英特爾 (INTC) 2006 Q4 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Intel fourth quarter 2006 earnings conference call.

My name is Letitia and I will be your coordinator for today. [OPERATOR INSTRUCTIONS].

At this time, I will now turn the presentation over to Mr. Alex Lenke, Manager, Investor Relations.

Please proceed, sir.

Welcome to the Intel fourth 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 fourth 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 rate, interest and other income, R&D spending, capital spending, and depreciation.

These statements do not reflect the potential impact of any mergers, acquisitions, divestitures, investments, or other business combinations that may be completed after January 15th, 2007.

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

With that, let me turn it over to Paul.

Thanks, Alex.

In the fourth quarter, revenue was at the top of the range we said in October, driven by a strong mix shift to our new core microarchitecture in servers, mobile, and the desktop.

Microprocessor units came in about as expected, and set an all time record, with unit and revenue records in both server and mobile.

For the year, Intel achieved record mobile and server processor units, flash units, and chipset billings.

Our 65 nanometer transition was superbly executed with more than 70 million units shipped during the year. 65 nanometer leadership allowed us to drive down the cost of dual core computing, boosting dual core volumes to over 50% of total Q4 microprocessor shipments.

In mobile, we took dual core from zero to over 90% of the performance mix in the course of one year.

We continued to advance our technology in Q4 by launching the industry's first quad-core processors for volume servers and PCs.

We launched 9 different versions, including the first Core2 quad processors for mainstream PCs, and we're well on track to deliver a million quad-core processors by the middle of next year.

We also completed the development of our 45 nanometer technology, which is scheduled for production in the second half of this year.

I'm pleased to announce that we have working samples of Penryn, our first 45 nanometer microprocessor, and gave booted four different operating systems on this first silicon.

In mobile, we saw record units and revenue on the strength of our product line.

We look forward to launching our new Santa Rosa later in the first half, which supports new integrated graphics, our Robson-NAND flash technology, and our 802.11-n WIFI technology, which has already begun to ship.

During the quarter, we demonstrated our first mobile WiMAX silicon and are developing products that work on both WiMAX and WIFI networks, automatically giving users the best available broadband connection.

Desktop revenues were higher on the strength of Core2 Duo, while overall units were below seasonal as desktop growth continued to be offset by the demand shift to Mobile.

The digital home business had record units and revenue, driven by the ramp of our Viiv technology.

Our server business had record units, while strong demand for new products led to record revenue of over $1 billion.

The energy efficient Woodcrest processor exceeded 50% of total server units and Itanium revenues were at record levels.

Our Clovertown quad-core processor extended the industry records set by Woodcrest with energy performance 60% higher than Woodcrest and up to 2.5 times the competition's best.

Chipset units were approximately flat from the third quarter, and billings set a record for the year.

Our flash business set a new unit record, with growth in NOR as well as in our new NAND business.

We also begin volume shipments of the first 65 nanometer NOR memories that can store a gigabit of data on every chip.

During the quarter, we continued to take significant actions to increase our operating efficiency and reduce ongoing costs.

Our employee head count ended the year at 94,100, down from 102,500 at mid year.

Our spending is trending lower, as Andy will discuss, and our capital forecasts now reflect savings identified in our structure and efficiency review.

In summary, 2006 was a challenging year, and our response was to drive for product and technology leadership, aligned with the fastest growing segments of the marketplace.

We were pleased to end the year with strong market acceptance of our advanced technologies, including the new Core II Duo and Xeon processors.

In 2007, we will continue to drive technology, including quad-core and 45 nanometer while further increasing our operating efficiencies across the Company.

Let me now turn the call over to Andy.

Thanks, Paul.

The fourth quarter was a strong ending to a difficult year.

Revenue was higher than the third quarter in all geographies, led by revenue per server, mobile, and desktop processors, which together were up 14% sequentially.

For the second quarter in a row, the server business delivered double-digit sequential and year to year growth.

Gains from a divestiture, less asset impairment charges caused by this divestiture, contributed a net gain of about $0.025 to earnings per share.

Diluted shares for the year were down 5% from the prior year.

Headcount is lower by nearly 6% for the quarter and the year.

As we look ahead to 2007, we are on target to generate savings in costs and operating expenses of approximately $2 billion.

Revenue for the fourth quarter was $9.7 billion, but the top of the range we forecast in October and up 11% from the third quarter.

The growth came from higher average selling prices and unit volumes of microprocessors and from higher revenue from flash memory.

As was the case in the third quarter, the server business led the percentage growth and revenue with higher units and higher average selling prices.

The mobility group led the dollar growth with higher revenue for microprocessors, wireless, and chip sets.

Microprocessor revenue and the digital enterprise group of $3.9 billion grew 9%, its strongest sequential growth rate in 8 quarters.

