日月光投資控股 (ASX) 2026 Q2 法說會逐字稿

Hello, I am Ken Hsiang, the Head of Investor Relations at ASE Technology Holdings. Welcome to our second-quarter, 2026 earnings release. I'm joined today by Dr. Tien Wu, our COO; and Joseph Tung, our CFO. Thank you for joining us today. Please refer to our Safe Harbor notice on page 2. (Event Instructions)

各位好，我是向坤，日月光投控投資人關係主管。歡迎參加我們 2026 年第二季財報發布會。今天與我一同出席的有我們的營運長吳田玉博士，以及財務長董宏思。感謝各位今天撥冗參與。請參閱簡報第 2 頁的安全港聲明。（活動說明）

I would like to remind everyone that the presentation that follows may contain forward-looking statements. These forward-looking statements are subject to a high degree of risk and our actual results may differ materially. For the purposes of this presentation, dollar figures are generally stated in new Taiwan dollars, unless otherwise indicated.

在此提醒各位，接下來的簡報可能包含前瞻性陳述。這些前瞻性陳述具有高度風險，我們的實際結果可能出現重大差異。就本次簡報而言，除非另有說明，金額一般以新台幣表示。

As a Taiwan-based company, our financial information is presented in accordance with Taiwan IFRS. Results presented using Taiwan IFRS may differ materially from results using other accounting standards, including those separately presented by our subsidiaries. For today's presentation, Dr. Tien Wu will begin with a mid-year business update, I will then walk through the Q2 results, and Joseph will close with our third-quarter outlook.

作為一家以台灣為基地的公司，我們的財務資訊係依台灣 IFRS 編製。以台灣 IFRS 呈現的結果，可能與採用其他會計準則（包括我們子公司另行揭露者）所呈現的結果有重大差異。今天的簡報將由吳田玉博士先進行年中營運更新，接著由我說明第二季業績，最後由 Joseph 以第三季展望作結。

With that, let me hand the presentation over to Dr. Tien Wu.

接下來，我把簡報交給吳田玉博士。

Good afternoon. I would like to give you the first half 2026 recap and also the full-year outlook. For my presentation will be all US dollar terms. Consolidated revenue grew 24% year on year in the first half of 2026, with ATM revenues up 35% year on year for the first half. Leading-edge advanced packaging and overall testing outpaced growth.

各位午安。我想向各位回顧 2026 年上半年表現，並提供全年展望。我今天的簡報將全部以美元為單位。2026 年上半年合併營收年增 24%，其中 ATM 營收上半年年增 35%。先進製程的先進封裝（Leading-edge advanced packaging）與整體測試業務的成長優於整體成長。

For ATM business, we expect to maintain the same growth momentum into the second half. For full-year, LEAP services revenue is tracking ahead of prior guidance of USD3.5 billion, while general segment expected to grow by 30% (sic - see slide presentation, slide 3, "20%") year on year versus previous guidance of 13%. So for the full year, we expect the ATM business revenue to grow by 35%.

就 ATM 業務而言，我們預期在下半年可維持相同的成長動能。以全年來看，LEAP 服務營收進度優於先前指引的 35 億美元；一般業務區塊預期年增 30%（原文如此—請見簡報第 3 頁投影片，「20%」），高於先前指引的 13%。因此，全年我們預期 ATM 業務營收將成長 35%。

Machinery CapEx was USD2.7 billion. Building, facility, automation was $1.4 billion in the first half. Joseph will give you more detail for the full year, stepping up investment R&D, human capital, advanced capacity, and also automation and smart factory infrastructure to support a multi-year growth.

機器設備資本支出為 27 億美元。上半年建築、廠務與自動化支出為 14 億美元。Joseph 將就全年提供更多細節，包括加大對研發、人力資本、先進產能，以及自動化與智慧工廠基礎建設的投資，以支撐多年期成長。

On the second page, I would like to give you some highlight on market dynamics and positioning. There are many moving parts in the market today. You're reading the same newspaper, watching the same news as I do. What I'm trying to present to you is the company view. We will try to present you the logic. Why are we making particular decision in this particular junction of time?

在第二頁，我想就市場動態與我們的定位提供一些重點。目前市場上有許多變動因素。各位看到的報紙、收看的新聞，與我是一樣的。我想向各位呈現的是公司觀點。我們會嘗試把邏輯說清楚。為什麼我們會在這個時間點做出特定決策？

AI enabled new applications with bigger scale and potentials. I'm not going to articulate the detail. If you're interested, we can talk more in the Q&A. That's our current view. I think very few of you will disagree with this statement that AI is a paradigm shift, and we do have potential bigger scale and multiple applications. And that's our current view. In terms of when, how we are monitoring the progress. I believe we are at the beginning of the AI paradigm shift. There will be multiple stages of transition. We can talk more in the Q&A.

AI 正在催生具更大規模與潛力的新應用。我不打算在此詳述細節。若各位有興趣，我們可以在問答環節再深入討論。這是我們目前的看法。我想各位很少會不同意：AI 是一次典範轉移，且確實具備更大規模與多元應用的潛力。這也是我們目前的看法。至於時間點，以及我們如何監測進展，我相信我們正處於 AI 典範轉移的起點。未來將有多個轉換階段。我們可以在問答環節再多談。

The second comment is how we feel. AI demands new hardware that does not exist previously in size, complexity, and integration. You can argue about a competition intensity, the memory, the power, the linkage, the bandwidth. All in all, what we're trying to develop right now is a brand-new platform to support the potential AI applications. The data center, the agentic, and future in the physical AI, humanoid, all of the hardware will be different than what we have been producing so far.

第二點是我們的感受。AI 需要全新的硬體，其在尺寸、複雜度與整合度上，都是過去未曾存在的。你可以就競爭強度、記憶體、功耗、互連、頻寬等面向提出不同看法。總而言之，我們目前要開發的是一個全新的平台，以支援潛在的 AI 應用。資料中心、代理式（agentic）應用，以及未來的實體 AI、類人機器人等，其硬體都將不同於我們迄今所生產的產品。

There is a growing need for industrial power, connectivity, and storage devices because of the AI transition evolution or paradigm shift. The company is seeing all three right now from all of our customers on the multi-year basis. Let me talk about ASE's strategic priorities Again, this is the company view.

隨著 AI 轉型演進或典範轉移，產業對電力元件、連結性與儲存裝置的需求正在增加。公司目前已從所有客戶端看到這三方面的多年期需求趨勢。接下來我談談日月光的策略重點；再次強調，這是公司的觀點。

We have to have a blueprint, a plan, a vision in order for the 100,000 employees to act on. And what I'm trying to present to you is the highlight for that blueprint. Hardware infrastructure is a bottleneck. With AI, the hardware requirement is new, insatiable, and more complicated and more complex. And today, there are very few manufacturers capable of producing hardware.

我們必須有一份藍圖、計畫與願景，讓 10 萬名員工能依此行動。我想向各位呈現的是這份藍圖的重點。硬體基礎建設是一個瓶頸。在 AI 驅動下，硬體需求是全新的、近乎無止盡的，而且更繁複、更高複雜度。而在今天，能夠製造這些硬體的廠商其實非常少。

Therefore, it is the bottleneck today for our capacity, for automation, and more importantly, from innovation perspective, where you can talk a little bit more detail about a panel, about co-op, about glass substrate, about VRM, about silicon photonics. I can go on and on. But all of these are tied to the infrastructure and your capability to ramp with the complexity, integration, and the design blueprint the customer are asking to do. All in all, I call all of this to be hardware infrastructure, and that is the new bottleneck. We have not experienced this for the last 40 years.

因此，這在今天成為我們產能、自動化的瓶頸；更重要的是，從創新角度來看，你可以更深入談到面板、co-op、玻璃基板、VRM、矽光子等議題。我可以一直講下去。但這些都與基礎建設相關，也與你是否具備能力在客戶要求的複雜度、整合度與設計藍圖下快速擴產息息相關。總而言之，我把這些統稱為硬體基礎建設，而這就是新的瓶頸。過去 40 年我們未曾遇過這樣的情況。

Packaging is moving up in system architecture value chain. For me, as a designer, by training, the system architecture always is at the top of the value chain. Packaging is approaching the system architecture value by providing the new complexity integration capability with the variable that I just talked about it.

封裝正在系統架構價值鏈中往上移動。以我受過設計訓練的背景來看，系統架構一向位於價值鏈的頂端。封裝透過我剛才提到的各項變數所帶來的新複雜度與整合能力，正逐步接近系統架構的價值。

Now if you believe hardware is the new bottleneck and the packaging is moving up in the value chain of the system architecture, then ASE has a unique position to support the AI migration, evolution, or paradigm shift and align with all customers' long-term objectives today. ASE's competitive position, I talk about this for many times, the ecosystem position, cluster or Taiwan cluster, scale or the AI data center, Taiwan scale, and other digital scale.

如果你相信硬體是新的瓶頸，而封裝正在系統架構的價值鏈中往上移動，那麼日月光就具備獨特位置，可支援 AI 的遷移、演進或典範轉移，並與所有客戶當前的長期目標一致。日月光的競爭定位，我已多次談過，包括生態系位置、聚落或台灣聚落、規模（例如 AI 資料中心的台灣規模）以及其他數位規模。

The efficiency, I want to single out the pure play. The pure play enable you to have a seamless cooperation with all supply chain players. In future, this could be one of the competitive advantage for ASE as a pure-play OSAT, that we have no conflict with foundry, with no conflict with substrate provider. We have no conflict with anybody. Therefore, we have a good way not only to collaborate with our customers long-term, also collaborate with all of the ecosystem players that will turn critical for the overall complex integrated nature for the AI transitional evolution.

