英特爾 (INTC) 2006 Q2 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Q2 2006 Intel Corporation earnings conference call.

[OPERATOR INSTRUCTIONS]

I would now like to turn the presentation over to your host for today's call Mr. Alex Lenke, acting Director of Investor Relations.

Please proceed, sir.

Thank you.

Welcome to the Intel second quarter earnings conference call.

Attending from Intel are CEO Paul Otellini and CFO Andy Bryant.

Before we begin, please bear with me while I read our Safe Harbor language.

The second quarter earnings report discusses Intel's business outlook and contains forward-looking statements.

These particular forward-looking statements and all other statements that may be made on this call, that are not historical facts, are subject to a number of risks and uncertainties and actual results may differ materially.

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

The specific forward-looking statements cover expectations for product mix and demand, revenue, gross margin, expenses, tax rates, interest, and other income, capital spending, depreciation and amortization of acquisition related intangibles and cost.

These statements do not reflect the potential impact of any mergers, acquisitions, divestitures, investments, or other business combinations that may be completed after July 18, 2006.

Lastly, if during this call we use any non-GAAP financial measure as defined by the SEC and Reg G, you will find on our website www.INTC.com, the required reconciliation to the most directly comparable GAAP financial measure.

With that, let me turn it over to Paul.

Thanks, Alex.

In April, I said that the second quarter would be a time to work with our customers on inventory levels and prepare to launch a new generation of microprocessors.

We achieved those objectives during Q2 and are well positioned for the second half.

Second quarter revenue was in the range we said in April with microprocessor unit shipments meeting our expectations, but with pricing more competitive.

We believe that we lost a bit of microprocessor market segment share on a billings basis, but that we gained a bit of a share on a consumption basis adjusting for inventory reductions.

Intel's factories continue to lead the industry in the conversion to new technology.

We opened our third 65 nanometer facility in Q2 with over 50% of our microprocessor wafer starts now on Dual Core products.

In June, we initiated a major transition to the new core microarchitecture.

We launched Woodcrest, the world's best processor for two way volume servers that is delivering clear, platform leadership in both performance and performance per watt.

For multi-processor servers, we pulled in the Tulsa processor launch by two quarters to Q3 and have begun shipping that product for revenue.

We expect Tulsa based platforms to deliver industry-leading transaction performance in X-86 servers.

Yesterday, we launched our next generation Itanium processor code name Montecito, which uses Dual Core technology to double Itanium processor performance at 20% lower power.

Additionally, we notified customers that we are pulling in both a desktop and a server version of our first quad core processors into the fourth quarter of this year from the first half of 2007.

In the PC segment, we are very excited about Conroe, our upcoming Core 2 Duo microprocessor for the desktop, which offers 40% more performance and 40% lower power than our previous best.

Conroe will be launched on July 27th and is already setting new PC performance and performance per watt records across the board.

Merom, our Core 2 Duo processor for mobile PCs will extend our lead in mobile computing and has planned to ramp very quickly as it fits into the existing Centrino Duo platform.

Both Conroe and Merom have now qualified and have shipped for revenue.

Each is approximately one month ahead of the schedules that we laid out in our analyst meeting in April.

In total, we have shipped 5 new Dual Core microprocessors in the last 30 days, delivering new levels of energy efficient performance into every segment we serve.

Intel's chipset business had a good quarter.

And as 65 nanometer ramps, chipsets are rapidly moving to 90 nanometer on 300 millimeter wafers, giving us both better products and greater capacity.

The Broadwater chipset for Core 2 Duo desktops began shipping in early June with very healthy yields and a record ramp.

This chipset is the industry's first on 90 nanometers and it includes a version with programmable fourth generation graphics, along with advanced features for manageability, power management and has significantly improved memory controller.

There has been much speculation about pricing changes, and I'd like to provide you some context.

The introduction of the Core 2 Duo processors enabled us to reset our entire microprocessor line-up, giving us a three-tiered brand strategy of good, better, best.

Core 2 Duo is the best PC microprocessor in the world, and we will ramp it as rapidly as possible.

Our leadership with Core 2 Duo allows us to lower the price of Pentium, which is the best know processor brand in the world, to system price points not previously addressed by this brand.

Celeron remains Intel's brand for extremely cost sensitive designs.

We believe that these three brands, combined with our lead in 65 nanometer technology and capacity give Intel a unique competitive advantage.

We will accelerate Dual Core penetration into more segments than ever before giving Intel a compelling technology advantage at every price point.

Turning to operations.

We have finalized our plans to take $1 billion out of Intel projected spending in 2006.

At the analyst meeting in April, I also spoke about a comprehensive structural review project designed to produce a leaner and more agile company for 2007 and beyond.

One result of this program is the planned sale of our applications and communication processor business to Marvell announced last month.

In addition, last week, we announced the difficult action of eliminating 1,000 management positions throughout the Company.

We are now a bit over half way through the structural review process, which is not about any singular announcement, but rather about a series of significant and well-considered actions that will improve our competitiveness and operating efficiency.

We will continue to discuss these actions with you as they occur.

In summary, we have a tight handle on spending and are making good progress on identifying ways to improve our efficiency and agility for the long-term.

At the same time, we are delivering the best new microprocessors in the world and we use our manufacturing leadership to drive the industry's transition to Dual Core in all market segments.

With that, I'll hand the call over to Andy.

Thanks, Paul.

The second quarter was the time for facing the challenges of the market while moving ahead with plans to improve our products and capabilities.

Competition, market softness, a product mix shift in customer inventory levels entering the quarter were among the challenges.

In growth and servers, chipsets for mobile computers, and communications infrastructure were among the bright spots.

While we have trimmed the outlook for gross margin for the year, we continue to tackle the cost structure with reductions in capital spending, R&D, and head count.

