科磊 (KLAC) 2025 Q2 法說會逐字稿

Good afternoon. My name is Todd, and I will be your conference operator today. At this time, I would like to welcome everyone to the KLA Corporation December quarter 2024 earnings conference call and webcast. (Operator Instructions)

午安.我叫托德，今天我將擔任您的會議主持人。現在，我歡迎大家參加 KLA Corporation 2024 年 12 月季度收益電話會議和網路廣播。（操作員指令）

I will now turn the call over to Kevin Kessel, Vice President of Investor Relations and Market Analytics. Please go ahead.

現在我將電話轉給投資者關係和市場分析副總裁凱文凱塞爾 (Kevin Kessel)。請繼續。

Welcome to our earnings call to discuss the December quarter and calendar year 2024 results and outlook. I'm joined by our CEO, Rick Wallace; and our CFO, Bren Higgins. We will discuss today's results released after the market close and available at ir.kla.com, along with supplemental materials. Today's discussion and metrics are presented on a non-GAAP financial basis, unless otherwise specified. All full-year references we make are to calendar years.

歡迎參加我們的財報電話會議，討論 12 月季度和 2024 日曆年的業績和展望。與我一起參加的還有我們的執行長 Rick Wallace；以及我們的財務長 Bren Higgins。我們將討論今天收盤後發布的結果（可在 ir.kla.com 上查閱）以及補充資料。除非另有說明，今天的討論和指標均以非 GAAP 財務基礎呈現。我們所提及的所有全年數據均為日曆年。

The earnings materials contain a detailed reconciliation of GAAP to non-GAAP results. KLA's IR website also contains future investor events, presentations, corporate governance information and links to our SEC filings, including our most recent annual report and quarterly reports on Forms 10-K and 10-Q.

收益資料包含 GAAP 與非 GAAP 結果的詳細對帳。KLA 的 IR 網站還包含未來投資者活動、簡報、公司治理資訊以及我們 SEC 文件的鏈接，包括我們最新的年度報告和 10-K 表和 10-Q 表的季度報告。

Our comments today are subject to risks and uncertainties reflected in the disclosure of risk factors in our SEC filings. Any forward-looking statements, including those we make on the call today, are also subject to those risks, and KLA cannot guarantee those forward-looking statements will come true. Our actual results may differ significantly from those projected in our forward-looking statements.

我們今天的評論受到我們向美國證券交易委員會提交的文件中披露的風險因素所反映的風險和不確定性的影響。任何前瞻性陳述，包括我們今天在電話會議上所做的陳述，也受這些風險的影響，而 KLA 不能保證這些前瞻性陳述會成真。我們的實際結果可能與前瞻性陳述中的預測有很大差異。

Rick will start with some introductory comments, followed by Bren with financial highlights and our outlook. Before I turn the call over to our CEO, Rick Wallace, I wanted to remind everyone that our 2025 Investor Day will be held on the morning of June 18 in New York City. Now over to Rick.

里克 (Rick) 會先作一些介紹性評論，然後布倫 (Bren) 會介紹財務亮點和我們的展望。在我將電話轉給我們的執行長 Rick Wallace 之前，我想提醒大家，我們的 2025 年投資者日將於 6 月 18 日上午在紐約市舉行。現在交給 Rick。

Thank you, Kevin. I will summarize KLA's overall performance for calendar '24 and the December quarter and cover company highlights and updates on the industry landscape. For calendar 2024, KLA again delivered relative growth outperformance, strong profitability, and healthy return to shareholders.

謝謝你，凱文。我將總結 KLA 24 年和 12 月季度的整體表現，並介紹公司亮點和行業情況的最新動態。2024 年，KLA 再次實現了相對優異的成長表現、強勁的獲利能力和豐厚的股東回報。

Specifically, 2024 revenue grew 12% to a record $10.85 billion, and the Process Control revenue grew by over 12%, which indicates increased market share, while the Services business grew 15% to $2.5 billion for the year.

具體而言，2024 年營收成長 12%，達到創紀錄的 108.5 億美元，製程控制收入成長超過 12%，顯示市場佔有率增加，而服務業務全年成長 15%，達到 25 億美元。

Also for the calendar year, KLA maintained industry-leading gross and operating margins at 61% and 41%, respectively. The company grew free cash flow to $3.4 billion and returned $2.9 billion in a combination of dividends and share buybacks. KLA's outperformance for the year was driven by a return to growth at the leading edge, which includes increased investments in AI, high-performance computing, and continued momentum in advanced packaging as well as sustainable performance for KLA's Services business.

此外，就本年度而言，KLA 的毛利率和營業利潤率分別保持行業領先，為 61% 和 41%。該公司的自由現金流成長至 34 億美元，並透過股息和股票回購返還了 29 億美元。KLA 今年的出色表現得益於前沿技術的恢復成長，其中包括對人工智慧和高效能運算的投資增加、先進封裝領域的持續發展勢頭以及 KLA 服務業務的可持續表現。

And turning to the December quarter results for KLA came in above the midpoint of non-GAAP guidance range, despite navigating through the business impact of new US government export controls which were released late in the quarter. Specifically, the quarter revenue topped $3 billion for the first time. Diluted GAAP was $8.20, finishing at the upper end of the guidance range for the quarter. GAAP diluted EPS was $6.16.

儘管美國政府在本季末推出的新出口管制措施對業務產生了影響，但 KLA 12 月季度的業績仍高於非 GAAP 指引範圍的中點。具體來說，本季營收首度突破30億美元。稀釋後 GAAP 利潤為 8.20 美元，達到本季預期範圍的上限。GAAP 稀釋每股收益為 6.16 美元。

The business landscape is performing as expected, and we're encouraged by the strong demand we're experiencing in leading-edge logic with specific memory customers supporting high-bandwidth memory and advanced packaging. KLA's differentiated portfolio of solutions aligns exceptionally well with enabling our customers to navigate increasing complexity, growing design starts, and larger semiconductor devices in an environment of rising semiconductor demand.

業務前景表現符合預期，特定記憶體客戶對支援高頻寬記憶體和先進封裝的前沿邏輯的強勁需求令我們感到鼓舞。KLA 差異化的解決方案組合能夠完美地幫助我們的客戶在半導體需求不斷增長的環境中應對日益複雜、不斷增長的設計起點和更大的半導體設備。

Specific highlights in the quarter include a combination of strong sequential and year-over-year revenue growth, demonstrate an improving industry environment. KLA is specifically positioned to benefit from accelerating growth at the leading edge. Across all sectors, there are technology development investments AI and HBM as well as strengthening supply/demand environment which positions the wafer fab equipment industry for growth in calendar 2025.

本季度的具體亮點包括強勁的環比和同比收入成長，顯示產業環境正在改善。KLA 的優勢在於能夠從尖端技術的加速成長中獲益。在所有領域中，都有對 AI 和 HBM 的技術開發投資以及不斷加強的供需環境，這將為晶圓廠設備產業在 2025 年實現成長做好準備。

AI continues to be a crucial catalyst for KLA. We are well aware of the recent revelations of DeepSeek and the implications that it portends a diminished demand for advanced semiconductor in support of the AI infrastructure build-out.

人工智慧仍然是 KLA 的重要催化劑。我們非常清楚 DeepSeek 的最新進展及其意義，即支援 AI 基礎設施建設的先進半導體需求將減少。

As a company that has been developing AI models for use in our own inspection systems for many years, our own experience supports the theory that increased compute efficiency enables more adoption of AI on our platforms. The demand is clearly elastic. As it pertains to the demand environment for advanced semiconductor, we see no reason to believe that the increased compute efficiency in AI will have an impact on the advanced demand environment in the foreseeable future.

作為一家多年來一直致力於開發用於我們自己的檢查系統的人工智慧模型的公司，我們自己的經驗支持這樣的理論：提高計算效率可以使我們的平台更多地採用人工智慧。需求顯然是有彈性的。就先進半導體的需求環境而言，我們認為沒有理由相信人工智慧運算效率的提高將在可預見的未來對先進的需求環境產生影響。

AI is both an important driver and enabler of KLA's business. Specific drivers connected to AI that are a positive for KLA's growth are higher volume and higher value wafer demand, more complex designs, accelerating product cycles, larger die size, and growing advanced packaging demand. These trends demonstrate the increasing value of process control and assisting our customers through managing a dynamic production environment as investments and complexity increase.

AI是KLA業務的重要驅動力與推動力。與 AI 相關的、對 KLA 成長有利的具體驅動因素包括更高數量和更高價值的晶圓需求、更複雜的設計、加快的產品週期、更大的晶片尺寸以及不斷增長的先進封裝需求。這些趨勢表明，隨著投資和複雜性的增加，製程控制的價值不斷增加，並透過管理動態生產環境來協助我們的客戶。

Exemplifying this momentum for our advanced packaging portfolio continued in the quarter, the growing demand for more powerful systems of chips is driving more complex heterogeneous chip integration enabled by advanced packaging, which increases the value of process control in the chip package. This is fueling growth for KLA and our broad portfolio of systems.

本季度，我們先進封裝產品組合的勢頭繼續保持，對更強大的晶片系統的需求不斷增長，推動了先進封裝實現的更複雜的異構晶片集成，這增加了晶片封裝中製程控制的價值。這推動了 KLA 和我們廣泛的系統組合的成長。

KLA Advanced Packaging revenue grew to approximately $500 million in calendar 2024 and is expected to exceed $800 million in calendar 2025, up from our last estimate of $750 million. KLA's Service business grew to $667 million in the December quarter, up 4% sequentially and 18% year over year. This makes 50 consecutive quarters of growth for our Services business on a year-over-year basis.