在效率方面，我想特別點出純晶圓代工以外的純專業（pure play）。純專業模式讓你能與供應鏈所有參與者進行無縫合作。未來，這可能成為日月光作為純專業 OSAT 的競爭優勢之一：我們與晶圓代工廠沒有利益衝突，與基板供應商也沒有利益衝突。我們與任何人都沒有衝突。因此，我們不僅能與客戶進行長期合作，也能與所有生態系夥伴協作；在 AI 轉型演進所帶來的高度複雜整合特性下，這些合作將變得至關重要。

The first mover's advantage, I'm very specific about the first mover's advantages: technology, speed, capacity, and most importantly, trust. So everything the company does circle around the long-term business objective as well as the seamless integration with ecosystem player and give you the speed, the efficiency, and earn the customer trust.

先行者優勢方面，我非常具體地指的是：技術、速度、產能，以及最重要的—信任。因此，公司所做的一切都圍繞著長期的業務目標，以及與生態系夥伴的無縫整合，為各位帶來速度與效率，並贏得客戶信任。

So that's the highlight for me for the second half. Okay, thank you.

所以這就是我對下半年的重點。好的，謝謝。

Thank you, Tien. As Dr. Wu highlighted, our businesses performed extremely well throughout the first half of the year. For the quarter, we saw strong growth within both our LEAP and our general businesses. Non-LEAP capacities like wire bond and traditional advanced packaging were also tight. For test, our wafer sort and final test capacities were also running near full.

謝謝你，Tien。如吳博士所強調，我們的各項業務在上半年表現極為出色。就本季而言，我們在 LEAP 與一般業務兩方面都看到強勁成長。非 LEAP 產能，例如打線（wire bond）與傳統先進封裝，也同樣吃緊。在測試方面，我們的晶圓測試（wafer sort）與最終測試（final test）產能也幾乎滿載運轉。

Effectively, outside of equipment lines being placed into service, capacities were generally very close to being full. Our blended utilization rate for the quarter was between 80% to 85%. Our capability for near-term incremental growth is being gated by our abilities to install capital equipment and build out our buildings and facilities.

實際上，除了設備產線陸續投入服務之外，整體產能普遍都非常接近滿載。本季我們的綜合稼動率介於 80% 至 85%。我們短期增量成長的能力，主要受限於我們安裝資本設備以及擴建廠房與設施的能力。

From a financial perspective, Q2 ATM revenues came in ahead of our original expectations driven by higher loading from both our LEAP and General Services. From the profitability perspective, we also saw our gross margin outpacing our original expectations driven by higher loading, resulting in more operating leverage.

從財務角度來看，第二季 ATM 營收在 LEAP 與一般服務載入量（loading）更高的帶動下，優於我們原先的預期。從獲利角度來看，我們也看到毛利率在更高載入量的推動下超越原先預期，進而帶來更高的營運槓桿。

Revenues for our EMS business were generally in line with our expectations. EMS profitability was slightly below our expectations due to product mix and higher component prices. Please turn to page 5, where you will find our second-quarter consolidated results.

我們 EMS 業務的營收大致符合預期。由於產品組合以及零組件價格較高，EMS 獲利略低於預期。請翻到第 5 頁，您將看到我們第二季的合併業績。

For Q2, fully diluted EPS was $4.61, and basic EPS came in at $4.80. We believe our earnings reflect strong core profitability and a few non-operating items we'll outline shortly. Consolidated net revenues reached $191.1 billion, up 10% sequentially and 27% year over year. Gross profit was $40.2 billion with a gross margin of 21.0%, up 1 percentage point sequentially and 4 percentage points year over year.

第二季完全稀釋後每股盈餘（EPS）為 4.61 美元，基本每股盈餘為 4.80 美元。我們認為本季獲利反映了強勁的核心獲利能力，以及我們稍後將說明的幾項業外項目。合併淨營收達 1,911 億元，季增 10%，年增 27%。毛利為 402 億元，毛利率 21.0%，季增 1 個百分點，年增 4 個百分點。

The sequential improvement stems from higher operating leverage in our ATM business. The annual improvement reflects both structural efficiency gains and a more favorable NT dollar environment. Operating expenses totaled $19 billion, up $3.5 billion year over year, primarily driven by higher labor costs and further R&D spend to support LEAP initiatives.

季對季的改善主要來自我們 ATM 業務較高的營運槓桿。年對年的改善則反映結構性效率提升以及較有利的新台幣環境。營業費用合計 190 億元，年增 35 億元，主要由較高的人力成本以及為支持 LEAP 計畫而增加的研發支出所帶動。

As a percentage of revenue, operating expenses remained flat at 10% and declined 0.3 percentage points annually. This delivered an operating profit of $21.1 billion, up 21% quarter over quarter and 107% year over year, with an operating margin of 11.1%, expanding 1 percentage point sequentially and 4.3 percentage points annually.

以營收百分比計，營業費用維持在 10% 持平，並較去年下降 0.3 個百分點。因此營業利益為 211 億元，季增 21%、年增 107%；營業利益率為 11.1%，季增 1 個百分點、年增 4.3 個百分點。

Non-operating income totaled $4.6 billion compared to $0.7 billion in the prior quarter. While this amount appears relatively elevated versus prior quarters, the increase primarily reflects some potentially non-recurring gains. Our net non-operating income includes mark-to-market equity gains related to our corporate investments of $4.2 billion, foreign currency hedging gains of $1.5 billion, and other items including equity method investment income, government grants, rent income, and dividends totaling $0.8 billion.

業外收入合計 46 億元，較前一季的 7 億元增加。雖然此金額相較前幾季看起來偏高，但增加主要反映部分可能屬於一次性的收益。我們的業外淨收益包含：與公司投資相關的按市價評價（mark-to-market）股權收益 42 億元、外匯避險收益 15 億元，以及其他項目（包含權益法投資收益、政府補助、租金收入與股利）合計 8 億元。

These gains were partially offset by $1.9 billion in net interest expenses. Tax expense was $4.2 billion. Our effective tax rate came in at 16.4%. We saw a lower effective tax rate due to higher R&D credits generated during the quarter. We continue to expect an effective tax rate of 18% for the year. Net income for the quarter was $21.1 billion, up 49% sequentially and 180% year over year.

上述收益部分被 19 億元的淨利息費用所抵銷。所得稅費用為 42 億元。有效稅率為 16.4%。有效稅率較低，係因本季產生較高的研發抵減。我們仍預期全年有效稅率為 18%。本季淨利為 211 億元，季增 49%，年增 180%。

On page 6 is a graphical view of our consolidated quarterly performance. For the second-quarter 2026, our ATM business represented 66% of our consolidated holding company revenue, while representing 94% of our operating profit. This is compared to 61% of consolidated holding company revenue, while representing 87% of operating profit in the second quarter last year. We see this being primarily driven by the growth of our ATM LEAP services over the last few years.

第 6 頁為我們合併季度表現的圖示。在 2026 年第二季，我們的 ATM 業務占合併控股公司營收的 66%，但貢獻了 94% 的營業利益。相較之下，去年第二季 ATM 占合併控股公司營收 61%，並貢獻 87% 的營業利益。我們認為這主要是過去幾年 ATM LEAP 服務成長所帶動。

On page 7 is our ATM P&L. The ATM revenue reported here contains revenues eliminated at the holding company level related to intercompany transactions between our ATM and EMS businesses. For the second quarter of 2026, we recorded record revenues for our ATM business of $126.1 billion, up $13.7 billion sequentially, and $33.6 billion annually, representing an increase of 12% sequential and 36% annual growth.

第 7 頁為我們的 ATM 損益表。此處揭露的 ATM 營收包含在控股公司層級因 ATM 與 EMS 業務間關係人交易而沖銷的營收。2026 年第二季，我們 ATM 業務營收創新高達 1,261 億元，季增 137 億元、年增 336 億元，分別代表季增 12% 與年增 36%。

Gross profit for our ATM business was $34.5 billion, up $5.3 billion sequentially, and up $14.2 billion year over year. ATM gross profit margin was 27.3%, up 1.3 percentage points sequentially, and 5.4 percentage points annually, driven by higher operating leverage In a more favorable revenue mix, particularly a higher mix of LEAP.

ATM 業務毛利為 345 億元，季增 53 億元，年增 142 億元。ATM 毛利率為 27.3%，季增 1.3 個百分點、年增 5.4 個百分點，主要受惠於更高的營運槓桿與更有利的營收組合，尤其是 LEAP 占比提高。

During the second quarter, operating expenses were $14.7 billion, up $1.4 billion sequentially, and $3.3 billion year over year. The sequential and annual increases in operating expenses are primarily related to higher overall labor costs and general R&D expenses. Our operating expense ratio was 11.7%, down by 0.1 percentage points sequentially and 0.6 percentage points annually. We expect our ATM operating expense ratio should continue to improve during the back half of 2026.

