And welcome to the LAM Research Corporation September 2016 Conference Call. At this time, I would like to turn the conference over to Sadhir Kumar. Please go ahead.
Yes, thank you and good afternoon everyone and welcome to the Lam Research quarterly earnings conference call. With me today are Martin Nantzis, President and Chief Executive Officer and Doug Behringer, Executive Vice President and Chief Financial Officer. During today's call, we'll share our outlook on the business environment, review our financial results for the September 2016 quarter, our outlook for the December 2016 quarter. The press release detailing our financial results was distributed a little after 1 pm Pacific Time this afternoon. It can also be found on the Investor Relations section of the company's website along with the presentation slides that accompany today's call.
Today's presentation and Q and A includes forward looking statements that are subject to risks and uncertainties reflected in the risk factors disclosures of our SEC public filings. Please see the accompanying slide in the presentation for additional information. Today's discussion of our financial results will be presented on a non GAAP financial basis unless otherwise specified. A detailed reconciliation between GAAP and non GAAP results can be found in today's earnings press release. This call is scheduled to last until 3 pm Pacific Time.
And as always, we ask that you limit your questions to 1 per firm with a brief follow-up, so we can accommodate as many questions as possible. As a reminder, the replay of this call will be available later this afternoon on our website. And with that, let me hand the call over to Mark.
Good afternoon, and thank you all for joining our regularly scheduled earnings conference call. Anticipating our upcoming investor and analyst meeting planned for New York on November 18, I will keep my prepared comments shorter than usual. The focus today will include financial, operational and market highlights for the September quarter and calendar year 2016, an update on recent changes in the industry environments and initial perspectives we have on calendar 2017. Doug will then provide details of our financials and summarize guidance for the December quarter before we open the call for questions. Demonstrating the constancy of our focus, quality of alignments in the company and commitments to invest in the future of the company for these last several years, Lam delivered another outstanding quarter, reporting record shipments, including a record high foundry shipments, record total company revenues and non GAAP net income.
As is always the case, how you do something is at least as important as what you do, more so in the context of our multiyear growth and value aspirations. Accordingly, we have a permanent focus on customer oriented core values and culture and an extraordinary commitment to operational execution and company scaling. Our supply chain, our factories, our field service We are on track to deliver our customers a combination of schedule flexibility and on time performance for the year that is a significant enhancement of capabilities. And as shown by our guidance today, we are gearing up for an additional growth in December. Our differentiated product pipeline momentum continues at pace, most recently with metal ALD in 3 d NAND and atomic level etch in dielectric foundry applications.
In addition, our business teams are achieving more than 90 percent success in critical defenses and new market application penetrations combined, which is extremely rewarding. This is an exciting time to be part of the semiconductor ecosystem, no doubts. In the world of foundry, logic, the demonstrated performance benefits of 10 nanometer technology and commitments of our customers and reportedly more than 20 of their customers to the near term 7 nanometer roadmap are strong endorsements of the diversity of cloud, mobility and high performance computing enabled consumer and enterprise applications. What is clear in this segment also is the importance of 3 d architecture at both the chip and systems integration level. Innovation in our technology and product roadmap for the transistor, interconnect and also advanced packaging applications is a critical priority for systematically strengthening our strategic relevance.
We believe Lam is one of very few companies gaining share through this round of foundry, logic, technology node transitions. Our segments exposure continues to be more balanced and an objective and results we are particularly pleased with. Worthy of notes, in a world where we estimate foundry wafer fabrication equipment spending grows 25% year over year, 2016 compared to 2015. We are communicating to you today with our presented results and December guidance, a view that our foundry shipments for the year increased by approximately 40%. Illustratively, 48 layer 3 d NAND is enabling the 2 56 gigabyte high density storage that is required to power high resolution image capture and dual cameras in smartphone devices.
Similarly, in the data center, 3 d NAND based SSDs allow for 100 of terabytes of high speed storage to be located within the server itself, enabling cost efficient scale from cloud based services. Looking ahead, the roadmap is well defined and looks very promising for 3 d NAND. At the recent Flash Memory Summit in Santa Clara, our customers highlighted the numerous design and integration options not previously available with planar technology such as scaling layer accounts and the ability to transition from 3 level cells to quad level cells and leveraging pitch and density scaling to target different storage end markets. Lam was privileged to be the only major semiconductor equipment maker invited to present at the conference, independent recognition of our critical role in the 3 d NAND inflection and the commentary both on the opportunity and responsibility for Lam that is shared with our customers. Long term WFE growth is only made possible with end market demand catalysts and process technology inflections.
Combined with these two drivers, our roadmap and multiyear execution make possible the 5th consecutive year in 2016 where our SAM has outperformed WFE. Looking forward, we have targets and strategies that we will specifically discuss in November to enable and participate in these outperformance trends for many years to come. Our ability to capitalize on this extraordinary opportunity is all about leadership, execution and is measured ultimately by customer trust and competitive differentiation. As illustrations, in conductor etch, our Kiyo family with its control technology provides the most technically advanced production proven etching performance critical for logic and DRAM multi patterning in the industry. Lam has the broadest and strongest conductor edge position, especially in 3 d NANDs where KEO has now been qualified for production at all NAND customers for a majority of applications, which we believe achieved position in more than 2 thirds of the available market.
