Good afternoon. My name is Saeed, and I will be your conference facilitator today. At this time, I would like to welcome everyone to Micron Technology's Third Quarter 2011 Financial Release Conference Call. [Operator Instructions] It is now my pleasure to turn the floor over to your host, Mr. Kipp Bedard. Sir, you may begin.

Thank you. Good afternoon, and welcome to Micron Technology's Third Quarter 2011 Financial Release Conference Call. On the call today is Steve Appleton, Chairman and CEO; Mark Durcan, President and Chief Operating Officer; Ron Foster, Chief Financial Officer and Vice President of Finance; and of course, Mark Adams, Vice President of Worldwide Sales. This conference call, including audio and slides, is also available on Micron's website at micron.com. If you have not had an opportunity to review the third quarter 2011 financial press release, this is available on our website, again at micron.com. Our call will be approximately 60 minutes in length. There will be an audio replay of this call, accessed by dialing (706) 645-9291 with the confirmation code of 76907886. This replay will run through Thursday, June 30, 2011 at 5:30 p.m. Mountain Time. A webcast replay will be available on the company's website until June of 2012. We encourage you to monitor our website at micron.com throughout the quarter for the most current information on the company, including information on the various financial conferences that we will be attending. Please note the following Safe Harbor statement. I think what I'll do is go ahead and read it for you. During the course of this call, we may make projections or other forward-looking statements regarding future events or the future financial performance of the company and the industry. We wish to caution you that such statements are predictions and that actual events or results may differ materially. We refer you to the documents the company files on a consolidated basis from time to time with the Securities and Exchange Commission; specifically, the company's most recent Form 10-K and Form 10-Q. These documents contain and identify important factors that could cause the actual results for the company on a consolidated basis to differ materially from those contained in our projections or forward-looking statements. These certain factors can be found on the company's website. Although we believe that the expectations reflected in the forward-looking statements are reasonable, we cannot guarantee future results, levels of activity, performance or achievements. Also, we are under no duty to update any of these forward-looking statements after the date of the presentation to conform these statements to the actual results. I would like to now turn the call over to Mr. Steve Appleton.

Thanks, Kipp, and I also want to thank everybody for joining us today. I thought I'd start with a very brief update on Japan on the supply side, in other words, silicon and other materials like specialty gases and targets. I just want to confirm that there were no interruptions, and I think we're past the difficult period. I think it's worthwhile to comment on how good the Japanese companies were in executing their recovery strategy, and we really don't see any issues with it moving forward. On the customer side, I think similar for the most part, but we do see some fallout from the challenges the country is facing in recovering. In particular, this impacted our ESG revenues, and it was about 5% for the quarter. On the operations and technology, I think we had a number of achievements that were worth noting. Clearly, from the media, you would have seen we sold the Micron Japan wafer fab to TowerJazz, and Ron is going to have some more comments on that in a little bit of a the detail during his segment. IMFS continues to perform very well. Again, Ron will also add some comment on that. But let me just say that we're shipping to customers qualified product from the facility. I will note that, as is typical with us in the past, when we bring on a new facility, it's focused on pretty much a singular process known as singular product and all that coming out of that facility as MLC. Our 20-nanometer NAND still looks good for ramp in the second half of the calendar year, so we're pretty pleased with that. And then let me add that we are sampling 30-nanometer DRAM from our Virginia facility, and that technology will spread to another DRAM…

Thanks, Steve. The company's third quarter of fiscal 2011 ended on June 2. As usual, we provide a schedule containing certain key results for the quarter as well as guidance for certain metrics for the next quarter. That material is presented on a few slides that follow as well as on our website. As was previously announced, we sold our Japan fab operations to TowerJazz, as Steve mentioned, in the third quarter. We will continue to source wafers from that operation through a wafer supply agreement which extends over approximately the next 3 years. The sales transaction resulted in a gain in other operating income of $54 million in the third quarter. In addition, there was a tax provision of $74 million stemming from the gain and the write-off of deferred tax assets, netting to a $20 million loss for the total transaction in the quarter. Other operating income also includes a gain of $35 million from the final installment from Samsung under the cross-license agreement we entered into earlier this year. For the whole full fiscal year, we recognized all of the $275 million gain from the cross-license. Recall that there is a withholding tax on these payments that is reflected in the income tax provision of $6 million in the third quarter and $45 million year-to-date. The total tax provision of $104 million in the third quarter includes the effects of the sale to TowerJazz and the Samsung cross-license fee in addition to our regional taxes in other jurisdictions. As you see from the chart, we expect the tax provision in the fourth quarter to be back to more normal levels. IMFS continues to ramp ahead of schedule. As Steve mentioned, we began to charge most of the manufacturing costs in the inventory as we have qualified the…

Thanks, Ron, and with that, we'd like to take questions from callers. [Operator Instructions] With that, let's please open up the line.

