Aug 17, 2023
Welcome to the Applied Materials Earnings Conference Call. During the presentation, all participants will be in a listen-only mode.
Afterwards, you will be invited to participate in a question-and-answer session. I would now like to turn the conference over to Michael Sullivan, Corporate Vice President.
Please go ahead, sir.
Good afternoon, everyone, and thank you for joining Applied's third quarter of fiscal 2023 earnings call. Joining me are Gary Dickerson, our President and CEO, and Brice Hill, our Chief Financial Officer.
Before we begin, I'd like to remind you that today's call contains forward-looking statements which are subject to risks and uncertainties that could cause our actual results to differ. Information concerning the risks and uncertainties is contained in Applied's most recent Form 10-Q filing with the SEC.
Today's call also includes non-GAAP financial measures. Reconciliations to GAAP measures are found in today's earnings press release and in our quarterly earnings materials, which are available on the IR page of our website at appliedmaterials.com.
And with that introduction, I'd like to turn the call over to Gary Dickerson.
Thank you, Mike. In our third fiscal quarter, Applied Materials delivered results at the high-end of our guidance range.
Across the business, our teams are executing well. We are making incremental improvements in our operations as we productively scale the company, and we're driving our technology roadmap, introducing enabling new products and solutions for our customers.
In my prepared remarks today, I will start with our big picture view of the IoT-AI era and how this is driving growth and innovation across the industry. I'll then summarize Applied's strategy and how this is enabling us to outgrow the industry this year while also positioning us for sustained outperformance over the longer-term.
And finally, I'll cover our near-term performance and outlook, including some recent business highlights. In the summer of 2018, at our AI Design Forum, Applied laid out a thesis explaining how we expected IoT and AI to drive a new wave of semiconductor growth and innovation.
Five years later, IoT and AI inflections are having a profound impact across many sectors of the economy, as well as within the semiconductor industry. We view IoT and AI computing as two sides of the same coin.
At the edge, consumer devices, vehicles, buildings, factories, and infrastructure are all getting smarter and more capable. Our customers that serve these IoT, Communications, Auto, Power and Sensors markets, or ICAPS, are growing, and represent the largest portion of wafer fab equipment sales in 2023.
Increasingly intelligent edge devices are fueling an explosion of data generation that can then be transmitted and combined to create very large datasets for training AI models. High performance AI computing is primarily enabled by hardware innovations.
As a result, AI is becoming the key driver of the leading-edge logic and DRAM roadmaps, as well as heterogeneous integration, which creates exciting new innovation opportunities for device designers. In summary, the first part of our thesis is that the combination of IoT and AI drives demand for significantly more chips as well as next-generation silicon technologies.
The second part of our thesis relates to how those chips and new technology innovations will be supplied. In terms of technology, as the benefits of traditional Moore's Law 2D scaling slow down, the industry is moving to a new playbook to drive improvements in Power, Performance, Area-Cost and time-to-market.
The PPACt playbook has five key elements: new system and device architectures, new 3D structures, new materials, new ways to shrink, and heterogenous integration. The transition to the new playbook is positive for Applied in two key ways: First, as the roadmap evolves, it is increasingly enabled by advances in materials engineering where Applied has differentiated capabilities.
Key examples of this include the move to Gate-All-Around transistors and Backside Power Distribution in advanced logic. Second, the PPACt playbook is inherently more complex, and we can help customers manage this complexity by providing more comprehensive solutions, which include integrated products and advanced services to accelerate R&D, technology transfer, ramp, as well as yield and cost in volume production.
In parallel to the new PPACt playbook being implemented, we are also seeing regionalization of semiconductor supply chains, as countries seek to build resilient local capacity to support industry verticals that are essential to their economies. As a result, hundreds of billions of dollars of government incentives will be deployed globally over the next five years.
At Applied, we recognized these trends early and made important changes to our strategy, organization, and investment profile. In the past five years, we created a dedicated organization to focus on the ICAPS market and released more than 20 new products for ICAPS applications.
We also formed a team focused on co-optimized and integrated products to accelerate solutions for leading-edge logic and memory. This has resulted in much deeper strategic relationships with our customers, new highly differentiated products, and increasing market share.
We developed a very strong portfolio of products to enable multiple generations of DRAM technology that is fueling share gains in this market segment and contributing to our outperformance. And we established a clear leadership position in heterogenous integration and advanced packaging.
In fact, we just announced five new heterogenous integration products at SEMICON West in July. This strategy and increased focus on IoT and AI-driven inflections has enabled us to deliver more value to customers and strong performance in 2023, even during a period of very low investment by memory makers.
It also positions us for on-going outperformance in 2024 and beyond. Let me highlight a few key areas.
In DRAM, our revenues in the first half of 2023 were higher than our two closest process equipment peers combined. Our strength in DRAM is underpinned by multiple factors.
We have gained significant share in DRAM patterning, both for EUV-based and multi-patterning. We have developed unique, co-optimized hardmasked solutions, which are a key enabler for capacitor scaling.
