really what we do here in my group. So that's the demand side.
Now in response to the demand, these networks are changing dramatically. So let's take a look at each of these networks and see what's going on. So let's start with the service provider network, and let's look at the mobile network. Okay? So if you look at the picture a few years ago, most people were on their phones primarily doing voice traffic. And from your cell phone, your signal will go into a base station, to a cell tower, and then a base station. And from the base station your traffic will be backhauled, as they say, through the rest of the carrier network at about 1.5 to 2 megabits per second over a traditional TDM connection called a DS1 or an E1 link.
And as the carriers and service providers saw a need for more data and smartphones started arriving at the scene, what happened was they started offering data plans. And for $30 a month, you could get unlimited data. And before they knew what was going on, they had managed to clog the plumbing in their networks.
So they realized that the network that they had in place there with 2 megabit-per-second of TDM-based backhaul, is not going to cut it in terms of supporting the demand for all the smartphone users. So the first thing they did was, "Well, let's start charging for bandwidths." So now, today, for $30 a month you can get 2 gigabytes of bandwidth, right? That's all you can get. And beyond that, you've got to pay more. So they started restricting the use of the data traffic in their network because they just couldn't support it. The reason they couldn't support it was because that infrastructure that they have in place was just not geared for it.
So in battle with charging it more, they also started investing in higher data rates, so 2G to 3G to LTE, taking up the cellular data rates. And they also started investing in the backhaul portion of that network and increasing the speeds.
So if you look at the transformation that is going on right now as we speak in that service provider backhaul, mobile network, so the first thing you'll notice is macro base stations are, of course, being upgraded with 3G and of course now 4G with LTE. The backhaul is on move to Ethernet at 1 gigabit per second and going from 2 megabits per second or 1.5 megabits to 1 gigabit per second, right? You're looking at orders of magnitude increase in bandwidth. That by itself is still not sufficient. That's the first stage of the upgrade cycle.
Now as carriers roll out LTE, initially it's about just providing coverage. And you can get the coverage through macro base stations that they have, already have in place. But what's going to happen again is that's very soon, that's going to get clogged because if everybody goes to that portion [ph] on the same cell tower you're going to be -- again, this backhaul is going to be on a bottleneck, you have very limited spectrum available. And so the only way to get more capacity is to go to smaller cells. So rather than deploying these big macro base stations, they're going to continue to remain in place, but going forward, you're going to see deployment of a lot more smaller cells, or you may call them picocells, you may call them microcells and depending what people call them as. And they're going to be deployed just the way Wi-Fi access points are deployed today. And this is creating a huge opportunity for a company like us, to Broadcom.
So if you look at a macro base station today, it sells for about $20,000. And if you look at the small cell base stations, they're going to sell for maybe $1,000. They're going to request highly integrated solutions, highly integrated single logic solutions that people like us are good at building. And it represents an opportunity for us to go in and build not only base station sets but also be a complete solution provider for the backhaul. As you go to smaller cells, you're also going to need more backhaul. So all of these is creating opportunities for us, a lot more Ethernet being driven inside the service provider network and it's a global phenomenon not limited to any particular geography, and we've seen significant build-out already on the wide side of infrastructure across the world, and we've seen all this transformation of the mobile infrastructure over the next few years.
So good opportunity for us to gain traction just to give you some numbers here in terms of other opportunity. If you look at analyst data, looking at 2015, people are talking about 15 million or 16 million small cells being deployed every year. 16 million small cells, but maybe about $100 in each of those you can capture would be the size of that opportunity is. So that's the transformation happening in the service provider.
Let's talk about the data center, which is also going through its major set of transitions. So if you look at data centers a few years ago, typically most of this stuff is enterprise-based. And if you went to a typical large enterprise, they would -- data centers are consisted of, from a hardware perspective, servers, networks and storage. And these things will be deployed in what I could call application silos. So every application that a typical enterprise would run, maybe it's a database application, a little database, maybe for sharing applications, maybe it's an email application, would have its own set of dedicated servers, dedicated storage, dedicated network.
That's all changing as well. Processors are getting faster. They're going to multi-core. They're being virtualized. There's now broad spread adoption of virtualization in the enterprise. And all this stuff is changing. There's a lot more content being deployed in cloud even enterprise clouds.
So what's happening is that changing data centers to become much more larger, consolidated, very large scale virtualized and fully distributed data centers, where any application running on any server and the ability to migrate applications from servers within data centers across data centers. All of this is also creating a massive demand for bandwidth and scalability and power efficiency inside those data centers. So before most of the server connections were 1 gigabit per second. Going forward, they're all migrating to 10 gigabits per second.
So about 8 million servers are sold worldwide, over 6 million of them today are sitting still at 1 gigabit per second, the so-called mainstream rack-and-tower servers.
Now starting with this next CPU cycle from Intel, Romley, all the major server OEMs are going to be offering a 10-gigabit option on their mainstream servers. So about 6 million severs are going to make the transition over the next several years. And you could argue about what the attached rates are going to be a little. We're being a little conservative. We're thinking maybe 10% in the first year going to maybe 50% attached rate in 3-plus years.
And as that transition happens, there's massive transformation of the networks that sit behind the servers. All those networks go to 10-gigabit switching. And in fact, the deployment of the network in some case is actually done independently over the deployment of servers. And for 10-gigabit network deployment has already started happening in the data center. All of this, again, presents a huge opportunity for us.
Massive scale, you're talking about data centers with thousands of servers, and the network architectures itself within these Data Centers is changing as well, and going from more traditional hierarchical kinds of networking to a much more flatter, larger scale network.
So both of these, again, driving business our way. This is what we do from a silicon perspective provide solutions for the networking inside all these data centers.
Now when you look at the enterprise network, not as dramatic in terms of changes happening there, but still significant nonetheless. You have more users, you have more data, they all want more capacity in their networks, but your desktop is not going from 1 gigabit to 10 gigabits anytime soon. It's still going to remain at 1 gigabit.
On the other hand, what's happening is more and more deployment of wireless inside the enterprise. And most enterprise admins these days want to unify their network. They don't want to have a separate wireless network and a separate wired network. They want to bring them together. They want to have the same unified policy security across those networks because they want to provide access to users wherever they are. They also want to get better visibility of what is running inside those networks and figure out where this bandwidth is being used. Are people using the network to go look at YouTube? Or are they looking at enterprise applications? So the ability to classify applications and look at applications inside the network and police bandwidth appropriately is another important factor.
So what this means is this drives typically the collection of all of these things driving a slow but steady enterprise refresh in terms of the LAN technology that's out there. And we are well-positioned here to capture a lot of these enterprise LAN refresh targets across the majority of the customer base.
So how do we win? As you look at all these events, what’s going to create a winning strategy for Broadcom? And this is again very similar to what you already heard from Scott, Dan, Bob. How do we win? We win by having a complete solution and a leadership roadmap. We have broad IT portfolio, both depth and breadth. We're pretty strong in terms of integration and migrating across process. We're always looking to expand our footprint both in existing markets and also grow our addressable markets.
And finally the last piece for us, partnering with our tier 1 customers. So let me walk you through each of these pieces and how we win. So let's talk about the complete solution here. So today, if you look at how our business has evolved, we have gotten to the point as I said this is a very diversified, pretty complex set of networks here that you see in each of these enterprise data center service provider. And today, we are at a point where we got breadth and scale and execution. We have dedicated specific products for each place in that network and a complete product portfolio with a software SDK that's consistent from generation to generation. That provides a lot of stickiness because our customers continue to build multiple generations of products. And as they go from one generation to another generation of that product and go from our chip to our next generation chip, they can leverage and reuse a lot of their software going forward. So this provides again a lot of stickiness, a lot of barrier to entry in this market for us.
So across end-to-end solution, across really starting from the server and the data center, all the way through the network, and starting from all the individual access all the way up to the core and the service provider and the backhaul, same thing in the enterprise, we have that today.
Now we don't stop by just having a solution, we make sure that it remains leadership, maintains leadership. So if you look at some of the products, and I'm going to go through each of these products that we introduced this year, but what we've been able to do is successfully out-execute our competition in the markets that we're in.
