超微半導體 (AMD) 2009 Q2 法說會逐字稿

Good afternoon.

I'll be your conference operator for today.

At this time, I would like to welcome everyone to AMD's second quarter 2009 earnings conference call.

(Operator Instructions) After the speakers' remarks, you'll be invited to participate in a question-and-answer session.

As a reminder, this conference is being recorded today.

I would now like to turn the conference over to Miss Ruth Cotter, Director of Investor Relations for AMD.

Please go ahead.

Thank you, and welcome to AMD's second quarter earnings conference call.

Our participants today are Dirk Meyer, our President and CEO; and Bob Rivet, our Chief Financial Officer, and Chief Operations and Administrative Officer.

This is a live call and will be replayed via Webcast on AMD.com.

There will also be a telephone replay.

The replay number is 888-266-2081.

Outside of the United States, the number is 703-925-2533.

The access code for both is 1373624.

The telephone replay will be available for the next 10 days starting later this evening.

Please note, we are providing non-GAAP financial measures for AMD Consolidated and AMD Product Company today.

Reconciliation to the most directly comparable GAAP financial measures are provided in the tables in the press release.

Management believes these financial measures make it easier for investors to compare current and historical period operating results.

Management also believes it's important for investors to have visibility into AMD's financial results, excluding the foundry segment and intersegment eliminations.

And to better understand AMD's financial results, absent the requirements to consolidate the financial results of GlobalFoundries.

Before we begin today's call, I'd like to caution everyone that we will be making forward-looking statements about management's expectations.

Investors are cautioned that those statements are based on current beliefs, assumptions, and expectations; speak only as of the current date; and involve risks and uncertainties that could cause actual results to differ materially from our current expectations.

The semiconductor industry is generally volatile and market conditions are particularly difficult to forecast, especially, in light of the current state of the economy.

We encourage you to look at our filings with the SEC, where we review the risk factors, setting forth information that could cause actual results to differ materially from our expectations.

You'll find detailed discussions about such risk factors in our most recent SEC filing, AMD's quarterly earnings report on Form 10-Q for the quarter ended March 28, 2009.

Before this afternoon's prepared remarks, for scheduling purposes, I'd like to call your attention to two items.

Firstly, our Q3 2009 earnings quiet time will begin at the close of business on Friday, September 11.

Secondly, our financial analyst day is scheduled for November 11.

The event will be held at our corporate headquarters in Sunnyvale, California.

We hope to see you there and we will be sending out additional details over coming weeks.

And with that, I'd like hand to hand the call over to Dirk Meyer, President and CEO.

Dirk?

Thank you, Ruth.

And thank you, everyone, for joining us today on the call.

In the second quarter, despite softness in the global economy, end-user PC demand stabilized but the characteristics of the economy are playing out quite differently across customers, channels and particularly, geographies.

Therefore, we think it makes sense to evaluate the situation over a slightly longer window than just a quarter.

So, in looking at the first half of 2009, I'm pleased with how we're managing what's in our control.

We've successfully transformed our Company, significantly reducing our cash requirements for capital, while cutting operating costs and enhancing our platform road maps.

Our second quarter top-line performance was steady, flat with respect to the prior period.

And as I mentioned on the last call, end customers continue to search for maximum value for their dollar.

A good opportunity for us, given the AMD value proposition.

Gross margin in the quarter was below target due to capacity underutilization, lower industrywide ASP's and our selldown of 65-nanometer inventory.

But with the strong flow of new products and a leaner cost structure, coupled with the assumption of modest seasonal growth, we are positioned for stronger financial performance in the second half of this year.

We continued our pattern of solid execution in the past quarter.

In June, we began shipping our new six-core AMD Opteron processor, months ahead of schedule.

Top tier customers including Dell, HP, IBM, Cray and Sun announced servers based on the great power, performance per watt and easy upgradability of the platform.

We will see our first full quarter of revenue impact here in the third quarter.

At Computex, we introduced the first 40-nanometer desktop graphics processor and the first 1-gigahertz graphics processor.

We also were the first to showcase working silicon and drivers for Microsoft's DX11 technology, featured in the upcoming Window 7 release.

DX11 fully showcases the superior AMD visual experience and we expect Win 7 to cause an unprecedented adoption rate for this important new technology.

We are steadily growing our footprint in the professional workstation market, as Dell, HP, and Lenovo expanded their ATI FirePro based offerings.

The unique combination of our low power AMD Neo CPU's and ATI rating on GPU's is attracting customers interested in building affordable, full-featured, high-def capable ultra-thin notebooks, and all-in-one desktop systems.

In addition, we refreshed our Dragon platform for desktops with the Phenon II X4 955 Black Edition processor.

A part that really showcases the stability of our high-end processors in extreme overclocking conditions.

We closed our first full quarter of distinct operation as AMD The Product Company and GlobalFoundries.

GlobalFoundries is really off to a good start.

Building a strong team and engaging with prospective customers.

Fab execution continues to be excellent and I'm pleased to report that our two companies are working together smoothly and effectively.

In that context, GlobalFoundries is poised to break ground, this week in upstate New York, on what will be the most technologically advanced manufacturing facility in the world.

The project will create over 6,400 new, high-quality jobs, all of them here in the United States.

Going forward, we are positioned well to see improvements in gross margin.

I am pleased with our ability to control spending, while improving the productivity of our go-to-market and product development activities.