We have successfully ported key technologies developed for logic to DRAM, where they are used in the peripheral circuitry to significantly increase I/O speed. And, we are the largest supplier of advanced packaging solutions with leadership positions in micro-bump and through-silicon via that enable multiple generations of high-bandwidth memory.
As DRAM investments increase, we feel very positive about our opportunities, especially in high-performance DRAM for the datacenter. High-bandwidth memory is less than 5% of DRAM capacity today, but it is expected to grow at a 30% compound annual growth rate over the coming years.
If you look at a HBM2 die compared to DDR5, it’s about 25% larger because of additional I/O routing and the area needed for the TSVs. On top of this, the extra processing steps needed for die-stacking further increase our total available market by approximately 5%.
Another key growth driver is our ICAPS business. We see ICAPS demand as sustainable as these customers are delivering enabling technology for large, global inflections that will play out over the next decade.
These include industrial automation, electric vehicles, and vehicles with advanced driver assistance systems, solar and wind energy, where each megawatt generated requires $3,000 to $4,000 of power chips, and the smart grid, which could drive $50 billion of annual silicon demand by the end of the decade. ICAPS investments are also expected to be underpinned by government support around the world, and we expect these markets to be a significant beneficiary of regional incentives.
The increasing complexity needed to enable the PPACt playbook combined with a broadening of the industry's geographic footprint, are both key growth drivers for our service business. AGS delivered record revenue this quarter and is on track to grow in 2023 even with this year's low fab utilization rates and after absorbing the impact of new U.S.
trade rules. This year, more than 60% of our service revenue is generated from subscriptions in the form of long-term agreements.
These agreements are growing at a faster rate than the installed base and have a high renewal rate of more than 90%. With strong customer pull for our services and a robust pipeline of new capabilities, we believe we're on-track to achieve low double-digit AGS growth in the coming years.
While I am excited about the opportunities ahead, it is important to recognize that to deliver this future, the industry must also overcome significant challenges relating to complexity, cost and our combined carbon footprint. At Applied, we believe the way to do this is through closer collaboration and higher velocity innovation and commercialization of next-generation technology for energy efficient computing.
In the past quarter, we announced two major initiatives: our EPIC platform in the United States and a Collaborative Engineering Center in India. These investments will support even faster and better relationships with customers, universities, suppliers and government partners to accelerate time-to-innovation and time-to-commercialization while increasing our combined R&D productivity.
In addition, we are also driving a collaborative approach to reduce carbon emissions as the industry grows. In July, we rolled out our collaborative Net Zero playbook and we announced two major new products that help customers with carbon reductions: our Vistara platform, which lowers platform energy consumption by up to 35% and increases throughput density by as much as 30%, and EcoTwin, that enables customers to monitor, model and optimize chemical and energy consumption by tool and by recipe.
Before I hand over to Brice, I’ll quickly summarize. Advanced chips are at the foundation of major global inflections and, as the IoT-AI era takes shape, it is driving a new wave of growth and innovation for the semiconductor industry.
At Applied, we have focused our strategy and investments to deliver high-value technologies that enable key IoT and AI-driven inflections. We have strong leadership positions in ICAPS, leading-edge foundry-logic, DRAM, and heterogenous integration using advanced packaging.
This strategy is enabling us to consistently deliver strong results in 2023, despite lower overall wafer fab equipment spending, and positions us for sustainable outperformance over the coming years. Now, Brice, it's over to you.
Thank you, Gary. On today's call, I'll discuss the business environment, summarize our Q3 results, provide our guidance for Q4, and frame the investments we're making in our R&D infrastructure.
Before covering the near term, I'll remind you of our longer-term thesis. First, we believe the semiconductor industry is on track to grow faster than the overall economy over time and reach $1 trillion in sales by 2030.
Second, Applied's leadership in materials engineering is increasingly critical to our customers' roadmaps. Third, Applied's broad portfolio of differentiated products, balanced market exposure, and growing services business make us less volatile today than in the past and more likely to grow faster than our markets.
And fourth, our efficient business model generates strong profitability and free cash flow, which enables us to both invest in profitable growth and deliver attractive shareholder returns. Moving now to the Q3 business environment, we continued to see strength in ICAPS which largely offset weakness in leading-edge foundry-logic and NAND.
ICAPS demand was broad-based, generating record revenue in 200mm systems and record revenue in Europe. In fact, the United States, Japan and Europe are the fastest growing ICAPS regions this year.
Around the world, customers and governments are making long-term investments to ensure future supply of a wide range of semiconductors needed to support growing demand in key industries such as automotive, medical equipment and renewable energy, to name a few. Turning to our operational performance in Q3, our teams delivered strong results.
We exceeded our revenue guidance across Semiconductor Systems, Services and Display. We improved our delivery performance in systems and services, and made further progress reducing inventory.
Cash from operations and free cash flow were both the second highest in our history. Now I'll summarize our Q3 financial results.