We have also reduced a level of cash and a number of shares outstanding.

As Paul stated we made excellent progress in pushing forward with new products to support a stronger second half.

Revenue for the second quarter was $8 billion.

At the low end of our forecast in April, and down 10% from the first quarter.

Most of the sequential change in revenue came from lower average selling prices and unit volumes of microprocessors for the desktop and notebook segments.

Average selling prices declined as the market moved a bit to lower price mix within both performance and value segments in an unusually competitive pricing environment.

In all geographies and in the channel, revenue was lower than seasonal, indicating some overall softness in the PC market segment.

Some of the decline also can be contributed--attributed to customers working through first quarter inventory.

We believe we've met our goal to get customer inventories of our products in balance with business levels and in an appropriate mix for our new product ramp.

Chipsets and other products in the mobility group delivered revenue growth of 16%.

In the server business, revenue was up on improving average selling prices and unit volumes.

In the year to year comparison, quarterly revenue was down approximately $1.2 billion or 13%.

Revenue in the digital enterprise group was down 23% while revenue for the mobility group was up slightly.

Among the geographies, the year to year revenue decline was highest in Europe, which was down 24% and Asia-Pacific which was down 14%.

Japan's revenue was up 3%, marking the 15th consecutive quarter of year to year revenue growth.

Gross margin dollars of $4.2 billion, including share-based compensation of $66 million were down 16% from the first quarter.

Gross margin percentage was 52.1% on a GAAP basis, without share-based compensation gross margin percentage was 52.9%.

A decrease from the first quarter of 3.4 points on a non-GAAP basis came primarily from lower average selling prices for microprocessors, lower overall unit volume for microprocessors, and a mix shift of lower margin product in certain areas.

Offset somewhat by lower inventory write-downs.

In April our quarterly forecast for gross margin percentage, excluding share-based compensation, was 50% plus or minus a couple points.

The improvement of 2.9 points from the mid-point of the forecast was a function of better than expected inventory valuations, including qualification for production for some new products and improvements in unit cost for chipsets and microprocessors, offset by the mix shift in microprocessors toward products with lower average selling prices.

In a year to year comparison, gross margin percentage, excluding share-based compensation, is 3.5 points lower than the second quarter of 2005 primarily as a result of lower revenue, offset somewhat by lower start-up costs.

Spending, R&D and MG&A was $3.1 billion in line with our forecast in April.

Excluding share-based compensation of $266 million, spending decreased by approximately 3% from the first quarter, but is up 12% from a year ago.

Spending has grown as a percentage of revenue, however, cost-cutting efforts are in the initial phases.

The number of employees declined during the quarter to 102,500 which does not reflect the loss of 1,400 people we expect to be employed by Marvell, and 1,000 people whose positions were recently eliminated.

With these actions and attrition, we expect the number of employees to be below 100,000 by the end of the year.

Fully diluted earnings per share were $0.15.

Excluding share-based compensation, earnings per share would have been $0.19.

Quarterly average shares outstanding are down nearly 6% from a year ago.

On the balance sheet, inventories of $4.3 billion were up over $750 million from the first quarter as we ramp production of microprocessors on a 65 nanometer process and produced associated chipsets.

The Conroe microprocessor qualified for production during the quarter and contributed to higher inventories as did the Broadwater chipset.

Broadwater ramped nicely and had good demand which provided better unit cost and lower than expected write-downs.

We continue to reduce the levels of cash.

Total cash investments, comprised of cash, short-term investments, and fixed income trading assets ended the quarter at $7.2 billion, down by $1.5 billion from the first quarter.

Stock repurchases were $1 billion.

Capital spending was $1.7 billion, and dividend payments were nearly $600 million.

As a reminder, we indicated in April that we expect to reduce the level of our stock repurchases over coming quarters to less than half the first quarter rate as we continue to manage our cash flow.

As we turn now to the outlook for the third quarter, please keep in mind that the forecast data do not include the effect of any new acquisitions, divestitures, or external transactions that may be completed after July 18th.

I will use a mid-point of forecast ranges when making comparisons to specific periods.

We are planning for revenue in the third quarter to be between $8.3 and $8.9 billion, an increase of approximately 7.5% in line with seasonal patterns.

If revenue for the second quarter is adjusted for the inventory burn, then the growth rate for the third quarter forecast would be at the low end of seasonal patterns.

This forecast for Q3 assumes higher revenue than the second quarter for microprocessors, chipsets, and flash memory.

Our expectations for gross margin percentage in the third quarter is 49%, plus or minus a couple points.

Without share-based compensation of approximately one point of margin, the forecast for gross margin percentage is 50%, plus or minus a couple points.

Higher microprocessor unit costs, as we ramp production of the Dual Core product, and some erosion in average selling prices will lower gross margin percentage, while higher unit volumes will partially offset the decrease.

Keep in mind that gross margin in the second quarter benefited from the qualification of new products.

Spending, R&D plus MG&A should be approximately $3 billion.

The forecast includes approximately $300 million of share-based compensation.

Without the impact of these charges, spending should be approximately $2.7 billion.

This level of spending is flat to down slightly from the second quarter on higher revenue and indicates some progress at reducing spending as a percent of revenue.

Looking beyond the third quarter, we anticipate a fourth quarter with seasonal growth from the third quarter.

In the last ten years, growth from the third to fourth quarter has varied widely and averaged approximately 10%.

This is likely to result in annual revenue for the year that is lower than our outlook in April when we forecasted a decrease of approximately 3% compared to 2005 within a wide range of variability.

We expect gross margin percentage for the year to be 51% plus or minus a few points.

Without share-based compensation of approximately one point of margin, the forecast is 52% plus or minus a few points.

