Good afternoon. My name is Kate, and I'll be your conference facilitator today. At this time, I would like to welcome everyone to Micron Technology's fourth quarter 2014 financial release conference call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there'll be a question and answer period. If you would like to ask a question during this time, please press star, then the number 1 on your telephone keypad. If you'd like to withdraw your question, press the pound key. Thank you. It's now my pleasure to turn the floor over to your host, Kippen Ballard. Sir, you may begin your conference.
Thank you very much. Welcome everyone to Micron Technology's fourth quarter 2014 financial release conference call. On the call today is Mark Durcan, CEO and Director, Mark Adams, President, and Ron Foster, Chief Financial Officer and Vice President of Finance. This conference call, including audios and slides, is also available on our website at micron.com. In addition, our website has a file containing the quarterly operational and financial information and guidance, non-GAAP information with reconciliation, slides used during the conference call, and a convertible debt and capped call dilution table. If you have not had an opportunity to review the fourth quarter 2014 financial press release, again, it is also available on our website at micron.com. Our call will be approximately 60 minutes in length. There will be an audio replay of the call. You may access that by dialing 404-537-3406 with a confirmation code of 2237916.
This replay will run through Thursday, October 2nd at 11:30 P.M. Mountain Time. A webcast replay will be available on the company's website until September of 2015. We encourage you to monitor our website at micron.com throughout the quarter for the most current information on the company, including information on various financial conferences that we will be attending. You can also follow us on Twitter at MicronTech. 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 in 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.
I'll now turn the call over to Mark Durcan. Mark?
Thanks, Kip. We had another strong quarter benefiting from robust market demand as well as solid operational execution. We set a new record for revenue of over $4.2 billion for the quarter. Net income was $1.15 billion, or $0.96 per diluted share. For the fiscal year 2014, we generated record revenue of $16.4 billion, record net income over $3 billion, and record free cash flow of $2.6 billion based on record operating cash flow of $5.7 billion, less CapEx of $3.1 billion. As we enter fiscal 2015, I'd like to touch on a few key areas of focus, as well as provide a brief industry update. Ron Foster will follow with a financial summary, and before turning to the Q&A, we'll close our prepared comments with Mark Adams covering additional details of our operational performance and market conditions.
Beyond ongoing advanced component and technology development, we have three main operational focus areas for fiscal 2015. The first is technology deployment and includes continued 25-nanometer DRAM conversion and 20-nanometer DRAM ramp, completion of 16-nanometer planar NAND conversion and introduction of 3D NAND, TLC NAND deployment for cost-sensitive applications, and building out our capability to deliver advanced packaging solutions and controllers. The second is optimizing manufacturing capacity and includes improving our manufacturing efficiency through line balancing, cycle time, yield, and metrology initiatives. Managing our product mix in order to generate the best possible long-term margins and returns for our business. Finally, in 2015, we'll focus on growing our memory systems and subsystem solutions, including expanding the market penetration and offering breadth of our advanced bit addressable memory solutions, building additional storage solutions such as enterprise SSDs, and designing new and innovative mobile memory system solutions.
As we outlined at our Analyst Day last month, fiscal 2015 CapEx is expected to be in the range of $3.6 billion-$4 billion. Roughly 30% of this CapEx will go towards non-supply expanding investments, including manufacturing efficiency improvements, emerging memories, back-end capability, and additional system manufacturing and engineering tooling to support system and subsystem-level products. This compares to about 25% of our $3.1 billion CapEx bill in fiscal 2014 and highlights our belief that these investments, which are designed, among other things, to enhance our value-added product portfolio, will generate some of the highest returns going forward. In terms of the more traditional investments in memory process technology, DRAM will represent about 50%, and NAND will represent about 20% of our total CapEx in 2015. This compares to roughly 45% and 30%-35% respectively in 2014.
We expect the bit growth we generate from technology in calendar 2015 to be in line or below the market for both DRAM and NAND. Our bit growth profile is reflective of a strategy to grow the mix of premium and value-added products while maintaining and improving operational efficiency and costs. While this will lead to less pure supply growth, we expect we will generate more attractive and sustainable margins over time with this approach. Our long-term outlook for the memory industry remains favorable. For DRAM, we're forecasting 2014 industry supply growth of around 30%. In 2015, we expect DRAM industry bit growth in the low to mid-20% range, and beyond 2015, we expect industry supply growth to slow somewhat, ranging from the high teens to mid-20s.
Part of this encouraging supply trend is due to the fact that 20-nanometer and sub-20-nanometer process technology ramps significantly reduce the wafer production for existing clean room space. We believe the DRAM industry supply-demand balance will continue to be favorable for the foreseeable future. Demand elasticity has declined from historical levels, and cost per bit is no longer the major factor driving demand growth. For NAND, we're projecting industry supply growth in a high 30% to low 40% range for 2014. This includes an increase in industry wafer production of just over 10%, with the remaining supply growth coming from technology. We expect 2015 industry supply growth to be in the high 30% to mid 40% range. Beyond 2015, when 3D becomes more prevalent in the market, supply growth will depend on new capacity investment.
