Please welcome Vice President, Investor Relations, Mauricio Lopez-Hodoyan.
Good morning everyone, and welcome to Qualcomm's 2021 Investor Day. It's great to be back in New York and to see everyone in person. Before we start, I'd like to thank the multiple teams that worked on today's program and our executives for their commitment to investor communications. Now for some housekeeping. We'll make forward-looking statements in today's program regarding our business and financial expectations and other future events. I would like to refer you to our SEC filings for a description of our businesses and associated risks and other important factors which would cause actual results to differ materially from those in the forward-looking statements. Today's agenda includes presentations by Cristiano Amon, Jim Thompson, and Akash Palkhiwala . Then we'll have Alex Rogers join the question- and- answer session. With that, please join me in welcoming Qualcomm's President and Chief Executive Officer, Cristiano Amon.
Good morning, everyone. Thank you very much for being here. Actually, we're super excited, you all came in person. Full house, so this is make the day even more special. Before we start, I just wanna say big thank you to team Qualcomm that made this happen. This has been a big team effort because we have a lot to tell you. This is the time of Qualcomm right now. We are gonna have a lot of information to give you. We promise to make it all make sense, but the key message you're gonna see is we're truly diversifying. There's so many new end markets for the company right now, and the market is really moving towards our technology. Before I start, I wanna highlight a few things. Here's what you're gonna see today. Number one, demand for technologies continue to accelerate.
We are now diversified across many large addressable markets. The time for Qualcomm is now. We have one technology roadmap that we're very proud of, and that extends across every single growth opportunity we have, which we can leverage. We will have focused execution, as you have seen for the past few quarters, and financial discipline. That is gonna drive a company with strong operating cash flow and consistent capital return to our shareholders. Today, we cover our vision, our strategy, our incredible technology position, and our financials. With that, I would like to start the presentation, and I want to show you that we will always be the company defining the pace of innovation in mobile. You know us from mobile. We're no longer defined by a single end market and a single customer relationship.
While we're always going to be the company focused in driving innovation in mobile, there's more to Qualcomm. It is really an opportunity to connect everything that is supported by cloud computing in what we call the connected intelligent edge. It's about billions of devices that are becoming smarter, they're becoming connected, they have efficient processing, and they're enabling cloud computing. That is no different than the vision we had for 5G. While this is not enabled by 5G alone, it is clearly how we design 5G to make sure it's the technology that connects everything to the cloud 100% of the time. We are the company that will power the connected intelligent edge, and this is what we would like to demonstrate that to you and help you understand what are the opportunities we have ahead.
Our company is really focused on this vision, enabling a world where everyone and everything is intelligent, connected. That's actually supported by data. I just wanna highlight, there are a number of industry predictions out there. 64% of the data will be created outside the traditional data center. More data at the edge will require local intelligent processing. This is what we can do. Before we get to the presentation today, I know it's a busy slide, but I promise there's nothing in here that I believe you're going to disagree with me. When we look at the situation today, we see an incredible number of changes in the industry that are driving the need for technology.
Those industry trends, I'm not gonna be able to cover them all, but it's really been at putting Qualcomm at the intersection in creating demand for technology virtually across every industry. Accelerated digital transformation. That is happening at every enterprise. It got accelerated by the pandemic, but it's not pandemic related. It's about increasing the percentage of digital in the economy. Every single company is going to the process of becoming digital, accelerating digital transformation. Working from anywhere. Working from anywhere is changing computing. It's accelerating the connection between physical and digital spaces. It's changing home broadband. It's changing IoT at the home and at the enterprise. Convergence of mobile and PC is real. This is an area of incredible excitement for Qualcomm. At the end of the day, connectivity, advanced, efficient processing in AI is really gaining scale at the edge.
The Metaverse at different versions of what is going to be the connection between digital and physical is going to enable the next computing platform. In the automotive industry, it's going at an incredible pace of change where the car companies are becoming tech companies. We are well-positioned at the intersection of all these trends. When we look at that, it creates opportunities for Qualcomm across many end markets. Automotive, which we will talk more about it, and I like to point that some of you may have not seen, please look at the press release we issued this morning. It's very meaningful. Qualcomm has just been awarded the ADAS by BMW, and we're gonna talk about that during the presentation. Huge opportunity automotive and opportunity in the consumer, the industrial, enterprise, in edge networking within IoT.
That's what we're gonna spend most of the time during the day today, identifying, you know, how big those opportunities are for Qualcomm and our strategy to make that a reality. What is unique about our model is that we have one technology roadmap that is scaled to address all growth vectors. The reason is because mobile is winning. Our mobile heritage and DNA put us in incredible position to provide high performance, low power computing, on device intelligence, everything wireless. The leadership across not only AI processing connectivity, but camera, graphics, and sensors, will scale to support every single device at the edge, from earbuds all the way up to connected intelligent vehicles. That has been true as we look at the performance of the business as the diversification starts to gain scale. It's accretive to margins.
It's changing the operating margin profile of our semiconductor business, but we're just at the beginning of this incredible opportunity. The next slide is actually the most important slide. If there is one thing I would like you to take away from today's presentation and our vision for the future, is we have an incredible opportunity to grow. I wanted to look at this and I'll give you a notion of the past, the present, and the future. You're familiar looking at our business, about MSMs. We don't have the metric anymore. MSMs, MDMs or what some would call the thin modem in licensing. We have done well within that addressable market. Today, our technology is driving premium and high tier Android. We have an incredible opportunity for growth as a result of a changing OEM landscape.
RF front-end , it's one of the biggest success stories of a new entrant in RF front-end space, automotive and the broad IoT. That creates a $100 billion addressable market opportunity for Qualcomm. As we look at every single end market that we are powering, and we're just at the beginning of that transition with the full dimension of the connected intelligent edge, we see the opportunity to address a $700 billion addressable market in the next decade. That's one of the single expansion and create really secular growth opportunities. That's why this company can no longer be defined by a single market and a single end customer. That's the most exciting part of where Qualcomm is going and how we think about the future of this company.
I also wanted to highlight one other thing, and hopefully this will come clear to you in this presentation. There is belief in the market today in the cloud growth. I think it's undisputed. You just look at the valuation of some of the hyperscalers, cloud providers, and there's this belief that the cloud will continue to grow exponentially. Just if you look at the CapEx projections, 35% year-over-year growth. If you believe in the growth of the cloud, you believe in the Qualcomm growth plan. Because the, for the cloud to generate all of this data, you need billions of devices to be connected, to be intelligent, to have contextual information, and AI will get scale at the edge. We're very aligned, and you're gonna see that through this presentation, with the pace of digital transformation and cloud economy.
With that, I would like to summarize it. We're truly at the beginning of one of the largest opportunities in our history. The industry trends are creating new vectors of growth for Qualcomm, and we're seeing demand for technology virtually across every industry. We have one technology roadmap that is highly leveraged, is designed to scale, and is driving earnings expansion for our business. We're focused on customers in end markets that drive stable long-term revenues. This is an example of expanding our TAM and diversification while increasing margins and shareholder returns. At the end of the day, in intelligent connected edge, virtually all roads virtually was put in for legal purposes. Virtually all roads leads to Qualcomm. We're very excited about our future, and we have a lot to share with you today, and I promise this is all gonna make sense.
Now what I'd like to do is to invite on stage is our CTO, but is my friend. I have been working side by side with Jim for almost two decades. Jim is responsible for what's the best part of Qualcomm, which is our technology roadmap, and he's gonna tell you about it. Please welcome to stage Dr. Jim Thompson, our CTO. Very good.
Thank you. I've been at Qualcomm for 30 years. I started when I was a youngster. I started out on the CDMA project. I'm sure a lot of you are aware of what that is. I ended up in QCT over 20 years ago and have been the head of engineering for that for the better part of that 20 years. I've seen a lot of basically all of the technology development within Qualcomm. My career path kind of follows almost everything that we're talking about today. I've been really lucky in that respect. I think that our technology roadmap is really pretty spectacular, and it's been developed over, you know, a couple of decades, really from mobile.
You know, just going back, you look at Qualcomm, when I started, we were a mobile company. Starting with CDMA, that was about voice, but then we went to something called EV-DO, which was about mobile internet. Our vision was really about the mobile internet. At the time, this is going back 20 years, in developing these various technologies, we were looking at, or I should say when we were trying to drive the mobile internet, we're looking at, well, what is gonna generate use cases that are interesting. We're trying to imagine what the smartphone was gonna be. At the time, digital cameras were kinda new, so we started working on digital cameras.
At the time, it was simple enough that you could have two people assigned to it, and you could do a digital camera. Same thing with graphics, Game Boys, things like that, video. We started doing video, audio, MP3 players, things like that. We started getting involved in all of these different things that we thought would drive data and would drive the wireless internet. Our intention really was or our thoughts were really about modems, but we started doing all these technologies about 20 years ago. And so we've had this long history of developing them, and you end up with a. They're quite complex now. We have, you know, hundreds and even thousands of people working on some of these technologies at this point.
Very proud of our technology portfolio and, you know, we do competitive analysis quite often. In at least my opinion, there's lots of different ways you can look at these technologies, but I think we have the best portfolio in the world, and I think there's only a handful of companies that can compete with us. Now it's about you're looking at our future going forward, and it's about extending that technology roadmap. It came out of mobile, but extending that into the intelligent edge. I'll give you just some examples of when Cristiano talks about one technology roadmap, what we really mean by that, and I'll start with automotive. With automotive, there are cameras. There was just the BMW announcement about us winning ADAS, and that involves cameras.
If you look at a smartphone camera, it's about kinda making, for example, a face look good, a person look good, their hair look good. I mean, that's basically what it's about, and the sky looks very blue and nice. A car doesn't really care about that so much. I mean, you want to have some good colors so you can show things on the display. What you care about is computer vision and computers don't care about what they look like. It's about identifying a pedestrian very quickly so that you can make some decisions based on that. Computer vision for a camera is very different than what you would do for a smartphone. Other things for auto, supporting multiple OSs.
We don't do that in a smartphone, but all of our hardware, all of our IP needs to be able to run multiple operating systems in auto and other areas as well. Maybe talk a little bit about XR and their graphics plays a very, very big role. For example, whether it be VR or AR, when you render something with graphics, so you've got this very high resolution display that you want because you want everything to look super real, so there are lots and lots of pixels. Then you don't wanna render every pixel in the resolution that exists on the display, what you wanna do is only render those pixels that you're looking at.
If you can track using a camera, you can track your eye, what you're looking at, then you only render what's relevant and everything else is in low resolution. You can just imagine that by, you're looking at me, and I'm in high resolution, and then if I hold my hand out there, your peripheral vision can't really. The resolution of my hand is probably not as high. I got. Depends on where you're sitting in the room. That's an example of things that we do in XR. We're extending the technology. In that case, it's graphics. We're extending it into capabilities that are relevant to XR. Finally, PC. The thing I wanna say about PC is that it's actually moving towards the smartphone architecture.
Instead of us extending to it's kind of coming towards us. If you look at, like, operating systems like Windows, for example, now they have the capability of using accelerators. What I mean by that is that in a smartphone, we put all these accelerators like video, audio, for example, cameras, that the operating system understands how to use, and you can operate, and it works much more efficiently. Like, doing a video codec, for example, in the CPU or the GPU is much less efficient than doing it in hardware. The modern operating systems are starting to understand that, and they're really moving. That's really what smartphones are all about, the smartphone operating systems. Then on top of that, what we've done when you think of extending our roadmap.
We believe the future of the PC SoCs are going to be a fully integrated SoC. I think that that's something really important to understand. That means you're integrating CPU, GPU, all of these accelerators, connectivity, and so forth. That's all being integrated into one thing, one compute platform. For us, what that means is that we need to have a best-in-class CPU. I'm sure you're aware of us acquiring NUVIA, and so they're designing the next generation Qualcomm CPU. And then of course, graphics. We have the ability to extend our graphics all the way up to kinda gaming class, desktop gaming class graphics. Right now, you've only seen it in the smartphone, but you will see that in PC as well.
I can't talk about all of the technology. Don't have enough time to all the technology we're doing, but I'm gonna cover a handful of areas here. I think one of the key messages throughout this whole presentation is that we have this one technology roadmap. We've taken that mobile roadmap, and we've extended it, and that gives us a lot of advantage when it comes to things like time to market or just the cost of doing this IP is very expensive, and the fact that we have this base already gives us a fairly big advantage. I'm gonna start with AI, where we are with AI. I'm gonna make a claim that the gravity of AI processing is moving towards the edge.
