Advanced Micro Devices - Earnings Call - Q1 2011
April 21, 2011
Transcript
Speaker 8
Good afternoon. My name is Huey, and I'll be your conference operator for today. At this time, I would like to welcome everyone to AMD's First Quarter 2011 Earnings Conference Call. All lines have been placed in a listen-only mode. After the speaker's 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, Vice President of Investor Relations for AMD. Please go ahead.
Speaker 13
Thank you, Huey, and welcome to AMD's First Quarter Earnings Conference Call. Participants on today's conference call are Thomas Seifert, our Chief Financial Officer and Interim CEO, and Rick Bergman, Senior Vice President and General Manager of AMD's Product Group. 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 U.S., the number is 703-925-2533. The access code for both is 1522169. The telephone replay will be available for the next 10 days, starting later this evening. This is the first quarter AMD is accounting for GlobalFoundries under the cost method, and AMD's ownership of GlobalFoundries on a fully diluted basis decreased to approximately 12% as of the end of the first quarter.
Reconciliation for all non-GAAP financial measures discussed today is included in the financial tables accompanying our earnings release, available in the Investor Relations section of amd.com. Written executive commentary, which contains additional information regarding AMD's results, is posted on AMD's website at quarterlyearnings.amd.com. I would also like to highlight a few dates of interest for you. Thomas will present at the JP Morgan Global Technology, Media, and Telecom Conference on the 17th of May in Boston. Our second quarter quiet time will begin at the close of business on Friday, June 17th, and we will announce our second quarter earnings on Thursday, July 21st. 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 review our filings with the SEC, where we discuss the risk factors that could cause actual results to differ materially from expectations. You'll find detailed discussions about such risk factors in our most recent SEC filings, AMD's annual report on Form 10-K, for the quarter ended December 25th, 2010. With that, I'd like to hand the call over to Thomas.
Speaker 2
Thank you, Ruth. AMD's performance in the first quarter of 2011 was solid. We launched the first of our revolutionary AMD Fusion APU platforms based on our low-power Brazos processes, and we're pleased to report excellent OEM adoption and sell-through to end users. In the first full quarter of availability, roughly half of our notebook shipments were based on Brazos, a product rapidly gaining credit for redefining the user experience in both the network and thin and light notebook segment. We also began shipping Llano for revenue in the first quarter, and it is very simply the most impressive process in history. Featuring a modern graphics architecture, Llano gives mainstream PC users something they have never had before: a brilliant visual experience coupled with all-day battery life. It delivers a better end-user experience than anything else on the market, and our customers have told us that.
You should expect to see Llano-based systems widely available in this quarter. Beyond its unique performance characteristics, Fusion is also a key part of our overall profitability strategy. We delivered on our top-line guidance. We were right on target with our non-GAAP gross margin guidance at 45%, and we managed operating expenses carefully. As a result, we delivered solid free cash flow in the period. In the commercial market, improving performance in our embedded and professional graphics segments was masked by continued softness in our overall server business. While disappointed with our top-line trajectory in the server business, we've taken a number of steps that we believe should lead to improved results in the second half of the year. First, we've onboarded a larger team of experienced customer account engineers who are now engaged with customers.
In addition, we are working more closely with our OEMs and ODMs to drive the new AMD-based systems into the hands of key cloud virtualization customers. Finally, we continue on track to ship Bulldozer-based server platforms towards the end of the summer. Our new Bulldozer core delivers substantial improvements in floating-point performance in a platform featuring superior memory utilization and I/O throughput. In graphics, we continue to advance our leadership in a number of important ways. First, we launched the HD 6990, truly the world's fastest graphics card. Second, we set an all-time record for mobile discrete revenue. Third, we are now the exclusive provider of discrete graphics solutions for Apple's iMac and MacBook Pro platforms. In short, we believe that we continue to gain market share in the mobile GPUs in the quarter.
Our continued leadership in discrete graphics and our exciting new Fusion APUs prove the world's best PCs have AMD Radeon graphics onboard. Turning to our first quarter financial summary, first quarter revenue was $1.61 billion, a 2% sequential decline and within the guided range for the quarter. Non-GAAP net income was $56 million, or $0.08 per share, calculated using 740 million fully diluted shares. Among other adjustments, it excludes a non-cash gain of $492 million related to the dilution of our equity interest in GlobalFoundries. Non-GAAP operating income was $92 million. Non-GAAP gross margin was 45% and flat quarter-over-quarter. Lower than anticipated 32 μm manufacturing costs and the ramp of margin accretive APUs offset the negative impact of seasonally lower game console revenues, lower ASPs, and a weaker product mix in legacy microprocessor products.
