Intel Developer Forum, Spring 2003

Anand Chandrasekher, Vice President, GM, Mobile Platforms Group
Gadi Singer, Vice President, Wireless Communications and Computing Group; GM, PCA Components Group
Louis Burns, Vice President, GM, Desktop Platforms Group
San Jose, Calif.
Feb. 19, 2003

ANAND CHANDRASEKHER: Good morning. We're going to talk this morning about convergence in the clients. Our vision here is really simple. We believe all computing devices will communicate and all communication devices will compute. This is the vision that we've been talking about for the last two IDFs, and in particular, the last one, we spent quite a bit of time talking about the standards that are necessary to facilitate this vision becoming reality.

Six months ago, on this stage, Intel's president and COO, Paul Otellini, talked about the need for software to also make this vision a reality, software that spans across the range of devices to make this computing environment seamless. And Craig Barrett reinforced that yesterday. I want to give an update on that.

Software is critical to enable the goal of convergence of computing and communications. We believe that with our tools, we need to be able to enable applications to be written once and run best on all Intel processors, whether they are processors for the handheld, cell phones, notebooks, or desktops.

We believe that the applications need to be optimized for occasionally connected computing. And all of this has to be done on a foundation of standards that are comprehensive and widely adopted.

To give you an update on where we stand on this and the progress we have made, I'd like to invite up on stage with me Dr. Richard Wirt. He's our senior fellow in charge of our software initiatives for Intel. Besides being a senior fellow, you do a couple of other things right?

RICHARD: Yes. I co general manage our Intel software and solutions group. It's about 1500 people, to enable software for our processors. I focus on the engineering and keeping technology leadership for our software.

ANAND CHANDRASEKHER: Brilliant.

RICHARD: We have been talking over the last year about unifying our tools to provide a consistent set of tools that you can write once and run across our processors, whether it's a communicating processor or a computing processor. Today, I'd like to deliver an update on where we are with this.

We support these tools across Windows NT* and Linux as well as the other targeted operating systems that are popular for the PCA. As you can see from the chart, Intel Vtune™ technology now combines all support for all four processors into one tool. We'll be releasing shortly the compilers for Intel® XScale® microprocessors. These take advantage of the optimizing technology that we've used on our Itanium® 2 as well as our Pentium® 4 processors.

Today I'd like to announce our integrated performance primitive libraries. These libraries support multimedia and imaging, video, and really set you up for the focus on the Digital Home that Intel is beginning to have.

ANAND CHANDRASEKHER: That's fantastic, Richard. What's unique about our tool set?

RICHARD: I think what's really unique is, we have one set of tools that's consistent across all of these processors. This is supported by about 500 engineers doing the work on developing these tools with our compilers and our performance libraries.

We have a strong commitment now that we've had for a number of years to provide these and carry them forward.

ANAND CHANDRASEKHER: That's great. So all of this enables the goal of write once, run best on Intel processors?

RICHARD: Yes.

ANAND CHANDRASEKHER: Can you talk a little bit with occasionally connected computing as well?

RICHARD: Occasionally connected computing is important. As Intel is bringing out Intel® Centrino™ mobile technology, we've created hot spots all over the United States and around the world. It's important as the user moves between these hot spots, their desktop, or even a cell phone, to have a consistent and persistent environment, much as if they were always connected.

In order to do this, the application needs to keep track of its connection state and provide synchronizing software that is determined by the bandwidth that the connection has to update the data that the user is using.

One of the other things that you need to worry about is standards. We've talked before about universal plug and play. This keeps track of the characteristics of the device and presents it to the application. But manageability and security are also very important in this environment.

ANAND CHANDRASEKHER: So building on standards, essentially, to allow the software innovation to take place on top of a ubiquitous standard base?

RICHARD: Yes.

ANAND CHANDRASEKHER: What has the acceptance been?

RICHARD: Large corporations are already beginning to enable their applications for their mobile work force. Oracle just announced applications for the wireless, PDAs as well as cell phones. Siebel has had their CRM application enabled for wireless support for some time. J.D. Edwards and others are quickly moving to this environment.

ANAND CHANDRASEKHER: That's fantastic. Can you show us something that puts all this together so we can visualize what this looks like?

RICHARD: Yes. We've talked about occasionally connected computing. I'd like to show you another user usage model. This is mobile collaboration. Invite Bill and Diane up to show you a demo of this.

(Demo begins and ends.)

ANAND CHANDRASEKHER: What you've seen are Intel tools with the ability to write once, run best on a range of Intel processors, optimizations for mobile platform computing across desktop, notebooks and handhelds and a number of tools to enable occasionally connected computing.

Convergence has the opportunity to drive enormous innovation at the platform level. Craig talked about, yesterday, the opportunity that lies in front of all of us to use this innovation facilitated by convergence to drive demand.

What I want to do is take a look at this from a mobile standpoint.

The mobile marketplace is continuing to grow nicely. The consensus industry forecast talks about a compound annual growth rate of about 16 percent for the notebook market segment through 2006. This is being fueled by the thin and light product category, and also by the increasing acceptance of wireless LAN capability, which untethers the notebook, not only for the connection capability but also increasingly for battery life.

We're finding that consumers, whether they're in the work environment, the home environment or the academic, student environment are demanding connectivity, untethering from the wall, being able to compute wherever they might be, whether it's in the quadrangle or the classroom, which in turn fuels the need for longer battery life. Once you untether your PC from the network connection, you also want to untether your PC from the wall. And of course none of these consumers are willing to give up the need for leading edge performance so they can be productive wherever they are and for a lightweight system so they can carry it with them wherever they might be.

We've been engaged on developing a platform, it's been codenamed the Banias platform up until now. At the last Fall IDF I gave a lot of details from a technology standpoint as to what that solution entailed. It was a processor, a chipset, as well as a wireless component that was all designed together and validated together to deliver against all four of those vectors, breakthrough performance and capabilities.

The processor itself is called the Pentium M processor proper, the chipset is the Intel 855 chipset and the wireless component is the Intel Pro 2100 wireless component.

All of these components have been not only designed together; they've been validated together. This is Intel's best silicon ingredient technology for the notebook marketplace. And it's going to be marketed under the brand name of Intel Centrino Mobile Technology. And this brand name allows us to articulate to consumers, again whether they're in work or business, what they should be looking for in the best of mobile technology.

Now, last time I talked a lot about the ingredients from the technology standpoint that went into the processor and the chipset to enable us to deliver this kind of breakthrough performance, battery life, connectivity. What I want to do this time is show you what we've been able to achieve. So to do that I want to invite up on stage Don MacDonald. He's director of mobile marketing, and I saw in an article this week in the San Jose Mercury News that he "has a problem."