第二季營業費用為 147 億元，季增 14 億元、年增 33 億元。營業費用的季增與年增主要與整體人力成本上升及一般研發費用增加有關。營業費用率為 11.7%，季減 0.1 個百分點、年減 0.6 個百分點。我們預期 2026 年下半年 ATM 的營業費用率將持續改善。

During the second quarter, operating profit was $19.8 billion, representing a sequential 25% increase of $3.9 billion and a 124% annual increase of $11 billion. Operating margin was 15.7% up 1.6 percentage points sequentially and up 6.2 percentage points year over year. The NT dollar had a positive 0.1 sequential and a 0.6 annual percentage point impact to our gross and operating margins.

第二季營業利益為 198 億元，較前一季增加 39 億元、季增 25%，並較去年增加 110 億元、年增 124%。營業利益率為 15.7%，季增 1.6 個百分點、年增 6.2 個百分點。新台幣對我們毛利率與營業利益率的影響為：季對季正向 0.1 個百分點、年對年正向 0.6 個百分點。

On page 8, you'll find a graphical representation of our ATM P&L. The chart highlights the improvement in our gross profit margin. It should be noted here that our second- and third-quarter 2025 margins were heavily impacted by NT dollar strengthening. Over this timeframe, our margin improvement has been largely driven by the recovery of our utilization rate related to our general manufacturing capacities and increasing LEAP product mix.

第 8 頁可看到我們 ATM 損益表的圖示呈現。圖表突顯了毛利率的改善。需要注意的是，我們 2025 年第二季與第三季的毛利率受到新台幣升值的顯著影響。在此期間，我們的毛利率改善主要由一般製造產能稼動率回升以及 LEAP 產品組合占比提升所帶動。

On page 9 is our ATM revenue by the three C market segments. LEAP services are primarily included within our computing applications, with a lesser amount being included in the communications applications. As can be seen here, the computing application percentage continues to grow steadily. At this time, we see this trend continuing into 2027 and 2028.

第 9 頁為我們 ATM 依三個 C 市場區隔的營收。LEAP 服務主要歸類於運算應用，另有較小部分歸類於通訊應用。如圖所示，運算應用的占比持續穩定成長。目前我們預期此趨勢將延續至 2027 年與 2028 年。

On page 10, you will find our ATM revenue by service type. Despite the overall growth in our business, we did not see substantial shifts in service types during the quarter. All business lines appear to be keeping pace outside of small adjustments between materials and others.

第 10 頁為我們 ATM 依服務類型的營收。儘管整體業務成長，本季我們並未看到服務類型出現顯著變化。除材料與其他項目之間有些微調整外，各業務線看起來都能同步跟上成長步伐。

At the beginning of the year, we believed that our test business, led by rapid expansion of our wafer sort business, would outpace growth in our assembly business. While our test business has shown the strong growth we initially expected, our assembly business has been showing even stronger than expected growth, especially as it relates to legacy wire bond services. We now believe that both our assembly and test businesses will grow at similar rates during the year.

在年初時，我們認為由晶圓測試業務快速擴張所帶動的測試業務，成長速度將超越封裝組裝業務。雖然測試業務確實展現了我們最初預期的強勁成長，但封裝組裝業務的成長更超出預期，尤其是在傳統打線（wire bond）服務方面。我們現在認為，今年封裝組裝與測試兩項業務將以相近的速度成長。

On page 11, you can see the second-quarter results of our EMS business. EMS revenues grew 6% sequentially and 12% annually to $65.8 billion. Sequentially, our EMS business's gross margin decreased by 0.6 percentage point to 8.9%. This change was principally the result of product mix differences.

在第 11 頁，您可以看到我們 EMS 業務的第二季業績。EMS 營收較前一季成長 6%，較去年同期成長 12%，達到 658 億美元。就季比而言，我們 EMS 業務的毛利率下降 0.6 個百分點至 8.9%。此一變動主要是產品組合差異所致。

EMS operating expenses increased by $0.3 billion sequentially and annually. Our second-quarter EMS operating expense ratio of 6.5% was flat sequentially and down 0.4 percentage points annually. Operating margin came in at 2.4%, down 0.6 percentage points sequentially and 0.2 percentage points year over year. The sequential margin decline is the result of product mix and a higher component cost environment. Our EMS second-quarter operating profit was $1.6 billion, down $0.3 billion sequentially and up $0.1 billion annually.

EMS 營業費用較前一季及較去年同期均增加 3 億美元。我們第二季 EMS 營業費用率為 6.5%，較前一季持平，較去年同期下降 0.4 個百分點。營業利益率為 2.4%，較前一季下降 0.6 個百分點，較去年同期下降 0.2 個百分點。季比利益率下滑係由產品組合以及較高的零組件成本環境所致。我們 EMS 第二季營業利益為 16 億美元，較前一季減少 3 億美元，較去年同期增加 1 億美元。

On the bottom of the page, you will find a graphical representation of our EMS revenue by application. Generally, the moves in the consumer and communication categories are related to the seasonality of the underlying products we service. The growth in the computing category is largely driven by business related to our AI accelerator products.

在頁面下方，您會看到我們 EMS 依應用別的營收圖示。一般而言，消費與通訊類別的變動與我們所服務之底層產品的季節性有關。運算類別的成長主要由我們 AI 加速器產品相關業務所帶動。

From the holding company perspective, we continue to pursue synergies between our ATM and EMS businesses to co-develop system-level solutions, particularly in key areas such as optical interconnects, power delivery, and thermal management. By integrating critical EMS competencies with ATM technologies, we have the potential to optimize entire systems end-to-end, applying the same co-engineering disciplines that successfully scaled SIP architecture.

從控股公司角度來看，我們持續推動 ATM 與 EMS 業務之間的綜效，共同開發系統級解決方案，特別是在光學互連、電力傳輸與散熱管理等關鍵領域。透過整合 EMS 的關鍵能力與 ATM 技術，我們有機會端到端優化整體系統，並套用與成功擴展 SIP 架構相同的共同工程（co-engineering）方法論。

On page 12, you will find key line items from our balance sheet. At the end of the quarter, we had cash, cash equivalents, and current financial assets of $107.4 billion. Our total interest-bearing debt increased by $40.9 billion to $306.2 billion. Total unused credit lines amounted to $396.2 billion. Our EBITDA for the quarter was $45.8 billion. Our net debt to equity this quarter was 47%.

在第 12 頁，您會看到我們資產負債表的主要項目。截至本季末，我們的現金、約當現金及流動金融資產為 1,074 億美元。我們的有息負債總額增加 409 億美元至 3,062 億美元。未動用授信額度合計為 3,962 億美元。本季 EBITDA 為 458 億美元。本季淨負債對股東權益比率為 47%。

On page 13, you will find our equipment capital expenditures relative to our EBITDA. Machinery and equipment capital expenditures for the second quarter, in US dollars, totaled $1.7 billion, of which $840 million was used in packaging operations, $804 million in testing operations, $49 million in EMS operations, and $2 million in interconnect materials operations, and others. In addition to spending on machinery and equipment, during the quarter we also spent $658 million on facilities.

在第 13 頁，您會看到我們設備資本支出相對於 EBITDA 的情況。第二季機器與設備資本支出（以美元計）合計 17 億美元，其中 8.40 億美元用於封裝作業、8.04 億美元用於測試作業、4,900 萬美元用於 EMS 作業、200 萬美元用於互連材料作業及其他。除機器與設備支出外，本季我們亦在廠房設施方面支出 6.58 億美元。

It is worth reiterating what Dr. Wu spoke of earlier. We are investing in capacities and facilities because what we do has immediate impacts on key bottlenecks in semiconductor supply, performance, and efficiency. The AI build-out, regardless of open or closed model weighting, will require unprecedented hardware capacity and capability expansion. At this point, we are just trying to keep up.

值得重申吳博士先前提到的內容。我們投資產能與設施，是因為我們所做的事情會立即影響半導體供應、效能與效率的關鍵瓶頸。AI 的建置，不論開放或封閉模型的比重如何，都將需要前所未有的硬體產能與能力擴張。就目前而言，我們只是努力跟上需求。

With that, I'll hand the presentation over to Joseph to walk through the company's outlook.

接下來，我把簡報交給 Joseph，請他說明公司的展望。

Okay, now for third quarter of 2026 outlook, based on our current business outlook and exchange rate assumption of USD1 to TWD31.9 versus last quarter TWD31.6, management projects overall performance for the third quarter of 2026 to be as follows.

好的，接著是 2026 年第三季展望。根據我們目前的業務展望，以及匯率假設為 1 美元兌新台幣 31.9（相較上一季為新台幣 31.6），管理階層對 2026 年第三季整體表現的預估如下。

On consolidated basis, in NT dollar terms, our consolidated third-quarter revenue should grow by 21% to 22% quarter over quarter. Our consolidated third-quarter growth margin should be between 20.5% to 21.5%. Our consolidated third-quarter operating margin should be between 11.5% to 12.5%.

以合併基礎、以新台幣計，我們第三季合併營收預期較前一季成長 21% 至 22%。我們第三季合併毛利率預期介於 20.5% 至 21.5%。我們第三季合併營業利益率預期介於 11.5% 至 12.5%。

For ATM, in NT dollar terms, our ATM third-quarter revenue should grow by 11% to 13% quarter over quarter. Our ATM third-quarter gross margin should be between 28% to 29%. Our EMS -- in NT dollar terms, our EMS third-quarter revenue should grow by around 40% quarter over quarter. Our EMS third-quarter operating margin should be between 3.2% to 3.4%. With that is the '26 third quarter outlook.

