英特爾 (INTC) 2011 Q2 法說會逐字稿

Good day, ladies and gentlemen.

Welcome to the Q2 2011 Intel Corporation's earnings call.

My name is Chastity, and I'll be your coordinator for today.

All lines have been placed on mute to prevent any background noise.

After the speakers' remarks, there will be a question-and-answer session.

(Operator Instructions).

As a reminder, this conference is being recorded for replay purposes.

I would now like to turn the presentation over to your host for today's call, Mr.

Kevin Sellers, VP of Investor Relations.

Please proceed, sir.

Thank you, Chastity, and welcome everyone to Intel's second-quarter 2011 earnings conference call.

By now you should have received a copy of our earnings release and the CFO commentary that goes along with that.

If you have not received both documents, they're currently available on our investor website, INTC.com.

I'm joined today by Paul Otellini, our President and CEO, and Stacy Smith, our Chief Financial Officer.

In a moment we will hear brief remarks from both of them.

Followed by Q&A.

I'd like to tell you about a change to our Company's guidance and quiet period practices.

Beginning this quarter, the Company's quiet period will be reduced by two weeks to enable an extended period of dialogue with investors.

Intel will observe the quiet period from September 16th until publication of the Company's third-quarter earnings release.

Additionally, certain of our smaller business outlook items, the forecasts for amortization of acquisition-related intangibles, impact of equity investments and interest and other, and tax rate, will be pulled from guidance at the close of business July 26th, and not subject to an update by the Company after that time.

We also want to make everyone aware of the annual Intel Developer Forum taking place September 13th through the 15th at the Moscone Center in San Francisco.

In addition to keynotes from Paul and other senior leaders, we'll be holding a number of investor briefings and invite you to come and hope to see many of you there.

Before we begin, let me remind everyone that today's discussion contains forward-looking statements based on the environment as we currently see it, and as such, does include risks and uncertainties.

Please refer to our press release for more information on the specific risk factors that could cause actual results to differ materially.

Also, if during this call, we use any non-GAAP financial measures or references, we will post the appropriate GAAP reconciliations to our website.

So with that, let me hand it over to Paul.

Thanks, Kevin.

I'm very pleased to report another record revenue quarter for the Company.

In May, we met with many of you at our annual investor meeting, and laid out our strategy and the opportunities we have before us.

The trends that are driving the growth of our business that we described then are playing out as expected.

The data center business remains strong, our embedded business is seeing rapid growth, the enterprise PC refresh continues, and the emerging markets continue to be significant drivers of our broad-based momentum.

In the comparison of our second quarter results versus Q2 of last year, we saw strong double-digit revenue growth across every business segment.

For example, the data center business is up 15%, embedded is up 25%, NAND is up 15%, and our PC client business is up 11%.

If we look at the channel, which is an excellent proxy for emerging market strength, our channel revenue is up 17% versus last year.

In summary, this was a very strong quarter across all of our product lines, and around the world.

Now, let me take just a moment and provide a bit of color on performance of our major business segments.

The Data Center Group had another very strong quarter, with CPU revenues topping the $2 billion mark for the third consecutive quarter.

Demand for our Westmere EX family, that was launched in April, is very healthy, and this helped to strengthen server ASPs in the quarter.

At our Investor Day, we talked a lot about how the data center business was more than just servers.

To highlight that, microprocessor unit shipments for storage applications set a record, and were up 38% from a year ago.

Also, we grew our networking revenue 40% from a year ago.

The non-server parts of our data center business are growing at an exceptional rate.

As for servers, the story is much the same.

The market remains very strong, with demand from cloud-based customers leading the way.

The cloud segment is up 50% in the first half of 2011, versus the first half of last year, demonstrating how fast that business continues to ramp.

We believe that we are very early in the cloud build out, and that Intel remains extremely well-positioned to profitably grow from the explosion of mobile devices and Internet-based services.

In our embedded business segment, I mentioned that revenues grew 25% from a year ago, on record microprocessor unit shipments.

We shipped over 1 million units of Atom processors into embedded applications for the first time ever, which is an increase of 76% from a year ago.

Overall, we witnessed very broad-based strength in our embedded segment, with the fastest growth coming from the medical imaging segment, up 50%, print imaging, up 48%, communications, up 40%, and industrial applications up 20%.

We are starting to see many of the long cycle design wins that we've been tracking turn into revenue, and expect Q3 to continue this trend.

McAfee turned in a very solid quarter, with revenue beating our expectations.

It was a second-quarter record for McAfee in terms of revenue.

Additionally, the number of large deals signed was almost double that of Q1.

McAfee is executing very well, and the importance of security has never been more evident.

We are excited that later this year Intel and McAfee will be shipping the very first product co-developed by both companies, that will greatly enhance security for computing applications.

This new solution is a combination of enhanced software from McAfee and security instructions built into the hardware from Intel.

We look forward to providing more details of this product at our developer forum in September.

Lastly, the PC client group also had a very good quarter.

Demand for Sandy Bridge processors has been very strong, with the ramp rate of Sandy Bridge ahead of the previous generation of products by 20%.

Today, two-thirds of all the products that we ship into PC clients are Core i3, i5, or i7.

