英特爾 (INTC) 2005 Q2 法說會逐字稿

Ladies and gentlemen, thank you for standing by and welcome to the Q2 2005 Intel Corporation earnings conference call.

My name is Carlo and I will be your coordinator for today's presentation.

At this time, all participants are in a listen-only mode, and we'll be conducting a question-and-answer session during today's conference. [OPERATOR INSTRUCTIONS] It is now my pleasure to turn the presentation over to your host for today's call, Mr. Doug Lusk, Director of Investor Relations.

Please proceed.

Thank you.

And welcome to Intel second quarter earnings conference call.

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

Before we begin, please bear with me a minute 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 rate, interest and other income, capital spending, depreciation, and amortization of acquisition-related intangibles and costs.

These statements do not reflect the potential impact of any mergers, acquisitions, divestitures, investments or other business combinations that may be completed after July 18th, 2005.

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

Finally, beginning with this conference call, we are making a slight change in the format of prepared remarks.

We will first start with Paul, who'll make a few brief comments about the quarter and then turn it over to Andy for a more detailed review of financial and product highlights.

With that, let me turn it over to Paul.

Paul?

Thanks, Doug.

Intel revenues were a record for a second quarter, posting very solid 15% growth over a year ago.

We continued to see healthy growth worldwide in the PC market, driven by two major trends.

First, continued strength in emerging markets, where we set a number of revenue records including China, and had nearly double year-over-year revenue growth in Latin America.

Our APAC region continues to be our fastest growing region, and reached a milestone by representing 51% of Intel sales in the quarter.

Secondly, we continued to see a rapid transition to notebooks in more established markets like the U.S. and Western Europe.

Our mobility group grew 50% over last year with record shipments of mobile CPU's, chipsets and wireless LAN components.

Along with strong notebook growth, our mobile microprocessor business is also benefiting from a move away from the use of desktop microprocessors in laptops, as customers appreciate the benefits of our mobile platform, such as smaller form factor and longer battery life.

We're also continuing momentum in the handsets with 80% year-over-year growth in application processor units.

Our digital enterprise group launched a number of new platforms for server and desktop, including our first products featuring the Intel Active Management technology.

This technology will help enable IT managers to monitor, install security patches or diagnose problems for their PC's on their networks, even if they are turned off or have a failed hard drive or operating system.

For the digital home, we introduced a new platform for home PC's based upon the new dual-core Pentium D processor that supports consumer electronic-like features, such as surround-sound audio, high-definition video and enhanced graphics.

We've had good market acceptance for this platform, meeting our goal of shipping several hundred thousand dual-core microprocessors in the second quarter, with plans to triple that number in Q3 and ship millions for the year.

We also announced an investment in ClickStar with Morgan Freeman's Revelations Entertainment, intended to accelerate the availability of premium content, including first-run pre-DVD released movies over the Internet.

On the manufacturing front, our industry leading 65 nanometer process technology remains on track for production later this year, and we've begun sampling dual-core 65 nanometer mobile, desktop and server products to our customers.

In summary, our investments made over the past few years in emerging markets, new platforms, such as Centrino Mobile Technology and capacity editions are paying off with strong growth in revenue and profits.

Our dual-core ramp is on track and we look forward to launching a number of important new platforms.

Before I turn this over to Andy, I want to briefly address the AMD lawsuit.

Once I do, I would ask you to refrain from any further questions on this or any other matter related to ongoing litigation.

I want to be absolutely clear about where Intel stands.

Intel competes aggressively and fairly around the world.

This formula has led to Intel's success and it will not change.

We will continue providing customers with great products at competitive prices.

Over the years, Intel has been involved in similar legal issues.

Every one of these matters has been resolved to our satisfaction.

The same is true in Japan, where we agreed to certain recommendations from the JFTC, while firmly disagreeing with their factual and legal conclusions.

We did this because it's allowed -- it allowed us to continue serving our customers without disruption.

We unequivocally disagree with AMD's claims and are confident this latest suit, like the others, will be resolved favorably to Intel.

With that, here is Andy.

Thanks, Paul.

Mobile computing and chipsets led the company to a strong finish in the first half of 2005.

Revenue, operating income and net income in the first six months each registered double-digit growth from a year ago.

Factories remain full and inventories for logic products are tight.

In addition, for the second time this year, we raised the forecast for capital spending.

Stock repurchases and cash dividends during the quarter returned $3 billion to stockholders and average shares outstanding are down 9% from their peak in 1998.

We enter the third quarter planning to sustain the double-digit year-to-year growth achieved in the first half and looking ahead to Intel's fourth consecutive year of progress and gross profit margin.

Revenue for the second quarter was approximately $9.2 billion, just over the top of the range we forecast in the April earnings release, and also just over the midpoint of our June update.

This quarter was 13 weeks following a first quarter of 14 weeks, as this is a 53-week year for Intel.

Considering the extra week, we believe the decline of 2% from the first quarter is better than underlying seasonal patterns for this period.

Revenue grew 15% in the second quarter a year ago.

On a geographic basis, as Paul mentioned, the growth leader was Asia-Pacific, where revenue grew 28% over last year, on strength in both the export and local markets.

A favorable economic environment contributed to the growth in Latin America and continuing growth and local consumption moved China towards its record.

Japan remains a bright spot with 15% year-to-year growth.