Operating expenses for the quarter came in below the guided range of approximately $650 million due to tight expense management. R&D was $360 million, and SG&A was $261 million. Adjusted EBITDA was $198 million, down $43 million from the prior quarter. Now, switching to the business segments. Computing Solutions segment revenue was $1.2 billion, down 2% compared to the fourth quarter of 2010. We are excited about our APU platforms, which are gaining traction in the market. We tripled unit shipments over the prior quarter. In addition, we started shipping Llano, our high-end APU, late in the first quarter. Computing Solutions operating income was $100 million, up $9 million from the previous quarter, primarily due to improved gross margins from a higher mix of APU sales compared to the prior quarter.
Graphics segment revenue was $413 million, down 3% compared to the fourth quarter of 2010, primarily due to a seasonally driven decline in game console revenue. Graphics segment operating income was $19 million, down $49 million from the fourth quarter of 2010, primarily due to seasonally lower game console revenue. Turning to the balance sheet, our cash, cash equivalents, and marketable securities balance at the end of the quarter was $1.75 billion. Accounts receivable at the end of the quarter was $797 million, down $171 million, primarily due to the timing of sales during the quarter. Inventory was $648 million exiting the quarter. Long-term debt as of the end of the quarter was $2.2 billion. Non-GAAP adjusted free cash flow was $154 million, up $143 million from the fourth quarter of 2010 due to higher non-GAAP net cash provided by operating activities.
Turning to the outlook, the following statements concerning AMD are forward-looking and actual results could differ materially from current expectations. For the second quarter of 2011, AMD expects second quarter revenue to be flat to slightly down sequentially. Operating expenses are expected to be $620 million. As mentioned in the last call, our priorities remain clear. I said we would stay the course of execution and we continue to do so, bolstering our discrete graphics leadership, tripling unit shipments of our first APU Brazos, and shipping Llano for revenue in the quarter. I said we would increase our pace, and in addition to meeting or exceeding all of our major engineering commitments, we've launched a number of key cross-functional initiatives designed to improve productivity and scalability, optimize our go-to-market activities, improve R&D productivity, and generate greater supply chain efficiencies.
To close, we know we have the best IP portfolio in the processor business, and with our new Fusion APUs and Bulldozer on the way, we have our strongest product portfolio in the company's history, and we have a business model tuned for agility and profitability. I would like to thank our employees for their continued dedication, creativity, and commitment to delivering on our commitments. Before we turn it back to Ruth for the Q&A, I would like to make a brief comment on the state of the CEO search at AMD. The board is very happy with the interest we've received and is actively interviewing candidates, and we are pleased with our progress to date. Ruth?
Speaker 13
Huey, we'd be very happy for you to poll the audience for questions now, please.
Speaker 8
Thank you, Ms. Cotter. Ladies and gentlemen, if you would like to ask a question at this time, please press the star key followed by the one on your touch-tone telephone. If you wish to remove yourself from the queue or your question has been answered, you may press the pound key. Again, to ask a question at this time, please press star, then one on your touch-tone telephone. Our first questioner in the queue is JoAnne Feeney with Longbow Research. Please go ahead.
Speaker 14
Hi, thanks, and congrats on a nice solid quarter. I have a question about your interpretation of the aggregate situation in light of the results of your competitor. Did it cause you to change your assessment of sort of the global PC environment? Did the results sort of match your views? Is this purely a timing issue between you and Intel, or perhaps you can give us a sense of your visibility to the second half, given what we're hearing regarding the global PC situation?
Speaker 2
Yes, very good questions. I think I said in the last couple of calls that our expectation is around 11% for PC growth for this year. To be very honest, at this point in time, we don't see a reason to deviate very much from that opinion. Actually, data is moving more towards the view we had on the market for some time now. You can discuss whether it's now 10.5% or 10.7% or 11%, but we don't see any reason to deviate from that expectation on the floor for this year.
Speaker 14
Perhaps as a follow-up on Llano, where we understand the volume is coming in the third quarter, one question that comes up a lot is how compelling to the consumer will your advantage in graphics be? I'm just wondering if you could fill us in on your efforts with the OEMs and other resellers to help them understand the difference in graphics and whether they're able to translate that to customers in your view.
Speaker 2
Yeah, Rick, do you want to take that question?
Speaker 1
Sure. Just to be clear, the volume is actually coming in in Q2. That's certainly when we expect to ramp to production and have platforms launch on Llano. I think the best answer to your question is the success we've seen with our Brazos platform. It was just fantastic success in Q1, and our OEMs are now really on board with bringing even a better, more compelling, higher-performance solution with Llano. To a certain degree, we hear the reception around the great visual graphics video experience even from our competitor and how important that is in the marketplace. We think we're right on target with our value proposition.