So if you look at data center as an example today, we announced that tried [ph] back in April of 2010. And today, it's in volume production. It's got, I don't know, 10, 20 customers deploying this, starting to ramp in volume. And I'll complete that they finally came out with their solutions late this year. So that is a 18 months lead over our nearest competitor. And there's really no in-house ASIC solution to compare with that even today.
That's just one example. There are many set of examples here. If you look at the service provider market, we now have optimized chips for mobile backhaul that are being used by cell site routers by pretty much every tier 1 customer. And again, that's because we build our product that's very much optimized for that place in the network, that has all the specific feature set requirements needed for that market. So we continue to execute, and as long as we continue to execute and maintain that leadership position, we are able to anticipate the transitions that are happening and be ready for those.
Next, let's talk about the IP portfolio. So here you see us comparing our portfolio with that of some of our competitors. And as you can see, we have a number of players that are also more, what I would call in the point product category. They're pretty good at some things but not good across the board. And if you look at where we are and particularly if you combine Broadcom along with NetLogic, you can see that we have a very, very broad segment in this market. And what this allowed us to do is to pick and choose building blocks from these different products and start putting them together. We can start making individual [ph] process of course into switches. We can combine switches and microwave backhaul modems, for example.
And we do all these things. For example, as we acquired Teknovus, we acquired EPON technology with Teknovus. What we did a year after that, we announced the product that combines the Teknovus solution along with our in-house switch solutions to create an SoC. So this is just -- the fact that we have this IP breadth allows us to go selectively integrate and build the right products for each place in that network.
So looking at integration, process technology. So let me give you an example. This is a product from a couple of years ago, 65 nanometers, this is an example of somebody building a mobile backhaul router, okay, or a cell site router as they call it. This is what they will be building with. They will be building with a couple of Broadcom chips and Ethernet switch and a traffic manager and a bunch of non-Broadcom chips. I like the engineer who came up with the word BFT, but I'm not going to tell you what that is. But anyway, so you see that this is what they had to do. They had to go put together these 7 sets to create a cell site router. If you look at what we did, this year in 40, by the way, we've been shipping 40s since 2010, so it's not new to us.
This year we introduced the 56440, which is a single-chip solution for cell site routers. That integrates everything that's shown on the left. And by doing that, you're also able to address more of the building materials on the product and essentially increase -- maintain or at least increase our average selling prices.
So not only are we able to do better integration that use our power and that use our footprint, we're also able to capture more of the solution as we go through this process.
And this is something that we do across the board. And as I said this is just one example. We've done that in data center, we've done that in service provider enterprise, all the markets that we play in.
Okay? Let me talk about the footprint expansion. So I think we made a lot of progress on this, this year. If you look at last year, what I showed you last year was we had a baseline addressable market of about $3.4 billion. And that market is growing, and driven by all the things that we talked about, service providers who are driving more backhaul, service providers deploying more Ethernet, 10-gigabit transition in the data center enterprise, all this stuff is causing our market to grow at about an 8% CAGR based on market data forecast. And if you look at 2015, that puts us at close to $5 billion in terms of addressable market.
Now we are the market leader in that baseline, right? Today, we are the #1 share leader in pretty much all the segments that we play in, switches, controllers, all of them. So what we will do, of course, typical Broadcom fashion is we will continue to grow in and gain share even in the markets we play in. But what we've done this year is significantly expanded our addressable market.
How have we done that? We've done that through a couple of strategic acquisitions. And let me talk about each of those. So at the bottom, you see microwave backhaul. And microwave backhaul is a market that it's kind of a -- you would think it's a mature market by now, and in fact there's more fiber being deployed, but actually it turns out that 1/2 of this backhaul market that we talked about, forget the traffic back from the cell towers or into the rest of the carrier network, 1/2 the links are microwave links because it is difficult to get fiber to all those cell towers. And by the way, as the cell towers get smaller and it's going to be more difficult to get fiber to everywhere. And so microwave is here to stay. And that business again is transitioning from TDM to gigabit Ethernet.
And we bought this company, Provigent. They have a highly integrated system-on-chips targeting this market. And today, if you look at the top 6 equipment players which pretty much control the microwave markets between them, many of them are doing their own in-house solutions, not really ASIC solutions, but a lot of them with FPGAs and Cobol blocks put together. And Provigent is making great traction at penetrating their market with a very low share today but really getting good traction with all the tier 1 equipment players in that market. So here we have access to about a $650 million [ph] market with low share but we have a lot of opportunity to grow.
If you look at the NetLogic acquisition that we've announced, and you look at the embedded processor market, that's about a $3 billion market spread across, of course, multiple segments within the embedded processor control plane, data plane, multi-hybrid controls and data plane processor.
And again, we're starting out with a great product from NetLogic, the XLP product, which is a 40-nanometer, which is the best-in-class CPU with accelerators and the best-in-class power performance metrics. And we have the ability to drive with execution significant share gains in that market over time. So that's the second market where have again the last market that we have entered in with relatively low share with ample room for share growth.
The last piece, Scott talked about this, wireless infrastructure. That's another area where I talked about the transitions that are happening in the service provider network and wireless infrastructure, leading to more and more of these wireless small cell deployment as well as these more kinds of backhaul deployments. And that is a pretty large market opportunity for us in total. Now this is going to be a much slower market for us to capture. It's going to take us time with some organic investments combined with the pieces that we get from NetLogic. They had -- they bought this company called Optichron which does a digital front-end solution that sits quietly in the base station. Along with some in-house ASP technologies and processor technologies, along with backhaul pieces is what constitutes that market. And over time, we feel we have a very good approach towards getting into that market and getting traction.
But again, like with the infrastructure business most of these products are typically long sale cycles, long development cycles and sales cycles, but they lead to highly profitable and differentiated products over time.
So all in all, what we've done in this -- in our space this year is substantially increased our addressable market. And we've increased them in areas where we have ample room for share growth. If you take this, combined with the rest of the execution that we talked about, allows us to grow significantly better than all our markets for the next few years. That's really what we positioned ourselves to do in this business.
Okay, the last piece in terms of partnering with customers. So if you look at today, our business is fairly concentrated among the large equipment companies and we focus a lot on all of them. And you can see, this is just the who's who in the Ethernet networking space. And we're in pretty much all of them in some form or fashion to varying degrees. And we see continued ample opportunities for us to grow along with them in this space. Again, as we expand the markets that we cover, that presents more opportunities for sockets at these customers. There, we have dedicated support in place for each for these customers. And there's always this constant can we capture some of the business that many of them are still doing ASIC, and we have an opportunity to slowly but surely get additional sockets where they are potentially using -- previously using ASIC before. I'm not saying that's going to be an overnight success, but it's going to be something where we'll work our way slowly one socket at a time and get some more ASIC socket.
So if you look at that combination and this partnership, and at the same time as we do that, we're also committed to helping them differentiate with their solutions. And this is of course a concern that every one of our tier 1 customers has, whether "If I buy a chip from you Broadcom, how am I going to differentiate myself?" And our response to that is this. At the silicon level, we can customize features that we do in the customized features. You have differentiated at the silicon level. But above that, it's about how the silicon gets put together into a system and the system architecture. And if you look at that, again as an example, let me quote Trident. They got many, many customers shipping Trident. And there are some of them that are highly differentiated from an architectural perspective, highly differentiated in their solutions, all using the same underlying chips in the market. And so people are still able to take the same silicon and substantially differentiate themselves from a product perspective in their market.
And then above the system market architecture level is the software, in terms of the key differentiation, continues to be a key differentiation. And then beyond that solution [indiscernible]. So there's still ample room for our customer to differentiate as we come out with these products in the market.
So let me conclude here by the reiterating the points that we covered here. So first of all, there is strong demand growth in the infrastructure, more users, more contents, more bandwidth. All of that is driving need for more ethernet, and so the demand is there for the next several years. The secular demand is not going to go away, although there might be pauses in spending here and there, but ultimately, fundamentally, people have to invest in their network. And if you look at what the networks are doing in response to all these demands, service provider networks are moving to Ethernet and they're migrating to smaller cells, really changing what type of equipment they're going to be deploying in those network. The data centers are becoming a lot more scalable. They're becoming flat. They're becoming virtualized. And again presenting a lot of opportunities for us to differentiate silicon. Enterprise networks, unifying wired and wireless, becoming more application of that.