Europe saw year-to-year growth and revenue of 9% with exceptional progress in mobile computing.

On the other hand, overall revenue for the America's region was 5% lower than a year ago, primarily driven by the movement of sales from manufacturing to Asia-Pacific.

Compared to the first quarter, the Europe and Asia regions followed seasonal patterns, while the Americas and Japan were lower than typical.

Revenue was down slightly in the channel where the normal seasonality was partially offset by good growth and mobile processors and wireless components.

As is usually the case in the second quarter, microprocessor sales were slow in mature markets in Europe leading to an overall decline for the broader regions of revenue of 14%.

Unit shipments of total microprocessors were approximately flat with the first quarter, while the average selling price was slightly lower.

Unit shipment of chipsets, flash memory and wireless connectivity were higher, while motherboard shipments were down.

Gross margin dollars of approximately 5.2 billion declined 7% sequentially and increased 9% year-to-year.

On a percentage basis, gross margin was 56.4%, within the range we forecast in June.

Gross margin was 3 points lower than in the first quarter.

The largest component of the decline from the first quarter was lower revenue, driven by both unit mix and pricing.

In addition, start-up costs related to the ramp of 65 nanometer process technology were, as we anticipated in April, approximately $100 higher in the second quarter than in the first.

In a year-to-year comparison, gross margin percentage is 3 points lower than in the second quarter of 2004.

The decline reflects higher start-up costs and lower margins in non-logic products.

Spending on R&D, marketing, and G&A was $2.5 billion, lower than we expected in April and June.

This is flat with the first quarter's 14 weeks of spending.

Compared to the second quarter of 2004, spending grew at half the rate of revenue and was also a smaller percentage of revenue.

The number of employees grows during the quarter to 91,000 at the end of June.

A year-to-year increase of 11%.

Operating income was $2.6 billion, a sequential decline of 13%, and a year-to-year increase of 11%.

As a percentage of revenue, operating income was down from the previous quarter and the quarter one year ago.

Before looking at non-operating income, I will highlight results for Intel's two largest operating segments. 65% of total revenue came from the Digital Enterprise Group.

That is down from 73% a year ago.

Revenues from this group are declining as a percentage of Intel total, as mobile computing penetrates the enterprise and other market segments and as notebook customers rely more on Intel processors designed specifically for notebooks.

Secular trends and desktop computing and a more competitive market in service have meant that growth rates for microprocessor revenue in the Digital Enterprise Group have been slower than those for the Company.

Total revenue for the group was down 6% from the first quarter and approximately flat in the second quarter a year ago.

Microprocessor revenue for the group was down 7% sequentially and approximately flat year-to-year.

Revenue from chipsets, on the other hand, is growing as we focus increasingly on platforms.

Revenue from chipsets, motherboards and other was approximately flat with the first quarter and up 14% year-on-year.

Sequential and year-to-year declines in operating income were largely a function of revenue trends and higher start-up costs, most of which fell to the Digital Enterprise Group.

Revenue in Intel's Mobility Group was 34% of total revenue, up substantially from 26% of the total in the second quarter of 2004.

We continued to see strong notebook demand around the world, especially in Europe and strong growth in both the consumer and business segments.

Revenue from mobile microprocessors of $2.1 billion, grew 7% sequentially and 68% year-to-year.

With the success of Intel's mobile platforms, chipsets led to growth rates in mobile, with revenue from chipsets and other products of 10% sequentially and 94% year-to-year.

Our flash business shipped record units that saw ASPs fall in a very competitive market.

We retained our number one NOR position in the quarter.

Revenue trends in the Mobility Group drove a sequential increase in operating income of 4%, and a year-to-year increase of 92%.

For the Company overall, the total in the second quarter, with interest income, other income and gains and losses on equity investments was $105 million, consistent with our overall forecast of 100 million.

Net losses from equity investments were 22 million, which were largely due to impairment charges offset to some extent by gains.

Also within this category of the income statement, interest and other income was $127 million.

Fully diluted earnings per share, which includes potential dilution attributable to employee stock options was $0.33.

Basic earnings per share, which does not include potential dilution was also $0.33.

Average shares when calculating diluted earnings per share were 6.2 billion.

Basic shares outstanding was 6.1 billion, down 1% from the first quarter and 5% from a year ago.

During the quarter, we repurchased 99 million shares at a cost of $2.5 billion.

On the balance sheet, inventories of $2.7 billion were down slightly from the first quarter by $69 million.

Inventories are leaner than we would like, and we will be challenged to make significant progress in building inventory levels in the third quarter at current level of demand.

Total cash investments comprised of cash, short-term investments and fixed income trading assets, ended the quarter at $14.5 billion, a decrease of 1.3 billion from the first quarter, after stock repurchases of $2.5 billion, capital spending of 1.4 billion, and dividend payments of nearly $500 million.

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

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

We are planning for revenue in the third quarter to be between $9.6 and $10.2 billion.

The midpoint of this range would translate into seasonal, sequential growth of 7% and good year-to-year growth of 17%.

Our expectations for the gross margin percentage in the third quarter is 60% plus or minus a couple of points.

This is more than three points higher than the second quarter, lifted by what we anticipate will be lower start-up costs and lower unit costs.

Spending, R&D plus and MG&A, should be between $2.8 and $2.9 billion, up from 2.5 billion in the second quarter.