Speaker 14
Great, thanks. Maybe you could elaborate on Brazos. Where are you seeing the greatest success? Is it tier one players? Is it white box? Is it mostly notebook? Are there some desktop activities?
Speaker 2
To be very honest, you know, what excites us the most is that we see the success across the board. We have seen very strong OEM adoption across the whole customer portfolio. We've seen great traction in the embedded segment and also now in emerging markets in the white box arena.
Speaker 14
OK, great, thanks. That's it for me.
Speaker 8
Thank you. Our next questioner in queue is Glen Yeung with Citi. Please go ahead.
Speaker 7
Thank you. Thomas, I think I heard you say that half of your notebook shipments in the quarter were Brazos. Is that accurate? If it is, when we think about second quarter, what proportion of your notebook sales do you think will be a combination of Brazos and Llano, and what are the implications of that for gross margin and ASPs?
Speaker 2
Yes, very good question. You heard right, 50% of the share we had in the first quarter. I think the implications for gross margin in the second quarter are that gross margin is going to be up quarter-over-quarter. We will see a balanced impact on ASP performance because, as you know, while highly margin accretive, Brazos ships at a lower price point, and we have to see how the mix is going to balance itself out. Flat is today a very good assumption, but with gross margins moving upwards.
Speaker 7
Interesting, that's great. I wanted to ask another question, which is you sort of talked about some progress you're making or some growth you saw in emerging markets. I wonder if you could address what you're seeing in the corporate market just from an industry perspective, and then perhaps more importantly, how you feel AMD is doing in the corporate market and whether or not these new products you think will help your chances.
Speaker 1
Certainly. Hi, this is Rick again.
Speaker 7
Hi, Rick.
Speaker 1
Obviously, we have much greater success in the consumer market, and that's to a certain degree our focus. However, the value proposition on Llano is strong as an example in the corporate market as well. Certainly, video playback, video conferencing, all those factors are becoming more and more important in the corporate environment. In addition, of course, as we're seeing, consumer devices and notebooks and desktops and so on are increasingly being also used in the commercial marketplace, certainly in SMB, but even in enterprise now, they are adopting those consumer platforms.
Speaker 7
Actually, one last quick one, if I might, just with respect to Bulldozer. I know you guys have deployed some field application engineers into the market, and I wonder if you could give me a sense of what the initial response has been as you sort of prep the market for Bulldozer.
Speaker 2
The first feedback is very positive and welcoming. People appreciate, of course, that we have a significantly higher visibility in the marketplace now in terms of feet on the street, but also in terms of trying to get mind share. The product is exciting on a platform level. As I said before, the value proposition is exciting, especially for cloud and virtualization applications, and that allows us to be optimistic for the second half.
Speaker 7
Thanks.
Speaker 8
Thank you. Our next questioner in queue is John Pitzer with Credit Suisse. Please go ahead.
Speaker 3
Yeah, guys, thanks for asking me, taking my question. Tom, can you talk a little bit about what the ASP trends were in the quarter in the core client business? I guess, given that the server continues to be an area of weakness, I wonder if you can help me understand the relative size of servers in the revenue line, in the profit line, and what do you think the trend will look like going into the second quarter? I guess, more importantly, at what point will Bulldozer start to be a positive impact? Is it really a this year event, or do we have to start thinking about 2012? Thanks.
Speaker 2
Yeah, I'm not going to get more granular in terms of the revenue lines, but I think it's fair to say that we pretty much, I think we've stabilized market share in the first quarter on the server segment, and we will see Bulldozer impact in terms of revenue in the second half as early. You said we would start to ship product in the summer, and that means that we see revenue in the third and, of course, in the fourth quarter.
Speaker 3
Thomas, as my follow-up, on the graphic side, you guys have done a great job on the market share side over the last several quarters. How do you see that trending going forward? With kind of the APU coming out, is there any change in your view of sort of discrete attach rates on the client side? Thank you.
Speaker 2
Yeah, very good. Of course, we are really proud that we are able to keep this technology lead. Moving forward, we have climbed to a high level on the market share side. The focus is now to make sure that profitability moves in the right corner too. I think that is going to be the focus moving forward. For this year, we expect attach rates not to change. There are early research indications that in the first quarter, it might have even gone up, which is also partly reflected in our strong mobile GPU shipments. Moving forward, we'll see that at least at the low end of the graphics stack, there will be some replacement because, especially on our side, the APU performance is really compelling. For this year, attach rates seem to be holding up or maybe even ticking up slightly.
Speaker 1
To elaborate a little more on that, obviously, our Llano processors have truly discrete-level graphics performance. In some cases, the market just will not need to add an additional GPU. On other platforms out there, certainly, if you're buying a $600, $700 notebook, you want the latest and greatest GPU architecture, which today is defined by DX11 solutions. In that case, you're going to have to add discrete graphics. That is actually pulling up the attach rate on the other platforms. For the summer 2011, we see it as basically staying the same.