KLA Advanced Packaging 的營收在 2024 年成長至約 5 億美元，預計在 2025 年將超過 8 億美元，高於我們上次估計的 7.5 億美元。KLA 的服務業務在 12 月季度成長至 6.67 億美元，較上季成長 4%，較去年同期成長 18%。這使得我們的服務業務連續 50 個季度實現同比增長。

Finally, quarterly free cash flow was $757 million in calendar 2024, and the free cash flow margin was 31% over the same period, putting KLA amongst the top companies in the S&P 500. Total capital return in the December quarter was $877 million, comprised of $650 million of share repurchase and $227 million in dividends. Total capital return over the past 12 months was $2.9 billion.

最後，2024 年曆年的季度自由現金流為 7.57 億美元，同期的自由現金流利潤率為 31%，使 KLA 躋身標準普爾 500 指數中的頂級公司之列。12 月季度的總資本回報為 8.77 億美元，其中包括 6.5 億美元的股票回購和 2.27 億美元的股息。過去 12 個月的總資本回報為 29 億美元。

KLA have used consistent and healthy capital returns as fundamental to delivering value for shareholders. KLA's December quarter results delivered strong sequential and year-over-year growth, which validates KLA's process control leadership and portfolio strength. KLA Operating Model and the dedication of our global team continues to be the foundation of our sustained success.

KLA 一直將持續健康的資本回報作為股東實現價值的基礎。KLA 12 月季度的業績實現了強勁的環比和同比增長，證明了 KLA 的製程控制領導地位和產品組合實力。KLA 營運模式和我們全球團隊的奉獻精神繼續成為我們持續成功的基礎。

I'll now pass the call over to Bren to cover financial highlights and our outlook.

現在我將把電話轉給布倫，讓他介紹財務亮點和我們的展望。

Thanks, Rick. KLA's December quarter results demonstrate market leadership combined with the consistent execution and dedication of our global team to meet customer commitments and drive sequential and year-over-year growth profitability improvements. Revenue was $3.08 billion, above the guidance midpoint of $2.95 billion.

謝謝，里克。KLA 12 月季度的業績證明了市場領導地位，加上我們全球團隊的一致執行力和奉獻精神，以履行對客戶的承諾並推動連續和同比增長的盈利能力提高。營收為 30.8 億美元，高於預期中位數 29.5 億美元。

Non-GAAP diluted EPS was $8.20, above the guidance midpoint. GAAP diluted EPS was $6.16. Gross margin was 61.7%, operating expenses were $596 million. Operating expenses were comprised of $342 million in R&D and $254 million in SG&A. Operating margin was 42.3%. Other income and expense net was a $31 million expense.

非公認會計準則稀釋每股收益為 8.20 美元，高於指導中位數。GAAP 稀釋每股收益為 6.16 美元。毛利率為61.7%，營業費用為5.96億美元。營業費用包括 3.42 億美元的研發費用和 2.54 億美元的銷售、一般及行政費用。營業利益率為42.3%。其他收入和支出淨額為3100萬美元。

The quarterly effective tax rate was 13.7%. Net income was $1.1 billion, cash flow from operations was $850 million, and free cash flow was $757 million. The breakdown of revenue by reportable segments and markets, major products and regions can be found within the shareholder letter and slides.

季度有效稅率為13.7%。淨收入為 11 億美元，營運現金流為 8.5 億美元，自由現金流為 7.57 億美元。報告分部和市場、主要產品和地區的收入明細可在股東信函和幻燈片中找到。

Moving to the balance sheet. KLA ended the quarter with $3.8 billion in total cash, cash equivalents, and marketable securities; debt of $5.9 billion; and a flexible and attractive bond maturity profile supported by strong investment-grade ratings from all three major rating agencies.

轉到資產負債表。KLA 本季結束時的現金、現金等價物及有價證券總額為 38 億美元；債務 59 億美元；其債券期限結構靈活且具有吸引力，並得到三大評級機構強勁的投資等級評級的支持。

During the December quarter, we retired our $750 million November 2024 bond at maturity with cash on hand. KLA's balance sheet provides the ability to fund our growth strategy, organic and inorganic, and offer attractive capital returns to shareholders.

在 12 月季度，我們以現金償還了到期的 7.5 億美元 2024 年 11 月債券。KLA 的資產負債表為我們的成長策略（有機和無機）提供了資金，並為股東提供了有吸引力的資本回報。

Turning to our outlook. The industry outlook continues to gain momentum in the near term, driven by an increasing investment in leading-edge logic, high-bandwidth memory, and advanced packaging. We expect the WFE market to grow by a mid-single-digit percentage in 2025 from the high-$90 billion level for calendar 2024.

轉向我們的展望。受前沿邏輯、高頻寬記憶體和先進封裝領域投資不斷增加的推動，產業前景在短期內繼續保持成長勢頭。我們預計，到 2025 年，WFE 市場規模將從 2024 年的 900 億美元高點成長至中等個位數百分比。

Growth in calendar 2025 is expected to be fueled principally by increasing investment in both leading-edge Foundry/Logic and Memory to support growing AI and premium mobile demand, offset by lower overall demand from China due to the digestion of elevated levels of investment over the past couple of years.

預計2025 年的成長主要將受到對尖端晶圓代工廠/邏輯和內存的投資增加的推動，以支持不斷增長的人工智慧和高端移動需求，而中國由於消化過去20 年高額投資導致的總體需求下降將抵消這一影響。

In an encouraging development, our top customer recently said in an earnings call that they expect the number of new takeouts for N2, or the 2-nanometer node, in the first two years to be higher than both N3 and N5 in their first two years, fueled by both smartphone and HPC applications. As communicated in early December, we continue to estimate the impact on KLA's revenue in calendar 2025 from recent export controls in China to be approximately $500 million, plus or minus $100 million, with roughly 70% of the impact affecting our systems business.

令人鼓舞的是，我們的頂級客戶最近在一次財報電話會議上表示，他們預計 N2 或 2 奈米節點的新產品出貨量在前兩年將高於 N3 和 N5 ，由智慧型手機和HPC 應用推動。正如 12 月初所傳達的那樣，我們繼續估計最近的中國出口管制對 KLA 2025 日曆年收入的影響約為 5 億美元，上下浮動 1 億美元，其中約 70% 的影響影響我們的系統業務。

While we are hopeful based on our interpretation of the regulations that there should be licensing opportunities that will mitigate some of this impact, we are taking a cautious view given the significant delays in processing license request by the US government over the past few years. However, given KLA's business momentum, market share opportunities, and higher expected process control intensity at the leading edge across all segments, we are confident we will continue to deliver growth outperformance compared with the WFE market in 2025.

雖然根據我們對法規的解讀，我們希望有許可機會可以減輕部分影響，但考慮到過去幾年美國政府在處理許可申請方面出現了重大延誤，我們對此持謹慎態度。然而，鑑於 KLA 的業務發展勢頭、市場份額機會以及所有細分市場領先地位的更高預期過程控制強度，我們相信，與 2025 年的 WFE 市場相比，我們將繼續實現優異的成長績效。

KLA's March quarter guidance is as follows. Total revenue is expected to be $3 billion, plus or minus $150 million. Our revenue guidance is up 27% year over year at the midpoint, further illustrating the improvement we expect to see in calendar 2025. Foundry/Logic revenue from semiconductor customers is forecasted to be approximately 73%, and Memory is expected to be approximately 27% of semi process control systems revenue to semiconductor customers. Within Memory, DRAM is expected to be about 75% of the revenue mix and NAND, the remaining 25%.

KLA 3 月份季度指引如下。預計總收入為 30 億美元，上下浮動 1.5 億美元。我們的收入預期中位數年增 27%，進一步說明了我們預計 2025 年將出現的改善。預計來自半導體客戶的代工/邏輯收入將佔半製程控制系統收入的約 73%，而記憶體收入預計佔半導體客戶收入的約 27%。在記憶體領域，DRAM 預計佔收入結構的 75% 左右，而 NAND 則佔剩餘的 25%。

Non-GAAP gross margin is forecasted to be 62%, plus or minus 1 percentage point or up approximately 30 basis points sequentially at the midpoint, despite slightly lower revenue, primarily due to more favorable product mix expectations.

儘管收入略有下降，但非 GAAP 毛利率預計為 62%，中位數上下浮動 1 個百分點或比上一季上升約 30 個基點，這主要歸因於更有利的產品組合預期。

For calendar 2025, based on expectations for business mix across systems and service, systems product mix and factory utilization, we expect gross margin for the year to be approximately 62%, plus or minus 50 basis points. Non-GAAP operating expenses are forecasted in the March quarter to be approximately $585 million as we continue to make significant product development and scaling investments to support expected revenue growth.

對於 2025 年，基於對系統和服務業務組合、系統產品組合和工廠利用率的預期，我們預計當年的毛利率約為 62%，上下浮動 50 個基點。由於我們將繼續進行大量產品開發和擴大投資以支持預期的收入成長，預計 3 月非 GAAP 營業費用約為 5.85 億美元。

Given our expectations for company growth over the next couple of years, we will maintain our operating expense trajectory. For the remainder of calendar 2025, we expect sequential increases of approximately $15 million in incremental operating expenses per quarter. This is driven by our priority around our product development roadmap requirements as well as revenue growth expectations.

