Q3 2024 Applied Materials Inc Earnings Call
Participants
Michael Sullivan; Analyst; Applied Materials Inc
Gary Dickerson; Chief Executive Officer, Director; Applied Materials Inc
Brice Hill; Senior Vice President, Chief Financial Officer and Global Information Services; Applied Materials Inc
CJ Muse; Analyst; Cantor Fitzgerald & Co Inc
Stacy Rasgon; Analyst; Sanford C. Bernstein & Co LLC
Srinivas Pajjuri; Analyst; Raymond James Financial, Inc.
Vivek Arya; Analyst; BofA Securities
Christopher Caso; Analyst; Wolfe Research, LLC
Krish Sankar; Analyst; TD Securities Inc
Joseph Moore; Analyst; Morgan Stanley
Joe Quatrochi; Analyst; Wells Fargo Securities, LLC
Toshiya Hari; Analyst; Goldman Sachs Group, Inc.
Harlan Sur from; Analyst; JPMorgan Chase & Co
Charles Shi; Analyst; Needham & Company, LLC
Brian Chin; Analyst; Stifel, Nicolaus & Company
Blayne Curtis; Analyst; Jefferies LLC
Jonathan Dorsheimer; Analyst; William Blair & Company L.L.C.
Mehdi Hosseini; Analyst; Susquehanna Financial Group LLP
Presentation
Operator
Welcome to the Applied Materials earnings conference call. (Operator Instructions) I would now like to turn the conference over to Michael Sullivan, Corporate Vice President. Please go ahead, sir.
Michael Sullivan
Good afternoon, everyone, and thank you for joining Applied’s third quarter of fiscal 2024 earnings call. Joining me are Gary Dickerson, our President and CE; 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 our website at ir.appliedmaterials.com.
And with that introduction, I’d like to turn the call over to Gary Dickerson.
Gary Dickerson
Thanks, Mike. With record revenues in our third quarter and earnings toward the high end of our guided range. Applied Materials continues to deliver strong results in 2024. Secular trends are growing our available market and our unique and connected portfolio of capabilities, products and services positions us to outperform the industry over the longer term.
At our recent investor event at SEMICON West, leaders from our semiconductor business shared their perspective on the powerful multi-decade technology trends driving the industry forward and explain the role Applied is playing to enable the next generations of semiconductor technology.
In today’s call, I will summarize some of the key themes we talked about at that event, including how advancing energy-efficient computing performance is critical to deploying AI at scale. Why the energy-efficient computing road roadmap is increasingly enabled by materials engineering and Applied Materials and how we’re working in new ways to accelerate this complex roadmap and create new growth opportunities for Applied.
Semiconductors, provide the foundation for tectonic shifts in technology that will reshape the global economy over the next several decades, including AI, IoT, robotics, electric and autonomous vehicles and clean energy.
These multitrillion dollar global inflections are increasing demand for chips and driving the need for significant advances in semiconductor technology. The biggest tectonic shift evolve is AI and the race for AI leadership will in large part be determined by which companies in the semiconductor industry are first to deliver substantial improvements in energy-efficient compute performance.
In our discussions with leading AI companies, they are telling us that reducing power per operation is now more important than increasing operations per second. They are also talking about the need to drive a 10,000 times improvement in performance per watt over the next 15 years.
Evolutionary innovation is insufficient to deliver improvements of this magnitude, and we’re seeing the emergence of a new industry playbook made up of major device architecture inflections in logic, memory and advanced packaging.
These device architecture inflections are increasingly enabled by material science and materials engineering, where Applied is the clear market and technology leader. For AI, the advanced chips in the data center used for training AI models are built upon four critical categories of semiconductor technology, leading edge logic, high performance DRAM, high-bandwidth memory enabled by die stacking technology and advanced packaging to connect the logic and memory chips together in a single integrated package.
In leading edge logic, key device inflections in both transistor and interconnect are currently moving from chip makers, R&D pilot lines to high-volume production. The transition from FinFET to gate-all-around transistors grows Applied’s available market for the transistor module from around $6 billion to approximately $7 billion for every 100,000 wafer starts per month of capacity.
We also expect to gain share through the gate all around transition, and we’re on track to capture more than 50% of the process equipment spending for the transistor fabrication steps. For the interconnect module, our available market is also about $6 billion for every 100,000 wafer starts per month.
We forecast, this will also grow by about $1 billion with the implementation of backside power delivery, and we expect to win more than 50% of the applications we address in interconnect when backside power ramps in volume manufacturing.
In DRAM. We’ve also established clear leadership in process equipment and are in a great position for future growth. Over the past decade. We’ve grown our market share in DRAM by around 10 points as DRAM plays a critical role in energy-efficient computing performance, there’s a huge focus on advancing the roadmap. The next major DRAM inflection from 6Fsquare to 4Fsquare or vertical transistor architectures is materials enabled, and we expect our market opportunity to grow by approximately 10% to around $6.5 billion for each 100,000 wafer starts per month of capacity.