Speaker 2
Thanks, guys.
Speaker 8
Thank you, sir. Our next questioner in queue is Tim Luke with Barclays. Please go ahead.
Speaker 6
Thanks so much. Just to clarify, Thomas, when you suggest down flat to down slightly, is that down 1 percentage point or 2 percentage points? Is that how we should think about it? When you articulate that you believe that the gross margin will be improved in the calendar second quarter, is that sort of 100 basis point? Is that how we should think about that? Perhaps you could just tackle this, Huey.
Speaker 2
Good question. The box we would draw around a flat or slightly down is 0% to -5%-ish, probably. It's the right range. I think you're very much on target with your gross margin expectation. We think it's going to be up around 100 basis points.
Speaker 6
Thank you. Perhaps just on the ramp of Llano, can you give us a feel for the availability of product and really what sort of metrics we might be able to frame in terms of thinking about how that ramp is likely to progress now?
Speaker 2
We achieved, we made great progress, which allowed us to ship for revenue. We think we have ample product available in the second quarter. On the other side, you know, we saw on Brazos that once the product was out and available on shelf, demand really went up significantly, and we'll try to prepare for such a situation.
Speaker 1
Again, our key for Llano is to hit the critical cycle in the industry, which is the 2C or the back-to-school cycle. We have broad-based OEM platform adoption for that time frame, and we're well positioned to take advantage of that cycle.
Speaker 13
Can I do a follow-up? Go ahead, operator.
Speaker 8
No problem. Our next questioner in queue is Srini Pajjuri with CLSA Securities. Please go ahead.
Speaker 17
Thank you. Thomas, just one more clarification to the previous question. On the OpEx, you said it's going to be about $620 million. Just wondering if that's going to be a temporary decline or if that's the range that you expect going forward.
Speaker 2
We obviously worked hard to keep our OpEx under tight control. In the second quarter, we see a couple of 28 μm tapeouts hitting R&D costs, but the $620 at this point in time seems to be a good indication on the cost level moving forward.
Speaker 17
OK, great. A couple of questions on the ASP front. Obviously, you know, Brazos looks like it impacted the ASPs. I mean, should we expect ASPs to increase in the next few quarters as Llano ramps, or is that a function of the server business coming back at this point?
Speaker 2
Fair question. I think for the second quarter, it's fair to expect that, or at least we expect that both ramps equal each other out, so to speak. In the third and fourth quarter, ASP performance should go up, a higher share of Llano shipments, and the additional impact of the server shipments.
Speaker 17
OK, and then on the server business in general, Thomas, at this market share, I think you have about 5% or 6% market share. Are you profitable? That's my first question. If not, what's the sort of market share you need to justify the investments longer term? Thank you.
Speaker 2
Fair question. We are sitting today at 6.6% market share. Profitability at such a point is borderline, but we feel comfortable that the market share lift we need in order to reach profitability in this segment is actually very small.
Speaker 17
Is it double digits, single digits, or would you like to give us a bit more color on that?
Speaker 2
It's very close.
Speaker 17
OK, great. Thank you.
Speaker 8
Thank you, sir. Our next questioner in queue is David Wong with Wells Fargo. Your line is open.
Speaker 18
Thank you very much. Could you tell us what proportion of Brazos sales in the March quarter went Ontario versus Zacate, and does it look like Ontario is being used primarily in netbooks while Zacate goes into value notebooks?
Speaker 1
I don't know if we have an exact split on the two parts. Good adoption on both. We are seeing Ontario in some very exciting thin and light form factors, but adoption in a whole variety of different platforms from netbooks up into desktops.
Speaker 18
OK, great. When you expect to bring Bulldozer platforms out at the end of the summer, do desktop and server platforms come out at about the same time, or does server come out later than desktop?
Speaker 1
In the last call, we indicated early summer for desktops and late summer for servers, and that's still where we're at.
Speaker 18
Great. Thanks very much.
Speaker 1
Thank you.
Speaker 8
Our next questioner in queue is Jim Covello with Goldman Sachs. Please go ahead.
Speaker 12
Thanks so much for taking the question. Could you guys just go by segment, desktop, notebook, server, and talk about what your share expectations are for the year?
Speaker 2
As you know, we don't give market share guidance for the year, and I'm certainly not going to start here now. I think we always said at our analyst day, you know, we expect our desktop market share to stay about flat, slight gains. Long term, we don't exact, there's no reason why our notebook market share should be materially different from our desktop share. From where we are today on the server side and with the optimism we have on the Bulldozer platform, I think it's fair to say that we expect market share gains in the second half.