DON: Several.

ANAND CHANDRASEKHER: So, Don, you're going to tell us about how you're going to overcome that, right?

DON: I am.

ANAND CHANDRASEKHER: What are you going to show us here?

DON: Many of us have been working on the Intel Centrino Mobile Technology program for many years and it is my distinct privilege to show you the very first performance and low power demo of Intel Centrino Mobile Technology here at Intel Developer Forum.

So what we'd like to do is load these three notebooks with demanding applications, that include intensive spreadsheet applications, adobe Photoshop* rendering and video encoding and we'll see which one finishes first and give you a real application usage of high performance.

But because we care about performance, we also care about battery life as well, we have them hooked up and they'll come up in a second, we have all three of these notebooks connected to a power analyzer so you can see as they do their performance test how much power they're consuming and that's relevant for all of us who use wireless and care about battery life. Anand, if you can help me here. On one, two, three, go, we'll go.

ANAND CHANDRASEKHER: All right.

DON: One, two, three, go. Let me explain what the notebooks are and what you're seeing in the background. The notebook on the left is a Pentium® III mobile processor running at 1.2 GHz. Beside it is a similarly configured notebook but powered by a 2.4 GHz mobile Pentium® 4 processor-M. And this notebook enjoys an Intel Centrino Mobile Technology Pentium® -M processor running at 1.6 GHz. It's doing its thing, doing its calculations.

What I'd like to do now is draw your attention to the charts at the top. The first thing I'd like to point out is the Pentium 4 processor-M is the dark blue line. These are the power traces, how much energy it's consuming. The green line is the Pentium III processor. Today, the low power version of the Pentium III is the world's lowest power consuming processor for tablet PCs. And the Intel Centrino mobile processor is in the magenta color.

The first thing you can see is that when it's in idle state, Centrino consumes a lot less power than either the Pentium III processor-or Pentium 4 processor-M. For wireless users and notebook users that's a big deal because that's what your notebook spends a lot of time doing.

ANAND CHANDRASEKHER: I also see the blue line is taking up a lot of power here.

DON: Yes, you can see two things. One, the Pentium 4 processor-M, the highest performance mobile processor, consumes the most power. That's logical. But what you can also see is the trace there is showing the Pentium-M processor has finished far in advance of the mobile Pentium 4 processor -M processor. But in addition, it's been faster but consumed less energy and that translates into longer battery life, really important for all notebook users, especially wireless notebook users.

The other thing is you can see is that the Pentium III processor consumes less power than Pentium 4-M, but because it takes longer to finish the job, you actually end up consuming more energy.

And so the summary here is that for notebook users using Centrino Mobile Technology, if you care about performance you get the highest performance using Centrino Mobile Technology. But as a double bonus, if you care about battery life, you also get longer battery life. And this is a fantastic effort and I'd like you to give a round of applause to our architects and engineers who designed this product and I think they deserve a round of applause. So please join me.

(Applause).

ANAND CHANDRASEKHER: Thanks, Don. So let's put that in perspective. What he just showed you was breakthrough performance across a range of applications, very complex tasks, and the slide behind me shows you that data. The green is a mobile Pentium III processor, top speed available on that today. The orange is the mobile Pentium 4 processor, top speed available on that today, 2.4 GHz, and the Pentium-M processor at 1.6 GHz. You can see the Pentium-M processor provides breakthrough performance in the notebook environment.

Even more cool, you can see you can get, no pun intended, you get outstanding battery life. You get about 215 minutes of battery life out of the Pentium III M system. Remember, these are all equivalently configured systems. You get about 174 minutes battery life on the mobile Pentium 4 system, and you get a whopping 316 minutes of battery life on the Centrino Mobile Technology. That's five hours battery life, running the kind of applications you saw up on stage. Breakthrough performance, breakthrough battery life.

So if you're interested in mobile technology that delivers uncompromised performance and uncompromised battery life, Centrino Mobile Technology is it.

That's two of the vectors consumers care about, performance and battery life. The other two are connectivity and form factor, sleek, slim, form factors that they can carry with them.

We work pretty intensively to make sure our wireless solutions were not only designed together to deliver a robust experience but also secure. They all support industry standard security solutions in addition to the security support standards from Cisco, and we worked a lot to make the wireless solution easy to use from a consumer standpoint.

One of the aspects of easy to use is actually to take advantage of and resolve issues like coexistence between Bluetooth* technology and 802.11b. This is a technology I showed you about a year ago at spring IDF. It was a technology that was sitting in our labs, and I showed you an algorithm that we had developed to essentially address the coexistence so did you not have interference when both these radios were active at the same time. I want to now show you it's not just a lab technology, it's productized and in a product and Don, you're going to help me with this right?

DON: Sure.

ANAND CHANDRASEKHER: What are you going to show us?

DON: Two notebooks, similarly configured, both Intel Centrino Mobile Technology, both have Bluetooth and 802.11 radios. In this notebook, I'm going to demonstrate the problem visually which is of coexistence between 802.11b and Bluetooth both on the 2.4 GHz spectrum colliding with each other on performance.

I was at a pharmaceutical company last week, using their 802.11b wireless LAN and somebody asked me to send a file to a PDA.

What we're showing here is a file being transmitted over 802.11b. Typically it would be a file transfer. To see the problem, I'm using an AVI file. Everything is great, traffic is streaming well. If we turn up the music, you can hear that the sound is actually pretty darn good.

However, if I now send at the same time a file across to a colleague, in this case a PDA, wirelessly over Bluetooth, you can begin to see the impact. I'm simply copying a file. Once the file begins to transfer, the radios are colliding with each other, the performance is going down and you can see, in fact that, the 802.11b is beginning to hang and is barely working.

That's an example of the problem of coexistence between different protocols operating on the same spectrum.

Now, in good old Intel fashion, it's great to identify a problem, but it's our job also to come up with a solution. And what the labs have done in the mobile group here have been to develop, and as Anand mentioned, we showed a technology demonstration, it's now a real product that will show up in real Intel Centrino Mobile Technology systems starting in March.

So here we have the same demo. Can I get the audio up, please? So we have the same configuration and system on the left. Now I'm going to replicate the same Bluetooth file transfer, in this case to this PDA. And what you should see is once it begins to kick in, nothing happens. Basically the quality of the service remains very high. The file transfers happen on the Bluetooth radio simultaneously with the 802.11 transmission and the quality and experience is far superior.