And I am particularly happy with our engineering execution, where we continue to meet or beat all of our major milestones, including those for our next-generation programs.

Over you to, Bob.

Thanks, Dirk.

In a difficult operating environment with limited visibility, AMD's second quarter revenue and cash balance performance improved from the first quarter.

It appears that end-user demand has stabilized and we are pleased with our current inventory position.

And while industry dynamics in the first and second quarter were quite different, when we look at our 2009 first half performance, we think we held our own in both of the businesses.

Second quarter revenues was $1.184 billion and flat sequentially.

Revenue was down 13% from the same period a year ago.

Comparing the first half of 2009 to the same period a year ago, revenue was down 17%.

Second quarter net loss for AMD The Product Company was $158 million.

After backing out the benefit from the sale of inventory written down in the fourth quarter and charges mainly associated with Arc, the non-GAAP net loss was $244 million in the quarter.

The second quarter operating loss was $125 million.

Excluding the unusual items I just outlined, the non-GAAP operating loss was $205 million.

Adjusted EBITDA for the quarter was a positive $14 million.

We reduced our operating expenses by $29 million from the first quarter.

We continue to drive toward our break-even cost structure, based on the $1.3 billion revenue model.

We are now within $28 million of the $500 million operating expense target we set earlier this year.

Gross margin for AMD The Product Company was 36%.

This includes a 9 percentage point benefit from the sale of inventory written down in the fourth quarter of last year.

The non-GAAP gross margin was 27%.

The combination of across-the-board ASP decline and fab utilization continues to put pressure on gross margin.

In addition, we shipped a large volume of 65-nanometer parts in the quarter and sold down aging inventory.

Let me switch to the business segments.

Computing solution revenue was $910 million in the second quarter, down 3% compared to the first quarter.

We saw strong sequential growth in Asia, while sales in the Americas were flat and declined across Europe.

Microprocessor units were flat.

Overall ASP's were down quarter over quarter, mainly driven by notebooks.

Operating loss for the computing solution group was $72 million.

In the graphics segment, revenue for the quarter was $251 million, up 13% sequentially and flat year over year.

Much of the strength in the graphics business came from the Asia-Pacific and greater China regions.

Units were up, with significant growth coming from mobile.

ASP's declined across the board.

The graphics segment reported an operating loss of $12 million.

Now turning to AMD the Product Company balance sheet.

Our cash and marketable security balance at the end of the quarter was $1.6 billion, a slight increase over the first quarter due to reduced cash expense and a full-quarter impact of our Asset Smart strategy.

We used $15 million in the quarter to repurchase debt.

Depreciation and amortization was $103 million.

And capital expenditures were only $15 million for the quarter.

Now, let me turn to the outlook.

The following statements are forward looking and actual results could differ materially from current expectations.

In light of continuing uncertainty in the macroeconomic environment, AMD expects revenue to be up slightly.

It expects gross margin to improve from the current levels, driven by improved factory utilization; increased 45-nanometer CPU shipments and 40-nanometer GPU shipments; and improved product mix from the introduction of our Tigris notebook platform; our new performance-leading DX11-ready GPU family; as well as a full quarter impact of shipping the six-core AMD Opteron server processors.

We lowered our capital expenditures for 2009, as we continue to evaluate capital needs for the year.

CapEx for AMD The Product Company will be about $100 million, down from our prior plan of $150 million for the year.

And about $690 million for GlobalFoundries, down from the prior guidance of $760 million.

Warrants to purchase 35 million shares of AMD common stocks will become exercisable by West Coast High Tech, an affiliate of Mubadala Development Company, this Friday upon the public groundbreaking of GlobalFoundries Fab 2 in upstate New York.

For modeling purposes, the fully diluted share count for the third quarter, including the warrants is expected to be approximately 705 million shares.

In closing, we enter the second half with an increasingly competitive product line, strengthening financial performance and the hope of an improved economic environment.

One in which we know we can succeed with our new leaner business model.

At this point, I'll turn it back to Ruth for some Q&A.

Thanks, Bob.

Operator, we're happy if you could poll the audience for questions, please.

Thank you, Ms.

Cotter.

(Operator Instructions) Our first question comes from Uche Orji, please go ahead, from UBS.

Thank you very much.

Let me just start off by asking you about the ASP and the gross margins.

In the quarter that just ended, given that almost 10% of the revenues came from the sale of already written-down products, can you just help us quantify the ASP decline and how it will improve next quarter and similarly for gross margins?

Can you tell us x the impact of the inventory of written down products, just give us an idea.

You have a sense of the revenue.

You know all the drivers.

Give us a sense of how the gross margin should trend for the next quarter.

Yes, I'll give you the big movers and Bob can maybe chime in.

Sure.

Three big elements that give us an opportunity to improve margins next quarter are, one, capacity utilization.

Two, an improvement in unit cost, as a result of selling more 45-nanometer product.

And three, would be mix and the positive improvement that a richer mix would have on ASP's.

I actually think the two items I listed are going to be -- we plan to be the bigger movers of margin improvement.

And Bob, maybe you can provide some color on the just closed quarter.

Well, yes, I think Dirk summarized it quite well.

Clearly, those were the big impacts in the quarter.

Again, that inventory that we sold that reserved in the fourth quarter, again, we took a cautious position in the fourth quarter and reserved a lot of inventory that we'd built up at that point in time.