Overall, the trends in PCs remain much the same as the last few quarters.

The corporate business remains consistent and healthy.

Consumer markets are mixed.

The mature market consumer segment is still soft, but the emerging market consumer segment is healthy and growing.

Our guidance for the third quarter assumes a modest growth in the mature market consumer segments of the business.

At this point in the year, we believe that PC unit growth will be around 8% to 10%, down a bit from our earlier view, but above that in revenue, as enterprise PC purchases continue to drive a very rich mix.

I previously mentioned that our channel revenue was up 17% from a year ago.

Let me add some color to our emerging market results.

In looking at the last three months of reported PC shipments, you can see why we're so optimistic about the emerging markets opportunity.

For example, Turkey and Indonesia are up over 70% each.

India is up 17%, Russia is up 15%, and China is up 14%.

The latest data on Latin America also showed growth of 12%.

Brazil remains the key driver of this growth, and is poised to become the third-largest market for computers in 2012.

In general, all of these are large, high-growth regions where PC penetration remains very low while affordability of PCs continues to improve.

In closing, I want to highlight two important announcements made this quarter.

The first was the process technology breakthrough that we made public back in May.

Intel will be the first company to productize 3D transistors.

Our 22-nanometer process technology remains on-track for high volume production later this year, and will provide substantial improvements in power consumption and performance.

This technology is very complex and very innovative, such that we expect that the lead we have in process technology will expand going forward, providing Intel with important competitive advantages in the markets we're targeting.

Second was the Ultrabook announcement we made at Computex Ultrabooks are the next evolution in personal computing that will combine the best of the consumer PC with tablet functionality, as well as consumer electronics features, like instant on, and always-on, always-connected functionality, all in a stylish, thin and light system.

We are very pleased at the industry response and customer commitments around this new product category, and we'll provide more details at our September development forum.

With that, let me turn the meeting over to Stacy.

Thanks, Paul.

The second quarter was strong and better than expected, resulting in our first-ever quarter with greater than $13 billion in revenue.

From a market standpoint, the quarter played out as expected, with strength in emerging markets and enterprise, offset by softness in the mature market consumer segment.

We continue to benefit from the strength of our product lineup, as mix within both servers and clients was better than expected.

Demand for our latest product, Sandy Bridge, remains strong, and is the fastest-ramping product in our Company's history.

As a result of the acquisitions of McAfee and the Infineon wireless division, we will provide non-GAAP financial information in addition to GAAP for 2011 to provide additional visibility into the operational results of the Company.

On a non-GAAP basis, we achieved record revenue of $13.1 billion.

Gross margin of $8.1 billion was also a record, with a gross margin of 62%.

Operating profit of $4.2 billion was up 6% from a year ago.

Net income of $3.2 billion and earnings per share of $0.59 were up 10% and 16% respectively, from a year ago.

The rest of my comments will use GAAP financials, unless otherwise called out.

Our business growth has accelerated over the past few years.

Second quarter revenue of over $13 billion is up approximately $2.3 billion when compared to the second quarter of 2010, equating to 21% growth.

The acquisitions of McAfee and the Infineon wireless divisions contributed approximately $1 billion of the $2.3 billion increase from a year ago.

Gross margin in the second quarter was 61%, slightly down from the first quarter.

As expected, our 22-nanometer startup costs peaked in the second quarter and our costs were slightly higher as we ramped incremental factories on our 32-nanometer process technology.

This was mostly offset by the reserves taken in the first quarter for the Cougar Point product issue, which were significantly less in the second quarter.

As we look forward to the third quarter of 2011, we are forecasting the midpoint of the revenue range at $14 billion, up 7% to the second quarter, and in-line with historical seasonality.

We are forecasting the midpoint of the gross margin range to be up from the second quarter at 64%.

We are raising the midpoint of our outlook for R&D and MG&A by $500 million to $16.2 billion.

Some of this increase is due to a significant increase in our business, but a larger portion is driven by investments we are making in innovation to drive future revenue growth.

Our investments are focused on redefining the PC category with Ultrabooks, our server product offerings, strengthening our product offering in tablets and phones, and further extending our process technology leadership.

Our long-term goal here is to make these important investments that contribute to our growth, while still maintaining the progress we have made on efficiency and keeping spending as a percent of revenue below 30%.

The cash generation of our business remains strong, with cash flow from operations of $4 billion in the second quarter.

On top of investing in our business, we are increasing the amount of free cash flow allocated to the dividend, and actively repurchasing shares, consistent with the philosophy we outlined in May.

This was apparent in the second quarter, as we purchased $2.5 billion in capital, and are forecasting CapEx of $10.5 billion for the year.

Completed a $2 billion buyback, and announced a 16% increase to our dividend.

Our Intel-owned inventory was approximately flat to the prior quarter, and the inventory that sits at our distributors was down slightly this quarter.

As we look more broadly at the worldwide PC supply chain, we believe inventories are at a healthy level, and did not grow this quarter.

As Paul mentioned, our business results in the second quarter continued to reinforce the key themes highlighted at our May investor meeting.

