Kipp A. Bedard - Vice President of Investor Relations Ronald C. Foster - Chief Financial Officer, Principal Accounting Officer and Vice President of Finance D. Mark Durcan - President and Chief Operating Officer Steven R. Appleton - Chairman and Chief Executive Officer Mark W. Adams - Vice President of Worldwide Sales

John Pitzer - Crédit Suisse AG, Research Division Alex Gauna - JMP Securities LLC, Research Division Glen Yeung - Citigroup Inc, Research Division Harlan Sur - JP Morgan Chase & Co, Research Division William Dezellem - Tieton Capital Management Christopher J. Muse - Barclays Capital, Research Division David M. Wong - Wells Fargo Securities, LLC, Research Division James Schneider - Goldman Sachs Group Inc., Research Division Uche X. Orji - UBS Investment Bank, Research Division Betsy Van Hees - Wedbush Securities Inc., Research Division Ryan Goodman - CLSA Asia-Pacific Markets, Research Division Hans C. Mosesmann - Raymond James & Associates, Inc., Research Division Shawn R. Webster - Macquarie Research Bobby Gujavarty - Deutsche Bank AG, Research Division Unknown Analyst - Krishna Shankar - ThinkEquity LLC, Research Division

Good afternoon my name is Yuwi, and I will be your conference facilitator today. At this time, I'd like to welcome everyone to Micron Technology's Fourth Quarter and Fiscal Year End 2011 Financial Release Conference Call. [Operator Instructions] And now it's my pleasure to turn the conference over to your host, Kipp Bedard. Sir, you may begin your conference.

Thank you very much, and welcome to Micron Technology's Fourth Quarter and Fiscal Year End 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 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 (sic) [fourth] quarter 2011 financial press release, it is available on our website at micron.com. Our call will be approximately 60 minutes in length. There will be an audio replay of this call, accessed by dialing (404) 537-3406, with the confirmation code of 11557620. This replay will run through Thursday, October 6, 2011, at 5:30 p.m. Mountain Time. A webcast replay will be available on the company's website until September 2012. We encourage you to monitor our website, again 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. During the course of this meeting, 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 Investor Relations section of Micron'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. We are under no duty to update any of the forward-looking statements after the date of the presentation to conform these statements to actual results. And now I'd like to turn the call over to Mr. Steve Appleton. Steve?

Thank you, Kipp. I want to make a few comments, and then Ron Foster will make a few comments. And as Kipp said, we'll open it up for questions. It may seem a lot, but I thought I'd just do a quick refresh where we started the last quarter, which was we're all concerned about Japan. At the time we forecasted that we didn't think there would be any disruptions of significance in our supply chain, and in fact that turns out to be the case. And then it all went pretty smoothly, just as a point worth noting. On the operations and technology front, we feel like we had a pretty good quarter. As you can see from some of the data we posted from our bit growth and our cost reductions. On the execution side, we had a good operational quarter. In conjunction with that, our 20-nanometer NAND is ramping, continues to go well. But obviously, we're continuously adjusting the mix to take advantage of higher margin opportunities. IMFS is on track and also continues to perform pretty well. We're on track to reach wafer start capacity sometime in the late fall. And we continue to make investments in R&D as evidenced by our new research facility that will be ready -- it will be ready for tool installs in the beginning of calendar 2012. So I might add that the facility itself will be 40 -- 50-millimeter capable, but obviously we're very focused on 10 millimeter at this time. One other comment on IMFS, I think it's worth noting is that in our Q4, it accounted for about 25% of our trade NAND revenues, so that's obviously going to be pretty significant for us. On the CapEx front, we came in at about where we thought we…

Thanks, Steve. The company's 2011 fiscal year ended on September 1. As usual, we provided a schedule containing certain key results for the quarter, as well as certain guidance for the next quarter. That material is presented on a few slides that follow as well as on our website. The fourth quarter was financially challenging for us in our industry, which is reflected by the company's net loss for the quarter. Although we were able to meet or exceed all of our output and cost reduction goals in this quarter, DRAM price decline put the overall business in a loss position. However, as Steve mentioned, in the fourth quarter, revenue from sales of NAND products surpassed sales of DRAM, which had a favorable effect on the company's financial results as margins on sales on NAND flash products have remained more stable when compared to the margins on DRAM products. Comparing fiscal 2011 to 2010, we saw a nearly 40% decrease in per bit selling prices for DRAM products compared to about 15% decrease for NAND Flash products. In addition to the DRAM to NAND shift, we continue to have an increasingly diversified product portfolio within DRAM and NAND. Consequently, Micron has achieved significantly higher average selling prices for the year and most recent quarters compared to industry averages. On the operational front, you may recall that in the third quarter of fiscal 2011, IM Flash Singapore qualified its first product for sales to customers, which triggered the start of depreciation of production equipment at that site. We continue to invest in the IM Flash Singapore venture as it continues its production ramp. As of the capital call that was funded earlier this week, Micron's ownership interest in IMFS is 82%. Due to the timing of changes in wafer allocation relative…

Thanks, Ron. We'd now like to take questions from callers. [Operator Instructions] And with that, please open up the phone line.