Company revenue was nearly $6.43 billion, down 1% year-over-year and in the upper end of the guidance range. Non-GAAP gross margin was slightly above our guidance, and operating expenses were slightly below.
Non-GAAP EPS was $1.90, down 2% year-over-year, and near the high end of guidance. Turning to the segments.
Semi Systems revenue of $4.68 billion was down 1% year-over-year. Segment non-GAAP operating margin was 34.8%.
Applied Global Services generated record revenue of over $1.46 billion and non-GAAP operating margin of 29.3%. This was AGS's 16th consecutive quarter of year-over-year revenue growth.
While 200 millimeter system sales contributed to the growth, the team also made strong progress driving the leading indicators of our subscription business. For example, tools under service agreement are up 5% year-over-year to over 16,000 systems, and tools under comprehensive service agreement, which have the highest revenue per tool, are up 7% year-over-year, reaching 12,000 systems.
In Display, revenue grew sequentially to $235 million, and segment non-GAAP operating margin was 15.7%. Turning to cash flows, we generated $2.58 billion in operating cash flow during the quarter, which was 40% of revenue.
We produced over $2.3 billion in free cash flow, which was 36% of revenue. We distributed $707 million to shareholders through quarterly dividends and share buybacks.
We paid $268 million in dividends and the dividend per share was $0.32, reflecting the 23% increase announced in March. We used $439 million to repurchase nearly 3.4 million shares at an average price below $130.
Now, I'll share our guidance for Q4. We expect Q4 company revenue to be $6.51 billion, plus or minus $400 million.
We expect non-GAAP EPS of $2.00, plus or minus $0.18. Within the guidance, we expect Semi Systems revenue to be around $4.75 billion.
We expect DRAM revenue to be up by around $500 million quarter-over-quarter, primarily driven by trailing-edge shipments. We expect AGS revenue to be about $1.42 billion and Display revenue should be around $290 million.
We expect Applied's non-GAAP gross margin to be about 47% and we expect non-GAAP operating expenses to be around $1.17 billion. We continue to model a tax rate of 12.3%.
Finally, as we said last quarter, we plan to make a multibillion-dollar investment in new R&D infrastructure over the next several years to significantly expand our capacity to collaborate more closely and productively with our customers as we develop next-generation materials, process technologies and equipment. The scale of the investment will depend on our ability to secure government support.
The EPIC Center is expected to come online in fiscal 2026. And while we expect our capital expenditures to be higher over the next several years, there is no change to our strong commitment to shareholder returns.
In summary, Applied Materials continues to execute well. We are making good progress against our internal goals and outperforming our markets.
While wafer fab equipment spending is lower in calendar '23, our Semi Systems revenue in the first three calendar quarters is trending slightly up year-over-year and our Services business remains on track for year-over-year growth. We've aligned our business with the fastest-growing end markets and are winning key decisions across leading-edge foundry-logic, DRAM, ICAPS and advanced packaging.
We are in a great position to invest for technology leadership and growth, generate strong free cash flow, and increase shareholder returns. Now, Mike, let's begin the Q&A.
Thanks, Brice. Our goal is to help as many of our analysts as possible.
With that in mind, please ask just one question on today's call. If you have another question, please re-queue and we'll do our best to come back to you later in this session.
Operator, let's please begin.
[Operator Instructions] Our first question comes from the line of Stacy Rasgon of Bernstein.
Hi, guys. Thanks for taking my question.
I wanted to ask about the DRAM in the quarter and in the guide. So, DRAM in the quarter was strong, and you've got a lot of growth in the next quarter.
You said it was mostly from trailing-edge. Am I right in reading that is primarily the Chinese customer driving that?
And maybe you could sort of talk a little bit about your strength in China that is from DRAM versus foundry. It almost sounds like the DRAM piece is stronger than the ICAPS piece in China right now.
Thanks, Stacy. Thanks for your question.
This is Brice. Yeah.
So, we highlighted in Q4 that we'll see sequential growth in DRAM. It is related to confirming technologies that we can ship to customers in China.
We've gotten lots of customers about that -- or questions about that during the quarter. So, we wanted to make sure that people got that answer.
So you'll see that growth in Q4. It probably won't be the only quarter we'll -- where we'll have shipments along those lines, but we wanted to make that clear on the trend quarter-over-quarter.
And one more thing that I'd add, we think that's probably something that's not well understood by the investors that while the memory market has been weak for our customers, the DRAM market actually has been fairly strong this year from an equipment perspective, and this is just a continuation of that profile for us. So, we've got a good position in DRAM, and that is manifesting in Q4.
Stacy, it's Gary. I'd add...
Is that -- go ahead. I'm sorry.
Yes, maybe just, let me add some color and then you can ask the follow-on question. So DRAM, we've added about 10 points of overall DRAM wafer fab equipment share in the last 10 years.
So, if you look at the big inflections in DRAM, the periphery moving to high speed I/O for high-bandwidth memory, we're taking logic technologies into DRAM. That's a really big driver for our share gains.