We also expect to increase our share based on our position to enable the 4Fsquared inflection. In addition we believe the subsequent transition to 3D DRAM. will grow our addressable market by an incremental 15%. Further compounding Applied’s opportunity.
In the die-stacking technologies that enable High-bandwidth memory, we also have strong leadership positions, both in micro bump and Through-Silicon Via. We have seen demand for High-bandwidth memory accelerating in 2024 and expect to generate more than $600 million of HBM packaging revenue this year, which is approximately six times 2023.
Overall, including HBM, we expect revenue from our advanced packaging product portfolio to grow to approximately $1.7 billion in 2024. We believe this business can double in size over the next several years as heterogeneous integration is more widely adopted and we introduce new solutions that grow our addressable market.
Advanced logic, high performance, DRAM, High-bandwidth memory, and advanced packaging are all great examples of how future device architecture inflections are increasingly enabled by materials engineering. As a result, we expect materials engineering as a percentage of total wafer fab equipment to grow in both logic and memory through the coming node transitions.
At the same time, thanks to our inflection and focused approach to R&D and the strong positions we’ve established at these future device architecture inflections, we expect to capture more of the expanded opportunities we serve.
The value of bringing next-generation semiconductor technology to market faster has never been greater. At Applied, our strategy and investments are focused on accelerating the industry’s roadmap to support the highest growth rates, global inflections, spanning: AI data centers, edge-AI and IoT, robotics, electric vehicles and clean energy. This strategy is enabled by three pillars.
First, we have built a broad unique and connected portfolio of highly enabling technologies. As well as providing traditional, best-in-class unit processes, we can co-optimize, combine and integrate our technologies to deliver more comprehensive solutions that address higher-value challenges for our customers.
These integrated fab-in-fab Solutions have grown from approximately 20% of our semiconductor products revenue in 2019 to around 30% today. we expect demand for our integrated products to continue growing both at the leading edge and from our ICAPS customers who are serving specialty markets.
Second, we are changing the way we work inside and outside the company. Over the past five years, we have built new capabilities and dedicated teams focused on module integration, device design and simulation, data analytics and AI, advanced packaging and ICAPS.
At the same time, we are driving earlier deeper and more extensive collaboration with our customers and partners to win the device architecture inflection races, accelerate mutual success rates and increase investment efficiencies. To further support these collaborative partnerships, we will build out our global EPIC platform over the next several years, which is specifically designed for a high-velocity innovation and commercialization of next-generation technologies.
And third, we are helping customers transfer new technology into high-volume manufacturing faster and then optimize performance, yield output and cost in their factory operations. This is supporting double digit growth in services with a high percentage of our service revenue coming from subscriptions in the form of long-term agreements. Overall, AGS delivered another record quarter, which is their 20 consecutive quarter of year-on-year growth.
Before I hand over to Bryce, I’ll quickly summarize.
Applied Materials is delivering record results in 2024, and we’re in a great position to benefit from secular growth trends over the longer term. Semiconductors are the foundation for a tectonic shifts in technology, which will reshape the global economy over the next several decades. This is driving increasing demand for chips as well as the need for significant advances in semiconductor innovation. The race for AI leadership depends on delivering significant improvements in energy-efficient compute performance in the range of 10,000 times over the next 15 years. The need for more energy-efficient compute is driving major device architecture inflections within the semiconductor roadmap that are enabled by materials engineering and Applied Materials innovations. This expands our served market and is accretive to our share, and the increasingly complex industry roadmap creates new collaboration and growth opportunities for Applied, enabled by our broad, unique and connected portfolio of capabilities, products and services.
Finally, as you may be aware, Mike Sullivan will be retiring at the end of this calendar year and handing the reins to Liz Morali, who recently joined Applied as our new Head of Investor Relations. I would like to say a huge thank you to Mike for his many contributions to the success of our company and congratulate him on an outstanding career.
Now over to Brice.
Brice Hill
Thank you, Gary. And I’d like to thank our teams for their strong execution this quarter, which enabled us to deliver record revenue, improved operational performance and healthy gross margins. Today, I’ll summarize the market environment, discuss our Q3 performance and share our Q4 outlook.
As Gary mentioned, Mike Sullivan plans to retire from applied at the end of this calendar year. Liz Morali has joined Applied as our new Vice President of Investor Relations. Mike and Liz are working on a smooth handoff of the IR function beginning with our November earnings call. I hope you’ll join Gary and me in congratulating Mike on his extraordinary career and leadership and welcoming Liz to Applied Materials.
Beginning with the business environment, our revenue in Q3 as well as our outlook for Q4 reflect the industry’s focus on the major inflections Gary highlighted earlier. We are seeing particularly strong pull related to AI and data center computing. Specifically, our DRAM system shipments remained strong even as DRAM sales in China declined as anticipated, adoption of high-bandwidth memory and other forms of advanced packaging continues to grow. And in foundry, logic, leading edge investment is growing each quarter and becoming a larger percentage of our mix, while high caps demand remained strong overall.