The mobility group achieved growth in microprocessors of 19%, its best sequential growth in over 8 quarters.

Revenue from chipsets, mother boards, and other products in the digital enterprise group was $1.3 billion, down 8% from the third quarter, while revenue for chipsets and other products and group was $925 million, up 14%.

In a year to year comparison,quarterly revenue was down approximately 5%.

Lower revenue in digital enterprise group more than offset growth in the mobility group.

Among the geographies, the only year to year revenue growth came from the Americas region.

Gross margin dollars, including the effect of share based compensation of $94 million were $4.8 billion, 12% higher than in the third quarter.

Gross margin percentage was in line with our forecast and approximately flat with the third quarter.

The percentage was 49.6 on a GAAP basis and 50.6 excluding share based compensation.

Under-load charges on the 90 nanometer network and the factory's lowered gross margin from the third quarter by about 2 points.

Writedowns and other gross margins reductions in the flash memory business took nearly an additional point.

These charges and flash results offset much of the contribution to gross profit provided by higher average selling prices in unit volumes for the microprocessors.

In a year-to-year comparison, gross margin percentage excluding share based compensation, is approximately 11 points lower than the fourth quarter of 2005, primarily a result of lower average selling prices for microprocessors.

We continue to bring spending down as a percentage of revenue during the quarter and we significantly lowered head count.

R&D and MG&A were approximately $2.9 billion, excluding share based compensation of $240 million, spending was approximately $2.6 billion, a little higher than in the third quarter and just above our forecast in October.

Compared to a year ago, spending excluding the impact of share based compensation decreased approximately 12%.

As Paul mentioned, the number of employees is down nearly 6,000 to 94,000.

During the quarter, we completed the sale of certain assets of the Company's communications and application processor business.

The financial impacts of this transaction are reflected in two categories of the income statement.

The gain of $483 million, is included in interest and other income.

With the sale of this business, we plan to deconstruct facilities and to sell a factory resulting in charges from asset impairments of $317 million that are included in restructuring and asset impairment charges.

Total restructuring charges and asset impairments were $457 million, consisting of the impairments relating to the divestiture and employee severance charges from the efficiency program.

Fully diluted earnings per share were $0.26.

Excluding share based compensation, earnings per share is $0.30.

The gain from the divestiture, plus the related impairments resulted in a net gain of earnings per share in approximately $0.025.

On the balance sheet, inventory declined during the quarter by $163 million, or 4% to $4.3 billion.

Work in process and finished goods are lower than the third quarter, driven primarily by reduced wafer starts to control inventories.

Total cash investment comprised of cash, short-term investment, and fixed income trading assets ended the quarter at $9.6 billion, an increase of 1.8 billion from the third quarter.

Most of the increase came from cash flow from operations and proceeds from divestitures.

Capital spending was $1.1 billion, stock repurchases were $150 million, and dividend payments were nearly $600 million.

As we now turn to the outlook for the first quarter, please keep in mind that unless otherwise specified, the forecasts do not include the effect of any new acquisitions, divestitures, or similar transactions that may be completed after January 15th.

I will use the midpoint of forecast ranges when making comparisons to specific periods.

We are planning for revenues, which is typically lower in the first quarter to be between 8.7 and $9.3 billion, a decrease of approximately 7%.

Our expectation for gross margin percentage, including the effects of share based compensation in the first quarter is 49% plus or minus a couple of points.

This is nearly a point lower than the fourth quarter.

Lower revenues in the fourth quarter will lower gross margins.

Start-up costs related to the 45 nanometer process will reduce gross margin percentage by just under three additional points from the fourth quarter.

Fewer writedowns versus the fourth quarter and lower unit costs of microprocessors will partially offset the margin declines listed above.

Spending, R&D plus MG&A, including share based compensation, should be approximately 2.6 to $2.7 billion, down approximately 7% from the fourth quarter, due primarily to lower operating and marketing expenses.

In addition, in a separate category for restructuring and asset impairments, we expect expenses of approximately $50 million, down from 457 million in the fourth quarter.

We are on track with the previously announced plan to reduce headcount to approximately 92,000 by mid-year, which will be 10,500, or 10% lower than in 2006.

Looking ahead to the full-year, we are planning for gross margin percentage of approximately 50%, plus or minus a few points, including share based compensation.

We expect start-up costs to amount to about 2 points worth of gross margin compared to 2006 and underload charges beyond the first quarter to be minimal.

Cost saving actions in 2006 will be reflected and significantly lower spending in 2007.

The forecast for research and development is approximately $5.4 billion, almost $500 million lower than 2006.

Total spending, including R&D and MG&A,, but not restructuring charges is forecast to be $10.7 billion, down 11%, or $1.3 billion from 2006.

Efficiency efforts will also lower capital spending, which is targeted at $5.5 billion, plus or minus 200 million.