考慮到我們對未來幾年公司成長的預期，我們將維持我們的營運費用走勢。在 2025 年剩餘時間內，我們預計每季的增量營運費用將連續增加約 1,500 萬美元。這是由我們對產品開發路線圖要求以及收入成長預期的優先考慮所驅動的。

Our business model is predicated on ensuring 40% to 50% incremental non-GAAP operating margin leverage on revenue growth over the long run. Other model assumptions include non-GAAP other income and expense net of approximately $36 million expense for the March quarter and expect this to be roughly consistent throughout the calendar year. The tax assumption for March remains at 13.5%, and we expect this to remain through the June quarter.

我們的商業模式是建立在確保長期內非公認會計準則下的營業利潤率能夠提高 40% 至 50% 的基礎之上。其他模型假設包括 3 月季度非 GAAP 其他收入和支出淨額約 3,600 萬美元，並預計這一數字在整個日曆年內大致保持一致。3 月的稅率假設仍為 13.5%，我們預計這一比率將持續到 6 月季度。

Beginning in the September quarter, which is the first quarter of our fiscal year, our tax rate will reflect the adoption of global taxation pillar 2. Based on our current modeling, we think pillar 2 implementation will drive the tax rate slightly higher to approximately 14% in the second half of the calendar year. We will provide an update on this planning rate midyear, if necessary.

從九月季度（即我們財政年度的第一季）開始，我們的稅率將反映全球稅收支柱 2 的採用。根據我們目前的模型，我們認為第二支柱的實施將使稅率在下半年小幅上漲至約 14%。如果有必要，我們將在年中更新此計劃率。

For the March quarter, GAAP diluted EPS is expected to be $7.77, plus or minus $0.60. Non-GAAP diluted EPS of $8.05, plus or minus $0.60. EPS guidance is based on a fully diluted share count of approximately 133.3 million shares.

3 月季度，預計 GAAP 稀釋每股收益為 7.77 美元，並上下浮動 0.60 美元。非 GAAP 稀釋每股收益為 8.05 美元，上下浮動 0.60 美元。 EPS 指引是基於約 1.333 億股的完全稀釋股數。

In conclusion, our near-term revenue guidance points to relative stability around current business levels. Based on the strength of our backlog and market position, we see growth in calendar 2025 and expect to outperform the mid-single-digit growth rate we expect from the WFE market. KLA is focused on delivering a differentiated product portfolio that addresses customers' technology road map requirements and drives our longer-term relevancy and growth expectations.

總之，我們的近期收入指引顯示目前業務水準相對穩定。基於我們的積壓訂單和市場地位，我們預計 2025 年將出現成長，並有望超過我們預期的 WFE 市場中個位數的成長率。KLA 致力於提供差異化的產品組合，以滿足客戶的技術路線圖要求並推動我們長期的相關性和成長預期。

With the KLA Operating Model guiding our best-in-class execution, KLA is focused on implementing our strategic objectives designed to drive out performance. KLA's focus on customer success, innovative solutions and operational excellence drives industry-leading financial free cash flow performance and allows us to return capital consistently.

在 KLA 營運模式指導我們一流的執行下，KLA 專注於實現旨在提高績效的策略目標。KLA 專注於客戶成功、創新解決方案和卓越運營，推動業界領先的財務自由現金流表現，並使我們能夠持續回報資本。

That concludes the prepared remarks. Let's begin the Q&A.

準備好的發言到此結束。讓我們開始問答。

Thank you, Bren. Operator, can you please provide the instructions for Q&A?

謝謝你，布倫。接線生，您能提供問答的說明嗎？

(Operator Instructions) Vivek Arya, Bank of America.

(操作員指示) 美國銀行 Vivek Arya。

This is Michael on for Vivek Arya. To start, we heard from two of your peers yesterday, the peer that is action deposition focused issued a pretty similar WFE view to you guys for indexing on digits, while the other reiterated their annual guide suggesting that those spending remains healthy.

這是 Michael 為 Vivek Arya 表演的。首先，我們昨天聽取了兩位同行的意見，一位專注於行動的同行對數位索引發表了與你們非常相似的 WFE 觀點，而另一位則重申了他們的年度指南，表明這些支出仍然健康。

So the question is if total WFE is increasing something like $5 billion or so this year, but within that, litho WFE is also increasing (inaudible) also growing pretty strong, so consuming a good part of that incremental growth.

所以問題是，如果今年總的 WFE 增加了 50 億美元左右，但在這其中，光刻 WFE 也在增加（聽不清楚），而且增長相當強勁，因此消耗了增量的很大一部分。

What exactly is happening to the process control part of the market this year? And if process control WFE is growing solidly, which it seems like it is, does that suggest that (inaudible) for WFE could be conservative? Or are there other parts of the market that are shrinking by that much?

今年市場的過程控制部分到底發生了什麼事？如果製程控制 WFE 穩定成長（看起來確實如此），這是否意味著（聽不清楚）對 WFE 來說可能比較保守？或者市場的其他部分也出現瞭如此大的萎縮？

Yes, this is Bren. I'll take that one. And the guidance was clear that we think it's somewhere in that range of about $5 billion to use your number into 2025 versus 2024. We feel pretty good about KLA's share of overall WFE opportunity into next year.

是的，這是布倫。我要那個。指導意見很明確，我們認為按照您的數字計算，2025 年而非 2024 年，這一數字應該在 50 億美元左右。我們對 KLA 明年在整體 WFE 機會中的份額感到非常樂觀。

Obviously, as we move into '25, we've got more investment in leading edge, and that's certainly a nice driver for our business, and we're already seeing KLA's share of WFE at the N2 node being meaningfully greater than what we saw at N3.

顯然，隨著我們進入25 年，我們在前沿技術方面投入了更多資金，這無疑對我們的業務產生了良好的推動作用，而且我們已經看到KLA 在N2 節點的WFE 份額明顯高於我們所看到的在 N3。

What's happening in the DRAM, first advanced DRAM is good for us with scaling in EUV, but also as you look at high-bandwidth memory, high-bandwidth memory is also driving process control intensity due to the lack of redundancy, more complex logic circuitry and the base die. The need for more reliability, bigger chips and so on. So we feel pretty good about all of that.

DRAM 領域正在發生的事情，首先，先進的DRAM 對我們在EUV 方面的擴展很有利，但當你看到高頻寬記憶體時，由於缺乏冗餘、邏輯電路更複雜，高頻寬記憶體也在推動製程控制強度和基座模具。需要更高的可靠性、更大的晶片等等。因此，我們對這一切感到非常滿意。

And then I think, finally, the growth that we referenced in the letter in advanced packaging is accelerating for the company. It seems like every quarter, I keep raising the numbers.

最後，我認為，我們在信中提到的公司在先進封裝領域的成長正在加速。似乎每個季度，我的數字都在不斷提高。

So I'm pretty excited about the opportunities that are there, and it seems to be accelerating as we move into this year. So for all those reasons, it looks like process control intensity, KLA share of market looks to increase in 2025 based on our assessment. And in an environment where memory is probably a higher percent given expected growth in DRAM, I think the dynamics I talk about more than offset slightly changing mix that's still Foundry/Logic heavy, but a little bit more DRAM in terms of our views on '25.

因此，我對那裡的機會感到非常興奮，隨著今年的到來，這種機會似乎正在加速。因此，基於所有這些原因，根據我們的評估，過程控制強度和 KLA 的市場份額似乎將在 2025 年增加。在記憶體可能以更高的百分比出現在 DRAM 預期增長的環境中，我認為我談論的動態足以抵消仍然以 Foundry/Logic 為主的組合略有變化的影響，但就我們對「 25.

Got it. That's helpful. And for my next question, just could you help us with the linearity for revenue this year, the best extent you can, just because we're kind of at a high order mark to revenue this quarter? So should we expect maybe it to be more first-half weighted versus second half, especially given China is normalizing into the year and there's this impact of the export restrictions that we should consider?

知道了。這很有幫助。我的下一個問題是，您能否盡可能幫助我們了解今年營收的線性關係，因為本季我們的訂單與營收比率較高？那麼，我們是否應該預期上半年的權重會大於下半年，尤其是考慮到中國今年正在恢復正常，我們應該考慮出口限制的影響？

Yes, the second half, I'm not going to comment on it. As we said in the letter, we feel pretty good about relative stability as we look at the funnel here moving forward. We'll see as we start to move through the year, what happens. But in terms of how we're modeling the company, it seems that we're bouncing around this $3 billion level, plus or minus, at least as we look at the first half of the year.

是的，後半部我不打算評論。正如我們在信中所說，當我們看到未來的發展趨勢時，我們對相對穩定感到非常滿意。隨著新的一年的到來，我們將看到會發生什麼。但就我們如何塑造公司而言，至少從今年上半年的情況來看，我們的收入似乎在 30 億美元左右波動。

(Operator Instructions) Harlan Sur, JPMorgan.

（操作員指示） Harlan Sur，摩根大通。

Process Control, a strong outperformance for the team in calendar '24. I think your Process Control Systems business was up 15% year over year, right? That's versus WFE at up kind of mid-single digits. But if I look back over the past five years, the team's process control business has outgrown WFE on average by about 500 to 700 basis points per year.