Obviously, cloud's super important for that, but I think it's moving towards the edge now. Maybe I'll just explain that. If you look at, like, how you experience AI today, for the most part, the way you experience it is through your experiences on Google and Amazon and so forth. Really it's AI is used, they collect data on you, and then they essentially market to you, whether it's a product or a website or whatever. But that's really the way you experience it today. As we go through this digital transformation where there are sensors that are basically measuring everything about the world, whether it be a city, a business, your house, whatever, making things more efficient, that's generating an incredible amount of data.
You look at, like, for example, the most prolific sensor would be video. That's obvious one that, you know, easy to understand. You're generating all this data, and you just cannot send the amount of data that is being generated all back to the cloud. That's number one. Just can't do it. Some of that has to be processed on the edge. Then on top of that, there are other reasons like, for example, privacy, reliability, all these things that you probably think about when you think of AI that. There's many, many reasons why you want to keep that data. Whoever's generating that data will wanna keep that and process it on the edge of the network versus going into the cloud. I wanna give you an example of that.
This is just an example, I think it's a very intuitive example. It kinda goes with our ADAS announcement this morning. We have a number of different, not just BMW, but other ADAS customers, and if you look at the kind of data that is generated in the car, it's about, between our customers, range from about 5 Gbps to 15 Gbps. That is a lot of data generated, okay. You can't possibly send all that back. Imagine every single car on the road sending that data back. Not possible. Okay, there are other reasons too. You don't want to wait the time that it would take to send all that data back to the cloud. The cloud might be 100 milliseconds away, and then it has to come back, and you end up.
You could have moved 20, 30 feet in that kinda timeframe, and you can't afford that kind of lag and reliability. That just gives you an idea of why that processing has to be done on the edge. One of the things I wanna make clear, though, is that I'm not saying that there's no more cloud computing or whatever, you know, or AI in the cloud. That's not the case at all. It's just what I mean by that is that in a system design there's gonna be a lot of it that should be processed on the edge, and there's gonna be, you know, also a lot that needs to be processed in the cloud.
The point here is that there are certain things that must be done on the edge of the network, and that's what we're focused on. Okay. We've been working on AI for about 10 years, more than 10 years. You know, initially it was a research topic for us, and then it started to move into all of our technologies. If you think a lot of our technologies are signal processing based, and these signal processing-based technologies can be enhanced by AI. I mean, camera is a great example of that. It's really pervasive. It's affected almost everything that we do. We're also developing, and this is pretty, I would say a big spend for us right now or a big increase in spend for us right now, is on the platform itself.
When I say platform, it's what our customers use to develop AI in our products. Well, why don't I start. This slide talks a little bit about the base of that platform and how good the hardware is. What you see, the chart on the left is performance versus power, so performance on the vertical. High is good. And it shows us, this is for smartphone compared to our next best competitor. You can see that from a. This, again, this is a public benchmark that's run. We're about twice the performance and about half the power. That's really our focus. It's about making our accelerators for AI energy efficient.
I feel really good about the hardware we have. On top of that, we have a pretty good research effort. I would say that research effort, we write a lot of papers, but our objective really is to capture the results of those papers in tools that our customers can use, and its focus is really energy efficiency. I won't go through all these examples, just the top example there, where by going from floating point computation to integer computation, you can reduce the power consumption dramatically, so over an order of magnitude improvements that you can make. I wanna bring this up because, you know, we have a lot of activity here because we think that that is really how you get power efficiency.
Going to the next node, you know, process node helps. Moore's Law helps. But really where you're gonna get orders of magnitude kind of improvement comes from algorithmic improvements and us capturing that in software that we can provide our customers. Speaking of software actually is the single most area of highest growth, I would say, from a technology perspective. A lot of spend going into software. On the edge, it's a pretty complex ecosystem. A lot of different customers, a lot of different requirements, a lot of different models that they use, a lot of different what are called operators, the low-level operations of the hardware that are required.
At the core of it is something we call Snapdragon Neural Processor SDKs software developer kit. That's really the core of it. What we do for our customers is then we provide the ability to plug in, for example, the framework that they like to use, like PyTorch or TensorFlow or something like that. We have, I consider it to be the best hardware, the most scalable hardware going from low to high in terms of performance. Then we have this software developer kit that is quite. That's basically what everyone uses, and it's spending a lot of money on it and improving it every day. I'm gonna move on to camera.
Camera is one that is super easy for everyone to understand. My opinion is we're the best in the world at digital imaging. There's a lot of ways that you can look at it, measure it. There are benchmarks, like for example, DXOMARK is a benchmark that is used for measuring camera performance. Basically, kind of the pattern here that you're seeing here is that every year we come out with our premium tier product, Snapdragon in mobile. Then every year we have a customer. I mean, it looks like Xiaomi is typical. Xiaomi, OPPO. But we work with a particular customer, and we come out with the best camera in the world. We have that capability.
Really it comes down to from a customer perspective, you know, do they choose to, you know. How much effort are they willing to put into it? We work with them very closely at getting the best results from the camera. By the way, this isn't just image quality. That's part of it, but it's also video quality. There's a lot of different features, aspects to this benchmark. I wanna go to a video, and if we could run that video, that would be good. There we go. Okay. This is an example of a video that was taken with Snapdragon. It's an 8K, 30 frame per second video. We're the only ones in the industry that support that right now. It uses H.265.
That's a number. We also know that it's a standard, but HEVC is another term that's used in the industry, and it gives you about 240 x compression from the actual data, and it is visually lossless. Then going forward, we have other capabilities, like, for example, we'll announce this shortly in two weeks, I believe, but high dynamic range. What that means is we just essentially support more bits in the pipeline, which allows us to get blacker blacks and lighter lights. We can make things like this. They look very realistic looking, and you can do even better. It just continues to improve.
Not only that, but just even from a compression point of view, we're leading the standard, called H.266, and it's called VVC, and that'll increase the compression by about a factor of 2. This is just an evolving technology, and it keeps moving. What? Let's. Oh, there we go. Okay. Just talking about cameras and how AI intersects with cameras. I pulled out this one example. It's very simple example, and the kind of the point I wanna make is that it's not all gonna be AI, but AI can make your camera much better. There's a lot of conventional signal processing that you do.
Look at the picture up top, and what we can do is you take that picture, and then the phone or your device can look at that picture and determine what's what. For example, what is the sky? Then what is the face? What is the hair? For example, with sky, you like it to be blue, right? So we can make that blue or make it bluer than it is. You like a little definition in the clouds. You like the skin to be smooth, but you like, for example, the buildings in the background to be sharp.
That's all very different processing that you do, but it allows us, by doing what's called semantic segmentation, we can determine what is what, and we can process everything differently in that picture to make it look the best. That's an example of an AI feature that it will become very prevalent in smartphones. On the left there, I guess that was running before. I didn't watch it, but this is just a night shot. With the night shot of video, I guess this was a video. Was it running before? I don't know if you had a chance to see it. Anyways, you can see on the left is the improved and on the right is the original.
On the left, what we do is we take multiple frames, and we can blend them together to get better signal-to-noise. On top of that, we have an AI algorithm that we run that can eliminate things that are artifacts because you know, just like you know when something doesn't look right, AI can know when something doesn't look right. Also, like, just colors, like with green, for example, you know that the tree should be green, so AI knows that too. You can make changes to the video on the fly and just make it look a lot better. That's just examples of how we use AI. This is about extending our roadmap with camera.
You look on the left, and this is about computer vision, so the computer vision portion of our camera. For auto for ADAS, we needed to do something that was called dense optical flow. That's kinda confusing, but what it means is that you track the motion of every pixel. On a pixel-by-pixel basis, you track that motion. You've got these motion vectors associated with every single pixel. That's why you want that in automotive, so that you can very quickly see a pedestrian, for example. You can identify it, so you're not tracking an object. You're tracking pixels, so you can identify things super fast. Then that, in turn, can go back into our other, like for example, for XR or other. That feature is not just for automotive.
It flows back into our roadmap for everything else. Other obvious one is in automotive, the camera can be looking straight into the sun, but you still have to see that pedestrian on the side, so you can't have it just wash out your camera. Supporting very high dynamic range is another example. Again, it's just an extension of our existing roadmap. If you look on the left or the right, sorry, for XR, I mentioned foveated rendering and eye tracking. If you can see that only the things that we're looking at are in high resolution, and when we move out of high resolution, you can't see it so well. That just allows us to reduce the processing that we need to do in the graphics.
Then on the far right, that's just an example of camera understanding what the room is. Mapping the room and then also understanding what's a surface, what's a wall, you know, and in this case, it's just being used for a game. That general concept is used over and over and over again in for our products. Okay, moving on to graphics. Okay, we've been shipping graphics for about 10 years, and if you look at our mobile graphics, we're the best in terms of power performance. I think that's really a key thing, especially as we move into this the PC era, where we scale our graphics up. Being power efficient and doing an integrated SoC, that's super important.
Let me go to this chart, and this chart just shows the sustained performance of our graphics. That's really what's important because, for example, graphics is most heavily used with for gaming. In gaming, you don't play for one minute, you play for, you know, I guess I don't play, but probably not many people in this room play, but kids definitely play. It's, you know, it's 10 minutes, 20 minutes or more, and it's really about sustained performance. When you get a benchmark like ours. This is our nearest mobile competitor. You'll get a benchmark that's really high, but that benchmark doesn't mean anything. The benchmark only means something if you run it over and over and over again, and that's what this chart is.
Our graphics also scales. It scales from supporting milliwatt kind of range, a watch, a smartwatch, so tiny little display. Then it's, it will scale, like I said, to automotive, which has some compute requirements that use our GPU, but then also to PCs. I just wanna make it clear that our graphics will scale up to desktop class gaming capabilities. The architecture allows us to do that. Okay, I'll move on to processing. Really, I've just got one slide here, and this is about the purchase of NUVIA. What they're working on our next generation ARM CPU, and you think about that they...
They're pretty far along at this point, and we'll sample a product. Let's say nine months from now or something like that. I can't remember the exact date. Anyways, we're pretty far along with that development. Our objective is to have the highest performing low power CPU in the industry. If you think about our overall product roadmap or our. I'm sorry, our technology roadmap, that was the one weakness that I felt that we had for quite a while. By acquiring NUVIA, we've filled in that weakness. I'll go on to the next. Connectivity. Start with just all of the connectivity technologies that we have.
I consider us to be leaders in all of them. We're leaders not just in products, but also in standards. I wish I had more time, but I see I'm running a little bit long, so I'm going to go right into 5G. Okay, so the basis for 5G today is Release 15. Release 15 was designed to be very flexible. Think of it as being forward compatible so 'cause the whole point of 5G was to be able to have this, call it platform, that allows us to build all sorts of capabilities on top of it. The next release that's coming out is Release 16. Release 16, are the first device that will come out will be in December, so it's maybe a month away.
Most of the devices, pretty much all the devices that are coming from Qualcomm, that are using Qualcomm chips will be Release 16 next year. Release 17, the standards are almost done, so it's pretty much baked, and that will be about two years from now. That'll be in 2023. End of 2023, you'll see Release 17 come out. If you look at what's in Release 16 and 17, there's a lot of features associated with a smartphone. A lot of coverage enhancements, mobility enhancements, capacity enhancements, band aggregation to provide, for example, aggregating millimeter wave and some sub-6 bands together to improve the coverage of millimeter wave, for example. There's lots and lots of features associated with smartphone in Release 16 and 17.
If you are a competitor to Qualcomm, you better keep up with that roadmap. Very, very kind of complicated, dense roadmap. There's also a lot of new device types that are coming that are being supported in Release 16 and 17. Just quick examples of that would be like C-V 2 X. I should say cellular vehicle to vehicle to pedestrian, vehicle to infrastructure, and supporting that industry. There's also like for example I would call it wireless Ethernet is another example. Very high throughput, low latency, high reliability kind of communications. Those are just some quick examples. Finally, Release 18 and beyond, that's what we call 5G-Advanced.
That's where you're gonna see, like, for example, a bunch of features come out for extended reality, augmented reality. It's a little different requirements than what you would use in a phone. A lot of AI capability is gonna be introduced at that point. There's a lot of talk about the industry where 6G is AI. Well, you know, in reality, 5G is too. It was built that way to be flexible. You're gonna see capabilities that come in, like that we'll be introducing that will, for example, improve the capacity by 60%, by just using some AI capability and determining the channel, what are called the channel conditions to the various users in a sector. Lots of interesting stuff coming.