The adjustment and the forecast reflect lower revenue primarily due to lower ASPs.

The main cause of the decline from 59.4% in 2005 are lower prices of microprocessors, higher unit costs as we move to Dual Core, and higher cost for chipsets mitigated by lower start-up costs.

As a result of several cost savings actions, we have again lowered the forecast for R&D and capital spending for the year.

The forecast for R&D is now $100 million lower at approximately $6 billion.

This brings us to our previous goal for total spending for R&D, marketing, and G&A of $12.1 billion.

Without share-based compensation of approximately $500 million, the forecast for R&D is $5.5 billion, up 7% from 2005, and without share-based compensation of approximately $600 million, the forecast for total spending is $11 billion, essentially flat with 2005.

This includes spending for the IM Flash technology joint venture.

We have also lowered the capital spending forecast by $400 million to $6.2 billion plus or minus $200 million.

The estimated tax rate for the third and fourth quarters of 2006 is unchanged at approximately 30.5%.

In summary, we look ahead to our business improving in the second half with outstanding new products and seasonal growth.

While business remains competitive, we are focussed on--in doing the right things, delivering compelling products, competing for sales, proceeding with systematic and comprehensive cost cutting and executing superbly with 65 nanometers while laying the ground work for the next generation.

With that, let me turn it over to Alex for Q&A.

Thanks, Andy.

We will now open the conference call for Q&A.

We will attempt to take questions from as many participants as possible.

To help in this process, we ask that you please limit yourselves to only one question and no more than one brief follow-up.

Thank you.

[OPERATOR INSTRUCTIONS]

Your first question comes from the line of Samit Dhanda with Banc of America Securities, please proceed.

Yes, hi, Andy, I had a couple of questions.

First, it seems at least on a sequential basis the notebook business was off fairly significantly and conversely, the mobile chipset business did very well.

Could you help us understand what occurred there?

And then secondly, you've indicated that cost on Dual Core ramp--on the Dual Core ramp will impact margins significantly for Q3.

Is the expectation still that this will start to be a little bit of a tail wind in Q4 as you have indicated at the Analyst Day, or has that changed?

Thank you.

Sure, on the cost one, I would expect us to start to see the turnover in cost begin in the fourth quarter and continuing into next year.

My guess is the expensive Dual Core product peak will be in the third quarter and we'll start to see the newer products have a meaningful effect in the fourth quarter.

In terms of the notebook--

Why don't I take that one?

We saw a--in general what we're seeing is an expansion of the notebook segment as a percent of the overall PC market.

And in the second quarter shipments, we saw quite a bit of expansion in the--a shift if you will in the mix from the high end to the low end, principally driven by retail purchases, consumer notebooks tend to sell for a lot less than business notebooks, and they tend to use a lower-end microprocessor.

There may have also been some hesitation in front of our price move and in front of our Merom launch.

So I think those three things combined to get you.

The chipset business is up in mobile, which indicates the continuing growth in mobile, we've got a fairly high market segment share in mobile chipsets anyway.

So I think that one is a good leading indicator on where the notebook business is going in terms of what you see.

Okay, thank you.

Your next question comes from the line of Tim Luke with Lehman Brothers, please proceed.

Thanks.

Andy, with respect to the inventory levels, could you give us a sense on how you would expect them to look going forward based on hand and maybe what you perceive in the channel?

I was also wondering if you could give any color with respect to your guidance of a mid-point '07 of how we should view the unit growth versus the ASP direction?

Thank you.

Sure, hold on, let me capture this.

In terms of inventory levels, I'll give a long winded answer because I also want to explain a little bit about the second quarter (inaudible) was up over $750 million.

If you remember at the end of the quarter we expected a little less than half of that.

We knew we'd be trying to ramp Broadwater in that period of time.

We did not expect Conroe to qualify until the third quarter.

So what happened if you look inside the quarter, the surprise increase in inventory is several hundred million dollars for the Conroe product line, Better Health and Broadwater, so I took lower inventory reserves there, a third element is in the second quarter my mix was a little lower, so I have a little more Dual Core content in inventory, so it's a little higher.

Essentially what I'm trying to paint to you is what happened in the second quarter is I built a lot of product that is 65 nanometer processors and 90 nanometer chipsets, which is why I think is an okay thing.

Going into the fourth--the third quarter, I would expect profits to be about flat.

We had a big pop in valuation in the second quarter for the Conroe and Woodcrest.

I would expect chipsets to still grow a little bit.

I still think we're digging ourself out of the chipset hole.

At some point we may start to turn it over, but not in the third quarter.

So overall expect a little increase in the third quarter, but I don't think it's a big increase.

Just sort of $200 million, the same kind of range that you advocate--you sort of thought before, that kind of framework or how should we view that?

Well, I hate to put a number on it.

That's an acceptable magnitude.

If I were to see $200 million it wouldn't startle me.

And then you'd expect it to go down obviously in calendar fourth quarter.

I would expect fourth quarter to start to turn that down, yes.

Can you give us any framework there?

As bad as I was at forecasting inventory in Q2, I want to wait.

But yes, I would expect a slight increase in the third quarter, a slight decrease in the fourth.

Your other question was?

Was just that with respect to the mid-point of your revenue, your expectations in terms of how we should think about unit growth and ASPs, should we think about ASPs coming down a little bit?

I think ASPs do come down a bit in the third quarter, we do see a pretty tough pricing environment.

If you look inside the second quarter, we came in at the low-end of our range and most of that was ASPs.

If you looked at the units we shipped into the marketplace, if you look at the units we think our customers burned out of the inventories, it came in almost exactly as we had forecast to you guys in April.

The surprise was it was a little bit lower mix, a little bit lower price.