The largest part of the anticipated increase is tied to growth in revenue and profits.

In addition, R&D will increase as we begin production on 65 nanometer and redeploy resources to development of 45 nanometer.

We are also increasing R&D investment in new platforms.

Depreciation should be between $1 and $1.1 billion, approximately flat with the second quarter.

We expect amortization of acquisition-related intangibles for the third quarter to be approximately $30 million.

Our estimate for gains and losses from equity investment, and interest and other income, is a net gain of $130 million.

The entire year we have narrowed our outlook for gross margin percentage to reflect first half performance and third quarter outlook.

The revised forecast for the year is approximately 59%, plus or minus a couple of points.

The midpoint would be the fourth executive year of growth in gross margins.

We have also raised our forecast for capital spending for the year and are now targeting $5.9 billion, plus or minus $200 million.

Our prior forecast at a midpoint of $5.6 billion.

The increase is driven primarily by capacity investments to support 2006 and 2007 volume forecast of micro processor and chipsets to support our platform initiatives.

The outlook for R&D spending for the year is unchanged at approximately $5.2 billion.

Depreciation forecast is now $4.3 to $4.4 billion, a slight change from 4.4 billion, plus or minus, $100 million.

The estimated tax rate for the third and fourth quarter is approximately 30.5%.

Slightly lower than the previous forecast of 31%.

This does not reflect the possible impact of any potential repatriation of cash under the American Creation Jobs Act.

Please keep in mind that we expect to complete our analysis of what, if anything, we will do with regard to this in September.

The decision could impact a third-quarter tax rate.

In summary, this was a good second quarter.

A period when business is typically slow.

The momentum of the first half appears to be continuing as we enter the third quarter.

Demand is strong.

The factories are full.

We are ahead of our cost targets.

And business is generating high level to cash.

While we can't speak for the economy, we feel confident that if we stay focussed and disciplined, we will be able to continue to earn our customers' confidence in the marketplace.

With that, let me turn it back to Doug.

Okay.

Thanks, Andy.

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

We'll 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?

[OPERATOR INSTRUCTIONS] Sir, our first question is from the line of Christopher Danely from J.P. Morgan.

Thanks, guys.

Just a quick question around the gross margins.

I think you originally guided for 57, and for the midpoint we came in a little bit below that, Andy, can you talk about why we came in a little bit below the midpoint and then I have a brief follow up.

Sure.

I did expect around 57, that was my midpoint, so about a half a point light on that. $40, $45 million.

There is basically three reasons, a couple ASP centers and one reserve center.

In the ASP side, we were a little bit less than we expected because we shipped more Xbox than we expected, so that mix hurts a little bit.

We had a little bit of pricing pressure in computing processors, and then we had a few reserves scattered around a few different product lines.

Okay.

And just the follow up is, you guys are keeping the year gross margin guidance at 59%, and if I just plug in some typical Q4 seasonal growth, I get about 50 points of -- 50 basis points of gross margin improvement in Q4.

A, can you just go into the reasons why we wouldn't get a lot of incremental gross margins in Q4.

And, B, do you think Q4 could be the peaking gross margins, or short story long, where do you think your gross margins can peak and why?

You got me on one, it's going to be hard for me to say anything meaningful about, Christopher.

If you look at the first three quarters using the midpoint for the third quarter, at average would be below 59% just slightly.

It takes an awful lot of margin percent in one quarter to offset the other three quarters if you're doing a weighted average for the year.

As a result of that, I actually feel pretty comfortable at 59 plus or minus.

Would I like to do better?

I always would like to do better you, but you look at the business and you get what you get.

So I think things are going well.

Certainly we pick up -- if we hit the midpoint, more than 3 points going into the third quarter, start-up costs are lower, just as we expected.

We're seeing progress on unit costs.

The factories are still full.

So business is good.

Things are going well.

I'll make a Q4 forecast when we get to it.

Picking your peak gross margin is always a dangerous thing to do.

Every time you think you pick one, the economy will change or the mix will change or something.

Right now, I am pretty satisfied with what the business is generating.

I see full factories, I see competent schedule.

I see the move to mobile products, which is very good for us.

I think the business is doing well.

That's about all I can say.

Okay.

So no peak gross margin estimate.

I can't do that.

That's fair.

Thanks, guys.

Sir, our next question is from the line of Jim Covello, with Goldman Sachs.

Thanks so much.

Andy, question -- back in the May analysts meeting, you had talked about capital intensity as the 300 millimeter ramps, and you'd talked about a 30% average from '03 to '06.

You cautioned us at the time not to make that too granular.

But as you've raised your CapEx a couple times here, does that estimate change or do you think over the three or four-year period it still stays roughly similar to that estimate you had provided us back in May.

I'm certainly hoping it stays roughly similar, that the increase we're seeing now is a function of the first half of business leasing.

Let's not kid ourselves, chipsets are still extremely tight, profits are the bottom line.

The first half demand was stronger than we expected going in.

What we're looking at is if -- the worldwide economic strength continues, without a little more capacity, we could end up in the same situation for the next 18 months.

So we're trying to give ourselves a little head room assuming the economy stays healthy.

Terrific.

And then quick follow up, on the flash business, units doing very well, continued pressure in ASP's, are you happy with the cost model right now in flash or do you think we need to take a look at that?