Turning to our Q3 performance, we delivered record revenue of $6.78 billion, which was up 5% year-over-year with growth in all three segments. Operationally, we saw improvements in a number of key metrics such as manufacturing cycle times, linearity and on-time delivery. These improvements give me confidence that we are preparing the company to more efficiently support the growth we are forecasting in the years ahead.
The strong operational performance helped us deliver non-GAAP gross margin of 47.4%, which is up 100 basis points year-over-year. Non-gaap operating expenses of $1.26 billion were up 8% year-over-year in over 70% of the increase was driven by R&D programs aimed at the technology inflections Gary described. Non-GAAP EPS of $2.12 was up 12% year over year, and $0.01 below our highest EPS quarter ever.
Turning to cash flows and profit distributions in Q3, we generated nearly $2.4 billion in operating cash flow and over $2 billion in free cash flow. We distributed nearly $1.2 billion to shareholders, including the first $0.40 per share dividend and $861 million and stock buybacks.
Turning to the segments. Semiconductor Systems sales were $4.92 billion in Q3, up 5% year-over-year. Segment non-GAAP operating margin was 35%, up 130 basis points year over year. From a device perspective, our DRAM sales grew nearly 50% year-over-year to $1.16 billion DRAM sales in China declined sequentially as we anticipated, and this contributed to our company revenue in China declining by 11 percentage points sequentially to 32%, which is in line with our longer-term average inclusive of semi systems, AGS and display.
Our NAND memory sales grew 10% year-over-year to $203 million. Our Foundry logic sales were down 4% year-over-year to $3.56 billion. Leading-edge foundry logic demand was lower year-over-year but continued to strengthen on a sequential basis. We continue to expect to generate more than $2.5 billion in system revenue from gate-all-around nodes this calendar year, with the potential to more than double next year.
Our ICAPS business remains strong overall, with pockets of weakness in the auto and industrial end markets. Longer term, we expect the ICAPS market to remain around half the foundry logic market as major inflections in IoT, autonomous and electric vehicles and the global energy transformation are expected to drive mid- to high single-digit through cycle growth in ICAPS semiconductors well into the future.
We are investing in new products to compete in more areas of the ICAPS semiconductor and packaging ecosystem. We are also working with our customers to enable new power and sensor device architecture inflections using our co-optimized and integrated material systems.
Next, Applied Global Services delivered record revenue of $1.58 billion in Q3, which was up 8% year-over-year. AGS recurring parts services and software revenue grew more than twice as fast as overall segment revenue during the quarter.
AGS non-GAAP operating margin of 29.6% was up 230 basis points year-over-year and non-GAAP operating profit was a record $467 million. From a business perspective, customer factory utilization continued to strengthen during the quarter across memory, foundry logic and advanced packaging.
Our leading indicators of future AGS growth remain positive. Our installed base of systems and chambers increased 7% year-over-year, and our average revenue per unit increased even more. Our average subscription agreement length increased to 2.8 years, and the renewal rate was above 90%.
We continue to expect AGS to grow at a low double-digit rate over the long term. As a reminder, the consistency of our profitable growth in services gives us confidence in our ability to continue to increase our dividend per share. In fact, over the past 10 years, we have increased the dividend per share at a compound annual growth rate of approximately 15%.
Moving to display. Q3 revenue of $251 million was up 7% year-over-year in segment non-GAAP operating margin was 6.4%. While LCD equipment spending remains low, we are becoming more confident that the OLED technology found in smartphones will be adopted in notebook PCs and tablets, whose larger screen sizes will require a significant increase in capital investments.
Applied has built a leadership position in deposition and eBeam metrology technologies for the display industry, and we are well positioned to enable our customers to convert the notebook PC and tablet markets to OLED technology over the coming years.
Now, I’ll share our guidance for Q4. We expect company revenue of $6.93 billion, plus or minus $400 million and non-GAAP EPS of $2.18, plus or minus $0.18, both up 3% year-over-year at the midpoint. Within this outlook, we expect semi systems revenue of around $5.1 billion, which is up 4% year-over-year; AGS revenue of about $1.61 billion which is up 9% year-over-year, and display revenue of around $200 million. We expect non-GAAP gross margin to be approximately 47.4% and non-GAAP operating expenses to be around $1.275 billion. Finally, we are modeling a tax rate of 12.5%.
Thank you. And now, Mike, let’s begin the Q&A.
Michael Sullivan
Thanks, Bryce. (Event Instructions) Operator, let’s please begin.
Question and Answer Session
Operator
(Operator Instruction) C.J. Muse, Cantor Fitzgerald.
CJ Muse
I guess for my question, I was hoping you could speak to how your outlook for WFE in the second half of the year and for all of 2025 has evolved over the last 3 months? Obviously, tremendous strength, anything AI related, but elsewhere kind of seasonal at best.
So I guess, first part of the question is, can you share with us how you’re thinking about your silicon business in the second half of calendar ’24 versus the first half? And then the early signs that you see today and how that informs kind of your view both positive and negative into 2025?