When I look at my full year, even though our full year, if you do the math is going to be down a fair amount.

It's mostly pricing, the unit environment, the demand for product seems pretty normal in terms of--in purchasing.

A little weak, so don't change your mind, again, we know Q1, Q2 was a little weak, but I would paint a normal second half off of the first half in terms of units.

ASPs in mid-single digit decline, is that what I should think about in the third quarter?

Just some kind of approximation.

I can't go that level of granularity.

But, so despite the normal units that you have the extra inventory and you think the inventory goes up in the third quarter, what do you think is normal units then?

Again, go back and look at Gardner or anybody else's unit demand in Q3's of years gone by, and it's not too far out for revenue, as well.

You can get pretty close.

Thank you, guys.

Your next question comes from the line of Joe Osha with Merrill Lynch, please proceed.

Hi, Andy.

You're saying that you qualified Conroe more quickly than you thought so those wafers did become part of inventory as opposed to just flowing through the P&L, right?

That's correct and that was about $200 million worth.

Okay, I guess first question, then why does that not show up in the R&D which basically came in where you though?

As I look at this and I look at your--

Slow down, slow down, don't go so fast.

Okay.

If the product hadn't qualified, the expense would have not been in R&D, it would have been in other cost of sales, similar to our start-up cost.

So what it does is it flips out of cost of sales into inventory.

Okay.

So you look at my margin being a couple of points high, this is part of the reason.

Look at my inventory being a couple hundred million high, again they kind of balance.

Okay.

Second question, as I look at the timing and the mix of your ramp, it seems to me like an awful lot of this inventory addition has got to be Pentium D right which is the expenses to manufacture part that you're going bell and then end of life over the next couple of quarters, right?

Some of it is, yes.

There was an increase in inventory of that product.

Okay, and then as I look, I do some back of the envelope math here, I see your SG&A and your R&D falling off in Q4, as I think about 2007, should I think about--how should I think about the run rates for those operating expense lines?

I would think they're going down again, as Paul--it's going to be hard for me to give you a magnitude, as Paul said we're going through deficiency project evaluations now.

We'd be very disappointed not to see a significant dollar savings as we get into next year.

But I really would like to finish that project before I give you a number.

But just to clarify, down from the Q4 run rate?

Or down '07 on '06?

Because the numbers--as I look at that Q4 run rate, are you saying that it goes down from there?

Again, since we're not done with the work yet--I'm giving you a bias but I don't have studied numbers behind, okay?

My belief is it's down from the Q4 run rate and down year-over-year, both.

Okay, and Paul, may I ask one philosophical question before I go away?

Certainly.

You have a great new product line-up and I think everybody knows that, why on earth run up inventory like this?

It creates a problem that appears to me to be obscuring the progress you'd otherwise be making?

I don't think that we're running up inventory or creating a problem.

We're building--if you look at this history of Intel, we generally take care of our inventory pretty well.

And if I've made one mistake in the last two years, it was that we didn't build enough chipset inventory at the right time in second half '04 early '05.

And that got us in trouble in terms of market share and got us in trouble in terms of our customers and everything else.

We think that we're starting wafers on the right products now.

And I'm very comfortable with the current bill plans and factory loadings.

Thank you.

Your next question comes from the line of Krishna Shankar with JMP Securities, please proceed.

Yes, Andy, you talk about the billion dollars in operating expense and savings.

Should we be looking at a billion dollars taken out of OpEx between 2007 versus what the new guidance is for 2006?

Or can you give us some time frame for that OpEx reduction?

The billion dollars out of this year has essentially kept us flat to last year, so I want everybody to understand it isn't like we cut a thousand--a billion out of last year's total, we cut a billion out of our run rate, which got us to flat.

What will happen next year is going to be more a function of what happens in the efficiency project.

I do expect it to go down, I do expect to go down from Q4, but I really need to finish the work to know how much.

So I know you'd like to have a number, I have to resist at this point.

But again, I expect to make real progress next year, I expect it to be a lower number.

Okay, and just a quick follow-up.

Paul on the three tiered pricing structure that you spoke about, does this imply a long tier to be paying for (inaudible-heavy accent) where you have the sort of three-brand structure, or should we anticipate a pretty quick sort of flush out of the Pentium 4 network architecture?

Well, we had the opportunity to use a number of microarchitectural dye in that brand.

So I don't think it implies a long taled net burst as much as it implies a long tale to using Pentium and bringing that into newer lower price points.

In particular, we find this an intriguing opportunity in emerging markets where Pentium is quite well known, but as a percent of the overall sales because of price points of systems it hasn't been the predominant line item in the Intel lineup.

By having new system, new CPU price points, we enable new system price points for Pentium in these markets and we think that is an advantage to us.

Thank you.

Your next question comes from the line of Mark Lipacis, please proceed, sir.

Two brief questions, first.

Andy, can you give us a sense of the cash flow from operations for the quarter?

Boy--cash flow from operations for the quarter?

I don't know, I don't have that number with me.

We did a billion dollars in buy back, we pulled cash down a little bit--it's probably some positive, but not a big number, but I would have to get that, I don't have that.

Okay, fair enough.

Second question, of the inventory that you have on the books, can you give us a sense of what percentage of those are from the new microarchitecture products?

Well, if you're looking at the total, it's an odd answer.

If you look at the growth of $750 some million it's most of it.

If you look at Dual Core and you look add Broadwater, we grew in those spaces essentially all of the growth.

You still have flash communications products, you still have motherboards, all of those products are still there at about the same level they have been.

Okay, thank you very much.

Your next question comes from the line of John Lau with Jefferies, please proceed.

Great, Paul, Andy.

In terms of questions about your relationships with your major customer and the rumors about expanding usage from your competitor, what are the key issues that you think are driving the decisions?