In terms of the Intel business, taking me down to a level I hate to go to.

I'm reasonably happy with the cost model inside Intel's factories.

It is full factories.

We continue to look at the process to take steps out to get the cost down.

It is not the same cost structure you had if you're in a low cost geo.

So if there are opportunities to improve it, but inside the facilities I have, which are mostly handed to the flash business depreciated from the logic business, I think I have a pretty competitive business.

It is just at this point, in that business, it is real intense, tough market right now.

Terrific.

Thanks very much.

Sir, our next question is from the line of Ben Lynch with Deutsche Bank.

Hi, this is a question for Andy, actually.

Andy, when I look at some of the numbers, the record process of volumes, so this means they were up Q and Q, and if I take the processor revenues within digital enterprise and mobile products, it looks like they were down about nearly 3%.

So that would suggest that ASP's were down more than 3%.

It doesn't sound like Xbox alone would be able to explain that much given its overall portion of the mix.

Can you maybe give a little bit more color as to what else might be causing that effect on the ASPs?

And I have a follow up, please.

There was a decline quarter to quarter in ASPs, I would say half Xbox and half computing.

If you look at the computing half it is a variety of things.

We do have price pressure in the service space right now.

But also keep in mind as mobile is exploding, a big part of why it's exploding is mobile Celeron.

So it is not the same ASP as the Centrino platform.

Again, one of the things you count on is the mix improving to offset, what's a standard decline in the ASP if you don't find reasons to help it.

In reality, look at the quarter to quarter, and assume about half of it's due to Xbox and the other half is due to a couple of things, little bit of price pressure and server, and mobile Celeron expanding in the low end.

Great.

And just the follow up question I have, Andy, is when can we -- beyond just the general comments you guys tend to make, IDF [ph] et cetera, when can we really expect you guys to get the momentum going with regards to server competitiveness where clearly you are facing some pressure?

And also could you comment on whether from your perspective the ramp of AMD's Turion product is impacting your business, for example, your mix of Celeron versus Pentium M that you refer to.

Let me take this part of it, Ben.

In terms of the mobile -- or the server product line, we had a fairly weak product line in 2004, as we all know and talked about.

In early 2005 we -- mid-first quarter we introduced a platform refresh which significantly increased our competitiveness in the Xeon space and DP and MP processors and a new platform around that.

Subsequent to that, we've also introduced our first dual-core uniprocessor platform just this last month.

And we are, as I said in my comments, sampling our dual-core processors for servers, both Xeon and Itanium to our customers as we speak.

The next big refresh was a dual-core DP and MP product called Paxville, scheduled for early 2006 and that is highly likely to see that come into '05.

And then there is a platform refresh after that in mid-'06 that is also on target.

So you'll see a new platform, not just processor, but platform from Intel kind of chunking out every six months or so along the typical server cadence.

I think we have done a good job holding our business through a very difficult competitive period and I think that as I look forward in the server product line, I see us getting increasingly competitive.

Great.

Okay.

So starting now or starting with Paxville, do you think?

Beg your pardon.

Starting now or starting with Paxville?

As I said, we've introduced a refresh in Q1, a DP product just this last month and the next jump will be Paxville followed by the Dempsey.

So it's in the process of turning around now.

Basically.

I can't -- IT buyers in this space are very conservative, and things don't move overnight and the preponderance of momentum is in our favor as we sample and feed our next-generation platforms.

Thank you.

And then on the Turion and notebook --?

Really haven't seen much.

Okay.

So what would you account the Celeron and Pentium M shift to?

That was more of a -- there were a lot of desktops -- a lot of notebooks being built with low-end desktop products.

And what we saw was a shift here to Celeron M-based products, particularly for some of the low-end consumer notebooks.

They were more notebook in their flavor and it changed the profile of our mobile processors, if you ramp that Celeron M against the base of Centrino.

Great.

Thank you very much.

Sir, our next question is from the line of Tom Thornhill with UBS.

The gross margin guidance that you're giving were 22.

Andy, can you quantify how much the 65 nanometer start-up cost going away is impacting that, to what degree this is driven just by revenue increase, or is there also some mix being factored into that going into Q3?

Sure, Tom.

I would have said, if you only give me the choices I gave between the 65 nanometer diminishing and the start-up cost, and lower unit cost, those will explain all of the increase in margin and those are comparable in size.

Probably the start-up cost going away is little bit larger.

If the -- follow-on to this then would be if the 65 nanometer going away helps step it up, and some of this is the same but not quite the mix in units, then wouldn't it follow that Q4 on a little higher revenue would show a little more leverage?

Because if we only average 59 for the year, it assumes very limited gross margin improvement in Q4, which is seasonally a stronger quarter.

It is certainly possible, Tom.

Keep in mind though that I've got my factories running full out now.

So normally as you enter that fourth quarter you're getting the benefit of factories getting fuller and costs getting even better.

I have gotten most of that cost good news.

Now on the other hand, I do have my new 65 nanometer factory starting to ramp and those could give me some extra benefit.

I really want to wait until I get out a quarter to give you a Q4 number, but there are more positive than negative going into the fourth quarter.

Can I do one follow-on on ASP's?

Sure.

In -- Paul, as the Paxville product starts to ramp and you mentioned you might pull that into '05, will that change the ASP trend in servers, as that mix shifts and Paxville ramps followed by the Dempsey.