While the accounting says we actually have to show you and display how much we moved in the given quarter, that was still very good product in that period.

Our plan is we'll replenish that product.

We've built new inventory, so it's not a -- we don't have the product to sell in the go-forward quarters.

So let me just continue down, Bob, to some specifics of this.

So, the utilization rate goes from what level in Q2 to what level in Q3?

And just for me to know whether that implies 200 basis points, extra gross margin.

And then the mix on the 45-nanometer, it needs specifics as to what the differential is between 45-nanometer product margin and 65-nanometer project margin.

So that we can just quantify the delivery better.

Well, I'll kind of -- I'm not going answer that level of specificity of how much is associated.

What the move in utilization will move, was a fairly dramatic move, I'll call it from a factory utilization rates of more than 10 percentage points quarter on quarter of utilization improvement.

So, that will definitely be a pickup.

And then, if you think about 45-nanometer product, again, it's very product specific.

It depends on the additional functionality that was put into the dye at the next technology node.

But roughly, your dye cost is about half going from 65 to 45-nanometer.

The bulk of what we shipped in the second quarter was 65-nanometer version product versus on a go-forward basis, that will clearly drift to be dominated by 45-nanometer product.

Okay.

That's helpful.

And just one other question.

In terms of what you're seeing in the graphics market, you talked about ASP's being down.

If we look into next quarter, do you still expect ASP's to continue to decline in graphics?

And give us a sense of when they -- you think that might stabilize or how that will trend relative to the decline we saw in Q2?

And also, if you can by extension also talk about what the progress has been in the work station market?

When should we expect that to translate reasonably into some revenue for AMD?

Thank you.

Sure.

I'll do the last one first.

On the work station side, we've got cards, new SKU's, I'll say, showing up on the current generation of platforms from Dell, HP, and Lenovo, which just became available really in Q2.

So, we're at the very beginning stages of the ramp.

And as you know, it takes some period of time for us to penetrate substantially on enterprise markets, such as Workstation.

So I think you're seeing the beginning of what will become a multiquarter growth story.

So slow ramp but we intend to participate meaningfully in that business over time.

On the GPU side, more generally, we saw a lot of, I'll say, competition and price competition at the bottom end of the market.

I don't expect that dynamic to substantially change in Q3.

What I do expect to be an opportunity for AMD in the second half of the year is our introduction of DX11-compatible products and effective new product stack.

We expect to be out ahead of our competition.

And we think that will have a positive effect on our mix and hence, our ASP in the back half of the year.

Great, thank you very much.

Thank you, sir.

Our next question comes from David Wu from Global Crown.

Your question, please.

Yes, I hope to get an idea, Bob, when you answered the question about the mix, are you saying that the 45-nanometer would be the dominant part of the shipment mix in Q3?

And that the existing written-down inventory, 65-nanometer part has essentially been eliminated by the end of Q2?

No, we will still be shipping 65-nanometer in Q3.

I would expect, we will cross over to be dominated by 45-nanometer in Q4.

The crossover takes place in Q3 but a significant portion of the shipments in Q3 will be 45-nanometer.

Probably not 60%, but more in the 40% to 50% kind of zone.

So a significant piece versus it was immaterial in the second quarter.

I see.

So if you were to look at the dials -- of major dials to get the gross margin improvement, number one is factory utilization absorption.

And number two is essentially that mix you talked about.

Right?

Correct.

It's a balancing act between the two.

I don't want to be specific and put my hat on one topic.

It's really a lot of knobs we continue to try to turn to make progress to get back to the 40% target we need to run the business at.

The other thing I was wondering is if your utilization rate could be up as much as 10 percentage points between Q2 and Q3, are we talking about building inventory of the newer 45-nanometer part as you exit Q3 for Q4?

Yes.

Our view is -- as Dirk said, we'll adjust accordingly but right now, our belief is we're at a stabilization period.

We anticipate the second half, and this is what our OEMs are telling us, will be better than the first half.

And fourth quarter will be better than the third.

So, we need to start building some inventory in anticipation of the stronger fourth quarter.

We'll dial that up or down accordingly as we get through the quarter but that's -- your assessment is correct.

Third quarter is a period of time where you try to build some inventory to go into fourth and that would be a majority of 45-nanometer.

Okay.

Last question is this -- the handheld is $23 million now, it's a part that is not sold to Qualcomm, I guess it's going to stay with you.

Can you give us an idea how big that number will be for the rest of this year?

It will stay in that current zip code.

It's really on an end-of-life, supporting the existing customers that were designed in from that product.

We're not putting any new engineering to continue to elongate the product line.

But it will be in the $20 million per quarter zip code for awhile.

I see.

And the last one is, I assume that we are looking for third quarter $500 million OpEx target, right?

We're working our damndest to get close to it, yes.

Okay.

Thank you.

Thank you, sir.

Our next question comes from JoAnne Feeney from FTN Equity.

Your question, please.

Yes, thank you.

I was hoping we could talk a little bit more about the graphics side of the business.

So if I understood correctly, it sounds like the 40-nanometer GPU's will have both higher ASP's and higher margins than the processors you were selling in the second quarter.

Is that correct?

Yes, JoAnne, to be precise, we'll -- the new DX11 stack I referenced will be in 40-nanometer.

And clearly, that will give us an opportunity to kind of regain the high ground even more substantially and hence, improve our mix and ASP and hence, margins.

Okay.