就 ATM 而言，以新台幣計，我們 ATM 第三季營收預期較前一季成長 11% 至 13%。我們 ATM 第三季毛利率預期介於 28% 至 29%。至於 EMS——以新台幣計，我們 EMS 第三季營收預期較前一季成長約 40%。我們 EMS 第三季營業利益率預期介於 3.2% 至 3.4%。以上為 2026 年第三季展望。

Now adding a bit of color for the full year and next. First, on CapEx, given a stronger demand for LEAP in 2026 and beyond, we will need to add another USD1 billion each this year for facilities and equipment. So the total is adding another $2 billion for CapEx. While the additional investment for facilities and most of the equipment are for LEAP, we also need to add capacity for mainstream advanced packaging and testing to support the general market demand.

接著補充一些關於全年以及明年的說明。首先在資本支出（CapEx）方面，鑑於 2026 年及之後 LEAP 需求更強，我們今年在廠房與設備上各需要再增加 10 億美元。因此合計將再增加 20 億美元的資本支出。雖然新增投資中的廠房與大部分設備是為了 LEAP，我們也需要擴充主流先進封裝與測試的產能，以支援整體市場需求。

On LEAP, well, this year's LEAP service revenue tracking ahead of our prior-quarter guidance of USD3.5 billion. As business momentum continues to be very strong, we are aiming to double our LEAP revenue in 2027.

關於 LEAP，今年的 LEAP 服務營收進度已超前我們上一季所給出的 35 億美元指引。由於業務動能持續非常強勁，我們目標在 2027 年將 LEAP 營收倍增。

Lastly, on ATM profitability, with expanding margin-accretive LEAP and test businesses, our second-quarter ATM gross margin of 20.3% came in ahead of our guidance. We continue to expect sequential margin improvement with fourth-quarter ATM gross margin likely to exceed our structural margin ceiling of 30%. At such point, we will start to review if we would adjust our structural margin range.

最後談 ATM 獲利能力。隨著具備提升毛利效果的 LEAP 與測試業務擴張，我們第二季 ATM 毛利率 20.3% 優於指引。我們仍預期毛利率將逐季改善，第四季 ATM 毛利率很可能超過我們結構性毛利率上限 30%。屆時，我們將開始檢視是否調整我們的結構性毛利率區間。

With that, thank you.

以上，謝謝。

During the Q&A session that follows, we would appreciate if your questions could be as clear and concise as possible and asked singularly. We will start by taking questions from live participants and then alternate in questions from our online participants.

接下來的問答環節，懇請各位提問盡量清楚、精簡，並一次只問一個問題。我們將先接受現場與會者提問，之後再與線上與會者的提問交替進行。

I, as the moderator, will be receiving each question and repeating and directing each asked question. After an initial question, the participant may ask a follow-up question, clarifications of the earlier question, or another question entirely. Then we'll move to the next participant. Participants may return to the queue for any additional questions and/or clarifications. Thank you.

我作為主持人，將接收每個問題，並重述與引導每個被提出的問題。在第一個問題之後，提問者可以再追問、針對先前問題補充釐清，或提出另一個全新的問題。接著我們會進到下一位提問者。如有其他問題及／或需要釐清之處，與會者可再次回到排隊序列。謝謝。

Dr. Wu, so just to follow our conversation, your industry peers seems to be more aggressive in the US operation, not just at Amkor, but also KYEC announced to do that, right? I want to ask you about your partnership with the US customers and also your operation in the US. I guess it is a little bit separate question, but can you also comment a little bit about the Intel EMIB involvement by ASE? Thank you.

吳博士，延續我們剛才的對話，您的同業似乎在美國營運方面更積極，不只是 Amkor，KYEC 也宣布要這麼做，對吧？我想請教您與美國客戶的合作關係，以及您在美國的營運布局。我想這是稍微分開的問題，但也能否請您稍微評論一下 ASE 參與 Intel EMIB 的情況？謝謝。

Charlie, you're asking about the competitive landscape that we're facing, and secondly, EMIB? Okay. Dr. Wu?

Charlie，你是在問我們所面臨的競爭態勢，其次是 EMIB？好的。吳博士？

Well, let me answer the US operation first. When I'm going to talk about it, I have repeated several times for the last two years, we do have US operation, ISE, test development, also services in California. Well, one factory in Fremont, California. We have one factory in San Jose. We're in the process of expanding to number three and number four factories.

好，我先回答美國營運的部分。談到這個，我在過去兩年已經重複提過好幾次：我們確實在美國有營運，包括 ISE、測試開發，以及在加州提供的服務。我們在加州 Fremont 有一座工廠。我們在 San Jose 也有一座工廠。我們正進行擴充，規劃第 3 與第 4 座工廠。

These are particularly required by our customers during the -- well, for the last 20 years in Bay Area for upfront chip design, test development, also technology development. And that is ongoing right now. The agreement that we have with our leading customer is the following.

這些在過去 20 年於灣區進行前端晶片設計、測試開發以及技術開發期間，特別是我們的客戶所要求的。而這項工作目前仍在持續進行。我們與主要客戶之間的協議如下。

We will develop and build fully automated, efficient line in Taiwan. When we are comfortable with our resources and efficiency, in due time, we will migrate and move the operation to the other places in the world. It could be United States, could be somewhere else. That has always been the case. So it's not like we're not supporting. We are supporting development, R&D, also architectural design.

我們將在台灣開發並建置一條全自動化、高效率的產線。當我們對自身資源與效率感到足夠成熟、在適當時機，我們會把營運遷移並移轉到世界其他地區。可能是美國，也可能是其他地方。一直以來都是如此。所以並不是我們沒有在支持。我們正在支持開發、研發，以及架構設計。

In terms of manufacturing process development, for now, we are focusing on Taiwan. Until we build the appropriate scale, having the appropriate resources and know-how and efficiency, upon that time, we will work with our customers to move to the other part of the world for better logistics.

就製造製程開發而言，目前我們聚焦在台灣。直到我們建立起適當的規模、具備相應的資源、專業知識與效率之後，屆時我們將與客戶合作，為了更好的物流而移往世界其他地區。

The second comment is EMIB. EMIB, I'm going to repeat what TSMC has responded. Right now is really capacity constrained. If there's any other alternative technology that can offer the same yield to resolve the bottleneck that we have for the AI infrastructure, we welcome to see that. For ASE, we're also collaborating with the other customers also along the same line, including EMIB.

第二個評論是關於 EMIB。針對 EMIB，我會重申台積電的回應。目前確實受到產能限制。如果有任何其他替代技術能提供相同良率、以解決我們在 AI 基礎設施上的瓶頸，我們樂於看到。對 ASE 而言，我們也正與其他客戶在同一方向合作，包括 EMIB。

The following question is the -- if EMIB posting potential threat or competitive advantage over a CoWoS, right? That's always a possibility. I have been working in the packaging industry for 40 years. I've gone through about a thousand different packaging design. At the end of the day, probably 20 will leave, but we'll be here. So there's always a competing technology, alternative materials. That's always been the focus of R&D.

接下來的問題是——EMIB 是否對 CoWoS 構成潛在威脅或競爭優勢，對吧？這永遠都有可能。我在封裝產業工作了 40 年。我經歷過大約一千種不同的封裝設計。到最後，可能只有 20 種會留下來，但我們會一直在這裡。所以總是會有競爭技術、替代材料。這一直都是研發的重點。

For EMIB, we're happy to see if EMIB can ramp up. In terms of efficiency, performance, that's up to the system as well as the market to decide. We're not going to make a judgment call. Who has the superiority? But our focus is to support the CoWoS. We're trying to ramp up the co-op scale efficiency as fast as we can.

對於 EMIB，如果 EMIB 能夠順利放量，我們也樂見其成。至於效率、效能，則要由系統以及市場來決定。我們不會做價值判斷。誰更具優勢？但我們的重點是支持 CoWoS。我們正盡快提升 CoWoS 的產能規模效率。

If customer asks for other alternative technology, ASE will also include that in our roadmap. I don't think there's a potential threat. Again, we are a pure play. If the EMIB substrate becomes the right alternative, which is by EMIB substrate, and we do the assembly, there is no conflict. Anything is welcome. It's not like a zero-sum game like you can only choose one. It's never been the case. The world is a big place.

如果客戶要求其他替代技術，ASE 也會把它納入我們的路線圖。我不認為這是潛在威脅。再說一次，我們是純代工（pure play）。如果 EMIB 載板成為合適的替代方案——也就是由 EMIB 載板供應，而我們負責組裝——那就不存在衝突。任何方案都歡迎。這不是零和遊戲，不是只能二選一。從來都不是這樣。世界很大。

More question from the floor?

現場還有更多問題嗎？

Hi, Dr. Wu and Joseph, Ken. Thank you for taking my question. My first question is, can you give us a little bit more detailed guideline for your Q3 ATM demand driver for both for LEAP and also for general packaging and testing, the more detailed demand driver across the key applications or products?