With the new products out, is it performance anymore?

Is it more pricing, supply chain management?

What are the key levers for the decision process now that you have your new products out, Paul?

Well, I'm--I tend to be pretty monolithic in this thinking, I think it's a bit of all of the above, but technology and good products tend to be the overarching drivers of success in our industry.

And so I'm--you've seen the write-ups on the new products and the microarchitectures, you've seen customers talking about them in a very positive fashion, you've seen the press talking about them in a very positive fashion.

I think at the end of the day, our job is to make sure we build the best products.

Great new products very often ignite markets, example Centrino a few years back.

And I think that with this new class of products we can perhaps lift the entire market as we go forward.

Independent of that, the addition of those products gives us a much better competitive posture, particularly when you look at the three-brand strategy now.

And along the lines of expanding the new markets, Paul, you do have the Dual Cores now, Quad Cores coming by the end of the year.

In terms of how the migration path in the marketplace, when you have these new products here, what is the typical time in which you're going to see those next applications, or is it about a year away or nine months from now?

Well, it depends on--it's segment by segment.

In the server segment, Quad core doesn't need new software.

Server software today is all threaded so it just takes advantage of that as long as you can keep the pipelines full and we've done a pretty good job we think on architecting that to get high efficiency out of this Quad Core machine.

These are dual processor quad core machines.

In the extreme gamer community, the Quad Core also can be readily adopted quite quickly because the software is tuned for it, ditto on the workstation business.

When you start talking about mainstream microprocessors, I agree, we've got a ways before the applications can take advantage of all of that horsepower, but it's not that far away.

Most people are running multi-tasking environments today already.

Great, thank you.

Your next question comes from the line of Michael McConnell with Pacific Crest Securities, please proceed.

Thank you.

Paul just kind of a longer term question with respect to pricing for the microprocessor and it's pretty obvious, we've seen some of your larger customers begin to use your competitor on a more broader basis than in any time we've seen in history.

If we look at the new product line-up, which looks like it's going to be much more competitive, are we expecting ASPs to begin to move up as we progress with the ramp of Core Duo into next year?

I don't think we're giving out ASP guidance today, Michael.

So I'm going to avoid that.

But I think the criteria is going to be what is the mix--how does the mix settle out?

Geography by geography, quarter by quarter in terms of the percent of our products that are going to be Core microarchitecture based versus Pentium versus Celeron.

I think that we haven't had this kind of line-up in a long time.

And we certainly have never had a three-brand strategy.

So to the extent that we can create pull and demand around these new products, you can get some lift.

I will tell you that we have shifted virtually all of our direct advertising in the second half of this year to focus on Core 2 Duo, and you'll start seeing those in billboards and print ads and television ads throughout the second half, particularly as we move into the peak selling season from Labor Day on.

And then one follow-up just on the restructuring.

You said in the prepared comments that head count could fall below 100,000 by the end of the year.

There's been some pretty large numbers going around in terms of the amount of employees that could be--we could be seeing reductions for.

Not trying to read the tea leaves here, but should we be thinking that the restructuring or the employee reductions will not be on a very broad basis?

Or maybe more tempered?

Don't make that connection, yet.

What you should be thinking of if we did no new restructuring activity I'll get it below 100,000.

We believe between now and well to the next two months, we will make a variety of decisions, some of which may have expending in head count applications yet this year, which will take the number down even lower.

So I'm not making that next statement yet.

On what we've done today, I'll get below 100, we may have actions that will take it even lower before the end of the year.

Thank you.

Your next question comes from the line of Michael Masdea with Credit Suisse, please proceed.

Thanks a lot, when we looked at your '06 estimate you talked about how we've seen it go from the high-single digits to the low, and increase to a high-single digit decrease.

And you stated a lot of that was really due to the ASP side.

Help me understand, kind of how that kind of happens.

What kind of environment really led to that kind of pricing environment, and what gives you confidence that we're not going to see that again in your ability to forecast that we're not going to see that again?

Well, it's possible it will happen again.

If you go through the math, we have a relatively wide range in Q3, and we said Q4 will be seasonal off of whatever that final number would be.

So right now I'm not picking a number for the year and saying thank God I finally figured it out.

The environment we found ourselves in the second quarter ASPs were down to (inaudible) price for two reasons.

One is the competitive price environment, and one it was a little bit lower mix than we expected.

What we've done is assume the competitive price environment stays.

As Paul said the product line-up gets stronger and stronger.

Hopefully that allows us to gain some traction there.

But we've made a forecast that says we think that there'll be a quote normal Q4 off of the third quarter which is 7.5% off of a weak Q2.

I don't think it's a particularly aggressive forecast for the rest of the year, but I certainly don't have a forecasting track record this year that says I can bet on it.

That's fair enough, and then on the inventory side given where we are now and given what you said about production and everything else.

Are you thinking about any sort of difference in your inventory target in terms of days going forward?

Has it changed at all?

And should we read anything into the CapEx in terms of what you think production should be versus in demand?

A couple things there, so in terms of quote inventory strategy, no, I think we're in the middle of product transition, and in the middle of a product transition, you want to aggressively build a new product, and then carefully manage the older products.

That's what we're doing.

Hopefully up a little in the third quarter, down some in the fourth quarter.

See where the market is and then figure out what we need to do for the rest of next year.

I would--if you asked me, would I like to see it be lower, of course.

If you asked me if I'm concerned where it is, the answer is no because we need to make this product transition.

On the capital spending side, think of it in two ways.

One is, of the reduction of the mid-point by $400 million, half of it is essentially slower construction spending.

So we have all of the same construction projects we thought we'd have, we've just slowed down some, which means some spending spills into next year instead of this year.