[Operator Instructions] Our first questioner in our queue is Alex Gauna with JMP Securities.

I was wondering if you could go into the gross margin. A pretty steep fall off here, maybe some of the puts and takes. You had some pretty good bit growth, but I'm wondering about what's happening on the gross margin line and maybe what you're expectations are going forward here.

You bet. It primarily was captured in both Steve and Ron's comments. The biggest quarter-to-quarter change was in commodity DRAM and that was not only from an ASP change, which was pretty dramatic Q-to-Q, but also from an enriched mix from Inotera, where they start with more of the standard 2 gigabit DDR3 products, which as Ron mentioned in his notes have the biggest quarter-to-quarter downward pricing pressure. So primarily look forward in that DRAM area.

Well then going forward here, can we expect with some of the enriching mix of NAND that we have some leverage going the other direction?

Well we're not going to make a prediction. As you know, to gross margins but I think about the call, you'll get our feel for what our cost reduction outlook is and of course we have you overlay that with what you think the market dynamics from an ASP standpoint will bring. But yes, we do have some opportunities in terms of product mix that I think Mark Durcan would like to chat to you in that.

Yes. Let me just jump in on the NAND mix specifically. While, we did comment last quarter about a richer mix of SLC and we continue to shift larger volumes of SLC in the marketplace. With the ramp that's going on at IMFS continuing to be weighted heavily towards MLC see currently, then I wouldn't look for some big uplift in terms of the richness in mix in NAND over the next quarter or 2. And so we'll see continued cost reduction and bit growth, but that will be primarily in the MLC area and then as we look for additional opportunities to enriching that mix as we move forward.

And our next questioner in queue is John Pitzer with Credit Suisse. John Pitzer - Crédit Suisse AG, Research Division: A couple of questions. First, can you talk a little bit about the inventory levels in commodity DRAM? I think at the Analyst Day, you talked a little bit about inventory levels having come down and price bits per box starting to see an acceleration. I'm kind of curious how we ended the quarter.

Yes. This is Mark Adams. The inventory in that channel seemed to play out favorably towards the end of our Q4. The speculation was that through most of our Q4, inventories were up as we commented on in our last earnings call. But we saw some activity, as Steve noted in his opening comments, toward the last week of fourth quarter. And so fourth quarter-to-date that suggests that inventories are in a better condition, a better balance, and it's reflected in a very modest but still upward trend on commodity DRAM pricing. John Pitzer - Crédit Suisse AG, Research Division: And then, guys, as a follow-up, can you talk a little bit about the revenue falloff in the Wireless Solutions Group and the operating loss there? And kind of how do we think about that segment of the business moving back into an operating profit position?

Yes. I think, again this is Mark, the Wireless business, as many of you know, over the course of the year has been interesting to say the least in terms of the competitive landscape on the unit side, those selling the phones and those manufacturing and making the phones. And we've experienced our own transitions around products serving those businesses and trying to optimize the capacity that we're allocating to each of the bit use. We're being pretty careful with the growth around Wireless. It's a good business for us long term. But as we service our Embedded business with light products in our DRAM and NAND business, we're trying to be careful with the volatility around the Wireless business against the demand profile. So the top line growth is kind of somewhat limited by our ability to serve other business units and other segments more profitably for Micron. The gross margin and operating results that you're referring to are somewhat weighted down by us transitioning third-party supply of products from the old Numonyx channel to Micron-enabled silicon and that transition is almost done, which will help us. But in the past, this has somewhat limited our top line gross margin there. So we're a little bit more optimistic going forward, and we're a little bit more careful about how we grow in that space. And again, to reflect on that business, the #1 player 1.5 years ago had a close to 40% market share is now down to 20% market share, somewhere in that range. And the nature of the type of products being sold into that business is changing. So we're watching that to make sure we make the right decisions around production capacity, products and then actual customers themselves.