In dielectric etch, we continue to make several technology and productivity enhancements to our Flex family, including most recently the release of atomic level etch capability. Our ALE innovation features a proprietary advanced mixed mode pulsing capability, delivers a 2x improvement in selectivity versus conventional etching while providing atomic level control over variability, critical to address issues like RC delay for advanced foundry and logic process nodes. The Flex Series products are seeing substantial momentum in 3 d NAND and are now qualified as production tool of record for critical high aspect ratio applications at all major customers. In deposition, our Altus Tungsten Fill and Vector Dielectric Deposition Systems are seeing very strong momentum that is set to continue into next year. 2017 will be the year of 64 layer 3 d NAND and LAN's product innovations and application share momentum position us well to improve our market share as we head into next year.
Complementing our leading edge and critical application strength, the performance of more than 40,000 LAN process modules installed provides a strong platform for overall business momentum and is a key focus of our customer support business group. This business is on track for another record year. More important than reported growth in any one year, however, is the increased confidence of sustaining long term profitable growth trajectory of the company. In the last several years, we have fundamentally improved our capability to support our customers through local manufacturing and warehousing, innovative technology and productivity upgrade solutions, refurbished equipment sales and service and product solutions to overall fab productivity improvements. These investments are paying dividends, again, more to follow at our investor event.
Now let me provide a short overview of customer equipment spending trends for calendar 2016 and our preliminary expectations for 2017. Outlook for global economic growth and unit shipments for PCs and smartphones have remained stable since our last reports. Increasingly more importance than unit demand for traditional consumer computing devices, however, is the significant growth in content for leading edge memory and logic devices in smartphones and servers. This dynamic is being driven by the high end by the end market demand inflections I spoke of earlier and is powering our SAM growth outperformance opportunity. WFE spending for 2016 is now tracking slightly above the high end of our previously communicated range, driven by stronger foundry and 3 d NAND CapEx investments.
Sequentially for calendar 2016, NAND WFE is now on track to grow over 40%, foundry over 20% and expectations for DRAM CapEx remain unchanged at down approximately 40% year over year. Momentum and commitments continues to build for 3 d NAND where we now expect total industry shipped capacity to reach approximately 425,000 wafer starts per month by the end of this calendar year. We are confident of sustainable positive long term trends in this segment. At the same time, our strength in foundrylogic continues to build where we are realizing the benefits of multiyear efforts through customer spending associated with additional qualifications and PCOR wins for 10 and 7 nanometer devices. Despite the challenges of DRAM and certain pockets of logic spending in 2016, despite also the memory versus logic segment headwinds for the company relative to our peers this year.
We are on track to grow several company shipments meaningfully for the 5th straight calendar year at a CAGR of approximately 18% over that extended period. As we look into 2017, we are encouraged by the strong momentum in 3 d NAND. For numerous reasons, we are also more convinced and convicted in our belief that DRAM supply and DRAM sorry, DRAM supply and demand conditions will continue to improve as we previously indicated. As a result, we have an upward bias to spending expectations for DRAM and NAND in 2017. We expect flat to a modest pullback in logicfoundry spending as customers digest the 10 nanometer investments made in 2016 and commence initial 7 nanometer roadmaps.
Overall, we are modeling slightly higher total WFE spend in 2017 with memory versus logic segments tailwinds for the company once more. Again, our objective will be to outgrow WFE in 2017. To summarize, fundamental to our strong performance has been our strategy to partner closely with our customers to help address their most tough and difficult challenges. Despite, as we conveyed a couple of weeks ago, the disappointments at not being able to deliver to our customers the potential of unmatched capability and innovation from Lam and KLA 10 Core combined, we're very excited by the demonstrated multiyear strength and inflection driven outperformance future opportunity for Lam Research. With continued execution, a robust product pipeline, overall stronger installed base and market share positions and increased strategic relevance to our customers, we believe the innovation to support long term value creation for all stakeholders is a valid ambition for Lam Research.
Before turning the call over for additional details on our financial performance and outlook, I would like to extend my genuine thanks to our customers and employees who together make the exciting story of Lam Research possible. Their support and encouragement inspires us each day to achieve our full potential. Doug?
Okay. Thank you, Martin. Good afternoon, everyone, and thank you for joining us on the call today. The September quarter results reflect another quarter of solid execution and record results. In the quarter shipments, revenue, net income and cash from operations were at the highest level we delivered in the history of Lam Research.
We're obviously very pleased with what we achieved this quarter. Shipments in the September quarter were $1,708,000,000 up approximately 8% compared to the June quarter and at a record level for the company. The combined memory segment made up 56% of system shipments and that was down from 66% in the prior quarter. Non volatile memory shipments contributed 43% of the system shipments and this was down from 51% in the June quarter. 3 d NAND investments are continuing at a healthy level with customers committed to their plans for scaling the technology to reduce cost, enhance performance and increase output in response to strong customer demand.