In summary, outstanding new products, leadership and manufacturing technology, comprehensive cost savings, and disciplined execution have delivered a solid second half and built a strong foundation for 2007.

Business remains competitive, and we remain focused on doing the right things in the year ahead.

Delivering compelling products, competing for sales, attacking costs, executing superbly with 45 nanometers, and pushing ahead with 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 yourself to only one question, and no more than one brief follow-up.

Thank you.

Operator?

[OPERATOR INSTRUCTIONS] And your first question comes from the line of Glen Yeung with Citigroup.

Please proceed.

Thanks.

Thanks for letting me ask the question,here.

I guess I have two questions.

Both of them are focused on gross margin, so Andy, it may be best for you.

Looking at your '07 guidance of 50%, really not a lot of improvement, even off of the fourth quarter number that you just printed.

I wondered if you could talk to us as you have in the past.

You mentioned start-up costs, what start-up costs are expected to be, what unit costs are expected to do and what you think ASPs will do in '07?

I'll almost answer all of those.

On start-up costs, first a little tutorial, in the Intel business, you always have some level of start-up costs.

We always have a new process or a new factory, or a process changing from the factory.

So in general, you would expect us in a year to have about 2 points of margin dedicated to start-up costs.

In 2007, we will have approximately 4 points of margin deterioration because of start-up costs.

Because of the startup of the 45 nanometer process.

So that's step 1.

The other thing I would tell you even though it was not asked, 70% approximately of those start-up costs will be in the first half of the year.

So it's a front-end loaded cost.

Now, in terms of unit costs, we are seeing lower unit costs and microprocessors in the first quarter.

That's one of the things that keeps margin actually a little higher than you expect with this revenue drop.

I expect to see unit costs continue down through the year.

So I think I'll have a very good progress through 2007 on unit costs of microprocessors.

The one I can't give you is we don't do an ASP forecast today.

What I would tell you is we do anticipate a continued competitive business environment.

Okay.

And part of your answers segues into the second question, which is the pattern of gross margins in '07.

Particularly given your statements about where we'll see the start-up costs over the course of the year.

It sounds like it's relatively safe to assume it's low in first half, much stronger in second half, and is it fair to say that start-up costs kind of revert back the normal in '08?

Yes, but let me be careful.

If you talk about start-up costs, it's high in the first half, lower in the second half.

I'm sorry, gross margins is what I'm referring to you.

Okay.

Gross margin then flips on that.

So I wanted to make sure we got that one.

And at this point, I would expect '08 to revert to normal, yes.

Great.

Thanks.

And your next question comes from the line of Tim Luke with Lehman Brothers.

Please proceed.

Thank you, just as a follow-up to that, as you plan your 50% gross margin guidance for the full-year and given your commentary on the first half impact to the start-up costs, should we logically see some sequential profession through each quarter of the year, or with seasonal decline in revenue would it be sort of flattish or lower in the second quarter?

Well, I really don't want to make a second quarter forecast yet.

What we did say was in the first quarter we should see just under 3 points margin deterioration for start-up costs.

We said the full-year has a 2-point effect, 70% of the profit generated within the full year is in the first half.

And other than that, I really don't want to do much more talking about the second quarter.

As a follow-up, I might then ask you.

Refer to your expenses being somewhat above your forecast and could you give some clarification there and on the assumptions for the $2 billion in cost cutting?

Thank you.

Sure.

The expenses were a bit above the range I gave at the beginning of the quarter.

There were 3 reasons for that.

One is with higher revenue, we have slightly higher Intel and size accrual expenses.

Two with higher profits than we anticipated we have a little bit higher profit-dependent expenses.

And the third effect is even though I'm 1,000 heads lower than I expected to be, the rate at which they left the company was later in the quarter, so I had a little bit more expense because of that.

What was the second?

$2 billion.

On the $2 billion.

You can actually see it if you do a year-over-year on the spending, you see it down $1.3 billion.

And tougher part to see is in manufacturing.

You'll have to take my word for it.

We could actually show you some of the programs at the analyst meeting give you a sense for the very specific programs that lead to easily a half a billion dollars and I can give you some more detail on the savings that we expect to see out of the capital budgets, as well.

I'm fairly confident I've seen $2 billion, and it wouldn't surprise me if I even get a little bit more.

Thank you.

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

Hi, Andy.

I guess the question is for you again.

I just wanted to be clear on the start-up costs impact.

Are you suggesting the gross impact is 4.0, which equates to something close to $1.5 billion of start-up costs, and is that all 45 nanometer-related?

Or does that include the start-up costs from IMFC, and if so what's the rate count?

So you're thinking of the full year?

Yes.

So in the general running of Intel's business, there are about 2 points of gross margin. 2 percentage points.