Well, it certainly, I think, increases our competitiveness.

It really depends on how much volume we get into the second half of this year.

That's where I'm not comfortable making a comment at this point in time of how much -- and what the mix is of MP versus DP.

I think though it's fair to say, and Andy said earlier, we're seeing a generally more competitive set of pricing in servers.

Okay.

Thank you.

And, sir, our next question is from the line of Adam Parker with Sanford Bernstein.

Andy, I didn't quite understand your response to an earlier question.

A quarter or two ago you had said that you thought you would get 200 to 300 basis points of gross margin expansion Q3 versus Q2 because of lower 65 nanometer start-up costs.

What do you think that is now, all else equal?

What I think it is now, it's in the neighborhood of 200, yes.

200.

And then you said that would be half of --?

I said they were approximately the same, but that the start-up costs were a little bit larger.

Got you.

You have 3.5 points, little more than half, as approximately the same size.

Great.

Can you talk about what the -- I'm trying to understand a little bit more color on the incremental operating margins in that Mobility Group.

What was the quarter-to-quarter change in losses in flash memory Q2 versus Q1 and was that different than your expectations at all?

We don't break out the flash at the operating profit line.

Anymore.

I don't think we ever did, to be honest.

We've reported revenue for a long time, but not operating profit.

But in reality, it was a tough quarter and it was not as good as I expected.

We did have a little bit of loss of revenue because of the fire at ASE and the pricing environment was tougher than we expected going in.

So can you help me with the number, 50 million less or 100 million less?

I really can't give you the number.

That is the best I can do for you.

One last follow up then.

How should I think about the incremental profits then, in that Mobility segment?

Should it result in Q2, be -- how I should think about the incremental operating margins or can I expect a bit more lift, or is it just too tough to call with the mix within notebook.

It is difficult to call for a variety of reasons.

Keep in mind that the Mobility Group has a number of different groups in it.

It has the cell phone chipsets in it.

It is an emerging business for us.

And as we try to drive that business in a ramp phase, it is still an investment.

If you've got a lot of revenue there, it's going to have a different margin profile than if you got the revenue in mobile microprocessors.

It's going to have a different profile if it's chipsets going into or connectivity chips.

So I understand we need to get some history for you guys.

I think we've put some in.

I have six quarters of history on there, so you can start to create some models.

I can't give you much more sub-level than that at this point.

Okay.

That's fair.

Paul, one other question, sorry.

Are the 65 -- it seems like the 65 nanometer dual-core product ramps are better than your previous expectations?

Is that fair?

How would you compare the ramp so far versus the Prescott ramp for the fall of '03?

At the risk of jinxing things, it feels significantly better than Prescott took on at this point in time.

As you are all well aware, we did a reset of our whole engineering, scheduling, resource, confidence factors and so forth, sort of this time last year, and went to high confidence scheduling and I think you're seeing us execute on the results of that high confidence scheduling program, number one.

Number two, the product health, since we've had them longer than we had the 90 nanometer products at this point in time, feels better, and so we're more confident with the ramp.

All right.

Great.

Thanks much.

And, sir, our next question is from the line of Mark Edelstone with Morgan Stanley.

Good afternoon.

First off, nice quarter overall.

Paul, just some other questions on servers, if I could.

It sounds like the Xeon revenues were down quarter to quarter and I wanted to get a sense of what the change was like there.

And then can you give us a sense of what the power dissipation is going to look like as you bring out Dempsey and Paxville platforms.

I don't know if I want to get into that much detail, but I'll tell you, Mark, on server -- on Xeon quarter to quarter is that server units were up double-digit percent quarter to quarter, and -- sorry, year-over-year.

And there was a difficult pricing scenario.

So you can work from there.

The power envelopes, we have a number of products coming out, including some low-power products for blades.

I don't know how much detail you want to get into.

But in general, what you'll see us draw is a boundary condition on the power envelopes, it jumps up a little bit with Paxville, that sets an envelope for us to be able to operate in over the next few years, with low power blade versions coming out underneath those.

Paul, on the Xeon, I was looking really more for a sequential compare?

I don't know that I can give you that level of detail right now.

Sorry, Mark.

No problem.

And then just one other question on some of the segment details.

Andy, I wonder if you can just give us what the sequential compare would look like for chipsets and motherboards combined, both the digital -- basically the desktop, as well as the notebook side if you were to pull out some of the platform initiatives, if you pulled out things like the Wi-Fi adder and just looked at strictly logic chipsets and motherboards, what would that sequential change look like?

Boy, we had them combined what you see is chipsets growing faster than processors right now.

In the first quarter motherboards, and the second quarter motherboards were constrained a little bit.

So I guess what I'd say is you would see faster growth in motherboards and chipsets than you would see in processors.

The wireless connectivity in mobile will be similar to chipsets.

It's coming from a lower base of penetration to a higher base of penetration.

Okay.

So I guess the decline in revenues then really comes from things in networking and other products that you classify as "Other" in that Digital Entertainment Group?

Digital Enterprise Group.

Digital Enterprise, excuse me.

Again, which quarters are you talking about?

Just looking at Q1 to Q2.

Looking at Q1 to Q2, the biggest decline drop would be motherboards, which are somewhat constrained by our constraint in chipsets.

We want to make sure the customers got the part they needed first.

As a result, our motherboards were lower in quarter-to-quarter comparisons.