Speaker 12
If I look at the issue from where you are today, because if we look at the full year, it'll be tough just because of Intel shipments growth and ASP growth in the first quarter versus yours. From here is where you would expect to gain the share, kind of mark to market for the end of Q1. Is that the idea?
Speaker 2
I think that's fair.
Speaker 12
OK. Could I just, on the foundry side, obviously there's been a lot of discussion. Obviously, you're aligned well with several foundry partners. Could you give us a little bit of an update on who you think you're going to use for what going forward and what the different parameters are there? Thank you very much.
Speaker 2
Yeah, I think we were quite open on our last call. Our foundry strategy has not changed. We are pleased with the progress we see. As I said, we expect several 28 μm tapeouts during the current quarter. We balance our loading according to risk, performance, and price. There is, of course, a natural incentive for us at this point to work closely with GlobalFoundries, and that is, of course, something we continue to do.
Speaker 12
Thanks very much.
Speaker 8
Thank you. Our next questioner in queue is Ross Seymore with Deutsche Bank. Your line is open.
Speaker 16
Hi, Thomas, could you talk a little bit about in your Computing Solutions Group, the difference between what happened in NPUs and chipsets? You've done a great job at gaining share in chipsets. I just wonder if that's continued in the first quarter.
Speaker 2
Our chipset business in the first quarter was, we are pleased with the performance we had. Let's put it this way.
Speaker 16
How did it compare versus the Computing Solutions Group as a whole? Better, worse?
Speaker 2
Better.
Speaker 16
OK. The gross margin, when you talked about that going up roughly 100 basis points sequentially, is there anything driving that other than the positive mix of more APUs?
Speaker 2
At this point in time, it's very much mix-driven. The cost impact is there, but slightly only in the current quarter.
Speaker 16
Gotcha. My last question, you had an extra week in the first quarter. Any impact either in what happened in the first quarter or your guidance into the second quarter that we should think about, and more so how your guidance would be different if you weren't losing that week?
Speaker 2
To be very honest, the additional week was part of the guidance for the first quarter, and we took that into account. The guidance for the current quarter takes into account that it's one week less.
Speaker 16
Absent that, you would actually have grown nicely in your guidance in the second quarter?
Speaker 2
Yes, I think that's fair to say.
Speaker 16
OK, great. Thank you.
Speaker 8
Thank you. Our next questioner is Vivek Arya with Bank of America. Please go ahead.
Speaker 11
Thanks for taking my question. Thomas, I just wanted to revisit this PC unit growth expectation for the year. I think you mentioned the 11% unit growth reasonable for the year. Could you perhaps give us some more color on where you are seeing the positive sell-through? Is it consumer? Is it corporate? Is it developed market? Is it the emerging markets? Thank you.
Speaker 2
Yeah, you know, I think our opinion here doesn't, and the visibility we have does not differ materially from what you hear from others. North America and Western Europe was weak, and we don't expect that to change materially. We saw strong demand in the emerging countries all over that spectrum, China, India, especially Latin America for us. We have been putting resources into those regions. I think we mentioned that on several calls in the past that we ramp up our infrastructure in emerging markets. Of course, our exposure to this consumer segment is significantly higher than to the commercial sector. We have been seeing that the products and the price points we hit, especially with Brazos, are right on the spot of what those market segments demand. Yeah, that's pretty much it. Very strong general business.
Speaker 11
Got it. As a follow-up, Intel recently raised their CapEx, and they're planning to be very aggressive at the next generation, 22 μm and 14 μm nodes. How do you look at that, Thomas, from a competitive point of view? What are the implications on AMD? What do you need to do to respond to that, if anything?
Speaker 2
The good thing is we don't have to invest. From my CFO perspective, I think that is very positive. Joking aside, you saw that our partners, GlobalFoundries and TSMC, spend an aggregate amount of money that I think even outperforms the Intel investment. I think it's $8 million on both sides, so $8 million and $7 million. Don't quote me here, but a significant amount of money. Of course, we work with our partners in terms of the technology we need to be competitive. I think we have also reached the point where competitiveness in this market segment is defined by more than just the technology roadmap. We have just seen that on Brazos.
Moving forward, we see a lot more levers that help us to differentiate ourselves and bring highly competitive products into the market that just take into account the business model we have and the skill set we have, the IP we have, and especially the agility that this business model offers us.
Speaker 11
Got it. Just one last question, if I may. I think in the past, you have given long-term targets of getting to 50% gross margin, perhaps better than that. What about the APU gets you better margins? What are really the main drivers from here on to get to that kind of target? Thank you.