嗨，吳博士和 Joseph，我是 Ken。謝謝讓我提問。我的第一個問題是，能否就你們第三季 ATM 需求驅動因素提供更細的指引，包含 LEAP 以及一般封裝與測試，能否按主要應用或產品更詳細說明需求驅動？

Rick, you're asking for maybe segmented drivers of what's helping the industry or our results pick up at this point. Is that correct?

Rick，你是在問，可能有哪些分項驅動因素在此時點帶動產業或我們的業績回升。是這個意思嗎？

Yes, right, demand driver.

是的，沒錯，需求驅動。

The question is the Q3 driver. Again, we have a very awkward and peculiar position because we're capacity constrained. So we're talking about 12% to 13% growth. That means we have to add 12% to 13% capacity. The demand overall is strong. I'm not going to comment about the memory pricing or any consumer devices, but overall, all of our customer are asking for more devices for Q3 and Q4.

問題是第三季的驅動因素。再說一次，我們的處境很尷尬也很特殊，因為我們受到產能限制。所以我們談的是 12% 到 13% 的成長。這代表我們必須增加 12% 到 13% 的產能。整體需求很強。我不評論記憶體價格或任何消費性裝置，但整體而言，我們所有客戶都在要求第三季和第四季更多的出貨量。

Rick, follow up?

Rick，要追問嗎？

No, not follow up, the second question. Can you elaborate your development of -- for CoWoS like -- and this year, roughly about how much is still coming from outsourcing and what's your development and progress into next year of your internal for CoWoS development?

不是追問，是第二個問題。你們能否說明你們在 CoWoS 方面的開發——像是今年大約有多少仍來自外包，以及你們內部 CoWoS 開發在明年的規劃與進度？

Rick, you're looking for the composition of our LEAP services between what is potentially OS-type services and full process and test.

Rick，你是在看我們 LEAP 服務的組成，介於可能屬於 OS 類型的服務與全製程及測試之間的占比。

I think basically, we are on track with our full process business development. But this year, we said that we're going to have about $300 million worth of revenue coming from that space, and things are on track. We are aggressively expanding that capacity, and by next year, I think we will have pretty substantial growth in that area as well.

我想基本上，我們的全製程業務開發進度符合規劃。但今年我們說過，該領域將帶來約 3 億美元的營收，目前一切都在軌道上。我們正積極擴充該產能，而到明年，我認為該領域也會有相當可觀的成長。

And again, this is something that under development, we have not fully, I think margin wise, has not fully reflected at this point. But going forward, I think the full process will also be another margin-accretive business for us and with pretty good substantial growth coming into next year.

再者，這仍在開發中，我認為就毛利而言，目前尚未完全反映出來。但往後看，我認為全製程也將成為另一項能提升毛利的業務，並在明年帶來相當可觀的成長。

We have our next question coming from online.

接下來我們有一個線上提問。

Sunny Lin, UBS.

UBS 的 Sunny Lin。

Thank you very much. Good afternoon. Congrats on the very strong outlook. Sorry for not being able to attend in person. So my first question is maybe to follow up on 2027 LEAP outlook. And so maybe take a step back. You are guiding for this year LEAP to exceed $3.5 billion. So should we be looking at maybe $4 billion or maybe between $3.5 billion to $4 billion? And then for 2027, for it to double, have we already had an idea in terms of breakdown by full process, outsourcing for substrate and also test?

非常感謝。午安。恭喜你們展望非常強勁。抱歉我無法親自到場。我的第一個問題可能是追問 2027 年 LEAP 的展望。先退一步看。你們指引今年 LEAP 將超過 35 億美元。那我們是否應該看作大約 40 億美元，或是介於 35 億到 40 億美元之間？另外，對於 2027 年要翻倍，你們是否已經有按全製程、載板外包以及測試的拆分概念？

Sunny, you're looking for an update on our LEAP guidance for this year and then looking forward into 2027, whether we have any nuggets of information for you. Is that correct?

Sunny，你是在尋求我們今年 LEAP 指引的更新，並展望到 2027 年，看看我們是否能提供一些額外資訊。是這樣嗎？

Right. Thank you, Ken.

是的。謝謝你，Ken。

Currently. Well, I think Tien mentioned that we are ahead of our lead revenue this year. And by ahead, I think I think we will be adding another $200 million worth of revenue coming out of this year. And on top of that, for next year, we still see very strong momentum, and we believe at this point that we should be able to double the revenue of the revenue next year.

目前。我想 Tien 提到，我們今年的 LEAP 營收進度超前。所謂超前，我認為今年我們會再增加約 2 億美元的營收。此外，對明年我們仍看到非常強的動能，我們相信以目前來看，明年應該能把營收再翻倍。

In terms of -- we are going full speed ahead with both our assembly as well as tests. And as Ken mentioned, at this point, assembly is also catching up in terms of its growth momentum with tests. So next year, I think the combination will be pretty much similar to what we're having seen this year.

就——我們在組裝與測試兩方面都全速推進。如 Ken 所提，目前組裝在成長動能上也正在追上測試。因此明年我認為兩者的組合將與今年看到的情況大致相似。

Well, but -- sorry, maybe let me clarify, if I may. So I just want to get a bit more color in terms of LEAP breakdown going to 2027. I do assume that full process should account for a much larger portion for 2027. So any color will be very helpful.

嗯，不過——抱歉，也許讓我再澄清一下，如果可以的話。我只是想更了解一下，關於 LEAP 拆分到 2027 年的細項。我確實假設到 2027 年，full process 應該會占更大一部分。所以任何補充說明都會非常有幫助。

Probably two quarters from now, we'll give you better color. Thank you.

大概再過兩個季度，我們會給你更清楚的說明。謝謝。

Sure. So maybe if I may, second question, Dr. Wu, I want to double click on your earlier comment regarding AI driving lots of new applications with different complexities and scope. And so if I may, maybe based on your current engagement with the clients, what type of devices you are seeing better visibility that -- and then if we could see a more meaningful rep in the coming few years?

好的。那如果可以的話，第二個問題，吳博士，我想針對您先前提到 AI 會帶動許多新應用、且複雜度與範疇各不相同的評論再深入追問。所以如果可以，基於您目前與客戶的互動，您看到哪些類型的裝置有較高的能見度——以及未來幾年我們是否可能看到更具意義的占比提升？

Sunny, you're looking for a little bit more color in terms of what potential applications and devices that we may see coming in the future.

Sunny，你是想更了解一些未來可能出現的應用與裝置的潛在方向，對嗎？

Yes, due to AI.

是的，因為 AI。

I would like to separate the technology statement versus business statement. Technology would take 15 to 20 years to develop. And the business statement, typically, you're interested in four quarters, probably the longest lead time. So let me comment on the CoWoS, for example, or EMIB and other alternative technology.

我想把技術面的陳述與商業面的陳述分開來談。技術通常需要 15 到 20 年來發展。而商業面的陳述，一般你關心的是四個季度，可能已經是最長的前置時間。所以我先以 CoWoS 為例，或是 EMIB 以及其他替代技術來做一些評論。

The AI data center is driving computational intensity. Therefore, the reticle size becomes bigger and bigger that we know already. Everybody demands more bandwidth from logic, ASIC, as well as memory. Therefore, the immediate up round will be the panel or more complicated CoWoS to accommodate the bigger chiplet or reticle size. That's one dimension that we're dealing with.

AI 資料中心正在推升運算強度。因此，我們已經知道光罩（reticle）尺寸會越來越大。每個人都要求邏輯、ASIC 以及記憶體提供更高的頻寬。因此，近期最直接的升級方向會是採用 panel（面板級）或更複雜的 CoWoS，以容納更大的 chiplet 或 reticle 尺寸。這是我們正在處理的一個面向。

We also comment that somewhere along the line, towards the end of the year, the CPO will start launching. Initially, it could be in small volume. However, we will have a critical benchmark information in terms of the bandwidth and the system performance and also the thermal dissipation that can resolve. Pending on the cost performance ratio and also the yield, we will learn critical information. When the reticle side becomes bigger and the optics becoming in a different hierarchy, then the power delivery becomes the next immediate questions. This will be deployed in the next two to three years.

我們也提到，在某個時間點、接近年底時，CPO 會開始推出。一開始可能是小量。不過，我們會取得一些關鍵的標竿資訊，包括頻寬、系統效能，以及可解決的散熱能力。取決於成本效益比以及良率，我們會學到關鍵資訊。當 reticle 尺寸變得更大、光學進入不同層級時，供電就會成為下一個迫切的問題。這會在未來兩到三年內部署。

On a much longer term, there are other things people are working on. For example, today we're primarily dealing with digital and digital. There's a lot of activity going on to start accommodating sensor, analog, mixed signal with digital. If you think about the humanoid applications, other than the brain, the eyes, ears, fingers, they're all analog. So what are the technology that can provide the low power, the bandwidth, as well as the fully automated, high volume, integrated capability to do digital-analog mixed signal?

在更長期的角度，還有其他事情是大家正在投入的。例如，今天我們主要處理的是數位對數位。目前有很多活動在進行，開始要把感測器、類比、混合訊號與數位整合在一起。如果你想到人形機器人的應用，除了大腦之外，眼睛、耳朵、手指，這些都是類比。那麼，什麼樣的技術能提供低功耗、頻寬，以及全自動化、高量產、整合式的能力，來做數位—類比的混合訊號？

These are the infrastructure a pure-play OSAT supplier should work with the customer trying to develop. The technology takes 15 to 20 years to develop. All of the business we're taking or deploying now are the result for the past 15 years of effort. So I would like to make that separation. But if you're interested, I think the AI has a very long leg. You can listen to all the smart people about infrastructure, AI data center, about agentic, and also the physical interface.