過程控制，這是團隊在 24 日曆年表現優異的表現。我認為你們的製程控制系統業務年增了 15％，對嗎？相比之下，WFE 的增幅則為個位數中段。但如果我回顧過去五年，團隊的流程控制業務每年平均比 WFE 成長約 500 到 700 個基點。

So given this year is going to be more leading-edge technology inflection driven, which is where obviously you guys have a strong leadership position, is it fair to assume that if WFE is up mid-single-digit percentage points this year that your process control business should grow kind of low to mid-teens percent in calendar '25?

因此，考慮到今年將更多地由前沿技術推動，而你們顯然在這方面擁有強大的領導地位，是否可以合理地假設，如果 WFE 今年增長了中個位數的百分點，你們的流程控制業務在25 年應該會成長10% 左右嗎？

Harlan, thank you for those comments. Obviously, we're not going to guide for the year. But clearly, we're feeling great about the position we have. And a couple of things have changed, as you well know, in terms of the dynamics. One, because we've resumed scaling, there's more opportunity for more inspection.

哈蘭，謝謝你的評論。顯然，我們不會為今年做出指引。但顯然，我們對自己的地位感到非常滿意。正如你們所知，從動態角度來看，有幾件事已經改變了。一是因為我們已經恢復了擴展，所以有更多的機會進行檢查。

So I've always viewed the opportunities as being twofold. One, there needed to be an opportunity, and then we had to have a solution. And sometimes we've had the case where, for example, if you go back years to 3D NAND, where, of course, if we could have looked through and found defects, there would have been an opportunity, it was tough to do.

因此我一直認為機會是雙重的。首先，需要有機會，然後我們必須有解決方案。有時我們會遇到這樣的情況，例如，如果你回顧幾年前的 3D NAND，當然，如果我們能夠仔細檢查並發現缺陷，那麼就會有機會，但這很難做到。

But now, we're really at an interesting point where the leading edge every dynamic is going kind of in our favor. In terms of higher-value wafers, you've got larger die size. I talked about this in the prepared remarks. And you've got accelerated technology nodes and more layers that need to be figured out. The other dynamic, of course, that you know is HBM, is looking more like logic than it used to, less redundancy, more valuable die and, of course, the dynamic around packaging.

但現在，我們確實處於一個有趣的節點，前沿的每一個動態都對我們有利。就更高價值的晶圓而言，晶片尺寸更大。我在準備好的發言中談到了這一點。而且您已經獲得了加速的技術節點和更多需要解決的層。當然，您知道的另一個動態是 HBM，它看起來比以前更像邏輯，冗餘更少，晶片更值錢，當然還有圍繞封裝的動態。

So we feel great, as Bren said, about process controls position and overall the spend for our customers. And that's the conversation we're having with our leading customers is very focused on getting that availability and being supportive of the technology ramps as they make these big investments going forward.

因此，正如布倫所說，我們對製程控制狀況以及我們為客戶帶來的整體支出感到非常滿意。這就是我們與主要客戶進行的對話，重點在於獲得這種可用性，並支持他們在未來進行這些重大投資時的技術提升。

That's great. And then on the 60% growth outlook for your advanced packaging business this year, can you just kind of help us unpack that a bit? How much of that mix is 2.5D packaging technology like CoWoS HBM versus other packaging types? And what is the rough mix of process control versus semi manufacturing systems? And then we're already starting to see some future AI designs moving to 3D SoIC technology starting next year.

那太棒了。那麼，關於今年先進封裝業務 60％ 的成長前景，您能否幫我們稍微解釋一下？與其他封裝類型相比，CoWoS HBM 等 2.5D 封裝技術佔比為何？製程控制與半製造系統大致的組合是怎樣的？然後我們已經開始看到一些未來的 AI 設計從明年轉向 3D SoIC 技術。

Is this a further tailwind for the team, given the higher complexity of these next-generation 3D architectures?

考慮到下一代 3D 架構的更高複雜性，這對團隊來說是否會是進一步的順風？

Yes. Great questions. I mean I think two things have happened, and we got this early indication from our leading customers 1.5 years, 2 years ago, that the challenges in packaging we're going to look a lot more like what was going on in the front end, and they ask us to make some of the platforms that we use for the front end available for packaging.

是的。很好的問題。我的意思是，我認為已經發生了兩件事，我們在一年半到兩年前就從主要客戶那裡得到了早期跡象，即我們在封裝方面面臨的挑戰將更像前端所發生的事情，他們要求我們將一些用於前端的平台用於打包。

And so back to the question, what is this? A lot of it's inspection and metrology, derivatives of the projects and programs that we have and have many years of experience with. There's clearly some plasma dicing so SPTS as part of that overall solution. But there's no question that our customers are driving.

那麼回到問題：這是什麼？其中許多都是檢查和計量，源自於我們已有並有多年經驗的專案和計劃。顯然存在一些等離子切割，因此 SPTS 是整體解決方案的一部分。但毫無疑問，我們的客戶正在推動我們。

As you know, it's a very expensive when you have these high-end chips along with this complex packaging in the HBM, the risks are very high if there's a yield loss. So there's more inspection opportunity there, and we feel great about the continued growth as we go forward. Right now, it's mostly 2.5, but -- 2.5 HBM, but we see it's going to go forward. And our customers are -- this is an area that's moving very quickly. And because they need solutions, they're very focused on making sure we understand them as we go forward.

如您所知，當您擁有這些高階晶片以及 HBM 中的複雜封裝時，成本非常高，如果出現產量損失，風險會非常高。因此那裡有更多的檢查機會，我們對未來持續的成長感到非常高興。目前，它主要是 2.5，但 - 2.5 HBM，但我們看到它會向前發展。我們的客戶—這個領域發展非常迅速。因為他們需要解決方案，所以他們非常注重確保我們在前進的過程中理解他們。

Harlan, it's Bren here. It's about 65%, 70% semi PC versus process.

哈蘭，我是布倫。其中約 65% 到 70% 為半 PC 與流程。

CJ Muse, Cantor Fitzgerald.

CJ Muse、康托·菲茨傑拉德。

I just wanted to dig a little bit deeper in your outperformance relative to WFE. Within that, you're including that $500 million China hit. And so would love to hear, I guess, beyond the rising process control intensity at 3-nanometer and HBM, are there other drivers? Are those the two principal ones we should be thinking about?

我只是想更深入地了解您相對於 WFE 的優異表現。其中還包括了 5 億美元的中國損失。所以我很想聽聽，除了 3 奈米和 HBM 製程控制強度的提升之外，還有其他驅動因素嗎？這是我們應該考慮的兩個主要問題嗎？

Yes, CJ, those are the two principal ones we should be thinking about.

是的，CJ，這是我們應該考慮的兩個主要問題。

Yeah, those are the two principal ones. Well, it's also the sure game.

是的，這兩個是主要的。嗯，這也是穩操勝券的比賽。

Yeah. You've got higher intensity at the nodes. And to Rick's point, we feel very good about some product momentum in a number of our markets. So that's -- and then I think finally, as you look at that and you look at what's driving growth within Process Control, you've got an acceleration in certain products where we have a strong market position. So they're influencing the growth rate, obviously, more relative to the overall.

是的。節點處的強度更高。正如里克所說，我們對一些市場上的一些產品發展勢頭感到非常滿意。所以這就是——然後我認為最終當你看到這一點並看到推動過程控制增長的因素時，你會發現我們在具有強大市場地位的某些產品上實現了加速增長。因此，它們對成長率的影響顯然更多是相對於整體而言。

So that also drives an improving its share. Optical pattern inspection being one of those areas.

這也推動了其份額的提高。光學圖案檢查就是其中一個領域。

Well, and we saw some improvement in some of the work in medical. And including the Gen 5, CJ, for or print check, which obviously shows up in the optical, but it's part of that solution. So look, we're feeling pretty good and there was some investment made by our customers to support prior nodes once they realized that there was still a yield opportunity there.

嗯，我們看到醫療領域的一些工作取得了一些進展。包括 Gen 5、CJ 或列印檢查，這顯然出現在光學中，但它是解決方案的一部分。所以看，我們感覺非常好，一旦我們的客戶意識到那裡仍然有收益機會，他們就會進行一些投資來支持以前的節點。

And just to follow up on that, Rick, you talked about share Can you elaborate on that? And then my second question would be on Service. You talked about hitting kind of the long-term growth rate over time, but would be curious given kind of the China impact, how you're thinking about growth for overall Servicing calendar '25?

為了跟進這個問題，里克，你談到了分享，你能詳細說明一下嗎？我的第二個問題是關於服務的。您談到了隨著時間的推移達到長期成長率，但考慮到中國的影響，您如何看待 25 年整體服務日曆的成長？

Sure. So on the share side, I think there's a couple of areas that might be more obvious than others. One is optical simply because optical grew disproportionately perhaps in the rest of the market, and we have a large share there, so that creates a greater overall position there. We had some really strong momentum in e-beam, but then the other area where we really saw some strong performance was in packaging.

當然。因此，在共享方面，我認為有幾個領域可能比其他領域更明顯。一個是光學，因為光學可能在其他市場中成長不成比例，而我們在那裡佔有很大的份額，因此在那裡創造了更大的整體地位。我們在電子束領域有著非常強勁的發展勢頭，但我們真正看到強勁表現的另一個領域是封裝領域。

And so that's the one where the teams have really done a great job focusing in the last couple of years, and we've been able to see continued momentum there. So we feel pretty good about it.

所以這就是團隊在過去幾年中確實做得非常出色的地方，而且我們已經能夠看到持續的勢頭。因此我們對此感覺很好。

Obviously, the numbers for the year aren't going to come out, but we had gained a lot of share in the last couple of years. And the question was would there be any retrenchment, we feel pretty good about where we are.