Although 3D technology drives a significant increase in bits per wafer, without capacity investment, this is offset by substantial decreases in wafer output per square foot of clean room space. The demand profile matches up well with the existing supply growth rate, and we also believe that due to elasticity of demand for NAND in client and enterprise storage applications, the demand growth could be significantly higher over time. At Micron, we'll continue to manage our NAND business focused on long-term returns. I want to congratulate the whole Micron team for a tremendous year. We look forward to building on this success in 2015. The memory industry continues to look very attractive, and we've never been better positioned to deliver differentiated value for our customers and shareholders. I'll stop here and turn it over to Ron and Mark before turning for Q&A. Ron?
Thanks, Mark. Our fourth quarter 2014 and fiscal year ended on August 28th. We posted to our website a file containing the financial information I will cover, including GAAP and non-GAAP results, certain key metrics for fiscal 2014, the fourth quarter, as well as guidance for the first quarter of fiscal 2015. For fiscal 2014, we posted record revenue and net income, as Mark mentioned, with a net income of $3 billion or $2.54 per diluted share on net sales of $16.4 billion. This compares to the fiscal 2013 results of net income of $1.2 billion on net sales of $9.1 billion. Recall that the 2013 results include the $1.5 billion non-cash gain in the fourth quarter from the purchase accounting for Elpida and Rexchip, which we now refer to as Micron Memory Japan, or MMJ, and Micron Memory Taiwan, or MMT.
The results for fiscal 2014 reflect continued healthy market conditions, particularly for DRAM products and our focus on maximizing long-term returns in the business. Non-GAAP net income increased significantly comparing 2014 to 2013. Now focusing on the fourth quarter results, we posted net income of $1.15 billion or $0.96 per diluted share on net sales of $4.2 billion. The fourth quarter gross margin of 33% includes a $66 million charge associated with a patent license with Tessera. The go-forward license will have cost of goods sold charges that are not anticipated to have a material impact to any future quarterly period. In addition, gross margin in the fourth quarter was adversely affected by $38 million from a last-time sale of legacy architecture mobile PCM or phase-change memory products, which was mentioned last quarter as part of our guidance.
Without these two items, gross margin for the fourth quarter would've been higher by approximately two and one-half percentage points. On a non-GAAP basis, net income for the fourth quarter was $961 million or $0.82 per share. Non-GAAP adjustments netted to a reduction of income of $189 million or $0.14 per share in the fourth quarter and included the following: the charge related to the Tessera license mentioned previously, restructure charges in the quarter of $22 million, which were primarily related to employee termination benefits from a workforce reduction in Italy. Amortization of debt discount and other costs of $37 million in the fourth quarter includes imputed interest on our convertible notes and the discount on the MMJ installment debt.
The loss on restructure of debt of $17 million related primarily to the repurchase and conversion of convertible notes in the fourth quarter, including mark-to-market adjustments of the 2031B Notes to be settled in the first quarter of this year. Non-cash taxes from the MMJ operations were $118 million benefit in the fourth quarter, which includes a benefit from a change in the estimated utilization of net operating loss carryforwards for MMJ and MMT. A couple of significant gains recognized in the fourth quarter also adjusted out of the non-GAAP results as follows. A $93 million gain was recognized as a result of Inotera's equity GDR offering in May, that was executed at a price above the carrying value of our investment. This gain was recognized in the fourth quarter, since we account for our equity method results from Inotera with a two-month lag.
A $119 million gain was recognized from the sale of our remaining interest in Aptina to ON Semiconductor. The transaction closed just shortly before the end of the fiscal year. Finally, there is a 27 million share anti-dilutive effect of capped calls based on the average stock price during the fourth quarter of $31.91. In the first quarter of fiscal 2015, we expect the following non-GAAP adjustments. Between $40 and $45 million for amortization of debt discounts on the convertible notes and the Elpida installment debt. We are anticipating a Q1 net loss on restructure of debt of approximately $25 million as the conversions of the 2031B Notes are completed. Non-cash taxes related to the Elpida acquisition are expected to be between $20 and $30 million in the first quarter. Also, the anti-dilutive effect of our capped calls will be based on the average share price for the quarter.
Please refer to the convertible debt dilution table included in the earnings call data file posted on our website. Turn now to our results by technology and other guidance for the first quarter. On a consolidated basis, we are guiding total revenue for the first quarter in the range of $4.45 billion to $4.7 billion. Recall that our first quarter of fiscal 2015 will include an extra week due to the periodic adjustment of our weekly fiscal calendar to coincide with the end of August year-end, which will have a direct impact on operating expenses. However, revenue may vary differently. Turning now to DRAM. DRAM revenue increased nearly 5% compared to the third quarter, primarily due to an increase in bit sales volume and stable ASPs. DRAM gross margin remained in the high 30% range with flat cost per bit.
Cost per bit would have been down 2%, absent the effect of the Tessera license. Our reported income from equity method investments for the fourth quarter was $119 million, substantially all of which is attributable to Inotera. DRAM gross margins for the first quarter, using quarter to date ASP and projected mix for the quarter, should be up a couple points compared to Q4, based on bit production up mid to high single digits, ASPs up low single digits, and cost per bit down low single digits, which includes the effect from the Tessera license in Q4. Key items affecting our DRAM guidance for the first quarter are continued favorable market conditions, like-for-like pricing, generally flat to up for PC and server DRAM, and trending up for mobile DRAM, which Mark will comment on in a few minutes.