Our deposition in edge products are critical to enabling the scaling for our customers. We are extremely well positioned with the PTOR decisions made by our customers in what may be the largest multiyear growth opportunity in semiconductors. Customer discipline remains intact in the DRAM segment with shipments making up 13% of total system shipments compared to 15% in the prior quarter. There appears to be a stabilization of the supply demand balance and we're seeing positive momentum in the marketplace as a result. Demand is strengthening led by content growth in both smartphones and servers.
Spot pricing is showing improvement. System shipments into the foundry segment was a bright spot as Martin mentioned and increased to 36%, which was up from 27% in the June quarter. The system shipment dollars in foundry represented a record high level for Lam Research. Foundry spend was biased toward investments to execute 10 nanometer ramp and 7 nanometer pilot projects. We also saw continued spend predominantly in the China region at the 28 nanometer and above nodes.
And finally, the logic and other segment came in at 8% of system shipments, up from 7% in the June quarter. Logic spending was largely in support of conversions to the 10 nanometer node. We set a new record for revenue in the September quarter. Revenue came in at $1,632,000,000 which was a sequential increase of 6% and a little bit above the midpoint of our guidance. Gross margin was a little lower than the midpoint of guidance at 45.2%.
As I previously stated, we expect to see some quarter to quarter variability in our gross margin due to factors like business volumes and mix. And I'll remind you the financial model is the best reference to help you analyze our ongoing financial performance. And I look forward to sharing an update to the financial model with you at our Investor Day in November. Our operating expenses for the quarter were $372,000,000 which was an increase of 3% from $361,000,000 in the June quarter and it held steady at about 23% of revenue. OpEx spending allocated to R and D was again at benchmark levels of 63% of total spending in the quarter.
Operating income for the quarter was again strong at $366,000,000 an increase of 2% compared to the 3 $59,000,000 in the June quarter. Operating margin decreased slightly to 22.4%, down from 23.2% in the prior quarter, primarily due to the lower gross margins. The September quarter tax rate came in at approximately 12%, which was a little bit higher than the 10% rate last quarter, primarily due to changes in the level and mix of jurisdictional income. A rate in the low to mid teens for the December quarter would be a reasonable number for you to use in your modeling. Earnings per share for the September quarter were $1.81 based on a share count of approximately 178 1,000,000 shares.
The share count includes dilution from both the 2018 and 2001 convertible notes and the remaining 2016 convertible note warrants, with a total dilutive impact of about 15,000,000 shares on a non GAAP basis. As the stock price continues to rise, the convertible notes will continue to add to the diluted share count, albeit at a decreasing rate of increase. Dilution schedules for the convertible notes are available on our Investor Relations website to help you with your modeling. We returned $0.30 per share for a total of $48,000,000 in dividend distributions to our shareholders in the quarter. There were no share repurchases during the quarter.
Let me now turn to the balance sheet. At the end of the September quarter, we had $7,470,000,000 in cash on the balance sheet. In line with the plans we stated during our call earlier this month, at this point, we have redeemed the $600,000,000 20.23 senior notes and the $1,000,000,000 2026 senior notes under the special mandatory redemption provision of those notes. The $800,000,000 2021 senior notes will remain outstanding and will be used for general corporate purposes and other purposes as can be found in the prospectus. If our pro form a for the cash we returned to bondholders in October, we would have finished the quarter with $5,830,000,000 on the balance sheet.
Roughly 40% of this cash is currently domestically available. And I'll remind you, we continue to generate between 20% 25% of cash domestically on an ongoing basis. Cash generation continues to be healthy at the company. In the September quarter, we generated $473,000,000 in cash from operations, which was an increase of 12% sequentially and an all time quarterly high for the company. Day sales outstanding for the period improved to 72 days from 74 days with the shipment profile slightly more linear than it was in June.
Inventory turns improved from 3.5 to 3.9 times. At the end of the September quarter, the deferred revenue balance was $704,000,000 a 27 excuse me, a 24% increase quarter over quarter. That number excludes approximately $65,000,000 from shipments Japan shipments remain as inventory carried at cost on the balance sheet. Company non cash expenses during the quarter included the following: $39,000,000 for equity compensation, dollars 39,000,000 for amortization and $36,000,000 for depreciation. Capital expenditures were $42,000,000 which was down from $52,000,000 in the June quarter.
We exited the quarter with approximately 7,800 regular full time employees. We added headcount during the quarter primarily meet customer needs for additional factory output and field support. Additionally, we added resources to continue our investment in product development. Let me now turn to our outlook for the December quarter. I'd like to provide the following non GAAP guidance.
We expect shipments of $1,850,000,000 plus or minus $75,000,000 We expect revenue of $1,840,000,000 again plus or minus $75,000,000 We expect gross margin of 46 percent plus or minus 1 percentage point. We're forecasting operating margins of 25 percent plus or minus 1 percentage point. And finally, we forecast earnings per share of $2.18 plus or minus 0 point This earnings per share number excludes roughly $37,000,000 associated with the interest expense, one time fees and accelerated costs associated with the retirement of the 2023 2026 notes as well as the term loan commitments. With our year to date results and the strong guidance I just provided for the December quarter, we anticipate that calendar year 16 will be another record year of financial outperformance for the company, a 5th consecutive year of growth outpacing that of the industry as a whole. Before transitioning to the Q and A session, I wanted to pick up on Martin's earlier comment regarding our Investor Day, which is now on November 18.