The explosion of devices that compute and connect to the Internet continues to drive the build out of the cloud, and fuel our Data Center Group growth.

Our embedded business is growing at a rapid pace, as the demand for both performance and low power from the Internet of things drives architectural conversions for Intel.

Emerging markets already make up more than half of our revenue, and as the technology that we sell becomes more affordable to billions of consumers, we will continue to see robust growth from emerging markets.

These are the market forces that drove first-half revenue growth of 23% from the first half of 2010, and these are the market forces that give us confidence as we enter the second half of 2011.

With that, let me turn it back over to Kevin.

Okay.

Thanks, Stacy and Paul.

We'll now move to Q&A.

And as we usually do, we would like to have each participant ask just one question and one follow-up if you would to allow as many participants as possible.

Chastity, we'll take our first question.

Thank you.

(Operator Instructions).

Our first question comes from the line of Ross Seymore with Deutsche Bank.

Congrats on the strong results.

Just a question on your total PC unit forecast that you're looking at.

I guess if you're not really looking for anything different than seasonal, and your full-year guidance appears to be about the same, how do we reconcile a weaker addressable market to the fact that your guidance has not changed?

Our guidance is in revenue dollars.

And I said pretty clearly that one of the things we're seeing is less strength in the netbook segment and stronger strength in the overall mix of PCs.

In fact, our PC client business, notebooks and desktops, non-netbooks, is running double digits.

Great.

I guess on that front, in the PC client side of things, looks like the Core mix keeps rising.

It's about 20% I believe of mix now.

How high do you think that can go and how long do you think it would take to get there?

I think I said that two-thirds of our PC shipments were Core i3, i5 and i7, the branded portions of the product line.

In other words, it wasn't 20%.

It was 6.5%.

My error.

Sorry.

Answer your question of going forward, you can kind of see it in the gross margin recount for Q3, we're expecting pricing to be relatively flat.

It says kind of stable from here.

It surprised us with our strength, but we believe that strength continues through this year.

Thank you.

Thanks, Ross.

Thank you.

Our next question comes from the line of Patrick Wang with Evercore.

Patrick, are you there?

Can you hear me?

Go ahead.

Sorry about that.

Hey, so congrats on the great results here.

I just want to see if you could maybe give us some milestones we should look for over the next couple of -- over the next quarter or so, in terms of how your core business is growing in the these emerging markets, geographies.

The next quarter, I mean, the next milestone is our -- is three months from today.

Not specifically next three months but next few months but just over the next couple quarters here, maybe help us get a better sense of how to actually track those -- the growth in those areas.

Well, I mean, there's no better way of understanding than putting feet on the street, to the extent that you have the opportunity to go look at them.

I was just in Brazil, and the fact that Brazil will become the third-largest country market in the world for computers next year is pretty astounding.

It wasn't too long ago when they were number six or seven.

And so it's the real-time dynamic of these markets waking up, increase in disposable income, a decrease in the cost of computing, a decrease in the cost of bandwidth and connectivity, all coming together.

There's no one indicator I could point you to, because as we've discussed before, some of this stuff is below the radar screen of some of the third-party trackers, but we see it real-time, because we've got feet on the street, and we're serving the market.

Where you see it pretty clearly is in our overall results.

The emerging markets have been the big driver of incremental units.

The last few years, our prediction is that just continues as we go through the next few years.

That's helpful.

I know you guys have also shared that, the graph in terms of number of weeks it takes to actually buy a PC.

I guess my second question is also on the same topic here.

In your emerging markets, what kind of product mix are you looking for the strength there, and I know that you guys are also pushing pretty hard for Ultrabooks.

Just curious how that impacts your mix over time.

The strength, the mix in emerging markets really historically is not much different than our overall mix, and every emerging market is different.

For example, the Tier 1 and Tier 2 cities in China have a slightly richer mix than the United States on average.

When you average China out across all the cities though, you get a closer to average mix, maybe slightly down as you move into the tier 5 and tier 6 cities.

We really haven't seen the mix go down, or the pricing be dramatically different in these emerging markets, because just like as you would expect, when it's your first time to buy a television, or the first time to buy a computer, you want value.

You want something that's going to last, that's going to be good for your family for more than a year or two, and that tends to have you buy up a bit.

I see.

Got it.

Thanks so much.

Good luck, guys.

Thanks, Patrick.

Thank you.

Our next question comes from the line of Sumit Dhanda with Citadel.

Yes.

Hi, Stacy, hi, Paul.

A couple of questions.

First, on the geographic trends, seems like I hear on the mature markets that the Americas seem to do a lot better than Europe in the quarter if the you just look at sequential sales growth.

Any thoughts on the dichotomy there, especially given that North America had a tough comp, as did Europe with the extra week in Q1?

You have to be a little careful of that data because it's billings data, Sumit, and now we're in a situation that the worldwide supply chain builds and builds most of the notebooks in APAC, so I wouldn't draw too much of a conclusion from the billings data.

I do think we would say that the consumer in North America was a smidgen stronger than the consumer in Europe in the second quarter, and emerging markets across both regions was strong.

And enterprise PC sales --

Is strong.

And the server sales in North America are good.

Yes.

And that's a good point that Paul made.