Maybe just talk a little bit about our capability. We are a company that supports the entire world, and I think that makes us unusual. When I say the entire world, the world is not all 5G. 5G is not the same everywhere, 4G is not the same everywhere. There are all these different band combinations that you have to support. They're different everywhere in the world. Like for example, you know, there are about 10,000 different band combinations that we have to design for. It's pretty complicated if you're doing that. And then there's also 180, I believe that's how many that have launched 5G at this point. But everybody's got a little bit, you know, different bands. They've got different infrastructure providers. Their feature sets are different.
There's a lot of idiosyncrasies associated with the bands or the global nature of mobile, and it's something that we do very well. I think it shouldn't be underestimated that the value of being in our fourth generation of our 5G baseband. It's just a lot of complexity. Every single generation that we come out with improves from a throughput point of view, from a feature point of view, and from a power point of view. In fact, they're pretty dramatic improvements that you'll see that we're achieving now and you'll see in the future. Okay.
Now the way we think of modem is not just baseband or baseband and transceiver, but we think of baseband all the way to the antenna. To start, you need to have best-in-class components, okay? Like for example, with our front-end, that's the latest. We've been working a lot on that, and we're actually super happy with how things are going in our front-end. The set of components that we have, it's like PAs, LNAs, switches, filters, pretty complex to kinda go, well, who is the best? I would say we are, you know, you compare us across all the different players and we're extremely competitive is the way I would describe it.
The advantage that we have is that because we control the whole system from baseband all the way to the antenna, we can optimize, and there are features that we can provide that nobody else can provide. Those features end up giving us a better performance at the system level, so better thermal performance, actually better throughput, better power, all sorts of advantages that we get out of designing it as a system. The first thing I wanna help you understand or make you believe that we actually, from a component level, we are the best. The one area that's really easy to look at, so you can kinda look at a chart and see, is filters. We have the most comprehensive filter portfolio.
We released something we call ultraSAW, SAW surface acoustic wave filter, you know over a year ago. It's pretty much every single one of our customers that, I should say, our Android customers use ultraSAW at this point, and it is clearly the winner in terms of performance. You can see the chart. Our next best competitor is below us. ultraSAW very successful this past year. We're going to a technology called ultraBAW for frequencies above 3 GHz, and that is on the right side. BAW is bulk acoustic wave filter. You can see there on the chart that that's comparing us to our next best competitor.
That's the area that I'm actually really happy about at this point, that we've closed the gap with that very strong competitor. You look at the launches, virtually every one of our customers, again, if you look at this coming year, they're gonna be launching with ultraBAW. It's been very successful from a component point of view. I'll talk a little bit about system here, but before I do, this is an example with the power amplifier. If you look at one of the things that we've done over the last number of years is we've developed our own transistor. We have our own epitaxial stack up for that transistor.
That's got us to the point where we're one of the best, if not the best in PA. It's complex how you measure it. What we can do is not just have a very competitive PA, but we can have a much better PA because we can tie it in with the system. One of the things we do is what we call DPD, digital predistortion. We have digital predistortion in the baseband with the knowledge of the PA, so we can do a much better job. Then on top of that, we use something called envelope tracking. Because we know exactly the signal that's going out, we can always bias the PA to be at its most efficient point.
This chart shows 30% advantage, but that advantage can be even higher depending on the modulation rate. It's really something that's, you know, an advantage that we have that's pretty hard to compete with. If you look at smartphone, or, I'm sorry, modem, there's I mean, obviously, it starts with the smartphone, but there's a lot of other areas just to give you an idea, like for example, in fixed wireless. Fixed wireless is something that is really taking off for us. For example, we support eight antennas, so eight different digital streams. We also support higher power modes. All those things that came out of the standards and standards that support other types of device types.
That's a fairly easy extension from our existing design that we have. It's not super difficult for us to do that. And then on top of it, there's other things I mentioned, wireless Ethernet for, think of the factory of the future where you're getting rid of a lot of the cables. Those are examples of things that we do for modem. I hope that I gave you a good feel for all these technologies and why we've taken the investments that we've made over the last few decades, and we've extended that with understanding of all the end markets we're going into and extended all these technologies to support different these different end markets.
We have this one technology roadmap to do that, and I think that's a big advantage that we have. That applies perfectly to all these different end markets that we have that Cristiano is gonna talk about. I'll say one more thing, and that is, we have a very healthy patent portfolio. Obviously, our focus there, lots and lots of work in 5G. All the subsequent releases that I talked about, there's lots of patents that are being generated associated with that. Then also just our very broad technology portfolio. It's the same thing. We've developed a lot of I think very important patents associated with that. So we protect ourselves and also our licensing business.
That, I'd like to thank you, and I guess we're going back to Cristiano.
Thank you very much, Jim. Hopefully, what you could see is we have a very unique technology roadmap. Pretty difficult to compete with Qualcomm in connectivity, because we continue to drive the roadmap forward. We have everything wireless. Think about our camera and our audio. You can see, you can listen, we can sense with RF, and we have the most efficient processing, whether it's CPU, GPU, and an AI. That's the combination that is really driving the scale of Qualcomm and the growth across a number of end markets. What I'm gonna do in this part of the presentation, I'm gonna walk you through our strategy and then provide you an ability to understand how we're executing on those new opportunities. I've divided the presentation in four key pillars of the strategy.
In handsets, we're very focused on the opportunity available to us in the market right now. Android is one of the fastest-growing revenue and margin expansion opportunity for us in handsets. The second one is our RF technology across everything. Number three is the opportunity that we have in automotive to be the preferred partner of the auto industry. And the broadest opportunity of them all is our IoT revenue stream, you know, where it's about all of those technologies that Jim outlined to you, connectivity and smart processing, enabling the cloud-connected edge. With that, let's just start focusing with handsets. Our strategy can be summarized in one single statement. Snapdragon is the platform of choice for premium and high-tier Android smartphones. Let's talk about Snapdragon a little bit. The phone market is changing. I think people need to get out more.
There is incredible opportunity within the Android space. There's a change in the OEM landscape. There is an opportunity for Android to move forward to becoming more premium. The Snapdragon has been the undisputed leader, setting the industry benchmark for premium experiences in mobile, period. Number one in a number of different categories, and doing that with all-day battery life. One thing that is probably gonna be new to you is the incredible effort that we have put in in such a short period of time in Snapdragon marketing. The reason I said that people need to get out more, 80% awareness in China and India about Snapdragon processor. It is now the number one preferred smartphone, mobile processor brand.
There's one thing I would like you to look in this slide. When our customers are launching their premium flagship devices, and how they communicate that to their customer base, there's one thing in common. You know, it's displayed front and center about the experience the consumers will see out of their premium smartphone. That's an incredible, valuable asset in, for Qualcomm, and it is incredible power in the industry today of the Snapdragon brand. I'm also gonna use this opportunity, hence it, to make a new disclosure. We have now two-year customer commitments for premium and high-tier handsets across all of our main customers. Xiaomi, Honor, Vivo, OPPO, we have two-year contracts with them. Akash is gonna provide a little more detail on the financials that provide incredible predictability for our handset revenues.
When we think about Samsung, this is a relationship that we have been expanding, the upswing trajectory. When you think about share, we have design wins across all of 2022. For the new flagships and new form factors that are getting traction in the market, Qualcomm has the worldwide volume in the Fold series and the Flip series, and a very stable position within the Galaxy S series. It's really, this slide is the validation that the strategy is working. It's a focus on premium and high-tier. We're highly differentiated with Snapdragon. It's consumer preference, and that is providing incredible stability of our revenue in the handset space. With that, I'm going to talk about how we feel about the RF front end market. As we enter the space, we've been redefining connectivity, as Jim Thompson outlined, modem plus RF.
It's not only an opportunity to generate one of the fastest growths in QCT in the mobile space, but also to take that across every industry. This is a summary of where we are. We're well-positioned to be the global leader in revenue. We're the number one in the mobile industry. We exceeded our 2019 analyst day target one year ahead of schedule, and we've been designed across every single OEM. There was a lot of skepticism about Qualcomm entering this space, as the fifth player, and we're very proud of the execution of our team and the fact that we're winning with technology. This is incredible position to be in, and we have opportunity to grow. Then I wanna give you an ability to understand our scale. We now have performance leadership across every component.
Jim outlined the filter. With our ultraSAW, with ultraBAW, we're now defining the benchmark for performance. Besides leading in performance, we have now incredible scale. We shipped 8 billion RF front-end units in fiscal 2021. The smallest unit is in excess of 300 million. This, it's a very good position to be in. We had executed well in handset, and going forward, we have an incredible opportunity to grow. There are gonna be three drivers of growth in addition to tracking or expansion in handset. Global millimeter wave adoption. Millimeter wave is deployed in the United States, is deployed in Japan, it's gonna be deployed in Korea, and we remain optimistic with all the activity in China, with the China Unicom millimeter wave for the Winter Olympics. But our...
As Akash will outline our assumptions, we don't have any assumptions in our long-term model other than the millimeter wave markets that are commercial today. The reality is, if millimeter wave gets traction, and it will get traction, and it will expand, that's an incredible upside opportunity for our RF front-end business. We're also gonna extend the RF front-end, modem plus RF, into the Wi-Fi, and we're gonna attach RF front-end to automotive and IoT. We're very proud that RF front-end is a success story by every metric, and it's gonna continue to be part of the growth of Qualcomm. With that, I wanna move away from handsets and front end, and talk about automotive and the IoT. Starting with the automotive. What is unique about Qualcomm is the ability to have capabilities across every single domain.
Our single strategy statement is we are the preferred partner for the digital chassis. We have leadership. We're the number one in telematics, in automotive connectivity, also the number one in premium infotainment. We're now working with 25 global OEMs. The last time you heard from us was 23. We're working with 25 global OEMs. They have select Qualcomm platforms. With the acquisition of Arriver, we're very well positioned for ADAS. What we have been building is a solution that really meets the demand for technology as the car is being transformed. You should think about the car with a completely new digital cockpit experience. The demand for that is very easy to see, and it also points why Qualcomm has been extremely successful in building a digital cockpit.
Users are driving looking at their phone, and at the end of the day, there's opportunities to bring a much more immersive experience, what's relevant to you within the cockpit of the car. The car is gonna be connected to the cloud 100% of the time. With 5G, the car is connected to the cloud. When that happens, the car becomes the center of distribution of media, gaming, personalized experience, a lot of data analytics and artificial intelligence. New OEMs create a completely new business model with services and upgrades over the life cycle of the vehicle and ADAS become table stakes. We built a platform for this. That's what we call the Snapdragon Digital Chassis, which is a platform for the future of automotive.
Our success in a very short period of time is not only because we have the relevant technologies that are required for this transformation, but the ability to have capabilities across every single domain and do what Qualcomm does best. We build a system. The Digital Chassis comprises of the Snapdragon Cockpit Platform, the Snapdragon Auto Connectivity, Snapdragon Ride for ADAS and autonomy, and Snapdragon Car-to-Cloud Services. What that does, it provides an opportunity for us to increase our silicon contents within new vehicles by 10x. We're not doing this alone. We're doing this in partnership with our customers. If you look at the valuation of the auto companies right now, and you look at valuation of a company like Tesla, it's very clear to every one of the OEMs that they're becoming a technology company.
Qualcomm become the preferred partner to help build this technology future, and that's why we build this open flexible platform, which is our Digital Chassis, in partnership with our customers. One of those great strategic customers and great example of bringing together Digital Chassis is our partnership with General Motors. That's the ability of creating a platform that enable not only new services and technology but makes the car an innovation platform as a service center. We're very proud of the partnership with GM, and we're working with them across every single domain from connectivity cockpit to the service platform with our Car-to-Cloud service Platform, as well as the Super Cruise. I think that's just the beginning of what's happening in our partnerships with the automotive sector.
When you think about having all of those technologies, this also has an impact outside the car when we think about the future of transportation. I like to provide this example because it's easier, easy to understand. Let's just talk about navigation today and navigation tomorrow. I think that's an example that'll tell you the difference about being a component provider versus a system provider, and how all of those technologies work together. Navigation today, you're driving your car, you're behind the wheel, you look at the map in your infotainment. You have the streets, you have what position, location, your position in that street. You have the direction of traffic. Looking information that comes, for example, from Google, you have the ability to see green, yellow, red based on number of cars in the road, the congestion.