That's a great example of performance in the context of wireless, but it also means that your notebook PC, as I showed earlier, gets quickly into hurry up and idle because it's not spending time to transmit.

So again, better performance and lower power consumption.

ANAND CHANDRASEKHER: Fantastic. Thanks.

DON: Thanks, Anand.

(Applause.)

ANAND CHANDRASEKHER: So what you just saw was the Intel Centrino Mobile Technology. Designing it together with Bluetooth technology allows us to bring cool ease of use features like the one you just saw.

It's also about form factors. That's the other vector that consumers care about, whether they're in work or business environment or in the home environment.

Last time I talked about a remote heat exchanger technology that allowed us to essentially cool our Pentium-M processor very differently in a much smaller form factor than the mobile Pentium 4 processor M.

What that enables are designs like these. There's a range of designs up here in front of me. This is an X Series* from IBM. This is a Samsung machine, about less than one inch thick. This is a Toshiba Portege* which will be available in the marketplace. This is a tablet from Motion Computing, ultralow voltage processors, no fans.

You can see a range of other systems here as well, one from Gateway, and two designs from Taiwan, one from FIC and one from Asuss. A range of designs available in the marketplace. This is just a sampling. We have more designs on the way, on Intel Centrino Mobile Technology by factor of four than what we had for mobile Pentium 4-M this time last year. Outstanding support from the industry on this technology.

Historically, when we validated our processors and chipsets, this is what we've done. We validated the processor, chipset, looked at the optimized drivers and software that go on top of it.

We extended this with wireless. We not only validated the system with our wireless hardware, but we also validated against the security solutions, tokens, and worked with companies like VeriSign, CheckPoint. There's an announcement that goes out today on the co-development work that's underway between Intel and CheckPoint on the Intel Centrino Mobile Technology platform as well as moving forward. We also validated against a range of access points. Of course, sitting on top of it our stable image software technology that covers the graphics drivers, the wireless drivers. We also validated, to some extent, with the VPN technology, the middleware that exists, so that we looked at the entire solution that an end user would be experiencing when they looked at a wireless environment.

We didn't stop there. We realized that that was a good part of the solution, but we also needed to engage with the service providers that deployed hot spots on a global basis, whether they be MegaBeam, Telia, or British Telecom in the European continent, whether they be China Netcom, China Mobile in the People's Republic of China or Korea Telecom in Korea, or NTT Broadband in Japan. Each complements different recipes and we wanted to make sure they all worked in a seamless manner on Intel Centrino Mobile Technology.

We also worked with Boeing's Connexion* technology to enable Centrino to get to new heights, literally. What we have done is engaged with these companies in two forms, one in a co-marketing arrangement so that when you are in a hot spot, you can recognize you are in a hot spot. But probably more importantly, working together with them so they can take advantage of our engineering together with theirs to enable a more robust experience for users. You don't just have to hear it from me. I'd like you to hear about it from T*Mobile, AT&T Wireless, and Connexion by Boeing themselves.

(Video playing.)

DAVE CABLE: T*Mobile sees its relationship with Intel as a huge win in being able to drive adoption and experience the growth that we believe is possible as we roll out wireless LAN throughout the T*Mobile network and throughout the nation.

T*Mobile sees that having verified the Centrino Mobile Technology on its network, that it will enhance our ability to provide that promise to our subscribers through a full capability service on our network.

We plan to have hot spots in places where people typically go: Coffee shops, airports, and other places where they may congregate but really need to use our data services.

It's really encouraging to see companies like Intel invest in growing the market with solutions that are rooted in giving our customers a great experience.

MICHAEL WOODWARD: AT&T Wireless is very excited about the launch of Centrino Mobile Technology.

What wireless LAN allows us to do is it allows us to offer choices to our customer and a new complementary service that goes with our traditional offerings. In particular, we will be showcasing this excellent customer experience at the Denver International Airport, where we'll be allowing our customers to come in and see a booth staffed by both Intel and AT&T Wireless that shows this solution.

AT&T Wireless and Intel have been working together for a long time on a technology road map and a strategy. And what's so exciting about Intel as a partner is that together, we share a common strategy and a common vision for how wireless data should be deployed and integrated. Both companies are very passionate about that. So that's something we really rely on.

SCOTT CARSON: With Connexion by Boeing, we have a vision about how you can enhance your life when you're remote and mobile. Centrino, with its power management capabilities, really opens up a whole new area for both Boeing and the airlines. But, more importantly, for the passengers on the airplane.

It offers them freedom and a way to choose what they do, a way to use their time productively, if they want. You can surf the Internet, be connected to your E mail, just be entertained to make Connexion by Boeing effective. You wire the seats or go wireless for access. But then you always have to have power at the seats.

It may free us from that completely. For the airlines, that means a less complex airline. For our passengers, a less complex world they have to live in.

ANAND CHANDRASEKHER: Some of you may have seen some familiar faces on that Boeing video. Many of were you on that plane, the maiden flight. That's what you get to see when we launch the product in March.

What I have here is a Dothan wafer. Dothan is Intel Centrino mobile processing technology taken to 90 nanometers. Craig Barrett yesterday talked about we're going to be in production in 2003 in 90 nanometer technology with 300 millimeter wafers. This is going to be the first processor on 90 nanometer technology going to 300 millimeters.

What you get with that is clearly increased performance but also a 30 percent reduction in cost. And in the corporate IT environment, we have heard repeatedly from CIOs that what's very important is platform stability. So we have created extra diligence in designing this processor. It will, essentially, fit into the Centrino designs that will get launched in March with no change, either from a hardware standpoint or from a software standpoint.

That enables IT shops, essentially, to do a single qualification, be able to take the Dothan processor as it comes out during the latter part of the year, second half of this year, and it also positions us for a high volume ramp.

So that's what is coming out on the mobile side in 2003. I want to talk a little bit about 2004.

Craig yesterday demonstrated Newport, a concept vehicle that's built out of our labs which showcases always on, always connected capability. It's a concept platform to showcase new capabilities that we and the industry can consider for 2004.

I have up here a sample Newport system. I'm also going to invite up on stage with me Ticky Thacker, who's the director of our mobile labs, to help me with one aspect of this demo.

You can see here that Newport can operate in a clamshell form factor. This particular system has four different radios on it. It's got dual band, 802.11a, b, GPRS as well as Bluetooth technology. I showed you how it operated in the clamshell mode. I can also detach it and use it as a tablet. This is a completely working prototype.

One of the cool things that's there on this platform is also a secondary display. And that gives us the ability or gives consumers the ability to get access information on a realtime basis. And I actually want Ticky to help demonstrate this capability. What does this do, the secondary display?