Our next questioner in queue is Shawn Webster with Macquarie.

A couple of questions. Going back to gross margins for a moment, what was the size of the inventory write-down that occurred? And what were the devices that got written down?

Shawn, this is Ron. We don't normally itemize inventory adjustments. But what I can tell you is sort of in line with the Mark Adams' comments, given some adjustments going on in the Wireless segment with some of our customized products to go into some of those businesses, we had some inventory adjustments notably in the Wireless and to a lesser extent, in the Embedded market. I would characterize this as being probably $20 million to $30 million above our normal quarterly rates that you'd see, that kind of range.

Okay. And then in terms of the bit production, not shipments, but what was your actual bit production sequentially for both DRAM and NAND in your fiscal Q4?

We had DRAM in fiscal Q4 up mid-single digits. We had NAND up low 40%.

Okay, and maybe one more, if I could squeeze it in. What are your customers telling you in terms of the demand outlook going into calendar Q4 by your various end markets on the PC side and handsets and servers? Are there any areas that are more stable or have notable strength than others?

Yes. Again this is Mark Adams. The application or segment breakout you're looking for, I would start by separating out Consumer and Corporate. The Consumer business still seems relatively off on the demand side against -- so the PC business is not super strong going into the holiday. And I think that impacts not just PCs. You can also make an argument that the Wireless business will -- could potentially see some negative in that demand pressure. Tablets are kind of a separate piece themself in terms of the demand and it still seems to have some pretty good growth through the holiday. Around things like server and networking, we still see pretty good strong demand. And as best as we can see going forward, it looks to remain strong. As well as in our embedded markets, the Automotive and Amusement segments are pretty strong around those types of NOR and DRAM-based solutions that we sell into those sectors. So again, I would go back to repeat that consumer still seems a little weaker and that the corporate landscape has maintained relatively stable around the application base that we serve.

Next questioner in queue is Krishna Shankar with ThinkEquity.

Yes. Can you give us the revenue percentages of PC DRAM, specialty DRAM, NAND and NOR? And can you rank gross margins for those 4 segments, please, relative to the corporate average?

I can break it down revenue-wise, by the business units for you. We had about 32% in DSG; around 30% in NSG; around 20% in WSG; about 11% in ESG; and around 6%, other.

Okay. And going forward, can you talk -- I know that PC DRAM pricing has continued to sort of -- it kind of picked up a little bit in September but can you talk about sort of the incremental impact of PC DRAM on the November quarter versus the August quarter?

Well I think what we'll say to you, Krishna, is our normal talking points, which quarter-to-date on DRAM were down about 10% to 15%. NAND is about flat quarter-to-date. And of course we'll -- it's up to you to look out over the next couple of months and determine what direction ASP will be.

Our next questioner in queue is Uche Orji with UBS.

Real quick, so if I look at the NAND ASP movements, obviously a lot of that was a mix from SLC to MLC. Can you talk about the breakdown of how that moved around between last quarter and this quarter in the mix of SLC to MLC? And related to that, what are your plans around ramping 3-bit-per-cell? Are you about to -- are there any design wins, which you are looking for to enable that categories to ramp?

This is Mark. Let me cover the NAND mix fees, anyway. The SLC as a percentage of mix dropped a couple of percent on a wafer basis. But the MLC grew roughly 6% as a percent of the wafer mix, with the remaining difference being a reduction in 3 level cell. So I don't think we want to pin the exact percentages for you. But on a wafer basis, it will give you a sense of how that moved around. And again the absolute value of the SLC was not decreasing. It was -- really, that was driven primarily by growth in total wafers out and total bit of. I'm sorry the second part of the question was what?

The second part of the question was how should we think about your growth in 3-bit-per-cell going forward? And will you be able to use that in the embedded market based on the percentage gives us right now?

Yes. I would look over the next few quarters for the 3-bit-per-cell to be relatively flat as a percent of our mix, although we are working on a number of solutions that over time could drive that number up in some of the embedded applications.

Yes. I think one more comment to tack on that when you refer to SanDisk, even if it's in the embedded market, I think that their overall capacity is still -- they've been saying they're more OEM-focused in their revenue in that way. They still have a fairly healthy retail business that can consume 3 bit and not pay accountably for performance. And so with the production cut that we're seeing in 3-bit, we're utilizing through our Lexar channel today. But as far as industrial strength applications, 3-bit is not showing up. And I think for us, a large part of our capacity is not necessarily just solely dedicated to retail, there's another unique value-added application markets.