Speaker 2
Yeah, we said that this pretty much comes from different levers. For the low end of our APU strategy, the Brazos products that we are shipping currently that are so hugely successful, for those products, it's much more a cost optimization. They are highly gross margin accretive because of good design and chip size efficiencies. On the APUs that are the spectrum that is covered by the Llano products, that is much more a margin accretive game through price performance and price points at which we can play now.
Speaker 11
Great. Thanks and good luck.
Speaker 2
Thank you.
Speaker 8
Thank you. Our next questioner in queue is Christopher Danely with JPMorgan. Please go ahead.
Speaker 0
Hey, thanks, guys. We asked your main competitor this. Might as well ask you. It seems as though the commentary or tone of business from you guys at Intel differs somewhat with some other companies in the PC food chain as far as some of the OEMs and the drive companies go. How would you guys explain that discrepancy?
Speaker 2
I don't think in that case our arguments are so much different from our competitor. As I said before, we have seen strong demand patterns out of the emerging markets. We have seen good demand out of the general business. I think we saw some uncertainty in some market research that came out early. I think those numbers will be corrected as we move forward. We saw early numbers from Mercury Research that are going to be published soon that also would support that trend that we have been seeing and move more the numbers in the directions that we have been talking about now for a quarter.
Speaker 0
Great, thanks. On gross margin, you talked about your expectations for Q2. Could you maybe give us your thoughts on gross margins in the second half of the year and what would be the drivers either way there?
Speaker 2
Yeah, I'm not going to give more granular guidance on gross margins than what we gave for the overall year. Of course, we are looking forward to seeing a higher mix of Llano-based products in the second half. Of course, we are looking forward to see stronger shipments into the server segment and getting our Bulldozer-based products into the market. This would have a substantial impact on gross margin development in the second half.
Speaker 0
Sure. Can you guys give us your take on sort of the tablet market or any kind of tablet strategy that AMD has, or will you wait until the new CEO comes?
Speaker 2
No, we cannot stop work because of that. I hope you saw that we work hard delivering on our promises. That is an exciting form factor for us. I think we have a lot of IP that is going to play nicely in this field. We are working with multiple OEMs on platforms. Once we are at a good spot to talk about it, we will become more specific on the products and the wins we have.
Speaker 0
Sure. Last question, you gave us an update on the CEO search. Any rough estimate on the timing of that?
Speaker 2
It is in the hand of the board. The management is, that's not a management topic, but I think Bruce, our Chairman, was explicit that finding the right candidate is more important than hitting a specific time window. They are happy with the progress they have been making, and they are actively interviewing.
Speaker 0
Great, thanks.
Speaker 8
Thank you. Our next questioner in queue is Chris Caso with Susquehanna Group. Please go ahead.
Speaker 5
Hi, thank you. I wonder if I could ask a bit about the graphics business going forward and specifically your market share assumption there. One of your competitors in that space has made some claims that they're attempting to gain some shares specifically in the notebook space this year. I wonder if you could talk about your assumptions.
Speaker 1
Certainly. As we mentioned in the first quarter, based on our shipments, it appears actually that we gained some share. We will see here at the end of the month when Mercury Research numbers come out. You also heard that we picked up the Apple business as well, which we're thrilled about. Of course, in light of their success in the marketplace, that's a big boost for our business as well. We will see how it plays out. You know, we feel like we can win in that notebook section in a very interesting way, both with our APUs, which is going to bring an entirely different kind of value into those platforms, as well as our GPUs. In some platforms, they actually play together where we can actually use dual graphics. We feel real good about our position in the overall graphics market.
Speaker 5
OK. If I follow on with a question on servers, understanding that you're talking about late summer for Bulldozer, obviously customers move more slowly in that segment. What's the time frame when we could expect to see some more meaningful share moves or volume moving in that segment, given the amount of time customers need to qualify the product and update their own platforms?
Speaker 2
Yeah, of course, we don't start working with the customers in the summer. We've been engaged with the customers now over a significant period of time, even on the new product. I said before that we expect significant shipments in the second half, third, and especially fourth quarter. While your statement in general is right, there are specific applications where the design in time and lead time is shorter, for example, in the cloud segment.
Speaker 5
OK, thank you.
Speaker 8
Thank you. Our next questioner in queue is Ambrish Srivastava with BMO. Please go ahead.
Speaker 10
Hi, thanks. Just a couple of quick ones. Thomas, what is the, and I didn't catch it, if you did give it, I do apologize, the expectation for graphics units and ASPs in the second quarter? The second question on the quick side is, what should we expect for the APU percent mix as we head into the back half of the year? Lastly, on the server side, is it a matter of just your share is so low that it's going to go up, or architecturally, can you help us understand what your offering will have an advantage over Intel's?
Speaker 2
Do you want to start with server?