顯然，今年的數字不會公佈，但在過去幾年裡，我們的份額增加了很大一部分。問題是是否會進行裁員，我們對目前的狀況感覺相當良好。

For service, anytime you lose access to a fab, you had the immediate headwind that you can't get access to that equipment. So as I look at growth this year, I think growth is probably in the high single digits for Service, which is below the long-term model. We outperformed the long-term model by a little bit in 2024.

對於服務，只要您無法進入晶圓廠，您就會立即面臨無法存取該設備的困難。因此，當我回顧今年的成長時，我認為服務業的成長率可能處於高個位數，低於長期模型。2024 年我們的表現略優於長期模型。

Over time though, it's generally our view at least in terms of how we run the company as we think about the efficiency of the market that, that if you have fabs that are inhibited from being able to supply that capacity has to get added somewhere else. And so that would create an opportunity for us to make some of that up over the very long term.

但隨著時間的推移，我們通常認為，至少就我們如何經營公司而言，當我們考慮市場效率時，如果你的晶圓廠無法供應這些產能，那麼就必須在其他地方增加產能。因此這將為我們創造一個在長期內彌補一些不足的機會。

And obviously, that would mean that you would end up with whatever was spent before would have to be replaced somewhere else. So I think over the long run, we feel pretty good about the growth trajectory in our long-term model. But in the short run, it does affect, obviously, your ability to get at that capacity, which puts pressure on the growth rate and also put some pressure on our ability to move resources around, and so we'll have to deal with some inefficiency as we've staffed up to support those fabs and now we have to move those folks to support other customers.

顯然，這意味著你最終必須在其他地方彌補之前所花掉的一切。因此我認為從長遠來看，我們對長期模型中的成長軌跡感到相當滿意。但在短期內，這顯然會影響你獲得產能的能力，從而給增長率帶來壓力，也給我們調動資源的能力帶來壓力，所以我們必須處理一些效率低下，因為我們已經配備了人員來支持這些晶圓廠，但現在我們必須將這些人調去支援其他客戶。

So there's a few moving parts. But in the long run, we feel pretty good about the trajectory given the higher value of offerings, what we're seeing in terms of pricing as it relates to new products, the opportunities in packaging for incremental service. So I think that the drivers for Service are all pretty compelling.

因此，有幾個活動部件。但從長遠來看，考慮到產品價值的提高、新產品的定價以及增量服務包裝方面的機會，我們對發展軌跡感到十分樂觀。所以我認為服務的驅動因素都非常引人注目。

Obviously, the installed base is growing, lifetimes are increasing. So in the long run, we feel pretty good about the long-term target.

顯然，安裝基數正在成長，使用壽命也在增加。因此從長遠來看，我們對長期目標感到十分滿意。

Joe Quatrochi, Wells Fargo.

富國銀行的喬‧夸特羅奇 (Joe Quatrochi)。

Just a follow-up on the services impact from China. Just given the fact that most of your services is highly recurring. Do we just take that, I guess, quarterly kind of run rate impact all in the March quarter and then grow from there? Or is there further kind of headwinds to think about in the out quarters?

這只是對來自中國的服務影響的後續關注。鑑於你們的大部分服務都是高度重複性的。我想，我們是否應該將季度運行率影響全部集中在 3 月份季度，然後從那裡開始成長？或者外部方面還存在哪些進一步的阻力需要考慮？

Yes, I think that's the way to think about it because you lose what you would have gotten at those fabs and then it grows from there. So I think that's a reasonable way to think about it.

是的，我認為這是思考問題的方式，因為你會失去你本可以在這些晶圓廠得到的東西，然後它會從那裡增長。所以我認為這是合理的思考方式。

Okay. Perfect. And then just thinking about capital intensity for process control on the DRAM side, can you talk about just the difference in HBM process control intensity relative to conventional DRAM, just how to think about that adoption? I know obviously, EUV being adopted across the board is helpful for you guys, too.

好的。完美的。然後，僅考慮 DRAM 方面製程控制的資本強度，您能否談談 HBM 製程控制強度相對於傳統 DRAM 的差異，以及如何看待這種採用？我知道，顯然，全面採用 EUV 對你們也有幫助。

Yeah. So as I said earlier, right, with an HBM device, you've got a few things that are happening. You've had bigger die because you have to drill the PSVs. They're bigger, so you have less redundancy, which historically has been pretty significant DRAM. And so that's been a headwind to process control intensity.

是的。正如我之前所說的，使用 HBM 設備，就會發生一些事情。因為您必須鑽孔 PSV，所以您的模具會更大。它們更大，因此冗餘度更低，從歷史上看，DRAM 一直非常重要。這對過程控制強度來說是一個阻力。

The logic circuitry is more complex. The reliability on all the guy in the stack is higher. So for all those reasons, it's very good for process control intensity.

邏輯電路更加複雜。堆疊中所有人的可靠性都更高。因此，由於所有這些原因，它對於過程控制強度非常有益。

I think overall for DRAM, it's moving the needle probably somewhere from, we'll call it, the 9% to 10% range where we've been historically as a percent of WFE that it probably moves up a good 100 to 150 basis points from there. Now obviously, mix will affect that. Most of the focus is on HBM in terms of new requirements. So mix dynamics could affect that. But we feel pretty good about these dynamics as they affect and drive the DRAM market.

我認為總體而言，對於 DRAM 而言，它可能正在從我們稱之為的 9% 到 10% 的範圍內移動，從歷史上看，WFE 的百分比可能會上升 100 到 150 個基點從那裡。顯然，混合會影響這一點。就新要求而言，大部分焦點都集中在 HBM 上。因此混合動態可能會影響這一點。但我們對這些動態感到十分滿意，因為它們影響並推動 DRAM 市場。

And it's most pronounced in the latest technology nodes, and that's where we're seeing it more. And so it's going to take a little bit of a time for us to really figure out what that overall looks like. I think by Investor Day, we should be in a pretty good position to talk about it on a longer-term basis.

這種現像在最新的技術節點上表現得最為明顯，我們也在那裡看到更多這種現象。因此，我們需要花一點時間來真正弄清楚其整體情況。我認為，到投資者日時，我們應該能夠以相當好的狀態來長期討論這個問題。

Timothy Arcuri, UBS.

瑞銀的提摩西·阿庫裡。

Bren, can you give us RPO? It was supposed to be up. Can you give us the number?

布倫，你能給我們 RPO 嗎？它本該升起來的。能告訴我們電話號碼嗎？

Yes. So RPO was down about $900 million. About half of that was related to debookings we took due to December 2 regulations. So about half of it related to that. And then the other half, shipment levels were higher. So that's how played out it before.

是的。因此RPO下降了約9億美元。其中約有一半與我們根據 12 月 2 日的規定取消預訂有關。其中大約有一半與此有關。而另一半的出貨量則較高。原來之前的情況就是這樣的。

Got it. Okay. And then Process Control Systems, you said pretty stable from here, but what about EPC. It was up a lot this quarter. It grew a lot in -- well, it didn't grow that much, but it grew a lot in Q4.

知道了。好的。然後是製程控制系統，您說從現在開始相當穩定，但 EPC 呢？本季度，該數字大幅上漲。它增長了很多——好吧，它沒有增長那麼多，但在第四季度增長了很多。

So how to think about it for this year? Can it grow 10%, perhaps low double digits this year?

那麼今年該如何考慮呢？今年它能成長 10% 嗎，也許是低兩位數？

Yes. I think overall, EPC is probably going to be about mid-single digits. You have to remember that what shows up in EPC is flat panel business. And so at the end of this quarter, we will be done shipping systems for flat panel after we announced end of manufacturing 12 months ago. So you have the flat panel revenue coming out. And so obviously, that in the niche year, it affects the overall growth rate of the EPC businesses as we report those segments.

是的。我認為總體而言，EPC 可能約為中等個位數。你必須記住，EPC中出現的是平板業務。因此，在本季末，我們將完成平板顯示器的出貨系統，而我們在 12 個月前宣布了停止生產。因此你就獲得了平板顯示器的收入。因此顯然，在利基年份，當我們報告這些細分市場時，它會影響 EPC 業務的整體成長率。

So overall, we feel pretty good if you look at SPTS growth, especially semiconductor, mostly driven by advanced packaging growth year-to-year. ICOS component inspection, again, a packaging centric business is also growing. PCB businesses are more tied to mobility and capacity, so less growth in those areas. And then, of course, you got the offset from losing the FPD piece.

因此，總體而言，如果看一下 SPTS 的成長，我們感覺相當不錯，尤其是半導體的成長，這主要受到先進封裝逐年成長的推動。ICOS 組件檢測，再次，以包裝為中心的業務也在成長。PCB業務與流動性和產能的連結更為緊密，因此這些領域的成長較少。然後，當然，您因丟失 FPD 件而得到了補償。

Now losing the FPD piece does enhance the margin ratios, right, gross margins probably 20 bps higher. I think operating margins are probably 30 bps higher because the revenue mix is a little bit richer. And certainly, that's factored into how we guided gross margins as we look at next year -- or look at this year, 2025.

現在失去 FPD 部分確實會提高利潤率，對吧，毛利率可能高出 20 個基點。我認為營業利潤率可能會高出 30 個基點，因為收入組合更豐富了。當然，這已經考慮到了我們在展望明年（或今年，也就是 2025 年）時如何指導毛利率。

Krish Sankar, TD Cowen.