Can I just ask one more question real quick? So which came from production cuts? And Powerchip has been talked about in the press, and perhaps also Nanya has been talked about as possibly cutting back. With the way demand is and given the level of inventory that you have, what will it take for you to cut back on production as a way to sort of help credit inventory? Are there any triggers that you monitor at this point? And then two, the module makers, what is their strategy right now? Are they buying? Or are they just sticking on their hand?

On the production cuts, we've seen the same data that you have or the same announcements. In terms of Micron, we are, I think, still pretty good in terms of your variable costs versus ASP. And as a result, we don't really have any plan for any production cuts. And I'll let Mark answer the next one.

With regards to the module manufacturers or assemblers, they were partially responsible for the uptick towards the end of August. And I think a part of it was they carry a fair amount of inventory into the summer, and that sold through slower than they had hoped. And they managed their inventory fairly well towards the end of the summer, which reflected in relatively tight supply that drove some of the modest uptick in pricing that we referred to. And we're now kind of in a wait-and-see mode to see how they fair going into the holiday season and the aftermarket. Of course we have our own Crucial business, which is in that, competing in that space and is doing fairly well. But we probably need more data point before we're able to tell you long term how that works out.

Our next questioner in queue is James Schneider with Goldman Sachs.

On the DRAM pricing side, you're talking about quarter-to-date pricing down about 10% to 15%, which seems to be a little bit worse than what we've seen in terms of the PC DRAM on the spot market. Can you talk about some of the mix that might be going on, on the server, especially area that might be contributing to those difference as it trends?

Let me just add one thing. That is relative to an average for last quarter, so keep that in mind. And then I think Mark Adams would like to add some commentary around your question about the server.

Yes. I think -- also, when you look at pricing trends over a longer period, say a year or 4 quarters, what typically obviously leads you in either direction down or up is the commodity pricing. And then there is some correlation between the commodity price and the direction of our other application ASPs. That being said, as Kipp noted, we saw some of those modestly decline, which gave us an average when compared to prior quarter isn't -- as kind of in trend line of the commodity pricing. I think commodity pricing wouldn't be down to your points nearly as much as, we're suggesting the overall ASPs, so 10% to 15% is our blend base on the average of last quarter, and some of our higher specialty markets, higher priced specialty applications coming down a little bit. And we think that, that's reflective in the calculation.

Okay, that makes sense. And then as a follow-up, on the NAND side, as you look at the environment right now, could you maybe comment and give kind of a first bogey on what you expect in 2012, your NAND bit production to do through calendar 2012, either a year from today or for the full calendar year?

We really like to stay away from a year-over-year comparison. I think as Mark Durcan has noted, we can make a lot of changes to the mix. So we'd like to stay with something more in the short term. We're looking at mid-single digits for our next fiscal quarter, and then something more in the double digits for the quarter after that. And please keep in mind, as we can move these bits around quite a bit.

And then your take on the industry growth for next year in NAND?

Yes. NAND is looking at around 65% to 75% depending on some of these things we're talking about, SLC, MLC, TLC mix as we go through 2012.

Our next questioner in queue is David Wong with Wells Fargo.

Have you taken any reserves that might be applied against the fine related to a negative [indiscernible] in the Rambus antitrust case? And how much of SG&A is currently going towards legal costs associated with litigation?

Well the reserves that we take are on the basis of anticipated litigation costs. We are not taking in the reserves with respect to any kind of judgment or verdict. And of course until we have that, we have no idea nor when that might be.

Okay, and can you give us any magnitude of what your legal costs are running on a per quarter basis?

Well Ron can jump in here in a second, but obviously they vary as to -- depend on when you're in trial. We just got through a trial, and some of our costs for this last quarter up to, really, about a week ago, are going to be, I think, a little bit higher than normal because we're actually in litigation.

Yes. We don't actually break down those pieces in our SG&A structure. Bear in mind that we try to estimate each quarter, give it a little bit more color on how this works. The cost to adjudicate or settle items, as Steve mentioned, in a particular quarter and we look forward to the estimated cost of adjudication through all the events we have. And we have a number of events that are detailed in our filings, our Q and K filings that you can look at. We estimate all those and then try to improve the cost estimates in each quarter. We true up those estimates. So obviously we're very involved in a significant suit. We will true up those things in a more timely basis. And we did have some increase in this quarter on a number of categories as we are looking at the go-forward adjudication cost to work through these numerous events, not just the one Ramus item.

Great. And my final question, for the commodity DRAM and for your NAND, are your current contract prices above or below spot prices?