The trade NAND side, revenue increased 6% in the fourth quarter with a 13% increase in bit sales volume, partially offset by a 6% decrease in the average selling price. Trade NAND gross margin was in the mid 20% range, down approximately three percentage points, with ASP declines outpacing cost per bit reductions of 2%. However, absent the effect of the Tessera license agreement, cost per bit would have been down approximately 5%. Trade NAND gross margins for the first quarter, using quarter to date ASP and projected mix for the quarter, are expected to be down one to two percentage points compared to Q4, based on bit production is expected to be up high teens, ASPs down low to mid-single digits, and cost per bit down low single digits compared to Q4, which includes the effect of the Tessera license.
Key trends for Q1 affecting this guidance are generally stable NAND market conditions, also strong growth in client and enterprise SSD volumes, with more pronounced price competition and initially higher BOM costs that impact margins. Looking at other P&L and cash flow results and guidance, SG&A expense in Q1 is expected to be up slightly from the normalized weekly run rate from the fourth quarter. R&D expense in the first quarter in fiscal 2015 is expected to increase compared to the fourth quarter, even when normalized for the 14th week in Q1, due to higher volumes of development wafers processed for new technologies and new solutions development. The company generated $1.35 billion in operating cash flow in the fourth quarter and ended the quarter with $5.4 billion in cash and marketable investments, approximately $1.8 billion increase over the fiscal year.
Expenditures for property, plant, and equipment in the fourth quarter and fiscal year were $1.3 billion and $3.1 billion, respectively. Our cash tax rate continues to be in the low single-digit range at current profit levels, while GAAP taxes are expected to run in mid-single digits this year. Throughout this past fiscal year, we've been in the market in each open trading window, revamping the capital structure of the company with a focus on reducing dilution associated with our convertible notes. We have been repurchasing or converting convertible notes with cash and replacing them with high-yield notes, including the fourth-quarter repurchase of the 2032 notes and issuance of $1.15 billion 5.5% notes. Approximately 90% of the free cash flow generated during the year was used for these dilution management activities. In addition, in the fourth quarter, we used $339 million of cash to prepay certain foreign loans and leases.
We have indicated that the pending conversion of the remaining 2031B Notes will be for cash, resulting in an outflow in Q1 of approximately $390 million of cash. I'll turn it over to Mark Adams for his comments.
Thanks, Ron. I will begin, as usual, with a review of our Q4 operating performance by business units, as well as share commentary on market insights, key segment trends, and memory industry dynamics. Our computing and networking business unit, referred to as CNBU, had another strong quarter, achieving $1.9 billion in revenues in Q4. DRAM pricing in CNBU was up quarter-on-quarter, highlighting a continued healthy demand supply balance in computing, server networking, and enterprise market segments. On the client side, we continued to receive strong demand signals from our PC customer base. We locked our pricing through the end of our calendar Q3 and with our three largest OEM customers and have received interest in locking prices again for Q4. We achieved record revenue and bit shipments in our server segment. Server DRAM growth is being driven by customers adding more memory per system.
In fiscal year 2014, we saw a 40% year-over-year growth in DRAM per server, while ASPs in this segment have strengthened over the same period. This growth in server-based memory is based on increasing server workloads, continuing to require DRAM performance and density, and is a great example of a high-growth segment with a demand profile that is not sensitive to price. Networking revenues in DRAM were up 19% quarter-over-quarter. Demand for our networking products remains strong, driven by LTE build-out in China and in other emerging markets. Next year, we expect to see significantly more 4G handsets sold utilizing this capacity, which continues to drive higher memory content per phone. I will address this more in the mobile section. The U.S. will fully be engaged in a 4G LTE build-out throughout calendar 2015, as a result, we anticipate seeing continued strong demand in this area.
Lastly, our graphics business continues to flourish and increasingly contributes to an uplift in ASPs and margin. We are starting to see strong gaming consoles demand ahead of the holidays and remain optimistic for a good quarter in graphics. We are pleased with the progress on the technology innovation front in DRAM. We commenced early DDR4 shipments in Q4 as Intel officially launched DDR4-enabled platforms in early September, and Micron is validated with their entire 4 gigabit-based portfolio. I want to recognize our R&D organization, our engineering teams, and BU product development organization for positioning Micron as first to market in what we feel can be a strong value-added business. We are in qualification at major OEMs for both server and client opportunities for DDR4. We anticipate a significant price premium for those customers looking to differentiate their solutions.
We have also launched into the non-volatile DIMM category, providing for DRAM content backup, which dramatically improves reliability. Our NVDIMM product launch, initially based on DDR3, is going well as we have received production orders from tier 1 customers. We have also signed a lead major OEM customer for the launch of our DDR4-based NVDIMM, targeting shipments in calendar Q2. Micron's Hybrid Memory Cube continues to gain significant traction at leading network and server customers. In addition, similar technology from Micron will be used in Intel's Knights Landing platform for high performance and low latency benefits in high-performance computing. Finally, our GDDR5 product continues to gain wide adoption in gaming consoles and is also being sampled in high-performance networking applications. Our storage business unit, or SBU, recorded $907 million in revenue in the fourth quarter, up 4% quarter-over-quarter.