I'm letting you pick your revenues, so you multiply that by two percentage points and you'll get the baseline.

There will be an additional 2 points of margin deterioration in 2007 due to the high cost of the 45 nanometer start-up.

So, yes, take whatever your revenue is, 4% is equal to, actually not whatever yours is, but whatever ours is, it will be close enough, we won't be that far off. 4% of that cost is the total start-up costs for the company approximately in 2007.

But the incremental deterioration is only 2 points versus 2006, correct?

That's absolutely correct.

It's an incremental 2 points, of which you start to see that in the first quarter.

The first quarter, itself, has as I said, just under 3 points increased deterioration of margin due to start-up costs.

The first half of the year has 70% of those costs.

And I guess as a follow-up to this, what I'm unclear on is you think about the cost side of the equation, clearly the start-up costs are a bit of a head wind, and you do have slightly higher depreciation.

But I would have anticipated more of a benefit from lower labor costs given the reduction in manufacturing head count and then the materials line at least as a percentage of comps trending down giving better unit costs.

I'm just trying to figure out why this doesn't equate to better gross margins despite the start-up cost head wind?

Again, we could each have different models.

We do believe it's still going to be a competitive business environment and we're going to have to fight to win orders.

Okay.

Let me just ask a final question.

If you were to think about where does the biggest delta to your gross margin forecast what line item?

The biggest risk to my forecast is that the question?

Risk, positive or negative?

So the start-up costs were pretty predictable, I won't bring that as a positive versus a negative.

As is always the case, price has the biggest effect.

Cost per unit, how full the factories is probably the second biggest factor.

And what kind of price decline have you built into your forecast?

At least you save the one for last that you knew I would duck.

We're trying not to make a full-year revenue forecast.

We're trying not to make an ASP forecast.

Okay.

Thank you.

and your next question comes from the line of Joseph Osha with Merrill Lynch.

Please proceed.

Hi, there.

Two questions.

First, are you able to say what you expect inventories to be down in the first quarter or not?

What's your thought there?

What I would expect, Joe with revenue being down some, if we can kind of stay flat, I'd be pretty satisfied with that result.

That'd be a go.

So flat, you think?

That's the goal.

Tough to hold in a declining revenue environment.

I think we can we have a shot at it.

Secondly on start-up cost, just for the heck of it.

I went back and comments from the fourth quarter of '05 and talked about start-up costs for 65 nanometer being an issue.

And I guess I'm wondering whether we're in an environment now where you're going to have to deliver process innovations at an increasing rate, in order to compete.

And whether we should necessarily think about there being a point of which this kind of start-up cost treadmill ends.

I don't think so, Joe.

We're pretty consistent on our 2-year treadmill that we've been on for years and years and years.

We think we stay on it, we don't accelerate it.

We also think we don't drift back from it.

So I do think that leaves us with every two years a new generation start-up spike.

And if we did look back at '05 to see if it's in the same magnitude.

Quite frankly it's similar to the magnitude we're seeing for '07.

I'd hoped it would be a little less, because we're ramping three factories into the four, the difference is the equivalent is a little more expensive so the cost is backing up a little bit, it turns out to be the same dollars.

So you think you can jump off of this treadmill in '08?

At least briefly?

We did in '06.

So yes, if you look at '06, it was what I call the standard load.

And if you look at '08, I see no reason we can't get back to the standard load.

Okay.

Thank you.

And your next question comes from the line Hans Mosesmann with Nollenberger Capital.

Please proceed.

Thanks.

Paul, a question on there's a lot of anticipation for this operating system.

What is your observation regarding the launch and how it has impacted Q4 and perhaps the first half of the calendar year?

Thanks.

Well, we're optimistic about the launch and the prospects for Vista, particularly in the consumer markets.

I think it's going to be instant adoption for new machines right out of the chute.

And that's particularly important when you think about the fourth quarter where the action was.

The action in the fourth quarter was in consumer notebooks and they sold at record numbers despite the specter of a Vista launch in January.

So I'm -- I'm optimistic that Vista will help us across the board in consumer sales, including the hottest part of the market, which is notebooks right now.

Has it changed the linearity at all for Q1 based on historicals?

Not from our perspective.

Loading with our customers is in normal linearity in terms of the backlog by a month.

Okay.

Thank you very much.

And your next question comes from the line of John Lau with Jefferies & Co. Please proceed.

Great.

Thank you.

Wanted to circle back to the server area.

You had mentioned that you had record units and ASPs for your server units.

And given, given the competition has indicated that they are much lower, it would imply that you had gained a significant amount of market share going forward.

The gross margin is a little bit lower, you're guiding down almost a percentage lower than you did last quarter.

Do we take that to assume that you will continue to be very price-sensitive and competitive in that marketplace to continue gaining market share?