這些就是一家純 OSAT 供應商應該與客戶合作、共同開發的基礎建設。技術需要 15 到 20 年來發展。我們現在承接或部署的所有業務，都是過去 15 年努力的成果。所以我想做這樣的區隔。但如果你有興趣，我認為 AI 會有很長的成長曲線。你可以去聽所有聰明的人談基礎建設、AI 資料中心、agentic（代理式）以及實體介面。

I also would like to give you some color on how I see the AI. I'm not the best person to talk about it. I have a very simple view to look at AI. AI is for a new pattern recognition in the domain knowledge you're familiar with. When people start crossing domain, IT versus pharmaceutical, IT versus medical, the new pattern get recognized. When you recognize the pattern, that's the potential application.

我也想補充一些我對 AI 的看法。我不是最適合談這個的人。我對 AI 的看法很簡單。AI 是在你熟悉的領域知識中，做新的模式辨識。當人們開始跨領域，例如 IT 對製藥、IT 對醫療，就會辨識出新的模式。當你辨識出模式，那就是潛在的應用。

We're at the beginning of the knowledge collection and the early stage of a pattern recognition. That's what AI does. But today we'll only talk about single field. Eventually in the world, I do not know how many, it's got to be a hundred domain. How do you bridge all the domain knowledge and start creating something hybrid? The medicine, the surgical, humanoid, all of this are the potential application. All of the technology, we use exactly the building block that we're developing today. So what we're doing, not just for AI data center.

我們正處於知識蒐集的起步階段，以及模式辨識的早期階段。這就是 AI 在做的事。但今天我們大多只談單一領域。最終在世界上，我不知道有多少，應該會有上百個領域。你要如何把所有領域知識串接起來，開始創造出混合式的東西？醫藥、外科、人形機器人，這些都是潛在應用。所有技術，我們用的正是今天正在開發的那些 building block。所以我們做的，不只是為了 AI 資料中心。

If all we're thinking about is AI data center for a specific customer, then we really miss the point of what engineering does. Engineering is the built elegant solution for the future demand regardless how difficult it is. It takes 15 to 20 years. This is not a stock trade. But what you're seeing is we are the first mover. We already have the cluster efficiency in the early stage of this AI transition. I think that means a lot if you really understand what I'm trying to tell you.

如果我們只把焦點放在某個特定客戶的 AI 資料中心，那就真的錯過工程的重點。工程是在不管多困難的情況下，為未來需求打造優雅的解決方案。這需要 15 到 20 年。這不是股票交易。但你所看到的是，我們是先行者。在這波 AI 轉型的早期階段，我們已經具備叢集效率（cluster efficiency）。如果你真的理解我想表達的意思，我認為這代表很多。

No problem, sounds good. Thank you very much for sharing, Dr. Tien.

沒問題，聽起來很好。非常感謝您的分享，田博士。

Thank you, Sunny.

謝謝你，Sunny。

Next online question is from Gokul Hariharan.

下一位線上提問者是 Gokul Hariharan。

Okay, good afternoon. Thanks, Dr. Wu, Joseph, Ken. The first question just to Dr. Wu, you have been working in very close partnership with the LEAP foundry partner for the last few years, which has been helping the LEAP revenue growth quite nicely. They seem to be transitioning a little bit towards 3D panel level kind of future technologies, at least pretty aggressively. It looks like in the next two, three years, compared to the last three, four years have been -- largely been about CoWoS and various forms of CoWoS.

好的，午安。謝謝，吳博士、Joseph、Ken。第一個問題是給吳博士：過去幾年您與 LEAP 的晶圓代工夥伴非常緊密合作，這也相當有效地帶動了 LEAP 營收成長。他們似乎正在有點轉向 3D panel level 這類未來技術，而且至少看起來相當積極。看起來未來兩、三年，相較於過去三、四年主要聚焦在 CoWoS 以及各種形式的 CoWoS，會有所不同。

So as they are embarking on that kind of transition, Dr. Wu, could you help us understand what is ASE's role? Is it getting elevated further in that partnership? And obviously, some of the full-process CoWoS is also a result of them leaving some of these older areas to you as well. So just could you outline what you talked about on the technology side in relationship to how this partnership with the LEAP Foundry is going to work out in the next maybe two, three years first?

所以當他們展開這樣的轉型時，吳博士，您能否協助我們理解 ASE 的角色是什麼？在這個合作關係中，ASE 的地位是否會進一步提升？而且很明顯地，一些 full-process CoWoS 也是因為他們把某些較舊的領域交由你們來做的結果。所以，您能否先從技術面，說明一下您剛才談到的內容，並連結到未來兩到三年 ASE 與 LEAP Foundry 的合作關係將如何運作？

Gokul, you're asking about as our foundry partner progresses further down their technology roadmaps, what our plans are to intermix and or interlink with them in this process. Is that correct?

Gokul，你是在問，當我們的晶圓代工夥伴在其技術藍圖上持續往前推進時，我們的計畫是如何在這個過程中與他們交織與連結，對嗎？

Yes. And do you get more value add out of it as they kind of migrate out to more complicated packaging?

是的。另外，當他們往更複雜的封裝遷移時，你們是否能從中取得更多附加價值？

The collaboration is long term. Once again, I want to focus on the pure play. The foundry ideally is a pure play for wafer. Testing, bumping, packaging are enabler to support the delivery of the wafer to the correct customer in the correct timing, also with the correct efficiency.

這個合作是長期的。再一次，我想聚焦在 pure play（純粹分工）的概念。晶圓代工理想上是專注於晶圓的 pure play。測試、凸塊（bumping）、封裝則是促成者，用來支援把晶圓在正確的時間、以正確的效率，交付給正確的客戶。

Packaging pure play is to develop a packaging architecture, all the Lego pieces to facilitate that. When the pure-play foundry with the pure-play packaging collaborate, that boundary is very clear. So today, there are packaging that needs to be taken by the foundry people because access to the leading-edge wafer is just simply not available to the others or some architecture requirement that has strict IP that would like to honor and respect the customer as was the foundry.

封裝純代工（pure play）的重點，是建立一套封裝架構，把所有像樂高積木一樣的元件都準備好，以利推進。當純代工晶圓廠與純代工封裝廠合作時，彼此的分工邊界是非常清楚的。因此在今天，有些封裝工作需要由晶圓代工端來承接，因為其他人根本無法取得最先進製程的晶圓，或是某些架構需求涉及嚴格的智慧財產（IP），必須遵守並尊重客戶要求，就如同晶圓代工端一樣。

That is a domain that the pure-play wafer people needs to decide. But preclude from that, the collaboration will cover all aspects in the packaging arena. In terms of where do we stop the line, when do we start the line, that depends on the wisdom of the management team of the ecosystem collaborator.

那是純晶圓代工端需要做決定的領域。但撇除那一塊，合作將涵蓋封裝領域的各個面向。至於界線要畫在哪裡、哪裡開始、哪裡結束，取決於生態系合作夥伴管理團隊的智慧判斷。

I'm not sure how to give you a better answer. But as you can see, the OS, the full process, all of the collaboration are the full spectrum. In terms of future, we do not know. But it really depends on the pure play, IP, as well as the customer's requirement. But we will try to navigate through all of the complexity.

我不確定要怎麼給你更好的答案。但如你所見，OS、完整製程，以及所有合作，都是全方位的光譜。至於未來，我們也不知道。但這真的取決於純代工模式、IP，以及客戶的需求。不過我們會努力在所有複雜性之中找到前進的路徑。

But the important thing is we all understand we would like to provide speedy -- the most elegant solution in the quickest amount of time to our customers collectively. That understanding is very clear. And it's very obvious for the next three years, as well as the ramp up in the next few years.

但重要的是，我們都理解：我們希望以最快的時間，為客戶共同提供快速——也就是最優雅的解決方案。這個共識非常清楚。而且在未來三年，以及接下來幾年的放量爬坡上，這一點都非常明顯。

Got it. That's very clear. Thank you, Dr. Wu. My next question is just on a couple of the guidance increases that you have. One is on LEAP for 2027. I think previously we were expecting, I think, $1.9 billion or so additional. Now we're expecting doubling of the revenue.

了解。這非常清楚。謝謝您，吳博士。我下一個問題是關於你們上調的幾項指引。其中一項是 2027 年的 LEAP。我記得之前我們預期大概是增加 19 億美元左右。現在我們預期營收要翻倍。

Is that primarily coming from having more line of sight into capacity availability? Because I think demand is probably still much higher than what you can really support even next year. And secondly, mainstream, us expecting the growth to be close to 20% this year from the, I think, 13% previously. Where is that upside coming from on the mainstream growth?

這主要是因為你們對產能可用性有更清楚的能見度嗎？因為我認為需求可能仍遠高於你們即使到明年也能真正支援的水準。第二，主流業務（mainstream），我們預期今年成長接近 20%，高於之前我記得的 13%。主流成長的上修動能是從哪裡來的？

Gokul, you're looking for incremental explanation related to what is driving our LEAP 2027 outlook? And also what has helped us drive our mainstream expectations up this year? Is that correct?