You tend to see more of the servers billed directly in the market.

That also probably drives the North America numbers a little bit.

Okay, and for my follow-up Stacy, your comment that ASPs will be flat in 3Q despite that being typically the stronger consumer quarter, is that more an enterprise-driven phenomenon, i.e., PC Client business, or do you think servers will be disproportionately strong in the third quarter?

No, it's more of an overall mix within the different businesses, within the PC business and within the server business.

If I look back at what I've been predicting over the last several quarters, what I've been wrong on is that our mix to the Core product line, Core i3, i5, i7, has been higher than I've anticipated, and so, even with a bit of a seasonal uptick in the consumer segment, I think that rich mix keeps me approximately flat from the standpoint of ASP.

Okay.

Thank you very much.

Thank you.

Our next question comes from the line of Glen Yeung with Citi.

Thank you.

If we look at your third-quarter guidance, can you give us a sense as to what parts of your business are above your average sequential growth, and what are below?

I'm just taking the major groups, PC client, data center, other IA and software and services.

I'd say across the major businesses, what we're expecting is seasonal and so what it is, is kind of a continuation of the trends we've seen so far.

We've seen relative strength in enterprise, and we've seen relative strength in emerging markets.

We think that continues into the back half of this year.

We've seen relative softness in the consumer segment, but looking back tat a decade of data, in every year, be it a strong year or a weak year, we see the second half stronger than the first half on a unit basis, just because that's the buying season, both for enterprise and consumer, and so off of that weak first half, we think there's a seasonal uptick in the second half.

So nothing significantly different from a normal trend.

Some just starting from a weak starting point, and some starting from a strong starting point.

Okay, fair enough.

And then, the other question I have is, when you look into the emerging market client business, is there a difference there in the way inventories are managed, either from the perspective of overall PC inventories or from the perspective of standalone microprocessors?

Are they harder to track, easier to rack?

Do you feel like they have control over that, or they don't?

Actually, it's probably slightly better than we have in North America or Western Europe, and we have the same visibility into our multi-national accounts whether they ship in China or in Cleveland.

That is, we run these off a big hub, it's a system that basically they only draw what they need for their near-term builds.

In fact, we're seeing a conservatism there in terms of people moving a bit more to the air lanes versus shipping lanes, just to be conservative.

In general, we feel pretty good about that part of the inventory.

In emerging markets in general, a larger fraction of sales comes through the channel.

A lot of that is white box and build to order, and by definition, build to order, someone walks in and gives someone $200, there's a computer that afternoon.

So we get a much more real-time feed there, and of course we don't claim that revenue until it sells out through our distributors.

So if anything, it may be slightly more real-time than the mature markets or the enterprise market.

That's great.

Thanks, Paul.

Thank you.

Our next question comes from the line of John Pitzer with Credit Suisse.

Yes, guys, congratulations.

Paul, you talked about the percent of the client business which is sitting at Core today.

I think you said two-thirds.

I'm kind of curious, where are we now with Sandy Bridge, as a percent of Core?

As you go from a Core i3 non-Sandy Bridge to Sandy Bridge, is that both an ASP and a gross margin accretive event for you guys?

We're taking Sandy Bridge in the second half of the year down below the Core brand and taking versions of that into Pentium and Celeron, and as a percent of our Core business today, it's over half.

What was the second part of the question, John?

I'm sorry.

As you go from sort a Core non-Sandy Bridge to a Core Sandy Bridge, is that both an ASP and gross margin accretive event for you guys?

If we sell the same SKU at the same price, it's just slightly better, because you're selling one chip instead of the CPU plus chipset, but what we're seeing with Sandy Bridge is a trend to sell up a bit for every SKU, because of the demand for the performance and the graphics capability.

And then Paul, as my follow-up, can you talk a little on DCG, some of the channel checks we've done around customers' expectations around Romley are pretty high.

I'm kind of curious, time line on Romley, and as you think about introducing that new server architecture, do you think that's going to cause some stall-out in the DCG group, or how should we think about that for the back half of the year?

There's really no change for the schedule for Romley, which for those of you who don't follow our code names is the Sandy Bridge version of the server products.

We are public in saying that we'll be in production this year.

Nothing's changed.

We expect -- still expect to be in volume production this year.

We have not gone public on the launch date, because our server customers are very conservative about putting the dates out for exactly the reason you point out.

I don't expect to stall-out, given the timing that we've built into the product.

Thanks, guys.

I would add that as we've seen the product, it looks very exciting.

The performance is really quite good.

Thank you.

Thank you.

Our next question comes from Daniel Berenbaum with MKM Partners.

Yes, hi guys.

Thanks for taking the question.

Just wanted to revisit ASPs real quick.

With ASPs up year-over-year, how much of that is being driven by the client platforms and the branding that you've driven there, how much is being driven by pure mix in client, and how much is being driven by shift in the enterprise platforms?

I don't think I can separate out the branding from the mix within client.

I think those things are synonymous.

I think that we've spent a lot of energy to be able to articulate, differentiate it, features and performance at different price points.

That's a tool that we hadn't historically used, and that, in conjunction with having a great product like Sandy Bridge and the kinds of performance that brings, has led to the richer mix within the client segment.