Capacitor scaling is an area where we have strength, patterning share gains, advanced packaging, especially high-bandwidth memory. So, all of those areas are really big drivers for Applied and have contributed to this 10% overall WFE share gain in DRAM over the last 10 years.
And then, you have another question, Stacy?
Yes, I think I got it. Stacy -- yeah.
Yeah. I just wanted to follow-up on -- for the quarter, DRAM was pretty strong in Q3.
Was it the same drivers?
In Q3, there was a small amount of shipments to Chinese customers, so relatively small. Q4 would be larger.
Got it. Thank you so much.
Thank you. Our next question comes from the line of C.J.
Muse of Evercore ISI.
Yeah, good afternoon. Thanks for taking the question.
I wanted to focus, I guess, on sustainability. You're talking about significant outperformance versus WFE here in calendar '23.
And so, a two-part question. So, if you look out to the January quarter, representing calendar year, are you suggesting that you guys should be kind of in that flat to down 4% kind of year?
And then, as you look to calendar '24, can you speak to, I guess, what kind of pent-up demand in terms of deferred revenues is helping this year that won't be replicated next year versus gains where you see leverage to the right end markets and/or new technologies that can drive relative outperformance? Thanks so much.
Okay. Thanks, C.J.
Maybe I'll start with the 2024, because I think you were embedding a question there on, in '23, how much did we benefit from unserved orders in '22? We think that's about $0.5 billion to $1 billion.
All of that was served in the first half of the year. So, what you're seeing from us in '23 besides that $0.5 billion to $1 billion is largely tracking with the business that we see, the WFE business that we see during the year.
So, there shouldn't be an overhang. There shouldn't be a benefit from any previously booked business that we see now or going forward.
And, as far as sustainability, that's one question we get regularly. The second question that we get regularly is just about ICAPS and the sustainability from China.
And well, that's the most the largest country in ICAPS for China. It's not the fastest growing.
That's sort of a -- that's a global trend and a global demand function. We see that as very sustainable.
We think Gate-All-Around will start shipping in earnest in '24. We think our Services business will grow in '24.
We think that the Display business will be a little bit better in '24. So those are kind of the bits that we've given for '24.
Otherwise, we're not going to give a specific forecast yet for '24 or for the January quarter.
C.J., this is Gary. Maybe I'll add a little bit more color.
On the leading-edge foundry logic, what we see is more of that spending moving to materials-enabled technologies, which is the sweet spot for Applied. So, if you look at, again, the nodal spending more is shifting to materials engineering.
Gate-All-Around, we've talked about, that's a $1 billion opportunity for us. Backside Power, when that comes in, that really leverages all of our strength in interconnect.
And that's probably the same zip code in terms of size of opportunity for Applied. ICAPS, we've talked a lot about, formed the organization four years ago.
We've introduced 20 major new products since we formed the organization. And we still have room to grow in ICAPS share.
DRAM, I talked about. I think people really maybe don't understand the strength we have there.
I talked about those big inflections in gaining 10 points of overall DRAM WFE share over the last 10 years. Packaging, that's a $1 billion business for us now, and we see an opportunity to double that over the next few years.
And as Brice said, Services, we're going to be up in Services in a relatively weak memory market in terms of utilization, and we're still on track for low-double digit growth. So, again, those are some of the things that really will help us outperform over the longer term.
Yeah. Thanks, C.J.
Thank you. Our next question comes from the line of Vivek Arya of Bank of America Merrill Lynch.
Thanks for taking my question. I wanted to understand what measures do you have in place to ensure that there isn't a pull forward of your mature technology shipments because of geopolitical reasons or because of take or pay arrangements?
Do you have ways to measure the utilization of what you are shipping? And if I kind of just ask that same thing in a different way, do you think Applied will grow in line, above or below whatever WFE does in 2024?
Hi, Vivek. Thanks for the question.
I'll answer the second part first. So, we do think Applied will grow faster than whatever the WFE is for the market.
And that's a good way to think of it, because we think of -- we think and plan for long-term secular growth, and we think we'll grow faster. On the question of pull forward, in specific to China, we've got many ways to test that equation.
One of the things that we do is we look at the amount of local capacity relative to local consumption. We think that there's strategic direction in the country and strategic incentives in the country to make the investments that will at least equal local consumption.
And we think those things are tracking together and can rationalize what they've put in place from that perspective. We also triangulate the overall capacity being purchased relative to semiconductor market itself.
As you know, we test that with intensity and we make sure that equation makes sense to us. And then, from a micro perspective, we also look at the installation of our equipment.
So, we make sure when it lands in China. We understand that it's installed, and then we track the utilization also at a macro level in the market.
And we know that all that equipment is being installed and operated. And so, we don't see signs of unneeded or unwanted or unused equipment pulling up from that perspective.
And we think the buys are rational at this point.
Okay. Thank you.
Thank you. Our next question comes from the line of Krish Sankar of Cowen.
Thanks for taking my question. I kind of have a two-part technology question for Gary.
