超微半導體 (AMD) 2008 Q4 法說會逐字稿

Good afternoon.

I will be your conference operator for today.

At this time, I would like to welcome everyone to AMD's fourth quarter 2008 earnings conference call.

All lines have been placed in a listen-only mode.

After the speakers' remarks, you will 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 Ms.

Ruth Cotter, Director of Investor Relations for AMD.

Welcome to everyone to AMD's fourth quarter 2008 earnings conference call.

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

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

There will also be a telephone replay.

The number is 888-266-2081.

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

The access code for both is 1313309.

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

I would like to take this opportunity to call to your attention our first quarter 2009 earnings quiet time which will begin at the close of business on Friday, March 13.

Now, before we begin today's call, I would 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 turbulent economic conditions under which the entire world economy is presently suffering.

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

You will find these detailed discussions and our 10-Q filings for the quarter ended September 27, 2008.

Now, with that, I'd like to turn the call over to our President and CEO, Dirk Meyer.

Thank you, Ruth.

And thanks for joining us everyone.

The fourth quarter of 2008 is going to be remembered for the severe stresses placed on the global economy and on our industry.

The global economic environment led to a softening in end user demand for PCs and servers in what is usually the year's strongest quarter.

In addition, we saw the beginning of a severe inventory correction across the IP supply chain, particularly acute in notebooks and one that is continuing into this quarter.

The effect on the economy is going to continue to dampen end user demand.

The degree to which is uncertain.

This combination makes the future particularly murky.

With that backdrop we have established a clear set of priorities.

We want to preserve and generate cash.

We want to protect our core investments.

We want to manage our P&L by reducing our breakeven point in line with the new economic environment and do all of these things while continuing to serve our customers.

Consistent with these priorities, we are focusing on completing our Asset Smart transaction in February, and with the exception of the shareholder vote scheduled for February 10, all major closing conditions have been met.

In particular, we have received clearance from the committee on foreign investment for the US, or (inaudible), the transfer of development incentives from the Empire State Development Board and New York Authority's Control Board has been approved.

We are documenting the agreement with the holders of the Dresden debt to transfer the obligation from AMD to the Foundry Company.

And the SEC has cleared the proxy for distribution to shareholders.

In addition, we completed the sale of our DTV business to broadcom for $142 million in cash, and as detailed in Tuesday's announcement, we sold certain hand held division assets to Qualcomm for approximately $65 million in cash.

We have accelerated our program to reduce our cost structure.

We will take the AMD product Company headcount down by another 9%, primarily in manufacturing, SG&A, and through the hand held division transaction.

We have put in place a temporary comprehensive salary reduction program and we are reducing our manufacturing output to keep inventories down while completing our transition to 45-nanometer technology and developing 32-nanometer technology consistent with our plans.

These actions will allow us to protect our core investments, specifically, our product road map and the great people who innovate and bring them to life.

For our customers, we expanded our performance graphics product line with the ATI mobility Radeon 4000 for notebooks and the Radeon HD4830 which delivers exceptional PC game performance for the mainstream markets.

We shipped our 45-nanometer Quad-Core AMD Opteron or Shanghai processors, ahead of schedule.

And systems based on this new family are winning performance best benchmarks across the most relevant server workloads, offering superior virtualization, energy efficiency, and platform stability.

We announced the availability of our new Yukon platform for ultrathin notebooks and the first customer offering based on Yukon, the HP pavilion DV2 was named best notebook of CES by Laptop magazine.

We launched the Dragon platform for desktop PCs featuring the new AMD Phenom 2 X4 processor.

HP, Dell and others plan to offer systems based on the Dragon platform in the current quarter.

Measured against the elements of our plan that are within our control we're performing well and our unique graphics, video and platform value propositions are helping our customers deliver more differentiated offerings.

Crucial in this kind of market environment.

The reality of today's global economy requires that we redouble our focus on cash management and an expense control and react swiftly in response to our customers' changing needs.

We will do so while protecting the assets and road map that make us well-positioned for success and with that I'll turn it over to Bob.

Thank you, Dirk.

Good afternoon, everyone.

I'd like to remind everyone that for the purposes of my remarks this afternoon, I will refer to continuing operations unless stated otherwise.

In addition, a reconciliation of GAAP to non-GAAP financial results is available on the press release.

Fourth quarter revenues were $1.162 billion, down 28% sequentially, excluding third quarter process technology revenue, license revenue.

Fourth quarter net loss was $1.424 billion, or $2.34 a share.

Losses from continuing operations were $1.414 billion, or $2.32 a share.

This includes the negative impact of $996 million in net charges, of which $75 million were cash related.

Items to note include $714 million of charges related to impairment of goodwill and acquired intangible assets and amortization of acquired intangible assets.

An incremental write-down of inventory of $227 million.