TICKY: It allows us to have realtime messaging capability to a very rich client. And you can access all kinds of information, such as your Outlook messages, calendar messages, and personal information.

We're also going to show you an exciting capability this morning. We're demonstrating this first-time capability that we have developed with Verizon. And it allows us a Verizon network, desk phone, to be transferred in realtime to the notebook. The way it does is there is an always on Verizon client on the notebook. And, Anand, we're going to demonstrate this. I'm calling you. So now you can see on my always on display, this is a realtime machine connected to the Internet, that Anand called me. I can decide to pick up that call. And let's try. It'll allow me to transfer the call to my cell phone.

And, shortly, what you'll see is it'll connect me first to the to the network.

(Phone ringing.)

TICKY: And then allow Anand connect to the network and then both of us can talk. So we're just momentarily waiting for Anand to connect.

ANAND CHANDRASEKHER: Takes a few seconds. There we go.

TICKY: Hey, Anand, that's a good show.

ANAND CHANDRASEKHER: That's brilliant. Thanks

(Applause.)

ANAND CHANDRASEKHER: Now, what else can you do?

TICKY: Let me show you some of the other capabilities of this messaging functionality. So what we're going to show is some of the capability that exists on the notebook. Example, it shows you the connectivity, and I'm going to show you what you can other things you could have done with that call.

So here I'm going to go into the call menu, and it'll show that there were some other calls that I had on that day.

Let's see. So there's a call from my desk. It associates the call from my desk to some of the activities on my calendar, for example, I could have one on one with Craig as well as asking send a report out to Anand or Craig.

So here I'm basically allowed to then send the message to Craig in realtime. So you're seeing a realtime capability, allowing a realtime forwarding with the rich capability of the notebook and the and the client.

ANAND CHANDRASEKHER: Great. Fantastic. Thanks, Ticky.

(Applause.)

ANAND CHANDRASEKHER: The technology you just saw is embedded in Verizon's network, and that's an embedded application that they ported over to this Newport platform to demonstrate this capability. Their solution is called Verizon Digital Companion, and it's to be launched in a couple of weeks here.

To enable that kind of a platform, you also need long battery life. This particular system gives you six hours of battery life when it is connected to the keyboard. When you detach it and use it in tablet mode, it gives you about two and a half hours of battery life. And we're using some alternative sources of battery life in addition to the standard battery that's built as part of the system.

Last IDF, I talked about the need for extended battery life and getting to eight hours, requiring not only components innovation, software optimizations, but also cultivating alternative fuel cells. I want to give you a quick update on where we stand on that.

In terms of components development, I have here a display. It's a low temperature plasma looking display. This display uses three watts of power, a significant advancement in terms of reducing power on the display front. I would expect these kind of displays to be in place in the 2004 time frame, systems like Newport concept vehicle. In effect, Newport does use an LTPS display system.

We are also working on alternative power sources. I have here another concept platform, this is a tablet that we showed last year. And it is fueled by a fuel cell, this blue box here that I am holding in my hand. And this is a working prototype of a fuel cell for a notebook. This particular fuel cell is developed by PolyFuel, an Intel Capital investment portfolio company. In a couple of years, a three year time frame, in this kind of a form factor, PolyFuel anticipates being able to deliver a 150 watt hour battery capacity. That's huge.

The five hours of battery life that you saw Don demonstrate was based on 48 watt hours of battery. This is three times that amount in about the same form factor.

We're well on our way to getting to older battery life working with all of you as an industry. More on that as we make additional progress.

I want to put this opportunity in front of us in perspective. In 1721, George Graham, a British scientist invented a balance spring mechanism that essentially allowed clocks, chronometers, to do away with the pendulum, increase accuracy and reduce size. It's not that accuracy didn't exist, it's not that portability didn't exist. They existed. But he was able to significantly improve the reliability, the accuracy of the clocks, and shrink the size further.

That improvement in technology, which in hindsight looks pretty simple, revolutionized the field of naval operations because it made longitude calculations much easier to do, and it had unintended consequences in fields far outside of the field of watch making. It had impact on transportation, it had impact on financial industry in terms of transactions. Huge.

And I think in front of us what we've been able to do with Intel Centrino Mobile Technology, we have a similar opportunity as an industry. To take advantage of this opportunity, we need to accelerate innovation on the four vectors that we know consumers care about: performance, form factor, seamless wireless connectivity, and battery life.

And by delivering new usage models, like we were able to demonstrate with Newport, drive demand and ultimately grow our mutual businesses. So we have a tremendous opportunity in front of us here.

What I'd like to do now is shift focus and have my colleague Gadi Singer, vice president of the Wireless Communications and Computing Group, and general manager of the PCA Components Group, talk about the other side of mobility, handhelds and cell phones. Hi, Gadi.

(Applause).

GADI SINGER: Hi. Thank you. Good morning. Once upon a time, all phones were for voice only including the cell phones. The big change happened in the transition from the second generation cell phones to 2.5G, when data was enabled. When we have packeted data, we have the ability to be always connected, always on the Internet. And this major transition enabled a new category of usage models.

This transition from the voice-only to voice + data is happening. And as you can see on the left, it's expected to take a large portion of all phones sold in the world within a very short time.

If you look at the installed base, the subscribers of 2.5G and 3G are expected to reach about 200 million by the end of this year. And reach one billion subscribers within the next five years. That's a very large base.

This transition doesn't happen in all geographies at the same time. If we take, for example, Japan or some other countries in the Asia Pacific region, they have already made much of the transition. If you look at the graph on the top, data and color screens are a given. You cannot sell a phone in Japan, or it's very difficult to do so, if it is not data enabled and color.

And there's a very rapid transition to Java and richer application base.

We can see the same phenomenon moving into Europe and yes, even in the United States it's happening. If you listen to the commercials on TV or get those glossy brochures, it is obvious to know where the value and excitement is.

So this transition to data rich phones is happening. And what are the best architectures to allow this transition, to allow this device to become this converged computing and communication device?

Intel introduced PCA - the Intel® Personal Internet Client Architecture -- a couple of years ago. And the concept behind PCA was simple as it was revolutionary.

Rather than merging it all together, building separate modular subsystems, having a compute subsystem, a communication subsystem, and a memory Flash, all of them modular but well integrated, well communicating to each other.

Now, by having a separate modular computing subsystem, you create an environment for the developer that is very familiar. That's a compute environment. We know how to develop for that.

So it creates a better development process, shorter time to market, and it allows building in a very scalable manner. As the compute subsystem grows, advances with products, with time, you have your application growing on that compute subsystem.