We continue to make progress in both the client and enterprise SSD market. We set records for total SSD revenue, gigabytes shipped, and overall units in our fourth quarter, while also achieving record revenue for both client and enterprise SSDs individually. We increased client SSD shipments to tier 1 OEMs by 23% quarter-on-quarter. For our fiscal year 2014, SBU revenue was up 23% when compared to 2013, coming in at $3.5 billion. Gross margin dollars were up 17% year on year, reflecting the overall health of the NAND market. We remain focused on improving the fundamentals of our NAND business as we believe there is upside to our performance relative to our competition in the coming quarters. I would like to discuss some of the elements which we feel can bolster our NAND business. First, we continue to invest in process innovation.
We executed the fastest ramp in our company's history for our 256 gigabit 16-nanometer planar product. TechInsights recognized Micron with the Semiconductor of the Year Award and Most Innovative Memory Device Award for this technology. We are making progress in our 16-nanometer TLC product with engineering samples still on track for the end of this calendar year. We are targeting late spring shipments of TLC components to consumer applications and expect to be shipping a TLC client SSD drive into the market during our fiscal fourth quarter. Micron will continue to increase our leadership in overall NAND scaling with our 256 gigabit 3D NAND device, which we believe will have the highest density per square inch of silicon in the industry. We remain on track for calendar Q4 samples and currently forecasting volume production by the second half of calendar 2015.
Beyond innovation at the technology level, we continue to add controller and firmware resources that are helping to accelerate product development, enhance the quality of our PCIe, SAS, and SATA-based SSD products. Led by the launch of our M500DC enterprise product, which targets data center applications, our enterprise SSD revenue was up 79% quarter-over-quarter. This investment in system-level solutions has led new customer wins in server, network security, and cloud and video streaming segments in Q4. In addition to chip and system-level investments, we continue to recruit strong level sales and marketing personnel to position our go-to-market engine for future success. We are investing in expanding our customer engagement support for what is increasingly becoming more of a design and custom solution-type relationship with key accounts. We are continuing to diversify our NAND business into more attractive end market applications.
As an example, revenue for NAND sold in the mobile segment was up 18% quarter-over-quarter. Coupling NAND with DRAM in the form of eMCPs is a high-growth opportunity. eMCPs are essentially low-power DRAM packaged with a managed NAND product all behind a controller. We believe this solution is a great example of the value of Micron's portfolio. We continue to grow share in our consumer product segments with Lexar-branded memory cards and USB products, gaining share in the U.S. and international markets. We remain bullish on the long-term market opportunity in NAND, both in terms of the market outlook and our ability to improve our operating performance. Upside potential in our NAND business will be driven by the chip-level innovation, including TLC and 3D NAND, investment in system-level capabilities, additional organization capabilities, and focused execution on growth and diversifying end markets.
We are very pleased with the performance of our mobile business unit. MBU revenue came in at $909 million, with a 22% operating margin. We continue to see strong demand in mobile. The iPhone 6 launch appears to be going quite well and reflective of consumer appetite for improving smartphone application features and performance. On the high-end, the Samsung Note 4 is now shipping with three gigabytes of low-power DRAM, as is the Xiaomi Mi 4 product line as well. In NAND, the higher-end configurations have shifted from 32 gigabytes and 64 gigabytes now to 64 gigabytes and 128 gigabytes. We are currently forecasting NAND growth in the mobile segment of greater than 3x year-over-year for fiscal year 2015.
The low to mid-range price smartphone market has driven additional memory content as well. Even the entry-level segment is evolving from phones with virtually no DRAM to new products such as the Android One, which is 1 GB of low-power DRAM. This is a $100 smartphone targeting 5 billion users in emerging markets. We are focused on our diversified set of mobile customers and continue to balance our PC DRAM and mobile DRAM capacity to optimize for long-term profitability. On the product front, we are growing our managed NAND business with increased shipments of eMCPs, as I mentioned earlier. The rapid adoption of eMCPs for the low to mid-end smartphone, which is the fastest-growing segment today, represents a unique opportunity for Micron.
Our known good die flow from our MMJ fab, formerly Elpida, coupled with Micron's industry-leading 16-nanometer NAND technology, enables us to capture share due to the accelerating market shift from eMCP, so to the eMCP from eMMC. Over the next year, we are forecasting that roughly 25% of our overall mobile revenue will come from memory combined with a controller. We believe that companies like Micron, who have the portfolio of managed DRAM, will have a distinct advantage in the long run in the mobile segment. In addition to managed NAND, we are currently working with our chip partners to enable low-power DDR4 for future qualification in high-performance applications for mobile. I want to congratulate the mobile team for their successful turnaround of a business that within the last two years was losing money on the bottom line.
We feel that the combination of increasing memory content in mobile devices and a declining rate of mobile industry supply growth due to the attractiveness of other relatively higher-margin segments will lead to a healthy overall supply and demand outlook for mobile. We look forward to the team's continued strong execution. Our Embedded Business Unit, or EBU, set a record of $1.774 billion in fiscal year 2014, which represented a 39% year-over-year growth. Our fourth quarter revenue of $476 million reflected a seventh straight quarter of record shipments for embedded. Automotive revenues were 40% increase year-on-year. The automotive segment continues to benefit from memory content growth fueled by both infotainment and advanced driver assistance systems. Our commitment to the unique needs of this market in areas such as quality, reliability, longevity, and service have enabled us to strengthen our market leadership in Q4.