Restructuring charges of $50 million.

Foundry Company formation costs of $23 million.

An impairment of our expansion investment of $20 million a gain of $30 million on buyback of our convertible debt.

Fourth quarter operating loss was $1.274 billion, but on a non-GAAP basis was $260 million, compared with $24 million last quarter.

Gross margin in the quarter was 23%.

This includes a 20 percentage point negative impact associated with the incremental inventory write-down of $227 million.

We took this reserve due to the weak midterm outlook.

Let's switch to the businesses.

Computing solution revenue was $873 million in the fourth quarter, down 27% compared to the third quarter, excluding the process technology revenue -- license revenue.

Microprocessor units were down across the board.

Overall ASPs were slightly down, driven by a drop in client ASPs.

However, server ASPs were up significantly with the introduction ramp of our 45-nanometer Quad-Core Opteron processors.

Operating loss for the computing solution group was $431 million, significantly affected by the inventory write-down associated with this business.

In the graphics segment, revenue for the quarter was $270 million, down 30% sequentially.

ASPs were up as a result of our stronger product line, but demand in this segment was down significantly.

Operating loss for the graphics segment was $10 million.

I'm going to turn to the balance sheet.

Our cash and marketable securities balance at the end of the quarter was $1.1 billion, down $245 million from the prior quarter.

Capital expenditures were $112 million for the quarter, and $621 million for the year.

As economic conditions deteriorated through the quarter, we adjusted spending, coming in well below CapEx guidance of $800 million for 2008.

Let me turn to the outlook.

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

In light of current macroeconomic conditions, very limited visibility and continued inventory corrections in the supply chain, AMD expects first quarter 2009 revenue to decrease from the fourth quarter of 2008.

At our November analyst day, we told you that we were driving toward a $1.5 billion quarterly breakeven point for AMD, the product Company.

Now, given the environment, we are reducing our breakeven target to approximately $1.3 billion, based on the programs Dirk mentioned earlier.

We are heading down this path this current quarter and our goal is to be there in the second quarter.

From a modeling perspective, we are maintaining our gross margin goal of 40% or more, so no change from what we told you before.

R&D is going to be about $300 million a quarter, down from the $325 million we outlined in November.

This reduction does not compromise our road map.

SG&A is expected to run around $200 million, down from the $240 million previously guided.

Depreciation and amortization will be approximately $100 million per quarter.

Capital expenditures have been reduced to about $45 million per quarter, roughly half of what we discussed at analyst day.

Now, that's the foundation of AMD, the product Company.

Based on the expected February close of the Asset Smart transaction.

Based on lowering the breakeven point, we will record a restructuring charge in the first quarter.

And we will communicate the magnitude later in the quarter.

In this environment, we are laser-focused on the priorities Dirk mentioned at the onset, preserving cash, preserving and generating cash, while protecting our core investments, and managing our P&L and of course, while we continue to serve our customers' needs.

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

Thanks, Bob.

Operator, we'd like you to poll our audience for questions now, please.

Thank you, Ms.

Cotter.

(Operator Instructions).

Our first question is from Tim Luke from Barclays Capital.

Your question, please.

Thanks so much.

Couple of quick things.

Bob, just in framing the environment, the seasonally last year in the first quarter you were down around 15%.

While you're just saying that you expect the revenue to be lower, given that it was down 33 this quarter, do you feel that's likely to be more or less than the seasonality that you may have seen last year?

And just in terms of the way you're likely to report going forward, if you close the transaction in February, how should we think about you reporting the first calendar quarter of '09?

Will you provide fully consolidated numbers or the separate elements?

And just also on the OpEx, it looked like the R&D and SG&A actually went up a little bit sequentially in the calendar fourth quarter.

Do you have a feel for how we should be modeling that for the combined entity at the beginning of the year?

Sure, Tim.

Since you had so many questions in there, if I miss one, circle back.

But first, your first question was on the forecast of us just forecasting down.

We're not going to give any further guidance than that.

As Dirk mentioned and I mentioned, the current environment is pretty murky.

Visibility is pretty low.

Seasonality didn't -- clearly didn't hold in fourth quarter so who knows for first quarter.

Just kind of leave it at that at this point in time.

Reporting was your next question.

We will report AMD the product Company, Foundry Co, The Foundry Company and AMD consolidated results starting in the first quarter.

So we will give you the granularity of those three pieces starting with the next reporting cycle.

And then I think your last question was on operating expense, kind of guidance.

The best way to think about it its, again, AMD the product Company is the $1.3 billion breakeven target.

We will get there by the second quarter.

We will make significant progress in the first quarter, from that perspective.

So we won't get there fully, just based on the time lines of executing to people reductions and salary cuts and other activities we're doing.

But we'll make significant progress in the first quarter.