And our formula for success remains unchanged throughout all this. And it's consistent with what we do at Broadcom overall. And it's that set of things that's shown there, right? The complete product solution, execution and maintaining the leadership, the ability to have the broad IP portfolio and be able to integrate selective pieces of technology as appropriate for each point product in the network. And again, a continued relentless focus on expanding our footprint in the markets that we're in, and also looking at other markets to expand. And finally, our customer partnerships.
So let me stop here and open the floor for questions. Right here.
Vivek Arya - BofA Merrill Lynch, Research Division
Rajiv, Vivek Arya from BofA. My question, actually 2 questions. One is if, let's say, if I just add up the sales of the top networking OEMs, service provider and enterprise, both with Cisco [ph] and [indiscernible], let's say if they grow 6% to 8% next year, conceptually is there any reason why you should not be able to grow at that or higher base? That's one. And then the second is, I thought that you did not list the PLD [ph] guys on the competitive slide. And if they are able to move to 28-nanometer quicker versus the ASSPs, then why shouldn't they gain share versus the ASSP?
Okay. What a good question, Vivek. So the first question is yes, absolutely there's really no reason why we should be growing any slower than our main customers. We would hope to gain and grow faster than what they would grow at. The second question on the FPGA guys, we've been, I would say, competing with FPGAs for a long time in this business, and give I'll you several examples. I don't think just that going to 28 nanometers is going to change anything in terms of the competitive dynamics of the market. They'll go to 28, we'll also go to 28. Like they went to 40 we went to 40, and it didn't really change the dynamics in any way. Now what we see FPGAs being applied to in our segments of the market is end markets where volumes are relatively low or the featured content does not mature and people, customers want a lot of programs ready are where people think they want [indiscernible] they still want to differentiate and they want to write their own codes. Okay? So those are the 3 kinds of applications piece for FPGAs. And a great case in point for example is service provider Ethernet. Okay? So if you look at Ethernet and the service provider network, it's been on for a while, it's not completely new here, and the first wave of Ethernet solutions that people built were built on top of FPGAs. And that's because Ethernet standards were still emerging, initially enterprise Ethernet going to service provider. And over time, what happened was volumes of Ethernet started picking up. The standards got nailed. Now there's a whole carrier Ethernet set of standards that are differentiated from the enterprise Ethernet standards, and that's pretty much. So volumes went up, standards became nailed. There was a great opportunity for us to build multi [ph] silicon solutions. So over the last 5 years, our service provider business has grown from 0 to 40% of our overall business. And a lot of that, by the way, was in FPGAs before. And so I see the same dynamics continuing. I don't think that going to 28 is going to change anything. It's just going to be the same dynamics. They'll have a place, we'll have a place.
Question about your acquisition strategy. So if you look at in a very simplistic level, you tend to acquire technology leaders and then kind of proliferate the market. If you look at NetLogic, NetLogic is obviously a technology leader in service for search in small market. And embedded processors, at least as of what they're shipped today, they're pretty far from being a leader. Could you talk about what you're going to do and why you acquired that particular company? Because there's obviously other alternatives there.
Sure. So when you define a leadership, yes, you could define it in terms of market share leadership or product leadership. And clearly, embedded processors, is what I think you were talking about, NetLogic is not a market share leader today. Now, however, if you look at the state of the market and we look at all the products out there, these are the XLP products from NetLogic in 40 nanometers was going to be a killer product. And again, they've been sampling that product since late last year, late 2010. We've actually been that product now for multiple years, from its inception in PowerPoint slides to the time it got delivered. And today, they have -- what they've done is they've created a high-performance CPU core, that is an Intel helium-class [ph] CPU, combined that with the appropriate off road agents in a very attractive power footprint with a very compelling product that's getting a lot of design win and traction. So when we look at the potential for gaining share in the market going forward, we thought that it was pretty good, with the products that they had. And that's really why we like what we saw there and bought NetLogic. Of course, all the other businesses are pretty solid, right? So they have the DCAM [ph] business which is they are the market share leader in stretched [ph] or knowledge-based processors, and that's going through the transition from IPV4 to v6. Larger are the spaces, so bigger memory required.
And then they also have this interesting product called the Digital Front-End, which again sits right in the base station, which could be combined with a lot of other things that we bring to the table to play more in wireless infrastructure. So a very good complimentary to the products with what we had, as well a leadership product in processors.
Srini Pajjuri - CLSA Asia-Pacific Markets, Research Division
Srini Pajjuri from CLSA. Rajiv, 2 questions, first on 10-Gig, you said the attach rate this year is likely to be 10%, growing to maybe 50% in the next few years, I'm just wondering given that there's clearly a need for 10-Gig and also the technology seems to be pretty mature, why is it going to take that long for the adoption. Is it simply pricing? If so, can you compare and contrast where the 10-Gig pricing is versus Gig, that's my first question. And the second question, on the base station front, you talked about the picocells having about $100 of silicon content, how much of that $100 do you address today? And also if can you talk about the macrocells, what's the typical silicon content and how much do you address today?
Okay. So on the first question on the 10-Gig attach rates, frankly I think we're all sitting here guessing. And we certainly are being conservative in terms of what these attach rates are going to be. I don't think this is a pricing issue, necessarily. It's about -- I don't think there's going to be -- of course that still remains to be seen in terms of what the -- it's certainly not a pricing from a chip perspective, but from an OEM vendor perspective, what are HP, Dell, IBM are they going to do in terms of pricing their 10-Gig adapter solution and versus a 1-Gig solution. My suspicion is, this probably won't be a significant difference. So it could be a lot better than 10%. I'm just saying, the first year after something launches, it's going to take time. And going from 10% to 50% in 2 years, by the way, for me is pretty aggressive in my space, typically. It doesn't, transitions don't happen overnight, it takes time. And really for us, the question is, at what point does it really become the default? That 10-Gig today is a default on blades, right? That's all blade servers come with. And the question is, at what point will that become the default on racks? And certainly, that's not true in the coming cycle, but maybe in the next cycle or the cycle after that. And these cycles tend to happen roughly every 2 years. To address your second question in terms of base station content. So today if you look at these small cells, we don't address it at all. It's a hardly -- it's not a market that we're in today. We're just about to get into it, which is why it's a new addressable market for us. And the small cell market itself is very minimal today. We do have a femtocell solution today in-house, which is primarily targeted at essential markets. And we will build on top of that as well, and that's sitting in Dan's group. So the answer is, out of $100 content or so, today we do have the femtocell, which can of course be extended, and we'll probably capture a little bit of the service provider market, but bulk of it we don't have. On the macro side, clearly, the silicon content is much more. If you can imagine that a macro base station goes for about $20,000, there's probably $2,000 worth of silicon content in there. We are in a good position to address the processor content in there with NetLogic, as well as the Digital Front-End. And so, we can address that piece of it. Where we are really not there is in the DSP piece of that base station.
We're hearing about what a lot of pricing pressure in the China fiber-to-the-home market, given that there's probably 8 semiconductor companies vying for sockets there right now. So the question is, is Broadcom winning in that market? And other than price, how is Broadcom able to differentiate their solutions relative to the 7 other competitors?
You're spot on there. I think the EPON market has gotten extremely competitive in China. And as you said, there's just too many players competing for too small a market, frankly in my opinion. And where again, so therefore, playing in this market just by providing a standalone optical line terminal and a standalone ONU is not a recipe for success in our view. And so, our view is to deliver a solution that other people cannot deliver. And to do that by achieving integration on both ends of that wire, okay? So we've already done the integration on the ONU side. So if you look at these ONUs, they for the most part go into multi-dwelling units. And so, in front of the ONU, you have a switch to distribute the traffic. So we already integrated the switch on that side, with the ONU together into a single SoC, not something that our competitors can easily do. So we got that on one side. We're working on the other end, obviously, as you can imagine. So on the OLT side today, we have a standalone OLT, and you can imagine, just like the way we do things, we'll -- right in front of that is a switch. And we have the ability to integrate the switch, and by the way going forward, we have the ability to integrate processors too. So it's really seen as a classic integration play and that's really how I think we can differentiate ourselves in that crowded market. And of course, the other thing that could happen to the end market is, yes, there's a lot of competition for Gigabit EPON, and we're also trying to drive 10 Gigabit EPON. And to the extent that 10-Gig EPON gets adopted, there's going to be fewer players at least initially in that market.