Think about the future of navigation even with the driver behind the wheel. With technology such as 5G C-V2X, the car is connected to every other car on the road. It's connected to every pedestrian, to every bicycle. It's connected to the traffic lights, so you start to get information on how much you should adjust your speed to get a green. You have information about what's happening at the intersection. Is there a car coming, and is supposed to be stopping, and based on the rate of speed, you need to get an alert. The front-facing camera looking at you from the dashboard powered by the digital cockpit, it will understand if you're paying attention or not based on recent events. A pedestrian moving straight in a parallel to you in the sidewalk, you're 45, 65 miles per hour, that's a safe environment.
If the direction of speed of that pedestrian is towards the street, that's not a safe environment. Ability to predict, use artificial intelligence and processing in the car, and make sure that all those systems are integrated together. It's the ADAS system working with the digital cockpit with the cloud connectivity. That's what makes the Qualcomm position very unique, and in addition to building, you know, a lot of technology in the car, building the future of transportation. We're very happy with how we can make this come to life through the enhancement of our platform provided by the acquisition of Arriver. Qualcomm Technologies plus Arriver allow us to have a scalable, open, proven platform for ADAS and autonomy. I'll give you a glimpse of how that's gonna look like.
The Snapdragon Ride Platform, front and surround vision provided by Qualcomm and Arriver. Driver policy provided by Qualcomm. Many years of research, some of you have seen our cars driving around in San Diego, plus Arriver, plus the OEM stack. Open platform with a broad set of partnerships for parking and driving monitoring, and then multiple software platforms. The reason we're getting a lot of traction already, and the market is really paying attention to what we can do in ADAS and autonomy, is because in addition to having the most comprehensive platform, it is open and allow the OEMs to also innovate. It's the Qualcomm horizontal business model that has proven to be very successful across all the business and what we have done in mobile.
With that, I'd like to show a quote from a great partner of Qualcomm, which is BMW. As part of the announcement we made this morning, we've been awarded the next generation ADAS in autonomy from BMW. We're gonna be helping BMW build next generation ADAS and autonomous driving platform, and we're very excited to continue our expansion automotive with a great strategic partner such as BMW. I think that's just the beginning. This is no different than what happened to us in telematics. In digital cockpit, we have an incredible opportunity to scale very fast, and that which is reflected in our automotive contract pipeline right now. With that, I'm gonna go from automotive to IoT. I want to take you back to the many trends we discussed earlier, and I'll pick a few.
Convergence of mobile and PC. How do you build new computing devices at the edge? Mixed reality, XR, virtual reality, augmented reality is the next computing platform. Connection of physical and digital spaces, this can be as big as mobile, and will enable the Metaverse. 5G is driving next generation infrastructure and fundamentally reshaping how we think about edge networking. In the broad digital transformation of enterprises, back to the message. Growth of the cloud is directly associated with the growth of the edge. That is driving demand for our technology. That creates opportunity for us to do connectivity and smart processing across the entire cloud-connected edge. Across consumer, edge networking, and industrial IoT. As you reflect upon those trends, and I just highlighted few of the trends I showed this morning.
There's one company that shares our vision across all of those areas, and that company is Microsoft. Here are a few words from Satya.
Hello. Thank you so much for having me here at your Investor Day to talk about our deep and long-standing strategic partnership with Qualcomm. Looking back, there is no question that in the past 20 months have been a catalyst for an unprecedented wave of digital transformation that's creating important new opportunities for our customers. Fundamentally, we are moving from a mobile and cloud era to an era of ubiquitous computing and ambient intelligence. To put this in perspective, by 2030, there'll be 50 billion connected devices, more than double the number today. This means the places we go to, the things we interact with, will increasingly be digitized, creating new opportunities and new breakthroughs, from precision medicine to precision agriculture, from personalized e-commerce to personalized education, from connected manufacturing floors to connected homes. That's why our partnership with Qualcomm is so important to us.
Qualcomm is a leader at the intelligent edge, driving advances in efficient computing, wireless connectivity, and on-device AI. Your vision for a future of technology where everyone and everything is intelligently connected is aligned with our own. Together, we are helping our customers apply the power of our entire technology stack to meet the real-world needs of today and tomorrow. There are so many examples from how we have collaborated to bring ARM to Windows ecosystem, to how we have designed silicon and systems for category-creating devices like Surface Pro X or Surface Duo and HoloLens 2, to how we have advanced new IoT platforms that bring AI to the edge. We are also working together in many areas that will be increasingly critical to our customers going forward, including building the next gen 5G infrastructure and private networks.
This is all just the beginning, and I greatly look forward to what we will accomplish together in the years ahead. Thank you so much.
We're incredibly proud of the partnership with Microsoft. Few companies are as strategically aligned with Qualcomm as Microsoft. This is a very long-term relationship for Qualcomm. We're really committed to their success. What is interesting, it exactly expands across each and every one of the opportunities we have in the connected intelligent edge. As we think about each of those opportunities, and I know one of the questions is: How should we think about the Qualcomm IoT revenue stream? I'm gonna provide a summary to you as I go to each one of those opportunities. IoT is changing. The way to think about IoT, we think about IoT next generation. It's moving away from traditional microcontrollers with embedded software and connectivity. It's really about always connected. If it's not connected, it's not useful.
It's about high performance computing that can run a full operational system with very complex computational requirements, and doing this with very low power. It's about scaling AI at the edge with on-device intelligence. Our IoT revenue stream and growth opportunity is divided in three categories, consumer, edge networking, and industrial. You're gonna see a lot of opportunities and they look very different, but there's one thing in common. I would like to use this time in the presentation to connect it to you where is this coming from, because it's really about building on those industry trends. They're moving the demand towards the Qualcomm roadmap. On consumer, it's about the convergence of mobile with PC and the upcoming opportunity with the Metaverse. Wearables is about extension of what you do with your smartphone and the devices that are around your smartphone.
Consumer electronics is about going from the IoT of the past into the IoT of the future, leveraging the technology roadmap that comes from mobile. Networking, it builds on Qualcomm connectivity 5G DNA. We're not ashamed to say that we're the number one cellular and Wi-Fi company in the world. We really understand connectivity. We really understand connectivity and RF, and we can take that to the other side of the link. We can leverage our technology to build networking as networking is being transformed. Industrial is just this very broad digital transformation which started and is gonna grow significantly within the next few years. That's how we think about this IoT revenue stream. Now I'm gonna take you to each one of those, and I'll tell you what's really important at the end of my presentation.
The first one, within the consumer space, is to focus on delivering next-generation consumer experiences. You can easily understand what this opportunity is looking at this slide. On the left, you see the opportunity to expand the Snapdragon platform with attached devices. For the investment community, the easy way to think of this, there's an incredible opportunity in a very large SAM created by Apple for watches and AirPods. We're the company providing those devices for everybody else. That's the opportunity within attached wearable devices to the Snapdragon platform. The other one is really focused on category-leading devices for the new consumer electronics that are representatives of the capabilities of the edge. Example of this is the Peloton Bike+ , the Tread, is the Amazon Astro robot. Here's a quote from David Limp from Amazon, highlighting the role of Qualcomm.
Qualcomm Technologies help us realize this vision, delivering world-class solution that make this reality across multiple products." This is just the beginning of a great opportunity within the segment. This is a very important partnerships expanding. We're working with Amazon with a number of devices under Home, from the Eero device to Ring, all the way to the Astro robot. We're excited about what we can accomplish together. That's a great example of what's happening with Qualcomm in the IoT consumer segment. Now I'd like to talk about the convergence of mobile and PC. This is one opportunity that we are very bullish on, and we are super excited about the recent developments in the industry that is moving the architecture to a mobile architecture. Qualcomm is using our Snapdragon platform to enable next-generation laptops.
You can connect to all of those trends we talk about, the work from anywhere, the enterprise transformation of the home, and what's happening with the industry. I'll point to key things that defines the next-generation laptop. Always connected, enterprise anywhere that drives on-demand computing. 5G on-demand computing allow, if you have a workstation, you have a on-demand computing that you run workloads from the cloud on your laptop, and you can move, whether it's in the, in the work premise or from the home. Collaboration experiences. Camera, audio, super important in PC. I bet each and every one of you, number one use case on your PC right now is a collaboration tool, whether Teams or Zoom. It took a pandemic to create a killer application of video telephony. We've been trying to do that since 3G.
Now in every single one of the PCs, the number one application is communications. It's even changing phone form factors, as many folks are holding their phones like this to do a Zoom call. Gaming is moving to the cloud. It's a no-brainer to a gaming developer to be able to develop a game, host it in the cloud, have a compatibility with, instead of multiple consoles in the desktop gaming PC, have a compatibility with every single screen out there, whether it's a television, it's a computer device or it's a phone, do online gaming. As that happens, the most important thing is do you actually have a gaming PC that allow you to have an all-day battery life, it has the right connectivity for you to play games, instead of thinking a separate dedicated, you know, gaming desktop.
While gaming desktop will continue to be a market opportunity, laptops can now provide that experience as companies like Netflix and Microsoft started to bring gaming to the cloud. On-device AI accelerated experience. You didn't see a lot of AI processing in PCs. Very CPU-centric architecture. You see some of the announcements were made in Microsoft in the Qualcomm platform, and that's unique to the Qualcomm platform, the ability to use on-device AI. Great example of that is your eyes are always looking at the camera when you're doing a Teams call. The ability to continue to have efficiency processing with the leading performance per watt. This is an incredible opportunity for Qualcomm, and you should think about where we are. We have the winning team. We are very well-positioned to be the preferred platform for PCs for the inevitable transition to ARM.
When we think about next-generation CPU, it's about creating the benchmark in performance for Windows PCs. We're going to scale our GPU to desktop-like or discrete graphics performance capabilities, and we're gonna have the most power-efficient AI engine in addition to what we do in multimedia connectivity. That's another great opportunity that we have within the IoT consumer. Here's a quote from my dear friend Panos from Microsoft about the very broad partnership we have with Microsoft for the transition of Windows to ARM and the ability to build on this incredible opportunity for the future of personal computing. By the way, great job from Panos and Microsoft on Windows 11. It further validates the opportunity that we see right in front of us. Now I'm gonna go to the last category of IoT consumer, which is the next computing platform with mixed reality.
One thing that you're gonna see in every single one of those discussions about the Metaverse. Absolutely the majority of the devices that you see right now in virtual reality and augmented reality is powered by Snapdragon. We also recently announced the Snapdragon Spaces. Snapdragon Spaces is the XR development platform. We're getting a lot of traction, and we couldn't be more happy with what we have done partnering with Meta on the Oculus Quest 2. Snapdragon XR2 also named Time Magazine's 2021 Best Invention for Augmented Reality. That has the potential to be the next computing platform and could be as big as mobiles if we're thinking about carrying AR glasses with us in addition to a smartphone. Here's what I would like you to take away.
If you are going to spend time in the Metaverse, Snapdragon is going to be your ticket to the Metaverse. The technologies that Jim outlined is very difficult to do. How you connect the human into a virtual reality and augmented reality world, and that's what we have been investing early on. I remember talking about XR before it was popular a few years ago, and we're very excited about our position right now. I'll point out to you that this is already starting. Oculus Quest 2 was 10 million units. The success of Oculus Quest 2 had an impact on the company that is providing. High performance, super low power hardware and perception algorithms and software platforms are required to make this a reality, and that's a role that we can play, and with that, create an incredible growth opportunity.
Here's a quote from Mark Zuckerberg. The partnership that we have is very, very strong. It's multi-generation, and we're super excited with the work that we have been doing with Meta and the Oculus team. I honestly believe that's just the beginning. I summarized those three opportunities. Now I'm gonna talk about architecting the modern network. IoT edge networking, we divide that in a couple of new opportunities that are available to us. For the first time in wireless, we now have a technology that can truly compete with fiber. 3G, 4G for home broadband wasn't an ideal technology. 5G is now the fastest-growing last-mile broadband technology, and it's really gaining scale. Just to put in perspective, there's now 68 5G fixed wireless access providers across 32 countries.
We predict that by 2026, 25% of global mobile data traffic is gonna be provided by 5G. This is a growth opportunity to leverage their technology. It's not the same as mobile. It's about creating a very reliable, long range, high performance, and that is reflected in the design pipeline. Virtually everyone that is building on this opportunity is designing with Qualcomm. That's an incredible opportunity for growth, highly leveraged, and we see this happening at both developing economies as well as developed economies. Here's example for you of a quote from T-Mobile about our partnership. T-Mobile has launched 5G fixed wireless access, providing now home broadband in addition to mobile, and it is a very strong partnership with Qualcomm.