If I may, just for the consolidated Company, the gross margin this time, appears to be about 43%.

What are some of the variables that we should be thinking about and how do you think we should frame expectations for the first quarter?

Well, I'll kind of walk through pluses and minuses.

Obviously, the minus is volume, what is the volume forecast change quarter-to-quarter, so that's obviously I think it's a negative because we're forecasting it down.

Offsetting that is our lineup.

Our lineup is significantly better than it was in the fourth quarter, particularly in the microprocessor front.

So, we've got products moving in the right direction.

Most of what we'll be shipping will be in the 45-nanometer variety.

Therefore, enjoy the cost reduction of that part.

So there's pluses and minuses.

Factory utilization will be crummy, just based upon the kind of -- the demand environment we're currently staring at which is why of course we and everyone else are taking actions.

Does that help a little bit?

Thanks very much, guys.

Our next question is from Uche Orji from UBS.

Your question, please.

Thank you very much.

Two questions.

First is on channel inventory.

Just I understand the challenge with the visibility in the market, but do you have any sense as to where we are now in terms of the inventory destocking in the channel, and any sense at all when you think your production might be coming in in line with consumption?

Yes.

The question varies depending upon the product line.

First of all, if you look at our desktop products through the component distribution channel, there we saw the channel react pretty quickly to the economic environment early in the fourth quarter.

And we actually left Q4 with inventories pretty appropriate to sell-through in that channel.

So that channel responded pretty quickly.

The other extreme of course would be notebooks where our OEM customers often have these machines built in Asia and ship these machines on boats so the supply chain is long.

There, the channel takes quite a bit longer to respond and bring down inventory and we see that process still continuing into this quarter.

It's awful tough to predict when overall our customers' inventory levels are going to be appropriate, given where demand is, but that's a process that I don't expect to be fully complete in Q1.

And any comments on graphics?

If I look at your graphics number, how much of your results relative to Nvidia was a result of share gains, obviously if you can make any comments as to what you see in terms of channel inventory for graphics similar to what you said for desktops and notebooks, that would be helpful?

Sure.

Well, first, it's hard to know based on third party reports where we ended up in terms of GPU share in Q4, though I expect, given the extent to which we have a differentiated offering, both in terms of performance, power, efficiency and features, we did gain share in Q4.

The same commentary I would say applies to GPUs in terms of channel inventory of CPUs, namely our add in board channel can react a bit faster versus certainly the part of our GPU business that's associated with notebooks.

So in terms of color commentary, very similar to what I described on the CPU side.

Great.

Thank you.

Our next question is from John Dryden from Charter Equity.

Your question, please.

Hi, thanks for taking my questions.

Bob, could you give operating cash flow for the quarter and an update on the free cash flow outlook that you provided at analyst day for positive for 2009?

And secondly, with the shareholder meeting on February 10, once approved, what's the time line for the $700 million input from ATIC and $100 million plus input coming from Mubadala?

Okay.

First, your first question was on the current cash flow.

Think of it this way.

Obviously, cash balance is down 245.

We paid off about $95 million worth of debt.

We invested about $112 million in new capital.

Therefore, cash flow from operations was slightly negative in that equation.

So, we did burn cash.

That's definitely part of the reason we're taking a very aggressive actions in the first quarter.

If you think about it, with the forecast on a go-forward basis, again, with the breakeven model of $1.3 billion, depreciation of $100 million a quarter, CapEx of about $45 million a quarter, I think you can do the math and determine that that's actually making a little bit of free cash flow.

That's our model.

First quarter will be a little on the negative side because we won't get actually to the breakeven model but that's kind of the plan as we kind of move through time.

And then, again, the timing co-timing, shareholder meeting is scheduled for February 10.

We're expecting we'll close this transaction within 24 to 48 hours after the shareholder vote is affirmed as a yes and that will bring $700 million into AMD the product Company, $1.4 billion into Foundry Co.

and another 100 plus on the equity sale of the AMD shares to Mubadala.

That will all happen, again, I'll say 24 to 48 hours after we get shareholder approval.

Thanks for taking my questions.

Our next question is from David Wu from Global Crown Capital.

Your question, please.

Yes.

Can you give a little bit more color on the -- your Puma platform has been doing pretty well in terms of design wins and the ramps.

And I was wondering whether from your comments that it looks like notebooks is going to be weaker than desktop in Q1 and do you expect overall the rate of decline in your computing side to be slowing down in Q1 and in view of the fact that your desktop inventory seems to be clean?

That's the first question.

The second one I was wondering is this one wonderful letter that your friend across the street sent you yesterday about needing to set up a meeting for discussion on this cross-licensing thing, is there anything that would lead you to be a show stopper in terms of this transaction in a meeting of that sort?

And the last question I have is really that of graphics.