Brice Hill
Okay. C.J, thanks for joining the call. I think what we’re seeing in the business today when we look at Q3 and our Q4 outlook, very strong energy around the AI inflections that Gary talked about. So leading edge is accelerating. We highlighted that in the prepared remarks. We’re — our forecast for $2.5 billion of gate all-around related equipment in this fiscal year.
We haven’t changed that forecast. That stays the same. And when we think about DRAM and HBM memory, those kind of all go along with this energy we’re seeing around the AI inflections and investments in that area.
At the same time, our ICAPS business, in both Q3 and Q4, remains very strong. We think that’s robust, and we can dig into that a little bit more during the call. But if anything, over the last 30 days, you asked us how we’re feeling about the outlook, I would say the ICAPS business continues. It seems like every quarter, we sort of raised our expectations with respect to ICAPS.
So a lot of energy around leading edge. ICAPS is remaining strong. Very strong year for DRAM and HBM. Those are kind of the key themes. For ’25, we’re not giving specific guidance, but we would say we’re pretty enthusiastic about gate all around and the technologies we’ve talked about on the leading edge.
Operator
Stacy Rasgon, Bernstein Research.
Stacy Rasgon
I wanted to dig into the China revenues a little bit. So 32% of revenue this quarter. It sounds like a lot of that was DRAM. I guess that — was it all DRAM? And then in the guidance for next quarter, and I guess your expectations going forward from there, do you expect the China mix to be staying around these levels in the low 30 or does it go higher or lower? Is this is there anything else going on with China that we need to be aware of?
Brice Hill
Great. Thanks for joining Stacy. So yeah, on the 32%, I’ll shift your perspective on that just a little bit. So the 32% is almost no DRAM to China. So you’ve got it right. Our business mix to China declined to 32%. And if I think back in time, just to remind everybody, if we go just over a year back, it was like 17% mix to China. We were having supply chain issues. So it was very low at that point.
In the last three quarters, we bounced up to the mid-40s where we serve the DRAM that we could ship and that’s the prior three quarters. And we feel this quarter, 32% is probably normal for us, or in the normal range for us, that’s across our entire business for shipments to China. And what it really represents is ICAPS. And the ICAPS market in total has been very robust. We think it will be a record year in our fiscal year for ICAPS.
In China also, we would call it robust. There’s — we’re adding customers, utilizations are improving. It’s a wide variety of products and factories. So we think that market is robust, both in Q3 and in Q4. So that gives you a sense of what’s happening for us. And then next quarter, we’ll have a small amount of DRAM, but nothing like we’ve had the prior three quarters.
Stacy Rasgon
Is it fair to say that you still see the mix in the low 30s next quarter to China?
Brice Hill
Yeah. We expect that mix approximately normal, which we call 30 for the next quarter, yes.
Stacy Rasgon
Got it. That’s helpful. Thank you, guys.
Operator
Srini Pajjuri, Raymond James.
Srinivas Pajjuri
Brice, just to follow up to the previous question on China, DRAM. Obviously, you said it’s near zero levels. Can you talk to your visibility as to when — if you’re seeing any hope on the horizon? Do you expect DRAM to come back at some point in China?
And then also ICAPS, you said the demand is pretty strong in China. Can you talk to what kind of trends you’re seeing outside of China for ICAPS?
Brice Hill
Yes. Thanks, Srini. The China DRAM in Q3 and Q4 will be at nominal levels. We have speculated, you may see new investments in DRAM there. I don’t have any comments to make on that specifically for the road map. But I would say that DRAM globally is very strong. It should be a very, very strong year for DRAM.
Of course, that includes some of that China volume. But we think that goes along with the mix to HBM and utilizations in DRAM have improved, and there’s a lot of energy around the high-bandwidth memory for the leading-edge systems.
And then on the ICAPS side, no surprise here for probably anybody. We expect the edge technologies that we describe as ICAPS to grow mid- to high single digits over time. And that’s driven by the inflections that we talk about, which is clean energy, renewable power, electrification, those sorts of AI edge sensors, all those sorts of usages. So mid-to-high single digits is our view long term.
And when we look at China for both Q3 and Q4 and the entire ICAPS business, I said before, it’s probably going to be a record for us in our fiscal year, and it’s not just China. We’ve had regions, other regions grow during this quarter.
Srinivas Pajjuri
Got it. Thanks, Brice.
Operator
Vivek Arya, Bank of America Securities.
Vivek Arya
I’m curious if we should bake in the impact of any CapEx cuts that were announced recently at Intel. What does that imply for just the overall WFE market for the remainder of the year and into next year?
From your comments, it doesn’t seem like it’s having a big impact. Is the assumption that whatever CapEx has been cut there is being increased somewhere else? Or just what are the puts and takes? Because it is a fairly sizable CapEx cut, which we are saying is more WFE in terms of the mix.