I think that in general, that's the bigger driver, and then the fact that the enterprise market has been relatively stronger than the consumer market is a secondary driver to the ASP good news that we've seen.

Okay.

And then, you talk a lot about emerging markets and I just want to make sure I -- emerging markets versus enterprise.

How much of emerging markets is enterprise, or do you have more or less visibility into what the split is between, say, consumer and small-medium business in emerging markets?

We won't get to that level of detail in this call.

In general, the business segment in emerging markets is a larger percentage of the total, but as the affordability of technology comes down like we showed at the investor meeting, we actually expect a lot of the growth to come in the consumer segment of those markets.

It's going to be a faster part of the growth equation.

Okay.

Great, thanks very much.

Thank you.

Our next question comes from the line of Doug Freedman with Gleacher & Company.

Hi, guys.

Thanks for taking my questions, congratulations on the strong results.

Stacy, if I could dig in a little bit on the CapEx plan and if the little bit of pull-in that you're showing -- should I characterize the slight increase in CapEx this year as a pull-in from next year, given the slide that you put up at the Analyst Day?

Can you help offer some color there?

Sure.

Some of it is a pull-in.

Some of it is that we're going faster.

So this is really construction-related spending.

Some of it is that we're going a little faster on the 14-nanometer factories.

The larger portion of it, and keep in mind we're talking about a relatively small increment, but the larger portion of it is an increase in the scope of those factories.

As we got into the planning, we saw some opportunities to do some things in the infrastructure that we're putting in place that make it easier for those factories to support multiple generations behind 14-nanometer, so kind of putting in place the capability for 10 and 7-nanometer, and we think it's a kind of high ROI to do that now versus having to retrofit those factories later.

It's a combination of those two things but if I was ordering them, the capability is driving a little bit more and the speed is driving a little bit less.

Terrific.

Thanks for that color.

If I could look at what's going on near-term with the OpEx number, this is one of the few times I think I've seen you guys grow spending faster than you're forecasting revenue growth from that perspective.

How should we think about revenue growth and the fall-through to EPS if we were to just be modeling your revenue?

Is there a ratio we should think about that you're targeting to drop through, or anything you can offer as color as far as understanding those two numbers?

Yes, I mean, the simple answer there is what I said at the investor meeting, which is Paul and I are still committed to drive spending as a percent of revenue at or below 30% is our long-term goal.

And so I do think we will over the long-term get some leverage in that line item, but we have a couple of things going on now.

We see very fast growth in revenues.

Some of this is revenue-dependent.

When I just step back and look, we grew from 2009 to 2010, we grew the business $8.5 billion.

From 2010 to 2011, we're on track to grow the business in excess of $10 billion again, and so that revenue and profit-dependent spending is a piece of this.

But I think more importantly, we're making some important investments in our business that we believe give us an even better position as we go forward, and we talked at the investor meeting about redefining the PC category with Ultrabooks, so we're making investments there.

We're making investments in our server business to continue and accelerate the growth rate in that business, and we're making investments in our tablet and phone business, in order to improve the offering that we can offer to customers there, and Paul can talk a bit about that.

It's a combination of those two things, but right now, those investments are important.

They don't change our long-term view of the business, but they're important investments for us to make.

Great.

Thanks for the color.

Thanks, Doug.

Thank you.

Our next question comes from the line of Craig Berger with FBR.

Hi, guys.

Thanks for taking the question.

Can you, just on the CapEx, help us understand how much of that is sort of for your equipment, versus how much of that is longer-term halls and walls and how long are your halls and walls depreciated over?

Of the increment, it's almost all construction and that's -- you can have varying depreciation schedules, but think of that as depreciating over a long period of time.

We tend to use buildings for a decade.

That's part of the reason that we build in the capabilities to support multiple generations in a factory.

Equipment, it tends to depreciate on a much shorter period of time.

Again, it can vary, but think of that over a handful of years.

Can you talk about CapEx spending like that for the year?

Not the increment, but the total?

Probably not at the level of detail that you would want.

But if you look at the CapEx that we're doing this year, we said at the investor meeting more of that, much more of that is being driven by building factories than what is normal for us.

Andy talked about that in his investor meeting presentation.

I would refer you back to that.

It's been a long time since we've had to build incremental shelves, but the growth in our business now requires it, so we're putting in place some clean room space that we haven't had to do over the last couple of generations.

It's a larger percent, but I'm not going to break out the exact percent between the two.

Thank you for that detail, and then just a follow-up.

Gross margin, Q2 to Q3, I know you've gone through it a little bit, but can you just help us understand again why there's greater than 100% revenue fall-through?

Yes.

Why don't I walk you through sort of the gross margin.

The short answer to your question is, we're starting to get the benefit of the tapering off of other cost of sales and so that tends to give us higher than normal fall-through.

But if I step back, give you the reconciliation of Q2 to Q3 and talk a little bit about what I expect for the year, as we go from the second to third quarter, gross margin is going to be up about 3.5 points, so that takes us up to 64%.

The 3.5 points comes from -- we expect it to be a higher volume quarter.

We talked about that.