It also allows you to have an easier path of migration. Where is the best base of knowledge and content and application today? It's on Intel Architecture clients. This computing subsystem allows you to move from one compute subsystem on another platform.

So the Intel PCA provides the architectural frame. To talk about how it comes into reality, we'll look at it in three levels. We'll talk about the leading building blocks, we'll talk about integration, and I'll talk about two parts of integration -- the in package integration and on die integration. We'll talk about putting it together into full solutions. We'll start by taking a closer look at the building blocks.

At the heart of the compute subsystem is XScale microarchitecture. The compute engine that is a leading embedded processor, the best in class architecture that has the best performance. It is scalable, but it can reach the best performance of embedded processors at low power.

On the left is the performance levels on MPEG4 decoding which are very good, but what you can also see is we're creating a range of products that improve the performance at very fast clip.

But every time you hear me talk about high performance, you will hear me talk about low power or low battery life as well, because they have to come together.

So what you see on the right is the drive that we have to reduce power. In this case, it's the nonactive power, such as sleep and idle. We are reducing the active and nonactive power by combination of architectural technologies, circuit techniques, and process techniques.

So as we go up with time on performance, we stay or push down the power to extend the battery life and support those four vectors of mobility that Anand talked about.

But Intel Xscale technology, even though it's a very important building block, is not the only one. There is a set of leading building blocks that is provided as part of the whole portfolio. On top of the Intel XScale technology software are the drivers. There's also the communication system which is based on Intel Micro Signal Architecture, MSA, which has in it a world class DSP and other elements. And on top of it, a lot of the communication software & communication layers.

There's the high density on die Flash capability, and the basic technologies, the packaging and silicon technologies that are important.

So we have all those building blocks that are required to create the best systems. But having all the best in class building blocks is necessary but not sufficient.

The next element is putting them together with tight integration. Now, integration has a generic value, inherent value, in all four vectors. It's clear it has a value on form factor. It's smaller. It has a value on mobility, on ability to integrate all elements of connectivity. But it has inherent value on performance and battery life as well. And let's look at that.

So the first element is a smaller footprint and by stacking within a package. Integration in package means you're stacking multiple dies one on top of the other. And by stacking multiple dies, we get a smaller footprint, lower power, and ability to have higher performance.

Intel is the leader already in stacking. We have shipped over 100 million units of Flash, and we're moving from just Flash unit being stacked to Flash plus logic. And I'll show that as well as go into an Intel XScale technology demonstration. But before we go there, let's show the integration here.

What you see here is the Intel XScale microarchitecture-based Cotulla PXA 250. What happens when you take a Cotulla PXA 250 and integrate it with two Intel StrataFlash® dies? What is the resulting dimension? This is the result. This is the PXA262 - integrated three dies into a smaller space in the same profile, 1.4 millimeter high profile.

Just to see some of the power of the Intel Xscale microprocessor I have here with me Gary Forni, the director of the ISV enabling group. We talked about having this basic Intel XScale system, and what can you show us?

(Demo begins and ends.)

GADI SINGER: So if you can do all that with the in package integration, let's take it to the next level, to on die integration. What we've announced last week was the Intel PXA800F cellular processor, which was formerly and fondly known as Manitoba.

This is the industry first Wireless Internet on a Chip. The big story about this is about being small. It is about being able to put all the major ingredients of a cellular system into a single piece of silicon.

Talking about convergence of computing and communications, this is done on a single piece of silicon.

Now, what does that single piece of silicon have in it? It has a leading 0.13 micron process for logic plus flash, and it includes a computing system, as you would expect. At the heart of that computing system, there's an Intel XScale microprocessor running at 312 megahertz with its own SRAM. There's also a full communication system, including Intel MSA, its own SRAM, and some of the peripherals. It has a high density, four megabytes of flash on it. It also has multiple additional peripherals, such as a sophisticated power management system and RF interface for a variety of RF transceivers. The generous 4 Mbytes flash that's on this allows to put on the full GSM/GPRS protocol stack and the full Java operating system or realtime operating system, and a few applications. All of them on the same die.

So running your communications and your applications and your realtime operating system is done all execute in place without ever leaving the die. And there are major implications to the ability to do that.

If you look on the left, it creates a headroom for a new level of capabilities that translates directly to a new level of user experience. Being able to run Java games, being able to have better reach or Internet services, being able to run video decoding. All of those are enabled by the significant headroom that's created by the new levels of performance of the PXA800F.

Now, having the flash on die allows to drive performance up, but also at the same time drive power down. It is closer to where it is being used, so you need less time and need less power. It is wider path and a direct path. You don't need all the buffering. And you don't need all the cycles to go in and out of the die with all the wait states.

So what you see on the top right is that the latencies are much shorter. Getting data, getting instructions from the flash is three times or more faster than getting the same instruction from an external SDRAM. At the same time, it's much lower power, because you use the cycles only to do work. The instructions are there when you need them.

So you can also see on the bottom right that it provides lower power. Now, it all sounds good. But how does it look? How does it feel? Craig demonstrated some of that in his keynote yesterday. But to take a closer look at the PXA800F capabilities, I would like to literally call on to stage Mark Casey, the PCA marketing director.

(Demo begins and ends.)

GADI SINGER: So where does this PXA800F fit in the overall scheme of handheld devices? It is very well tuned and optimal device for the mainstream phones. Not only does it fit in the mainstream phones, it brings into the mainstream phones some capabilities that were associated only with the very high end.

So if you look at the model, and on the left branch you have the phones, the PXA800F is on the mainstream phone. And if you still are interested in a standalone application processor for the communicator at the top, phones such as this smart tech phone are using the PXA262, as we demonstrated earlier, for the communicators and the smart phones. And what about the rest of the PDA? The PXA250 is a very capable, scalable leading component application processor for the PDA entry level and the PDA mainstream.

So you're saying, this is what we see today. But what are you working on? What do you have in store?

Well, we have a range of future products that are in various stages of development and productization.

And we're driving along all vectors. On the communications space from standards such as GSM/GPRS, adding additional standards such as UMTS.

On the architectural front, adding architectural innovations such as the wireless MMX that we've introduced last IDF and is on track, as planned, for future introduction, and other architectural features. And also pushing the frequency, pushing the memory size, the density on die.

Packaging and silicon technology, you can expect to see forward progress. You can expect to see more dies being stacked, three and more dies being stacked into the future. We'll take the integration on silicon to the next level, integrated 3G products.