Romit J. Shah - Nomura Securities Co. Ltd., Research Division
Rajiv, Romit Shah from Nomura. We're hearing that CapEx is being held back from the T-Mobile AT&T merger. Do you see that as an issue impacting your business in the next year?
I think on a broader question on service provider spending and CapEx, there clearly has been a slowdown in service provider CapEx and spending. We saw very strong first couple of quarters, and then it started slowing down in Q3, and we said in Q4 we were [indiscernible] across the board. So how long will this softness continue? It's hard to say. Can we call a bottom? I don't know. But eventually I think CapEx is going to have to come back up. Now again lets be very clear, I don't think service providers are going to increase the overall level of their CapEx dramatically, okay? But what they are going to do is to deploy that CapEx in appropriate areas in their networks. There's clearly a decline for example in SDH, the [indiscernible] equipment being deployed. A significant decline and that's being offset by decline -- increased investment in mobile infrastructure, increased investment in ethernet. So we do expect, fundamentally the demand growth is there. They're going to have to go invest, make those investments, timing of those is hard to predict.
Romit J. Shah - Nomura Securities Co. Ltd., Research Division
Can you characterize the softness as being broad-based or specific to certain a region?
I will say it's broad-based.
Romit J. Shah - Nomura Securities Co. Ltd., Research Division
Okay. And then the second question I had for you is, big networking companies like Cisco have talked about margin pressure due to shift to mid- and low-range products. How does that play out for you?
So I think that is a very unique situation for Cisco, okay? And I don't think this is a broad-based marketing issue. I think as you know, switching is one of Cisco's largest businesses, and they do have a lot of competitors in there that are coming in there and offering products at lower prices, and they had to compete at some point and that's what putting the pressure on their margins. For starters, of course, given what Cisco is going through, they're obviously trying to put more pressure on their vendors to try and get them price reductions and so forth and we're continuing to work that with them.
Any other questions? Okay, thank you very much. Thank you.
Please welcome Eric Brandt, Broadcom's Executive Vice President and Chief Financial Officer.
Eric K. Brandt
Good afternoon. It's good to be home. I'm from New York. So I learned something, by the way, Bob and I -- I learned a second thing Bob and I had in common. We're both from New York. I have seen the Brooklyn Bridge, he hadn't as a child. That's because I was born in Brooklyn, but moved to Staten Island. The second thing we had in common is that our bridge is never shown, the Verrazano Bridge. Probably because where it goes, is never shown in pictures of New York either, by the way.
Anyway, what I want to talk to you about today is a similar thing to what I've done in prior years. And so, what I try to do when I come to Analyst Days is give you some perspective on how we as a management team think about our business, how we make choices and how we run the business by showing you a little bit of some of the analytics and some of the things that we review as a company. And in prior years, we've looked -- I talked about shareholder value and shareholder friendliness and portfolio management. I talked about R&D intensity. And the growing R&D intensity and the changing competitive dynamic o the industry, which at the time probably wasn't very popular, but I think today everybody agrees that that's in fact happening and that's what's causing the industry to begin to separate in terms of the people who can deliver products and grow through the changes in technology and cost, and those who can't. And I've also talked about sort of the volatility in managing our business through volatility. Today, what I'd like to talk about is strategic deployment of cash and how we think about deploying our resources as a company to drive value and support our strategy. Now what I hope to do also is stitch together some of the pieces you've heard from my various colleagues today about the way our business operates and try to draw together the scene economically as to why Broadcom succeeds the why the Broadcom does, and why we create the value and perform the way we do over time.
So let me start by talking about driving shareholder value. Now we think about this all the time. We do pay attention certainly to the stock price. We also pay attention to being a good company. But at the heart of thinking about shareholder value is sort of this three-step progression. First is, you need a strategy which drives value. And I think we do, and one that is time-tested. Second, you need to deploy your capital in service of that strategy and try to find a way to maximize the deployment of capital in service of that strategy to drive advantage. And then third, have the operational excellence and tools to make sure that you execute against that strategy on a repeatable basis, such that the company continues to perform.
So before I talk about those 3 pieces, I want to talk a little bit about shareholder value and how we think about shareholder value. And obviously, it's the performance the stock over time, but what drives shareholder value? And I've had this conversation with a number of investors over time, and I don't think there's a single person in this room who would disagree with me that says, the faster your growth rate and the higher your gross margin, the higher the valuation of a company. And so, if you look here at a chart, right? And you just simply look at gross margin multiplied by growth rate, right? You see a relatively consistent correlation and this could be any industry. It turns out that this is the semiconductor industry, and this is a chart we looked at while we were looking at the NetLogic transaction. And this chart applies in the industry certainly in periods of time where you don't have extraordinary volatility. Where people don't know what earnings are and growth rates are, et cetera. But what's particularly interesting about this chart when you look at it, is how tight it is. There are companies that have very high dividend yields, there are companies that have no dividend yield. There are companies that are in the restructuring phases, there are companies whose growth rates are distorted for one thing or another, but at the end of the day, when you look at this chart across multiple periods of time, and take out sort of periods of high volatility, like we experienced -- we're experiencing now, or as we exited '08 going into '09. The correlation of this chart can be as high as 70%. Typically runs about 50% to 55%, up to about 70%. So as a company, when we think about our business, we think about profitable growth. What does profitable growth means to us? It means growing the top line and at least maintaining the gross margin and preferably expanding the gross margin. Now I'll talk about why that matters in a moment.
So let's talk a little bit about building a value-creating strategy or value-maximizing strategy. And so, let me start by saying that at Broadcom and for those of you who've listen to us talk here today and in numerous investor meetings across multiple years, it's the same drum, right? We have a particular strategy. We are fortunate that our strategy has worked literally almost from the inception of the company all the way through to today, and I would argue when I get to the end of my presentation, it may even be accelerating. And so, let me start by talking about it and you sit here and listen to us talk about this thing IP, right? We talk about intellectual property. Who cares? Why do we talk about intellectual property? Why does it matter? It matters because the accumulation of intellectual property and the growth of intellectual property is at the heart of this business and the heart of what is driving technology going forward, such that we can meet the demand that Scott was talking about as communications continue to grow. And so, at the very heart of our strategy, the first step of our strategy is to foster a culture of innovation. This is a company where 76% of our employees are in R&D. We are an engineering company. We are a technical company. That's what we do. And then, we build or accumulate IP everywhere we go, okay? So we accumulate these various pieces of technology, not really sure how they apply economically, but I'll come back to that in a minute.
Second, we talked about integration. We talk about SoCs. We talk about integration. We talk about taking that IP and leveraging it into SoCs. We talk about executing at a level which drives new chips. Why? Because that changes the fundamental economics to our customers and sustains the economics to Broadcom. Because by integrating this technology and accumulating this IP, what we are able to do is at a much lower cost point for the customer. Deliver a product which is more integrated, lower R&D intensity for them, and it supports our gross margin because the alternative of a chipset, which costs more, which challenges the economics of our competitors. And whether you saw that in broadband, which is the start of Broadcom and in all of the chips that Dan showed, and the way Dan's business continues to separate itself from competition, both technologically and from a market perspective or Rajiv's business or on the combo chip or as we try to show you from the SoCs that Bob discussed on the cellular side, it is the same process. Changing the fundamental economics and performance through integration of intellectual property into products. And as intellectual property becomes more scarce and the ability to do individual chips becomes more difficult in a product because of the cost structure to the customer, the more the advantage plays to Broadcom.