The broad category of industrial and multi-market, otherwise referred to as IMM, achieved record revenue for both the quarter and the year. Year-on-year revenue growth was 10.8%, driven by continued growth in factory automation, machine-to-machine systems, and aerospace and defense. As we see strong demand growth in areas such as automotive, entertainment, consumer electronics, connected smart home devices, and machine-to-machine systems, we remain bullish on our EBU performance for fiscal year 2015. Operationally, we had a tremendous year. In a year where we shipped a more diversified set of products with more complex packaging and custom configurations to a broader set of end customers, we were able to reduce overall inventory and drive our turns 18% higher compared to the end of fiscal year 2013. This is all the more impressive given the significantly larger revenue scale of our business than which we left 2013.
Our DRAM and NAND front-end cycle times improved 25% and 18%, respectively, year-over-year. Our back-end test and assembly facilities reduced their cycle times 33% and 34% for NAND and DRAM. We see additional room for improvement in our operations during fiscal year 2015. The pricing environment for our portfolio of products remains favorable. PC pricing has remained strong with more than 90% of our computing capacity tied to OEM contract pricing, most of which with quarterly pricing agreements in place. This has been a positive effect on mobile pricing as well, having balanced our production to take advantage of PC DRAM ASPs and margins, which are currently above the mobile business. On the NAND front, pricing dynamics are relatively stable heading into the holiday build season.
We remain bullish on the long-term outlook for NAND, with increasing attach rates in both client and enterprise solid-state drives, with a strong content move from 128-gigabyte systems trending upwards to 512-gigabyte systems, as well seeing a big increase in phones. As the industry converts to 3D NAND, we think performance and costs will continue to improve, driving accelerated adoption of NAND in client, mobile, and enterprise market segments. We see a bright future in NAND and other emerging non-volatile memory technologies. In closing, I also want to congratulate our team for a great quarter and an outstanding fiscal year 2014. We look forward to a strong Q1, and we feel there's opportunity for improved performance in fiscal year 2015 as we continue to drive our memory solutions to higher value-added customers and segments. With that, I will hand it back over to Chip.
Thanks, Mark. We will now take questions from callers. Just a reminder, if you are using a speakerphone, please do pick up the handset when asking a question so that we can hear you clearly, and please open up the lines.
Ladies and gentlemen, if you have a question at this time, please press star then the number one key on your touchtone telephone. If your question has been answered or you wish to remove yourself from the queue, please press the pound key. Our first question comes from the line of C.J. Muse with ISI Group. Your line is open.
Yeah, good afternoon. Thank you for taking the question. I guess first question is around DRAM pricing. You talked about greater share in terms of contract and increased duration. Would love to get an update on what percentages is out two, three months, and what kind of visibility you have into the new calendar year from where we are today.
Well, I'm going to be careful about speculating too far out into the calendar year. My commentary earlier tried to address that we have our top three PC OEM customers on a three-month pricing agreement that ends at the end of September. The current indication is that we're going to have another discussion around and probably end on a three-month pricing agreement for the calendar Q4 period. As current market suggests, we think those conversations will be in line with stable or slightly improving pricing.
Okay, helpful. On the cost side, can you provide an update on where you are on the DRAM 25-nanometer ramp in terms of percentage output as you move into the November quarter and how we should think about that ramp beyond and the cost savings achieved?
Okay. Yeah. Just for review purposes, we currently have 25-nanometer product coming out of both Hiroshima, which is targeted primarily at mobile. We also have our Rexchip facility in production with 25-nanometer as well aimed at mobile. Today, as we sit going into the quarter, 25-nanometer product of DDR3 will be roughly about 25% in the quarter out of Hiroshima. Sorry, 50% in Hiroshima and out of Rexchip, it will be roughly 25%.
Great. Thank you.
Our next question comes from line of Betsy Van Hees with Wedbush. Your line is open.
Good afternoon, congratulations on another great quarter and guidance. I was wondering if we could talk a little bit more about the DRAM business and what percentage is PC DRAM, server DRAM, and mobile DRAM. Then maybe within that, we could talk about what 25-nanometer and how that's shaping up.
Sure. When you look at our DRAM business today, the best way to think about it is mobile is roughly about 25% of overall DRAM. Computes roughly around 30% of DRAM. Then, if you look at the specialty business, you asked about server. Server's roughly about just below 20%. Those are kind of the three areas you asked about.
Okay, great. Thanks so much. That's very helpful. As we look at the percentage of your DRAM mix today, how does that compare to last year at this time? As you guys are looking forward, you've been talking a lot about optimizing your mix. Are we going to see a shift more to PC DRAM, which seems to be the better lower cost for you and better pricing advantage?
Well, Betsy, I can tell you where we were last year. I said that compute was in the kind of mid to high 30%. It was slightly higher last year, but still under 40%. Mobile was lower last year coming out of fiscal year 2013. Increasing as we left the year. The server business, it was slightly higher. By the way, the question on PC capacity and how we're optimizing, it's not just the mobile to PC dynamic, it's also some of the low end of the server business that we've shifted capacity back to meet the demand of the PC business.
Thanks, Mark. That's really helpful. My last question, and I'll jump out, is on DDR4. You talked about a very nice pricing premium, I was wondering if you could give us kind of a range of what type of % pricing premium we're going to see. As we're looking at DDR3 and the conversion to DDR4, how are you guys thinking that's going to play out over the next couple of years?