Gokul，你是在尋求更進一步的說明：是什麼在驅動我們對 LEAP 2027 的展望？以及是什麼讓我們今年把主流業務的預期上調？是這樣嗎？

That's right, yes.

是的，沒錯。

We have clear line of sight in terms of who needs what. We also have a clear line of sight on the building, the facility that we're building today. When we make a comment that this year, by year end, we're tracking ahead of the $3.5 billion, that comment is made because we see our yield and execution that gave us the confidence to tell you that by year end, we will achieve that target.

我們對於誰需要什麼，有很清楚的能見度。我們對於目前正在興建的廠房與設施，也有很清楚的能見度。當我們評論說今年到年底，我們的進度領先於 35 億美元的目標時，是因為我們看到良率與執行表現，讓我們有信心告訴你：到年底我們會達成那個目標。

By the same token, we also give you a comment that whatever that target has been achieved by year end, with the next 12 months, building new facility and adding new machines, we will be able to double that. That line of sight is there. The uncertainty exists in terms of how can we execute. So the business demand is not a concern. It's our own capability to execute the two lines of sight, which is clear in front of us.

同樣地，我們也告訴你：不論年底達成的目標是多少，在接下來 12 個月內，透過興建新廠並新增機台，我們將能把它翻倍。這個能見度是存在的。不確定性在於我們能如何執行到位。所以需求面不是問題。問題在於我們自身把兩條清楚擺在眼前的路徑執行到位的能力。

In terms of the general devices, we see very strong demand in industrial, power, connectivity, and storage devices. I do not know what the general market from semiconductor this year -- I don't have that number, right? I believe we will outperform the general market simply because our association with a stronger AI infrastructure development, also our superior capability to have fully automated lines for most of the general devices.

就一般元件（general devices）而言，我們看到工業、電源、連接性與儲存裝置的需求非常強勁。至於今年整體半導體市場的數字——我手上沒有那個數字，對吧？我相信我們會優於整體市場，原因很簡單：我們與更強勁的 AI 基礎建設發展連動，同時我們在多數一般元件上具備全自動化產線的優勢能力。

So people want to build devices to go into electrical vehicle, AI data center, or any high-impact, high-risk applications, chances are they will use our fully automated line. That's why we're having a very good attraction. So in terms of capacity expansion, not only we need to expand the LEAP, we also need to expand the general market.

因此，當客戶想打造要用在電動車、AI 資料中心，或任何高影響、高風險的應用時，他們很可能會採用我們的全自動化產線。這就是為什麼我們的吸引力非常強。所以在產能擴充方面，我們不只要擴 LEAP，也需要擴一般市場的產能。

That's putting a tremendous amount of pressure on ASE, which is why I think during the shareholders meeting, I complained that ASE is building 13 new facility simultaneously this year. And we just bought another seven brownfield. So we're building, we're buying, and we're spending CapEx. We're not happy about it, but the question now is that is our obligation. This is why we're here. We want to make sure we are providing the critical capacity to the best of our capability to satisfy the long-term objective for all of our customers.

這對 ASE 造成了非常巨大的壓力，所以我想在股東會上我才會抱怨：ASE 今年同時在蓋 13 座新廠。而且我們又買了另外 7 個棕地（brownfield）廠區。所以我們在建、在買、也在投入資本支出（CapEx）。我們並不開心，但現在的問題是：這是我們的責任。這就是我們存在的原因。我們要確保在能力所及的範圍內，提供關鍵產能，以滿足所有客戶的長期目標。

Got it. Yes, that's very clear. Yes, thanks. Thanks very much, Dr. Wu. Thank you.

了解。是的，這非常清楚。是的，謝謝。非常感謝您，吳博士。謝謝。

Charlie, do you want a shot at another round of questions there? Yes, go ahead. Microphone?

Charlie，你要不要再來一輪問題？好，請說。麥克風？

Thanks, Ken. I do have two questions, but can I make a clarification on the previous Q&A?

謝謝，Ken。我有兩個問題，但我可以先釐清一下前一段 Q&A 嗎？

Please, go ahead.

請，請說。

So about the LEAP revenue next year, right? So Joseph, you talk about the mix will be very similar to this year -- next year's LEAP revenue mix, right? But according to our analysis, it seems like your full 2.5D, the end customer is an 86 CPU, that part is growing 3x, right? And I assume your testing piece is also growing more than double. So how come the rest of the substrate outsourcing can also double next year? Because NVIDIA, I think the Street consensus is next year is growing 50%, and TSMC CoWoS expansion is 70%.

關於明年的 LEAP 營收，對吧？Joseph，你提到明年的 LEAP 營收組合會跟今年非常類似——是指明年的 LEAP 營收組合，對吧？但依我們的分析，你們完整的 2.5D，終端客戶是 86 CPU，那一塊成長 3 倍，對吧？而我假設你們的測試（testing）部分也成長超過一倍。那為什麼其餘的載板外包（substrate outsourcing）也能在明年翻倍？因為 NVIDIA，我想市場一致預期明年成長 50%，而台積電 CoWoS 擴產是 70%。

I think I was referring to, in general, assembly and test seems to have the same momentum. But assembly includes not just OS. There are full process. There are other steps or process steps that we are entering. So -- also some new packages that are coming on stream. So I think that's a general description of what we are seeing next year in terms of between assembly and test.

我想我指的是：整體來看，組裝與測試似乎有相同的動能。但組裝不只包含 OS。還有完整製程。也有其他我們正在切入的步驟或製程環節。另外也有一些新封裝將開始量產導入。所以我想這是我們對明年組裝與測試之間走勢的一個概括描述。

Anyway, we were very happy to see that strong growth. I just want to make sure we get a breakdown right. And next question is a little bit joking, but I just see that your quarterly CapEx exceeding your EBITDA, right? So is that your free cash flow is also turning negative? Do you think share price will react negatively tomorrow?

不管怎樣，我們很高興看到那麼強勁的成長。我只是想確保我們把拆解（breakdown）弄對。下一個問題有點開玩笑，但我看到你們單季資本支出（CapEx）超過 EBITDA，對吧？所以你們的自由現金流（free cash flow）也轉為負值了嗎？你覺得明天股價會有負面反應嗎？

Yes, I think we will continue to have very heavy CapEx for this year, not only this year, but also going into next year. So I think the negative cash flow situation will remain for some time. But at the same time, we're still maintaining a very healthy balance sheet, and we do have multiple cost-effective funding sources to fund the upcoming CapEx requirement.

是的，我想我們今年會持續維持非常高的資本支出，不只今年，明年也會延續。所以我認為負現金流的狀況還會維持一段時間。但同時，我們仍維持非常健康的資產負債表，而且我們也有多種具成本效益的資金來源，可以支應接下來的資本支出需求。

At this point, I think we're pretty confident that we will be able to support our growth in a healthy manner. We are still at the early stage of this mega trend. Like I said last time, we are not going to be shy in making the necessary investment, not just to support our customer, but also to maintain our clear leadership position in this field.

在這個時間點，我認為我們相當有信心，能夠以健康的方式支撐我們的成長。我們仍處於這個巨大趨勢的早期階段。就像我上次說的，我們不會在必要投資上畏縮，不僅是為了支援我們的客戶，也為了維持我們在這個領域清晰的領導地位。

Okay, thanks, Joseph. Yes, we very like to hear you have a more funding need. So last one is pricing and margin. So we keep hearing from your customers that you hike price aggressively in second half. Can we confirm this is the case, whether it's more price hike in a traditional service or the advanced packaging? I think that's part one.

好的，謝謝你，Joseph。是的，我們很樂於聽到你有更多資金需求。那最後一題是定價與毛利率。我們一直從你們的客戶那邊聽到，你們在下半年大幅調漲價格。我們能否確認確實如此？這是傳統服務的漲價比較多，還是先進封裝的漲價比較多？我想這是第一部分。

And secondly, how does that translate into your long-term gross margin outlook? I think you were having this 26% to 29% margin range for several years. Do you think you can break through the 30% any time soon? Thank you.

第二，這會如何反映在你們長期的毛利率展望上？我想你們過去幾年一直維持在 26% 到 29% 的毛利率區間。你認為短期內有機會突破 30% 嗎？謝謝。

Yes, I think it's, like I said in the outset, we will be -- we're very likely to exceed the margin ceiling in fourth quarter. And at that point of time, I think we will be -- we will start to review whether we should adjust the margin, structural margin range. Of course, when I say adjust, I mean upward adjustment.

是的，我想就像我一開始說的，我們——我們很可能在第四季超過毛利率上限。到了那個時點，我想我們會——我們會開始檢視是否需要調整毛利率的結構性區間。當然，我說的調整是指向上調整。

I think we are still in a very, very friendly pricing environment. I think, of course, we saw a lot of inflationary pressure in terms of materials and components so far. So far, we can definitely pass on these cost increases to our customers through our pricing arrangements. And of course, we'll continue to seek for the most suitable pricing strategy considering the situation as well as our margin return requirement.

我認為我們仍處於非常、非常友善的定價環境。當然，到目前為止，我們在材料與零組件方面看到不少通膨壓力。截至目前，我們絕對可以透過定價安排把這些成本上升轉嫁給客戶。當然，我們也會持續在考量情勢以及我們的毛利回報要求下，尋找最合適的定價策略。

Are you okay? Okay. So the next question, we will go back online.