I'm sorry, could you repeat the question?

For commodity DRAM and for NAND, are the contract prices that you're getting from your big customers, are they above or below what we see on the day-to-day spot prices?

Well today, they're slightly below, and that's just normal for the cycle that go on in pricing in the market. The spot market normally leads you in one direction or the other. In this case, spot has come up a little bit. As we've noted a few times in the call, the OEM contract pricing is lagging slightly.

Our next questioner in queue is Kevin Cassidy with Stifel, Nicolaus.

This is Dean Bramos [ph] calling in for Kevin Cassidy. My question is can you elaborate on plans to accelerate mobile DRAM and other non-commodity DRAM products?

Sure. As you can -- might imagine, I think like a historical effect, we've sold a lot, a fair amount of DRAM capacity into the computing sector, and most notably, the desktop client, notebook market. Over the years, we spent a lot of time focusing on new application segments. And I think now more than ever, especially with the new business unit structure that we've set out as a company and the acquisition of Numonyx and the market segments they were serving in joining with Micron, we've got a number of new application markets that are serving that same capacity. And when you split that out, markets like amusement and the server and networking segment and PC graphics, if you look at automotive, there's just a number of different sectors that we serve that may be, a while back you would have said were primarily dedicated to the PC space. So the evolution of the applications we're serving are much more dynamic and are leading us to try to optimize our capacity for the most profitable long term and most profitable customer relationships and long-term business. And I think that effectively shows up when you look at our shares in some of these segments, automotive, networking server.

Our next questioner in queue is C.J. Muse with Barclays Capital.

I guess first question on the NAND side. I was hoping you could help me understand your thoughts on profitability there. Your cost on efforts came in better than what I was thinking. Intel mix is heading lower and that'a also a little bit by the SLC/MLC mix shift. But curious, as that trend goes forward over the next few quarters, how should we think about the uplift to earnings for just NAND by itself and then the contribution to your overall company?

Well again we don't like to give you guidance on margin per se. Let me maybe go back, as the previous question might at least help you put this in at least order of margin ranking for the quarter we're just reporting, specialty DRAM still the highest; NAND, second; NOR; and then PC DRAM. So as Mark noted, we do have some pretty nice operational cost stems [ph] coming. We do have some levers in different parts of our business to continue to look at a pretty favorable NAND business. But we're certainly not going to give you any guidance as to operating margin or gross margin.

This is Ron. One thing I could add to that, if you look at our NSG results, which is our commodity NAND business, we're still running a profit. And that's despite the mix shift that was already commented on in terms of the ramp of IMFS and the higher MLC content. So our core commodity NSG product line, as well as all the other NAND business that go through NSG is still showing a positive operating profit. And then of course we have NAND that goes through our other 2 businesses, Wireless and Embedded, which can have reasonable margins as well. So in addition to the IMFS performance, which we have improving and had already been recounted and how that's going to help us going forward, and the cost structure of IMFS is already below the average of our other fabs. We're showing profitable performance across all of our NAND businesses right now, if you look at the aggregate total. And the improving performance in IMFS will only help that. Of course, the self-predicated on pricing assumptions, which right now look flat.

Sure, that's helpful. And as a follow-up question, you talked about the industry bit growth for NAND tracking 65%, 75% for 2012. Curious, can you grow in line with the industry with the capacity you have at IM Flash Singapore and the $2 billion CapEx budget you've outlined? Or would you need to add capacity there?

Yes. This is Mark Durcan. Kipp commented on bit growth over the next couple of quarters. I would say, if you want to take a longer term view, keeping in mind the quarter-to-quarter, the numbers will be pretty bumpy. We should be averaging in the high teens in terms of our trade bits into the market, the bits that Micron sells for profit. So we should exceed in as we go.

Our next questioner in queue is Hans Mosesmann with Raymond James. Hans C. Mosesmann - Raymond James & Associates, Inc., Research Division: Can you comment on some of the supply dynamics in DRAM? There's some supply that we have come off, maybe some is coming back in next year. What's that dynamic as we look over the next 18 months in DRAM specifically?