The graphics and the CPU, both MPU side have both undergo inventory correction.

Should I infer from the comments that you gave that in fact the graphics is further along in the inventory correction by OEMs and channel partners than the MPU side?

Good questions.

First, on your question regarding the Puma platform and how to think about the two businesses, notebooks versus desktop.

First of all, notebook is still going to enjoy a higher growth rate in terms of end user consumption throughout next year, versus desktop, consistent with what we've seen in the past couple of years.

It's extremely hard to know what those growth rates are going to be or in fact whether they're going to be positive or negative over the course of the year.

The notebook supply chain, as I said, is much longer than desktop.

Therefore, the inventory correction takes longer and, therefore, our notebook sales are going to take longer to come into balance with where end user system consumption is.

Okay.

And it's awfully hard at this point to put numbers on anything that I said there, which is why our guidance is what it is.

Second, you asked about the letter we received from Intel.

As we said when we announced the Foundry Company transaction back in October, we constructed that transaction consistent with the terms and conditions of all of our IP license agreements.

That was true then and it's true now.

So we have complete confidence that this agreement is consistent with all of our IP terms and conditions.

And Intel's letter is in no way a condition for closing the deal and the deal is going to close on February 10.

Finally,-- assuming we get the shareholder approval as we expect to.

Finally, you asked about commentary about comparing GPU to MPU inventory correction and whether GPU is ahead of the other.

I wouldn't say so.

GPUs are down on notebook motherboards and have the same channel correction delay, I'll say, as CPUs do.

So first order, I can't characterize the situation that way.

Thank you.

Our next question is from Chris Danely from JPMorgan.

Your question, please.

Thanks, guys.

So you guys talked about the Yukon product offering going into thin and light notebooks.

Can you give us AMD's plan and what your thoughts are on the netbook market?

Sure.

The way I think about this market as it unfolds is the distinction that we use today in terms of using words that are different between netbook and notebook I think is going to go away over time.

I think what you're really going to see over time is a continuum of price points in the marketplace and a continuum of form factors in the marketplace.

So really, this distinction between what is a netbook versus a notebook is going to go away.

Having said that, what we identified in the market today is an opportunity that we've exploited and are going to exploit pretty well, and that is the fact that given away netbooks are configured today, consumers who want a notebook at those kind of price points have to make a compromise and as a result, don't enjoy a full PC experience, particularly around the graphics and media capability of the machine and likewise, people who wanted a really nice thin and light machine had to pay a lot of money, typically well over $1,000, up to $1,500.

And our Yukon platform offers really a full PC experience at really low price points and the market is really reacting positively to that value proposition.

Okay.

And Dirk, would you care to put a -- give us a guess of the relative growth rates of your CPU versus GPU business this year?

Boy, I don't think we want to provide a forecast with that kind of granularity, particularly given the murkiness in the marketplace.

No problem.

I'll try one last question.

What exactly is the exit strategy for the Foundry Co?

Tell me what you mean by exit strategy.

How do you get it off your P&L?

Oh, what are the conditions for not consolidating?

Is that your question?

Well, it still flows through your P&L so I think all of us are looking--?

I understand.

So you're asking about how we're going to account for it over time.

Bob, why don't you take that?

In the early stages, obviously as we're the only -- we're the anchor tenant and the only customer through typical accounting regulations, we'll have consolidation.

As Foundry Co executes its game plan and brings on third party revenue, we will have the ability at some point in time in the future to not consolidate and then as time goes on and capital calls are required to fund that business to make the large capital investments to build out different clusters, we will, as we've said in the past, most likely not participate in those capital calls and our ownership will decline as time goes on.

Through a combination of all those things it will become just an investment by AMD and then eventually it just kind of winds its way out as time goes on.

Okay.

Thanks a lot.

Thank you.

Our next question is from Doug Freedman from Broadpoint AmTech.

Your question, please.

Hi, guys, thanks for taking my question.

Can you offer some color into what's going on as far as the ASPs in the CPU segment?

You had a nice performance there in GPUs where ASPs actually increased but I know that you've introduced a bunch of new products and Quad-Core is in there.

I would have thought we would have seen some improvement in mix.

Can you comment on that at all?

Sure.

As Bob said, first on the server side of the business, we did see a pretty good increase in ASPs quarter on quarter, while across the client businesses we saw I'll say a small decrease in ASPs quarter-to-quarter.

Largely due to mix and really, this was a matter of us choosing to pursue some opportunities for single core design wins with a few of our OEMs that really -- and that solely was responsible for the ASP decrease quarter-to-quarter.

Overall, I'd say the pricing environment was consistent Q3 to Q4.

Can you comment at all as far as what you're seeing in the marketplace from a -- is there a demand shift or any noticeable movement in the marketplace towards less powerful processors at all?