克里什·桑卡爾（Krish Sankar），TD Cowen。

And Rick and Bren, thanks for quantifying the $500 million plus or minus impact from export control. You also spoke about China WFE dilution. I'm just kind of curious, if you lay in the digestion from China, how to think about your decline in China sales year-over-year on top of export controls in '25 versus '24?

謝謝瑞克 (Rick) 和布倫 (Bren)，你們量化了出口管製造成的 5 億美元左右的影響。您還談到了中國 WFE 的稀釋。我只是有點好奇，如果您關注中國的消化情況，那麼如何看待 2025 年與 2024 年相比，在出口管制的基礎上，中國銷售額同比下降的情況？

Yes. I'll try to help with that. I mean, obviously, we'll have to see how the year plays out. But if you look at how we finished the year, right, this last quarter was 36%. We finished the year at 41% of our business in China.

是的。我將盡力幫忙。我的意思是，顯然我們必須看看今年的情況如何。但如果你看看我們今年年底的表現，你會發現，上個季度的成長率為 36%。今年，我們在中國的業務佔比達到 41%。

As we look at 2025, I think that percentage drops to about 29% plus or minus, a point or two as we go forward here. And so when you do the math on that, assuming stability that we articulated about our top line as we think about where we are right now, that translates into the overall China business down somewhere around 20% or so.

展望 2025 年，我認為這一比例會下降到 29% 左右，上下浮動一到兩個百分點。因此，當你對此進行計算時，假設我們闡明的營收穩定，考慮到我們目前的狀況，這意味著整個中國業務將下降約 20％ 左右。

Got it, got it. That's very helpful. And then another question is going to China, I apologize to this. But when you look at your numbers compared to some peers, over the last two quarters, your China sales have been more resilient compared to your peers.

知道了，知道了。這非常有幫助。然後另一個問題是針對中國的，對此我深感抱歉。但當我們將您的數據與一些同業進行比較時，會發現過去兩個季度，您在中國的銷售表現比同業更具彈性。

Is this due to the wafer business? Or is it because China is building domestic critical capacity? Like what's happening there that kind of makes you relatively more resilient to your peers?

這是因為晶圓業務嗎？還是因為中國在建設國內關鍵產能？就像那裡發生了什麼事，讓你相對於同儕而言更具韌性？

Yes. I think the easiest way to think about it is you have to remember that KLA is really about helping customers qualify process and speed time to results, yield learning and so on. So as a result, you end up with -- particularly with greenfield fabs, a higher level of adoption is that opening and more continuous investment at lower levels.

是的。我認為最簡單的思考方式是，你必須記住 KLA 的真正目的是幫助客戶確認流程、加快獲得結果的時間、獲得學習成果等等。因此，你最終會得到 — — 特別是對於新建晶圓廠而言，更高水準的採用意味著在較低水準上開放和進行更持續的投資。

So when a customer goes to add a significant amount of capacity, obviously, more capacity-centric peers are going to participate, but then they'll get it in that quarter, and then it will fall off where ours tends to be a little bit more consistent in terms of the investment profile. And so it also, I think, tends to hold up because I think the value of process control, given the maturity of those operations, is pretty high.

因此，當客戶要增加大量容量時，顯然會有更多以容量為中心的同行參與，但隨後他們會在那個季度獲得它，然後它就會下降，而我們往往會下降一點投資組合狀況更加一致。因此，我認為這種觀點往往成立，因為考慮到這些操作的成熟度，我認為製程控制的價值相當高。

Tom O'Malley, Barclays.

巴克萊銀行的湯姆·奧馬利。

Mine is a little short-term oriented, so forgive me here. But the last two things we've seen you and a competitor kind of talk about better NAND pretty significantly into the March quarter. I was just hoping you can give us just a little more detail. It didn't sound like from a sequential basis, you had really called that out. I don't think a ton of detail that would make sense that we see coming there.

我有點短期導向，所以請原諒我。但是，我們看到您和競爭對手在三月季度就更好的 NAND 進行了相當重要的討論。我只是希望您能給我們更多細節。從順序來看，這聽起來不像是你真正提到了這一點。我認為我們看到的大量細節都毫無意義。

But just maybe describe what's happening there? Is that coming from a single customer? Is that coming across multiple customers? I totally understand it's a much lower base from these guys. But would love to try to figure out where the strength is coming from just on a sequential basis into March?

但也許可以描述一下那裡發生了什麼事？這是來自單一客戶嗎？這是否遇到多個客戶？我完全理解這些人的基礎非常低。但是，很想弄清楚，三月的連續成長力量來自何處？

The strength at pretty low levels. It's pretty broad-based. We have seen the NAND business tick up, right, over the course of the '24 and into '25. We expect a little bit more improvement there. I think overall for the industry off of a very low level.

實力還處於相當低的水平。它的範圍相當廣泛。我們看到 NAND 業務在 2024 年至 2025 年期間呈現上升趨勢。我們期望那裡會有更多的改善。我認為整個行業的水平總體來說還處於很低的水平。

There's likely to be some WFE growth there, but it's not significant. And as a percent, it's bigger, obviously, given the level of WFE at that presently. But I would expect to see that improve a little bit moving forward, but not a lot 2025.

那裡可能會有一些 WFE 增長，但並不顯著。並且，從百分比來看，考慮到目前的 WFE 水平，這個數字顯然更大。但我希望看到這種情況在未來有所改善，但到 2025 年不會有太大改善。

Got you. And then on the DRAM side, clearly, there's debate in the broader market. You guys called out AI in some of your prepared remarks, but it seems like there's some share jockeying that's currently taking place. It sounded more positive for the year, your kind of view on the DRAM side. Any commentary just when you think about 6 months ago is when you talk to your customers.

明白了。而在 DRAM 方面，顯然，更廣泛的市場存在爭議。你們在一些準備好的發言中提到了人工智慧，但目前似乎正在進行一些份額爭奪。您對 DRAM 方面的看法對於今年來說聽起來更加積極。當您與客戶交談時，請回想 6 個月前的任何評論。

Obviously, people were putting in capacity for kind of all of 2025. Have you seen incremental spend there in the short term or rush orders to try to catch up by customers? Anything on that would be helpful.

顯然，人們正在為 2025 年全年投入產能。您是否看到過那裡短期內的增量支出或緊急訂單以試圖趕上客戶？任何與此相關的事情都會有幫助。

Well, I think our customers -- it's Rick, our customers certainly set out their plans looking out for the year. So there's been no real short-term change. I do think the strength in terms of what they're seeing in demand and support of the AI infrastructure continues to grow, and we see momentum there.

嗯，我認為我們的客戶——是里克，我們的客戶肯定制定了他們今年的計劃。因此短期內沒有發生真正的變化。我確實認為，他們對人工智慧基礎設施的需求和支援力度持續成長，而且我們看到了發展勢頭。

So we are definitely in conversations and a lot of them about slot availability. Remember, we still have many products or a few critical products that are supply constrained. So we're in conversations about that.

因此，我們確實在進行大量關於時段可用性的討論。請記住，我們仍有許多產品或一些關鍵產品的供應受到限制。我們正在討論這個問題。

So we feel pretty good about the demand, especially at the leading edge. And the dynamics around advanced DRAM are playing to our strength because of the challenges, both the value of those devices, but also the yield challenges. And as we mentioned before, the die sizes are smaller, less redundancy and it's looking more like higher process control intensity as we talked about.

因此，我們對需求感到非常樂觀，特別是在前沿領域。先進 DRAM 周圍的動態正在發揮我們的優勢，因為這些挑戰既包括這些設備的價值，也包括產量的挑戰。正如我們之前提到的，晶片尺寸更小，冗餘更少，而且看起來更像是我們所討論的更高的製程控制強度。

Chris Caso, Wolfe Research.

克里斯‧卡索（Chris Caso），沃爾夫研究公司。

Just a follow-up question with regard to the China impact. And you've given some color on what you expect that for the year. From a quarterly basis, is there any sort of incremental headwind or benefit as we go into the second half? I know that you talked about some of the mitigations and licenses, which are taking some time. But I guess how do we think about this as we go sequentially through the year?

這只是一個有關中國影響的後續問題。您已經對今年的期望做出了一些說明。從季度角度來看，進入下半年是否會出現任何增量阻力或收益？我知道您談到了一些緩解措施和許可，這需要一些時間。但我想，隨著時間的流逝，我們該如何思考這個問題？

Yes, we're pretty cautious with it overall. We'll see how it plays out. As I said in the prepared remarks in terms of licenses that could mitigate the impact. But when we look at it over the course of the year or what we expected, it was pretty consistent across the year. So it wasn't -- maybe again, that could be the nature of how customers buy process control versus other types of products. But it was pretty -- half to half was more or less pretty consistent.

是的，我們總體上對此非常謹慎。我們將看看結果如何。正如我在準備好的演講中所說的那樣，許可證可以減輕影響。但當我們回顧全年的情況或觀察我們的預期時，會發現全年的情況都相當一致。所以事實並非如此——也許，這可能是客戶購買過程控製而非其他類型產品的本質。但它相當——一半對一半或多或少相當一致。

Okay. Got it. And just a follow-up on gross margins. Again, you're kind of starting out with 62, you're guiding to 62 for the full year. So sort of assuming that remains stable as you go through the year.

好的。知道了。這只是對毛利率的後續關注。再說一次，您從 62 開始，預計全年將達到 62。所以可以假設它在一年中會保持穩定。

And I guess at what point with regard to some of the operating leverage that you typically get with the fall-through, what's kind of the starting point from that, that we can to see some of the leverage kick in as revenue starts to grow?