[Operator Instructions] Our first question comes from Uche Orji from UBS.

So let me start by asking you about the DRAM expectation bit for the coming quarter. So one of the expectations we made [ph], your gross margins were [indiscernible] revenues fell short of expectations. Part of that is obviously mix. As we look into the mix for the coming quarter, what should drive the outlook to mid-single-digit? Should we expect that mix to also be positive for gross margin?

Yes, around specialty memory, those obviously have a positive, more positive impact to gross margin. Was that the bit of your question?

That's correct, yes.

Yes. Specialty memory continues to be our highest-gross-margin products.

Right. But in terms of how we expect -- the expectation for the coming quarter, are we expecting within that guidance, what is your view of what should happen within the PC markets? And how do you expect the PC market demand for DRAM to pan out?

This is Mark Adams responding to that. I think in general, we've all seen a softening of the desktop, notebook, PC climate. Partially offset by some growth around tablets, more helpful on the NAND front, obviously, than the DRAM front. But it's been pretty hard to kind of see beyond Q3 calendar year in terms of what the demand picture looks like as some of the bigger OEMs have kind of come out and made their forecast, the analysts certainly have made their forecasts, and overall, the PC environment is weak at this point, and hard for us to call too much further out in the future.

Just one more question. In terms of severance and networking, that's one area where we've seen strength, especially in your comments earlier. How do you expect that to trend within your guidance and for the rest of the year? And we've seen strength from servers for the past few quarters. And how much more runway do you think we can get from that?

Well, that's a good point, because my -- I should have qualified my comments. The PC comments are primarily around the consumer environment. The corporate refresh is still in process and going fairly well in terms of the NAND side for our more enterprise-corporate type products. And servers certainly is one of the products that Kipp was just referring to as being a better opportunity for us and more stable for us. And we think that continues not just around the servers, but even our networking product portfolio and some of our embedded products, more in the commercial applications. So the softening we're talking about is primarily around the consumer environment, and desktop, notebook. Even Steve made reference to the value lines market consumer -- value line mobile market. But the enterprise applications, especially DRAM, continues to be pretty strong for us, and we see no reason that's going to change in the coming quarter.

And just lastly, IMFS. With the plus 4 4Q [ph] targets, you talked about that ramping. Is it possible for us to know what the stats [ph] will be for the 4Q?

We've given you the bogey of 60k wafers per month, and the more current guidance is all baked into our bit growth guidance for Q-to-Q.

Our next question comes from James Schneider from Goldman Sachs.

Could you talk about the inventory levels at customers you mentioned with respect to DRAM, and when you expect those to kind of be back on normal levels? And within that, can you talk about what's the linearity of your DRAM sales were this quarter?

Sure. I think I'd break the inventory answer into 2 parts: the OEM component to that and the channel. The OEM component, I think, will probably be managed fairly well throughout this calendar quarter. So I think that -- I'm sorry, what I really referred to is calendar quarter Q3, our Q4. I think the OEMs have got a good handle on that, because I think that's already in process. The channel fees, as Steve mentioned, are somewhere in the 5-week range, 4 to 5 weeks. And I think that really was a reaction to Japan and some early accumulation on concerns over supply to the -- in the spot market and the aftermarket module market. So I think that is more of something that gets to be managed over time here over the next coming 90 days or so. But I feel that the OEMs are in better control of that and already actively working through that. The channel folks are getting into it now, and I think that's what's causing the differentiation between the spot market pricing and the OEM contract pricing.

Great. Thanks. That's helpful. And then with respect to the mobile and server DRAM spaces, can you tell me whether you see any of your competitors moving capacity to those areas? And what kind of impact do you expect on pricing, either what you saw at the last quarter or what you expect over the next couple?