Any commentary you can offer there or any expected impact maybe from the new operating systems that are possibly due out later in the year?

When you say less powerful, do you mean more power like notebook or--?

More single core versus dual core or Quad-Core.

Well, first, consistent with the way this processor business has unfolded over the years, we sell more capability into ever lower price points, so as an example, quarter on quarter, we had a higher mix of Quad-Core and dual-core processors.

In addition, consistent with what's been happening in the PC industry for a while, system level ASPs have been coming down year to year to year and we saw that certainly in Q4 and we expect that to continue a bit in the future.

All right.

Great.

Thank you.

You bet.

Our next question is from Ross Seymore from Deutsche Bank.

Your question, please.

Hi, guys.

It seems like the Foundry Company is going to stay consolidated for quite some time so can you give us a little bit of color on how we should think about that breakeven target from a consolidated perspective and then maybe even the OpEx in the first quarter from a consolidated perspective?

I'd like you to actually think about it differently from that perspective.

Because the Foundry Company is not going to be a cash flow implication to AMD, the product Company.

I would really like to get you to start thinking about it just from an AMD the product Company standpoint, which is why the guidance and the discussion we've had since we've announced the transaction has been in that kind of vein.

So don't want to go down that path.

Clearly, that's a discussion that will have to take place with Foundry Company because that's -- and Doug outlined a little bit of his goals and expectations for spending and capital expenditures at the analyst day.

So not ready to go into that kind of detail and trying to get people focused on AMD the product Company.

Maybe switching gears a little bit, you took that 200 odd million charge for the inventory in the fourth quarter.

What do you expect from an inventory charge if anything in the March quarter?

I'm expecting hopefully just normal type things, based on our current modeling of what we expect as we move out over time, based on the inventory.

But, it's appropriate time to be cautious because the outlook is so murky, so don't expect us to have that magnitude of charge again.

Okay.

Then the final question, on the asset sales to both Broadcom and Qualcomm, can you tell us when those are hitting the balance sheet, if the Broadcom one didn't already, from a cash perspective?

Sure.

The DTV sale did impact the fourth quarter, so we received the bulk of that cash in the fourth quarter.

The broad -- the hand held transaction just took place of course this week hand that cash will take place in the first quarter.

Great.

Thank you.

Our next question is from David Wong from Wachovia.

Your question, please.

Thank you very much.

What percentage of service sales do you expect will be from Shanghai in the March quarter?

And in line with that, what percentage of the sub sales are Quad-Core today versus dual-core?

Could you repeat the second question?

Quad-Core versus dual-core for service, what proportion?

Okay.

First of all, in the current quarter, we expect the Shanghai sales will be less than 50%, but, couple of tens of percent of the overall Opteron mix and your next question was what percentage of our server sales are Quad-Core?

Yes.

Roughly three-fourths.

Thank you very much.

Sure.

Our next question is from Glen Yeung from Citi.

Your question, please.

First question is on ASPs, just going back to some of the other responses you've already given.

Is it your sense that in this market, there's no point in trying to actively lower pricing, is there just no demand or price elasticity in the market today?

I guess what I'll say is we manage pricing to optimize our gross margin dollars effectively no matter what is the environment.

We've seen over time that the degree to which end user PC demand can be influenced by microprocessor component pricing is pretty limited.

So take those two statements together and there's your answer.

Okay.

And then also wanted to ask about servers in particular, recognizing that you've got -- you're still growing -- well, Quad-Core is already a pretty high proportion.

You saw an ASP increase in the fourth quarter.

If ASP can actually increase in the first quarter and whether or not there's any potential that server kind of bucks the trend in Q1 relative to desktop and notebook.

By bucking the trend, what trend are you referring to?

I hesitate to say the word up, but up.

What I will say is the dynamics of that business are different.

The product cycles are kind of driven by new technology introductions.

Given all of that, and given that we've got such a good position right now in Shanghai, where as I said in my opening remarks, we offer superior performance across the majority of relevant workloads, I do think that in Q1 our server business is going to be the strong suit relative to the other two.

But in terms of predicting absolute numbers or absolute growth quarter-to-quarter, we can't do that.

And then on inventories for Q1, just to clarify, I'm not sure if you actually said this or not but you expect inventories in Q1 to fall again, your inventories?

Yes.

We have lowered our manufacturing rates quite a bit and expect inventory will be flat to down, just depends on end market demand.

We're definitely slowing everything down pretty significantly.

Bob, is your sense that you've got utilization rates to a point now where you are undershipping the end market by a relatively wide margin?

Well, there's -- if you -- let's go back a little bit in time.

If you look at the fourth quarter data, clearly the PC end market sales was significantly different than the microprocessor sales, so there was a massive inventory correction that took place in the fourth quarter.