If you think about the comments that we had last quarter, I think they're still consistent for this quarter, which is why I made the statement that we didn't think that we're oversupply-driven challenged right now. It's mostly a demand-driven challenge. And if you look at the CapEx, the CapEx and the capacity going to DRAM has already started to decline. And in fact, if you look at the peak of the CapEx into DRAM, which would have been 2010 moving into the first part of 2011, that it did peak and it's starting to come down quarter-by-quarter. So it really got to a level that was probably just over half of what we had peaked before in the 2006-2007 timeframe. So we're just not going to get the kind of bit growth that we saw in that timeframe, which is why we made the statements that I think we're in a demand-driven challenge environment right now. Now with that in mind, if you look at the CapEx that is being spent and you look at all that data, almost 100% of that CapEx is going in to advancing facilities. Now there's some new silicon, but if you think of what we've been doing with Inotera, with our own facilities, if you think about what LP has been doing or Hynix has been doing, it's also been going into upgrading, if you will, to the next generation of technology, which does drive some bit growth. But it's not near the bit growth you'd get from bringing on new facilities. I think Samsung had some media commentary, where it has some new silicon. But for the most part, I don't think anybody expects any large amount of new silicon coming online. And we're not seeing it in the marketplace. And in fact, if you look at the equipment suppliers, they're confirming that their book to bills are below 1, and there's a lack of demand for equipment going on in that segment. So I think that by and large, it will probably not take much of an improving demand environment that would -- that pretty quickly soak up any excess supply in the marketplace. Hans C. Mosesmann - Raymond James & Associates, Inc., Research Division: Okay. And as a quick follow up, what's the bit growth supply number you guys are working off as you look into 2012 calendar?

For DRAM? Hans C. Mosesmann - Raymond James & Associates, Inc., Research Division: For DRAM, yes.

Looking like 40% to 50%.

Our next questioner in queue is Ryan Goodman with CLSA.

So just on the 65% to 75% growth number you put out, that's a little bit lower than you've talked about in the past. Just curious, how has this impacted the capacity expansion plans for IMFS next year?

Really no change. We've always talked about IMFS as being in that 60k start range by the end of the year. And I think Mark, if he hasn't updated, he's more than happy to. It's going really well. It's ahead of schedule, so we'll hit that pretty shortly. None of the budget in that $2 billion CapEx that we've bogeyed for fiscal 2012 includes a ramp beyond that. So the numbers that we used in what I quoted were industry supply numbers, not Micron-specific.

And so right now the expansion plan is pretty much holding at that 60k, there's nothing additional on the $2 billion to take that higher than that as of today?

That's correct. Mark would you like...

No, that's what I was going to say.

Okay. And then just one follow-on also. You'd mentioned a little bit of the Numonyx production was being pulled in-house as a cost driver -- or a cost saver, I should say. I'm just curious, where are we in that? Specifically on the NAND side of things, is that mostly pulled in at this point? Or is that one quarter, 2 quarters away from being fully captive at this point?

Yes, it's primarily in-house at this point other than a couple of very small volume items, where throughputs and large amounts are in.

Okay. I guess, and if I could just slip in one more really quick, just on the cash. The cash balance is looking pretty good, could you give an update just on the mindset on buybacks versus M&A? I know you did a little bit this last quarter on the buyback side, but going forward, how should we be thinking about that?

Yes. Well clearly the weaker the market is, the more conservative we get probably in terms of these or our cash. But the other core letter with that is the lower our share price, the lower we view it as being significantly below book where it sits today and maybe it might creates an opportunity for us to consider when we should do something like that. Obviously that's a board decision that will have to be made as we move forward. We've done some of it already as you've noted. But I would say that our balance sheet is pretty strong. And as of right now, we have, I think we have pretty good flexibility, whether it's M&A or any kind of equity restructuring alternative we want to pursue.

Our next questioner in queue is Bob Gujavarty with Deutsche Bank.

I was wondering if you could talk a little bit about -- you mentioned the retail SSD business. What's your sense of what OEMs are thinking about SSD adoption for maybe this holiday season and maybe more importantly next year? Is it still build materials challenge for them? Or have you seen some warming up from them?

This is Mark Adams. I've definitely seen some warming up from our major OEMs. I think on the build material sensitivity is obviously around the client, desktop, notebooks portion of that business. And I certainly think as people are looking to kind of differentiate [Audio Gap] focus and energy and the developments around the silicon continue to mature, i.e. controller development, software and firmware development and make the enterprise more realistic in terms of enterprise cloud performance. And finally, the noted channel fees, it's interesting to note, if you look right now in the sales of SSD units, in the notebook PCs are roughly 45% of all SSDs being sold. And second to that, one might think it would be enterprise but it's not actually. Roughly around high 20s to 30% of the SSDs are sold through a channel today. So when you combine all that, there's a pretty robust appetite for SSDs in the market, good growth in 2012 and thus, we continue to be a pretty strong good growth. And we think the OEMs will become more involved in the market opportunity.