Brice Hill
Yeah. Thanks for joining, Vivek. Our outlook, we’re not changing our outlook. We’re still saying $2.5 billion on the leading edge for gate-all-around in this fiscal year. And when we think about next year, we’ve just had — we have every quarter accelerating this year in the leading edge, and we think that’s being driven by a lot of the excitement around AI and AI data center.
So we get — as I think, we get regular updates from our customers. And our outlook that we’ve just given for Q4 is up to date with all of our customers. So I think we tend to think of the market as a macro level. Customers update us all the time, but no real change to our view for leading edge looking forward.
Operator
Chris Caso, Wolfe Research.
Christopher Caso
Question is on foundry logic. And perhaps you could give some color on the growth that you’re expecting into ’25 driven by some of the new process nodes such as gate-all-around as compared to capacity, because we know foundry capacity has additions have taken a pause here. Are you seeing any signs of that starting to improve either as you go through the second half of ’24 into ’25?
Brice Hill
Yes. Thanks, Chris, and thanks for joining. When we think about leading logic, actually, I’ll make a comment for utilizations. We’ve seen utilizations improve in every single end market this quarter, and our expectation is for that to continue next quarter.
So DRAM, NAND, ICAPS and leading edge, and no change to our expectation on absolute leading edge investment, gate-all-around technologies, not changing our number for this fiscal year. And we do think it’s a harbinger for acceleration next year as we look forward.
We’re not giving a guide for ’25, but we think there’s a lot of energy around the AI markets. And we see that in HBM. We see that in DRAM. We see that in leading edge acceleration. So we’ll hope that those trends continue.
Operator
Krish Shankar, TD Cowen.
Krish Sankar
Gary and Brice, I’m kind of curious about — I know you don’t want to comment on next year. But qualitatively, there’s optimism on DRAM and NAND, WFE growing. But if you ship out HBM, you’re beginning to see big demand in (inaudible) moderate for DDR and NAND compared to the first half of this year. So I’m curious how you think about DRAM and NAND WFE next year relative to this year? And what (inaudible) drive up spending next year?
Brice Hill
Okay, Krish. You’re right. We won’t give a guide for ’25 specifically, but we do think the DRAM, even outside of HBM, we do think the DRAM will put new capacity in place, so we expect investment in DRAM.
There has been additions in wafer start capacity in DRAM. And there’s more allocation of the DRAM capacity to HBM as you sort of highlight. So the investment level there is higher than NAND.
Looking forward in NAND, NAND still — it still remains fairly low, but we did see, as I highlighted, utilization is improving, I would call them probably normal range for both DRAM and NAND. We see prices improve. We saw inventory positions improve at the vendors. So we think it’s a more positive environment looking forward than prior quarters for actually both memory technologies.
Operator
Joseph Moore, Morgan Stanley.
Joseph Moore
Looks like you had some good solid growth from the services business. Again, are you benefiting there from the utilization increases in memory that you just talked about? Are you seeing offsetting utilization declines on the ICAPS services side? And then any indication of customers worry about export controls, stockpiling on spares, anything like that? Just any puts and takes around that service growth.
Brice Hill
Joe, yes, Services business, very exciting for us. We grew 8% year-over-year in Q3, 9% is our expectation for Q4. And just a reminder for everybody, we expect low double-digit growth going forward. We’re expanding our services book with customers kind of getting into information, AI-related services that help them ramp et cetera.
We actually did expect a little more growth in Q3. We had thought we would grow double digits this year. We’re just short of that. And it’s because utilization, while it grew, it grew a little bit less than we — or a little bit slower than we thought. So there’s good news here. Utilizations are improving across the whole system. 8% year-over-year growth for us, and we expect that to pick up, as I highlighted as we go into Q3 and looking forward.
And just one other comment, just for the rest of the investors. The other thing with this business since it’s mostly recurring revenue for us, about 85%. That gives us a very stable operating profit and growing. And so we think about our dividend as being enabled by the profits from the services business, and that gives us confidence we’ll be able to raise our dividend looking forward.
Gary Dickerson
Joe, just maybe a little bit more color. If you look at what really all of our customers are focused on, there is a race for these new device architectures and the complexity is going up a significant amount. And in the prepared remarks, I talked about 30% of our tools are these integrated platforms that have a high degree of complexity.
So as these customers are racing to bring these new devices — complex devices to market. And also, we’re shipping more and more of these integrated platforms with multiple technologies, that also gives us a really good tailwind.
Operator
Joe Quatrochi, Wells Fargo.
Joe Quatrochi
I wanted to ask about the advanced packaging. I think based on the revenue expectations you have for HBM and advanced packaging, it imply basically the non-memory advanced packaging relatively unchanged year-over-year. Is that the right way to think about it? And so why wouldn’t that grow this year?
Brice Hill
I think so, Joe. In our forecast, we highlighted that Advanced packaging was $1.1 billion last year, should grow to $1.7 million this year, and we highlighted that $600 million of the growth would come from HBM related equipment.