What I think we're all trying to understand is, is it done?

Is it halfway done?

Is it a quarter done?

Is it three-quarters of the way done?

Whatever that impact is.

We clearly don't feel like it's done in the notebook space.

So, kind of managing our way through that from that perspective, hoping over time obviously those two numbers will come into balance, which is end market and actually selling of microprocessors to service that markets.

But right now it's out of balance, as the supply chain is trying to get corrected with inventory.

But Bob I was--?

Sorry.

Well, I was just going to put a point on that by saying we think our CPU sales will -- out of AMD will be less than end user consumption.

So hence the channels are going to drain in Q1.

And likewise, as Bob told you, our own inventory is going to drain in Q1 so we're clearly manufacturing below our shipment level in Q1.

My question is then, you make the point that you may be better balanced in desktop than you are in notebook and I think Intel said the same thing on their call.

What's the clue that tells you that that is true?

Is it that order trends in the desktop products seemed to have stabilized or cancellations seemed to have diminished?

We can see the inventory levels of our master distributors, so visibility on inventory levels, along with sales out at the top of our distribution channel, gives us a pretty good idea of what's happening downstream on desktop.

Then the pull rates from OEM on desktop, actually started at low rates but actually started on a consistent basis right as the year started.

So, gives us a pretty good flavor from both angles that not much in the system, if the system has demand it need to be serviced.

That's very helpful.

Thanks a lot.

Our next question is from Adam Benjamin from Jefferies.

Your question, please.

Yes, thanks, guys.

Just trying to figure out some of the commentary regarding the breakeven at $1.3 billion.

If you do some math on that, it would seem to equate to roughly down 10% in the March quarter.

I know you're not giving specific guidance but it would require roughly a 30% sequential improvement in June to get there in the breakeven quarters you're talking about.

Your biggest competitor is talking about down 15% and not really even giving guidance.

I just want to throw those numbers out there and see if those are in the ballpark and can you walk us through the dynamics of how the mechanics of how you're going to get there in.

Yes, I think you're confusing a couple things.

We're not giving guidance, number one.

So we're not giving guidance for first quarter, we're not giving guidance for second quarter.

The guidance I'm trying to give you is our cost structure of what's our cost going to look like from a modeling perspective of how much money are we going to spend and, therefore, you can interpret what revenue we need to make to break even.

Whether we get there or not, will be when the dust settles we're figure that out.

We're trying to drive from a much higher rate than $1.3 billion to get down to $1.3 billion as quickly as possible.

We think that's doable by the second quarter and we're working like (expletive) to get revenue to match that number but we're not going to say we're going to be there or not at this point in time.

But we think that's the appropriate thing to do, to run our business, preserve cash, hold our road map intact and service our customers.

Okay.

But you are kind of saying you're working hard to get there on the revenue side to 1.3.

You may not get there and you don't want to guide to that.

But maybe help us understand the dynamics of how you're thinking about it, how you would get there in terms of the moving parts?

Dirk, here, let me -- I'll probably say the same thing as Bob but with different words.

When we talk about the $1.3 billion breakeven model, what we're talking about is where we're sitting with the cost structure of the Company, with the goal being given the severely limited visibility that we have on the business, to preserve and generate cash on the one hand while preserving our long-term investments particularly on the product road map and those two priorities land us at an operating expense level that Bob defined in his opening remarks.

Those remarks are not intended to forecast where revenue is going to be over the course of the next two quarters.

So different conversation.

Okay.

That's all I have.

Thanks for the clarification.

Thank you.

Our next question comes from Patrick Wang from Wedbush Morgan.

Your question, please.

Yes, thanks for the questions here.

First off, can you talk a little bit about the ramp you guys have done at 45-nanometer.

I think before you said that greater than 50% shipping, exiting Q1 here--?

Patrick, I'm sorry, but we can't hear you.

Hello?

Yes, could you just speak a little louder.

Sorry about that.

I'm a little shy, sorry.

Anyways, so first question is just the ramp of 45-nanometer, I know you guys had talked about greater than 50% shipments exiting Q1 here.

Just wanted to see how that tracked so far?

Yes.

I mean, we are in the fourth quarter, roughly half of the material is 45-nanometer from a starts perspective.

Obviously, that material comes out in the first quarter.

So we're moving quickly to -- by the second quarter, the majority of what we'll be shipping will be 45-nanometer.

Clearly the break point cross from shipments is in the first quarter kind of time frame.

So there's no pullback in terms of the ramp of 45 at this point?

No, no.

Okay.

Great.

And then second, I just wanted to know--?

We just need a demand pick up, that's all.

Further, to the extent we've taken down our utilization it would be on the 65-nanometer equipment.

Got you.

Got you.

Okay, that's helpful.