Great. And maybe a quick one on DRAM, is there any opportunity to kind of maybe upsell DRAM beyond just the capacity RDIMM perhaps, specializing on low-power PC DRAM or especially super small form factors or special packaging to get maybe some upsell over just plain capacity commodity DRAM?

Yes. And I think the market loses sight of that because I think the -- to your question, most people kind of say, well tablets are going to eat away at the DRAM markets. But in fact, we have a little bit different view on that. We think that low-power DRAM applications, which of course require us a bigger chipset and we sell it at a premium to the commodity DRAM, we think that, that's a good opportunity for us. And so the differentiation there allows us to penetrate tablets. And over time, we think tablets will continue to grow in content per box, obviously not the same as notebook today. But when you factor in the low-power premium and the bigger chipset around low-power, the impact on wafer utilization, wafer capacity is somewhat negated. And so we think there's a good opportunity for us there, and we're pretty bullish.

Our next questioner in queue is Betsy Van Hees with Wedbush Securities.

Kipp, on the DRAM side, you said 30%, net 30s was DRAM. Of that, how much was specialty and how much of that was commodity PC?

Commodity PC is now down to about 15%, 16% of total revenues.

Okay. And then, Steve, you talked about this is a demand-driven downturn in the DRAM market, and then Mark, you talked about the PC -- or you talked about the tablet market. And the fact that this is a double-edged sword for you guys because it was very positive with NAND, but then on the other side, its negative on the DRAM side from a bit growth perspective. So when you put those kind of into perspective on the DRAM side, is it really -- and we're heading into -- we're heading out of peak seasonality. So what's going to be the drivers to get the DRAM market back and DRAM supply and demand back into balance?

So I think -- this is Mark Adams. I think that well in general, we're still going to see some PC growth. I think the last number that I've seen on 2012 PC growth numbers is in the 10% range over 2011. And I think that as I was just commenting on around low-power DRAM, it's not exactly a one for one trade off in terms of capacity utilization. So combined with continued growth around corporate enterprise applications, if you look at, for example, Micron being a leader in the server DRAM business, that's a business where density continue to dramatically increase. I think today we approximate that on average, servers grow out with 25 to 30 gigabytes per box. In networking, we're seeing increased densities as well. So across our overall landscape, we think that there's plenty of application areas to invest on the DRAM development from a go-to-market perspective, and we think that, that will all work itself out in terms of overall supply and demand alignment.

Sorry, Kipp. One other thing, on the equation, also keep in mind that as tablets have clearly replaced some of the notebooks and the PCs, it's actually also driving quite a bit of server connectivity, quite a bit of connectivity to the Internet and the infrastructure that then drives demand. I mean if you look at the servers being put in place by the Googles and the Facebooks, et cetera, it's actually pretty stunning how much DRAM they're consuming now in order to put the infrastructure in place with which for all these tablets to access. So it's not quite the equation that you are describing. Although you're right, there's a piece of it that's happening there. But there's another side of it that I think people miss.

That was very helpful. And that actually leads to my next question. You were talking about the Google and how they're building their own servers and their need for storage. So that brings me to the enterprise side could you talk a little bit more about your customer engagements? What type of traction are you getting with the tier 1 OEMs? And are you shipping in volume production? And how are you looking for next year?

I'll let Mark answer most of that questions, but I did want to make the comment that it's an interesting scenario for us where not only are we engaging with what you would think of as the tier 1 server OEMs, but we are also now engaging directly with some of those companies that I mentioned that are becoming big consumers of this product and actually building their own systems. Mark, do you want to add to that?

The only thing I would add to that is those types of customers from a technical competency standpoint are kind of learning as they go, and they look to Micron to be able to add some value there. So if you think around SSDs as an example or overall system performance in DRAM and how the memory architecture amend itself to better performance for their servers, the engagements are a little bit more consultated and technical in nature and we get a value add for that.

Our next questioner in queue is Harlan Sur with JPMorgan. Harlan Sur - JP Morgan Chase & Co, Research Division: Based on your Analyst Day presentation, it looks like above 45% to 50% of your DRAM wafer outsourced is still at the 5x nanometer node and roughly 40% at 40. Could you just give us you view on where this mix will probably likely be at the end of the calendar year, maybe including 30-nanometer as well?