Can you give us a sense of what you anticipate in terms of the size of the ramp?

We've been hearing from some other components suppliers to video card makers, such as the router, a steep ramp coming in the quarter, tripling their business.

Is that the kind of range you guys are looking at?

Is that a 40-nanometer question or DX11 question?

That's a -- well, those are the same for you guys, aren't they?

No.

We actually introduced a 40-nanometer DX10.1 part here this quarter.

I think I'm referring to just the DX11 parts.

Yes.

So, I expect that to be a very quick ramp in the desktop AIB channel, as we refer to it.

And expect DX11 to start showing up in notebooks more like the first half of next year.

Okay.

So the 40-nanometer 10.1 part then, is that what we're seeing in the third and fourth quarter this year?

Yes.

Okay.

And would those then come with those higher margins?

Through cost improvements, yes.

It's less of an upsell, as opposed to a unit cost improvement going from 55 to 40.

Okay.

And so perhaps you could just refine this a bit more.

What's the split of your adoption by the PC OEM's versus the retail add-in board vendors for these new 40-nanometer products?

And how is your -- how are your margins on one or the other?

How's your visibility on one or the other?

What kind of ramp are you looking at for the third and the fourth quarters there?

Yes.

Good question.

First of all, it's a lot easier for us to get an understanding of the traction to expect on the OEM side because we're awarded design wins in advance.

And therefore, can anticipate the business going forward.

There's a lag from when we see the design win realized and from when the machines show up in the market from six to nine months depending upon the form factor you're talking about.

Versus the AIB channel, what you referred to as retail, I'll just more generalize it and say AIB because, of course, some of that goes to system builders or white box guys in effect.

There the ramp, the adoption of early technology is very quick but we don't get orders well in advance of delivery date.

So it's harder to predict.

The one thing we're pretty confident on is the product looks great.

I think DX11 is going to be a good selling point.

So, we expect to enjoy good success in the back half of the year.

And it sounds like then, that runs into the first quarter of '10.

So, Bob, do you see a path here towards break even, given the ramp that you're anticipating on the graphics side, plus the new systems coming out in server and notebook?

For this year?

Yes.

We're still -- we have not given up on our goal to be break even, cash flow positive in the back half of the year, etc.

I'll be very disappointed if we don't exit the year at a 40% gross margin.

So, we're working down that path through combination of all the things we've kind of highlighted to kind of do that.

And so it's -- one last question if I could.

So it sounds like you guys have an atypical foundry supplier here on the CPU side and that you're going to benefit as utilization rates ramp up again.

So, they're going to pass those cost benefits on to you guys directly, it sounds like.

Yes.

JoAnne, the way I would think about it is, while we're operating as two companies today, clearly at the instant we became two companies, there's a set of manufacturing capacity in GlobalFoundries that we're accountable for.

And we pay for it.

So that's the -- that's what we look like today.

Out in time, that relationship will evolve to one that's much more like a typical foundry relationship in terms of how the economics work.

Does that help?

Yes, that helps.

So, it sounds like you have -- for the time being, you can benefit from the return of utilization over there and then, in the future, as they add new customers, you'll revert to a more traditional foundry pricing relationship.

Right.

Correct.

Okay.

And just one other point, JoAnne, just to kind of give you again, I know our information is relatively new from being transparent between GlobalFoundries and AMD.

To kind of give you a flavor to everyone on the phone call, the foundry revenue actually went down quarter on quarter.

And that's kind of reflecting, we didn't buy a lot of material.

We didn't run a lot of material.

That translates into underutilization.

We anticipate changing that trend and buying more wafers out of GlobalFoundries, which clearly improves the underutilization part of the equation.

But you can see that on the selected corporate data sheet that was in our earnings release.

Okay.

Thanks very much, guys.

Thank you, ma'am.

Our next question comes from Doug Freedman from Broadpoint.

Please go ahead.

Great, thanks for taking my questions.

If we could look at your commentary as far as the growth in the third quarter and focus a little bit on what is your normal seasonal pattern for 3Q?

If I look back over the last four years, we've had 9%, 18%, 26% and 31%.

Would you like to give us some sort of idea as to what normal seasonal is going to be off of this level?

Well, if -- first, I'll tell you what we think of as normal when we kind of try to characterize history.

On the CPU and GPU business both, we kind think the normal seasonal growth, Q2 to Q3, being low to mid-teens.

And clearly with the variation above and below that.

And maybe, Doug, what you're asking about is; Okay, given that context, explain the guidance a little bit.

Yes.

And there I would say that while we do, as I said in the opening remarks, expect a return to sequential growth, I think it's too early to say, given the overall weakness in the macro economy, that we're going to return to normal seasonal growth.

And therefore, the guidance reflects lower than that.

Can you talk about what factors you may have under your control to sort of drive that revenue?

Are there important ramps in the quarter that you're looking to take place?

And what are going to be some of the gives and takes on the revenue line next quarter?

Specific to AMD, sure.

One will be the full quarter availability of Istanbul.

And the opportunity that we have to, at the very least, maintain the momentum we have in four-socket servers, while at the same time, enjoying the benefit of some of the buildouts that we think are going to start to happen again in the back half of this year around big cloud installations.

Where Istanbul is -- the low power versions of Istanbul are particularly attractive.

That's number one.

Number two, we've got two new notebook platforms being introduced in the market here in Q3.

One is Tigris, which is the mainstream notebook platform.