You spoke about high-bandwidth memory and how it's benefiting advanced packaging, which makes a ton of sense given the exposure to TSV, et cetera, but on the other side, it seems like hybrid bonding is slowing down, and some of your hybrid bonding customers are looking at things like thermal compression bonders. Can you give any color on that?
And along the same path, Gary, you've spoken about Gate-All-Around being like a $1 billion for every 100,000 wafer starts per month incremental opportunity and 5 point share gains for Applied. You're there in most of those core technologies like epi, vertical epi, ALD, [Selective epi] (ph), et cetera.
Are customers in Gate-All-Around making bundling decisions, or are they still going with best-of-breed solution? Thank you.
Thank you. Those are two big questions.
Let me start off with high-bandwidth memory packaging. So that uses two critical packaging technologies, Through Silicon Via and micro-bumping to enable the stack DRAM.
The great thing for Applied is that we have the broadest portfolio to enable all different types of packaging. CMP, PVD, CVD, etch and plating.
So this opportunity for us is meaningful. The TAM for HBM increases about -- the overall packaging by about 5%.
We're the overall leader in packaging. As we have talked about, it's a $1 billion business, and we're on track to double that over the next three to five years.
So that's definitely a good opportunity. On the hybrid bonding, what I would say is that, we absolutely see hybrid bonding as a great opportunity over time.
It's not something that is going to generate significant revenue in the short term, but chiplets are something that every single company is focused on. And I do believe all of the things that we've discussed relative to growth rates there, we're still believe that's on track.
And then Gate-All-Around, the other question that you asked, we still see that as about $1 billion opportunity for 100,000 wafer starts per month. In FinFET, we had close to 50% share of that overall FinFET opportunity.
We've indicated with the full adoption of Gate-All-Around, we have an opportunity to increase our share 5%. And I think one of the things that helps us a lot is that we're deeply engaged with every company in Gate-All-Around, even with their integration teams, because we have so many of those critical enabling technologies.
So all the things we talked about back to our Master Class when we size that around $1 billion is basically still on track. And relative to ramping Gate-All-Around, we believe that is toward the end of 2024.
Yeah. And then, well, the other thing is on the unit processes in IMS.
There are some areas of that market where we're combining different technologies together under vacuum, which is something that's completely unique for Applied Materials that has a big impact on electrical performance. But the main message is we have very, very deep engagements with all of those integration teams and Gate-All-Around is pretty much on track to what we had talked about previously.
Great. Thank you, Gary.
Thank you. Our next question comes from the line of Toshiya Hari of Goldman Sachs.
Hi. Good afternoon, and thank you so much for taking the question.
I had a follow-up on your ICAPS business, maybe for Brice. So, how big was ICAPS as a percentage of your business in the July quarter?
Where do you see that going in the second half of the calendar year? And maybe more importantly, yourself and Gary talked about sustainability in the business.
I think the rationale for China makes a ton of sense. How are you thinking about the U.S., Europe and Japan, where your customers, yes, they're strategic, but they also care about near-term profitability and returns.
With some of the end markets slowing, is there fear that the non-China business within ICAPS could slow going forward? Thank you.
Okay. Thanks, Toshiya.
Good afternoon. On the ICAPS, we haven't articulated exactly how big it is.
What we've said is, this year, it's our largest market. It grew very strongly last year.
It's growing strongly again this year. And we've said, over the longer term, as you think about the business, it's probably a third ICAPS, a third leading logic, and a third memory.
And that's about as much. But you can get a feel for it.
As we said all year, the ICAPS business growth has been strong enough to offset weakness in NAND and leading logic. So those -- that will let you probably get a ballpark.
On the regionalization, we do think that regionalization leads to a little bit more spending, globally for ICAPS. But we also look and see whether that's a sustainable trend from our perspective.
We look at the underlying businesses and the growth rates of all the device types that are being invested, whether it's China or whether it's the different markets across the world. And we rationalize that and we think all of the investment that's being made is consistent with the growth rates of those underlying devices and market.
So 6%, 7%, 8%, 9% CAGRs for the different device types that fall into that, like, computer vision, analog, power chips, those sorts of markets. And then, on the last piece.
Different geos and profitability.
Yes. One thing I would say on the -- Toshiya, this is Gary.
Relative to regionalization of the supply, one of the things that is extremely important is some of those ICAPS segments are tied to very large vertical markets in those different regions. So, having a secure supply chain is something that is a big focus for every country.
There are hundreds of billions of dollars of incentives there for supporting those different ICAPS markets. So that's another thing that certainly we're seeing.
And the last thing I'd say is relative to opportunities, the -- not only is it systems, but as companies are moving into new locations, that's a really great opportunity for our Service business, because there is no service infrastructure set up for those companies moving into new regions. And that certainly will help us.
We're on track for the low-double-digit growth in our Service business, but the opportunity is supported by this regionalization.
Good. And, Toshiya, yeah, I forgot the last piece of your question was on the business -- the health of the business today and whether customers are lowering starts or changing their plans.