We've decided to host the event in New York to make the event more convenient for a larger group of investors and analysts that are following the company. We look forward to seeing you in New York on 18th. That concludes my prepared remarks. Operator, Martin and I would now like to open up the call for questions.
And we'll take our first question from
Krish
Sankar
with Bank of America. Please go ahead.
Yes. Hi. Thanks for taking my question. I had 2 of them, Martin. First one is, if I look at your December guidance compared to your the model that you guys have given in the past, it looks like the quarterly WFE is running at close to like $37,000,000,000 $38,000,000,000 you guys are annualizing at well over $8 in EPS.
Is it fair to assume that if next year is only going to be a modest growth to the extent you can characterize it, should we assume sequentially the next couple of quarters should decline from the December quarter? And I also had a follow-up.
Well, Krish, as is customary for us, we really don't comment specifically about quarter to quarter transitions and inflections. What we tried to provide some perspective on today are some very significant and underpinning markets references for you and for us as a company and whether it's a very favorable outlook relative to a universally adopted 3 d NAND transition at this point, or perhaps more specifically to the company, our share momentum in logic and in foundry, we feel like the objectives of the company that are currently defined by the long term financial model, we're executing kind of well within that, maybe a little ahead of that. And certainly, the expectations we have at this point for Canada 2017, slightly stronger with a bias to memory compared to this year, which bodes well for outperformance for land.
Got it. Got it. And as a follow-up, if I can ask on 3 d NAND, if you look at the NAND installed base, it's about 1,400,000 wafer starts and you guys said exiting this year, 3 d NAND will get 425,000 wafers. If 3 d NAND is going to help replace hard disk drives eventually, is the potential installed base larger than the 1,400,000 that we have or had for planar or the fact that 3 d has more density or bits in a given way for it actually offsets actual installed base growth over time. I'm just kind of curious to your viewpoint on that.
I think our viewpoint is a bit of both. I mean there's density on wafer for sure, but our view is even in the time horizon of calendar 2017, we end calendar 2017 with more wafer starts than we began the year. And just in context, the 700 reference that I think most people have for next year now when they describe the shift 3 d NAND capacity by the end of calendar 2017. That 700 probably qualified as 600. That 600 it's probably no more than 40% of the installed base as 3 d NAND capable.
So we're in the very early stages of a multiyear transition, and we think that's a pretty compelling part of the Lam Research story.
Thanks, Martin.
We'll take our next question from Stephen Chen with UBS. Please go ahead.
Great. Thanks. Congrats on the execution, Martin and Doug.
Thanks, Chip.
Just a follow-up question on the shipments. So the second half of this year's shipments looks like they'll end up 17 percent higher than the first half. I remember when you first looked at the second half of this year, you thought it was going to be flat and it's improved over the last few quarters. Could you share any color on how you see the first half of twenty seventeen versus the second half of this year?
Well, on the basis that I think from the beginning of this year, I was the idiot of the industry that was talking about the stronger second half. I'm going to reserve judgment and claim that it's too early to have commentary on the first and second half next year. I mean we've kind of put on the table for you a reference for higher WFE. I think it's kind of modest. It's probably a single digits increments year over year.
And there are a lot of moving parts that could potentially make it a very evenly distributed year, make it first half a little stronger and make it second half a little stronger. So more maybe in the analyst meeting, much more likely, more commentary on first half, second half early next year.
Okay. Thanks. And then just a question on customer concentration. Could you just share your early thoughts on the December quarter shipments, how it looks memory versus foundry logic?
Yes, Stephen, it's Doug. I think both are continuing to be relatively strong in the next quarter.
Okay. Thanks, Doug.
Yes.
And we'll take our next question from Amit Dariannani. Please go ahead with RBC Capital Markets.
Thanks. Good afternoon, guys. I guess two questions from me as well. To start off, if you think WFE is going to be up modest, call it 2%, 3% or so next year, could you just maybe help us understand what do you think your TAM or your revenue opportunity would look like in that scenario where WFE is up 2% to 4%?
We got kind of a couple of ways to answer that. One of them is, you'll take a look at the long term financial model that will be updated in November for you at least as far as kind of giving you at least another year of visibility in terms of our expectations. Another reference for the company is the stated ambition from a market share growth perspective. So we've got 2 references for you. We've got the long term share growth, the 3 to 5 percentage point references we've talked about many times, 1 to 2 percentage points through technology nodes.
And there's also a commentary on SAM expansion. And the headline for the company, if you go back to a 2012 reference point is we had a portfolio of products that was competing for about 25 percent or so of wafer fabrication equipment spending. That's in the 33%, 34 percentage range probably this year. We expect in excess of 35% to be the competitive reference point for us and make available more products to our customers and some part of it is the technology inflection in patterning and 3 d NAND, which has a natural bias to the etch and deposition segments of WFE.
Fair enough. That's helpful. If I just follow-up, the December quarter guide, obviously very strong revenue and shipments. I'm curious, I mean, do you think there's some of the pull in that's going on maybe on the memory side from your customers for the CapEx plans and that's why is somewhat better? Or is this just a reflection of good end demand?