If it's okay, we're not going to comment on the actual pricing premium, suffice to say, our production decisions will be driven around ROA and margin, not any market share goals necessarily.
Okay, great. Thanks again, Congratulations on a great quarter and guidance.
Thank you.
Our next question comes from the line of Romit Shah with Nomura Securities. Your line is open.
Thanks, Great quarter. I just wanted to get your perspective on DRAM content in mobile. We're seeing the iPhone 6 use one gigabyte of DRAM. Some people were expecting DRAM content to be higher. Just your perspective on that. Is that just sort of specific to the Apple ecosystem because we're seeing in Android, DRAM content go up? Or is it because DRAM pricing has gone up so much that key OEMs are now trying to minimize the amount of memory in the phone?
I think it's a little bit of both, to tell you the truth. If you look at the overall, so for the category, for the mobile segment, DRAM content per unit is up pretty dramatically. The Apple ecosystem is more efficient, but I also would suggest that given how tight the memory business is, there probably wasn't a lot of room for them to be able to address higher-end configurations.
Okay, that's helpful. Then I guess, just back to your outlook for 20%-25% bit growth. What is your assumption for mobile within that?
As we've mentioned on prior calls, we think that mobile will be about mid-40s% next year in terms of bit growth. You also should remember that we've stated, and we'll state again today, we're going to manage this from a balanced portfolio perspective and really look at the returns piece of the business. Mobile, as we said, we don't necessarily need to be the biggest supplier of memory. We want to be the most profitable supplier of memory and helping our customers innovate. So it's going to be roughly in that area, but we'll see over time how it plays out.
Then just last question. Is there any revenue impact from the extra weeks?
This is Ron. There should be, and we contemplated that in our revenue guidance of $4.45 billion-$4.7 billion. Theoretically, you get about one-seventh additional cost on OpEx and one-seventh additional revenue, and the COGS moves with the revenue. Yeah, one-fourteenth, excuse me, 7%. All I'd add is that when it's on the revenue side, you don't necessarily get linear effect from adding one week to the quarter where you have some customers who are pulling in from stock, et cetera, on a monthly basis, for example. In general, you can think of it that way, about a 7% lift.
Okay. Thank you.
Our next question comes from the line of Mark Delaney with Goldman Sachs. Your line is open.
Thanks very much for taking the question. I was hoping you can help us think a little bit about the free cash flow dynamics in your business. You talked about some of the CapEx increases that you're contemplating into fiscal 2015. Maybe you can just help us think about your ability to also grow your operating cash flows, and if you could talk to where the free cash flow margins are between your DRAM and NAND businesses.
Sure. Mark, we don't specifically go through a detail in terms of what our view in projecting operating cash flow and free cash flow. We're giving you a view out one quarter. If you look at our current trajectories and the stability we're seeing right now in the marketplace, you can extrapolate that going forward and potentially see trend lines in the similar range. We don't give guidance beyond one quarter out.
We've also given you the CapEx guidance. You have a view of that. It is up some from last year, $3.6 billion-$4 billion. Given current trajectories, we certainly would have that well covered with operating cash flow. If you look at the DRAM versus NAND breakdown, we don't break out operating cash flow and free cash flow by technology. I don't have a view for you on that.
Mark, let me add, just from a business perspective, clearly we've got a lot of margin opportunities still in our business. Mark talked about some of the pricing trends. I talked about where we see supply going. Internally, relative to execution, we've talked about what we can do on the NAND side relative to adjusting our mix on TLC and eMMC, the mobile space, SSDs, et cetera. We think we got a lot of levers in our business here, and we're just very focused on executing in what's a pretty good market environment.
Yeah, I appreciate that perspective. For a follow-up, I was hoping to talk on the NAND business specifically. I know at the Analyst Day, and in some recent conferences, you guys have talked about some improvements that you're expecting in NAND over the next two to six quarters, and I understand TLC, for example, is part of that. Can you just talk in a little bit more depth about how you're progressing with that? Maybe you could tie in some of the comments you made in your prepared remarks on SSDs and how SSDs are impacting both pricing and cost, and then the overall profitability levels of your NAND business as we think out over the next few quarters.
Well, I'll turn it over to Mark Adams here in a second. Let me just, on NAND, we'll have 16-nanometer samples out in calendar Q4 for TLC. We'll have a TLC SSD in the marketplace in the summer, maybe slightly before that. We continue to have pretty good growth in our penetration of the client SSD market, as well as the mobile segment generally. Mark, you want to add any color beyond that?
Well, I think that it's important to know that in the cost curves in our NAND business, as SSDs continue to grow as a percentage of our overall NAND business, that'll add cost of goods impact, because we get things like controllers and PCBs and the rest of the bill of materials of an SSD that go into our COGS calculation. Secondly, I'd like to emphasize that, as you can see by the growth quarter-over-quarter for the last two quarters, market access is not a problem. People are leaning on Micron to drive their client SSD and enterprise opportunities. We continue to invest in, as these businesses are high-value businesses, they're not ones that you necessarily can develop overnight. We're very happy about the progress in the quarter, especially as I mentioned, enterprise sales up 79% quarter-over-quarter, well ahead of market growth.