Sure. I think server is -- so we need to just run over and do that on a quarter-to-quarter basis, there's a lot of qualification and supply variables that go into actually getting into the server business. I think on the mobile side, a lot of people have talked about that, and we've seen one of our competitors talk about going back to the commodity business. So that's a harder one to gauge. But I also think that comes with the challenges of not just redirecting capacity, but getting qualified at some of these bigger OEMs. So it's not a perishable capacity choice. You have to kind of be in the business and be committed to it over time. I think the comments might be more fluff than reality in the short term.

Understand. And one last one for me. Can you talk about where PC DRAM pricing is relative to your own cash cost today?

I'm not aware of anybody that gives out their cash costs, sorry.

Okay. Fair enough.

Our next question comes from Vijay Rakesh from Sterne Agee. Vijay Rakesh - Sterne Agee & Leach Inc.: Yes, I guess just on the DRAM side, obviously, it looks like you had some, at least some issues on your data site [ph]. Is that a result of, you think, that should help you on the year margin in the output side as you go out?

Vijay, you were breaking up a little bit. Could you repeat the first part of your question? Vijay Rakesh - Sterne Agee & Leach Inc.: Yes. In the May quarter, you guys probably had [ph] some impact from Inotera. Is that a real issue and [indiscernible] should help you going forward in the margin end of business [ph]?

This is Mark Durcan. On Inotera, we continue to make progress there not only on the 50-nanometer yields, but also on the 42-nanometer ramp. And we're looking for an improving situation as they move forward quarter-over-quarter. Vijay Rakesh - Sterne Agee & Leach Inc.: And when I look at your mix in DRAM now, what's the mix between PC, server and mobile? And where do you see that end of the year?

Today, personal systems are about 18%, and that looks like that will come in about where we will average for the year. As Ron mentioned, we're getting awful close to having NAND surpass total DRAM revenues for the first time in our history as well. Vijay Rakesh - Sterne Agee & Leach Inc.: Got it. Okay, and it looks like your CapEx goes down pretty nicely next year. What are the puts and takes to that $2 billion that you gear out for next year? Is there a chance that comes back more the last 2 years kind of in the $600 million range CapEx?

Well, I think that our CapEx next year and the variability of it will really be based upon how we view the market. So our CapEx for our fiscal year, keep in mind that we still have IMFS we're paying for as the equipment's been delivered or under qualification, there's somewhat of a lag delay. So a lot of that CapEx is associated with IMFS, of which we're already committed to, and ramping, and the rest of it is just sprinkled throughout the rest of the network. That's not likely to change much unless for some reason, we feel like we ought to try to get more production out of IMFS, in which case we would spend more to do that. But that's all market-dependent at this point. Vijay Rakesh - Sterne Agee & Leach Inc.: So all the CapEx is for NAND and IMFS, nothing in -- mostly all for NAND, not for DRAM, right?

No, that's not true. I didn't say that. I said a lot of it was for IMFS and NAND, and then the rest of it is sprinkled throughout the rest of the network. Remember, we got many fabs. It doesn't take a whole lot per fab to add up to a total of $2 billion. Vijay Rakesh - Sterne Agee & Leach Inc.: Got it.

Our next question comes from Daniel Amir from Lazard.

How could you -- you view the NAND side here as you ramp up IMFS in terms of the impact on gross margins here. So how should we look at kind of the margin profile of that business here in the next 6 to 9 months?

Danny, in terms of gross margin ranking, trade NAND is one of the better ones we have. It's about our second-highest gross margin. So to the extent that we're shipping more of those bits, that will certainly help on the gross margin side.

So as you -- and IMFS, I'm assuming, will just have a positive impact as the volume increases there?

There's a couple of moving pieces, so you have to be a little bit careful. Obviously, we're adding more bits. We get more of that output, so that's good. There's obviously an ASP component to it, which we don't want to get into try and predicting today. But steady-state today, it's our second-highest gross margin product.

Okay. And then the other question is kind of on the wireless side. I mean, you made some commentary that it is a little weak right now. Kind of what do you predict here in terms of kind of the next quarter? I mean, do you see seasonality starting to come back, helping the wireless segment? Or do you have the same visibility in the wireless side as you do in the PC side?

I think our visibility in the wireless side is pretty good. I think part of what's going on in the wireless space is a little bit of a shift in the industry structure, with market share shifting from company A to company B. I think that's contributing a little bit to the fogginess in terms of that business. Having said that, I would think with some newer models coming out, and it is a bit of a seasonal uptick, that it'd be reasonable to assume that there could be an improvement in terms of the market conditions around wireless coming into back-to-school and the holiday. And remember, our comments were around the segment of the wireless market. The smartphone market continues to be very strong, and I also think that might be something that contributes to some higher-end gross for the back half of the year.