It doesn't mean March will be down more dramatically than normal?
Well, it's
kind of really hard to kind of give opinion on kind of pull or push. I mean the reality is the year is a little stronger than we expected when we last talked to you in our earnings call. And it's a little stronger in 3 d NANDs and also in foundry. So if that's your definition of kind of pull, then I guess the answer to your question is yes. But are we communicating to you an expectation that this is pull that is over exuberant and it's at risk of creating a hole shortly thereafter?
I think the answer to to that question is no. We see compelling value propositions in 3 d NANDs and compelling value propositions in the 10 and 7 nanometer logic foundry investments. They're going to go up and down a little bit quarter to quarter and pull in and push out. But fundamentally, the long term headline is the most relevant one as far as the value of our company is concerned.
Yes. And that's why we suggested to you, we see next year a little bit stronger as well.
Fair enough. That's perfect. Thank you guys.
Yes. Thanks Amit.
We'll take our next question from C. J. Muse with Evercore. Please go ahead.
Yes, good afternoon. Thank you
for taking my question. I guess first question trying to, I guess, dig a little bit deeper into the 3 d NAND side. And curious, when you think about greenfield versus conversions in calendar 2016 and your current outlook for calendar 2017, is that a headwind or a tailwind for you, particularly when you think about next year being a year really for 64 layer count?
I don't know, it's C. J. It's terrifically difficult to answer a question like that because there's so many moving parts. This year is the headwinds most significantly because of memory and logic foundry kind of spend mix for the company. At a 3 d NAND level, the story is really positive for the company in conversions and vertical scaling, but it's also really positive compared to overall WFE and greenfield as well.
So probably it gets a little bit better as people step from greenfield to conversion. But it's really hard to give you a simple answer to a very complex modeling question.
Sure. Okay. I guess I'll take a stab at the second one then. Deferred revenues have grown considerably over the last year. And just curious, I guess, as part of that, does that really speak to your new products and new wins and timing of revenue rec?
And as part of that, when do you think you'll start to catch up and we'll see that flow through to the revenue line?
Yes, C. J, a little bit of it might be some aspect of new products, but more importantly, it's the fact that we've been growing shipments fairly consistently. And when you have a profile where shipments are on a growth trajectory, deferred revenue typically builds and revenue lags as a result. When shipments level off, you will see revenue then catch up with shipments. And so that's the way you should expect it.
The fact that shipments have been outpacing revenue is why deferred revenue is built.
Very helpful. Thank you.
Thanks, C. J.
We'll take our next question from Timothy Arcuri with Cowen and Company. Please go ahead.
Thank you very much. I guess I had two questions. First for Doug, really on cash, it seems like there's maybe $1,000,000,000 worth of excess onshore cash today. And if I look at your free cash flow, you're going to probably generate $1,500,000,000 roughly a year and that would imply maybe another $350,000,000 to $400,000,000 of onshore coming per year and only half of that is being consumed by your dividend. So clearly, you have a lot of flexibility there on the cash.
Can you talk about how you think about repurchases versus dividend?
Yes, Tim, why don't you come see me on November 18th, I'll give you some clarity at that point. I don't have anything to tell you until then.
Okay. All right. Thanks. And then I guess, Martin, I'm going to try to come at this wave of equipment run rate question a little bit differently. This year, if WFE is going to be 34,000,000,000 dollars which is now what you're saying, you're going to be about 18.5% share that's flat despite a very difficult year in DRAM.
So that's great for you. But if I use the 18.5% number in your guidance in Q4, it implies that the industry is run rating WFE at roughly 40,000,000,000 in the 4th calendar quarter, which is a pretty heady number. So I guess my question is, do you argue with that math on where the WFE run rate is for Q4? I'm just trying to see how shipments can't come down here even if next year is a good WFE year. Thanks.
I guess one of the advantages of the gray hair I have is I know how to answer that question now. So I don't even try to argue or defend or rationalize or justify because as we've said many times, Tim, we understand our SAM really well. We understand our share of SAM really well. We don't understand, frankly speaking, the SAM of the rest of the industry for products that we don't have. And we just kind of back into a WFE reference to the best of our abilities.
So I don't really have an ability to kind of defend, argue, answer a question about WFE because it's just kind of an extension of a very well thought through SAM and quite exactly what the cycles of spending are for other parts and quite exactly what their pricing strategies are. I guess you got to ask everybody in this really to kind of pull it all together. Our view is the best one we've articulated.
Okay, Martin. Thanks. You guys are usually very good at WFE, so that's why I asked. Thanks.
Thanks. No problem, Tim. Thanks.
We'll take our next question from Farhan Ahmad with Credit Suisse. Please go ahead.
Hi. Thanks for taking my question. I just had a question on NAND. In terms of your planar NAN in terms of the planar NAN installed base, it's about 70% of the overall capacity. What kind of productivity improvement or bit growth are you seeing on that installed base?
And is it fair to assume that the vast majority of bit growth going forward needs to come from 3 d NAND?
I think yes is the answer to that question. I'm sure every customer in the world is working to make more productive their planar capacity, particularly in the context that at the end of this year, the majority of the installed base for NAND is still a planar capable investment. But certainly, I think kind of the growth of the industry and the chosen technology that is now kind of universally adopted, has the focus on 3 d now.