In terms of microprocessors, what we said was units kind of flat, maybe down a little bit in the Digital Enterprise Group, ASP's, little bit of pressure there.

Okay.

Great.

Thanks a lot, guys.

Sir, our next question is from the line of David Wong with AG Edwards.

The increase in capital spending you talked about, can you give us a feel for what change in production capability that would lead you to sort of making sure you have more capacity?

Is there a percentage you can give us?

It is difficult to give you what the percentage change related to the approximately 5% income in this year's capital spending.

It has been for a wide variety of different projects.

It goes some to the back end, it goes some into the fab network, it goes some into the engineering labs.

So it's, again, a wide variety of projects.

By itself, it wouldn't be meaningful.

It's got to be an added base on top of the growth in the 12-inch network that was already underway.

Great.

Thanks.

Sir, our next question is from Glen Yeung with Smith Barney.

Thanks.

I guess I have two questions.

One is a bit shorter term and one is a bit longer term.

On the short-term side, if we look into the third quarter, we're looking at seasonal strength.

Your factories are all running full, as you suggest.

Are there concerns as we go into that stronger period in the next couple of months that you may, in fact, be short of parts?

My concern is the chipsets side.

On the processor side I'm in balance today where capacity has been coming on line.

I think I'll be okay.

On the 8-inch network where the chipsets are made, it will continue to be a struggle for us through the third quarter.

Okay.

The longer-term question was on gross margins.

You've been higher than this year is 59% bogie a few times in the past.

If we look at, for example, in the last six years, three of those years you had record CapEx, just as a data point.

If we look at structurally how your business is composed today, versus those years in the past when you've had higher than 59% gross margins, is there anything different that could prevent you from being higher, whether it's next year or two years from now or ten years from now than the 59% number?

The thing that would prevent us from -- of course, there are a lot of things that prevent and effect margins.

Number one will be how full are the factories, how strong is demand.

Number two will be, how strong is the road map, for example, servers.

How fast and how long does the transition to mobile continue?

So the questions are endless.

Could it be better?

Of course.

Could you have a worldwide recession and have it be worse?

The answer is, unfortunately, yes, of course.

I really don't want to make a long-term or even an '06 forecast at this point.

If we can continue to expand the margin this year, I'll be delighted.

Okay.

Thanks.

And, sir, our next question is from the line of Michael Masdea with Intel.

I'm with Intel now, maybe I should change the question then.

To follow up to that last question, maybe ask a little bit more specific, 4Q '03 you did 63.5% gross margin.

Your factories were running pretty full back then.

What is different between then and now, given your factories are running pretty full right now when you look at your gross margin?

I'll be honest you, I did not do a gross margin reconciliation between the midpoint of the Q3 out to two years ago, so I don't want to try to answer that question here.

If you're asking, could margin grow to 63%, the answer is of course it could.

But, again, it could also decline to 55%.

What I've given you is a best-guess midpoint of 60.

That includes start-up costs getting better, that includes unit costs getting better.

If my factories stay full and demand stays as it is, it is a pretty good number.

If you compare, if you go back to the first half of '03, you guys substantially outgrew the end markets and there was no inventory problem.

You had nice ASP effect from Centrino.

Last year in the second quarter you outgrew the end markets quite a bit and then you had inventory problem.

This year we're outgrowing the end markets, it looks like Intel is in the first half pretty robustly.

What gives you the confidence that we're not pulling in some demand and we're not seeing an inventory build out there and given the shortages, we're not seeing some aggressive ordering?

First of all, we don't think we're outgrowing the end markets.

We think we're growing pretty much with the end markets.

So I don't know where you get that data point.

We do look at unit growth year-over-year, we look at first quarter, we look at second quarter, I don't sense growing faster than end markets.

Second thing is, we do go out and look at the channels.

You can see our inventory.

We look at our own inventories.

We talk to our customers.

There's definitely no evidence of an inventory build anyplace right now.

Does that mean it could be hidden from us?

Of course it could be.

Last year we were surprised at the end of the quarter.

There had been inventory build, we missed it.

Believe me, every sign we could have seen last year, we have looked for this year, and we don't see anything like that.

Is there any increased risk given that in the past you said chipsets are a good indicator that you guys use, and given the shortages on the chipset side, does that decrease your visibility at all or do you still feel like there is enough other data points out there for you guys to use?

Chipsets are a good indicator, the requests from a customer for chipset orders have not changed.

We get a pretty good sense of what they would like to have and what the market would like to grow.

We also know there are other companies out there who can meet some of the demands.

We're pretty confident we'll be satisfied some place.

To us, the biggest problem is we're losing some revenue we could have had if we'd have had a little bit more capacity.

So, no, there's no sign looking at the chipset side, no leading indicator that says the demand is about to fall.

Last question on that other piece of your business.

You went from $62 to $80 million, but your loss went from $430 to $483 million.

What is going on there and what changes are you making to address that, if any?

That is where we -- the way we run a budgeting process, we put a profit dependent above a normal target rate.

So when Paul sets a plan for us, if we meet our plan, here is how much bonus that should do go with it.

If we do better than that plan, so that we're creating more profit dependent expenses, and that's retirement expenses, that's bonus expenses, they will be put into that other bucket.

So what it means is we're doing better than we expected.

Great.

Thanks a lot.