Which brings me to the last point on the putting together as a full solution. Putting it all together for the end user. And part of it is Intel enabling creation of the full solution. But also part of it is what the industry is doing, how it refines and fits in the ecosystem, how it brings a range of compelling products to the end user.

If we look at the development environment, it should look familiar. If you worked with other IA platforms, it has the basic development tools such as reference board and development boards, the basic drivers for all the hardware elements that we have on silicon. It has the operating environment and operating systems, and every single major operating environment in the handheld space has been tuned and optimized for the Intel XScale systems, whether you're working with Windows* or palm or Symbian* or Linux or in the Java environment.

Then you have the familiar tools that Richard talked about in the beginning. If it's the hardware debuggers, simulators, video, or the IPP primitive library. All of those are there. And it's deliberate. It builds on the experience that Intel has on creating those development environment over many years. It always allows the expertise to come into play with those environments, because many of the developers are familiar with them. And it eases the porting from one client platform, to another.

You, as developers, have a tremendous opportunity here. The handheld device is coming of age as the emerging gateway to the digital universe. It will be the converged device that a billion or more people are going to carry with them in not that long. This is your opportunity as the architectural landscape is shaping. Use the PCA as a basis to innovate, to excite, and to delight your customers.

Now, our user has been mobile for the last two keynote speakers, and she reaches her digital home or digital office. This convergence of computing and communication is happening also on the desktop.

And to talk about the desktop PC segment, I would like to invite Louis Burns, Intel vice president and general manager of the Desktop Platforms Group.

LOUIS BURNS: Good morning. It's always a pleasure to come back and talk at IDF with you, the industry, share our plans, and as important to us is also hearing your ideas and your plans for what we're going to do from an innovation point of view.

My agenda today is actually really pretty straightforward and simple. We are going to spend some time and explain what's going on in our products and our platforms and our initiatives for 2003. Then we are going to give you a peek at what we're looking at in the 2004 timeframe.

So what we've said many times is Intel thinks about these things in probably a little bit different way than most. We think about the entire balanced architecture of that platform. It's not just about CPUs or about chipsets or about I/O. It's about how all those interact and work and deliver an outstanding experience for the end user.

Gordon Moore last week basically confirmed that Moore's Law is secure for the next decade. So it's not just about Moore's Law and CPUs. It's about the other parts of the platform, like I/O, audio, and memory, and other parts of that.

So let's get started and give you an update on what we're going to be delivering in 2003.

These building blocks we talk about, these initiatives we talk about are really focused on bringing end user valued experiences into the marketplace. Intel does a lot of research with end users, with our customers, with academic environments to try to understand what the needs are. We build that into our planning.

So first up for 2003, the processor. A year ago, I talked to you and introduced the name of Prescott. What I want to do today is give you an update of where we're at with Prescott.

Firstly, this product will be built in our 300 millimeter factories. Intel will have a number of these around the world, which brings tremendous cost advantage to our products and a number of other benefits. It'll also be based on 90 nanometer technology. Now, what you may not know about this is because we have such advanced technology folks, they work very closely with our architects and designers of the processor. So what you see is the absolute best performance possible by the integration, the close working of process development, the 90 nanometer guys, targeted with our architects and product development guys, to deliver that into the marketplace.

Moving forward, Intel's truly unique in this capability. You'll see enhancements in the Intel® NetBurst® microarchitecture. It's a tremendous enhancement to the widely accepted and very well thought of Hyper-Threading Technology.

Thirteen new instructions, an L2 cache of a megabyte, and its first instantiation will start with an 800 MHz front side bus.

So tremendous move forward, continuing Intel's clear leadership in the processor space.

What I have here is actually a wafer. This is actually the first few runs through the factory of the Prescott wafer. And what you'll see here is taking advantage of that 300 millimeter capability and delivering the next generation process and product into the marketplace.

What I'd like to do now is talk about the platforms we'll be delivering into the marketplace in the first half of this year.

The first one up is codenamed Canterwood. Canterwood is about the ultimate in high performance platforms in the marketplace. There is a number of really cool, new innovations in this product that deliver on that ultimate performance experience. First up is Performance Acceleration Technology, code named PAT. Not codenamed after Pat Gelsinger, but he likes to think that and we let him believe that.

PAT -- the way you should think about it is a turbo mode in our controller. And what our engineers have done through some very clever invention is to enable this turbo mode, and in working with DDR memory, absolutely the fastest capability in DDR on the planet.

If you move around to the right side, we think that leadership in both mainstream and performance platforms, moving forward, include the following things. At least an 800 MHz front side bus. The establishment of dual channel memory in all of the performance and mainstream platforms moving forward. In this case, DDR 400.

We've done this while also paying attention to the cost, because all this capability, this awesome performance is to be delivered in a four layer motherboard, which has had some very interesting engineering things to solve.

We've talked in the past about Serial ATA. This platform will bring Serial ATA native in the platform. And then again, through some clever invention that we've done with our software guys, we're delivering soft RAID capability on that platform.

So tremendous performance in a well balanced, well thought through, overall architecture, from the processor, to the front side bus torques, the memory subsystem and the I/O.

So what I'd like to do now is actually give you a demonstration of the Canterwood platform, the first time in public. And the way I'd like to do that is actually by giving you an update on the usage model we talked about, Digital Home.

The last couple of conferences, we spent a lot of time talking about the vision and where this thing is going. What I'd like to do now is actually provide some proof points for you as to where we're at.

So what I have here is a Canterwood based product, a 3GHz Pentium 4 with Hyper-Threading Technology turned on, and we're going to show you some pretty awesome performance capabilities.

Now what we talked about as a first proof point in the Digital Home vision is really the integration of CE capability into a PC. Two simple examples would be live TV and personal video recorder, PVR functionality.

So let's drive here. What you'll see here is a very interesting, simple, easy to use interface. And what you can see is it will deliver both a two foot experience, with either a mouse or keyboard, or if you want you can sit back with a remote control and basically run this.

Let's look at some of the capability. First let's bring up live TV. What timing to have Mike Tyson come up. I guess that's the trouble with live TV. I don't know if we're going to be able to digitally enhance his tattoo or not. But what you can see in this live TV capability, if I get a phone call in my office or one of my kids comes in to do homework, I can easily pause it. And while it's paused, it's basically recording and encoding that information. I can take the phone call or help my kid with his math class, and then I can basically pick back up and have that run again.

Pretty cool capability. PC, this awesome performance enables that in a very smooth, easy to use fashion.

I do a lot of digital photography. And there's this really cool new application that I want to show you. So I'll basically pause the TV. I'll come over here to an application which is called 3-D album that's shipping in the marketplace today, and what's interesting about this is it takes your photo library and lets you display it in a fun way.