On 3 d, yes, for sure.
Got it. And then just one question, in terms of how the WFE is tracking for last couple of years versus the semi revenue growth. I mean like last year, U. S. Semi cap companies were all up close to double digit on an if you combine all the revenues.
And this year also is like very healthy growth and next year is also very healthy growth and we aren't seeing that much semi revenue growth. Have you thought about like what's the main drivers which are giving you more of the growth relative to the semi industry?
Yes. I mean, at a company level, obviously, there are 2 basic headlines. One of them is the proportion of WFE that we're competing for, which to my earlier answer has kind of transitioned from the 25% level to the 35% level since 2012. So we're competing for $3,000,000,000 more business each year if you just want to pick a $30,000,000,000 WFE reference. We're competing for $3,500,000,000 if you want to pick a $35,000,000,000 reference.
And some part of that is a product portfolio message for the company and some part of it is the technology inflections, the patterning and 3 d NAND or 3 d device architecture transitions for the industry. That's kind of message number 1. And message number 2 is share gain. And we've had some pretty significant messages on share gain. We talked in our last earnings call, if my memory serves me right, about double digit share gain in the planar to industry, in the logic and foundry space, at least as best we can analyze, we're one of very few companies that has demonstrated share gain in logicfoundry combined in recent technology node transitions.
And we're just beginning to see the evidence of that show up in our financial statements.
Thank you. That's all I have.
Thanks, Martin.
We'll take our next question from Jagadish Iyar with Summit Redstone.
Yes. Thanks for taking my question. Martin, if I take your NAND shipments this year, give or take, you're up like almost about 75 percent. And if bit growth for next year is similar to this year and I layer in your commentary on spending, how should we think about your growth in NAND shipments as you look at 2017 given where your market share is and where the industry is transitioning to 64 layer? And will it be consistent with this year or could it be higher or lower?
Well, I mean the overall kind of story for the company in 3 d NAND, we're targeting to be more positive proportionately next year than this year and that's a commentary on our best understanding of the conversion strategies of our customers and our targets and plans to systematically introduce more products to compete for more SAM and to make more competitive the products that we have to get a bigger share of it. Now there are no kind of pendulum swings in this industry. It's pretty rare that you accomplish major swings in any one year. And so I think the performance of the company will be a continuation if we keep executing of the last couple of years and that continues to be our focus.
Okay. I have a brief follow-up. You talked about foundries trying to digest some capacity for 2017. Clearly, there is one leader there, But how should we think about other foundries trying to catch up to this preeminent leader here as they try to ramp to 10 or 7 nanometer? Thank you.
Yes. I mean, obviously, you have to ask them, not us. But clearly, the opportunity that exists from demonstrated performance benefits at the 10 and 7 nanometer technology nodes make it a very strategic and legitimate target for every foundry. And certainly from the seat we have, Everybody is focused on doing their best to get their fair share. However, they define that technology inflection and none of this is easy.
And so we support all of our customers in whatever ambition they have to the best of our ability. So I sense people are broadly focused on it and doing the best they can.
And Jagadish, I'd just point out that we're continuing to see nice investment profile at 28 nanometer and above. And obviously that stuff is much broader
based. Thank you.
We'll take our next question from Ramesh Shah with Nomura Securities.
Yes, thank you and congratulations on the strong results. Martin, I just wanted to ask as it relates to China, we've been sort of all looking for evidence that the local vendors there will start spending in both foundry and memory. And I just wanted to get your sense on that region and if it's something you're anticipating to at least hear about as we get into next year.
Yes. I mean I think for sure there are ambitious plans stated and there are active discussions, I think, between customers and potential new domestic customers in China with Lam and presumably with every other equipment company. Our position and our presence in China is historically very strong in etch and deposition both. So I think we have kind of good platform. And as best we can tell, there are up to 20 active projects certainly that we're focused on tracking in China.
And I think the spending or at least the significant spending is more likely to be a calendar 2018 statement than calendar 2017, but I think there's a decent amount of statements that towards the end of 2017, we could start seeing something emerge. So that's about the best I can give you at this point.
That's helpful. Thank you.
Thanks, Roland.
We'll take our next question from Edwin Maugh with Needham and Company. Please go ahead.
Hi, thanks for taking my question.
So I want to kind of circle
back on your foundry logic share gain that you talked about and maybe dig a little deeper and can you kind of give us some color where you are driving the share gain? Is it on the you guys have very strong patterning. Is that one area that is driving share gain? Is it around other area? I think historically you guys are very strong at Thompson.
Maybe give us some color, one step more detailed color around the shaking around foundry logic?
I think you just did a pretty good job answering your own question actually. I mean clearly the single biggest SAM expansion in foundrylogic is patterning. We've got some pretty important headlines that are company unique, I think, relative to microprocessor share gain momentum. And our presence and share gains are both etch and deposition and they're kind of front end and back end as far as interconnect as well as kind of just a level of footprint. So pretty holistic, pretty comprehensive and patterning is obviously a key driver.