We start to see as consistent with our historical patterns, we start to see the falling off of start-up costs, so that adds about a point.

We get about a point because we're now projecting to be finished with the impact of the Cougar Point reserves, so we don't have to repeat that in the third quarter and then there's about a half a point as we continue up the ramp of the various 32-nanometer factories, our costs start to improve.

We talked about that on prior calls.

So that's what gives us the benefit into the third quarter.

And then as I think about the fourth quarter, and you can kind of get the math when you look at our annual forecast, I expect that we get a little bit more benefit in terms of start-up costs.

You tend to see that come down over a couple of quarters.

And so as I think about the fourth-quarter gross margin, I would put it in the range of a bit higher than the 64, probably based on what I know today, below the 65% that we've articulated at the high end of our range.

So kind of in that zone.

Thank you so much.

You're welcome.

Thank you.

Our next question comes from the line of Michael McConnell with Pacific Crest Securities.

Thank you.

Stacy, I had a question just looking at the segmented revenues here on PC Client.

Looks like you had growth in this business of about 11% year-over-year, yet operating profits were down about 1% year-over-year.

Could you talk about the disconnect there with ASPs being as firm as they are?

It's start-up costs, Michael.

If you think about our trend, you're on a two-year cadence, so doing this year-over-year comparison from Q2 2011 to Q2 2010, what you're doing is you're comparing our lowest point in start-up costs to our highest point in start-up costs, because they start coming down the pier.

If you think about it on a Company-wide basis, some of that is DCG as well, but if you think of that on a Company-wide basis, that's kind of 3.5 points of gross margin on a year-over-year comparison, so it's pretty significant.

Right.

That makes sense.

Okay.

And then I wanted to ask, just looks like the upside driver relative to your expectations within the other group, the other Intel architecture group, and I think Infineon was about $90 million upside versus your expectations, so understand that point and the driver there.

What was the other areas I guess that were considerably strong, and what's the outlook for that particular segment as we move into the back half of the year?

Yes, and I apologize, I'm not sure I understand your question.

If I look at what was different the expectations when we started the quarter, Infineon actually came in right where we expected.

They had a good quarter, but in line with the expectations.

The other elements in there were pretty much in line.

The difference from expectations was our ASP was a little bit higher, our mix was a little bit better, and so -- that was more of a PC and server comment than an other Intel architecture group comment.

Okay.

Thanks for the clarification.

Thank you.

Our next question comes from the line of Shawn Webster with Macquarie.

Great.

Thank you very much.

A simple one, and then a more complicated one on CapEx.

So on the simple question, what did your you units do sequentially for processors?

Were they down 3%?

Down 1%?

Flat?

On a sequential basis it's a little bit complex, Shawn, because remember, we had the 14th week in Q1.

If I strip out the 14th week, I would articulate it as units versus seasonal, and you normally see at Q2, that's just a little bit down from the first quarter.

Okay.

Thank you.

On capital expenditures, high-level question.

If your outlook for PCs is coming down I'm wondering why we didn't see CapEx be flat or maybe down some, and then maybe you could share with us your expectation for capital intensity, do you expect it to get back to the historical kind of 35% rate next year?

Thank you.

Yes, remember that capital -- the capital forecast that we put in place is in support of multiple generations.

We don't buy capital in Q2 to support what we're doing in 2011, so capital we're buying is 22-nanometer peak.

It's 14-nanometer factories.

You'll soon see us moving into the cycle of 14-nanometer equipment.

It's a longer-term view.

In terms of capital intensity, it's the same data that we shared with you at the investor meeting.

Thanks, Shawn.

Thank you.

Our next question comes from the line of Uche Orji with UBS.

Thank you very much.

First of all, congratulations, Paul and team for the work done.

Paul, let me just talk -- ask you about Ultrabooks.

Sean was quoted at Computex as saying they should be about 40% of the consumer mix by the end of next year.

If one were to go back and compare this to CULV from a couple of years ago, what makes you more confident that this will achieve this level of success, beyond the form factor.

I think CULV had performance issues.

If you could talk about the level of confidence you have with Ultrabooks, and how you see that ramp from now to next year to get to that 40%?

As I look at this, I don't think that the Ultrabook strategy is anywhere near equal to the CULV strategy, as you call it.

That was really kind of a point product.

It was focused on form factor.

We didn't really put a lot of engineering into it with our customers, and we didn't look at other features.

If you will, it was kind of a trial run in hindsight is the way I would look at it.

The Ultrabook project is much more akin to Centrino.

It's a very holistic approach to moving the entire market to a different kind of form factor, not just in terms of its thinness, but in terms of the feature set.

I talked about always-on, always-connected.

The machine is always aware of the networks around it.

We talked about instant-on, instant-boot capability.

We talked about building in integral touch into it, another feature set.

This is as much about the features around the skin or inside the skin as the shape of the skin, and as we look at this with our customers, we also see that there's a great deal of engineering that has to be done.

One thing we know, today these feature sets cost more money.

We don't think the PC prices are going to go up over time.

So what we have to do is work with the ecosystem to cost-engineer these features for high volume price point displacement.