The other half, about $200 million, actually comes from, I hate to use this word, it's not quote the efficiency project, but when a company starts to focus on getting more efficient, you start to find ways to increase utilization of tools.

In this case, almost all of the second $200 million comes from the back-end processor (inaudible) where we found a way to increase the loadings into the equipment in those factories and we can save a couple hundred million dollars.

We don't think it changes our capacity, we think it's just better use of the capacity we plan to have.

So in other words you're going to reassess at the end of the year based on what you've done in the second half of this year and what '07 looks like in terms of your--

In terms of inventory, yes, in terms of capital, I hope that through this project I find six more ways to increase my utilization of my equipment and save some more money.

Thank you very much.

Your next question comes from the line of Adam Parker with Sanford Bernstein, please proceed.

I'm just wondering on inventory, wondering on wireless.

I'm still not sure.

I heard Paul say you don't think you're running up inventory.

And Andy you said this is kind of what happens in the middle of a product transition.

I'm wondering what level would you be concerned about?

If I look at days of inventory, it's basically among the highest it's ever been so why is that the wrong thing to look at?

And if it's only slightly up and it's slightly down the next couple of quarters, doesn't it create a problem heading into '07?

So is there a risk you have to take here a big reserve here?

Just help me understand why you're so comfortable when history implies that you're in a danger zone here on inventory.

So I don't think history implies I'm in a danger zone on inventory.

It does show it's higher than it's historically been, but at $4.3 billion of inventory that's about one quarters worth cost of sale, that's not a danger zone.

Especially when you have a 13-week prepare time anyway.

I'm not uncomfortable with the level of inventory.

Would I like it to be less?

Of course, I would always like to have managed it more tightly.

But it doesn't lead to big write-offs.

I'm pretty comfortable with that situation.

Now if you're telling me demand's going to fall off the table next year and I'm going to have unload factories, then it will be low.

Keep in mind what Paul says, the last time we built inventory, not quite--it wasn't quite this magnitude, but it was a big number, we jerked down the factories, created shortages in chipsets, affected our microprocessor business, we're going to try to make sure this time we don't over react to a one quarter inventory build.

Particularly since we think we built all the right stuff.

Okay, so you're viewing this as just kind of a one quarter build, because it seems like it's more than that.

When you guys had guided on your Analyst Day, you had said look the net value over microprocessor inventory will be down in Q2.

And now it looks like, because of this qualification, it's up.

So I'm just trying to figure out, could you have structurally--maybe it's a follow-up on Masdea's question, could you have structurally higher days of inventory for the next year?

You could, but that's certainly not a conscious decision we've made.

In the Analyst Day we did say microprocessor inventory would be flat to down, no question.

Again, I didn't count on Q2 qualifying Conroe and Woodcrest for $200 million.

So that wasn't in my thinking envelope before.

The other thing that happened was we shipped from previously reserved inventories, so some of the lower-end mix products, I didn't expect to ship those things.

When I shipped those and replaced them with higher value Dual Core inventory, again, it grows a little bit.

So yes, I would say the first 200 is a good news surprise, the fact that my inventory mix is a little richer is--I'll call it a good news surprise, but I'll call it an understood surprise.

And in essence, it's not a place I'm concerned about today.

Now if it grows another $800 million next quarter, then you can--

Have you ever been concerned about it in the past?

Have I been?

Yes, I was concerned two years ago.

I was the one who convinced Craig and Paul to pull down our inventories, I take responsibility for that and I was wrong.

So I guess I'm saying it looks so much worse now than it did two years ago, particularly because of demand outlook if you look at GDP projections from economists, it looks weaker now than it did at the same point two years ago.

So it seems hard on the--unless you're just explaining it largely as the new products, or all as the new products and the chipset ramp, it seems--the optics of it seem more difficult to understand for us dumb guys.

I beg to differ, I think two years ago we were looking at a potential weaker economy, we also were looking at an overbuild that we had put in place for Prescott, which we knew was going to be a weak-yielding product to begin with, then it yielded well.

We can relive the history, if we want.

But if you look--the facts are if you look in Q2 of '04 at what the GDP projections were for the biggest 20 countries for the second half of the year and you compare them to what they are right now for the second half of this year, it looks weaker at this point now than it did two years ago.

That's what I'm worrying about is are you going to get capacity growth exceeding revenue growth to have a further inventory build and then your margins are kind of impeded for longer than you currently expect.

But it sounds like you have comfort around that.

It is possible.

We--what we have built in for the rest of the year is a--like we said it's a seasonal back half off of a weak first half.

If you tell me you think the worldwide economy is going into recession, I will have inventory problems.

Well, we'll all have problems.

The other thing I wanted to ask about was just the wireless sale.

Is there anything about you embedded into your Q3 guides in terms of revenue or anything?

We think the business closes in the quarter, so we kind of have it for the fourth quarter.

I'm sorry, say that again, I couldn't hear you.

We think the sale closes in the fourth quarter, we have it in the fourth quarter.

Okay, so the sale of this business, can you talk now that it's kind out, it's gone.

Can you talk at all, kind of remind us what--I know, Andy, you guys do these sort of ten-year MPVs on your businesses, and obviously the assumptions there dramatically change, but in retrospect, what can we learn from what happened to your original expectations given a couple of years ago this was looked at as a real big contributor to your future growth.

How did--how did that really change?

Well, I think that our view of the markets evolved.

And remember we were aiming, we always aimed that business at the smart phone market, and our ability to penetrate.

And then we did quite well on the PDA portion of the smart phones, we just didn't do that well with a couple of notable exceptions in the generic smart phone.

Mostly because the generic smart phone business did not take off as a percent of the market.

The cell phone market is still very much low-end driven in terms of unit growth.