Maybe I'll answer the first part of the question, Andy can speak again to the gross margin.

We don't give and we aren't going to give a gross margin by product segment in terms of forecast.

You're right we did have record units and servers.

It wasn't a record ASP, it was a sequentially increased ASP.

And we're very comfortable with that.

That reflects rapid adoption of Woodcrest and some other quad-core products.

In terms of market segment share, I think it's a little premature to answer that question and we're optimistic that our surge in Q4 and servers will reflect on market share as the numbers are calculated.

And going forward, are you with the gross margin overall, I'm not asking about the specific server area, but if you're guiding gross margins to be lower about a percentage point, some of that has start-up costs, of course.

But are you going to maintain that competitive pressure to gain back the market share that you may have lost in 2006?

Is that part of your goal?

Well, I continue to believe that our shareholders are best served by Intel using its capacity to retake market share that we lost this year.

Nothing's changed there, except that we perhaps have a much better product portfolio to do that with.

And the flexibility and the breadth of that product portfolio gives us a number of pricing options.

Great.

Thank you very much.

And your next question comes from the line of David Wong with AG Edwards.

Please proceed.

Thank you very much.

Can you give us an indication of your gross margin guidance, how much of that is impacted by NAND start-up costs with any other assumptions associated with NAND?

Are you assuming any NAND writedowns in 2007?

Boy, I don't know that I'm assuming any NAND writedowns in 2007.

The product is just launching and getting started.

There are start-up costs and the routine part of the business associated with NAND.

If you -- one of the things that happened from the third quarter or the fourth quarter, the reason margin didn't rise as much as we would have expected with that kind of revenue growth is we did have additional NAND start-up costs and we did have some writedowns of some flash inventory.

There's nothing baked in that's remarkably different.

It is a ramping business, it is a new business, new factories are being brought to bear and as those come online, you'll see start-up costs with those, as well.

One third one if I might.

Could further divestitures change either revenue or your gross margin or OpEx guidance for 2007 or if you've pretty much done all the things that might have a big impact on 2007 income statements.

Everything we've done is reflected in the outlook we provide you and I can't comment on anything that is being considered.

Right.

Thanks very much.

And your next question comes from the line of Cody Acree with Stifel Nicolaus, please proceed.

Thanks.

Andy, can you talk about where you stand on the crossover for the new microtecture?

As far as production?

Yes.

It's essentially done.

If I look at the inventories I have in the processor space, it's almost completely transitioned.

Think of processors on the new 65 nanometers chipsets rapidly getting to be 100% of the 90 nanometer.

What variable, I guess what incremental impact do you have as you go into the next couple of quarters of utilization rates, yields, and manufacturing, if we're already all the way over to the new architecture?

How much impact can that have outside of these fluctuations and the start-up costs?

Well, show you the traditional quarterly cost unit chart at the analyst meeting.

I think what you'll see is a pretty solid progress throughout 2007 on that.

Same thing will happen in the chipsets base as we bring Crestline into the 90 nanometer 12-inch factory, we'll start to see some benefit there.

If demand went up and the factories were more loaded, you'd see some benefit there, quite frankly if demand softens, you'll see more underload charges.

There's one other factor, which I think is the mix of quad versus dual versus single core processes over the year.

And our road map would have us continuing to ramp the quad-core, which effectively utilizes more than the dual core and the dual core will continue to move down in our price stack over the course of the year, which effectively uses more capacity than single core.

We think that's a very cost effective and competitive move given the 65 nanometer die sizes.

And so you mentioned, Andy, the internal inventories in the Q1, the target is to keep them about flat regardless of the revenue trends.

Are you comfortable, then at that inventory level or as we move on into a revenue growth period are you hopeful that you can work those down any further?

Boy, I hate to make an inventory forecast after the year.

I guess it would still come down a little bit to be honest.

I'm in a situation, now we said we took about 2 points of underload charges in the fourth quarter, which was more than we expected as the quarter began.

We did cut back wafer starts in the 90 nanometer factories.

We have underload charges in the first quarter, as well again in the 90 nanometer factories.

From that point through the rest of the year, I think I'm in pretty good shape.

So, yes, I'd like to manage my inventories not to grow, I'd like to manage them to be down some.

If revenue's off seasonally in the second quarter, it makes it a little bit tough, that'd be a good goal and then to get some benefits in the back half of the year.

You want to take a shot of where you stood market share at the end of the year?

Not really, we've frequently speculated before, we've seen final numbers, it's a guessing game.

We'd just as soon wait and see what actually gets published from the people who have the numbers.

Any places you feel you definitely gained share?

Again, the speculation game hasn't worked very well.

I think we'll wait and see what gets published.

Thank you.

Next question.

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

Please proceed.

Good afternoon, guys.

Thanks so much.

Andy, can you update us on the plans for the strategy with the NOR flash business at this point going forward?