So I think you’re thinking about it right. It was probably stable this year with HBM, high-bandwidth memory driving the growth in that area. And again, we would point back to the energy around high-performance systems and sort of the race Gary talks about to develop systems optimized for the AI workloads. So there’s a lot of energy there at this point.
And then just the last thing is we have highlighted that, that total packaging business, sort of to your point, $1.7 billion, we think it has the opportunity to double over several years looking forward given the energy around advanced packaging technologies. So we’ll keep investing there.
Operator
Toshiya Hari, Goldman Sachs.
Toshiya Hari
I was hoping you to speak to gross margins and how you’re thinking about profitability going forward? The puts and takes price. And Gary, you talked about complexity growing and how things like MS contributing toward 30% of your business and that number growing over time.
I would think your ability to price would improve going forward? What am I missing in terms of, again, the puts and takes as it pertains to gross margins?
Brice Hill
Okay. Toshi, I’ll start on that. So 47.4% gross margin for us in the quarter, we’re actually very pleased with our performance in cost and pricing this quarter. And I’ll just highlight, as our mix to China, which are generally smaller customers declined during the quarter to 32%, that gave us some headwinds on the gross margin side. We were able to get some pricing performance and also some cost improvements and improvements in managing our inventory that helped us offset that mix decline and deliver 47.4% in the quarter.
Since the China mix is 32%, we call that normal. I’ll say now that 47.4% should be approximately our baseline level. And then we look forward, we’ve talked about a goal of getting to 48% or higher next year. That’s still our goal. We think we can make improvements going forward.
There’s a couple of headwinds if services or display grow faster than the core business, that can be a headwind. But that’s still our goal, and we think we’ll make — continue to make cost and pricing improvements. And then, Gary, on them.
Gary Dickerson
Yes. Toshiya, again, our focus is really to enable our customers to accelerate their innovation for these new architectures, especially around AI data center. That data center will pass smartphones and PCs relative to wafer starts. Everybody is focused on energy-efficient computing.
So these inflections are incredibly important for the entire ecosystem. Whoever gets there first, wins big, and everybody else is left behind. So we’re driving tremendous innovations in foundry logic leading edge, high bandwidth memory, DRAM. There’s new architecture inflections. I talked about 4Fsquared advanced packaging.
So our positions in all of those inflections are very strong. We’re on track to capture more than 50% of the inflection spending as those new devices ramp. And our goal is to move the needle for our customers and for Applied, and that includes how we drive value for them and how we drive our margins higher.
Operator
Harlan Sur, JPMorgan.
Harlan Sur from
So as we look at those companies that are pushing the absolute limits on advanced manufacturing technologies like two-nanometer, and three-nanometer and they’re maxing out their reticle field limits. They’re also pushing advanced packaging technologies, their design — these are the guys that are designing the AI and accelerated compute GPUs, accelerators, for example.
So what we’re hearing is that most of today’s designs on these next-generation chips are targeting tape out second half of ’25 and beyond, they’re specifically focused around these new 3D SOIC architectures, so using hybrid bonding, chip stacking approach, 3-nanometer die, 2-nanometer die on 3-nanometer die, the manufacturability challenges here are pretty significant, right?
But similar to your front-end integrated solutions, I mean, the team has been working on an integrated hybrid bonding system with some of your partners. What’s the time line for introducing these integrated solutions? And will you be trying to intercept these next generation of 3D SOIC solutions coming to market let’s say, in the 2026 time frame?
Gary Dickerson
Harlan, thanks for the question. So heterogeneous integration is one of the biggest areas of focus for Applied and for the entire industry. At our AI event, along with AMD at SEMICON, there was a discussion around driving a 100 times improvement in energy-efficient computing. So again, this is the big race that everybody is focused on. And certainly, packaging is a key part of that. So we’re in deep engagement.
It’s really multiple technology nodes five, 10 years out into the future. And there’s going to be tremendous innovations. If you look even at new DRAM technologies like 4F squared, or high bandwidth memory, you’re going to see adoption of hybrid bonding technologies that are going to be really important for those parts of this ecosystem, and you mentioned what’s happening from a logic standpoint.
So I think the great thing for Applied is we have this broad portfolio in packaging. We have, as you mentioned, the hybrid bonding, digital lithography. We have other new technologies in the pipeline that even expand our portfolio further. We have an advanced full-flow packaging lab in Singapore, where we’re driving co-innovation with our customers to accelerate those architecture inflections.
So I think this is going to be one of the most exciting segments of the market going forward. We’re at about $1.7 billion today. We’ve talked about doubling the packaging revenue over the next few years. And I think it’s going to keep going from there. Again, this is a really key part of the industry drive for energy-efficient computing.
Operator
Charles Shi, Needham & Company.
Charles Shi
I want to follow up on the China question. It looks like your China revenue as well past the peak around almost $3 billion per quarter level, now it’s down to like the low $2 billion. And based on the commentary, it sounds like you are still expecting roughly 40%-ish of the China revenue growth year-on-year.