Speaker 1
Sure, we'll start with the back half here on the server question. One important point to keep in mind is we have a two-year cadence on our platform where the processor can be used in the same platforms as the prior year. As we move into the Bulldozer-based solutions, they're going to leverage that platform capability. That allows a shorter time to market as well as broader platform availability right out of the gate. Some of the key advantages that we have with Bulldozer certainly are our floating point, our memory bandwidth, and obviously performance per watt, performance per dollar is very compelling with these products. We've been sampling for multiple months now with Bulldozer-based product, and the inception is fantastic. On the former part about graphics, do you want me to take that?
OK, on the graphics portion, we're not going to give you that guidance on the units, but on the ASPs, roughly flat, no big change in our product line going forward there.
Speaker 10
I'm sorry, the graphics business is expected to be flat in the second quarter?
Speaker 1
In terms of ASP, yes.
Speaker 10
OK. What about the third question I had on the APU as a percent of total mix by the end of the year?
Speaker 2
I'm not sure I want to give guidance to that front alert.
Speaker 10
OK, fair enough. Thanks, guys.
Speaker 8
Thank you, sir. Our next questioner in queue is Stacy Rasgon with Sanford Bernstein. Please go ahead.
Speaker 15
Hi, guys. Thanks for taking my question. I just want to revisit the PC unit growth outlook just briefly again. I know you said there may have been some issues potentially with the market data. If you just take it as is, I'm just trying to figure out what's wrong with the data. Is the data just like massively negative versus reality, or are you anticipating a hugely above seasonal Q2? It kind of would take that sort of a profile through the year to get anywhere near double-digit PC shipments per year. I'm just trying to get a better feeling for where you think the discrepancies in the data are, and what do you think kind of Q2 seasonality for PC shipments actually would be in order to get to the kind of unit guidance that you're talking about?
Speaker 2
Yeah, I mean, the guidance we gave for the second quarter still is consistent with where we think the market is going to go for the year. Our own model has been rather stable around this value now for a significant period of time. I think that a significant discussion came about what really was expected for PC shipments to happen in the first quarter, and part of the negative sentiment came from an expectation that year-over-year, the first quarter would be negative. I think the numbers that we start to see now are not reflecting a negative growth year-over-year in the first quarter.
Speaker 15
Sequentially, though, in the first quarter, that data showed it down 10%-13%. You're saying it was actually maybe down 5% sequentially, or was it flat? I mean, literally, the way it is, you would need to have PC growth up 10% in Q2 from the numbers in Q1 to get anywhere near, even close to a double-digit growth. I'm just trying to get a better feeling for where the discrepancies are, because it just seems like a big mess.
Speaker 2
The numbers I have in my head now are year-over-year, so let's see how we align that. I think the negative expectation was that it would be down by -3%, -4% year-over-year on the first quarter. We've seen now research that will, I think, come out more in the next days that now will change that into a probably up 7% range. Our own expectations, I think, are somewhere in the middle between +3% to 4% of shipment growth in the first quarter. I think that is, you know, if you put that on a line and keep our guidance for Q2 in mind, that is a consistent statement.
Speaker 15
You're saying that you think the market data shows a 10% point discrepancy in year-over-year PC unit growth versus the data that was published, up 7% versus down 3%?
Speaker 2
That is what the indication is currently, yes.
Speaker 15
OK. For my follow-up, quick question for you. In terms of the wafer purchase agreement range for this year, $1.1 billion-$1.5 billion, can you give me some feeling what drives the variance in that? Is that just strictly a demand-driven thing? Is that a cost-driven variance? What is actually driving that $400 million range for this year?
Speaker 2
It's a demand-driven statement primarily, and maybe some incentive-driven adjustments to that.
Speaker 15
Can you elaborate on that, incentive-driven? I thought the incentive-driven didn't cut so clinically well.
Speaker 2
Yes, we said on the call when we went through our WSA agreement that we have some arrangements in place that incentivize GlobalFoundries to outperform current plans and targets. Of course, you know, we would benefit in more volume or better yield.
Speaker 15
I thought that came out in 2012. I thought that was the $400 million between the $1.5 billion and $1.9 billion in 2012.
Speaker 2
Some light parts and some flexibility in the first year, but primarily, it's a demand statement.
Speaker 15
Got it. I guess one last question for you. In terms of that wafer supply agreement, my understanding is you give them a fixed wafer forecast, and you're on a die by this year. You go to a cost plus in 2012. What that means is if the yields are bad in 2012, you get fewer good dies, but you don't make the fixed payment. If yields are really good, same way because you get more good dies, so your die cost goes down, but you make the fixed payment. It's sort of a buffer to your, you don't benefit from yield upside in 2012 versus the way you would in 2011. I'm just trying to figure out on top of that. That sort of implies if yields are very good, you get a lot more good dies.