The plans and the NOR strategy on the Paul and I keep reminding them we expect to see quarterly improvements and pretax.

I'd say we made some last year, we kind of flipped back a little in the fourth quarter, which was disappointing to us.

There's no question the team's dedicated to turning that thing around and we're driving as hard as we can.

What do you think will ultimately drive the go or no go decision there?

Right now there's only a go decision and the go says make money and if you don't make money, we've got to figure out whether we have to resize the business or if there's some way to restructure it to make i more sense for us.

And you think at some point during '07 you would have to make that decision if the results didn't turn around?

Well, I guess I could say yes to that, but I'd rather say, I think Brian and the team are going to turn the results around.

That's the goal.

Got it and if I could ask you a quick question, follow-up on the CapEx.

Obviously a lot of the CapEx spending is for 45 nanometer.

Any break down between equipment percentages in '06 and '07 versus just non-equipment?

I'll give you that later, but the equipment percentage is higher than normal.

What you'll see when we show in the analyst meeting, it's one of the higher percentages we've had towards advanced 45 nanometer-type equipment than we've seen in a long, long time.

We've got savings across almost every other bucket.

We saved as head count came down, I can save on office, as the efficiency program really targeted heavily, I can save on some of the equipment we buy back there.

What you're seeing is we'll see is a higher percentage in equipment than we've historically had.

And historically we've said it's about a third fab equipment.

Right.

Terrific.

Thanks so much.

And your next question comes from the line of Adam Parker with Sanford Bernstein.

Please proceed.

Yes, hi.

Do you guys think you undershipped consumption in the fourth quarter?

I know you talked about having some channel inventory you needed to burn down.

Did you accomplish what you needed to there in Q4?

Do you think given the overall demand environment, economically speaking, that supports having the best NPU quarter in the last eight or more quarters?

This is a two-sided question.

And we talked -- you talk to our customers and see what they think about their inventories.

All information we have says inventories are under control and no issue.

If you look at our channel inventory, it's normal, it's dead flat with the third quarter, about where we expected to be.

What we really are concerned about is there may be somebody else's parts in the channel that could cause it to slow down a little bit in the first part of the first quarter.

So, okay, and just economically, you're confident that you didn't overship at all?

We don't think we did at all.

So no, we don't think we overshipped at all.

The only other question I have would be about pricing for processors.

Can you talk at all how you would characterize the pricing environment now versus a few months ago?

Was it more competitive in the second in say the second quarter of '06 than it was in the second quarter of '04, and maybe a little bit on desktop versus notebooking in that regard.

It remains competitive, Adam.

You have to remember over the course of this year, we probably had the biggest single series of price moves we've ever had.

That had to do with us resetting our entire price stack in servers, desktop, and notebooks for the introduction of the new products for all three lines that came in essentially over the summer of the first quarter.

We've never had that broad base of new products hit us all at once.

And so, as you bring the new products in, you drop the prices on the old products and price the new ones to hit the sweet spots of the volume.

So the effective thrash in pricing was really much more a function of new product technologies coming in than anything else.

I will, from a competitive perspective, I think that --

But you've had new products in the past and not taken the same cut?

Not across the board.

Not in all three segments at once.

It's the first time we've introduced top to bottom.

That implies a much more benign environment going forward, then?

There were 40 new products this year.

I said part of this was a competitive environment, and I don't think that's changing.

Part of it was also the magnitude of bringing 40 new products out.

Any desktop versus notebook distinction there or no?

It was much more acute and more competitive in desktop.

And that's your forecast going forward as well?

We're not giving a forecast.

The reason people keep asking is because the single most important variable for your company is pricing for processors and that's why we think about it.

We recognize that.

All right.

Thanks, guys.

Your next question comes from the line of Chris Danely with JP Morgan.

Please proceed.

Thanks, guys.

Nice job on the share gains.

Hey, with the, I guess $5.5 billion in CapEx, can you give us a sense of how much you expect your unit output to increase?

So I'll duck that one.

What I will tell you is we're kind of directing people to look at the market forecast to see what they think overall units will do.

Most of them are saying 8 to 10%.

And then the question becomes can we take some share or not?

But it gives you a ballpark for what we're thinking of.

And then on the inventory, in the past it's trended between, sort of 60 to 70 days and it looks like we, we're going to set the bottom here around 75 to 80 days and it was up to 100 days.

Do you feel comfortable with this 80 to 100 day new range of inventory?

Or do we expect to look at a different range, I guess next year?

As always, I'd like to be a little bit lower.

I want to be, we have to be careful in the past, I've gotten it too low and we've been burned by it.

Like I said my goal would be to kind of hold it flat to a down revenue period, a seasonally down revenue period.

And hopefully as revenue kicks up, I can have inventory down a little bit.