Okay, great. And then just kind of focus on the foundry logic area as well. I think some of your peers are kind of more upbeat about the coming year thinking foundry logic can grow a little bit more. And you mentioned some digestion on 10 nanometer. Do you think that the trailing edge growth will not be big enough to offset any digestion in 10 nanometer and some of the customer hasn't been spending too much, you think they won't be a big spender in 'seventeen?
Well, actually our view and it's still forming, we're a long way from kind of being definitive on it. We said kind of flattish to maybe slightly down is the reference we have for foundry. We do expect that 28 nanometer and above investments in 2017 will be greater than they were in 2016. But to your point, that's a capacity addition that is much cheaper than a leading edge capacity addition. Whether it turns out to be sufficient to make a neutral or grow, kind of time will tell.
As best I can tell from customer disclosure, there's kind of a common view that capital intensity in foundry is assumed to be kind of flattish. And certainly, we're not in a position to message any different from that. There's clearly a large amount of investment going in this year for 10, and there's commitments and significant interest in what I think now is generally accepted to be a more valuable and probably more significant technology in Ota 7 and that looks like that's more of a kind of second half statement than first half at this point.
And Edwin, it's still kind of early for us to have a formulated view on 2017 as we get closer to 2017 itself. And we would normally give you more color when we're a quarter further down the road. So check back with us in a quarter, we'll give you an updated view.
That's very helpful color. Thank you.
We'll take our next question from Joe Moore with Morgan Stanley. Please go ahead.
Hi, thank you. I had a couple of questions on smaller product areas. Can you talk about advanced packaging? It seems like we're starting to see some more consumer oriented products coming out again in the end of this year. Could that be a decent driver for you guys next year?
Well, certainly systems integration and packaging are very prominent parts of the strategies of our customers to overcome the limitations of physics in the front end. And I think demonstrated performance and cost benefits kind of showing up there. It's one of 4 technology inflections that we identified 3 or 4 years ago now that are relevant to the future of our company. It was then and still is the smaller of the 4, but as time passes, the through silicon via edge position and strength that we have becomes more valuable. And obviously, the presence of the company in a connect copper position is also very strong.
So over time, I would expect the advanced packaging SAM to be more relevant. And I think from prior disclosure where we're strong and where we're not.
Okay, great. Thank you. And then also on the image sensor side, have you seen any signs of life science recovery in that segment? Thank you.
I would say not as strong right now as it has been, but reasonable evidence that strength is beginning to emerge. So whether that's the first half or second half kind of reality, we will see.
Great. Thanks so much.
Thanks, Jeff.
We'll take our next question from Weston Twigg with Pacific Crest. Please go
ahead. Yes. Hi, thanks. Just wondering first on DRAM, you said you expected DRAM WFE to be up a little bit next year. Wondering if you're seeing any signs of new DRAM capacity activity or just really just no transitions at this point?
I think it's more of a no transition statements. I mean, arguably, there are kind of 50,000 wafer starts per month contractions in 1 year and maybe additions in another year. And if I were to place a bet, I would say we probably end next year at or about the same in terms of wafer starts that we began the year. But as you know, this has been a year of significantly reduced spending. Our estimate is that DRAM investment is down 40% this year.
By the end of this year, we believe that approximately 50 percent of the installed base is 20 nanometer capable. So there's a pretty significant opportunity for value creation for the customers in terms of cost reduction and performance from continued conversion. And as it's also true, the 1x world for DRAM is just beginning in calendar 2016 and we think will be kind of the primary change to the installed base in calendar 2017. And if I were to guess, I would guess that at the end of next year, about half of the installed base in DRAM is a 20 nanometer, maybe a quarter of the installed base is 1x and a quarter of the installed base is at 25 or above. So I think that approximate segmentation of the 1,000,000 or 1,100,000 wafer starts, depending on how you calculate it is kind of what we're looking at right now.
Okay, good. And then similarly in 3 d NAND, you said WFE was biased up. And you mentioned 700,000 installed wafer starts by end of next year, but do you see that number bias higher as well, the way things are tracking?
I said just on the 700, I said that was kind of the shift capacity statement that we're coming up with and I'm seeing other people kind of say the same thing, slightly above 700, and that compares with the 425 that I referred to today in my prepared comments for the end of calendar 2016. And our estimate is that probably that's there's about 100,000 or slightly more than 100,000 wafer starts unqualified at the end of this year, probably a similar number at the end of next year. And obviously, that exists in the context of 1,400,000, 1,500,000 maybe 1,000,000 wafer starts per month of capacity by the end of next year. So a little bit more capacity and a lot more 3 d NAND, but in relative terms still 3 d NAND installed base is the minority of the installed base by the end of next year.
All right, very helpful. Thank you.
Thanks, Wes.
We'll take our next question from Harlan Sur with JPMorgan. Please go ahead.
Good afternoon and solid job on the quarterly execution and on the outlook. On the on your view on DRAM spending being higher next year, I mean, clearly, the customer base here remains pretty disciplined, very focused on profitability. I appreciate the view from the prior question on the
applications
is that SAM growth coming from?