I think we're just at the beginning of this transition, especially as more countries are deploying 5G. Bigger is going to be the addressable market for 5G wireless fiber. There's also a lot of excitement on Wi-Fi access point. What we saw happening, first, there was very high demand from the enterprise transformation of the home. Now that demand continues to be very high as the workplace is preparing for the future of work that includes collaboration and video. We see Wi-Fi anywhere, as not only we're connecting computers and phones in the home, but every other device out there. The mesh continue to evolve and be deployed everywhere. The Wi-Fi access point is becoming a hub for IoT in the home. Is not a coincidence. You saw the success of the Amazon acquisition of Eero.
The evolution, the access point with more technology and more value. You have the opportunity to create the modern network connecting the access point directly to the cloud. You see a very vibrant technology roadmap with Wi-Fi 6E and Wi-Fi 7, with carrier upside expansion as Wi-Fi access point become connected to 5G wireless fiber. We have 30+ immersive home designs with Wi-Fi and 300+ customers, for enterprise-grade designs with our Wi-Fi access point platforms. Here is a testimonial for one of our great customers, NETGEAR. NETGEAR has been first with the majority of our technologies, a very innovation-focused company that is helping us commercialize next generation Wi-Fi access point as well as access point with 5G. Then I'm gonna go to the last part of the IoT edge networking.
We don't talk much about this because we have a lot of things that we're talking about, but the cellular infrastructure is changing. It's changing fundamentally because the RAN is getting virtualized. Many of you have heard about vRAN. You may have not paid attention, but Qualcomm has been leveraging our own roadmap to build what's the world's best solution for the future of vRAN. Not only we have solution for the small cells, but we have solutions for the large capacity radio units, and we have an inline accelerator card that plugs directly into a data center.
What the feedback we have been receiving from the market compared to every single alternative they have is 50% lower total cost of ownership and 50% more energy efficiency, which is particularly important as many of the operators now have aggressive sustainability targets. This is becoming a reality, and I just wanna show, and we're very proud of showing that, we had made an announcement with NEC. Here's a testimonial from, NTT DOCOMO. Japan is one, and this operator in particular, is one of the most advanced, most difficult to get in operators in the world in terms of the quality requirements and the feature requirements they require for 5G. We're partnered with them and NEC for next-generation vRAN architecture using the Qualcomm platform for both the radio unit as well as the digital unit, for vRAN and Open RAN.
That's yet another growth opportunity. Before I summarize it to you, I wanna spend some time on the broader IoT industrial. The transformation of industry is really tying Qualcomm to the cloud. We become, and you saw that clearly from Satya video. As the cloud company and Microsoft as the enterprise cloud company continues to grow with the cloud economy across a number of enterprises, we become the partner of choice at the edge. It's not only unique to the cloud, it's about changing all the companies with the technologies we have at the edge. I wanna start listing some of the different areas. I won't have time today to cover each and every one.
I'm gonna pick one example, but hopefully, this example will give you an idea of what's driving a lot of the fast growth for Qualcomm, already significant revenues in this space, and an incredible opportunity as IoT matures. There are a number of different verticals. I'll spend time today providing you examples of smart cities and retail, but even the verticals that we're showing you today are not all the verticals we're focused on. We're excited because at the end of the day, the market is moving towards our one technology roadmap. We're making an announcement of the smart Times Square experiences. If I believe it, the announcement is going out today. We're really partnered with Times Square to create a whole new experience for smart cities.
134 million global visitors per year in Times Square, and we're gonna help it become the epicenter of a brand-new smart entertainment hub and a premium platform for some of the world's most innovative tenants. State-of-the-art connectivity infrastructure with indoor and outdoor 5G and Wi-Fi 6, transforming visitor experiences. This is a great showcase of smart cities. There's a lot more happening in smart cities, but we couldn't miss to show this example, especially as we're here in New York. I'm gonna go to retail. Retail is one sector that is an incredible transformation happening, especially as we think about e-commerce. You have to really reimagine the store at the edge. The retail store today has the ability to apply technology at the edge to fundamentally change the experience. Operations become way more efficient, data available to retailers.
There is a number of technologies from making shelf labels digital with electronic shelf labels, all the way to computer vision for inventory tracking or touchless points of sale. One of the great partnerships we have is with Walmart, and here's a testimony provided by Doug. They are actively pursuing the digital transformation of retail, and they really see a company like Qualcomm with capability across all domains as the key partner that look for Walmart as they become a very advanced technology company for retail. We're very proud. Thank you, Doug, for the incredible partnership. That's just an example of the many opportunities we have in industrial IoT. I know I spend a lot of time telling you about a lot of opportunities. As a recap, we talk about what we're doing in handsets and front-end. We have automotive. We have IoT consumer.
We have IoT networking, and we have IoT industrial. While it looks a lot, we're incredibly focused because it leverages our one technology roadmap. From everything you heard today, there are a number of opportunities that I would like you to take away. Those are some of the most significant opportunities for us in IoT. If I could go back to slide. I accidentally clicked forward. All right, if we can't go back, I'll just talk to it. Number one is the convergence of mobile and PC. That's very significant. The number two opportunity is enabling the Metaverse. The number three is wireless fiber. Then what is happening at the edge based on Industry 4.0.
This will create an incredible opportunity for growth beyond what we're seeing today into the next decade. At the end of the day, we're uniquely positioned to grow across multiple industries, multiple end markets. At the end of my presentation, we're truly going to enable a world when everyone and everything will be intelligent, connected. Hopefully, this was helpful. I know we had a lot of information, but I want you to really take away that we're having a 7x expansion of the addressable market, and we never had so many end market opportunities for Qualcomm as we see today. We're super excited about that, and we're gonna provide it to you, a view, in what that means financially, from our CFO after the lunch break.
Thank you very much for listening to our presentation today, and we look forward to continue the conversation with you after lunch. Thank you.
At this time, we'd like to invite everybody to please enjoy your lunch. We'll give the room an update on when we'll begin the next session in just a few minutes. Enjoy your lunch.
Please welcome Qualcomm's Chief Financial Officer, Akash Palkhiwala.
All right. Hope all of you had a great lunch, or should I call it brunch? We started a little early. This is great. Thanks for attending the show. We are so glad to be doing this in person to meet all of you after a long time. Also people who are online, shout out to you as well. Thank you for joining. I know there are a lot of people who are online, and unfortunately, everyone could not be here in person, but we're super excited to be doing this. Last time we did this was two years ago. I had been CFO of Qualcomm for two weeks, so that was a little rushed.
I'm just very happy that I get a redo today, and hopefully I'll do a better job today than I did then. Very, very exciting time for Qualcomm. Last couple years, you've all been tracking us. We've done great. Really, all of us believe that the best is yet to come. Cristiano and Jim talked through our technology portfolio. They talked through the businesses we have, our overall vision. What I'm gonna try to do next is put a financial framework around all of that. Let's get to it. Let's do this. All right. Through my presentation, I'll talk about these four key priorities. They're the same ones that I talked about two years ago, and we're still very focused on all of them. Revenue growth, diversification, operating discipline, and capital allocation.
These are all extremely important to us. As we execute going forward, these are gonna be front and center for us. There are tremendous opportunities in front of us. We have the intersection of all these trends that Cristiano talked through, and really looking forward to take advantage of all of them. Before that, before we go there, what I'd like to do is give a scorecard against our 2019 Analyst Day performance, the targets we set at that point, and then quickly talk through record results in fiscal 2021. Well, it's great to see a lot of familiar faces in the audience. It's great. I've been seeing all of you on Zoom calls and Teams meetings, so this is awesome. All right. This is the picture of the slide that I presented two years ago.
What I'm very happy to report is that we have significantly exceeded these targets. What's more, we did it one year in advance of when we said we'd do it. It's been fantastic, and this was across all objectives: growth, diversification, fiscal discipline. Also during this time, we did not sacrifice investments in technology. We've continued to make investments where appropriate in select areas, and a lot of what Jim and Cristiano laid out is a result of the investments we made. Today, I'll lay out targets for the next three years at the end of my presentation and through it. The key message we want to convey is we're positioned to drive diversified growth while significantly reducing reliance on any specific customer or product. That's a key objective we have, and you'll see that in the numbers I'll show. All right.
The next three or four slides, I'll quickly go through our fiscal 2021 results. I think they're important because they set the tone for what I'm gonna say after that. Fiscal 2021, a record year for us. Revenues up 55% to $34 billion, and we also guided $10 billion for our December quarter. That's a big quarter for any chip company, any fabless chip company that you've seen. Really, excited about the scale we bring to solving big problems. Operating margin nearly doubled, and we more than doubled EPS. If you think about a company at our scale growing this fast, it's not every day that you see that happen, and we're excited about where we are, but more excited about where we are going. On this slide, I wanna quickly talk through RF front-end, auto, and IoT.
In these three areas, these were very small businesses for us some time ago. In these three areas, we had record revenues in each one of them in fiscal 2021. We grew by more than 50% in each one in the fiscal year. RF front-end, as Cristiano said, we're the largest player in handsets today, and we are positioned to keep growing beyond that. In auto, we have revenues of approximately $1 billion in fiscal 2021, and we think of this as a 10-year revenue growth run rate. We're in a tremendous position based on the announcements and the technology we have, and I'm gonna translate that into what we think it'll happen for us financially.
The last one is IoT, $5.1 billion in fiscal 2021. That is. If you look at our peers and the scale of IoT companies, that's a very large number. What's attractive is all these markets use the technology. The one technology platform that we have created is used across all of these. Our diversification strategy, it's working. If you look at what has happened over the last 2 years to our revenues across RF front-end auto IoT, it's more than doubled. We went from $4.8 billion-$10.2 billion over the last two years, and the growth rate was 1.5 x faster than handsets, which by itself was growing very strongly. If you look at how much of QCT revenues is contributed by these businesses, it's 38%.
It used to be a lot less before, and it's grown over to 38%. We're going the right direction. We're achieving the objectives we wanted to achieve. Also, this diversification is margin accretive. As we talked about earlier, we are reusing the R&D that we have in mobile in all these areas, and that makes it financially attractive to grow in these areas. Okay, operating margin, definitely my favorite chart. In the last two years, we've shown very strong operating margin leverage. I think when we were here the last time, the most number of questions I got was on this topic. It feels gratifying that we executed based on the targets, much better than the targets we had set out. OpEx as a percent of revenue over the last two years has declined from 34% to 23%.
That's 11-point decline, extremely strong. Again, we were able to leverage our technology. At the same time, QCT margins expanded by 14 points from 15% to 29%, greatly surpassing the target we had set two years ago of greater than 20% operating margins. Of course, the gap between the 11% decline in OpEx and 14% increase in EBT margins, this was driven by improvement in our gross margin structure as well. Pretty powerful margin story, operating leverage story across the board. As I mentioned earlier, while we did this, we continued to invest in the key technologies that were needed for us to have a very strong path going forward. In addition, this is an important point that we haven't talked about before, is we've continued to invest in go-to-market channels for auto and IoT.
As we get into all these markets, that's something that we needed to build, and so we got some of those talent from other companies, and we've built the team so that we have a go-to-market channel in those areas. Similarly, we've built the organization within Qualcomm focused on each of these areas so that we can have a dedicated team that's focused on executing in each one of them. Also these metrics, of course, show the benefit of the one technology roadmap that Jim and Cristiano discussed. Over the last couple of years, we have significantly expanded our commitment to ESG as well. We announced a goal for net zero global emissions for Scope 1, 2, and 3 to be zero by 2040, which is defined by Science Based Targets. Qualcomm's heritage is all about low power, right?
The entire business is actually built on, and Jim talked about this in some detail, built on technology that consumes the least amount of power. That's our competitive advantage, very aligned with our ESG objectives. In addition, if you look at 5G is going to drive growth across industries that really help from an energy efficiency perspective as well. Our business is naturally inclined towards some of these initiatives. All right, that was the past. Now we're gonna talk about the future and how we see each of our businesses play out. We have a strong platform for growth. I'll talk through the QCT revenue streams, each one of them that Cristiano addressed. I'll talk through stability in QTL. As Cristiano said, there's a tremendous expansion of our addressable market, and I'll cover that in our three-year financial forecast.