Our next question comes from Kevin Cassidy from Stifel, Nicolaus. Kevin Cassidy - Stifel, Nicolaus & Co., Inc.: Thanks for taking my question. Can you give some of the puts and takes on the trade NAND for next quarter being down low double-digits? Is it less demand in the SSDs or is that a steady-state? Or maybe just give a description of what's happening there.

I think -- let me make sure you're reading the right lines there. We look at bit production up double-digits, and currently, the ASP is down low doubles. Is that what you're... Kevin Cassidy - Stifel, Nicolaus & Co., Inc.: Right. Yes, I was asking about the -- it was up 15% in the third quarter, and now you expect it to be down low double-digits. So I was wondering...

That's correct, Kevin. As Ron mentioned in his comments, we're going to see -- we're obviously executing pretty well at IMFS, and those will be more discrete units here in the short term. So expect more OEM-type business during the short-term quarter. So it's mix-related. Kevin Cassidy - Stifel, Nicolaus & Co., Inc.: Okay. It is mix-related. And are you allowed to say what -- or do you know what Intel's mix was going to be?

No. Kevin Cassidy - Stifel, Nicolaus & Co., Inc.: Okay. And maybe on the DRAM side, can you say what your percentage is of 2-gigabit DRAM now?

Kevin, I don't have that breakout in front of me. But I'm happy to follow up with you. It's something that we present from time to time, and so I'll get it to you.

Our next question comes from Daniel Berenbaum from MKM Partners. Daniel Berenbaum - Caris & Company: Can we come back to DRAM shipments in the quarter? Just kind of running through the math, it looks like your DRAM bit shipments were actually down somewhere in the 10% range. Is that somewhere in the right ballpark? And can you help walk me through why they were down so much?

Yes, sure. This is Mark Adams again. The DRAM shipment was more of a carryover for us, especially around some of our OEM customers, for Inotera output. They've got us timely in the quarter, but we could not convert it into modules in time to get to our OEM agreement. Secondly, towards the end of our quarter, we saw a pretty wide gap between the spot markets and the OEM contract market and certainly didn't want to play into the game of end-of-quarter negotiations when the spot market's driving price down further. So we chose to keep some of that inventory for our larger OEMs going into the Q4 fiscal. Daniel Berenbaum - Caris & Company: So then some of the increase in the inventory in your balance sheet was due to holding onto some of that NAND -- sorry, holding on to some of that DRAM, sorry?

Absolutely. I mean, absolutely the case. And again, if you track the spot market pricing over the last 2 to 3 weeks of our quarter, it started to get pretty volatile. And for us, given our OEM demand curve for Micron memory, it was a pretty logical choice not to play in the spot market game at the end of the quarter. Daniel Berenbaum - Caris & Company: Okay. And then can you comment on the gross margin in the Wireless and Embedded Solutions business? How did those trend? Either -- if you're willing to give absolute numbers, that would be great. If not, how did those trend from last quarter?

We're not going to give you the actual numbers, but they did come down a little bit in those segments. Daniel Berenbaum - Caris & Company: I'm sorry, gross margin came down in those segments?

Yes. Daniel Berenbaum - Caris & Company: And are we past sort of all of the purchase accounting from the Numonyx acquisition in terms of gross margin?

No. There was an impact, a little larger than we expected, actually. It's about a $20 million impact in this quarter that we just reported, and that's about equal to what's left. We do anticipate the rest of that will come in over the next couple of years. Daniel Berenbaum - Caris & Company: You had a $20 million negative impact in the quarter, okay. And if then the rest will come through, you would then expect all of the Numonyx accounting treatment to be done sort of by your fiscal Q1 of next year?

It's all related to the certain parts that did get written up in the inventory, and so some of those are legacy parts that will sell out over the next year, even. But it won't all clear up in one quarter. Daniel Berenbaum - Caris & Company: Okay. And then while we're sort of on those kind of charges, were there any other inventory adjustments in the quarter?

No.

Our next question comes from David Wong from Wells Fargo.

On your summary sheet, you talked about your trade NAND cost per bit being up approximately 12% in the quarter. To what extent do startup charges play into that?