And that's the only way you can achieve sort of a 40% number, as Sean predicted, in that time frame, is by doing price point replacements and then looking forward a year later into the next generation silicon.

It gets cheaper to do it, so we can penetrate more of the market.

Okay.

Thanks for that.

Can I ask a follow-up on another way to look to at ASP resilience, in your CPU business on the client side, how much of that is beyond just mix?

How much of that is also the ability to take more material share, and specifically I just want to know if the things like rapid attach rates has changed within the integrations you're seeing and therefore you're capturing more of that value, or is it a capability of Intel to take advantage of the fact that components like memory are a lot softer, therefore there isn't much pressure to take CPU pricing down by your customers.

Well, no, it's really demand-driven is the way I would look at it.

End customers are demanding a higher mix of products, which gets into the branding and feature-set question that came up earlier.

In terms of bill of materials, we had a very strong position in integrated graphics prior to Sandy Bridge.

That just got stronger with Sandy Bridge.

It satisfies more and more of the market.

Customers are using it particularly in smaller form factor machines, because of the power advantage.

There's no compromise on pixels, as you get a significant advantage on battery life, and as a result, we're seeing a very strong utilization both in desktops and notebooks for Sandy Bridge graphics.

Thank you very much.

Thank you.

Our next question comes from the line of Christopher Danely with JPMorgan.

Hey, thanks, guys.

Hey, Paul, I was wondering if you could just kind of take a step back and give us your latest thoughts on tablets and ultimately what you think it will mean and do to the PC market, what it means to Intel and what sort of share goals you have in tablets?

Chris, my thoughts really haven't changed from April.

Even after I saw the numbers yesterday from Apple, and we get engaged in lots of Win 8 and Android tablet design activity across the board, I believe this category is additive to computing.

I don't think it's going to replace any one category.

I think it may replace some discretionary sales at any point in time, and I think that's one of the reasons the netbook's down year on year is the tablet strength to some extent.

The netbook's probably also down because of the very good values you can get in low-end notebooks nowadays, as compared to say, a netbook.

There's a number of pressures on that particular segment.

Going forward, I really don't see that.

I think it's a device that some people will use as a standalone device, but many people use it as a companion device and that's good for us and that seems to be the model, even if you look at the Apple numbers, it seems to be the model that's consistent with their base.

Thanks.

As my follow-up, just some quick modeling questions.

Going forward, what should we be thinking about share count and McAfee/Infineon revenue?

Those are two very different follow-up questions.

You're picking up from our esteemed Tim Luke.

Next time I'll say it with a British accent.

I'd love to hear you try.

From a share count standpoint, I think you know that's something we're not going to comment on.

I can kind of tie back our philosophy to our actions in Q2, if you think that's helpful to you.

We articulated the philosophy in the May investor meeting of first investing in our business, no change there, dividends second, and we articulated that we wanted to allocate more of our free cash flow to the dividend, making that approximately 40% of our free cash flow, and then using the remainder of cash generation, one of the things we look at there is the buyback.

I think you can see that played out nicely in the second quarter.

We kept our cash balances relatively flat.

We increased the dividend 16%.

We made significant investments in our business, including the CapEx number, and we bought back $2 billion worth of stock with the free cash flow that we generated that was in excess of what we needed for those other uses.

I think that's a nice model for us and very consistent with our philosophy.

In terms of McAfee and Infineon, what was your specific question there?

I guess what you guys think or what should we be thinking on the revenue going forward, if there's any sort of seasonality there?

It doesn't change the seasonality for the Company as far as I can tell.

So I haven't seen anything there.

If you look back, again, I commented on this earlier, if you look back across the last decade, we see in every environment the second-half revenue is a few points higher than the first-half revenue and McAfee and Infineon don't change that pattern for us.

Okay.

Thanks a lot.

That's very helpful.

Thank you.

Our next question comes from the line of Vivek Arya with Banc of America.

Thanks for taking my question.

Paul, I'm curious, how will Intel be impacted by the launch of Windows 8?

Is that a net positive, or do you think that changes competitive landscape, like what are the puts and takes that we should think about?

Well, we talked about this an awful lot at our investor meeting, so I'm not going to give you a lot more color than we gave there, except, in general I think it's good.

Whenever Microsoft has a new operating system, that's good for us.

We've been working with Microsoft for several years now on the feature set of Win 8 and optimizing our silicon around it in PC space.

I think the jury is out on how that will be received by the public and tablets.

And we expect to be hyper-competitive for Win 8 tablets.

We're involved in the program as aggressively as any of the silicon vendors and our job is to make sure that we can out-perform, out-battery and out-compatibility the Arm guys in Win 8 tablets.

And just as a follow-up on the near term, I think you mentioned the 8% to 10% unit forecast, slightly different than before.

Could you talk about the fact that -- I think you mentioned the netbooks, but those have been weak for some time.

Are there any other dynamics at play?

Is it a macro issue?

Are there other factors we should think about?

The principal change is the net book.

As I said earlier, our core, our main stream PC and notebook -- notebook and desktop business is still forecasted for us at double digits, and we're very comfortable with that part of the product line.

The biggest change from six months ago is a slower growth rate on netbooks.