As we begin to evaluate our opportunity to participate in broader segments of that market going forward, we just didn't see numbers that gave us a reasonable return on incremental investment going forward, hence the decision.

On the other hand, as we talked about in the Analyst Day, we're very, very excited about ultra-mobile PCs, and low-power Intel architecture moving into these things and having computers get smaller, with the communication capability, that is where we're putting our energy at this point in time.

Was there any coincidence, or was it pure coincidence that the same $600 million was the amount you invested in WiMax the very next week?

It's pure coincidence.

Because on the surface you can say okay well, we did a ten-year MPV that suggested it's going to be much better than we thought DSP --

No, no.

Two ships passing in the night may happen to be the same number when they passed.

All right.

Thanks guys for your time.

Your next question comes from the line of Glen Yeung with Citigroup, please proceed.

Thanks, Andy, when I look at your gross margin forecast for the year, I think I'm reading it right that it's 51% for the year and if that's right it implies that fourth quarter gross margins will actually be below third quarter gross margins, I want to make sure I'm understanding that correctly.

Well I'm not making a fourth quarter gross margin forecast right now.

What you have to recognize is the full year has three years of baggage behind it, of actually --

Three quarters.

Three quarters of actual behind it.

So if you look at that fourth quarter, you're going to find you have a wide range of outcomes that still leads to a mid-point of 51% plus or minus a few.

So don't overanalyze it at this point.

Do what you think makes sense.

And i.e. if revenues are up and utilization is up, in theory it should be up?

I believe cost per unit should get better in the fourth quarter.

I believe the new products will be a higher percentage of our revenue in the fourth quarter.

So there are some reasonably positive things out there.

But again, let me wait a quarter before I give you a fourth quarter margin estimate.

Okay, fair enough.

I wanted to ask, also some questions on the pricing trends for the third quarter, recognizing that you guys are making some changes.

And most notably in notebook because there aren't any sort of specific changes that you're making for notebook.

If I'm understanding it correctly, you actually cut prices on notebook in March, and now your new--your refresh in the third quarter actually comes back to the old ladder.

How should we think about notebook pricing across the board for the third quarter?

The difference in the degree of reset of the price stack in desktop and notebooks reflects the strength we have in the notebook business.

We are the performance leader, the volume leader by far there.

We are--I've designed the new products to come in and sit on top of the other ones.

And yet we're driving Dual Core farther and farther down the stack particularly as it becomes appealing in to consumer notebooks.

It really reflects the competitive strength of the product line more than anything else.

Sure, that makes sense.

So, Paul, actually if we think about just the desktop part of it, you guys are seemingly putting the single core chips on fire sale for Q3 fine because that's going to end of life at some point.

Within the Dual Core processors, only, is there a strategy here for you guys in trying to move the mix?

Are you trying to get us to buy more high-end in the--in the Dual Core market starting in Q3?

Heck, yes.

Wouldn't you?

Yeah.

I mean, that's, it is always Intel's strategy to try and sell the richest mix possible.

The thing that we're--that we're really focussed on right now though was given that we have Dual Core running on a number of new microprocessors and older ones, is really the penetration of Dual Core and making that much more pervasive in desktops in all segments.

And did you say what you thought Dual Core was going to be as a percentage of desktop sales in Q3?

No, but I showed you, I showed you Q3, I showed you a line through year end at the analysts meeting, and if anything it's gotten a little bit richer than that.

Okay, that's good to know.

All right.

Thanks a lot.

Your next question comes from the line of Jim Covello with Goldman Sachs, please proceed.

Thanks very much.

Question on the margins for Q3 and then its sort of implied margins for Q4, is all of the weakness in margins in Q3 ASP related or are there other factors going on?

No I think we'll have a richer mix of Dual Core, the early Dual Core products, which means the cost of units will go up.

In fact, probably, I would guess cost is a bigger element than average selling price.

And then--then relative to Q4 to Glen's point, there are scenarios that Q4 is down again.

If it was cost in Q3, wouldn't that improve as we went through to Q4?

Yes, I believe my cost units should improve in Q4 versus Q3.

I believe the number of units shipped out of the new architecture products will be greater in Q4 versus Q3.

So I think I have a few good things happening in the fourth quarter for us.

But again, then, I don't want to push it too much, but the margins are theoretically down for both quarters.

What--if there's positive things going on, why isn't that showing up in the margins then?

Again, if you do the math with three quarters of actual at our mid-point and look at what the range of Q4 could be to still round to unit 51, I think you'll still see a wide range of possible outcomes.

Sure, I don't want to get wrapped up in semantics, but it's not up in conjunction--if revenue's going to up 10%, you would think margins would be better than flat to up slightly.

Yes there are scenarios mathematically that the margins are up, but there are scenarios down too.

Just trying to get a handle on why the margins wouldn't be up consistent with the revenues.

Okay, I don't know how many times I can say I'm not giving a Q4 forecast.

There's a wide range of outcomes, I agree there's no 10-point up margin outcomes.

Okay.

I guess that's--I guess we'll have to deal with that.

Then, on the inventory, what would it take for the Company to do an inventory write-off?

Under what scenario would you decide to write off some of the inventory?

Well, I want to be clear, we always have some inventory write-offs.

If you're asking an extraordinary inventory write-off?

Yes.

It would take most likely a big change in demand.

If suddenly demand dropped 10% or 15% and we decided to push all of the new products and stop selling any of the single core stuff, then we take an inventory write-off.

Just to be clear, a change in demand relative to normal seasonality, because clearly demand is just okay now.

Change in demand from our expectations.

Okay.

Recognize, again, I know we are in very different spaces on this.

I have the equivalent of one month worth of constant sales in my inventory--one quarter's worth of constant sales in my inventory.