This is Ron. We had startup charges, as I mentioned, last quarter, and we had some reduced level of startup charges overall this quarter. In total, for the company, it's about $36 million in the second quarter, dropped to about $20 million in this most recent quarter. Some of those costs are now being put into inventory and will go out with the shipments of IMFS product in the coming quarters.

So that then wasn't the primary driver of the cost per bit for trade NAND being up.

No.

So was there some other factor?

Yes. If you look at it, the ASPs were up about 15%, and our costs were up about 12% in the quarter and the impact was related to mix, as Steve commented, specifically in the area of higher SLC mix. SLC is a higher-margin product. It's also higher cost. And so that affects the top line, but we end up with more margin, as Steve mentioned in his comments, going forward. Also, there was some SSD growth that he mentioned that helps drive up both the pricing and the cost a little bit.

Right. I see. Then on the DRAM side, what was the pattern of your contract pricing through the quarter? Was contract pricing rising or falling as you progressed through the May quarter?

I would say that early in the first half to 2/3 of the quarter, the contract pricing was rising. And I'd say it kind of plateau-ed and flattened out towards the end of the quarter.

So if it was rising through the quarter, why does your ASP estimate for the current quarter, the quarter today, why is it down so much? Was there a huge plunge at the beginning of the current quarter? Or is there some other factor that's playing into your expectation for DRAM ASPs looking like it's going to be down high single-digits?

Well, there's 2 things there. The guidance we're giving is quarter year-to-date. So I'll come back to that in a second. But your question earlier was about Q3 and the trend during Q3. And I stopped short by saying that it was flat during the last month, which was May. Our June contract pricing is down slightly, and then we also have some mix effects there as well, causing the net impact of the quarter-to-date guidance we're giving you.

Our next question comes from Doug Freedman from Gleacher. Doug Freedman - Gleacher & Company, Inc.: Thanks for taking my question. Can I question you on what you think the bit demand for the full year is now looking like it's going to be, given where we are, halfway through the year now, on both DRAM and NAND?

Let me start, Doug, with the bit supply, and then Mark can talk you through maybe some of the different segments. We look at DRAM somewhere in that 40% to 45% range. We see NAND in that 70% to 80% supply range. Duly noted that on recent equipment conference calls, they have referenced equipment pushouts. So it does -- in our analysis, it does look like we've probably come through our or are in the highest sequential bit growth quarters now. And that going through the rest of the year, especially in DRAM, they tend to tail off pretty good. We're fairly -- looking at the NAND numbers in terms of supply, looks like you're going to run kind of low double-digits here for the next couple of quarters. And with that, I'll let Mark maybe talk you through some of the segment demand things that he sees.

Yes, it's pretty consistent with earlier comments. I think it's -- the consumer business overall tied to kind of the macroeconomic environment, it just seems a little bit hard to predict and a little bit softer than we might have thought going into July. But certainly, you've got some seasonality effect in there as well. So around DRAMs, for example, the PC business, to earlier comments, is weaker than we had hoped. And right now, we don't have line of sight on anything in terms of a recovery back to an upward-priced environment in DRAM on the consumer side, especially DRAM and our better margin product areas. We continue to remain pretty bullish, because the demand there has remained strong. And I referenced this earlier, but the server products and the server business overall is pretty good. The growth there on units is still in line with, I think, what the analysts kind of suggested we would be for 2011. Networking. Again, networking products remain strong for us as well. And our AIMM business, which is part of the Embedded business unit, is very stable also. So it's not that we can make one conclusive comment about the overall memory landscape. I think the commodity consumer-oriented businesses are seeing some softness of demand side for DRAM, and the specialty businesses around enterprise and corporate are pretty good. When you look at NAND, we're not seeing anything that would change our perspective on a pretty strong demand environment for NAND. SSDs have been a continually good story for us at Micron. Steve referenced earlier slightly below 40% quarter-over-quarter growth, new technology launches, and good reviews for us and customer qualifications on enterprise performance drives. So SSD could be a strong place for us. If you look at the smartphone piece of the business, where a lot of that NAND is consumed in the mobile market, still very strong. And the tablet business is, again, is another good environment for us for NAND. So we think the drivers on NAND are pretty consistently strong through the back half of the year, especially [indiscernible]. And I don't think anything's going to change that going through the holidays, minus some significantly different environment as far as the economy. Doug Freedman - Gleacher & Company, Inc.: If I could, for my follow-up, focus in on the mobile DRAM side of the business and your aspirations to ship phase change by the end of the year. Can you give us an update on where you stand and what the mobile DRAM market is bringing to you guys in terms of an opportunity?