Second, just in terms of moving parts for gross margins, can you help us think about some of the moving pieces behind gross margins as we go into the first quarter?

I know you said 40%, but -- or greater than 40.

Sure.

Again, I'll kind of -- there's kind of positives and negatives.

On the positive side is really just the discussion we just had, 45-nanometer, continues to be a bigger piece of the equation.

That yields a better cost structure, therefore better gross margin.

Our product offering, with the introduction of 45-nanometer microprocessors, both in the server variety and the desktop variety continues to be a full quarter offering in the first quarter.

That's the good news.

We play higher up in the stack, different areas to play in, so that's good news from that standpoint.

The offsetting pieces, so the negatives are utilization of factories are definitely lower.

And demand is down.

So that part of it -- and then finally, a piece that needs to be considered after the Foundry Co transaction is we'll pay a markup for that material.

But again, that's later in the quarter and we'll disclose that information from that perspective.

So is it fair to say that if we look at the shape of margins going forward here, that it could be choppy, especially with the -- with this transition to Foundry Co?

Yes.

I mean, like I said, we'll give you the pieces starting in the first quarter, even though it will be partial AMD, 100% and partial less than 100% from that perspective but and we still believe 40% or more is a very reasonable target as we move through time for AMD the product Company.

Okay.

Okay.

And then also another question on Yukon.

I know that you guys had touched on this a little bit earlier today but any traction here you can talk about?

Any milestones that we should be looking forward to?

Well, first, we had -- of course, HP announced with us at CES the availability of the Pavilion DV2 platform.

That will start shipping here in the current quarter.

The milestones you want to look for are more design wins which our customers will announce at the appropriate time.

Great.

Then just the last question here.

If we think about enterprise and consumer trends out there, can you just provide a little bit more color in terms of what you're seeing there, who's weaker than what and some of the trends you're seeing?

Yes.

As -- seeing through our server business, certainly, we started perceiving softening in enterprise purchases earlier in 2008, frankly, so before the Q4 meltdown.

During that time, consumer spending and hence consumer PC consumption was still healthy and growing.

Interestingly, Q3 to Q4, PC end user demand really didn't decrease very much Q3 to Q4, of course, the fact that it decreased at all is noteworthy.

Usually consumption increases by greater than 10% Q3 to Q4.

Going into 2009 it's awfully tough to forecast where consumption is going to go across either market segment which is why we're so reluctant to create a forecast and I think the real wild card is going to be to what degree the current economic climate has an effect on consumer spending patterns and how that affect, whatever it is affects PC consumption.

It's just too early to tell.

Thanks so much, guys.

Thank you.

Our next question is from JoAnne Feeney from FTN Midwest.

I was hoping perhaps you guys could clarify the operational impact of splitting off Foundry in terms of the cash use, especially where it concerns operating expenses?

Is it the case that basically your Foundry operations will be funded out of the Foundry balance sheet at this point going forward once the deal is closed?

Yes.

Okay.

Think of it this way.

The CapEx, the operating cost of the factory, the process development type cost, and the SG&A to run Foundry Co will be borne by Foundry Co from a cash perspective.

Could you remind us what's the--?

Obviously, we'll buy material so there will be cash coming from us to pay for the material we buy but if there is a shortage of cash based on that scenario, what we pay for material, that will be borne by the Foundry Co balance sheet and funded accordingly by Foundry Co.

Could you remind us what's the starting cash balance for Foundry Co?

$1.4 billion.

And now that AMD product Company won't be required to put up money for the CapEx, what level, Bob, are you folks now comfortable with cash balances for AMD the product Company?

Kind of from my perspective of -- probably I'm not going to revise that forecast for AMD the product Company just yet, based upon this murky environment, so it's going to stay in that $800 million to $1 billion kind of range.

Probably could drift down but I'm just going to be cautious based upon what's happened in the fourth quarter and the kind of outlook we have.

So post the transaction, I will have excess cash over the comfort zone.

Okay.

Which is nice to have in this environment.

Sure.

And then one -- on the gross margin side.

You have a wafer agreement with Foundry Co that I understand to increase a pricing arrangement.

Based on that wafer agreement, do you -- are you comfortable with an outlook of a 40%-ish gross margin?

Is that where your comfort level comes from?

Yes.

Or do you need to see ASPs rise?

No.

I mean, based upon the Foundry agreement that we know, which as we've told people, it's a cost plus arrangement which we're very happy with, on microprocessors, that that drives even with that markup that's generated in that arrangement that drives a 40% or more gross margin for AMD the product Company.

Are you less exposed because of that agreement to the consequences of a drop in capacity utilization than you would be under the sort of pre-Asset Smart arrangement?

It's -- I'll call it -- think of it this way.

On day one, it's better than it is today by a significant margin.

But it's not as good as I'll call it of being a completely fabless company.