The other would be our second generation of ultrathin, affordable platform.

Both of those will start showing up from our OEM's at the end of Q3 here.

So, that's an opportunity.

Finally, I think we've for the past couple of quarters relatively underserved the market in terms of quad core desktop mix.

And I think we've got an opportunity to do continually better there.

That is Phenon II quad core alongside our discrete graphics strength.

And then finally, of course, the DX11 launches that I referenced earlier.

All right.

Great.

If I could move on and focus a little bit on the GPU business or just in general, the Company's royalty stream.

Can you comment, Bob, what impact royalties had in the quarter and what impact they may have going forward, both from a revenue contribution and a margin contribution?

Yes.

So, let me -- I'll not do them exactly in order.

The contribution to the bottom line is close to 100%.

So, it's from prior activity and basically, you collect the check and it falls through.

It's not exactly 100% but it's pretty damn close from that perspective.

Second is, no surprise, it's very much tied, gaming royalties is tied to the patterns of people buying gaming boxes and systems.

Second quarter is not a big buying quarter.

So, revenue actually declined in the gaming business from second -- from first to second.

Traditional patterns would be gaming starts to accelerate in the third and clearly, in the fourth, as people get more excited about buying game systems for the holiday season back-to-school, etc.

So, we anticipate gaming royalty to improve from a low point of the second quarter.

All right.

Did the results in Q2 meet your expectations or was that an area of weakness in the quarter?

No, that was actually quite close.

That was not our challenge.

Our challenge was much more of just what was the mix of product.

We had to basically sell a lot more units to get the revenue to the top line.

Gaming wasn't an issue.

All right.

And then lastly, Dirk, if you could just go back and talk about if you've acquired any new customers in the quarter and how you feel like your customer engagements have been going?

Yes, Doug, good question.

And we can never be specific on OEM's but the one thing I will say is, over the past couple of quarters, I would say that we're starting to see our customer base, again, show renewed interest and engaging more deeply with AMD on the strength of a couple of things.

One is, I think clearly, our engineering execution across all of our product lines has been very solid over the past 12 months.

I think the capability that we've developed in graphics and the prospect of delivering really balanced, cost effective, increasingly power-efficient platforms is really starting to resonate with our customers.

And that's a positive thing.

And then ultimately, and I'll say this is a much longer term opportunity, I think clearly the progress that we've made with regulatory agencies around the world is going to help us over the coming quarters and years.

So, I'd say increasing momentum.

My last question, same sort of concept on the new customers.

I know it's a requirement for GlobalFoundries, for the deconsolidation of those numbers, for them to acquire new customers.

Can you advise how their progress is and if there's any sort of a forecast or a timeframe in which we might see that deconsolidation?

Yes.

I can't be specific on the deconsolidation issue but I can give you some color commentary.

I would say first, that the first few month of GlobalFoundries life has seen them kind of reaffirm, in their mind and our mind, the value potential that they have as a new independent foundry.

And I expect they'll be announcing their first new customer, beyond AMD, within roughly the next 30 days.

Great.

Thank you.

Thank you.

I'm showing our next question comes from Jim Covello from Goldman Sachs.

Please go ahead.

Great, thank you very much for taking my questions.

Just two questions, both around GlobalFoundries.

One, kind of picking up on the last question about new customer activity at GlobalFoundries.

What kind of role is AMD playing, if any, in helping GlobalFoundries secure new customers or is it really just the designs that you're helping them with?

Yes, this is Bob.

Pretty much, we stay out of that part of the process.

We do anything pretty much they ask us from a -- I'll call it if they need to help, they need to understand how do we do it.

But pretty much it's an independent sales team they're trying to create and a customer engagement-type process to go acquire third-party customers.

So it's not necessarily a situation where you might lend them an engineer to solve a customer problem that they might be presented with or anything along those lines?

No.

Okay.

And relative to the capital call situation with GlobalFoundries, I would suspect that they would have -- there would be a capital call around that entity this year.

And the assumption is still that you guys would continue not -- to contribute $0 to that and let the ownership stake be diluted.

Is that right?

That's correct.

Through this first half of the year, there was no capital calls.

There will be a capital call in the back half of the year.

And yes, that is our current view from an AMD perspective is we will not participate in that capital call.

Okay.

Can you say what that capital call might be, just so we can get a send of the ownership stake pro forma for that?

No, not that specific but I will mention one thing is, as we've launched that company, we targeted that they should maintain around $1 billion of cash.

You can kind of see in the current quarter results, they're slightly below that.

So that kind of gives you at least a framework.

Terrific.

Thank you so much.

Thank you.

Thank you.

Thank you, Jim.

Our next question comes from Stacy Rasgon from Sanford Bernstein.

Please go ahead.

Hi, guys.

Thanks for taking my question.

I had a question on the gross margins paid by AMD to GlobalFoundries.

It looks like they went up fairly significantly in the quarter.

At the same time that the cost base was also going up due to underutilization.

I was wondering if you could give us some clarity on what that gross margin profile that you will paying to GlobalFoundries looks like going forward?

Will that continue to be increasing?

Again, we're not going to give that level of color.

It's based on the contractual arrangement we have with GlobalFoundries.

But I'm not even talking about just the dollar amount.

I understand that the dollar amount went up in some sense, probably because of the underutilization and the higher cost base.

Just the percentage basis.

What is that based on?