Thank you very much.
Our next question comes from the line of John Pitzer with Credit Suisse. Your line is open.
Good afternoon, guys. Thanks for letting me ask the question. Guys, the first question I want to talk to you a little bit about is the relationship with Inotera. They're clearly providing a lot of CapEx in this move to 20-nanometer, but you're providing a lot of IP to help them get there. I'm just kind of curious, as I think about the pricing arrangement of market minus, they're going to get pretty good cost savings with the move to 20 over the next year or so and probably grow bits at a much faster rate than the overall Micron DRAM business. How do I think about your benefit on the gross margin line as they bring down their cost curve?
Obviously, it's a complicated relationship. By way of background, you're right, we provide a lot of advanced process technology products, customer access, customer service, a lot of things to the relationship. What we get back is the full output of Inotera, the market minus transfer price. We also get back equity gains on the bottom line. Without going into too much detail, John, the supply agreement, it's a three-year agreement. It calls for annual renegotiation. Any changes to the agreement will be negotiated with the company, between Micron and Inotera and its major shareholders. We greatly value the strategic relationship with Inotera and Formosa, and we intend to work closely with them to make sure that it's a long-term sustainable relationship where each party gets a reasonable return given what they're bringing to the relationship. Not much to add to that.
Got it. Then guys, as my follow-up may be for Ron. Ron, can you remind me again on the currency front, the yen-dollar relationship, I think you've historically said every one-point move is about a $5 million to $7 million hit to operating margins or benefit to operating profit, sorry. Is that still the correct way to think about it?
Yeah, John. In terms of the yen, which is moving a lot just very recently, as we all know, the comment was on operating spending or general spending, yen-based spending. Today, our calculus is that our operating spend is impacted about, which is not just OpEx, but all of our yen-based cost of sales activities too, are impacted about $3 million to $4 million per one yen per quarter. Most of the yen-based spending is in COGS, actually, so the impact is delayed about a quarter based on our inventory flow-through, so you need to keep that in mind.
We have a lower level of yen-denominated costs today, especially now that we have converted MMJ to U.S. functional currency. We also have a yen-neutral balance sheet, so that the metric to use now is $3 million to $4 million per one yen change per quarter in our overall cost structure. We contemplated the current levels of the yen in our guidance.
Ron, if I could sneak a quick one in. The 30% of CapEx that's non-capacity related, how do we think about the depreciation schedule on that CapEx? Is it similar to the capacity-related CapEx, or is it different?
The CapEx that's not. You mean the infrastructure in our front end, back end? Yes, it's the same structure.
Perfect. Thanks, guys. Appreciate it, and congratulations on the good results.
Our next question comes from the line of Rajvindra Gill with Needham & Company. Your line is open.
Yes, thanks. Just echoing the congrats on good results and guidance. With respect to your outlook for DRAM supply growth next year, fairly favorable. Could you perhaps maybe elaborate on what you're seeing with your competitors, or if you have a view on what your competitors are doing with their respective capacity, and what's your thoughts on their overall projection of supply and how they're managing DRAM supply as well?
We don't have any better insight into what the competitors are doing than you do. We read what they say in their earnings announcements and in the press, et cetera. What we understand is that Samsung's had a relatively effective 20-nanometer ramp and pulled forward some of the gains that they otherwise were going to have a little bit later, in terms of bit production. We have a sense as to what they're doing to replace wafers lost through that 20-nanometer conversion based on what we've read in analyst and market reports. We continue to believe that the supply growth is in that range given those reported incremental wafers coming from all the various parties in the marketplace.
In terms of your overall gross margins, the DRAM margins continue to kind of drift higher in the pricing commentary. DRAM is very positive. When do you think we're going to start to hit an inflection point on the NAND gross margins? I know you talked about TLC kind of moving forward. What percentage do you think of the output exiting next year will be on TLC, given the fact that there's a big cost delta between MLC and TLC? More specifically, on the NAND gross margins, what's your long-term view?
We're still very early in the ramp and deployment. I just mentioned we'll have a component in the marketplace next calendar quarter or in calendar Q4. It kind of depends on how successful we are with market adoption and what goes on in relative pricing in all the various markets we serve. We're still going to be looking at margin and long-term sustainability of that margin. At least we'll be in a position to address those markets as we move through the year, and we think we'll see some continued deployment.
Thank you.
Our next question comes from the line of Vijay Rakesh with Jernity. Your line is open.
Hi, guys. Congratulations on a solid guide. Looks like your earnings are approaching $1 a quarter here pretty soon. I had two questions. On the DRAM side, how do you see the 20-nanometer mix ramp through 2015? I'll have the next question after you answer that. Thanks.
Our DRAM ramp in 2015 won't be significant volume until the second half calendar year. You should see some volume show up in Q4 of our fiscal year.
Got it. Okay. On the NAND side, obviously, you are going to TLC and 16-nanometer. When you look at the mix between client and enterprise, on the SSD side especially, how do you see that mix playing out? What's the mix today, and how do you see that playing out exiting next year?
Today, our mix is roughly two-thirds to 75% client and another one-third, 25% to the third enterprise. We think proportionally that's in the ballpark. A lot of it depends on market conditions, and it'd be tougher to forecast out that far how much that would change from where it is today.
Great. Okay, thanks.