You have to be able to sell all the good dies that you actually receive. Are you baking any kind of, what sort of, I guess, cautions or anything else in your wafer forecast are you giving them to make sure that no matter what the yield situation at GlobalFoundries is in 2012, you'll be able to actually sell everything that you get?
Speaker 2
Yeah, I think this sounds more complicated now than it really is. I think in 2012, we moved more towards a normal foundry relationship. You know, we also have to forecast wafer and product demand with TSMC, right? I mean, that is not significantly different. In normal foundry relationships, you have situations, especially when you ramp up new technologies and new products, that you start with a die-based purchasing agreement and convert to a wafer-based agreement once the yield is up or the product is mature. There's nothing that dramatically different from what we do with GlobalFoundries. We've extended that period of the die-buy for this year to protect, you know, when we originally started, as I said, to protect our downsides with the progress they have been making. It's really more a downside protection at this point in time.
We are not worried at this point in time by overforecasting demand. I think the degree of sophistication with which we handle such processes and have learned to handle such processes is well established within the company.
Speaker 15
Got it. That's helpful. Thank you, guys. Appreciate it.
Speaker 13
Operator, we'll be happy to take two more questions, please.
Speaker 8
Sure thing. Our next questioner in queue is Doug Freedman with Gleacher & Company. Please go ahead.
Speaker 9
Great, thanks, guys, for taking my question. Thomas, if I could ask you to have a little bit of a conversation with yourself since you're wearing two hats of the CEO and CFO, how do we think about you setting the goals for the company financially here? You know, we're running into, this is sort of the first quarter year-on-year where we didn't get earnings growth, and it's largely because of growth in OpEx. Can you discuss with us what actions you would take on the OpEx side to return to sort of a year-on-year earnings growth type of scenario, or how many quarters, how long do we need to search for that revenue growth, and how we should think about you trying to make some of those challenging decisions?
Speaker 2
Yeah, you know, they are quite straightforward, to be very honest. You saw some of the effects already in the first quarter. We provided a guidance of $650 million, and we have been really tough on ourselves and on how we managed operating expenses. This allowed us to come in significantly under the guidance we originally gave. I mentioned that we have a significant amount of cross-functional activity started in the company across our core processes to redesign them for scalability, but especially also for productivity across R&D, across sales and marketing, and also supply chain. We expect significant productivity levers out of those initiatives. I think you also see it already partly reflected in the guidance we gave for Q2. I said it's going to be around $620 million for operating expenses in total, but it includes, as I said, the first couple of 28 μm tapeouts already.
We will be tough on operating expenses, and I think the discipline that we try to instill in the company is something we will keep our grip on.
Speaker 9
Is it possible for you to have revenue growth in excess of your operating spending growth this year, or is this a year that we need to think about you investing a little faster than the revenue growth one?
Speaker 2
I would very much expect that our revenue outgrow our operating expenses.
Speaker 9
Great, thank you.
Speaker 8
Thank you. Our final questioner for today will be Cody Acree with Williams Financial. Please go ahead.
Speaker 4
Thanks, guys. Thanks for fitting me in here. Going back to an earlier question about kind of your comparison of order rates versus what Intel had seen, Intel definitely got some atypical seasonality heading into their product launch, a big burn in the channel, and then a big restocking. What are you seeing as far as typical seasonality of order patterns ahead of Brazos and Llano?
Speaker 2
Even if it's a difficult point, I cannot give you much different arguments, to be very honest. We saw very strong channel demand, partly driven by the products that we started to ramp, also in part restocking of their Q4 burndown. The visibility we have into their inventories does not indicate any alarming things. Their inventories seem to be lean and in line with what you expect for the business at this time in the cycles. We had a couple of emerging markets where our products or our platform-based products were sold out on the shelves. Not much more to add to that.
Speaker 4
I guess, did you see a material burn that now has had a benefit of replacement in the first quarter for Brazos? Do you expect in the June quarter, maybe June and into September, a similar restocking benefit with Llano?
Speaker 2
We are optimistic about our Llano launch.
Speaker 4
Great, thanks, guys. One last thing here. Gross margin trends into the second half as you get the mix of Llano coming in and then Bulldozer uptaking there. What kind of trends could we expect to see from a mix shift?
Speaker 2
I'm not going to get more granular. We gave guidance for the complete year, and with those trends, I think we moved to the high end of the gross margin guidance we gave in the second half.
Speaker 4
All right. Thank you, guys.
Speaker 8
Thank you. That concludes our time for questions and answers, and this also concludes today's program. Attendees, thank you for your participation and have a wonderful day. You may now.