Sir, our next question is from the line of Joe Osha with Merrill Lynch.

Hi, guys.

First just a housekeeping question.

The implication of your full-year guidance on expenses is that the R&D is going to decline in dollar terms in Q4, am I correct there?

You bumped the Q3 number a lot, but, the full-year number is unchanged.

So does that mean it goes down in the fourth quarter?

I didn't realize I gave an R&D Q3 forecast, so it's hard for me to answer that question.

You gave an operating expense number for Q3, but then you left your full-year R&D number unchanged.

Right.

The number you're seeing in the third quarter was essentially consistent with the plan we put together the beginning of the year, it's a little bit higher to be honest.

And I'll actually answer your question even though we provided no guidance consistent with that.

I see R&D in the fourth quarter being approximately flat to the third quarter.

For the record, okay.

It is tough to get to 5.2 billion for the whole year.

Looking at that.

Okay.

Enough.

Secondly, can you maybe give us a little insight in terms of facility ramps next year?

My understanding is that incrementally you're going to see capacity coming from this conversion of Chandler from 200 to 300, which is coming back up as well as the second phase of Ireland.

Is that correct?

That's correct.

Third and final, and just back to the server issue, I'm trying to understand is -- I guess I get a little bit tangled up here.

Paul, did you say that the business was up year on year on unit terms, but down in revenue terms because of pricing?

Am I characterizing what you said correctly there?

I said it was up year on year double digits in unit terms, I said it is a tough pricing environment.

I didn't give you a revenue number.

Okay.

Okay.

That's fine.

The last question before I go away, to the extent you can comment, is -- are you adding to your 300 millimeter manufacturing plant in 2006 at the same level that you did in '05, or more, or less?

Can you give me some visibility there?

In essence, it would be more than we had in '05.

Recognize in '05 we're just now ramping the process in the back half of the year with the Chandler facility.

My guess is a little bit more.

Okay.

And that is in sort of acres of silicon terms or unit output.

We think of it in terms of wafer starts.

Thanks a lot, gentlemen.

Our next question is from the line of Allan Mishan with CIBC World Markets.

I have a question about the accounting of the gross margin.

Andy, you talk about taking underutilization charges when the factories are not full enough in flash, you spoke about that in the past.

Now that you're running so full in chipsets and logic, are you taking benefits from that right now?

And, if so, do those reverse in the fourth quarter?

Accounting doesn't allow you to do that.

Accounting requires if you're not using a facility to take the excess and make it a period expense.

If you're running your factory at, quote, more than full, what you end up with a little bit lower unit cost, which gets expensed when the product gets sold and shipped.

So, no, there is -- there is no extra credit for running the factories real full.

Okay.

Thanks.

And just a housekeeping question.

The revenues from Xbox, is that recognized in the all other section, or is that part of Digital Enterprise, where do you put that?

It is in the "All other" section, yes.

Okay.

Thank you.

And, sir, our next question is from the line of Michael McConnell with Pacific Crest Securities.

Paul, question on the flash business.

Just trying to understand the dynamics there with respect to pricing.

Could you talk about what you're seeing in the market?

Are you seeing the density curve starting to slow, given some of the incremental data points you've heard from other companies talking about feature-rich phones and the slowness we've seen in that particular segment of the mobile phone market?

Could you talk about what is going on there, please.

Yes, for two reasons, at least from our perspective, we're seeing that one is that we are -- we have reentered in the last few quarters the embedded market which tends to have a lower density on average than the -- versus the cellular market.

And number two, as you pointed out, we've seen a slower than expected ramp on some of the 3G phones.

The feature risk 3G phones.

The combination of those two has the density curve a little less than we would have expected and therefore, the pricing a little more aggressive.

Okay.

And then one last one, thank you for that.

One last one for Andy.

Andy, what effect, if you could talk about it, what effect did flash memory have?

It looks like you had a 9% decline sequentially there off of another 10% decline in March sequentially in revenues.

What effect did that have in the Mobility Group overall with the operating margins?

I understand that you were shipping more Celeron M, and that was a definite component that compressed the operating margin expansion, but did flash also contribute to that?

Certainly, the place you actually see, I talked a little bit about that, is when you look at the year-over-year, I talked about the margin and the non-processing space as not being as good.

So again, I can't give you specific flash information, but it is a low margin product, and when you're under price pressure and revenue is dropping, it tends to be felt fully in the operating margins.

Would it be fair to say, then, that the effect that you saw in terms of the adverse mix in the processor business, both desktop and mobile, outweighed the effect that you saw from maybe a weaker flash environment on the pricing front?

From the first quarter to the second quarter?

Correct.

Just the overall effect on margins.

That would be fair to say, yes.

Thank you.

And, sir, the next question is from the line of Krishna Shankar with JMP Securities.

Yes, in the case of the mobile market you talked about some pressure on the ASP's because of the trend for desktop replacement notebooks to mostly value.

Do you think that we are at the beginning of a secular trend toward lower notebook pricing, which had a significant premium or desktop pricing or is this just a temporary phenomena as the shift occurs.

I think it is more the latter, although it has been clear that notebook prices continue to come down, which is one of the reasons we've seen expansion of notebook volumes in consumer segments around the world.

Our job, then, is to make sure that as the consumer notebooks expand in the market, that they represent a reasonable mix for us in terms of performance versus value products.