Can you give me some feeling for just how to model this in general going forward?

As we've discussed before and is publicly disclosed, it's a cost-plus arrangement.

Okay.

So, it's the cost of producing wafers and we would do that plus some more.

So why did the margin go up in the quarter versus last quarter?

To be honest, I'm not sure I understand how you have come to that conclusion that margins went up.

No, gross margins that you paid, as far as I can tell, went up in the quarter.

Do you have any -- can you give me any sense of why?

Was it just based on the agreement or was there something else in terms of mix or anything else?

No comment.

I'm not sure I understand your question.

Okay.

Also, regarding the foundry company OpEx, I note that both the R&D and the SG&A came down a fair amount this quarter versus last quarter.

Am I right in assuming that both of those ought to be increasing going forward?

Number one, as the process development ramps up.

Number two, as third-party business comes on.

And number three, as you're just ramping up a sales force to start attracting some of that third-party business?

I'll call it -- yes.

In general, think of it in two directions.

One, they need to create an independent company with all the appropriate functions, sales force, libraries, increased portfolio of technologies that service the foundry industry.

But they also have to deal with the economic reality, which is the world is pretty soft.

Our demand is down and so they've got to figure out the right timeline to go build out all those capabilities.

So, does that imply that the ramp of capacity for third party that you guys showed at your analyst day in November is going to get pushed out?

Sure, yes, which is reflective in the CapEx reduction, that we just talked about, from $750 million to sub $700 million.

Got it.

And so, will that correspondingly push out the date where deconsolidation activity can probably happen?

That's a to-be-determined.

Can't figure it.

There's not enough information to make the determination that's based upon having meaningful third party revenue.

Okay.

Got it.

One more question around the deconsolidation.

I understand under -- It's my understanding, under the current deal licensing agreement with Intel, the foundry needs to qualify as a subsidiary.

And I know there's disagreement right now on whether or not it qualifies or not.

But I'm just wondering, assuming that it eventually does get deconsolidated, under the current terms of the agreement, would the venture still have to qualify as a subsidiary?

And in that case, would the shareholders still be exposed to any profits or losses from that entity through minority interest line, if not directly consolidated into your financials?

Well, high level, sure, we would still account for our interest in GlobalFoundries through a minority interest line.

What I'm asking is --?

A noncash line item.

Yes, what I'm asking is, would it still be your intent to maintain a significant ownership stake in the venture even after it's deconsolidated?

Maybe is another way to ask the question.

He's asking -- (technical difficulty).

To your question, sorry, we had a little phone issue.

No problem.

To me, that's going to be a to-be-determined decision of what's our ownership stake on a long-term basis in GlobalFoundries.

As we've discussed, in a near-term basis, is we will continue to most likely not participate in the capital call.

It's structured that it's at our option.

But what's that going to be long term, say five years from now?

Too early to make that determination.

Just like any investment, whether you stay in it or double down, stay in it or get out.

Yes.

So, can you give us any idea of any specific milestones we ought to be looking for for deconsolidation?

Is it that Class A equity ownership needs to drop below 50%, so you're no longer exposed to more than 50% of the losses?

Or is it a certain amount of third-party revenue that has to ramp?

What are the types of things that we ought to be looking for going forward to have comfort that an activity like that might be more imminent?

To me, it's a real simple one and it just goes to the high level of don't get lost in the weeds.

It's really that GlobalFoundries is becoming a successful company in the foundry business and they get third-party revenue, non-AMD.

That's really -- and that at some point in time when that gets big enough, the mechanics will allow us to deconsoli -- to not consolidate.

But that's the singular thing to really pay attention to, is to see how many third-party customers come to the fold and then, how those people turn into a revenue stream.

Got it.

But again, so I'm clear, even after deconsolidation, likely still some exposure through minority interest, even if it isn't on cash, it's still going to be an accounting hit, correct?

Sure.

If we still have an ownership, we would have to account for our percentage of that ownership.

Yes.

Okay.

Great.

Thank, guys.

I appreciate it.

Like we did in Spansion.

So, it's no different than the Spansion-type investment.

Got it.

Okay.

Thank you very much.

Thank you.

Thank you, sir.

Our next question comes from John Pitzer from Credit Suisse.

Please go ahead.

Yes, thanks, good afternoon.

Bob, just a clarification.

Did you say that utilization rates went down in the June quarter?

They went up a little bit from first but not too much.

They were relatively flat quarter on quarter.

We actually didn't turn wafer starts on, I'll call it, any material amount of difference until June.

Can you just help me understand, was there anything specific in your business?

Because for most semi companies that own fabs, they cut really hard in the March quarter, which made June fairly easy comparing.

Can you just help me understand what was going on with your business and why you didn't feel it necessary to cut more aggressively in March?

So, our timing was maybe potentially different.

We had inventory that we thought made more sense to burn off, that we built in the second half of last year than, I'll call it, building more inventory.

And then, any way to quantify how much more written off inventory will run through the top line, as we move into the September quarter?

There's approximately only about $20 million left to go.

So, there's very little left of the reserve we took in the fourth quarter, which to remind everyone was about $225 million.

So we've either scrapped or have turned around and sold that material.

So, there's only a little bit left.

And then, my last set of questions, just around average selling price within the CPU business.

Was the decline in the June quarter more a mix of what was going on in the business, or was there like-for-like selling pressure?