I'm not sure I have something to hand to answer that question, Harlan. I apologize. Obviously, at a macro level, when we last reported our kind of $3,000,000,000 SAM reference, we framed the SAM growth opportunity in DRAM at the 50% to 60% level through calendar 2018. And in the patterning area specifically, which is the single biggest kind of market expansion in DRAM that's relevant to our company, we characterized 15 to 25 steps at 20 nanometer, going to 25 to 30 steps at 1x. So that's about as helpful as I can be at this moment, and we'll take that question as long as as well as I think it was kind of CJ's question earlier around the SAM for each of the 3 d NANDs implementation scenarios and we'll try and get some more information for you at the analyst meeting.
Yes, we'll update that page for you, Harlan.
All right. Thanks for the insights there. And then on the record shipment outlook, I'm wondering if you're starting to stress your manufacturing capability. I mean, as I think about the potential for more growth for the team in 2017, I'm wondering if you can just give us a view on what is the upper limit from a shipment dollars perspective that you can accommodate from a manufacturing perspective on a quarterly basis?
Yes. We actually have more flexibility than perhaps you realize. I mean, it takes a huge amount of effort. So these are not kind of 5 minutes investments. We are constantly assessing not just capacity in our own factories, but our supply chain and in our field service organizations for the installation because if you can make something, we can't install it, you have a problem.
If you can install it, but not make it, you have a problem. If you can make it, but not buy it, you have a problem. So I mean a lot of effort focused on that. We have nor will we ever have disclosure that speaks to the maximum capacity of the company, but it is more flexible. We are challenging and stretching the entire company, to be honest, but we believe we are executing with appropriate discipline and we're making investments to make sure that we have capacity to meet the we're doing we're doing a pretty good job all things considered.
Yes, Harlan, we have a phenomenal supply chain organization. These guys are just super good at what they do. So the execution of the company has been really, really good from those guys.
Thanks for the insights.
Thanks, Harlan.
We'll take our next question from Toshiya Hari with Goldman Sachs. Please go ahead.
Great. Thank you for taking my question and congrats on a very strong quarter. I just had one on gross margins. I appreciate the quarter to quarter volatility here, but I was hoping you could discuss a little bit more what drove the slight miss in Q3. And for the current quarter, you're guiding gross margins to improve, marginally despite a pretty big revenue number.
Is this as good as it gets from a gross margin perspective or is there a leverage left in the model? Thank you. Toshi, I'd just point you back to the existing disclosed financial model is the best way to be thinking about the company. And obviously, I'll give you an update on that when we get to the 18th November. I'm not going to apologize for a 30 basis point, slight below midpoint on gross margin.
It's really hard with all the moving parts and different customer mix things you have going on. There's always going to be variability. And we've been pretty consistently in that 45 to mid-forty 6 gross margin range and that's kind of where the business runs and runs pretty effectively.
Yes. I think in context, again, just to kind of put some more substance around what Doug's just said, I mean, our long term model for calendar 2015 2016 in a world of $35,000,000,000 WFE, which we think is greater than the world we're actually living in. We said we would aim for a $5,800,000,000 to 6,300,000,000 range. The sum of our reported results and guidance are at 6.3. So at a revenue level, as a commentary on SAM and share, we're performing higher than we indicated in our long term financial model.
In terms of targeted profitability levels, which includes gross margin and the investments in operating expenses, and as we said many times, our focus more is on opings and gross margin. Our long term model said we were targeting a 22 percentage point operating income and the sum of the results and guidance we've just given are 23%. The long term model also said there was a range of EPS targeted between $6.75 on by the way a lower share counts because when we put the model out, we weren't signaling to you that we would be in essentially a blackout period for 12 months because of the KLA transaction. But against that $6 to $6.75 EPS reference, the sum of our results and guidance today are about $7 So I kind of would like it in context, if I may ask for that. I think the company's performance financially is really nicely balancing growth and profitability.
And the trick here is balance. It is super easy to drive profitability up and not grow. It is really hard to architect a vision and then perform consistent with delivering profitable growth. That's our vision. And that is what we believe we're executing and delivering.
And of course, if you're a member of Lam Research, long may that continue. Thanks, Joshi.
I appreciate that. Thanks very much.
We'll take our final question from Patrick Ho with Stifel. Please go ahead.
Thank you very much. And I just got one question regarding some of the foundry logic commentary you've highlighted today. Within the context of the 10 to 7 nanometer node that appears to be one big node again, some of the share wins that you've talked about at 10 nanometers, are those simply going to transition to 7 nanometers or are there more potential application wins that you can get at the 7 nanometer on its own?
Well, as I know that you know, Patrick, nothing is quite as simple as it just transfers, But obviously, demonstrated performance of 10 nanometer for positions 1 is a very strong reference point for selection at 7, and we feel very confident about that. We are investing to achieve more than more market share 7 than we have at 10. And I believe there are some pretty compelling reasons why we can deliver. That's again consistent with our long term aspirations of the 3 to 5 and the 4 to 8 percentage points that we have for each of the product lines of the company.
Great. Thank you. Thanks, Patrick.
And I will now turn the program back over to our presenters for any additional or closing remarks.
Yes. Thank you for joining us for the conference call today. Just a quick reminder that our Analyst Day will be held in New York on November 18 and we look forward to meeting you all of you then. Thank you.
And this does conclude today's program. Thank you for your participation. You may now disconnect.