Hence, the continued shift of our Centrino advertising towards consumers.

Okay.

And then could you talk about the initial response to the dual-core processors this last quarter, and how that is shaping up?

We exceeded our internal volume goals for Q2 and we have the -- triple that number forecast for Q3 with millions in Q4.

It is virtually all of the materials going into the high-end desktop, gamer, high-power user environment, and it is selling quite well.

And are you sort of committing that through aggressive pricing or are you able to get the price premium for performance there?

It is still priced as a premium performance product.

Thank you.

Operator, we'll take two more questions, please.

Yes, sir.

Our next question is from the line of Apjit Walia with RBC.

One big picture question, you talked about your revenue around the base getting bigger from the Asia-PAC region, do you believe that could potentially -- as the end consumer that is different than what it might be in the developed world, is there a change or difference in behavior, are they refreshing their PC's more often or are you seeing any of these trends which could potentially change your profile going forward?

Initially.

We haven't seen that change.

I think that I commented a couple years ago that the first country market, to shift to Pentium 4 from Pentium 3 was China, and China was also one of the leading countries in the adoption of Centrino mobile technology when that was launched.

In the tier 1 and tier 2 cities of the Asian markets, the profile is very much like, and in some cases slightly ahead of that of the United States and Western Europe in terms of the mix of products and the ramp of new technologies they buy.

As we go to tier 3, tier 4, and tier 5 cities, it is a little bit different, because you're dealing with people who are buying their first computer.

So you have a higher mix to desktops.

Sometimes prices, the most essential matter, and so you see a little bit more of a volume desktop, value desktop trend there.

Our job is to market aggressively in those subsequent tiers in cities to provide the value that we've gotten out of the first two tiers.

One thing I guess for Andy, you mentioned some questions ago, there were two obviously extreme scenarios of gross margins, being 63% and then it could also be as low as 55, you're also saying unlike last year, you've taken much more critical steps to make sure you don't repeat the same issues as last year, what could you envision could be scenarios where gross margins can come in as low as 65%?

What could go wrong in the next couple of quarters.

I can't do that one.

I was trying to give two extremes and say we can build a case around either extreme.

Can we build it?

My guess is that you can build an extreme case better than I can, where demand falls off, factories are underutilized, you have reserves.

You can build that case.

Do I believe it?

Absolutely not.

If I did it would have been inside my range of forecast.

It is not there.

I was just trying to make an extreme example that it is really not usually much value for us to make up the story at 65 or make up the story of 55, which any of us could do if we wanted to.

I agree with you there.

Lastly, overall in the dual core is there any way you can guesstimate for the second half of the year or next quarter, how much overall in terms of product mix it could be.

It really hasn't changed much from the slide I showed at the analysts meeting.

Remember I showed end of '05, end of '06, end of '07, for each of the major segments.

It is fundamentally on that track.

Okay.

Great.

Thank you.

Sir, our final question comes from the line of Tim Luke with Lehman Brothers.

Thank you, Doug, for squeezing me in.

I was just wondering, on the inventory side, you were a little lower this quarter, how do you think we should think about that going forward?

Do you expect that you'll be able to build inventory or how do you see it?

It's a several questions.

I'd like to build inventory.

I think we're lower than we should be.

It is making it somewhat difficult to respond to customer needs.

It makes it difficult to adapt to short-term swings in the marketplace.

So we're lower than we would like to be.

I would like to think I could reestablish some inventory in the processor space.

But even that, like I said, we're kind of imbalanced.

It is not like I have a whole on whole lot of extra capacity in that space.

In the others, all of the 8-inch networks, not just chipsets and flash, and just any product we make, it is extremely tight.

So I don't think we'll make a lot of progress, but it is the goal, if I can find a way to get a few extra wafers, if we can find a way to get a little bit of extra yield and give ourselves some buffer, we would love to do that.

My goal is to increase it, it's going to be a challenge to do that in the third quarter.

Just to get back to Adam's question on the third quarter gross margin guidance, where previously, with respect to 65 nanometers costs coming off, I think you said you would be up 200 or 300 basis points in the third quarter from the second level.

Now it seems that your incremental lift would be 3.5, 400 basis points and I was trying to get a sense of what factors there are that are contributing to that reasonably significant lift?

Like I said, little more than half of it comes from reduction of the start-up costs.

The other is lower cost per units on the products that we're selling.

Again, as you run factories full, as you get a little bit better yield, you get a little bit lower unit cost and those things do count in the long run.

So it is a combination of the two, which reduction and start-up costs is a bigger factor.

And the curve of pricing pressure, particularly in the server arena is in what sort of direction, vis-a-vis your expectation.

It seems like it was slightly more acute than you had thought at the beginning of the quarter.

Do you now see it stabilizing or how should we view that?

I would've said very slightly, it is a little different, but not much than at the end of the quarter.

And it continues to be a tough environment.

It is business, our competitors do have a decent product.

As Paul said, we are growing units by double-digit factors, we think we're holding onto the market and it's a tough fight.

Thank you very much.

Okay.

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

A recorded playback of this call will be available at approximately 5:00 p.m.

Pacific time tonight.

Those interested can dial 1-888-286-8010 and reference pass code 13800942.

Thank you.

Ladies and gentlemen we thank you for your participation in today's conference.

This does conclude your presentation and you may now disconnect.