Let me start with the 5G unit forecast. All right. So this is gonna be a very long-term driver for us. I think there's a conversation about 5G being a very short-term driver, and that's a very handset-centric view. I think if you step back and look at the various opportunities that Jim and Cristiano outlined, this is a very long-term driver for us. Let me break it into two parts. Let me first talk about handsets and non-handsets. Before we look at the forecast, the key premise to remember is we are the leader in 5G. We invested early. We accelerated 5G deployment, working with our partners by one year. There's a very dense roadmap. We're gonna keep investing, and we're going to lead in 5G solutions. It's a key priority for us going forward.
If you look at handsets, it is playing out exactly as we had laid out two years ago. If you look at our forecast back then, if you look at our forecast now, it's playing out exactly in line. We gave that forecast even before the initial 5G networks were deployed. As you can see, we are only 525 million out of 1.3 or 1.4 billion of handset market, and there is a long way to go within handsets. Within handsets, we see this as a three to four year curve, and we have a forecast that you can see there. In non-handsets, 5G is not the same as 4G. 5G was designed for low latency, high data rates, high security.
These are qualities that are required in a lot of the applications that will require 5G. That is why we believe that 5G is going to have a very strong long-term growth outside handsets. You should think of this as a 10-year curve with the scale eventually being similar to handsets. Okay, the next three slides, I'm gonna talk about the handset business, then go to RF front-end, auto, and IoT. In the handset business, I'll start with what happened over the last year. The SAM over the last year expanded from $19 billion to $33 billion, and there were three key drivers for this. First is 4G going to 5G. When 4G goes to 5G, a lot more content is needed, and we expand the revenue we get per device. Second is increased capability of Snapdragon chipsets.
As Jim outlined, between camera, video, audio, GPU, CPU, every single vector is being pushed to the limit in terms of device performance, and that benefits our ASP. The third driver, which is the most important one for that this year, is we saw SAM expansion from the change in the OEM landscape. This means there was a share shift in China where our customers, OPPO, Vivo, Xiaomi, Honor, they all picked up share and they became bigger. A lot of the share that they picked up was in premium and high tiers, which is where Qualcomm is very strong. It was a very attractive expansion of our SAM. The key takeaway for this page is that the SAM expansion that we saw because of this is significantly larger than the modem-only opportunity, than the Apple opportunity.
It puts us in a very fortunate position as we go forward. As we've said in the short term, as our supply improves, the benefit of the improvement is reflected in our first quarter guidance, and we expect that to benefit the rest of fiscal year as well. Okay, I think Cristiano made this point briefly. I wanted to reiterate it and talk about the scale of Android. If you look at the scale of Android accounts for 85% of the total worldwide phones sold. It's a very large number. 15%, of course, is Apple. 85% is Android. When you take that number and you convert it into addressable market for us, 87% of our addressable market is Android.
That is why, as Cristiano said, our primary strategy in handsets is to be the platform of choice for premium and high-tier Android devices. That's where the market is as well. As the SAM continues to grow, the ecosystem innovations in Android will continue as well, and that will expand our SAM even when the total unit curve matures. Then I'll finish with just a data point that we gave at our earnings as well. In fiscal 2021, our Android revenues was 40% larger than our largest customer. Competitor, sorry. With that, I'm transitioning now to the forecast for handsets. Over the next three years, we expect a CAGR, SAM CAGR of 12% in handsets. This is between 2021 and 2024.
The drivers for this are the same as the ones you've seen before, which is you have 4G to 5G, and you have increased Snapdragon content. Our forecast for revenue is that it will grow at least in line with total SAM. The assumption we use for this forecast is that for Apple 2023 launch, our share is down to 20%. This is the planning assumption for the forecast purposes. Just to be clear, there is no new data point that makes us do this forecast versus our discussions in the past. We just wanted to set a base for this forecast, and so we've used that as a planning assumption. Let me step back and make sure you understand our leadership in 5G as we talk about this assumption. We started early investing in 5G.
We are clearly the leader. If you want to launch a global phone, we are going to have the best modem, and we're gonna be significantly ahead as we go from Release 16, 17, 18 that Jim talked about. We're going to offer that product to all of our customers, so you could have higher share, but the forecast that I'm giving you today is based on a 20% share of iPhone 2023. Anything better than that would be upside. If you look at just Android revenue, we expect that to grow significantly faster than SAM. Finally, the last point that Cristiano talked about, we have commitments from our Android OEMs, right? OPPO, Vivo, Xiaomi, Honor, Samsung, we have commitments from them that Cristiano outlined, and so that gives us the confidence in the forecast we are providing today.
With that, I'll transition over to RF front-end. RF front-end success has been tremendous for us over the last several years. As you know, we are at $4.2 billion now for RF front-end revenue, and we are the largest player in handsets. We are not done. We are poised for growth going forward. The reason for that is we have the broadest product portfolio, and the modem to antenna development that we're going to do is gonna be a competitive advantage. People who only have the modem are not able to do that. People who only have RF front-end are not able to do that. We are gonna have the best individual components and a system advantage in bringing the two together. Within handsets, and I'll talk through growth opportunity in each of these areas.
Within handsets, there are three key drivers for growth for RF front-end. First one, 4G to 5G transition. As I mentioned earlier, we are only partway through the transition. There's a long way to go. As that transition continues, the RF front-end opportunity for us continues as well. Second is OEM landscape shift. As that shift happened, a larger portion of the market became available to us for RF front-end solutions. Then finally, millimeter wave. As millimeter wave gets broadly deployed, we'll have an opportunity to expand our revenues as well. Three very strong drivers for growth within handsets. We'll take our RF front-end products and apply it to auto and IoT. We'll have an opportunity to grow in both of those areas. In auto, as everyone knows well, all cars are getting connected. 5G is gonna be in all cars.
As that happens, as 5G gets attached to all cars, we will be able to add RF front-end to our solution. The total amount of content available per car in certain cases could be as high as $30, so it's a very large number, and we're happy to provide a new disclosure today that our design win pipeline for RF front-end and auto is already greater than $600 million. Going over to IoT, there are two inflection points for the RF front-end business in IoT. First is the adoption of 5G. Cristiano talked about various places where 5G would be used in IoT. As 5G gets used, you should assume our RF front-end gets used with it. Second is Wi-Fi 7. That'll be the second inflection point.
When Wi-Fi 7 gets deployed late 2022, early 2023, we're going to make sure that our RF front-end solution is available for Wi-Fi 7. That, just like 5G was the entry point for Wi-Fi 7, for our RF front-end, Wi-Fi 7 will be the entry point for RF front-end in Wi-Fi. Now looking at the forward growth forecast, very similar approach to handsets. We see the SAM growth rate consistent with handsets at 12%. It's a combination of two things, 4G to 5G, and then auto and IoT coming in. As I said earlier, modem to antenna, we expect to have a technology advantage as we go into these markets and going forward. We're forecasting similar to handsets, total revenues to grow at least in line with SAM and non-Apple revenues to grow much faster than SAM.
Similar to the customer commitments we have on the chipset, we also have customer commitments on RF front-end. Cristiano did not cover this. This is an update to what he said, and that is reflected in our forecast. That's what gives us confidence that we'll be able to achieve this plan. All right, now I'll transition over to auto. This is the same slide that Cristiano showed. There is a tremendous opportunity for us to sell a system rather than individual components. This is an advantage we have in mobile, selling a system. This is an advantage we're gonna have in auto, selling a system. That's a tremendous benefit when you go to a customer. Arriver acquisition, Veoneer, and the software stack part of Veoneer that we're going to keep is Arriver. That is a key part of our strategy going forward.
With that, we're gonna have an integrated software plus hardware solution for ADAS, and that's consistent with the win we announced today with BMW. Here's our forecast for auto. We expect the SAM to expand from $3 billion to $15 billion over the next few years. The drivers are pretty simple. It's digital cockpit and telematics both expanding, and now we've added ADAS as an incremental opportunity for us. If you think about ADAS is moving towards level two plus, level three. As more and more vehicles go there, the opportunity in ADAS will continue to expand. We're very excited to have just a tremendous product line that goes from the lowest end car to the highest end car, addresses the entire spectrum. As we look at our design win pipeline, last time we gave you an update was three months ago. We had $10 billion.
We're at $13 billion . Design win pipeline has expanded to $13 billion. This does not include the RF front-end that I mentioned on the other page. $13 billion design win pipeline. From a revenue forecast perspective, we expect to grow from $1 billion in fiscal 2021 to $3.5 billion in two years, in five years, and then 10 years of $8 billion. The reason why we have confidence in this forecast is, of course, our product roadmap, customer relationships. Over the next five years, 70% of our revenue forecast is covered in existing design wins. This is cumulative revenue over the next five years, 70% of that is covered in existing design wins. We expect several OEMs to make decisions on what platforms they pick over the next 24 months.
We're in a great spot with the product portfolio that we have to be able to win a significant portion of it. As we look forward, we're in a position to expand our design win pipeline. Okay. Next, I'm gonna transition over to IoT. Cristiano talked you through all of our IoT categories, so I'm not going to talk through it again. The key points I wanna make on this page is we have a diversified product portfolio and revenue base. Total revenue of $5.1 billion in fiscal 2021 that's spread over nearly 14,000 customers. The top 10 customers is 42% of our revenues. It's a very long tail. As we get into all the areas Cristiano outlined, the tail will get longer.
This is a business that will continue to be diversified as we move forward. Similar to the discussion earlier, we still have one technology portfolio, right? One technology roadmap that we're gonna be using across all these opportunities. Financially, that's obviously very attractive to us. The last point on this slide is just revenue split across these three. You should think of each one contributing a very significant portion of the revenue that adds up to the $5.1 billion. When we talk about growth rates on the next page, all three of them are growing at strong rates as well. This is not a scenario where we are concentrated in one area. It really is diversified across the three buckets with strong growth rates for each one of them.
From a forecast perspective, we see the SAM in IoT expanding at a CAGR of 17%. Very strong growth over the next three years. This obviously is driven by digital transformation, cloud computing, creating a demand for our technologies, connectivity, low power compute, and Edge AI. Everything that Cristiano went through applies to this page, and that's what's driving the demand for our solutions. The strong momentum that we have in IoT exiting fiscal 2021, we expect that to continue and revenues to grow up to $9 billion by fiscal 2024. This is growth faster than SAM. Really, this is a question of how quickly can we grow and take advantage of the opportunities in front of us. As I said earlier, this IoT is more of a 10-year opportunity for us, not just a three-year strong growth rate.
All right, going over to QTL now. QTL, our forecast is we're going to maintain the current revenue scale and margin profile. QTL is the most successful licensing business in our industry. Over the last few years, through litigation, regulatory challenges, and FTC decision, that has validated our business model. That is a position of strength as we look forward. Within handsets, greater than 90% of our QTL revenues come from handsets, and all the major OEMs are licensed for a long period of time. The one key thing to note here is all OEMs are obligated to pay royalties independent of a chip relationship. You should think of those as distinct things. Going to non-handsets, we have an established program in auto and IoT.
As more and more cars get connected, especially as 4G goes to 5G, it's a tremendous opportunity for growth for us. In IoT, the way we approach that market is through a module licensing program. If you're making a module, you're licensing the module, and that's how we get paid. Again, IoT is a significant growth opportunity in the very long term for us. All right. If you go beyond the three-year forecast, I wanna step back and talk through the slide that I didn't skip, Cristiano ended up skipping over and provide some financial insights on it. We have opportunities for transformative growth because of the 7x TAM expansion over the next decade. This goes beyond kind of the three-year forecast that we're discussing. There are several of these opportunities, but I wanna highlight four today. First is PCs.
As you all know well, this is a very large unit and revenue pool market that exists today. Over $35 billion in revenue available. As this market transitions to ARM, and we have the next gen CPU developed by the NUVIA team, we combine that with the Snapdragon Technologies, all the technologies that Jim talked about, and make it available in an integrated SoC. We have an opportunity to be a significant player in this market. We're excited about what it means for us in the long term. Second is XR. As Cristiano said, there are more than 50 devices that have already been launched with our chipset in it. We have approximately 90% share of chipset design wins. So if you're convinced this is the next computing platform, and the market could be very large over a period of time, it's a tremendous opportunity for us.
There's obviously a lot of variability on how this market happens, how quickly. Regardless of what your view is on the scale or the pace of that change, as the market grows, we're gonna benefit from it. Third is Industry 4.0. Because of the connectivity to the cloud, as everything in various industries gets connected, this is gonna play out over the next 10 years, and there's tremendous opportunity for us to leverage mobile technology and sell into these markets. Finally, ADAS. I've already laid out the financial forecast for the total auto business, so I won't repeat that. This is a market that is expanding very quickly with a lot of demand for technology, and we like the cards we have, especially with the Arriver transaction. We're positioned to benefit from it.