The third step is to leverage that and leverage those economics into a way to outgrow the market. So we profitably try to bring gain market share. We try to find ways to enhance our product position, shrink die sizes, tailor product to the market, strip in IP, add IP, take IP out, et cetera, so that we are tailored and we offer a product set which is unique and capable to the market and optimizes the economics both for our customer and us. In addition, we are constantly on the search for new IP. We're looking for new intellectual property which will expand our markets. And so, as we expand our markets and grow our market, we not only gain market share and grow with our market, we create more opportunity for the company to expand its footprint. And so, for a strategy to work, it needs to create a virtuous cycle. And if you look at this strategy, and it is the same strategy, and we say this over and over again, I'm sorry if you're tired of hearing about it, okay? Because it is the one drum we have. It is a drum that drives advantage.
Let's talk about deploying cash. So as a company, the way we think about it, the next step is, if you have a strategy, the question is, are you deploying cash in the right places and are you doing the right thing? So I've taken the liberty of doing a slightly different cut of this for the purpose of increasing transparency to how we think about it. And so, what I've done is on the bottom, I've put together what I will call the operational components of the business. That includes the SG&A component and the overhead component of our operations, or COGS component of our P&L, and I've taken out the purchase of wafers and packaging, that is because that's not where we necessarily spend our money, because that money has to go there to make the product. It's how do we allocate our money. So that's what's in that bottom category and that's about 17% of our spend over the last 5 years. In the middle category is what I will call investing for growth and that's the out of money over the last 5 years that we have spent on R&D, and to be fair I've included stock-based compensation, so it's not just cash, it's actually the equity we grant it. As well as the M&A we've done. Because the M&A we do is really R&D, right? Somebody else's R&D that we spent balance sheet dollars to bring into the company, to bring intellectual property and technology into the company. So in essence, it's really an R&D spend. And so, that's about 63% of the spend. And then, what's left over the last 5 years, about 20% of that has returned to shareholders. And so, if he asked us how we think about it, the bottom part of this chart, right, is optimize and minimize, the middle part of the chart is maximize within the return criteria and return objective, and both in terms of ROI or whatever the return metric you happen to use, cash flow, return on investment, et cetera, and growing margin and the growth rate of the company, and what remains is returned to the shareholders. So let's look a little bit deeper into each of these, so you'll understand how we think about this.
So starting with the bottom, and sort of what I'll call the operational component, I've put 2 pieces here. And on the top, I've put the gross margin of the company on a product basis, and below I've actually put the SG&A component. Now why is that important? Well, first of all, I started at the point that value is driven by gross margin and growth. And so, if you're in a business where your gross margin is declining, but your growth rate is continuing, you have to grow that much faster to sustain your value, right? And the reason for that is very simply, right? Is that if a company is running a business and it's growing and it doesn't have to sacrifice price and margin to grow, it's a healthy business. If you are trading price and margin to grow your business, and you guys all know this, right? The business potential is at risk and may not be sustainable. And so, what's interesting about this chart is that over the last 5 years, our gross margin is relatively rock solid, okay? Sort of the one period in that 2009 time period where the growth rate dropped meaningfully and you had some overhead issues, the gross margin is relatively constant about 51%. What's even more interesting is that the consumer component of our company, which is the yellow line on top of that, has grown from about 25% to 30% of our business to 50% of our business. So what's embedded in there is the fact that we have improved the margin of our products, we have integrated new technology, we have driven down process nodes and found a way to sustain our gross margin in the face of an increasing mix of slightly lower gross margin business overall, which is not an easy thing to do for a company and something that we're quite proud of and enables us to continue the level of investment in R&D to drive that top line growth.
Looking at the bottom piece. SG&A. Our SG&A on a GAAP basis has dropped from about 13% to 9%, so about a 30%, 35% drop. Now that's scalable. That is scale. That is the growth of the company that's driving scale is efficient operations and we intend to continue to do that, and I suspect that number continue to drift down over time, excuse me, as the company grows. Why does that matter? Because if you go back to the what I said a couple of years ago about the R&D intensity and the competitive barriers in this industry, if you don't have points on your P&L, you can feed back into that growth engine, you will fall behind. And the reason why Broadcom doesn't fall behind is the product of these 2 things: our ability sustain our gross margin and the ability to drive leverage back into the R&D model.
Let's talk about driving growth. So over the time period, Scott talked about how large we are in R&D investment. I think from an R&D reinvestment perspective, excluding I think 2 biotech companies, we're #3 in the Fortune 500 in the terms of R&D reinvestment rate. And that's because we're trying to do that to drive growth. And the proof is in the pudding on whether you actually drive growth and deliver the margins, which I'll talk about as I get to the end. But if you look at our R&D reinvestment rate, again, including stock-based compensation, we have grown our R&D over the last 5 years by about 55%. If you take out stock-based compensation, which has been coming down, it’s actually closer to 65%. So we continue to put as much investment as we can, as makes sense to drive the growth of our business. In addition, I add the M&A and I included the NetLogic acquisition on the bottom, but you can see, we continue to invest in external R&D. Now we've done one large transaction in the case of NetLogic, but you can see this number sort of runs $200 million to $600 million, probably another couple of $100 million under NetLogic in there as well. But as a company, we are trying to both leverage our own R&D and external R&D to accumulate IP and add it to our portfolio and drive value as a company.
Let's talk about the last piece, shareholder return. So over the last 5 years, about 20% of this cash analysis, we have returned to our shareholders. The vast majority of that in share repurchases, about $4.2 billion total over the time period. And last year, we initiated a dividend. And this year we grew the dividend, and we anticipate we'll continue with the dividend. Our objective is to continue to grow the dividend. But the way we think about it is, optimize and minimize at the bottom, maximize the middle to the return requirement and return the remainder to our shareholders, okay? And if you look at our balance sheet today and you take out the money for NetLogic and you take out the debt, we have returned most of the remaining cash to our shareholders, either in the form of share repurchases or dividend. Now the question is, does this work? Does this add value? So let's talk about whether this adds value.
So if you look at our market in 2006, 5 years ago, we probably had about a $25 billion SAM, and we had about 15% market share in that SAM. Now fortunately we're in markets that grow. They grow typically faster than the broader semiconductor market, but we added additional markets by making certain acquisitions, and I'll talk more about them in a moment. And so, if you look today, we reckon our SAM is probably close to $44 billion. Not only have we ridden the growth of our SAM between 2006 and 2011, we've actually grown our market share. We're now up to a point, and what that's done is that has doubled the size of the company. If you take industry estimates and you plug in the acquisitions that we've already done and have yet to close, principally NetLogic, and you look at those markets. In the next 5 years, that market should grow to about $76 billion. And so, just if we sustain our market share, which as you know our objective is never to sustain our market share, it's always to grow our market share, and we pay ourselves to do so. There's significant growth opportunity by actually making the investments we've made organically and the acquisitions we've made.
Let me talk a little bit about implementing the strategy and trying to implement the strategy in a way which has operational excellence and works. And so, I showed this chart, probably the first time I stood in front of this group, and this is our portfolio grid and for those who aren't familiar with it, every year we go throughout a portfolio analysis. We look at our business. We actually even look at a level deeper than this. This is at the line of business level. We actually look at the platform level, so we have a chart that has 31 of this on it. And we assess where we are in our business in terms of our investments because we are very committed to managing our business within our long term operating model. Despite the fact that this was a lower growth year than everybody thought, okay, we were able to manage the business within the operating model that we committed to, at between 20% and 22%. And so, we look at this, we going into the year and in periods of time where things are tighter, like 2011 is, and certainly we don't know where '12 looks like, we make tough choices. And somebody asked me over the course of this year, "You guys really use this?" Yes we do. "You actually make decisions." We do make decisions. We're not slavishly wed to this, but I assure you that those things at the bottom become very important discussions around allocation of the company's most precious asset, which is R&D dollars, to the right places in the company. And this year, we had to make a difficult decision. We exited out digital TV and Blu-ray business because we felt that we had much better investment opportunities in the other parts of the grid. There was about 400 people involved in that, about 100 people which were connected to other parts of the broadband business, stayed in broadband, and about 300 people which were essentially displaced and those resources, those dollars, were reallocated to the top part of that grid, which is what you'd expect us to do because we can't spend on everything, nobody can spend on everything, we have to make tough choices.