So let's take a look at this. Let's do photo cube inside, which is actually painting the pictures on the inside of the cube. In case you don't like that, you're a little claustrophobic, we'll paint the pictures on the outside of the cube. We can change the rate of transition. Let's take this and make it spin a little quicker, or a little slower. So I change the interval and you can change size. That's cool, right? You can do it in a number of different ways.

Let's go to this one I like, picture rack. This lets you kind of put them on a rack. And if you pay attention, what you'll notice is as it's spinning by, you're actually looking at the picture from the front side and the back side, which I think is pretty cool. And you'll notice this does it in a smooth fashion.

Pictures are cool, but let's add a little music to it. Van Morrison has some cool music, and we'll basically add that on top of it. And that's a little more kind of a laid back thing, so maybe we'll put it in a hot air balloon. And what you can basically see is that picture is pretty cool. We'll go back to picture rack, and let's find a different set of music. Try to change the pace up here a little bit.

So what you can see is very easily from my MP3 list and my picture list, you can actually put together a pretty cool set of interfaces and kind of play with it.

Now, to make sure that you don't think we just kind of made this up and it's all canned, we actually took some pictures earlier. I want to show you just how easy it is to add new pictures to this environment.

So I take my Flash card, plug it in, and what should come up here in about five seconds is an Windows XP* wizard. And it will basically allow me to add these pictures in. Oh, it's going to be one of those days.

Try this one more time. Maybe we can turn the music up really loud while I try to get this to work.

(Laughter.)

LOUIS BURNS: Thank you. It's actually voice controlled also.

(Laughter.)

LOUIS BURNS: Here's the best part is you clap on, clap off, it turns things on and off.

Laugh.

LOUIS BURNS: So the demo gods are not with us today. We'll try this one more time. Okay. Anyway, so you'll have to trust me, it worked earlier, but the demo gods are not with us today.

So I'll turn that music off. Now, what I want to do now is just kind of picture my in laws are coming over and they like to look at my digital pictures but they'd really not be happy crawling around my office and looking at an 18 inch monitor.

So what I'm going to do now is move to my family room and we talked about the need in the first proof point of the step of the Digital Home is to have a device that simply and cost effectively enabled you to stream pictures and audio to a legacy TV or legacy stereo.

Last fall, we actually put out a reference design, and that reference design included hardware and software we put into the marketplace. What I'm pleased to announce and show today is actually one of our partners work ongoing this, Linksys, will actually be shipping for production this digital media adapter very soon to stores and capabilities around the world. There will also be a large number of other companies providing this capability in the short term.

So let's give you a demonstration of what this does for you. We can kind of sit back in my family room here and my in laws are in town and I want to show them some pictures.

So what I've got is a very simple interface. What I want to do is first put some music on. And what you'll notice is the play list is the exact same play list I showed you when I was working with the 3D photo album, because the play list actually exists on the PC. It's wirelessly being remoted over to the TV through the digital media adapter. So let's pick one of these songs. What did I miss? Oh, yeah. I'll get to that later.

(Laughter.)

LOUIS BURNS: It's one of those days. So we'll put the music on. So this is actually music coming through the legacy TV and its speakers, being streamed over remotely from the PC in the other part of the house.

So then what I want to do is, again, very simple point and click, is go to pictures. Now, typically what I show my in laws is the pictures of my kids playing soccer, or family reunions, or whatever, but they're really weird. They actually like to look at IDF pictures, so what we've got here is pictures from IDF that have been taken over the last few days, and I'll put that into a slide show. All right?

So what you'll see is you can set the pace, so with some cool music playing in the background, they can basically look at the pictures of whatever photo album I want to put up, and consume that in a fashion that's much more comfortable for them.

Now, being that I work at Intel I usually excuse myself at this stage of the meeting with my in laws and tell them I've got to go back to my office because I have a lot of e mail to do.

Meanwhile, back at the office, I'll show you how easy it is to add pictures to this album since it's now working.

(Laughter.)

LOUIS BURNS: You have to be flexible in this process. So using a simple Windows XP wizard, this high performance platform pulls them in very, very quickly, and basically we're done with that. We'll shut this window up, and basically what I want to do here is I'll do an update, and what you'll see here is it's grabbed those pictures and there's Gadi from his presentation, the outside of the building, the picture of the stage that's coming in. And I think Anand is coming around the corner here right now.

You can see it took a little time delay but we easily added pictures to this album in a fashion with music. So really cool what you can do in this space. I'm really not going back to my office to do e mail. I'm actually going back to pick up on what's going on with TV, because I want to pick up on what's going on with the Mike Tyson fight. So I can start the TV again, and while I've been in the living room, spooling information, pictures and audio, this has basically been recording this TV show for me.

So let's start it up. And at this stage of the game, again, very smooth in capture and play back of this capability. It's still streaming audio and video over to the other room in the house via a wireless capability. And then the other part of this is I'm a very big Kings fan, Sacramento Kings fan, so I asked my guys to capture the game last night with the PVR functionality. And what you'll see here is it should bring the Kings game up. Kings playing Milwaukee. The Kings are in good shape here. For any of you Lakers fans out there, maybe you'll make the playoffs, maybe not. And you can speed it up, and then go back to playing.

So what I've shown you is that because of the high performance platform capabilities we've put in place, we have the ability to smoothly and easily stream audio and video to a simple digital media adapter over here, and let people consume it in a fashion they like. It's also been capturing video real time while I've been doing that, and I've been able to do a number of things with the pictures, although they were a bit delayed.

So it's a cool example of high performance, well balanced platforms, proof points in the Digital Home that we're on the right steps to deliver our vision we talked about, and some pretty cool capability delivered by these awesome products we're putting in the marketplace.

So with that, I'd like to transition over to the next platform, Springdale.

Springdale really takes many of the same features you saw in Canterwood and delivers that into the mainstream space. And the way it does that is taking those features and architectural issues and then what we've added is an integrated graphics capability, the next generation graphics from Intel, called Intel Extreme Graphics 2, a substantial performance improvement from where we were with the 845G product.

We've also added communication streaming architecture capability, which allows to you connect up with Intel's high performance Gigabit Ethernet components, and deliver outstanding performance again in the mainstream price points, the mainstream platforms.

So once again, I'd like to give you a first time public demonstration of the capability of Springdale with its integrated graphics. We'll do this in the concept of the usage model of a digital office.

So what I have here is two Statesboro reference platform designs from Intel and what you'll notice on the left is the Hyper-Threading Technology (HT) enabled box that gives you two virtual processors. You'll notice on the right is the same configuration a different platform configured exactly the same with the exception of HT turned off.