Overall, we're kind of at this intersection of several trends with the right technology roadmap, which will help us drive long-term growth beyond three years. All right, on the M&A side, our historical model has been very simple. We buy companies and make them number one. If you look at the history, and I'll pick TDK of course as an example, we bought the RF front-end asset from them. They were the number four, number five player, and now we're the leader in handsets. We've done that for several of these technologies. As you know, NUVIA, we just closed, and we are in the process of acquiring Arriver, and we see that fitting into the same model. Finally, if you look at going forward, our strategy remains unchanged. We'll acquire small teams or technologies.
We'll acquire businesses that accelerate our internal growth strategy, and large acquisitions will be opportunistic. We'll look at them, but we don't see it as an integral part, required part of our strategy. Okay, capital return. As all of you know very well, very strong capital return program. Over the last five years, we've returned $47 billion while still keeping a very strong balance sheet. Over the last two years, we have returned 90% of our free cash flow. Our forward strategy is very consistent with what we have done over the last two years. We're gonna grow dividends. We're going to buy anti-dilute, do anti-dilutive buybacks, and opportunistically consider incremental buybacks. We're committed to a strong balance sheet and our current debt ratings. Before I conclude, I wanted to summarize all the financial guidance we gave on different pages.
When you put it all together, we have an opportunity to be a company with revenue greater than $46 billion. Just looking at QCT, the guidance we gave implies mid-teens revenue CAGR over the last three years with the planning assumption that our share of iPhone 2023 is down at 20%. With this assumption, Apple revenue as a percent of QCT will be low single digits exiting fiscal 2024. We're guiding QCT operating margins of 30% +. I already talked through QTL, very similar to what I said earlier. OpEx as a percent of revenue, we're currently at 23%. We're guiding 21%-23% as the range going forward. That's the summary of all of our guidance points. I mean, needless to say, this is a tremendous growth for any company, especially someone our scale.
We're in a very good position, even with reducing reliance on a specific customer. As I conclude, I wanna quickly highlight some key takeaways. Cloud adoption is accelerating demand for intelligent connected edge. That's who we are. We're positioned to drive diversified revenue growth going forward. One technology roadmap across all growth opportunities. Prioritized, disciplined, and focused execution, and then finally, drive strong free cash flow and capital returns. This concludes my presentation. We will now go to Q&A. Thank you.
Please welcome to the stage Cristiano, Jim, Akash, and Alex Rogers, President, QTL and Global Affairs.
All right.
Hi, it's Timothy Arcuri at UBS. Thanks. Jim, you showed a slide. You showed a 10x incremental opportunity per car in autos. I was wondering, I had two questions. One, can you put numbers around that? Like, what's the baseline today and you know, can you give us some numbers on that? Secondly, what's the sort of front-end versus back-end monetization opportunity? Is there a significant back-end monetization? You know, one of your peers is talking about getting half of the you know, back-end monetization for over the year, downloads, and I'm wondering if that's contemplated in that you know, number as well. Thanks.
You wanna take that?
Sure. If you really look at cars, it's like phones. The variance of price bands or total silicon opportunity across a low-tier car versus a premium tier car, there's a very large gap across. Thinking of $100 as an average opportunity today and some multiple of that tomorrow, depending on which tier you're in, is a reasonable framework to think about it. Front-end versus back-end, I think that's just a business model question. You could monetize it multiple different ways. To me, that's not really what portion of the value do you get, it's just how you price it. I think you'll see a variance across various players in the industry for that.
Hi. Thanks for the presentations. It's Kevin Cassidy from Rosenblatt Securities. Wondering if you could give us a little more detail about the new NUVIA designers. Are they gonna be focused just on the ARM portion of the Snapdragon, or do they go to the GPU and the AI portions also?
If you want.
Maybe I'll start that. We have in that asset that team is developing our next generation CPU, and there's incredible scalability. Our very first commercial deployment of the CPU is for our next generation PC computing part. Then that's gonna be taken to automotive. It's gonna scale to the Snapdragon premium, you know, Android flagship. We're looking at the data center opportunistically. We're very focused on the edge. We have an asset that can go to the data center, but that's more of opportunistic. We're really focused on bringing that for PCs, for automotive, and for the Snapdragon premium tier device.
Thank you. Srini Pajjuri from SMBC Nikko Securities. Couple of questions. Great presentations, by the way. First, on the margins, Akash, you said your target is to maintain greater than 30%. Just wondering how you think about growth versus margins, especially given your scale. Is there an upper limit that we should think about when it comes to your growth margins and operating margins? I'm gonna ask a second question as well.
Sure.
QTL, one of the, I guess, struggles for us is, you know, how to think about the terminal value of QTL. Obviously the duration of these contracts and duration of this IP portfolio is not public. So if you could give us some framework about how we should think about the potential terminal value. I know you said it's gonna be stable in handsets, but is it sustainable longer term? I guess that's the question that most investors wanna understand. Thank you.
Maybe I can take the first one, you wanna take the second one.
Yeah. Sustainability.
On operating margins, as you can see, we made tremendous progress over the last two years, right? We've gone from 15% to 29%. We'd set a target of 20%+ two years ago, and we've beat it by a wide margin. Super happy with that. We think a prudent target to set going forward is 30% +. I mean, this is a, I think, a reasonable way to operate the company going forward, and we might have opportunities to go higher than that. The real question is, what we're trying to contemplate in our guidance is a normalized supply chain environment as well. We think at this point, this is a reasonable guidance point.
On the QTL side, first starting with the commercial side of the question, if you look at our contracts now, we're in a cadence of about five to 10 years for your typical SEP agreement, Standard Essential Patent License agreement. They typically cover all versions of cellular standards, 3G, 4G, now 5G. For QTL looking forward, as I said, we have every major handset OEM signed up to long-term agreements in that range that I just discussed. The first renewal is not until fiscal year 2024. We have a good position on the commercial side.
In terms of the IP portfolio, if you looked at the slide that Jim presented, and you look at the progression of 5G from the first release and all the kind of foundational innovation in Release 15, and then you look at Release 16 and then on to 17 and 18, the portfolio is really foundational to everything you're seeing on that slide that was articulated. Qualcomm's timeline leading up to those releases in 5G is a timeline that starts 10 years earlier. Qualcomm is really positioned, I think, better than any company in the industry because we continue to innovate through every successive generation. We're always focused on what's going to move the industry forward in the most effective way from a technology perspective.
The standards actually benefit from the key innovations that Qualcomm has driven into the standards. Because we're able to scale that technology worldwide, I think as Cristiano or Jim mentioned, we're actually able to bring that technology into the market in a way that's successful. That adds value to the intellectual property that's associated with all of that innovation. Now, it's a very large portfolio, and the portfolio has migrated over time to being primarily focused on cellular standard essential patents. It's very heavily weighted for what's driving the QTL revenue. That's the way we're planning on managing the portfolio and the business going forward.
Thanks, guys. It's Matt Ramsay from Cowen. One quick clarification. Akash, I think I've asked you on the last, I don't know, five or six earnings calls about operating margin in QCT, and the progress has been remarkable. You guys talked about 30% going forward as a baseline. Would that hold even if the Apple revenue goes to single digits of QCT as you've laid out?
Yeah. The margin guidance contemplates all the other assumptions, so yes.
Okay, perfect. My second question, Cristiano, we've been over the last 10 years down this path of Windows on ARM and the notebook market. I think it's crazy that every tablet I buy and put a keyboard on, I can get on the internet through 4G or 5G, and none of the notebooks do. So what's different this time? Could you calibrate for us a bit the size of the or the ambition that Qualcomm has in the notebook market in terms of market share or dollars or whatever metrics you wanna give? Thank you.
Look, it's a great question. Thank you, Matt. Well, it's different. It's not different than what we'd expect to happen, but it's different this time because the architecture is changing. If you look at some of the fastest, you know, PCs out there right now, it's provided based on an ARM architecture. The bringing an SoC to the PC market is very disruptive. If you look what Apple has done with the M series, on the other hand, it even has the performance of an RTX 3080 discrete graphics. So the transition to that architecture is happening. As Apple switched to ARM, we're seeing also the developer ecosystem switching. You know, you look at some of the companies like Adobe and part of their software development is this ARM first. So that is different.
There's another thing that is different. The change in productivity, work from anywhere, if you describe it that way, it did change some of the attributes of what is the computing device that is required and accelerated what we call the mobile conversions for PC. Those two things are really creating an accelerated transition to a new architecture, and I would argue we're probably the best positioned company to do this on the Windows side. Now, the second part of the answer is, we knew it would be a road to get there. That's why we started earlier with Windows on Snapdragon. I think that was, not only work for us, work for Microsoft. Now with Windows 11, Microsoft has the full, 32- and 64-bit capability on ARM.
Connected with the fact that we're now gonna have the technologies that scale on both the CPU and the GPU side. We're actively working to build, as we said, the benchmark of performance for this new architecture of PC or Windows. The answer to your question, if you look at 250 million devices right now on the PC, you can model. I don't wanna make share projections, but you can model even if we get a smaller share is a significant revenue opportunity for Qualcomm. As you think about the use case changing with streaming of gaming, and on-demand computing. I wanna explain the on-demand computing. I think sometimes this is not easily understood.
We demonstrated right at the beginning of 2020, 8K video editing on Adobe on a Lenovo laptop connected with 5G completely transparent to you. If you have a full workstation for computer-aided design video editing, you are not gonna take that workstation with you when you have mobility. That is gonna also pull 5G to the PC. It's an incredible opportunity for Qualcomm.
Hi, Samik from JP Morgan. Thanks for taking the question. Just one on the RF front-end opportunity that you've talked about. I'm a bit surprised to see that you're guiding to grow in line with the SAM, just given the track record that you have of share gains. If you can just dig into that a bit deeper as to what are the offsets there, and why shouldn't we think that's a conservative number? Just in relation to RF, but the non-smartphone markets, what are you seeing in terms of the competition or competitive landscape there related to smartphones?
Yeah. On the first one, as we said, non-Apple, we expect to grow significantly faster than market. You're what you said is accurate. The offset is just the assumption that we're making on what share of Apple we're gonna have. We as you know, participate on the millimeter wave side at Apple. That's the driver. Remind me of the second question again. Yeah. Yeah. If you think about IoT, 5G in auto or IoT, right? It's the same people we compete with in phones. You should think of it as just the same way we succeeded in phones. We'll succeed in those places. That's just extension of the 5G market from phones going into those areas and very similar landscape.
Wi-Fi, today we don't compete in it, and it's mostly driven by the same competitors. As we develop Wi-Fi end-to-end similar to 5G, that's gonna create an advantage for us.
Hi, it's Joe Moore from Morgan Stanley. I wanted to ask about the growth in Android SoC, to the extent that you're talking about more than 12% CAGR for three years. You know, the 200 million incremental 5G units are below the premium tier, I would think. You're coming out of a shortage, so there might be a little bit more competition. And some of the Huawei benefit already played out. Like, what gives you the confidence to project that kind of growth, you know, that far forward?
Yeah. As we mentioned in our presentation, Cristiano outlined, we have two things. One is design win pipeline. We have insight into what's going to happen through all of 2022. Second is, as we mentioned, we have commitments from customers that gives us confidence as well. It's really a combination of those factors. Of course, the roadmap's great. There's all the reason, especially in China with Snapdragon branding, for OEMs to use us, and supply has been the constraint that we think that opens up.
Hi, over here. Gary Mobley, Wells Fargo Securities. Thanks for hosting this event. I wanted to ask a question about your RFFE success. It's you've been a clear winner in your baseband to antenna strategy. However, I'm curious to know if you've done any analysis on your design win traction or the revenue generated outside of your baseband market share footprint. I guess related to that, you know, I guess you already addressed it a bit, you know, Apple is presumably gonna move to some different variety or some combination of somebody else's RF. That's part of your assumption, correct? Just clarification there. Thank you.
No. Thanks for the question. Let me address that. The first one is, we are winning design not only on the modem to antenna, but also winning on the discrete side. For example, some of our largest customers that choose to have both, Snapdragon in their devices as well as their own silicon, we're now winning our front-end components across the entire lineup of devices. One of the fastest growth areas we had over the few quarters has been the expansion of the modem to antenna into discrete. Having said that, the modem to antenna, the system level, because it saves R&D, because it's validated, it has performance benefits, is gonna continue to be the key driver of growth.