I guess the gist of my question, within this inventory range, do you think your gross margins can get back to the low 60% range?

I don't think I want to make long-term gross margin forecasts today.

Okay.

Thanks.

And your next question comes from the line of Mark Edelstone with Morgan Stanley, please proceed.

Thanks, a lot, guys.

Andy I might have missed it, but what are the underloading charges expected to be in the first quarter?

it was about 2 points of margin worth in the fourth quarter and I expect it to be approximately the same in the first quarter.

And then, I think that's it for the year.

Okay.

Great.

And then question for you, Paul.

Santa Rosa is obviously coming in the second quarter, but I think you've had a one month delay here as of late, can you just talk about the factors that cause that pushout?

We have not delayed the product.

It's on the same schedule it had, from the day we set it.

We've always planned to launch it.

In the first half, in the second half of the first half and we've never given a specific month, but there's no delay on it, Mark.

So basically it's a May launch now, is that correct?

I didn't give a date, I just said no delay.

Fair enough.

And I guess both from a marketing perspective as well as from a technical perspective, everything is on track to basically get that launch date?

Yes, the processor is obviously shipping.

The chipset is in the middle of the fining tuning on the graphics and I already told you we've begun shipments on the 802.11-n product.

Got it.

Good luck, thanks a lot.

And your next question comes from the line of Michael Masdea with Credit Suisse.

Please proceed.

Just kind of a high-level question.

I think you said a record number of times in your opening preamble but clearly you're getting some momentum on a lot of scenarios, but your EPS here is if I look at it on a comparable basis about 25% in December of '05.

So, I mean, what closes that GAAP beyond which you've already talked about on the cost side?

How can you really get there?

Or have we really seen at least the near term peak in what you guys can earn, given the market?

So, again, now we're looking to start making long-term EPS forecast.

I can't do that.

We know that we're down versus a year ago for two reasons.

One, revenue is down some.

Margins are down about 11 points, I said, that's almost all ASP.

We know it's been a very tough pricing year.

Hopefully we've laid a foundation, as Paul said, for 40 new products, we had 45 nanometer coming out later this year.

We have another set of new products coming out later this year.

We feel pretty good about the position we put ourselves in and we think it's our job to execute and the business will come out where the business will come out.

Maybe another piece of that.

You talked about moving more to quad-core and dual-core, and that will take up more of the capacity, et cetera.

Silicon size has inherently gone up in what you're saying there.

Do you think, do you feel confident that the ASP side will offset that and mitigate that?

In other words you can get higher prices for the multicore or does the strategy remain sort of to offer it at the high end of what the prior generation was?

Well, we brought the quad-core in the same price range of our prior dual core extremes.

We brought the quad-core mainstream products in at the same price as our prior dual-core high end products.

So that model will continue.

We'll give you an update of the overall product costs at the analyst meeting, but we remain very comfortable that the product costs will continue to come down over the course of the year.

Average product cost.

In line, at least with the overall ASP increase, I mean the size increase, at least?

Back to the mix and the average, which we really are not going to go at.

Okay.

Thank you.

And your next question comes from the line of Charlie Glavin with Needham and Company.

Please proceed.

Thanks.

Paul, going back to a comment that you made right at the end of your opening remarks about capital forecasts not being in line post your savings efficiency review.

Can you give a little more clarity given the divestitures that you've done, has there been a change to the prior 8-K statement about $1 billion of deferred CapEx for '08?

So what you're asking are we still on track to save $1 billion from the '08 forecast we never gave you, right?

Right.

Boy, that's an easy one.

The answer is: We have seen capital savings earlier than we anticipated.

But it doesn't mean we said it's pulled in.

Our expectation is to go and find $1 billion in addition in '08.

So whatever savings that I get before then, we're going to take to the bank and then we're going to go see where we can get another billion dollars.

We've been pleasantly surprised that as you start to accumulate savings in this program, more seem to start to appear.

We're pretty comfortable that we'll continue to drive capital efficiency as well as other types of efficiency in the company.

Paul, is there any disconnect between the mother board unit shipments being down?

Chip sets being relatively flat, but the higher end microprocessors being up in terms of either market share rebalancing shortages that occurred coming out of December?

Can you give a little more clarity?

Thanks.

I think there were a couple of things.

I think that there was potentially a little bit of chipset double ordering after the announced acquisition of ATI.

And that's sort of created a bit of an earlier surge than probably would have ordinarily happened for the fourth quarter, which always leads the PC sales.

And that combined with the fact that we're now seeing, people were buying in December, for January-February sales.

And that's seasonally down from the fourth quarter.

So it's really a bit of both things that hit the chipset that caused it to be flat.

Okay.

Thanks, Paul.

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

Please proceed.

Thank you very much for time.

Just a quick question.