The second question I've been asked about this, this year and I thought would be interesting to put up there, is people say, "Well, gee, do you put your M&A on that grid?" You'll actually see how M&A played out on this grid. And for a lot of the M&A, we do it because it's relatively small in R&D, and frequently the return horizon is longer than what our 3-year planning horizon is, we don't put it on there. But I thought it'd be interesting to show you that when you plug NetLogic, the NetLogic acquisition, why it was so attractive to us, you can see that as a business, it is right up in that top quartile of the portfolio grid. And so, we do think about both our internal capital and our balance sheet capital, our P&L capital and our balance sheet capital in terms of how we use it. So that's a look at the P&L. That's a look at the portfolio. It's a look at the choices we make in terms of implementing. Let's talk a little bit about M&A.
So what I put on this chart is I've given a picture of recent M&A the company has done. And I want to make a really important point, which is, it is right at the core of the strategy that I outlined at the beginning. So at the very, very start of this, the first bubble, right, is accumulating intellectual property, valuable, high-class, leading intellectual property. Every one of those deals, the first question we ask. The second question we ask is, what's the growth look like? Does it accelerate the growth of the company? Is it accretive to the growth of the company? Not accretive to EPS, because anybody can do a deal for cash today and make it be accretive. Is it accretive to the growth rate of the company? Will it expand the value of the company? And the third thing we ask is, what's the margin profile? And is it accretive to margin profile of the company? You can see 3 of the deals on that page actually adds value to the overall gross margin of the company, and hence we believe are value creators for the company. Now we go through all the other analysis in terms of return on investment and cash flow evaluation, et cetera. We run risk-adjusted models to try to figure out how wrong we can be. But at the heart of our M&A strategy are these 3 questions. Consistent with shareholder value at the very start of that strategy wheel that I showed you.
So where does that leave us? If you look at where we're going, right? Where we sit today is not terribly different than where we probably all sat 3 years ago at the end of 2008 going into 2009, I hope it's a whole lot better than that. But fundamentally, there's a high degree of uncertainty. And so, what our financial formula for success is, we will press our advantage in the face of uncertainty. Now we will do it in a very financially responsible way, and as you can see, we had to take the tough decision to exit our digital TV and Blu-ray business to put resources into other parts of our business, like mobile and wireless, like some of acquisitions, and we're going to make very focused investments as Scott mentioned in those 2 areas, and pretty much the rest of the company will be run tight. And by I mean by tight is essentially no new investment, just the rollover investment until we are comfortable that the economy is improving, the growth rate is improving, and that we're in a position to make larger investments. From a cash perspective, the strategy is -- as I started and doesn't change, right? Priority one is to begin to grown the dividend, priority two is to look for additional IP that we can buy and once the cash values begin to -- cash balances begin to pick up, we'll continue with our share repurchase. And to the extent we don't have other opportunities to grow our business in a value-creating way, we will return to capital to our shareholders, which is what you'd expect us to do.
So let me put this all together. Proof's in the pudding. And you guys have seen this chart before, and I want make some different points than we've made in the past about it, which is, here's a picture of the growth rate of the industry over the last 10 years. And you'll see that overall semis grew at about 9%, the communications component of semis probably grew by 40%, 45% faster than that, and so, because we're an attractive market which is good, right? So we benefit from that. How did Broadcom do? Well, Broadcom grew about twice the rate of the communications semi market. So I mentioned that between '06 and '11 we had doubled -- actually between '07 and '11 we've doubled. And you can see over time how Broadcom has gone from about $1 billion in 2002, north of $7 billion today. And to come back to the point around shareholder value that was asked by somebody in the back corner of the room, I think earlier today of Scott, we think about being a good company, we believe being a good company that is responsible to its shareholders makes it a good stock as well over the long term. We don't believe in trying to manage the stock price per se. We believe in managing our business. And so, case in point, between 2007 and 2011, our revenue has roughly doubled. Our EPS on a non-GAAP basis has gone up 2.5x. Our EPS on a GAAP basis, because of our commitment to stock based compensation, excluding one-time items, is up almost 5x. So as a company from a financial commitment perspective and performance perspective, we believe we're doing what we should do, which is, we are beating our competition, we are returning profits to the bottom line, we are reinvesting in our business and we are extending our competitive advantage. And so, we believe we are being shareholder friendly and cannot control the market and sentiment of the stock, and for those of you who know, right, every day somebody is going to kill Broadcom and that's why I have to sell the stock, right? And so, so let's talk about the competitive side. So in 2002, and talk about extending our competitive advantage and track record of competitive advantage. And so, here's the redline. In 2002, if you take out the memory players in the semiconductor market, we were #19. By 2007, we had moved up to #11. And in 2010, and I suspect it won't be terribly different in 2011 since the industry grew at 2%, and with our guidance we're in the range of 8%, which is 4x the growth of industry, I suspect this will be similar. You can see that we have moved up to #6 on the chart. So from #19 to #6 over a 10-year period. Why is that? It all goes back to all of the technical discussion and chip discussion, and Broadcom-is-everywhere discussion you heard earlier today, which is the competitive barriers in this industry are increasing. The ability to produce chips, which change the fundamental market and change the fundamental economics, is becoming more difficult. And the chasm between the larger and successful companies who can do this and the smaller companies who would like to do this is widening. And you've heard us say that many of these companies, while they are fighting to do this are not profitable, yet we are profitable across of all of our businesses. And so, as a company, we are acutely focused on the competitive dynamic. We are acutely focused on every one of the examples that you guys raised about this company is going to kill us and that company is going to kill us and trust me, this is about as competitive a group of management as you've ever met in your life, okay? And 5 years ago, the companies that we all talked about in investor meetings are not the same companies we talk about today. Many of those companies are either not on the list at all or have dropped down. Some of them are still above us, but don't compete with us anymore. And today, we have a lot of conversation and I heard it today again, about Qualcomm and Intel. So I will mention competitor names, whereas everybody else said, competitor A, B, C, D, et cetera, okay? I will tell you, we are flattered to be mentioned in the same sentence as those 2 terribly successful companies. Are we anywhere near where they are? Not really, okay? We have a ways to go, but you can bet that we have our eyes set on it, and we have a model and a recipe that we believe that we're going to use which worked in broadband, worked in switching, worked in connectivity, and we anticipate will work in wireless infrastructure and will in cellular to drive advantage. So why is the future brighter than the past for Broadcom? Because as a company, we are truly extending our competitive advantage, okay? Every year somebody is worried about somebody eating Broadcom's lunch and every year, kind of like a meat grinder, we continue to outgrow the market. And we climb. Why do we climb? We climb on the back of that IP and that R&D investment, and we climb on the back of the financial management and the reinvestment approach that we take as a company to run this business as a business. Thank you.
I'm going to invite Scott up for questions to join me.
Eric K. Brandt
Shawn. Yes, Shawn?
Shawn R. Webster - Macquarie Research
Now can I ask you a Q4 question?
Eric K. Brandt
Sure go ahead.
Shawn R. Webster - Macquarie Research
So Q4, I was wondering if you could maybe tell us about the moving parts for the quarter, what was better than expected for your quarter, I recognize this, like $50 million better, but just kind of curious as to what went a little bit better than expected? Are you prepared to share any -- your outlook on OpEx for next year? And a more quantitatively way, like how much percent range do you think...
Eric K. Brandt
Sure. So let me deal with the first one. So in terms of Q4, I think we took a pretty conservative stance coming into the quarter which you would expect. We didn't bite on the belief that it really was a bottom, we just didn't know. And so broadly speaking, I think our business performed better than we thought coming into guidance. I don't think there's a specific element that I can point to. There are various points and pieces in various places that performed better, which we were pleased with. In terms of next year, in terms of OpEx, so you heard me say on the call that I said that the accounting base charges, fringe and merit and stock-based comp last year was about $45 million -- $40 million, $45 million. And I thought that if you applied the growth rate of our employee base, which is up 8% or 9% in 2011, and the merit increase that you'd probably roll that number. It looked closer to $50-ish million in Q1 just as a step up from an accounting basis. I think in Q4 some of the favorably is related to hiring ramp and timing, so I suspect that will roll into Q1, as well as some credit. So that's probably about $10 million -- $5 million to $10 million of spending that probably would have occurred in Q4 that probably got pushed out to Q1. So if I start, it's probably in the range of that $50-plus million, that $10 million, okay, plus whatever incremental there is. Now for the rest of the year, because I said we are going to manage tightly, unless the market turns around, I suspect that the OpEx progression next year will be flat to slightly down across the remainder of the quarters, subject to pushing of expenses across that time period...