I would typically value a whole lot more than one quarter's worth of inventory looking out into the future.

It would take a pretty dramatic change to cause this to have a big inventory write-off.

Terrific, thank you.

Your next question comes from the line of Chris Danely with JP Morgan Chase, please proceed.

Thanks, guys, just a follow-up on some of the gross margin stuff.

Andy, it sounds like inventory's up a little this quarter and then down a little in Q4 and perhaps you want to take inventory down in the first half of next year.

And it seems as though depreciation should increase to make a short story long, should we expect gross margins to bottom in the second half of this year or the first half of next year?

Again, I'm not making long-term gross margin forecasts.

What I really believe is we've shown a cost per unit of microprocessors at the analyst meeting, and what it showed was the highest point on cost for the microprocessors I expect to be the third quarter.

And then I expect to see period by period by period by period improvements.

Now, there are other products that we sell, as well.

We sell flash, we sell motherboards, we sell chipsets.

It's not as clean as looking at that one chart and saying I can forecast Intel's margin, but microprocessors are clearly the biggest part of our revenue.

I absolutely see our cost units improving quarter by quarter by quarter through next year.

Okay, and then just to follow with that, Paul, you said you want to keep the fabs full, have you guys lowered utilization rates at all?

Or do you plan on lowering utilization rates?

You mean in--what do you--I'm sorry, factory utilization rates, is that what you're asking?

Yes, have they been lowered or do you plan on lowering them?

No the factories are still operating, for the most part at full capacity.

Okay, and so, let's just say sales are a little bit lower than expected this quarter--

Then we might have to--then one of the options we have is to take down wafer starts a bit.

But right now none of our projections show that.

I guess I'm just wondering given what's happened the last couple of quarters, why aren't you guys a little more conservative on your guidance and inventory?

You mean why arene't forecasting it to grow more?

I mean, why forecast normal revenue growth and inventory going up given what happened the last couple of quarters?

Because the fourth quarter is bigger than the third quarter in terms of shipments.

And you want to build, we need to be able to support that growth.

It's still a seasonal year and then first quarter things drop off.

Yes, I understand that, it's just us analysts remember what happened in 2004 so that's why we're all pretty nervous.

Okay, thanks guys.

Operator, two more questions, please.

Your next question comes from the line of Eric Gomberg with Thomas Weisel Partners, please proceed.

Okay, great.

I was wondering if you could give any more color.

Given the early qualification of some of the processors where you'd expect to be either by the end of 3Q or year-end in terms of Woodcrest, Conroe, and Merom as a percent of processors?

And also do you have any concern regarding being able to ship enough Conroe given how good the reviews have been?

Will you be able to meet all the OEM demand?

I showed a number at--in April at the analyst meeting on Woodcrest that said we would in terms of the percent of our Zion products that were Woodcrest based, it would be 75% by Q4, I think it was 50% in Q3, or whatever the number was.

Those numbers have only gotten a little bit better since then, they've not gotten worse.

Obviously the denominator is still at risk, but we're ramping Woodcrest as fast as possible and our intent is to convert the product line.

Similarly on Conroe and Merom, right now we have rolled up demand from our customers and from the channel.

And we believe we can satisfy all demand for both of those products.

Okay, and if I could just ask a follow-up.

As you go through the strategic review, this is something you've been asked probably many times over the last few years, could you talk a little bit about the NOR Flash business and why that remains a strategic and compelling business for Intel to be in at $250 million operating loss in the first half?

Well, our focus on that business is to not have that kind of operating loss going forward.

So we are going through a similar rigorous process inside of the NOR business.

You've already seen one action that came out of that about six weeks ago where we rolled for the first time ever, we broke out the manufacturing and technology development groups for Flash from the rest of the Company and put them under the Flash group.

And we have already seen some pretty interesting ideas.

Some--the beginnings of some efficiencies rolling out of that as they learn how to be much more competitive with their peer group companies in the Flash marketplace.

I'm--my focus inside the Company is to make sure that that business gets to be a profitable one as quickly as possible.

I just want to clarify something, the losses you see in that business are not just the NOR Flash, we are starting up the--

That's right.

The NAND Flash business recognized in that business you have factories starting up pre-volume production.

So you have pretty high start-up cost, you have pretty high R&D costs.

A big portion of the increase and losses in that business is that start-up, which we hope to turn profitable sometime in the near future.

All right, that's very--could you give any color in terms of what you'd expect NAND contribution to be in the back half of the year?

In terms of--in terms of revenue, no.

Relatively small at this point, starting to finally ramp in the back half of this year as it's been setup in the front half.

Hopefully next year will be meaningful enough we can talk about it.

Okay, thank you.

Last question, please.

Your last question comes from the line of Hans Mosesmann with Moors & Cabot, please proceed.

Thanks, Paul, you mentioned that the rolled up demand from customers suggests that you could satisfy all of this demand.

Can you satisfy it at launch?

I just want to get a sense of--

Yeah, in fact at launch you'll even see Conroe stocked in the channel.

Okay, so both OEMs and channel will have product?

Yes, sir, and we've got motherboards out there staged and so forth.

This one is being done as broadly as we know how to do it.

Okay and then a follow-up.

In terms of previous launches, how quick of a ramp pervasiveness did you see with this core?

It's not sitting in the NetBurst ramp.

Faster than NetBurst?

Much faster.

Okay, thank you.

We'd like to thank everyone for listening to today's call.

A recorded play back of this call will be available at approximately 5 p.m. pacific time tonight.

Those interested should call 1-888-286-8010 and reference pass code 23019996.

Thank you.

Thank you for your participation in today's conference.

This concludes the presentation.

You may now disconnect.

Good day.