Yes, this is Mark Durcan. We continue to make good technical progress on phase change, both from a liability perspective as well as yield improvement. The actual uptake in the market is going to be tough to call right now, but we're definitely moving more resources towards both products to make sure we're in a position to deliver for that market space.

Our next question comes from Ryan Goodman from CLSA.

Question on servers. We've started to hear some chatter that server ASPs were actually coming under more pressure than normal over the past quarter. I know you sounded actually pretty bullish kind of looking out through the remainder of the year. Just curious if you saw anything like that. I mean, the premiums are still there. But are you feeling any more pressure? And any thoughts on the impact of OMNI [ph] coming on in the second half?

I don't think that, that pressure has manifested itself back down to us. I do think, though, that when you look at what's driving the application here on the server products itself in terms of the markets, in terms of cloud and some of the other applications, you are seeing a broader value line of servers take share in the market. And that's the unit volume shifting that way and could potentially drive that. We're not seeing that implicate our server pricing too much on the memory side of the house, but I think that there seems to be a higher volume of distributed server technology in the market, and I think to that extent that you could see that being potentially commoditized downstream. But we don't -- we haven't seen any impact internally.

Okay. And then just one follow-up on the NAND side of the equation. In the smartphone market, just curious about your thoughts on content trends at the high end of the market. If you think -- how do you think the demand picture comes together if there isn't a lot of content growth at the high end of the market? And also, just curious about, for Micron specifically, how you guys are doing in terms of gaining some traction in that area of the smartphone market.

I think, overall, we're pleased with that. If your question was around if the content per smartphone on the high end doesn't materialize, is that the question?

Yes. If there's not a lot of growth in content at the high end, are you still comfortable with the overall demand picture for 2011 and then going into 2012?

Yes, I think absolutely. Just in terms of unit growth alone, I think there's an opportunity there for us as well as if you look at the evolution of what people are using the smartphones with, we feel pretty confident that the content piece will evolve favorably.

And then any update you can just give on how Micron specifically is doing in terms of getting traction in the high end? It seems like a lot of the embedded progress historically has been more in the consumer areas. Just curious how you guys are doing on smartphone.

To the extent that when we look at our NAND portfolio, to the extent that we're choosing to be engaged in that market, which is pretty strategic to us, we feel pretty good about it. We feel, needless to say from a cost perspective and technology perspective, it's a good place for us to invest. And our customer demand continues to remain strong.

Our next question comes from Bill Dezellem from Tieton Capital Management.

Relative to the SSD strength that you're experiencing, would you please provide some more details and color behind what's taking place there, and including who are you finding the primary adopters of SSD to be at this point in the adoption cycle?

Sure. Our experience has been that the OEMs are really trying to adopt us to establish a separate class of notebook users, if you will, and trying to add some higher-margin opportunities themselves in selling up to executives and consumer power users. The volumes are pretty significant around that. And you think about the sudden growth in terms of enterprise applications actually manifesting themselves up in terms of revenue in the category, those are still pretty much the 2 drivers, I would say, are the OEMs driving power SKUs for the enterprise executives as well as consumer power users. And then in the enterprise sector, I'd say the shipments basically doubled quarter-over-quarter in that sector, and we continue to see long-term growth there for us through the next 2 to 3 years.

And then following that, where are we at today with the cost differential versus the traditional hard drives?

I still think from a cost per gigabyte, we're not quite close to yet equality. I think what you're seeing from the shipment and, both in the client space and the enterprise space, would suggest that even though we're not at cost parity, we're not really even close yet on a dollar-per-gigabyte basis, the benefits of the SSD and total cost of ownership allow us to compete pretty favorably.

And one additional question. What is the rate of cost reduction that's taking place with SSDs versus hard drives?

Well, Bill, it's a different discussion point because we can, on the discrete device level, bring costs pretty down. But what you're really asking is the cost to our customers, and that's more market-driven. So you have to be a little more specific.

Your point's well taken, and I'll ponder the question more offline.

Our next question comes from Glen Yeung from Citi.

There's been some discussion about Micron's potential partnership with Elpida. What are your thoughts around balancing consolidation and potential needs for additional capacity?