我想，關於您通常會因失敗而獲得的一些經營槓桿，從哪個起點開始，我們才能看到隨著收入開始成長，一些槓桿開始發揮作用？

Yes. Look, we have mix issues that generally are the biggest impacting item to our overall gross margins more so than customers or segments. But I would expect as we start to see overall revenue accelerate, we'll start to see the kind of leverage that we've seen historically.

是的。你看，我們有混合問題，這些問題通常是對我們的整體毛利率影響最大的項目，而不是客戶或部門。但我預計，隨著我們開始看到整體收入加速成長，我們將開始看到歷史上所見過的那種槓桿作用。

So I said 62%, plus or minus about 50 basis points the reason. I think some of that is predicated on what happens moving forward. We do have -- depending on the mix, right? we do have markets like the packaging market, which carry a lower gross margin given the complexity of the tools than some of our higher-end systems, but obviously, the gross margin dollars are quite significant. And the relevancy to growth in KLA is significant.

所以我說是 62%，原因大概是上下浮動 50 個基點。我認為其中一些取決於未來發生的事情。我們確實有——取決於組合，對嗎？我們確實有像封裝市場這樣的市場，由於工具的複雜性，其毛利率低於我們的一些高端系統，但顯然，毛利率相當可觀。且與 KLA 成長的相關性非常顯著。

So we're pleased with that. But I think as we move forward, I think you're likely to see us continue in that 60% to 65% range as we as we accelerate revenue over time.

我們對此感到很高興。但我認為，隨著我們繼續前進，隨著我們的收入不斷增加，您可能會看到我們繼續保持 60% 至 65% 的成長率。

And as we talked about in our 2022 plan, we saw gross margins were around 63% or so, obviously predicated on a volume level of about $3.5 billion. So that gives you a sense of kind of where we're at from here to there moving forward.

正如我們在 2022 年計劃中所討論的那樣，我們看到毛利率約為 63% 左右，這顯然是以約 35 億美元的銷量水準為前提的。這樣您就可以了解我們從現在到未來的位置。

And I feel pretty good about our ability to achieve that, given the investments we've made that are still -- I think we're in a good position to deliver against that. I don't think we have to go and make incremental investments in terms of the capacity -- the hard asset capacity we have to execute to those business levels.

考慮到我們目前所做的投資，我對我們實現這一目標的能力感到非常滿意——我認為我們完全有能力實現這一目標。我認為，我們不必在產能方面進行增量投資——我們必須將硬資產產能提高到這些業務水平。

Srini Pajjuri, Raymond James.

Srini Pajjuri，雷蒙德‧詹姆斯。

One short-term question on your March quarter guidance. Just the Foundry/Logic, I think you're guiding for 73% of the mix to be Foundry/Logic. That is, I think, implies at least in a mid-single-digit type decline. We haven't seen a decline in that business in a while. And I'm just trying to understand how that reconciles with your comment about N2 demand being strong in the short term?

關於您三月季度指導的一個短期問題。僅就代工廠/邏輯而言，我認為您指導的 73% 的混合是代工廠/邏輯。也就是說，我認為，至少意味著中等個位數的下降。我們已經有一段時間沒有看到該業務下滑了。我只是想了解這與您對短期內 N2 需求強勁的評論如何相符？

I'm taking a look at it. I don't think it -- it doesn't look like it changes all that much. So I think given that the overall revenue guidance is what it was, I think for semi PC systems, I mean, we'll see how the quarter ends up. And we do have business that is infrastructure business, for example, that doesn't show up in those percentages. So when I look at the businesses that the semiconductor customers, it's pretty consistent.

我正在看一下。我不這麼認為——它看起來並沒有太大的變化。因此，我認為，鑑於整體收入預期，對於半 PC 系統而言，我們將看看本季的結果如何。例如，我們確實有一些基礎設施業務，但這並未反映在這些百分比上。因此，當我觀察半導體客戶的業務時，發現它們相當一致。

So I don't think it will change a little bit. But as we talked about, I think Memory overall is a higher percentage of the mix in 2025 than it was in 2024.

所以我不認為它會發生任何改變。但正如我們所討論的，我認為 2025 年記憶體總體佔比將高於 2024 年。

There were some other customers non-N2, N3 that showed up in December that aren't showing up.

12 月出現過一些其他非 N2、N3 客戶，但現在卻沒有出現。

Okay, okay. That makes sense. And then, I guess, as we go through the year, obviously, N2 is going to be relatively strong. Do you still have, I guess, material contribution?

好的，好的。這很有道理。然後，我想，隨著這一年的發展，N2 顯然會變得相對強勁。我想你還有物質貢獻嗎？

Are you still expecting material contribution from N3? Or is it at a minimal level? And then, I guess, just a follow-up to that, how does the, I guess, PC intensity change as we go from N3 to end N2?

您還期待 N3 能做出實質貢獻嗎？或者說還處於最低水平？然後，我想，只是對此的後續問題，當我們從 N3 到 N2 結束時，PC 強度會發生怎樣的變化？

Yeah. So most of the focus in terms of new investment is on 2-nanometer. There still is some incremental investment that's happening in N3, but the vast majority of it is 2-nanometer centric. Obviously, there's packaging investment that's also happening. And I mentioned it earlier (inaudible) been over the last couple of quarters or so, I said that at N3 versus N2.

是的。因此，新投資的大部分重點都集中在 2 奈米上。N3 領域仍有一些增量投資，但絕大多數都以 2 奈米為中心。顯然，包裝投資也正在進行中。我之前提到過（聽不清楚），在過去幾個季度左右的時間裡，我說過 N3 與 N2 相比。

At N2, we thought we were about 75 basis points higher in terms of KLA share WFE. I think that we're likely higher than that. It's probably 90-ish to -- 90-ish basis points, maybe 100 basis points. So trending in the right direction for sure.

在 N2，我們認為 KLA 股份 WFE 大約高出 75 個基點。我認為我們的數字可能會更高。可能是 90 個基點左右，也可能是 100 個基點。因此趨勢肯定是朝著正確的方向發展的。

Brian Chin, Stifel.

布萊恩·欽（Brian Chin），Stifel。

I'll ask a few questions. Yes, I was just curious, in terms of Logic/Foundry chip makers that are at the leading edge, but maybe not expanding capacity aggressively. Can you comment on the magnitude of residual spending you still see with them and technology development? Obviously, you're able to offset that or any drag there and outgrowing WFE this year. But just curious if you had any sort of commentary around that.

我會問幾個問題。是的，我只是很好奇，就處於領先地位的 Logic/Foundry 晶片製造商而言，他們可能沒有積極擴大產能。您能否評論一下您在他們和技術開發方面看到的剩餘支出的規模？顯然，你能夠抵消這一點或任何拖累，並在今年超越 WFE。但我只是好奇您是否對此有任何評論。

Are you saying -- I'm sorry, you said one that aren't at the leading edge?

您是說—抱歉，您說的是非處於領先地位的嗎？

At the leading edge, but not expanding capacity aggressively. That's kind of one guy doing that. But in terms of the other ones, you're sort of on the pace or maybe -- on the pace, but just not billing out aggressively. Maybe some sense of the spending engagement you still have with them?

處於領先地位，但沒有積極擴大產能。有個人就是這樣做的。但就其他方面而言，你的步伐還挺快的，或者可能--只是沒有積極地收費。也許您對他們之間仍然保持著某種消費關係？

So let's say, we derisked that in our '25 plan.

因此，我們可以說，我們在 25 計劃中降低了這一風險。

Okay. Fair enough. And then maybe just carrying for that last question, the process control intensity going from 2-nanometer around the (inaudible) because I think they're kind of meant to be somewhat closely coupled to some degree.

好的。很公平。然後也許只是繼續最後一個問題，過程控制強度從 2 奈米左右開始（聽不清楚），因為我認為它們在某種程度上應該是緊密耦合的。

Yes, we're a little early on that one. So I'd like to -- before we started making comments actually shipping to support that activity in a way to actually model it.

是的，我們對此有點早。因此，在我們開始發表評論之前，我想先實際發貨以某種方式支持該活動，並對其進行實際建模。

And one of the things, obviously, we've seen over the course of the last several nodes for intensity reasons, but also for share is that because of the design start environment, limiting reuse, customers are managing a much more dynamic design environment, you now have more designs that are driving the leading edge ramps, all these things have been positive factors, and then there's a share element as well.

顯然，在過去幾個節點的過程中，我們已經看到，出於強度的原因，也出於份額的原因，由於設計啟動環境的限制，客戶正在管理一個更加動態的設計環境，現在有更多設計正在推動前沿技術的發展，所有這些都是積極因素，而且還有共享因素。

This fundamentals, I think, shift moving forward in the composition of semiconductor revenue to larger, higher-value die with defect density is very problematic, I think plays to growing opportunities for process control. We have to execute on our programs to be able to deliver the right solutions for customers to solve their problems and solve the right problems to scale the production. But I think if we execute our own business, it does create an opportunity for us to see continued tailwinds in this area.