But I mean, should the current the dollar revenue level sustains into this year, I’m just hoping if you can give us any color because that would imply that China revenue probably is going down year-on-year, roughly 15%, 20% next year. I know you’re not guiding, but that will be the conclusion I draw on all the commentary you provided.
Plus I do want to get a little bit more insight. But if I compare the China revenue performance of Applied Materials with some of your process control peers like (inaudible) they were expecting China to be flat into the second half, but you’re obviously seeing half of the half decline. Since you have the PDC business, and I was hoping you definitely will have some insights into what drives the dispersion there?
Brice Hill
Okay, Charles. I think I would separate between ICAPS, our — the edge technology markets and the DRAM market for China. We had 3 quarters with elevated shipments that we described when we caught up on what was allowed from a DRAM perspective. If you separate that out, we’re currently not expecting that to repeat at least nothing near that magnitude for next year. So that leaves us with the ICAPS market.
And what we expect in the ICAPS market, our Q3 and Q4 and all of this year has been very strong in ICAPS. When we look at China, in particular, utilizations are improving. We have — we’re adding customers. We’ve got a large number of factory projects that are building out.
So if you strip out your estimate for those extraordinary DRAM shipments the last 3 quarters, our expectation over time is that the ICAPS market will grow. We say mid- to high single digits. We won’t make a call for next year, but we would expect growth over time.
Operator
Brian Chin, Stifel.
Brian Chin
So if I look at the past three years, past three 10-Ks, TSMC was your largest customer, but they weren’t 10% in either the April or January quarters. Taiwan, the geography was a little higher this quarter. Did TSMC cross the 10% threshold in fiscal 3Q? And then even bigger picture, this overall pattern would seem to suggest a pretty favorable year-over-year compare moving into 2025 when you think about the type of spending gate-all-around and expansions being discussed. And so I just kind of wanted to get maybe your view on that.
Brice Hill
Yes. TSMC was a 10% customer in Q3. And we think, obviously, they’re a large part of that leading edge investment when we think about the gate all-around technologies and lot of the energy being driven there as we look forward.
One of the other callers talked about the need for innovation, and Gary talked about the need for innovation on leading edge. There’s a lot of energy around that. We see acceleration each quarter on — of this year on leading edge and really haven’t changed our outlook as we’ve moved through the year.
Operator
Blayne Curtis, Jefferies.
Blayne Curtis
I may be a little confused on the history of ICAPS and the strength. I thought you were talking about last quarter that segment kind of being down slightly, and I think you just said mid- to high single digits. So maybe I have my reference point wrong, but also may it’d be helpful if you could just talk about the growth you saw in July, 15% sequentially for foundry logic. Can you just relate that to leading edge and ICAPS?
I think the reception is leading is very strong, but it sounds like your comments on ICAPS are suggesting that ICAPS is also strong. But then I’m trying to fit that all with like the comment that it’s going to be 50% of the mix, which I think — I thought that business is running a lot higher. So if you can kind of just help us a little history and then how July played out would be helpful.
Brice Hill
Yes. Thanks, Blayne. So in the quarter, we did see strength in both end markets, leading logic and ICAPS. Leading logic is accelerating every quarter this year, and we said ICAPS is very strong and continues to be strong throughout the year. This will probably be a record year for us in ICAPS for our fiscal year anyway.
And so leading logic is accelerating based on the gate-all-around investments and the AI trends that we’re talking about. And then ICAPS, the mid- to high single digits was a longer-term forecast. We’ve had utilizations improving.
And when we think about that market, we don’t give a guide for next year, but we think there will be continued growth in the mid- to high single digits as all those different end markets like AI sensors, electrification, autonomous vehicles, renewable energy, all those end markets continue to build out and grow faster than GDP. That’s our expectation
Operator
Jed Dorsheimer, William Blair.
Jonathan Dorsheimer
Just a two-part one on the ICAPS, Gary. So first, in that business, that was just over in some of your — some of the fabs or your customers over in Malaysia last week. And it seems like capacity expansion outside of China has started again after some of the EV pressures have shifted to opportunities in power density and data centers. So I’m just wondering if you could comment on that.
And then secondly, just as it relates to services for ICAPS, is that the same on leading edge? Or are there nuances in terms of the amount and timing there?
Gary Dickerson
Jed, yeah, so ICAPS, we’re very bullish on ICAPS longer term. Brice talked about kind of mid, high single-digit growth rates in ICAPS over time. Growth driven by all of those different segments, IoT, industrial automation, robotics, really edge computing also for AI. We think that’s going to be a really strong business. Over — I think many of the people on the call know that over five years ago, it’s actually April 12, 2019, we formed our ICAPS group.
And since we formed that group, just to focus on that market for IoT, communication, auto power and sensors, we’ve had share gains in that market. We’ve introduced more than 20 major new ICAPS products. And this group is just completely focused on innovation in those device segments. And there are races that are happening there.