Well, first of all, we don't have any partnership with Elpida at this point. And I think if you look at the entire industry picture, obviously, there are a few companies that are still struggling. The Taiwanese are struggling a lot, and then you draw Elpida into the mix. Of course, in terms of total size, they are also one of the smaller companies in the memory business. And I think over time, most of that will get resolved. But in the short term, typically, what we have to see is some kind of crisis in the industry and the more of a crisis, the quicker the resolution of consolidation industry. And whether we're in that time period or not, right now, I think it will depend on how long the market being down lasts. But short of that, we'll just have to see how it plays out.

Thank you, Steve. And as you ramp IMFS, is there a preferred mix of captive versus non-captive as Micron's ownership and output ownership of NAND from IMFS grow?

Well, is your question do we have a preferred use of the product in our own product downstream, as opposed to selling to the merchant market, or is it something different?

I guess the proportion of captive versus non-captive. As the ownership moves towards 80%, is there a...

I see, I see. Well, I think you can correctly surmise that the ability to ship our NAND product and downstream products, in other words S -- i.e., SSDs and other things, just has there buying proposition and more margin for us. So to the extent that we can do that, we'll be focused there. But I would say that our objective, in particular as our percentage of the total, as Ron mentioned, our percentage of the total output of the joint venture comes to Micron, our objective really is, though, to still serve all the markets in some form, and obviously, we'll direct that capacity to the best return for the company.

And just one last one for Ron. With the Samsung, the tax issues going away next quarter, what was the guidance for taxes beyond next quarter after the impact of Samsung taken away?

Well, yes, it's more normal rates of $15 million to $25 million kind of range going forward on taxes. That's our normal rate that we pay for the other regional jurisdictions.

Our next question comes from Bob Gujavarty from Deutsche Bank.

I have one question, maybe a little longer term. But as some of the notebook products kind of mimic more of the tablet functionality, I've talked to a few people that said they have to use mobile DRAM instead of commodity DRAM in kind of notebook, PC notebooks. What do you guys think of that as a longer-term opportunity to kind of move ASP up and then potentially move your dollars per PC system up?

You know, it's funny. We tend to think slightly the opposite of how you stated it, which is that the tablet not only serves in a mobile application environment, but starts to mimic some of the functions of PC. And I think there's a fair point that, that might be an opportunity for us, but it is on kind of a partner-by-partner basis in terms of how they look at their architecture and their system.

Our next question comes from Shawn Webster from Macquarie.

I have a question about the impact in Japan. So you mentioned that -- was that 5% of ESG or 5% of your overall revenues you think were impacted? And the inventory in the channel that accumulated. Is that all accumulated right after the earthquake? Or is that something that you think that was more -- there's maybe a buildup before, and it just spiked up after the earthquake, and it's come down a little bit after that?

Well, in terms of the Japanese impact to our revenues, I had mentioned that about 5% of our Embedded group was affected, primarily the entertainment industry over there, as you might expect. In terms of the inventory channel buildup, Mark can add a comment here if he would, but essentially, that happened after we had the Japanese event. And I would say to the extent that some of it did go, it's clearly trying to make sure that they were covered. But in addition to that, I would just add that, as Mark already did, that a couple of the markets were weaker than maybe people were anticipating, so they had some inventory that they accumulated.

Great. Thanks for that. And also, if you could talk about your DRAM supply. If you were to take Inotera's impact out of both Q2 and Q3, do you think that your supply in terms of their growth was roughly within -- in line with expectations, or a little bit below or a little bit above?

It might be a little bit below. As we have discussed in the past, we've been trying to enrich the mix, and generally, those are lower-density devices. And so you don't get quite the bit efficiency per wafer, for example.

Our next question comes from John Pitzer from Crédit Suisse. John Pitzer - Crédit Suisse AG: A couple of questions. Sorry if I missed it, but the server DRAM as a percent of overall DRAM, where did it stand in the quarter? And I guess, what kind of target should we think about as you exit the calendar year?

John, I've got it broken out to total revenues, and we ran server about 20%. So pretty good business for us. John Pitzer - Crédit Suisse AG: And I guess, Kipp, where do you think that could go as you guys continue to try to mix up the DRAM business?

I might have to give that one to Mark to do a little more in-depth discussion around how he sees demand. We know that bit growth per system will be up about 60% to 65%. We know units in general, when inclusive of all the OEMs that are building servers themselves, are up double-digits. And Mark, if you'd like to add anything to that?