And that certainly happens in the environment. We've seen it, but I would just remind you, the way people are doing equipment investments, it's a two-, three-, four-year planning horizon.
So they're really making investments for what the demand function they see that far out in time, and that makes their purchasing a little bit more stable than whatever the current demand function is.
Thank you for all the color.
Thank you. Our next question comes from the line of Harlan Sur of JPMorgan.
Good afternoon. Thanks for taking my question.
Team has been doing very well in process control. It's about 10% to 12% of your systems business.
I think it grew above 55%, 60% faster than your overall systems business last year. Gary, you talked about strong technology inflections and benefits across the wafer equipment portfolio, but I'm assuming this is also pulling strong requirements for more and better process controls.
So with that, I mean, can you just give us an update on process control portfolio? And how is that segment performing relative to your overall systems business this year?
Hi, Harlan. Yeah, thanks for the question.
So, PDC for us, there are really two major areas of focus. One is, as you've talked about, is growth within PDC, but the other thing that is increasingly important is the synergy with PDC and our process equipment, both for unit equipment and integrated material solutions, accelerating R&D for some of these major inflections.
So, on the PDC growth, the business grew about two times between 2020 and 2022. eBeam is one of the fastest growing segments there.
And we've maintained our leadership with around 50% overall share in eBeam. We have new products coming to market.
We've talked about in our Master Class, the cold field emission with the highest resolution and fastest imaging. So that's helping that business growth.
And then on the synergy, we had discussed the capacitor formation of DRAM where we're bringing new materials, new etching solutions together with eBeam technology, again, to accelerate those inflections nanosheets and Gate-All-Around. These technologies are incredibly complex.
And you can't fix what you can't see. In our eBeam, imaging is world-class, ahead of anyone else.
And so that synergy value is increasing. That will also help our growth in our market share overall in our semiconductor product group, but in PDC.
So again, we're really in a great position, strong pipeline of products coming in PDC and great synergies with overall Applied.
Great insights. Thank you.
Thank you. Standby for our next question.
And our next question comes from the line of Timothy Arcuri of UBS Securities.
Thanks a lot. I had a two-part question.
First of all, Brice, can you give us a sense of, of the $1.7 billion that's in July for China revenue, can you tell us how much of that is domestic China chip makers versus other multis that might have fabs in China, but they're not domestic Chinese chip makers? And then the part two of the question is, how do you handicap the potential that this China [DRAMs pull in] (ph)?
I mean, we have Micron that's been banned in China. We have a China memory maker now taking $500 million more worth of your tools in October versus July; that's $2 billion to $2.5 billion worth of extra WFE that's just in that one quarter.
So, how do you handicap sort of how you run production factory? Do you count on that being sustained?
Do you think that it's possible that it could be a pull in? If you can just comment on that?
Sure. Thanks for the questions, Tim.
So, on the $1.7 billion for China, about 27% of our revenue, it's almost all. It's not a 100%, but it's almost all domestic Chinese customers; very little on the multinational, at this point.
And then -- and just a reminder, it does include, a portion of our Services business. It does include a portion of our Display business in that number also.
And then on the handicapping, whether it's a pull in for the traditional or legacy DRAM, I certainly don't think it's a pull in. This is a technology that was viewed as not being part of the rules where technologies were restricted for China.
So, it's an existing technology. I think they'll put that equipment into operation as quickly as possible.
And just in general, if we go up to 100,000 feet on this question, for equipment like that that's being installed in China, our view is that it's not incremental if it weren't allowed to be shipped there, that it would be shipped somewhere else in the world for production. So, we just view it as mainly location and we triangulate the amount of equipment we're selling against the overall demand function.
So, we view it as location, not incremental.
Okay, Brice. Thank you so much.
Thank you. Our next question comes from the line of Joe Quatrochi of Wells Fargo.
Yeah, thanks for taking the question. When we look at your ICAPS portfolio, how do we think about your technology leadership over some of the emerging competitors, and specifically in China, just given some of the comments around their efforts to localize their supply chain?
I guess what other certain markets where you have broader leadership or more competitive advantage? I guess how do we think about that competition?
Yeah, thanks for the question, Joe. So I would say we're always focused on competition, even in non-critical process steps.
And the main focus for us at Applied is to keep innovating in all areas of our portfolio and create deeper strategic relationships with the leaders in all device segments. In ICAPS, as we've talked about, four years ago, we formed that group, and this has paid off with share gains, helped us build deep strategic relationships with all the leading customers.
And I'd say that those strategic relationships begin with the integration teams, whether it's in the leading edge or in ICAPS is super important, because those technologies are always moving very quickly and learning faster than others is really important. We've talked about in ICAPS, since we formed this group, we launched 20 major new ICAPS products.
And I would tell you that we also have a strong pipeline of new innovations that will continue to strengthen our positions. The last thing I'd say in ICAPS is there are still -- again, we have the strongest strategic position, broadest portfolio.