Our next question comes from the line of Alex Gauna with JMP Securities. Your line is open.
Thanks. Nice quarter. I was wondering if you could give us an idea of about how much of your DRAM capacity is moving through the spot market right now. In considering these contracts for the next quarter in the PC market, how you feel about leaving some flex capacity for price changes. Thanks.
Well, I mentioned in my script earlier, less than 10% of our capacity goes to the spot market in DRAM. So the other way to think about it is greater than 90% go to OEM customers. That being said, as it relates to kind of how we see pricing and our willingness to hold tight, we feel pretty confident where pricing is going in the quarter. We see stable-to-up pricing.
Holding inventory and all that stuff, we've commented before. We're in this business to make money, and it's a return decision. We're not looking to liquidate anything. We're in good shape, and that's the way we're going to act.
Okay. I'm wondering, you made some comments around the strength you're seeing in the server market, and I'm wondering your thoughts on how enduring that is, how much might be just the near-term benefit from the Romley upgrade cycle. Are there legs to this going forward that you can see? Thanks.
Yeah. The argument that you just posed is that there's more server units being sold, and our proposition is that there's more density, more memory bits going into every server. If, in fact, you're right, we'll see even faster memory growth in the server segment. We're talking about people wanting to put more capacity in a relatively modest single-digit growth server business today.
All right. One more, if I could. I'm wondering when we get into the back part of next year, and you've got your 16-nanometer TLC, where do you expect the competition to be, and how do you expect your products to stack up versus that?
It's hard for us to speculate because we're not inside those companies and their product discussions. Remember something, we have made significant investments, and we're very excited about our 3D offering, and we've talked about seeing some 3D production out towards the end of our fiscal year, calendar Q3 next year. We think there's a good combination in our product portfolio in the second half of the calendar year, next year, for consumer and mobile applications requiring 16-nanometer TLC, as well as entry-level client devices. We also are pretty bullish about our 3D development and efforts, and we think we're going to have a differentiated solution there on the high-end for high-performance NAND solutions.
Okay, great. Congratulations again.
Thank you.
Our next question comes from the line of Mark Newman with Bernstein. Your line is open.
Hi. Thanks for taking the question. Actually, I had a question on a little bit related to mix and how that impacts gross margin. Clearly, it looks like enterprise NAND seems to be the biggest area, enterprise SSD seems to be the biggest area of growth right now. You said 79% quarter-on-quarter growth. Can you talk about how much different the gross margin is for enterprise SSD versus the rest of the business, and how that might be impacting the gross margin going forwards? I have a follow-up question on your overall SSD business, both client side and enterprise side. What percentage of that uses your own controller versus some kind of external controller solution?
Let me talk to the first question around enterprise and the effect on the overall business. Without giving too much away for competitive reasons, just suffice to say that the enterprise SSD business is an above-average gross margin business for Micron, and we're continuing to develop products and try to maximize our opportunity there. As we grow that business, it should have a positive impact, all things being equal. As it relates to your second question about internal controllers, we continue to invest internally in resources for both the clients and enterprise market. Today, we have a hybrid development approach, which is primarily around in-house development for value-add controller systems. For the more commodity-type products, we look and work with third parties.
Remember, that doesn't mean we don't add value because we've got firmware development tests and all that that goes on in the background with these partners. If you're looking for a rough proxy, I'll ask Ivan and Kip to follow up, but my suspicion is we're somewhere in the 70%-75% of external controllers, maybe closer to 80, and about 20% internal controller development. We'll follow up, Mark, with you with the exact numbers.
That's great. Thanks a lot for that, Mark. Then on DRAM, considering the strong pricing guidance you gave for this current quarter we're going into right now, where is pricing strongest, and how do you think mix might be changing? I'm just wondering if the PC DRAM pricing seems to be pretty strong, and it seems to have higher margin than average, I understand, for DRAM. I wonder if that percentage of mix might change going forward. How do you think about that?
There's two components to that. There's the supply element, right, Mark? Then there's the overall end market behavior in terms of demand. As it sits today, we've talked about this in prior calls, we have a certain amount of flex capacity to move between PC and mobile, and we probably haven't emphasized enough, we also have the discretion to move some of that PC capacity and server capacity back and forth. Given the strength of the PC market, we've optimized around the PC business on a relative basis. Remember, we want to be in all three business for the long term. As it relates to going forward, it's a tougher call in terms of trying to predict that. PC strength clearly today is helping lift mobile pricing and server pricing, which led to pretty favorable results in the quarter.
Just follow up. Where do you think pricing is strongest now? We haven't really seen a huge increase in mobile DRAM prices yet like we have for PC DRAM prices, and I just wonder if we might see some further strength in mobile DRAM prices going forward.
Well, I think that given how we've looked at our, what I'd say, variable capacity, I think that's a distinct possibility, assuming demand plays out the way we think it will. I think that mobile pricing is likely to trend stable to upwards in the quarter given some of the new launches and some of the configuration and density improvements.
All right. Thanks very much, and great quarter.
Thanks, Mark.
Thanks, Mark. Appreciate it. We'd like to thank everyone for participating in the call today, and our apologies go out to those that are still on the line we didn't get a chance to chat with. If you'd 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 the 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 fourth quarter 2014 financial release conference call. You may now disconnect.