No, it's consistent with the demand picture that we get for our products. And I'd say in the server business, we're mildly constrained around the capacity of those parts that could go there. We're trying to ramp that up in the coming quarter. John Pitzer - Crédit Suisse AG: And then guys, in core PC DRAM business bits per box today, is price elasticity in that market done or we just dependent upon unit growth? Or how should we think about that?

Well, it's interesting. If you went back a couple of years, when the ASPs of DRAM were significantly higher, we went through this destocking [ph] period due to the supply and demand imbalance. Since then, obviously, it's recovered pretty nicely, and we were just short of 4 gigabytes per unit in the desktop notebook environment. And it's projected to go over that threshold, which I think is the first time, at least that I can remember, there have been 4 gigabytes per unit. So our sense is that continues to remain progressively increasing and then favorable. John Pitzer - Crédit Suisse AG: And then guys, my last question, and I appreciate this. There's probably not a lot you can say around this, but with the Rambus trial now moving forward, I'm kind of curious how we think about SG&A. And are there any milestones or timelines we should think about relative to that event?

Well, clearly, the Rambus trial started. It's a multi-month trial, and we'll continue to accrue litigation costs, as we have in the past. And that's really probably all we can say on the topic, because it's in flight. John Pitzer - Crédit Suisse AG: But Steve, now that it could move to trial, we shouldn't expect to step up in increments? Or it's already embedded in the guidance you gave for the quarter?

Well, that's already embedded in the guidance we gave for the quarter. The length of trials are unpredictable, and as a result, that doesn't mean that we'll have some additional accrued litigation costs the longer it goes. But we'll just have to deal with that realtime as it's happening.

Our next question comes from Atif Malik from Morgan Stanley.

If I look at your guidance for NAND trade cost reduction last quarter, low to mid-single-digits and you ended up doing up 12%, are these decisions being made on the fly in the quarter to change the mix? And if you can give the percentage of SLC, MLC and TLC mix right now and what it will be by the end of the year.

So I'll take the first part about decisions being made on the fly during the quarter and change mix. The answer is, of course, we're trying to optimize margins in the parts. And as we see opportunity, we'll do that. Now also, I think in correlation, make the comment that sometimes it's easy to change the mix, and sometimes it's more difficult to change the mix, depending on the product and the category and so forth. And we also have customer commitments that we have to make sure we make. So all of that goes into the hopper. But we're cautiously looking at it. And for the next part, I'll turn it back over to Kipp.

Yes, in terms of wafers probe, we saw a 26% increase in -- or 26% of our NAND wafers were processed in SLC in Q2. And of course, as those sell out, that would be into fiscal Q3, and that's why you saw the pretty nice increase in cost per bit. But again, referencing both what Ron and Steve said before, that also gave us the very nice margin lift and ASP lift as well. It's worth it for us to put as many wafers in that direction as possible.

Got it. And I agree, it's good for profitability. And what should we think about the kind of normalized SLC mix for the year? Higher than this level, about 26% for the year?

This looks pretty flattish. As we mentioned before, IMFS is primarily ramping right now on MLC. So don't think that the wafers aren't growing in SLC. But as a percentage of the total, they're about flat in that 20% range.

Okay. And for the second half of the year, are there any underutilization charges at Inotera that could come up if the demand remains lower? Or you think you can still use Inotera for server demand and other kind of non-PC DRAM demand?

This is Ron. Your underutilization question was about IMFS, in particular?

For Inotera. I mean, are there any implications -- I mean, can Inotera load up the factory? Is this [ph] 100% utilization right now in Inotera?

Yes, Inotera's ramped.

Yes, they're ramped in wafers. They're still working on coming up the yield curve, et cetera, and get more good bit yields out for production. So no, we don't get specific underutilization charge. We did when it was shut down in 2009. But now we're just paying on a wafer cost basis and, as I've mentioned before, there's a cost basis and then there's a profit-sharing component that gets spread back.

Thank you. And with that, looks like we're out of time. We'd like to thank everyone for participating on the call today. If you will please bear with me, I need to repeat the Safe Harbor protection language. During the course of this call, we may have made forward-looking statements regarding the company and the industry. These particular forward-looking statements and all other statements that may have been made on this call that are not historical facts are subject to a number of risks and uncertainties, and actual results may differ materially. For information on the important factors that may cause actual results to differ materially, please refer to our filings with the SEC, including the company's most recent 10-Q and 10-K. Thank you.

This concludes today's Micron Technology's Third Quarter 2011 Financial Release Conference Call. You may now disconnect.