We're working with those integration teams on their next-generation technologies. But we also have room to grow in certain segments like etch and PDC.
We have momentum in those segments, and that certainly is going to help us continue to gain share in the coming years.
Thank you. Our next question, please standby.
Our next question comes from the line of Joseph Moore of Morgan Stanley.
Great. Thank you.
I was surprised to see Services at a record level given the utilization issues in memory. Can you talk about that?
And I mean, are you still -- are you seeing headwinds from the memory side that are incremental and you're just offsetting that with strength elsewhere? And kind of what does that tell us about where that business can go when utilizations come back up?
Okay. Thanks, Joe.
This is Brice. So yes, the Services business, 16 consecutive quarters of year-over-year growth.
It was a record this quarter. And yes, the puts and takes there are that the transactional spares and the transactional activities that sort of are associated with utilization are significantly lower in the quarter.
And then what we see that's higher in the quarter that drove that growth was really the 200-millimeter equipment. So, we talk about ICAPS growth and the strength of those markets.
And the 200-millimeter equipment is really what's driving the significant upside in the Services business for the quarter. And then, on the second part of the question, incremental headwinds for memory.
When we think about the memory market, there's been discussion of when will it be turning around and when will we see upside. The indicators that we follow, price, inventory and utilization, we still see trending in the negative direction or flat.
So, we haven't seen a turn at this point. But we wanted to be careful to point out that in the DRAM market, we have seen strength this year.
And it's sort of a normal year from an equipment perspective on the DRAM side, even though the memory market itself has been weak.
Yeah, Joe, this is Gary. I think the other thing to add is that part of our business is based on service agreements, and that also helps us have a higher level of stability.
We're over 60% on the service agreements. And again, that's keeping that part of the business much more stable, even though the memory utilization is down.
And the other thing I would say is that there is a significant amount of complexity. If you think about all the inflections, whether it's in memory with DRAM or ICAPS or leading-edge foundry logic, about a third of our revenue are these integrated material solutions with multiple technologies combined in a single platform under vacuum, those are increasing, those steps are increasing, and the service opportunity for us is also increasing.
So, we've talked about the low double-digit growth in Service. Once you come back to more normal levels of utilization, and we have a high confidence we're going to be able to achieve that.
Okay. Thank you very much.
Thank you. Please standby for our next question.
Our next question comes from the line of Brian Chin of Stifel.
Hi, there. Good afternoon.
Thanks for letting us ask a few questions. Maybe firstly, it seems difficult to imagine AI not driving incremental foundry capital spending in the next upcycle.
So, have you thought about how much incremental WFE spend in future years could result from AI-driven wafer start expansion?
[Audit Starts] Hi, Brian, thanks for the question. We certainly have.
When we think about the WFE share of the different end markets, data center has approximately 20%, we think AI is about 5% -- AI-related WFE is about 5% of our overall market. And you've heard different peers talking about a high growth rate bearing between 30% and 50% for that workload and the amount of capacity.
So, if you view 5% as relatively small amount, we do think that it's growing rapidly and it’ll be an important workload going forward. So, that's our sizing for now.
And as a comparison, we would hold that next to 10% to 15% of wafers that would be associated with IoT, and we think those things -- IoT wafer starts and WFE will also grow at an elevated rate.
Yeah. Got it.
That's helpful. And heterogeneous integration is sort of supplemental to that as well, right?
And although it wouldn't be detectable from your results, did you see any pushouts from advanced foundry and logic customers? And also, when you think about the several domestic foundry shells that are either constructed or being constructed and ready to be filled, now until, I guess, into next year.
What's kind of your current expectations for the timing and significance for some of those expansions?
Yeah. There's still a lot of movement underneath the macro number in our results, there's still a lot of movement underneath as you're suggesting with that question.
So, we have seen a delay in installation of some of the tools, and we have seen push outs with some of the leading foundry customers. So, that's consistent with what we're seeing.
And then I would just highlight again the strengths in ICAPS and the strength in DRAM, both Q3 and Q4, our outlook is really what's offsetting that.
Thank you. Please standby for our next question.
Our next question comes from the line of Timm Schulze-Melander of Redburn.
Hi, there. Thank you very much for taking my question.
I wanted to ask about longer-term R&D intensity. You mentioned about the higher CapEx for the EPIC R&D center and just thinking about the mix of the business, a third of revenues from ICAPS probably intuitively a lower R&D intensity.
The other two-thirds leading-edge logic and memory, as you say, is more materials intensive and maybe higher R&D to sales intensity. Just wondering what we should expect the net of those two factors to be on a kind of multiyear view, please?
Yes. Thanks, Timm.
Well, our spending at this point is 18.1% of revenue. It's a little bit higher than our model, but we're very comfortable with our spending level.
We put about 66% of our spending, two-thirds towards R&D. And we are fully focused on the roadmap.
So when we think about the business, we think about the longer-term trajectory of the business. Our perspective is, we're investing in a secular positive trend.