Your comment on what’s happening on the different device segments. If you look at power electronics, there will be some major architecture inflections that will happen there that are important for electric vehicles and renewable energy. And we have an ICAPS architecture innovation group working — co-innovating with our customers on those new ICAPS architectures. That’s part of the strategy that we’ve built over the last 5 years.
We have new products that are in the pipeline that will serve large ICAPS new segments and new products for cost-competitive applications. So again, it’s been a big focus for us over a number of years. We’ve built tremendous capabilities.
From a service standpoint, ICAPS is also a good market for us. And that’s part of that overall growth opportunity for us, not only on systems where we’ve gained a significant amount of share. I’m very positive on how we’re positioned. The actions that we’ve taken I think really position us going forward in ICAPS.
Brice, I don’t know if you want to add anything else?
Brice Hill
No, I would just say that two of our regions grew during this Q3. So if the question is, is there investment in other areas of the world absolutely and we expect that to continue. Thanks for the question Jed.
Michael Sullivan
And operator, we have time for two more questions, please.
Operator
Mehdi Hosseini, Susquehanna.
Mehdi Hosseini
Just a very quick follow-up for Brice. If you could give us an update with the progression towards the 48.9% gross margin target. I assume that’s for fiscal year ’25. And how the funding for the EPIC R&D center, which I think is a $4 billion project is going to impact those targets?
Brice Hill
Okay. Thanks, Mehdi. So the update — the gross margin of 47.4%, we expect to make — we talked about very gradual improvements as we work towards that 48%. And that would still be my expectation. You see actually the movement between last quarter and this quarter. was modest. We fought those headwinds that we had from a mix perspective.
On the EPIC, we’re — we did get the notice that we’re not getting the grants, but EPIC, that’s an important platform for us, it stands for equipment process innovation and commercialization. It’s going to allow us to accelerate co-innovation with all of our customers. We are continuing with that investment. It will elevate our CapEx as we look forward. And we are benefiting from the investment tax credit that’s related to those types of projects. So we still do have some help from the government from that perspective.
And just a couple of other modeling notes since we’re on CapEx. So as we look forward, the CapEx will — for the EPIC center, and in general, we’ll be at a higher run rate than we’ve had in the past. If you look forward in your modeling, I just wanted to highlight that Q1 will be our normal step-up for pay-related increases and that you can look at the history to see what that step-up should look like if you’re modeling that.
And then the last piece would be our tax rate. Tax rate next year expectation will be 14% versus the 12.5% this year as more of our mix will be — more of our worldwide mix will be US related, so that will change the tax mix. So thanks for the opportunity to get those in there.
Gary Dickerson
Yeah, Mehdi, this is Gary. Maybe just a little bit more color on EPIC. As we’ve talked about, the whole industry is in a race to be first to market to deliver innovations for energy-efficient computing. So we — in discussions with our partners, our customers and partners, we really believe that EPIC can enable us to innovate — the way we innovate, really innovate in parallel.
So accelerating both for our customers and partners and for Applied, our ability to bring those energy-efficient innovations to market. So the concept we’re getting tremendous traction and pull with our customers and partners.
And we also think of EPIC as a global platform. So where we make investments is going to be based on incentives, talent, proximity to innovation ecosystem. So we are moving forward for sure with this EPIC concept. Very good traction with customers.
And again, over the last several years, we’ve built out this innovation engine inside Applied Materials. We have architecture innovation teams that are working on ICAPS. They’re working on foundry logic and DRAM and packaging. We have this unique and connected portfolio of materials innovations that are crucial for all of those major inflections.
So EPIC is a way for us to innovate — the way we innovate to move all of that in parallel to accelerate all of these major inflections that are crucial for AI, data center and all those major tectonic shifts in technology that we talked about earlier. But again, where we invest is going to be based on incentives, talent and proximity to local innovation ecosystem.
Michael Sullivan
I think we’re approaching the end of the hour. So Brice, I’m going to go ahead and ask if you have any closing thoughts for us.
Brice Hill
Okay. Thanks, Mike. From my perspective, I’m pleased that the investments we’ve been making in the technology road map inflections and also in our operational capabilities are showing up in our financial results this quarter on both the revenue and gross margin lines.
I believe we’ve put the company in a great position to grow along with megatrends like data center AI as a whole host of new technologies come to market across leading-edge foundry logic, DRAM, advanced packaging and ICAPS specialty chips.
We have the feel we need to keep driving the materials engineering road map with our customers while also distributing profits to our shareholders through dividends and buybacks. Gary will be at the Goldman Sachs Conference in San Francisco on September 11, and I hope to see many of you at the Citi conference in New York on September 4. Mike, thank you, and please close the call.
Michael Sullivan
Okay. Thanks, Brice, and we’d like to thank everybody for joining us today. A replay of today’s call is going to be available on the IR page of our website by 5:00 Pacific Time. Thank you for your continued interest in Applied Materials.
Operator
Thank you, ladies and gentlemen, for your participation in today’s conference. This does conclude the program. You may now disconnect. Good day.