我認為，這種基本面在半導體收入構成中向更大、更高價值且缺陷密度的晶片的轉變是非常成問題的，我認為這為製程控制帶來了越來越多的機會。我們必須執行我們的計劃，以便能夠為客戶提供正確的解決方案來解決他們的問題，並解決正確的問題以擴大生產規模。但我認為，如果我們開展自己的業務，確實會為我們創造機會，讓我們看到該領域持續的順風。

Yes. And let me give a little more perspective, too, because we've actually -- usually, if the spending is done in a node, that process control intensity is kind of set. But what we've seen happen is when we have new solutions that find new defect problems that are yield impacting, we've seen the backporting of that.

是的。讓我也給出更多的觀點，因為實際上 - 通常，如果支出是在一個節點上完成的，那麼這個過程控制強度就已經確定了。但我們見到的情況是，當我們有了新的解決方案，發現了影響良率的新缺陷問題時，我們就看到了這個問題的反向移植。

So in other words, you might see some systems going into prior nodes, which actually drives those intensities up for the prior node, which is the new baseline to go forward. So we feel -- we think part of the outperform is the fact that we actually have more solutions that solve the problems.

換句話說，您可能會看到一些系統進入先前的節點，這實際上會提高先前節點的強度，這是前進的新基準。因此我們認為——我們認為表現優異的部分原因是我們實際上有更多解決問題的解決方案。

We've always had more opportunities than we've had answers for in terms of customers trying to figure out how to learn quicker and adopt new technologies, but our technologies are really coming together in a way that we think there's -- it's both share, but also it drives adoption simply because we're solving more problems.

在客戶試圖弄清楚如何更快地學習和採用新技術方面，我們一直擁有的機會比我們得到的答案要多，但我們的技術確實以我們認為的方式結合在一起——這是共享的，而且它也推動了採用，因為我們正在解決更多問題。

So when we look at what we're seeing for N2, we feel pretty good about the potential to help our customers ramp those nodes, and that will be a basis on which to build going forward. For example, a lot of people didn't model early on the verification on wafer, the print check that we're using for Gen 5. That's essentially a new application. Once people value that, then they might even go back and back forth some of that capability when there's yield opportunity.

因此，當我們回顧 N2 的前景時，我們對其幫助客戶提升這些節點的潛力感到非常樂觀，這也將成為我們未來發展的基礎。例如，很多人沒有對我們用於 Gen 5 的晶圓驗證、列印檢查進行早期建模。這本質上是一個新的應用程式。一旦人們重視這一點，那麼當有收益機會時，他們甚至可能會反覆使用這種能力。

So we feel pretty good about where we are in terms of driving overall intensity, and that will be part of the message we share at our Investor Day is how we see that going forward, which will include the node you talked about.

因此，就推動整體強度而言，我們對自己目前所處的位置感到非常滿意，這也將成為我們在投資者日分享的訊息的一部分，即我們如何看待未來，其中將包括您談到的節點。

Charles Shi, Needham & Company.

查爾斯·施（Charles Shi），Needham & Company。

So I think that you guys don't want to call out the direction for the second half in terms of the growth relative to the first half of the year. But it sounds like the base case assumption from you guys is you're probably going to be around that $3 billion per quarter level, maybe throughout the year. Maybe some of that is contingent upon whether you can get some expo licenses for that $500 million impact from the latest export control.

因此，我認為你們不想根據相對於上半年的成長來預測下半年的方向。但聽起來你們的基本假設是每季的銷售額可能會達到 30 億美元左右，甚至全年也是如此。也許這在某種程度上取決於你是否能獲得一些博覽會許可證，以彌補最新出口管制帶來的 5 億美元影響。

But is there any other swing factors that you probably don't have a conclusion yet, but that could support some of the second half growth. Is there anything that you haven't mentioned?

但是否還有其他影響因素？還有什麼是您沒提到的嗎？

Well, the license is, as we said earlier, we haven't built that into the plan. And so we'll see how that plays out. And I think the stabilizing around current levels as we look forward, it seems like we're operating around this level. And as the EBIT go beyond the middle of the year, we'll see what happens.

嗯，正如我們之前所說，許可證我們還沒有將其納入計劃。我們將拭目以待事情將如何發展。我認為，展望未來，目前的水準將趨於穩定，看來我們正圍繞著這一水準開展業務。隨著息稅前利潤超過年中，我們將看到會發生什麼。

We mentioned earlier about derisking some opportunities. And so we'll see how those potentially play out around certain customers, but that could be a swing factor as well.

我們之前提到過降低一些機會的風險。因此，我們將觀察這些因素對特定客戶可能產生怎樣的影響，但這也可能是個決定因素。

And I think that back to what we said about certain parts of the market, we've been a little bit more cautious on. We'll see if there's more upside in China that I think we've tried to derisk that relative to the levels of investment we've seen over the last couple of years. But we'll see how that plays out as we move forward. But I think we're -- for now, it feels like around the current levels is the best that I can do from a guidance point of view.

我認為，回到我們所說的有關市場某些部分的問題上，我們已經變得更加謹慎了。我們將拭目以待中國是否還有更多的上漲空間，我認為我們已經試圖降低與過去幾年的投資水準相比的風險。但隨著事態的發展，我們將會看到事情將如何發展。但我認為—就目前而言，從指導的角度來看，感覺當前的水平是我能做的最好的事情。

Maybe a quick follow-up. What's the expectation for China revenue contribution into March quarter?

或許是一個快速的後續行動。您預計三月季度中國的營收貢獻是多少？

It will come down as a percent to be high 20s. We'll see. We'll see what ends up revenue in, right, because you've got different rev-rec policy issues from -- whether it's a new customer and a new fab versus an established customer, so that could either accelerate revenue to revenue shipment or extend it to an acceptance process. So we'll see how things play out. But in general, I would expect it to drop from the 35% level probably into the high-20s, maybe 30% at the highest.

其百分比將下降至 20% 左右。我們將會看到。我們將看到最終的收入情況，因為您面臨不同的收入政策問題 - 無論是新客戶和新工廠還是老客戶，這都可以加速收入到收入出貨量或將其擴展為接受過程。因此我們將觀察事情如何發展。但總的來說，我預計它會從 35% 的水平下降到 20% 左右，最高可能會達到 30%。

(Operator Instructions) Atif Malik, Citi.

(操作員指示) Atif Malik，花旗。

Rick, the question on foundry concentration comes a lot with investors. Obviously, you guys are doing very well with your foundry customer on N3 and all around, and there is a reference in Japan that's kind of ramping this year. How are you guys leaning into the two struggling foundries this year if that was a risk to your business?

里克，投資人經常問到有關代工集中度的問題。顯然，你們與 N3 及週邊的代工客戶合作得非常好，今年在日本的合作也正在增加。如果這對您的業務構成風險，那麼今年您將如何應對這兩家陷入困境的代工廠？

I'm sorry, say again. How are we dealing with...

抱歉，請再說一次。我們該如何處理...

How are you guys like leaning into the two struggling foundries and what impact that could have both this year and out two years in terms of your exposure?

你們如何看待這兩家陷入困境的代工廠？ 這對你們今年和未來兩年的曝光度有何影響？

Well, we obviously work with all our customers. And so if there's a way for us to add value, we're doing that. I think the bulk of the stated CapEx number pretty clearly head towards the direction of the biggest player in the market in terms of investment.

嗯，我們顯然與所有客戶合作。因此，如果我們有辦法增加價值，我們就會這麼做。我認為，所述的資本支出數字中的大部分顯然是朝著市場最大投資參與者的方向發展的。

But the other is we engage. I mean, certainly, everyone that we work with wants to improve their ramp-up time of new technology and improve their yield. And so of course, we're doing that, but that's not where the bulk of the business is today.

但另一方面是我們參與。我的意思是，當然，與我們合作的每個人都希望縮短新技術的啟動時間並提高產量。當然，我們正在這樣做，但這並不是我們目前的主要業務所在。

So I don't see a huge difference in terms of how we're engaging now relative to how we were in the past. It's just -- the dynamics have shifted much more towards the leader who's further ahead now than they've been in quite a while.

因此，我認為我們現在的參與方式與過去並沒有太大的差異。只是——與過去相當長一段時間相比，現在的情況已經更加傾向於領先者。

Fair enough. And Bren, on the $500 million restrictions impact, can you give some color? Were those like trailing-edge logic projects? Or was it in DRAM contribution in those sales?

很公平。布倫，您能具體談談 5 億美元限制的影響嗎？這些是像後緣邏輯項目嗎？或者這些銷售中的 DRAM 貢獻是什麼？

Yes, most of it was logic. Yes, very little. In fact, all of it was logic. They're very little. That was Memory.

是的，大部分都是邏輯。是的，很少。事實上，這一切都是有邏輯的。它們非常少。那就是記憶。

And it appears we have no further questions at this time. I would like to turn the call over to Kevin Kessel for any additional or closing remarks.

目前看來我們沒有其他問題了。我想將電話轉給凱文凱塞爾 (Kevin Kessel)，讓他發表任何補充意見或結束語。

Thank you very much, and thank you, everybody, for your time and your attention. We know how busy today is this weekend. So we appreciate it. We'll be speaking with you all very soon. I'll turn it back now to the operator for any closing instructions.

非常感謝，也感謝大家的時間和關注。我們知道今天這個週末有多忙。所以我們對此表示感謝。我們很快就會和大家交談。我現在將把它轉回給操作員以獲取任何關閉指示。

Thank you. This concludes the KLA Corporation December quarter 2024 earnings call and webcast. Please disconnect your line at this time and have a wonderful day.

謝謝。這就是 KLA Corporation 2024 年 12 月季度收益電話會議和網路廣播的全部內容。請立即斷開您的線路並祝您有美好的一天。