We think about semiconductors growing high mid-single digits, so 6%, 7%, 8%, 9%. We think there'll be -- it's becoming more complex to build those semiconductors, which drives more equipment, so that's an additive function.
We also think that Applied, looking forward, designing for the inflections with our customers will gain share on top of that. So we think there's an additive function between the market growth of semis, the higher complexity of the equipment, and then the increased intensity from materials engineering and Applied's participation.
So all those things add up to a significant growth rate. And so, what I would expect is, the spending level you see today will focus on maintaining approximately that spending level as we identify the best R&D projects to work on as we go forward.
And then I would also highlight that just for investors who are working on their models, we've held our spending flat for pretty much three quarters and we said we've been very conscientious and very focused on strategic hires only. As we move to Q1, so the first quarter of our fiscal next year, we'll see our normal pay rate cycle, and we'll see our normal raise of spending in that cycle.
So I just wanted to highlight to people that that's what they'll see if they're modeling that function.
Super helpful. Thank you.
Thank you. Our next question comes from the line of Charles Shi of Needham & Company.
Hey, thanks for taking my question. I want to ask, your largest customer in Taiwan talked about CapEx leveling off on a dollar basis.
Wonder what's the impact that you might be seeing in terms of your revenue coming from that particular customer in 2024 and beyond. And any preliminary thoughts on the relative outperformance of WFE maybe in '24 by segment, meaning leading-edge, foundry logic, ICAPS, DRAM and NAND?
Okay. Thanks, Charles.
All I can say about any particular customer, but especially leading-edge customer is that we're current on their forecast. So, we understand at an account level what they're forecasting in the future.
They've all started as we went through the supply chain crisis, we're getting better visibility into the roadmaps top to bottom. So, we have that -- we have a good perspective on that.
In general, we haven't called a number for WFE in '24 -- for our business in '24. I just highlighted in the prior question that our view of the long-term growth of the market, sort of what guides our investments, et cetera.
What we have said for next year is that we expect the ICAPS business to be stable. So that growth is durable, if you will.
It's grown at a high level for the last two years, and we expect that business not to fall off, but be stable. We expect our Display business to grow modestly.
We expect our Services business to grow at the low double digits as we've described. And I think that's the color that we've given so far for the market.
That leaves out memory and leading logic, and we'll just have to wait until we get closer to give a forecast. [Audit End]
Yes, Charles, this is Gary. Relative to longer term, as I mentioned earlier, we're in a great position.
If you look at this incremental spending for the future technology nodes, a higher percentage of that incremental spending is going to materials-enabled technologies. And as I said, that's a sweet spot for Applied.
Gate-All-Around, Backside Power, where you get power and performance benefits, but also you're going to get area savings with that inflection. So, we have been growing our position from -- for each technology node.
I think that gets even better going forward.
Yes. Thanks, Charles.
And operator, we have time for one more question, please.
Yes, sir. Please standby.
Our final question comes from the line of Sidney Ho of Deutsche Bank.
Great. Thanks for squeezing me in guys.
This is [indiscernible] on for Sidney. And I want to ask about Gate-All-Around.
Gary, you've been very clear about how big this opportunity could be, both in terms of revenue and market share. And so, maybe can you speak to the revenue contribution from this inflection in '24?
Yes. Thanks for the question.
I think that we're not going to give a specific dollar amount. But what I would say is that, that inflection is just starting to ramp in the latter part of '24.
So, it's not going to be a significant driver in '24, but certainly going forward, our opportunity there is significant.
That's great. Thanks.
Thanks for your question. And I'd like to just see if Brice, would you like to give us any closing comments before we close the call?
Sure. Thanks, Mike.
Just one content note that I wanted to add. So, we get a lot of questions about modeling gross margin, and I just want to highlight for the people that are doing their models.
Our 47% guide for Q4, that's a little bit higher than what we would have expected with a rich mix in Q4. We're probably running underneath that at a normal mix around 46.6%, 46.7%, somewhere around that.
So, as people think about modeling their next year, we think we'll continue to make progress from 46.6% or 46.7% to -- on our way to 48% to 48.5% in '25. So, if that helps from a modeling perspective, I just wanted to highlight that.
From a closing comments perspective, we're executing well. We're outperforming our markets.
The Semi Systems year-to-date revenue is trending up year-over-year. Our Services business is on track for year-over-year growth.
We're aligned -- we've aligned our businesses with the fastest growing end markets, and we're winning in leading-edge foundry logic, DRAM, ICAPS and heterogeneous integration. We're investing for technology leadership and growth.
We're generating strong free cash flow and increasing shareholder returns. I hope to see many of you at the Jefferies Conference in Chicago, and Gary will be keynoting at the Goldman Sachs Conference in San Francisco.
Now Mike, let's go ahead and close the call.
Thanks, Brice. And we'd like to thank everybody for joining us today.
A replay of the call is going to be available on the IR page of our website by 5:00 Pacific Time today, and we'd like to thank you for your continued interest in Applied Materials.
This concludes today's conference call. Thank you for participating.
You may now disconnect.