HT has had really good acceptance in the marketplace because it gives you instant benefits for what the user is trying to do.

So what we'll do here, we have a big Excel recalc, and then we're going to take advantage of the megabit and do a massive backup to the server in the environment.

This is on the non HT system we're running here. So we'll give that a head start. We'll do the same thing from Excel on this side running. We'll start the backup. And that's off and running.

Now, this is probably as exciting as watching paint dry to watch this thing run, but if you can think about how IT will use this, with the security issues that face many IT shops around the world, is in the background you could picture on an HT environment running all the security software, protecting that enterprise.

When I used to run IT at Intel, they actually had a Web site inside that users put up on how to take security software off because it slowed the system down too much. In HT enabled platform, it enables tasks to run very efficiently. So you will see, the HT system has finished in about 60 seconds.

I won't make you wait on the non HT system, which takes about two and a half minutes to run. Instant benefits today for corporations and consumers, taking advantage of multi tasking and multi threading, delivering real end user valued benefits into the marketplace.

So what you've seen is the first demonstration of Canterwood, ultimate performance platform, first public demonstration of Springdale, which is for the mainstream price points with many of the same features, both taking advantage of HT capabilities.

Now, another thing we're going to be announcing today is a capability called Granite Peak. Granite Peak is really targeted at the heart of one of the biggest issues that IT faces in the marketplace today. They have to reduce the number of images, they have to figure out a way to get the stable images. And that has a big impact on the costs of running the IT shop and the overall stability of that IT shop.

Intel's always been given tremendous credit for our work and stability in the products we've shipped to the marketplace the last few years. Granite Peak is a quantum leap forward in stability and a very strong endorsement of Intel's understanding of IT's issues in doing direct programs to deliver on that. So, simply said, Granite Peak is about being on Intel based systems, with Intel device drivers, the guaranty of no image change for six quarters.

Now, we've taken it one step further in order to try to improve and help IT and what they're doing. Not only is it zero changes on our device drivers and the image for six quarters. It's a single device driver for both notebooks and desktops. So it's not two stable images; it's a single stable image, starting with notebooks based on Montara-G and desktops based on Springdale.

This is a huge step forward, a very large commitment to really addressing the needs and the issues that IT faces.

So that's 2003. Let's take a quick look at where we're at from a 2004 point of view.

Yesterday, Craig Barrett showed you a sneak peek of a platform called Powersville. And the way you ought to think about Powersville, it's kind of a concept platform that we work with. What I want to describe is what we think platforms look like in the 2004 timeframe.

It'll be based on the next generation processor, codenamed Tejas. It'll have next generation I/O from PCI Express*, and it'll have next generation graphics based on PCI Express. It'll be delivered on DDR2, enhanced security. Also what you've just seen play with the video is we've really worked hard from the research we have done to deliver small, cool, quiet, no compromise platforms into the market place. So Powersville talks about all of the things that we'll see happen in the 2004 timeframe. I actually have a Powersville platform up here, which you can come take a look at after the talks are done today.

Now, in our work in talking with consumers and business users, it also became very clear that we need to take the next leap forward in audio standards.

So today we are announcing that leap forward in a program codenamed Azalia. What Azalia is about is really improving the experience for both consumers and for business users from an audio point of view.

So let's start with the consumer space first. In the consumer space, what they're really trying to do is enable a richer experience. Everything's moved to digital. The media has to be immersive in how people work with it. So what you'll see here as a simple example, Azalia will enable simultaneous audio streams and applications flowing around the home.

I showed you earlier a simple example of a single stream and pictures flowing wirelessly to a DMA or Digital Media Adapter. Imagine the ability to have that same high performance PC, but streaming to four or five different CE devices in your home in a seamless, smooth fashion, Azalia is about enabling that type of capability. It's about enabling a better, richer home theater experience with the technology we are providing.

What you really need to think about this in the consumer space, it's about improving the immersive experience that consumers are asking us for.

In the corporate space, it's really addressing collaborations and communications. It's about improving the capability as corporations, literally compete around the globe, they need the ability to remove time and space as a barrier in that process. There needs to be a significant leap forward to improve that experience. Azalia is the answer there.

It'll improve overall collaboration in speech recognition. It does this by a significant leap forward with cost effective, host based array microphone technology. It'll deliver stable audio stacks, standardized controllers, architecture, a number of other issues, again, addressing the stability of that software and that capability.

So the timeline on this is, the spec will be available this year to the industry and first products will be shipping on it in 2004 into the marketplace. Again, another example of driving standards with the industry, which we've done so many times with 3GIO, and Serial ATA with PCI, USB, so on and so forth. Working as an industry, we can deliver the next generation audio to improve the experience for consumers and for business.

The last thing I'd like to go to is Marble Falls. Craig showed a pretty cool demo on this yesterday and what you could do from a medical research point of view.

We have brought the platform back up on stage today. The way you ought to think about this is, this is a concept platform that we do a lot of research on as a test bed, if will you. So when we said in this case, it's focused on business users, we'll actually sit with business users and watch how they engage this. This'll be the platform we deliver the first working silicon to and the array microphone technology as an example. So we'll continue working with research, continue working with the industry, trying to find and meet the needs of consumers and business users.

So what I'd like to do now is, you know, we think about our industry. We like to think about it as we hope that we have a substantial positive impact on how people work, learn, and play around the planet. And oftentimes we get caught up in all the bits and bytes and the details. What we actually put together here is a video of how our industry collectively has had that impact around the world.

So let's roll the video, please.

(Video begins and ends.)

LOUIS BURNS: I just love that video for a couple of reasons. Whether you're a tie dyed child of the '60s out of Berkeley or in this example, a kid narrating this, what we do as an industry, I do believe, improves how people work, learn, and play around the planet.

I hope I've given you some examples in the desktop space of some really cranking technology. Prescott's going to continue Intel's lead in processors. Canterwood and Springdale giving you new capabilities to give you truly immersive experiences in how people work and engage these systems.

So to summarize for today, what you've seen from Anand and Gadi and I is really about a bunch of products and a bunch of innovations that we're providing into the marketplace.

But it's not just about those products. It's all the different aspects that ultimately deliver total client solutions.

Intel's view is through a handheld to a notebook to a desktop. That's how people think about clients. They want to get their data anytime, anyplace, anywhere in the world on any device.

So once again, thanks for your time and coming and spending time with us at IDF. There's a huge number of classes that go into immense detail on many of the things we've talked about. Have a good day and a good rest of the week. Thank you.