Well, first of all, without recharacterizing what you just said, you need to keep in mind that there's a lot of specialty and sort of value add legacy products associated with that mix. On top of that, I'll add that in the current quarter, we did crossover 42 to 50. So the majority of our bits are now 42-nanometer bits. And we will see really probably in the fiscal second quarter, not the quarter we're in now but in the fiscal second quarter of '12 for Micron a significant ramp up on the 30-nanometer node. So majority of the DDR3 42-nanometer, and you'll start to see that shift again when you get another roughly 90, 120 takeup [ph]. Harlan Sur - JP Morgan Chase & Co, Research Division: Okay. And then, so on your answer in terms of the crossover happening this quarter, that's on a wafer out perspective?

That's on a bit perspective crossed over in the quarter we just closed. Harlan Sur - JP Morgan Chase & Co, Research Division: Got it, okay. And as a follow-up question, maybe you can just provide us with an update on the 40-nanometer yields. Are they in-line with your cost reduction targets? And maybe you can just articulate what is the team doing differently at the 4x node from a yield perspective versus 5x where you did struggle a bit. And then maybe provide us with an early glimpse into your 30-nanometer yield improvement rates as well.

Sure, without getting too specific about yields, I will say, 42-nanometer node ramps had gone better than the 50-nanometer node. A lot of that -- we had commentary on the previous couple of calls about some of the transition issues we experienced in Inotera with mismatched tool set and cycle times around ramping that. Note that as we transition from trends to stack, but beyond that, I think we're pretty happy with where we're at, on a 42-nanometer node now. We still have some upside as we continue to reach what I would call ultimate material yields. But we're clearly executing much better now than we were 6, 9 months ago. On the 30-nanometer node, we're very happy with the just very early silicon coming out of Inotera. That's going really, very well indeed, although, we're in the early days there. So MTV ramping and Inotera still on the early lives out, but very impressed with the execution on the early material.

Our next questioner in queue is Bill Dezellem with Titan Capital Management.

What implications, if any, do you see with Hynix ownership structure changing, impact on the industry?

Bill, I don't really see a lot as you've probably noted or noticed from the media that they're having trouble finding someone that wants to buy them. And they've limited it to Korean investors only, which means I don't think you're going to see a lot of change in terms of the company's activities. Maybe a little more is being made of, it really exists around the controlling interest. But frankly, I just don't see a whole lot shifting.

Our final question for today's event is Glen Yeung with Citi.

I think in -- I guess it was in August, you saw a bit per box increase on the commodity [indiscernible]. And so I wonder if you can just talk to us about what you saw there and what are the instance to take away from that.

Yes. This is Mark Adams again. I think my comment at the time was that going to holidays, we continue to see the kind of the entry level value products around 4 gigabytes, both on the desktop side and the notebook side. And it's intriguing that we hear a lot about tablets getting into that and so on and so forth, and that might change strategies, the general overall strategy going in to the holiday. But the fact remains that it continues to get strong around that 4 gigabyte platform memory content, both on the desktop and notebook side. When you look at the building material cost on memory, it's pretty low these days. It's about 3% or so. And that's well below the average that we typically see. And I think the other thing driving that is I think the success that Apple has had kind of in the performance SKU for the holidays, they're not selling a lot of 4-gigabytes and 2-gigabyte units. And so I think the rest of the industry has to watch that and keep up with the kind of performance vendor, if you will. And I think that's driving the content to be in a better place. If you went back 18 months, I look at the data right before our call today, and if you went back 18 months, we were having the same conversation then, it was 3 gigabytes. And we're all saying it will never going to go up. And it stopped dramatically since then. So we don't see on the client side anything more than continued growth beyond where we're at today.

Does Ultrabook give you any reason to be optimistic for this holiday selling season?

Oh absolutely in terms of memory content, absolutely. And we think that 4 again is the basic minimum there. And one of the things -- it's intriguing us because you're not able to upgrade the Ultrabook. They're going to max out on point of sale of the unit itself. And thus, I think that you're going to see increased sale in the Ultrabook market of 4 and 8 gigabyte configurations.

Is too early to expect Ultrabook to be a meaningful contributor?

Yes, it'll be hard for me to call. I think it's definitely too early to tell at this point.

Okay. And then last question is when you look at 2012 CapEx, can you remind us what the split is between NAND-related and DRAM-related CapEx?

You bet. It's about 2/3 NAND wrapping up the MFS's fab and about 1/3 DRAM. And we would 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.

Thank you. This concludes today's Micron Technology's Fourth Quarter and Fiscal Year End 2011 Financial Release Conference Call. You may now disconnect.