Eric K. Brandt
Operating margin because anybody can squeeze operating margin per year, the real question is gross margin, can you drive gross margin. And so, I believe that as our competitive position increases and the IP and the value we add to the customers in the products we produce expands, like other people, and there are multiply examples that we can talk offline, but you know what I'm talking about, as that expands, I expect our operating margin would expand as well, but I would do it off the back of gross margin, not off the back of squeezing R&D or trying to squeeze SG&A harder, because I would much rather continue to outgrow the market, take advantage of that 70% to 75% growth rate over the next 5 years, and expand our market share. Other questions?
Christopher Caso - Susquehanna Financial Group, LLLP, Research Division
Chris Caso from Susquehanna. Given what you've seen so far this quarter with things being a little bit better than you thought, does that give you any more confidence going into next year. Obviously, customers are burning a lot of inventory, they're going into a correction right now. Anything you could say about where we are with the inventory correction and kind of the tone from your customers right now?
Eric K. Brandt
I think you can see from our guidance what we thought. I think, we think things were a little bit better in Q4. Hard to make a call in 2012 at this point. But I think as a company, I hope you guys can appreciate that we try to be conservative and give you a set of numbers we believe we can hit or hopefully do better than. And so, it's early to make a call on '12, I think the growth rate assumptions on '12 are all over the map. They're anywhere between negative a couple of points to positive 5 points, and I think that triggers a vastly different view of the market. So I just know at this point, if you have any...
Scott A. McGregor
Yes, a couple other observations. I would say customer inventory levels are pretty healthy, pretty lean right now. You've seen our own turns, quite healthy. So I believe any uptick that comes is going to put some strains on the operational system just to deliver the products. We already see customers who have ordered things in short periods of time, and we have to scramble quite a bit to satisfy those. I think also in terms of growth next year, it's pretty hard to forecast the macro side, but one thing you can be sure of is that we will goal ourselves on whether we outgrow the overall semiconductor industry and our peers. So we're really focused on that and we goal the management team on that, we goal the whole company on that, and we plan to outgrow the competition.
Eric K. Brandt
And if we don't, we don't get paid.
Just wanted to follow-up a little bit more on margins, gross and operating margins in cellular and connectivity. I'm thinking about the slide earlier, Bob showed with the 8 competitors, acknowledging that many of them are not making a lot of money. And you've been looking at the top guys in the industry, like Qualcomm, being very aggressive, maybe driving gross margins down a little bit, and other operating margins down a little bit. Intel, investing aggressively. I don't see a lot of, a little bit of rationalization, but not a lot of exits in the business. I'm trying to understand how this plays out going forward? It seems like the scenario where margins could get worse at some struggling company before it gets better?
Eric K. Brandt
I'll kick it over to Scott in a minute. I think there's always that risk, but it's not -- that risk is not really a risk of Qualcomm and Intel. It's more a risk of a smaller competitor that's trying to be irrational. And I think we've lived through that, certainly in broadband and managed it. You can see today, right, that many of those competitors, their strategies of being relatively irrational in trying to find a way to vector in on individual point solutions without the full platform, didn't work. I think we suspect the same thing will apply in the handset and wireless space.
Scott A. McGregor
Yes, a couple of other observations. The switching cost in the cellular industry of changing to a different baseband supplier is extremely high. Much more so in many of our others spaces. So the ability for a small player or two to woo a large handset customer in exchange for a couple of dollars off a baseband chip just aren't there. It's much work for them to go do it. They're looking for people who can supply long term. Another observation is that gross margin is a function of your competitiveness. And I hope you gathered from Bob's presentation that the competitiveness of our cellular SoC offering is going up dramatically. I think that the chips that Bob talked about are going to kick ass in the market. And I think that enables us to actually have some tailwind in terms of being able to push margin in that space, with just much more competitive products against where we might have been a couple of years ago struggling in the 2.5G market, stuff like that, which is much more commoditized. I think we're getting in the rare, rarer atmosphere as Bob's chart pointed out, there are very few competitors who can do the kind of chips that we showed you today.
Just a couple of questions. Eric, either for Eric or Scott. So if you look at that beautiful chart that you showed where your growing faster and faster than the market, there's actually -- every couple of years you kind of flatten out, then you grow faster and flatten out, grow faster. One question I have is that I think the reason you grow faster is because you keep investing R&D during those periods where you're not necessarily growing revenues that much. Now I'm not sure I understand properly, but you're talking about trying for the 20%, 22% operating margin model, yet you're obviously going to continue to invest. If you get one of those transitions, where get a situation where the choice is, maintain the model or keep doing R&D, what would be -- what should we be looking for in your behavior? And I have one more.
Eric K. Brandt
Sure. Look, I think we said this in 2009, and we haven't changed our view, which is, if we're in a down market, we will manage the business and cash flow. We will tighten our spending and if need be, if we believe our cash flow is threatened, we will cut, okay? I think as long as we believe that the market isn't headed to a significant downturn, and the threshold of our cash flow isn't threatened, and I want to give you an example. So in -- at the end of 2008, I stood up here and I told you guys that we were targeting $300 million cash flow from operations in this downturn, okay? Just to give you a sense of how much healthier Broadcom is today, we will probably do north of $1.7 billion -- $1.75 billion in cash flow from operations just a short 3 years later than 2009. And so, that's the first thing. I think the -- and in a growth market, the hardest one to figure out is that oscillation around the media. And even though we are flat at times, we actually outgrew the market in those periods, typically. Any in fact, 15 of the last 16 years we've done it. And so, it's always a hard guess because we make investment choices over the current year and half of our R&D growth in the following year is a result of the wraparound from the prior year. And so, as we go into '12, we made the point that we made a cut, we're going to reallocate dollars to another part of our business, and we'll make some very focused investments tightly for next year. And then we're going to sit and watch, okay? And so, I can't tell you, if next year is a down year for the industry and we make these investments, would our operating margins go down? Yes, they probably would. But I don't think we're going to make meaningful investments until we're reasonably comfortable that we can sustain ourselves within that range.
Scott A. McGregor
Yes, exactly. Building a high quality engineering team takes time. You can cut it instantly, but it takes a long time to build it there. So you want to make sure that you don't do a knee jerk reaction for something that's a quarter or 2, because it would take you a long time to rebuild that R&D team. So I think exactly we will model, we will manage the model. If we see a prolonged downturn, we would certainly look at possibly cutting. If it's a quarter or 2 blip, we wouldn't cut.
Maybe this is related to the previous answer. Companies change its sort of financing or capital structure quite a bit, most tech companies didn't use to carry any debt. Obviously, Broadcom is not raising debt, you're paying dividends. After the NetLogic acquisition, you're probably close to maybe no net cash or slight net cash. A big change for the company. How are you thinking about how much cash you need to have? How do you manage your debt? Is there a ratio you look for?
Eric K. Brandt
So look, we're not fans of debt. Fortunately for us and for everybody else right, cash was on sale, so we took advantage of it. And we'd much rather use cash to make an acquisition like NetLogic than use our stock. The P/E of the cash by the way, if you think about it, is really one over the interest rate, which is like 200. So it's almost like, using the go-go days, that's what's happened with debt. We're not fans of debt, we took advantage of the opportunity as it existed, and we're not particularly excited about increasing our debt level unless there happens to be something of very significant value to us. From a cash perspective, we try to run our business with about half of our operating expenses in cash, which gives us sufficient flexibility for any shocks to the system to manage the business without any real challenges and the ability to react.