And if you guys are talking about mix being a positive factor going into the third quarter for gross margins, is the implication there that we should actually see ASP's flatten to go up sequential?

Certainly, that's what we're trying to drive the business to.

If you look at Q1 to Q2, as Bob said in his opening remarks, we saw decreasing ASP's kind of across the product line but that was particularly acute in notebook.

Great.

Thanks, guys.

Thank you, sir.

Our next question comes from Patrick Wang from Wedbush.

Your question, please.

Great, thanks so much for the question.

Dirk or Bob, could you talk about if there's anything in particular that happened to drive the greater mix of 65-nanometer parts than you had expected last quarter?

Well, the mix of what we sold 65 versus 45 wasn't so much different than we thought.

As Bob said, we entered Q2 knowing our inventory position and therefore, were motivated to sell what we had on the shelf.

The thing that's always a little bit harder to forecast in advance is exactly how our sales will line up across the various price points in the market.

And where the opportunities will be in the market, as they show themselves.

I see.

So was there a particular focus in -- for Asia sales for the 65-nanometer parts or was it pretty broad based geographically?

Broad based.

Okay.

Got you.

And then, as we look to see the mix away from -- the product mix shift away from 65-nanometers, that's going to expect to benefit gross margins.

When do you think those tailwinds go away?

When did 65-nanometer become not meaningful enough to really start moving margins?

You mean, by what point will we have enjoyed whatever improvement we'll get by going from 65 to 45?

Right.

Sometime in Q4.

Okay.

Got you.

Third quarter, I'd say is transition.

We still have -- most of what we've been building is 45-nanometer.

We still have 65-nanometer in inventory.

Clearly, we'd like to move 65-nanometer first because it has a less long shelf life, if you know what I mean.

So it's in that kind of a thought process.

Okay.

That's helpful.

And then, Bob, how should we think about OpEx going forward?

Say you arrive at the $500 million target in the next quarter or two.

Should we expect that to walk -- back up slowly with the resumption of revenues or how do we think about that?

Our kind of thinking kind of goes like this is, we're going to get to $500 million and kind of stay at that level.

Except for, I'll call it, the program spend in marketing, which is associated with higher revenue levels or lower revenue levels, depending on what happens in any given quarter.

As Nigel talked about at the analyst day, we're trying to make that very variable.

So that, if revenues go up $100 million, you would see increase in marketing spend.

If revenues went down $100 million, you would see a decrease in marketing spend.

But the baseline of $100 million at $1.3 billion, is roughly the same, I'll call it, for the foreseeable future until we get confidence that the world has turned around and before we take the fixed cost structure up.

Got you.

Okay.

And then, either one of you could try to help me answer this one.

But can you give us a send of how you view the opportunity with Istanbul now that we're going to have a full quarter in the third quarter, we've seen all the kind of performance benefits, power benefits.

How do we kind of monitor the success of that and how do you think about that?

Yes, I'll repeat what I said earlier but with a different tone.

First of all, we've clearly got an opportunity with Istanbul to provide a nice incremental benefit to the four-socket platforms that our customers have out there on the market, where we've got a very strong position today.

And then, across the two-socket space, which is big, the bulk of the market and also fragmented, one of the places where that product, that Istanbul product, really shines is in the lower power bands of the product line.

That is our so-called HE and EE parts where those things are particularly well suited.

And we're particularly targeting at cloud installations, cloud buildouts, which we think is going to start to pick up again in the back half of the year and be one of the few, at least I hope, potential bright spots in the general enterprise space.

Okay.

Got you.

That's helpful.

And then last one.

If you could talk about bigger picture, if there's any good news that you could talk about in terms of in demand and maybe how you're feeling about back-to-school and Windows 7?

Thank you.

Yes, Window 7, first.

Awfully hard to predict.

Obviously, a good product, well executed.

I'm enthusiastic, particularly about the DX11 technology that's being dragged along at the same time.

So, I'm hopeful that that's going to motivate people to buy a little bit richer mix of graphics technology in their systems.

I think one of the wild cards is the degree to which the consumer channels stall a little bit waiting for Win 7 machines to show up on the shelf.

We're not modeling that as a big factor.

A lot of those machines are going to ship in advance of Win 7 availability with upgrade coupons.

So, I'm hopeful that that won't stall the channel.

We're not modeling for our own purposes Win 7 being a huge contributor to overall unit demand.

But as I said earlier, I hope at the very least there's some ability for us in the industry really to upsell the more richly configured machines, particularly around the GPU.

Outside of that, in terms of general commentary, I can't really do anything but reflect again on what I said earlier in response to a question.

And that is, I do think we're returning to a pattern of sequential growth here.

But we're expecting that to be relatively more muted than a typical seasonality again because the economy is generally pretty weak around the world.

Huge geographic variability, of course.

APAC and China in particular on the strong end.

I'll say North America, nominal and Europe on the weak end.

Great.

Thanks so much and good luck.

Thank you.

Thank you, sir.

Our next question comes from Hans Mosesmann from Raymond James.

Your question.

Thanks for taking my questions.

I'll be brief.

A question regarding 40-nanometer at TSMC, there's been rumblings that they've had a tough time with that node and perhaps you guys have as well.

To the degree that you had some challenges there, have they been resolved and where are we in that process?

Yes, we're not expecting any issues with the TSMC 40-nanometer technology to, I'll say, block our opportunities in market as we walk into the back half of the year.

So there's no issue going forward regarding the ramp?