If I look at the industry, and this is for Paul, for the PC industry for 2007.

Taking into account corporate spending and consumer situation.

Can you just share with us your thoughts as to how you see this industry developing for 2007?

Well, we're not going to give a forecast for '07 in terms of the units.

I'm personally very comfortable with the projections out of the big analytical firms in the 8-10% unit range.

I think it's going to be, the trends we're likely to see, that we're banking are Vista helping consumers, consumer sales, particularly in notebooks.

And the secular trend where the overall growth in the internet is driving more servers because it's more data to be stored.

Those two trends are not going to go away.

The question that we don't know yet is whether the combination of Vista plus the vPRO product line out of Intel will create a change in the refresh rate of enterprise-based PCs outside of notebooks.

And that's just a bit of an unknown at this point.

That's helpful.

Just if I look at 45 nanometer, can you remind us how much you think we'll get in 2007 percentage [that allow for] 45-nanometer space?

Boy, premature to give that since we haven't really announced the ship date of the first products.

Again, that's one of those things you expect to get some kind of insight on at the analyst meeting.

One last question, back to NOR flash, I know you said Q4 and you said you haven't decided.

That's a good decision at this point.

If I look into 2007 for NOR flash can you just give us an idea as to how this business will trend?

Should we expect it to kind of come back to the levels levels we've seen recently or what kind of trend are we seeing today given all of the issues in the handset sector?

Just give us an idea on how to think about it.

In general, we don't make forecasts by subsegments.

The NOR flash business revenue was up nicely in the fourth quarter, historically, it's seasonally down in the first quarter as you would expect.

We have, we have dedicated specific factory to that business, which we think should lead to an opportunity to have more market penetration.

The key is we have to make sure we have the products in the marketplace and we have to go and win orders like we do in every part of our business.

There's a possibility for nice growth there.

Again, it comes down to the fact that we just have to execute.

Probably more than any time in recent history, we feel like we're in a good position across the board and we have to go execute.

Thank you very much.

Operator, two more questions, please.

And your next question comes from the line of Mark Lipacis with Prudential.

Please proceed.

Thank you very much for taking the question.

You mentioned higher ASPs, I'm hoping that you would be willing to perhaps quantify how much better the ASPs were this quarter, and also can you help us understand to what extent was this higher ASPs driven by just having a phenomenal quarter with your servers, or within the individual segments are customers now walking themselves back up the price stack and paying a higher ASP in desktop and a higher ASP in notebook, for example?

Thanks.

Well, there's certainly was some improvement in mix to servers and notebooks in Q4 versus Q3, which gives you a better overall average.

ASPs in server were up sequentially.

ASPs and notebooks mobile were up sequentially.

ASPs and desktop were approximately flat.

Thank you.

That's real helpful.

The other follow-up question.

It feels like the Dual Core is ramping, is penetrating quite rapidly, how should we expect the quad-core to ramp?

Would you expect a similar pace of penetration of quad-core into the market as Dual Core or would you expect something a little bit slower?

I think in the performance intensive parts of the market, which are the gamers and the workstation parts of the market and DP servers, quad-core will have a very strong penetration.

It will not have a widespread consumer penetration until we get quad-core into the notebook, because that again is the fastest growing part of the market.

And it's still priced.

It's moved down nicely for the mainstream desktop.

It's not down into the highest volume price points of the retail segments yet.

And I don't really see it getting there over the course, at least the first half of '07.

Okay.

Great.

Thank you very much.

And your final question comes from the line Gus Richard with First Albany Capital.

Please proceed.

Thanks for taking my question.

Could you just talk a little bit about cutting server just slightly differently.

The Bensley platform, what percentage of revenue was that in the fourth quarter? and what do you see it being in the first?

I don't think we're in a position to give you that level of granularity, I'm sorry.

Can you, you had mentioned the shipments of Woodcrest in the fourth quarter, could you just give us a -- repeat that for me please?

I said Woodcrest was over 50% of the DP servers in Q4, which is the second quarter of production.

Okay.

And you expect that to continue to ramp going forward.

Oh, yes, it'll displace, it'll be our essentially our only DP product line, that and the quad-core versions of it very soon.

Okay and just to --

The only thing that's retarding that is customer conversion rates, and servers they tend to have a long tail in some of the conversions.

Right that switchover should finish up in the next two quarters, is that reasonable?

Sure.

Appreciate it, thank you.

Ladies and gentlemen, this now concludes the question and answer session.

At this time, I would turn the call over to Mr. Lenke for closing remarks.

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

A recorded play back will be available at approximately 5 p.m.

Pacific time, tonight.

Those interested should dial 1-888-286-8010 and reference pass code 53842511.

Thank you.

Thank you for your participation in today's conference, ladies and gentlemen.

This concludes the presentation.

You may all disconnect, and have a good day.