As far as Apple, look, I said many times before, as we continue to invest in leading our front-end technology, there's opportunity to supply to Apple. As it relates to Apple, we don't make any assumptions other than the assumptions we have in the plan and what we're doing with them on a millimeter wave. Everything on Apple, you should think of it upside.
Hi, Ivan Feinseth, Tigress Financial Partners. Very great event, and thank you for taking my question.
Thank you.
Thank you.
Can you give me the range of functionality that you contribute to an automobile? Like, what would be the minimum functionality and the maximum functionality, and then the cost range, and then how you integrate with, like, software updatability?
For automobiles?
For the Snapdragon Ride, for ADAS systems.
Happy to take that. One of the differentiations of Qualcomm, the Snapdragon Ride platform that we're getting very positive feedback from OEMs is the ability to provide scalability. We have this view that we should be thinking of ADAS. It should be as pervasive as ABS and airbags. Why not bring it to every car? We had took the same approach that we took to mobile, which is creating a multi-tier platform. We're doing that with ADAS as well with multiple SKUs. I wanted to emphasize when we talk about the whole platform having not only the Snapdragon Ride, but the connectivity, the service platform.
One of the services as part of the service platform was the Car-to-Cloud, the software SKU we're getting traction, which will allow OEMs to also upgrade from the minimum functionality to higher level functionality, selling directly to the consumer. That's gonna be commercialized within the coming quarters with one of our customers. The idea is to look at a multi-tier and create a business model that can actually upsell functionality at the lower levels, while we continue to do the full functionality for premium and high automobiles.
What would be the range in revenue to you from the minimum that, let's say, an OEM would adopt to the maximum?
The range of-
Revenue per car.
We did say around the 10x revenue opportunity overall for the whole Digital Chassis. You know, we kind of thought about putting a dollar amount, but you know, you can think about it. It can start in the multiple hundreds of dollars and you have a significant upside opportunity. At the end of the day, we have a lot of silicon opportunity in the car across all of those different domains.
I'd suggest that you just look at our overall revenue forecast, because I think being able to do math across tiers and ASPs is a little more difficult. There's not a direct answer to your question, but that's contemplated in the guidance we gave you.
Like Mary Barra spoke about the ongoing subscription service to upgrades and connectivity, would you participate, or do you provide a software upgrade kit to the OEM and then the ongoing subscription to the end user that pays either as included in the service or as an annual or point of time subscription upgrade?
Yes, some of those capabilities for our technologies enabled by our Car-to-Cloud service platform.
Thank you.
Thanks. Taking the question, Blayne Curtis at Barclays. Just wanna ask on the auto side, you did extend the TAM out to 26 to capture ADAS. So just kind of curious how to think about more in line with the other segments in the three-year TAM. Just trying to figure out how materially you see ADAS. I think the old TAM was growing at 12%. You've been doing much better than that. How do you think of auto for a shorter duration?
Yeah. The typical timeline on a design win and when a car launches is three to four years. The way you should think about ADAS is really a lot of the design wins today don't necessarily hit much in the 3-year forecast. It's really a tailwind beyond that for us or for anyone else.
Thanks. It's Brett Simpson at Arete. I just had a multi-part question on autos. The BMW deal, can you maybe just talk about. Does that flow through the whole fleet of BMW? Is it parts of the portfolio? Can you just give us a sense for the scale of that agreement and when it might start to kick in? The second question on autos is really about, I think in your slide deck, you talked about 70% of cars will be connected by 2024. What portion would be 5G, and what type of market share do you think Qualcomm can attain within that space? Then I have a follow-up for QTL.
Maybe I'll take that.
Yes.
You wanna take the first one?
Yes. It is really a technology platform. That's why I keep referring to the Digital Chassis. It can be applied across a number of cars within the BMW family. It's part of. It's really a technology design win. I think to your question on, if I understand it's 5G market share in automotive? Yeah. Our market share in automotive telematics is very high right now. It is. That could be a good proxy how we think about 5G. One thing that we like about opportunities beyond handsets. Jim Thompson, in his presentation, talked about we're one of the few companies that really enable the globe. When you're building some of those non-cellular devices, sometimes you don't really do regional SKUs. You just have one platform.
You want to future-proof that design as well because you're gonna have new frequency bands, or you're gonna have new versions of the standard while by the time you launch the car into production. That speaks to our strength in connectivity. As a result, it's a good proxy that we're gonna have high share when we think about 5G in automobile.
Essentially, it's a software upgrade over the life of the car. The modem keeps improving, and you get paid a follow on for operating the modems.
That's one of the concepts. It's a common concept that you started to see and why we build a Car-to-Cloud service platform. It's not only about connectivity, it's also about features in the digital cockpit or even different tiers of ADAS. That's the reason I think this is actually good, not only for Qualcomm, but also for our OEM partners as well.
Maybe just finally on QTL, you talked about the new program for autos and the IoT for licensing. I guess historically, you've always been part of patent pools, like the Avanci patent pool, and there's not a lot of revenue from those types of agreements is my understanding. Can you maybe just talk a little bit about, you know, the licensing opportunity in the car? Would it be similar levels to a smartphone? Do you think that the value you're adding is a lot higher? Just help us maybe with sizing that opportunity. Thanks.
Just put Avanci aside for a second. Qualcomm has been licensing independently in the automotive space for a long time. I think when GM had OnStar, you know, generations of cellular bill. We have literally hundreds of millions of licensed connected automobiles, connected units on the road. What we've done with 5G is we went ahead and transitioned from a percentage-based royalty to a dollar-based royalty. We're licensing at $5 per connected unit or connected vehicle. We went to market with that. We went public with that number, but we went to market with that some years ago to make sure that we were
Bringing our licensing program to the market for 5G in a way that was useful for the auto OEM so that they can plan their adoption of the technology in the units has been very successful. We have dozens of 5G agreements already signed up. We think that program will be successful going forward. On the IoT side, we've been licensing at the module level for quite some time. As Akash mentioned, the IoT curve is a longer curve. We talk about a 10-year curve. We have 3G, 4G license units out there. We have all the major module manufacturers licensed. We hope to see pickup on the IoT side from the QTL business as that 5G gets adopted across that curve.
Now, in the handset space, obviously the 5G adoption has happened much more quickly. We're looking at the IoT space optimistically as well.
Hey, guys. First of all, thank you for an extremely informative analyst day, and congratulations on the last 12 months of pretty incredible results. I had two questions. A lot of your growth going forward is hindered on Android. You talked about a two-year sort of understanding, deal, contract, whatever you wanna call it, with the Chinese guys. I think was Samsung a part of that as well, or maybe I'm mistaken? Could you give us some color on what that means. Is it inked? Is it an understanding? Is it a partnership? I have a follow-up.
Yeah. With the Chinese OEMs, it's an agreement that we have in place and which gives us a high level of confidence. I'll just stay at that level in terms of disclosure. For Samsung, as we said, we have high confidence in 2022 based on the design wins and conversations with them as well. That gives us confidence for the next year for Samsung.
A follow-up quick question on the edge opportunity. Could you help us think about a framework for competition? We typically see microcontrollers as something on the edge compute, but we don't see a lot of companies targeting it with sort of SoCs that you have. Is that a fair statement at this point in time? If you do see somebody, who is it that you see?
I'll take that question. It is a fair statement. What is really creating demand for Qualcomm technology, it's really what is happening in IoT as part of a much broader digital transformation that is driving a lot of growth of the cloud. It's also, we have been very focused on the opportunities that require the value of high-performance computing and on-device artificial intelligence. What makes Qualcomm unique is we actually have every single technology in-house. We control the roadmap, and we have the ability to quickly scale, as Jim said, from watch and an earbud all the way to, you know, a high-performance computing on the trunk of a car. What is also unique about Qualcomm versus other company is we have a direct relationship with the large cloud providers, the hyperscalers, and the large enterprises.
Think about Qualcomm being a company that is not dealing with the traditional, when you think about the space, with distributors and like a catalog-type business. It's really about transformative, you know, edge solutions. That's why what you see different of Qualcomm compared to some other companies is the direct relationship as, hopefully, I brought a few examples with some testimonials and quotes with some of the large enterprises, and I think that what makes it very unique.
Chris Rolland, Susquehanna. Thanks so much for hosting this day. My first question is for Akash and Cristiano. Maybe you can chime in here too. I just want to make sure that I understood the guidance correctly. That's better than 12% for handset, and that includes the wind down of Apple in that number?
Yes.
Okay.
The 12% includes the assumption on Apple.
Great. I think that takes us to high teens then, on that rate, ex Apple for you guys. During this time, 5G is maturing, pricing is maturing. I'm wondering if you can bridge us to that high teens percent. Is it all RFFE? Is it that move from 4G to 5G pricing? What accelerates that, bridges that, pretty big gap?
Sure. Overall, as I said, on the QCT level, we are seeing mid-teens is kind of the consolidated guidance, as I mentioned on my conclusion page. There are two key drivers within handsets. First is 4G going to 5G. As we've said in the past, we've typically, when that happens, we see a 1.5x increase for a like-to-like device. You take a high-tier 4G phone, high-tier 5G phone, a 1.5x increase in our revenue opportunity. A portion of it is RFFE, but the rest is chipset, so that's driving an increase.
Second, there is just we're continuing to see tremendous demand from the OEMs to increase the capabilities on camera, CPU, GPU, video, audio, all these features, and that increases the Snapdragon content that's available as well. Of course, Moore's Law plays into it as we go to advanced nodes. It's a combination of these factors that increases our monetization for a device.
Yeah, look, I'd like to chime in on that question because it's exactly important, and I wanna use this opportunity to make a clarification. I think we're also making a lot of clarifications during this day how we think about growth in mobile. When you look what is driving ASP, you know, of course, I think we're highly differentiated in 5G. You know, millimeter wave is a key differentiation of Qualcomm. But even as you think of 5G gets more mature, and we continue to see the 5G plus our front end, we've been talking in the past about the 1.5 multiplier that stays in place.
What is really driving ASP is the demand for much higher artificial intelligence computing, much bigger GPU, much faster CPU, and that's why we're bringing the next generation CPU design by the NUVIA team all the way to the mobile Snapdragon, and some of the new use cases we're seeing happening with 5G. It's very well-balanced. The other thing is the consolidation of Snapdragon brand as a must-have for premium Android, and especially as with the changing in the OEM landscape that had occur, some of our traditional customers like Xiaomi, OPPO, and Vivo, they're climbing up to the premium tier, and the Snapdragon brand is helping them to achieve that also globally. I remind everyone that Xiaomi is now number one in Europe.
I think all of those things are really great contributors to the QCT growth margin and growth.
We have time for one more question.
Yeah, thanks for fitting me in. It's Rod Hall with Goldman Sachs. I just wanted to ask two questions. One, back to the Apple exposure, Akash. Thanks, by the way, for the conservative estimate on 2023. Just, we're getting a lot of questions, we continue to get a lot of questions about what the current exposure is, and everybody's done their calculations. We would calculate about 16% QCT revenue exposure, so QCT only, not the royalty part, being 16% of revenues. And probably, and this is the tougher part, around 12% of EBT. That's the calculation we do. I just wanted to check with you and see if those kinds of numbers are in the ballpark.
I mean, as you know, we haven't provided a specific disclosure on what % of our business is them today. If you're looking at their overall scale and you're using kind of ASP numbers that aligns with our 1.5x and how 4G played out, that would be a reasonable way of thinking about it.
Okay. Okay, thanks for that. The second thing I wanted to ask is, in your, I think I'm right to say that in your IoT SAM, you're including the compute opportunity, the ARM-based NUVIA compute opportunity. Can you say when in that trajectory you might expect first revenue from those products? Or give us any kind of an indication of how that fits into that SAM, maybe what the, you know, proportion of the SAM is, if you don't wanna say what the timing is?
Yeah. Within kind of the next three years, we are including some benefit from the compute opportunity, but there is no heroic assumption in the forecast we outlined. Then I think on NUVIA, as we said, we'd be sampling the parts late in 2022 and commercial launches in 2023. Yeah. Just the other thing to note is just we already have chips that still serve that market, right? That is just when the CPU intersection happens, but we'll have products much before that.
I understand we've run out of time. I just wanna say a big thank you to all of you that came in person, really appreciate. Thank you for listening to us. Also the folks that follow the stream online, thank you so much. We're super excited about the future of this company. I think we have a great opportunity ahead, and we'll continue to be focused in delivering return to our shareholders. Thank you so much.