Eric, I just wanted to follow-up a little bit on that discussion about allocating resources, et cetera in down market, up market et cetera. Can you talk about, so guessing your R&D investment strategy, would you say that you've changed the regularity at which you portfolio reviews now using Blu-ray TV as an example. And is that something that we should expect? How do you look at the metrics, impairment of cash flows, et cetera, on a more regular basis?
Eric K. Brandt
Well, I simplified it, but the answer is not really. We still run an annual process. We watch the business in the current year relative to its objective. And we get about halfway through the year of the next year's projection. If we believe a business is falling substantially short of its objective. We will do a full business review and assess that business out of cycle. And there are numbers of cases of that where we will adjust investment. And so, it's interesting, I used the example of digital TV and I think Ross asked Bob what's different, right? Bob said, smartphones, smartphones, smartphones, which partially because Bob's been trained not to talk too much about this, but I'll answer the question right. There has been a variety of changes inside of the business that each of these executive vice presidents have done to themselves before we have done it to them at a corporate level. And so, what has happened in mobile and wireless is, in fact, Bob has discontinued the feature phone roadmap, has taken those resources off and reallocated them to other parts of the business. And so, we have those discussions on a regular basis, frequently they're managed by the general manager that runs the business, and we still run our corporate cycle because at the end of the day, we do have to run a corporate cycle and make larger decisions, particularly in periods of time of uncertainty as we face right now.
Scott A. McGregor
Stopping a line of businesses is rare, both historically and I expect will be rare going forward. So you won't see too much of that, and then to maybe address another question, we got a question, if that grid up there, there was like another one that was sort of red all at the bottom and would we look to that kill that one too? And I would say, well no, because it's tied with some of our other businesses, and so actually has a good return just outside of the scope of the time we looked at that one. But also I'd observe that if you look at the Olympic track team, even the slowest guy on the track team is a pretty fast runner. So that's a pretty good set of businesses we have.
Looks like your SAM growth on that slide was kind of 12% compounded over the next 5 years. And should we think of that as kind of a baseline level of growth and if you could gain share, should do better?
Eric K. Brandt
Given that, that data came from an external source, you can feel free to use that. We don't know. Do you know? I don't know. I mean, look, Bernanke doesn't even know.
Historically, you showed historical slides and how accurate I guess your SAM has been and you've gained share.
Eric K. Brandt
Yes, but you can see that it flows, right? So what's happened is, right, the industry has grown, communication's about 13%, this would say 12%. It might be a little slower than that because the denominator of the industry is getting fairly large. But I think that's a good starting point for a view over the long term.
Scott A. McGregor
With each new company we buy and each new market we enter, it increases that SAM dynamically too, and that's going to give you an acceleration on that.
Eric, a couple of questions. Where are we in the LTE R&D cycle? Can you help me understand sort of the size of that spend versus 2.5, 3G, and I guess given that you now actually have baseband revenue, is the LTE effort being self-funded within that business? And I have a follow-up.
Scott A. McGregor
So the LTE spend, I don't expect that we will dramatically increase it after the next quarter or so. We will increase the spending on that as part of the focused investment we've talked about. In terms of, does it pay for itself? Well, no, we don't have any LTE revenue yet at all. We tend to look at the whole wireless business as a whole and not break out for example LTE in particular as we see more and more SoCs, we are not going to sell an LTE chip, we're going to sell a cellular SoC that has a variety of components in it.
[indiscernible] Does this fund, the baseband revenues fund the efforts in LTE?
Scott A. McGregor
It certainly contributes to it. I don't know off the top of my head know exactly, again, we don't look at it that way, we look at it as a wireless business, a handset platform and do we make money on the platform? And that's...
Eric K. Brandt
We do move resources between wireless and cellular and apps processors in a fairly regular basis.
And then, as my follow-up on the apps processor side. A lot of other companies have put singular focus there. Trying to bring an apps processor to market for a whole bunch of functional applications. You guys seem to have focused in on connectivity as high value, kind of baseband SoC as high value, but kind of shied away from a standalone apps processor, has that strategy changed and if not, why not?
Scott A. McGregor
We would certainly do a standalone apps processor if a customer asked for it. I think you can speak, it's easier to do a standalone apps processor than to do an integrated one. We think the value is higher doing an integrated one. It's power savings. It's board space savings. You get dramatic performance and proven so by integrating with the baseband, this is for high traffic between those two. So I think from a differentiation point of view, and a value creation point of view, we're better off doing integrated apps processor. A lot of the guys who do standalone apps processors can't do an integrated one. If you can't do an integrated one, then you're going to celebrate the standalone one. We think the integrated one is better and again, it would be easy to spin one out, standalone if that made sense, we'd do it going forward, but that's not our focus.
Eric K. Brandt
Maybe 2 more questions.
Eric, you've talked in the past about taking balance sheet dollars and using that to drive earnings per share and you guys have done a couple of big deals this year, NetLogic, Provigent, can you frame for us EPS accretion from these acquisitions for 2012?
Eric K. Brandt
I believe we said in conjunction with the deal, I think we said at least $0.10, it really depends on the timing of the closing of the deal, but these deals are accretive. Accretive, again, for any company that has a profit, when you're using cash is easy. I think the real question is, have you created value over the long term by hitting your return metrics and the long-term growth rate and margin of the business? But ballpark, I think that's what we said, I don't know the exact number. Chris, do you remember the exact number? And it will grow meaningfully in the second year from that because there is about a 6-month window, probably 3 to 6 months on IT and G&A where we will have to manage with overlap systems and from an operations standpoint, potentially 9 to 12 months. So we sort of get everything aligned with our systems. And then once you wrap that around, the synergy savings in those areas, this is not about R&D synergy, those synergy savings are actually meaningful and we plan to increase that number.
Craig Berger - FBR Capital Markets & Co., Research Division
Craig Berger, FBR. Scott, can you help us understand in baseband, maybe what the impact of declining 2G exposure is in 2012 relative to the uptick of new 3G customers. And based on the design wins that you guys have in hand right now, when might we expect to see baseband be a 10% segment of your company's sub segments? If you can't answer that, when could you give us a more objective for that business?
Scott A. McGregor
So couple of things. On the wind down of 2G, the truth is, we don't know. It'll depend on how the models sell at our largest customers. How rapidly they move to other things, their own market shares. So there are a lot of dynamics on that we just don't control. What I can say though is that we expect that our 3G business will dramatically ramp next year. And in smartphones and other products, that's where our emphasis is. Think of that as our continuing business going forward in that space. So we'll have to see how that dynamic plays out. Could be stronger than we're forecasting, hard to say. But that was a difficult one for us to even get a handle on there. In terms of when will basebands get to be 10%? I don't care. I care when cellular SoCs become a really important business. Again, don't focus on basebands, basebands are chips. 5, 10 years from now those won't exist. They're going to be cellular SoCs because the 2 leading guys in this space are pushing that agenda for the reason that it delivers great value to customers. I think all these other guys who are pushing sort of these separate chips, they can't do these integrated chips, that's why they're pushing with separate chips. So I think we've got a great structure going forward there.
So with that, thank you very much for coming today. Before I wrap up, a couple of things. One is, please, please fill out your feedback form. This is our first time in New York, we did our agenda a little differently today. Slightly different lineup of speakers, we'd really like some feedback on that. And just for motivation, there's a giveaway, if you do this and I'll give you a hint. The giveaway contains Broadcom Bluetooth, Broadcom wireless LAN, Broadcom application processors, and Broadcom video processors. So it's a kind of an interesting thing you might want to check that out.
So with that, let me summarize. If I leave you with one point here. We've shown you today a formula for success. And it covers the 4 elements you see here, breath of IP, the ability to integrate, complete systems approach and execution. We think we are unmatched in many of these areas, and we have used this capability to change the economics of some of the businesses we've gone into, to improve the competitive dynamic for ourselves and the favorable economics for our customers. You've seen us apply this formula to areas like broadband, set-top box, switch, connectivity. We now have a strong #1 market position in each of those. What we're telling you today is we're going to apply everything we've got to go after cellular and wireless infrastructure with the desire to do the same in those spaces. So thanks again for coming, we'll see you at the reception.
Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.
THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.
If you have any additional questions about our online transcripts, please contact us at: firstname.lastname@example.org. Thank you!