As time moves on, it moves more and more to the fabless company model.

So it's kind of better than today, day one, and then continues to I'll call it the Foundry model as you move through time.

Okay.

Well, that's it for me.

Thanks a lot.

Thank you.

Our next question is from John Pitzer from Credit Suisse.

Your question, please.

Good afternoon, guys.

A couple questions.

Dirk, first, just given the new administration in Washington, is there any other boxes you have to check relative to the agreement or is that just all smooth sailing from here and already taken into consideration?

No new box, no new closing condition as a result of the change in administration.

And then Bob, I know that you want us to think about AMD the Design Company differently from the Foundry Company but in large part as we try to figure out the success of the Design Company, the Foundry's investments in 45 and 32-nanometer technology kind of matters.

I'm kind of hoping you can give us some sort of sense against that backdrop what the overall CapEx of the Foundry Company has targeted for in 2009?

I'd circle back to the number that Doug Grose talked about at our analyst day, which is was an approximately $1 billion of CapEx for next year for the Foundry Company.

And then I guess my last question, Bob, just relative to that 40% gross margin target, you said the comfort level has to do with the wafer agreement with the Foundry Company.

I guess I'm a little bit concerned, given where utilization rates are going to bottom, given your inventory levels, Intel's inventory levels, Nvidia's inventory levels that that pricing is never the first casualty in a downturn, it's the second.

So I guess at what point does -- at what point does pricing start to impact that 40% comfort level?

How much of a move do we need to see before you start to get nervous about that statement on 40% gross margins?

It's a dynamic equation of your product offering, where you sit in the stack, where the demand is, needs to service from the stack, where you are in the technology node migration.

I mean, there's a lot of of moving parts.

What negotiations you have with your vendors on either raw silicon, buying wafers, assembly, testing.

To me it's, I'm not sure I know exactly how to answer your question, just give you a sense.

That's the world we live in.

ASPs are always declining.

You've got to give more technology for the ASPs.

We've got to continue to reduce our costs.

We'll just have to continue to manage that.

And feel comfortable, based upon where we're at, what we see in the marketplace that a 40% gross margin seems doable.

And the current quarter's results kind of share -- if you look at it over a a period of time even in 2008 even though, quote, maybe we haven't always been in the best position from a product offering, we've been in that 40% kind of zone.

And then Bob, the last question I have for you, is just given the market downturn here, any change on how you might use third party foundries like TSMC relative to some of the graphics parts or is it still sort of a situation where you want to pursue a strategy of having multiple Foundry options?

Well, over time our goal is to have multiple options to build GPUs.

In the near term, in 2009, to be honest, the majority, almost 100% of our volume, will come from third party foundries, not counting Foundry Co, because they're just not prepared to pick-up bulk technology in the early stages of 2009.

Perfect, thank you.

Over time, as we get into 2010, '11, and beyond we clearly would like some balance in our ecosystem.

Great, appreciate it.

We'll take two more questions, please.

Next question is from Kevin Cassidy from Thomas Weisel Partners.

Your question, please.

Thank you for taking my question.

I just wanted to ask about the Intel licensing agreement.

What is the terms for that?

Is it up for renewal?

Terms meaning?

Most licensing agreements have a certain time frame.

Yes.

Hang on just a minute.

Well, first of all, the term -- our license to Intel patents are perpetual, meaning the patents to which we're licensed and they're licensed are licensed forever.

The agreement does expire and we expect we'll renegotiate it as we did earlier in this decade.

But the licenses persist.

Okay.

And do you know when will you extend it?

I expect we'll enter into those negotiations over sometime in the next couple of quarters.

Okay.

Thank you.

Our final question is from Eric Ruebel from MTR Securities.

Your question, please.

Hi, thanks for taking my question, guys.

On the convertible bond repurchase, what was the principal amount that was purchased in the quarter and how much cash did you use?

We took -- we retired $60 million worth of debt, used $20 million worth of cash.

Great.

And on the expected cash restructuring charges in Q1 and possibly Q2, could you give me some guidance on -- in the guidance here, take some more costs out, what I could expect for cash restructuring in the first two quarters of next year?

We'll give you that granularity later.

We're still trying totally all that up.

Then one last quick one on the guidance for the new Design Co, depreciation and amortization, $100 million.

Does that also include the $30 million that you've been charging per quarter for goodwill amortization or is that -- would that be incremental?

No, this is depreciation to run AMD the product Company doesn't include amortization of intangibles.

Okay.

Great.

Thank you.

Well, we would like to thank everybody for joining our fourth quarter earnings call today.

And those of you on the webcast and wish you all a very good evening.

Thank you.

Ladies and gentlemen, thank you for participating in today's conference.

This concludes the program.

You may now disconnect.

Good day.