Yes, going forward, we don't see any issues.

Okay.

And a more strategic question.

A lot of investors are very keen on trying to understand the netbook dynamic as it merges with the smartphones and there's operating systems wars that are going on.

And everybody has some kind of solution.

And you guys are oddly out of that discussion, at least at the moment.

And it strikes me that you have a lot of technology that is applicable.

What is the strategic direction, if any, that you have in that space as the netbook and Smartphone world converge?

Yes, good question.

First, we're aggressively working on more power efficient and lower cost or components-built, purpose-built for lower price points in the marketplace, like those associated with netbooks.

I continue and we continue to see kind of a gap between a netbook on the one hand and Smartphones on the other.

We're clearly targeting devices with reasonably useful keyboards and 10-inch kind of screen sizes.

And the continuum of PC-like devices above that, independent of our operating system, happens to be loaded on those machines.

The Smartphone space, a little bit of a different story.

As you know, we've sold our handset division, or at least some of the technology assets, a couple of quarters ago.

And while our technology is applicable to some of those form factors, that's clearly not a market focus for us at this time.

We're clearly focused on, I'll say, a 10-inch screen size and above.

Fair enough.

Thank you.

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

Yes, ma'am.

Our next question comes from Tim Luke from Barclays Capital.

Please go ahead.

Thanks so much.

Bob, just to clarify, you were saying that you are targeting a gross margin for the fourth quarter of 40% or you think you can exit the quarter with a -- at a run rate of 40% or something.

Could you just clarify that?

And the same thing with the operating profitability.

Yes.

What I said is I'll be very disappointed if we don't achieve 40% at the end of the year.

So that means, for the quarter, you'll get to 40% or what does that mean?

I'd love to say that's for the whole quarter.

That's really the goal.

That's what we're still driving toward.

And the operating profit goal?

Well, break even at $1.3 billion, you can do the math on that.

Okay.

Just off the -- in the quarter, were the servers down more than the broader 3% for the computing area sequentially?

And you spoke about ASP pressure in notebooks, any further color there?

Well, as -- let me circle back and just answer it this way.

We saw unit decline primarily in the notebook space.

So, I'll call it -- we did pretty well in total units with server or desktop.

Mobile being the weakest of the three of those pieces.

We actually made progress in the server space.

ASP's were definitely a dynamic situation in the quarter of, I'll call it, lower price points for a lot of the different areas.

So that's kind of the dynamic of the quarter.

Again, we didn't have much -- the six core Istanbul product in the quarter, that this came out at the tail end.

We will get a full quarter of that starting now in the third to change that dynamic.

When you think about up slightly in sort of your phrases, that's sub 5%, is that what we should think or up to that level or just if you have a framework for that?

Yes.

I kind of think of low to mid-single digit kind of growth would be up slightly, is my definition.

So that's my D code.

Very helpful.

And two last brief things.

Dirk, you were alluding to a new customer in 30 days for GlobalFoundries.

That would be presumably in the logic arena or any color there?

And lastly, Dirk, if you'd just frame your legal issues and ongoing discussion with Intel, one of the focuses is the big trial in March in Delaware.

Do you have any other thoughts for us on that and how much maybe the annual run rate of spending might be on legal currently and how we should think about that beyond the trial?

Thanks.

Yes.

Tim, good question.

I don't think we've provided any specificity on the legal bills.

And I don't want to start on that front.

You're right on the trial date.

And your first question, could you repeat it again?

Just you said you're going to have your first non-AMD customer for GlobalFoundries.

Yes, that would be in a logic space.

That's the focus for GlobalFoundries.

Thank you guys.

A bulk technology obviously, too.

Right.

Thank you very much.

Thanks, Tim.

Thanks, Tim.

Thank you, Tim.

And our final question for today comes from Ross Seymour from Deutsche Bank.

Your question, please.

Thanks for sneaking me in, guys.

I'll make it really quick.

The gross margin guidance, where you said it should be up sequentially is, that off of the base that is inclusive of the roughly 8 point benefit from writing off inventory or exclusive of that?

Upward 30's gross margin base or kind of upper 20's?

I don't want to give that level of specificity but I'd be disappointed if there wasn't a '3' in the first position.

Okay.

And then, a question transitioning quickly over to the gross -- or to the GPU side of things.

The operating profit dropped again to a loss there.

Can you just walk through some of the dynamics that could allow that to get better?

I know you've talked about DirectX11, those sorts of things.

Is there anything other than kind of mixing up that can help, such as cost cuts, anything along those lines?

That's a question on operating margin, right?

Yes, more operating than gross.

Well, two things, volume and mix.

So, going into the back half of the year, we ought to be able to move more quantities in total.

More quantity in total and also improve the mix on the strength of new product introductions.

Apples to apples, do you still expect pricing pressure?

It seems like it's pretty competitive between you and NVIDIA and that doesn't seem to be lessening.

Yes, I don't expect a substantial change.

Where we compete, I'll say line on line, I think the new dynamic in the back half of the year is we're going to be able to offer things that they won't be able to provide an answer to, which will be helpful.

Okay.

Thank you for sneaking me in.

Thank you.

Thank you.

We'd like to thank everybody for participating in today's call.

And that concludes AMD's second quarter earnings conference call.

Thank you.

Ladies and gentlemen, this does conclude today's program.

Thank you for your questions and for your participation.

You may now all disconnect.