They're still showing growth in netbooks, just at a slower rate, and that has impacts on ASP.

It also has impacts on overall volume.

Thank you.

Thank you.

Our next question comes from the line of Jim Covello with Goldman Sachs.

Great.

Thanks so much for taking the question and congratulations on the great execution.

The question I would have is, is there a way to quantify the impact of unit growth next year on the capacity side?

How much will unit capacity be up in 2012, as a result of the CapEx in 2011?

I'm thinking about how to answer that question, Jim.

As we've said, we're putting in capacity that's pretty consistent with the kind of TAM growth we've seen for the last several years, and we have some ability to flex up so we're making sure we're putting in some factory space to go a little bit higher than that, and we have the ability that if things end up being softer than that, as you saw in 2009, to respond pretty quickly and bring down the CapEx number and just upgrade and bring in more capital at the leading edge node.

Think of it as pretty consistent with historical unit growth rates with some ability to go higher than that, and some flexibility to respond if things end up being a little softer than that.

I think Andy did a great job of articulating that in the investor meeting.

If you want a refresh of that, we can take you through it.

I remember, he talked about how you can kind of shut it down pretty quickly.

So to that end, if I can use my follow-up on the same topic.

If the PC unit forecast that you guys see keeps coming down, is there a number that we could think about that we could then expect you to kind of go more on the slower side on the CapEx, as we head into next year, if it came down the five which is where the third parties are, maybe the third parties aren't right, but if your forecast kept coming down, is that -- is the market number, the third-party number a number where we would expect the CapEx to slow a little bit?

Well, it would certainly be much less than what it would be if the market growth is 12, but I realize that's not a terribly helpful answer to you.

But at this point in the year, that's as far as I'm going to go in terms of forecasting CapEx for next year.

Andy also articulated, when we're building factories and we're equipping those factories, those tend to be a couple-year cycle, so he gave you a strong hint to expect elevated CapEx next year relative to the historical trend line.

If the unit growth is significantly less than we're anticipating, that number would be less than what we're anticipating.

If it's more, it's more.

That doesn't help you, because we haven't told you the baseline.

There's a third factor, Jim, I would throw into here, which is that increasingly, our product line is more and more modular.

As we move from Sandy Bridge to Ivy Bridge next year, we have the ability to mix and match cores and mix and match graphic engines much more so than we've ever done before.

If we end up in a corner case that you're describing where the market short-term is much smaller than our ability to supply, we could up-crank our products pretty substantially, at very little marginal cost, and really have a much more across the board competitive offering than even we have today.

It's really helpful.

Thank you so much.

There's lots of tools.

It's interesting.

There's lots of tools here at our disposal that we haven't historically had.

The forward re-use of our capacity gives us the ability to move capacity quickly to the leading edge.

Five years ago, that isn't a capability we had.

As Paul said, the modularity of our architecture now gives us a lot of flexibility to kind of mix and match the capacity to the market.

Thanks.

Chastity, we're going to take one more question, if we could.

Our final question comes from the line of David Wong with Wells Fargo.

Thanks very much.

I would like to think that my accent qualifies me much better to be Tim Luke.

David Wong/Tim Luke.

Thank you.

Thank you.

David Luke.

Can you -- Stacy, can you go through the gross margin puts and takes as we go from the September to the December quarter, please?

Sure.

I won't do it at the level of detail, any more than I said in terms of, I expect to be kind of higher than what we are, and under the upper range of 65% that we put out there.

When I think about the fourth quarter, I think we benefit from start-up costs coming down.

A strong fourth-quarter gives us a couple of other benefits, and I think the big variable for us there is, we have a lot of product qualifications that are kind of teed up across the end of this year and the beginning of next year, and where those product qualifications fall could dictate where the gross margin ends up.

So in terms of variability, I think that's probably the biggest variable for Q4.

Great.

Thanks.

And for my follow-up, software and services, it looks to me like the operating margin's still negative despite the full quarter of McAfee.

Is that a normalized level of profitability, or are there special accounting effects that are still there that will go away over the next few quarters, and if so, can you give us an idea of the underlying operating margin that you've got in the division at the moment?

The revenue-related accounting impact is in that segment so you see that in the software and services group segment, the amortization of acquisition-related intangibles is outside of that segment, so it's a little bit of a mix.

That's hitting in the corporate line.

I think what you're really seeing there -- part of this will improve as we get the deferred revenue back from the accounting impact.

That should be sometime next year.

And you can see us already building up that deferred revenue balance on our balance sheet.

So I would encourage you to kind of watch that line item.

It's not something you've typically had to do within Intel.

The other pieces, remember, there's still a lot of investment inside that software and services group that's in support of tablets, and in support of phones and things like that, so there's a lot of investment in there that causes the overall segment results to be as you say.

McAfee, if you just isolated that, looks quite good.

Great.

Thanks very much.

You're welcome.

Thanks, David and thank you all for joining our conference call today.

As a reminder, our third-quarter earnings conference call is scheduled for Tuesday, October 18th, 2011.

Thanks and have a good night.

Thank you for your participation in today's conference.

This concludes the presentation.

You may now disconnect.

Have a great day.