你還好嗎？好。那下一個問題，我們回到線上。

Haas Liu, BoA.

Haas Liu，BoA。

Thank you so much for your time today. Yes, so just a few questions from me. I think first point is that when you commented on the gross margins, it's likely to exceed the ceiling of your structural gross margin range in fourth quarter. Would you be able to share with us how much of that would actually be contributing from your mature part of the business and how much of that would potentially be looked at by your growing mix in the LEAP contribution? Thank you.

非常感謝你今天撥冗。是的，我這邊有幾個問題。第一點是，當你談到毛利率時，你提到第四季很可能會超過你們結構性毛利率區間的上限。你是否能和我們分享，其中有多少是來自你們成熟業務的貢獻，又有多少可能是來自 LEAP 貢獻占比提升、產品組合成長所帶來的？謝謝。

Haas, you were kind of breaking up, but I think you're asking, what is contributing to our positive outlook in terms of maybe perhaps hitting our structural margins?

Haas，你的聲音有點斷斷續續，但我想你是在問：是什麼因素支撐我們在毛利率方面的正向展望，可能讓我們達到結構性毛利率？

Yes, that's correct, in the fourth quarter. And specifically, I wanted to know which part of the factory is more important. Is it the LEAP business contributing more, or is it because of the mature packaging and [advanced businesses], you are adjusting your price -- your pricing strategy to make that [organic]? Thanks.

是的，沒錯，是在第四季。而且我特別想知道，哪一部分的工廠更重要。是 LEAP 業務貢獻較多，還是因為成熟封裝與[先進業務]，你們調整了價格——定價策略來讓它更[有機]？謝謝。謝謝。

Haas is looking for the prime contribution for structural margins.

Haas 想了解結構性毛利率的主要貢獻來源。

Well, I think margin improvement is a combination -- is a result of many combination of different factors, including the margin-accretive business growth, including our improvement in efficiency, including the continuous expansion of our automated factories. Operating leverage doesn't -- of course, it plays a very important role as well as we continue to see volume growth.

我想毛利率改善是多種因素組合的結果——包括具毛利增益的業務成長、我們效率的提升、以及自動化工廠的持續擴張。隨著我們持續看到出貨量成長，營運槓桿當然也扮演非常重要的角色。

On the operating side, we are also seeing our passive investments start paying off. We are seeing -- on an annual basis, we're seeing OpEx ratio continue to drop. I think, overall, the efficiency is much, much improved and we have a much solid base in terms of our revenue coming on stream.

在營運面，我們也看到先前的被動投資開始帶來回報。我們看到——以年度來看，我們的營業費用（OpEx）比率持續下降。整體而言，效率大幅提升，而且在營收逐步到位方面，我們也有更穩固的基礎。

So that gives us the confidence that we should continue to see margin expansion on a sequential basis, at least for this year and next. And like I said, once we pass the structural margin range, we will start reviewing that and see how far we can go. Thank you.

因此，這讓我們有信心，至少在今年與明年，我們應該會持續看到毛利率逐季擴張。就像我說的，一旦我們超過結構性毛利率區間，我們就會開始檢視，看看能走到多遠。謝謝。

Yes, that's great. Thank you so much. And then just a quick follow up on your CapEx, would you be able to provide a breakdown? Since you also mentioned that the general market demand is also very strong as well, would you be able to provide a breakdown for your CapEx for this year?

是的，太好了。非常感謝。接著就資本支出（CapEx）快速追問一下，你是否能提供一個拆分？既然你也提到整體市場需求非常強勁，你是否能提供今年資本支出的拆分？

And for next year, I know you probably will not be able to give qualitative guidance at this stage for CapEx, but would you be able to discuss which part of the business is probably going to outgrow from the spending perspective? Is it going to be advanced or mature business is going to still be pretty solid from the spending? Thank you.

至於明年，我知道你們可能在這個階段無法就資本支出提供定性指引，但你是否能談談從支出角度來看，哪一部分業務可能成長更快？會是先進業務，還是成熟業務在支出上仍會相當穩健？謝謝。

Haas is looking -- you're looking to understand what the CapEx makeup is, what's driving the increase or what components are driving the increase this year, and then maybe if we have any type of nuggets of wisdom related to next year's CapEx.

Haas 想了解——你想了解資本支出的組成，今年增加的驅動因素或哪些項目推動了增加，然後也許我們是否能提供一些與明年資本支出相關的見解。

Well, I think, obviously we are seeing a stronger demand or demand forecast coming from not just customers, but also our foundry partner. And the requests that are coming in require a lot of the new investments, not just for this year, but also for next year as well as we continue to see very, very strong business momentum in both assembly and tests, I stress again.

我想，很明顯地，我們看到更強的需求或需求預測，不僅來自客戶，也來自我們的晶圓代工夥伴。而目前收到的需求，要求我們進行大量新投資，不僅是今年，明年也一樣；因為我們在封裝與測試兩方面都持續看到非常、非常強勁的業務動能，我再強調一次。

Also, there are new projects or new products, new process steps that we are entering into. We would not just put in the necessary equipment, CapEx. We also have to have the new facilities to house those capacity and also spending quite a bit of money for the R&D to support those new products or new projects that are coming on stream. So our CapEx is really based on what's necessary coming out of our customers' demand, and we want to put the right resources onto those high-potential revenue bases for us to continue to invest.

此外，我們也正在切入新的專案或新產品、新製程步驟。我們不只是投入必要的設備資本支出。我們也必須有新的廠房設施來容納這些產能，同時也要投入相當多的研發費用，以支援這些即將上線的新產品或新專案。因此，我們的資本支出確實是以客戶需求所帶出的必要性為基礎，而我們希望把正確的資源投入到那些高潛力的營收基礎上，讓我們能持續投資。

Okay, and your expansion going forward will probably be just pretty similar to your foundry here, right? You are going to probably focus more on the areas that you can convert capacity from one to the other. For example, you can both use it in the mature business as well as in the advanced business. So that is probably the area that you're spending more money.

好的，那你們未來的擴產大概也會跟你們的晶圓代工這邊很類似，對吧？你們可能會更聚焦在那些可以把產能從一種用途轉換到另一種用途的領域。例如，同一套產能既可以用在成熟業務，也可以用在先進業務。所以那大概就是你們投入更多資金的領域。

You're looking to -- well, Haas, we got to try to limit those questions to two in the future again. But I guess we can try to comment a little bit on where we're spending the CapEx as it relates to -- maybe this is an opportunity to talk about construction and our difficulties in terms of building buildings or getting enough buildings.

你想要——嗯，Haas，我們之後還是要盡量把問題限制在兩個。不過我想我們可以稍微評論一下我們的資本支出花在哪裡，這也許是個機會談談建廠，以及我們在蓋廠房或取得足夠廠房方面的困難。

Okay. But this year, we're raising the CapEx by another $2 billion. That brings up the total to about $10.5 billion. And out of this $10.5 billion, $4 billion will be for new factory buildings and facilities and $6.5 billion for equipment. Like Ken just mentioned, in terms of these new capacity that we're going to put in, we need to have the more advanced factory buildings and facilities to house this capacity.

好的。不過今年我們把資本支出再提高 20 億美元。這使得總額來到約 105 億美元。在這 105 億美元中，40 億美元將用於新廠房建築與設施，65 億美元用於設備。如 Ken 剛才提到的，針對我們將要新增的這些產能，我們需要更先進的廠房建築與設施來容納這些產能。

So at the same time, at this very moment, we are having 13 greenfield projects going on. We have another eight brownfield projects. By brownfield, we mean that we are buying existing factories and trying to renovate them to suit our needs. I think the current projects will be sufficient for us to carry ourselves into '28, maybe into some part of '29. And we will continue to look at the situation and find suitable new locations for further expansion going forward.

因此在同一時間、就在此刻，我們同時有 13 個綠地（greenfield）專案在進行中。我們另外還有 8 個棕地（brownfield）專案。所謂棕地，是指我們收購既有工廠，並嘗試翻新以符合我們的需求。我認為目前這些專案足以支撐我們一路到 2028 年，甚至可能延伸到 2029 年的某一段時間。而我們也會持續觀察情勢，並在未來尋找合適的新據點以進一步擴張。

And with these 20 projects going on at the same time, it puts a lot of challenge on us, puts a lot of pressure on us in terms of we really need to have a very, very efficient and very responsive construction partner for us to make sure everything is delivered on time, all the specs are -- all the quality is according to spec. And this is something that we are working on. And hopefully, I think Tien mentioned that execution is everything. Demand is there. What we need to do is really to execute whatever we're set out to do.

而這 20 個專案同時進行，對我們而言是很大的挑戰，也在壓力上帶來很大負擔；因為我們確實需要一個非常、非常高效率且反應非常迅速的建設合作夥伴，才能確保所有項目都能準時交付，所有規格——以及所有品質——都符合規格要求。這也是我們正在努力推進的事情。而且希望——我想田先生提到，執行力就是一切。需求是存在的。我們需要做的，就是把我們設定要做的事情真正執行到位。

Got it. Thank you so much, Joseph.

了解。非常感謝你，Joseph。

Do we have another question online there?

線上還有下一個問題嗎？

Gokul Hariharan, JPMorgan.

JPMorgan 的 Gokul Hariharan。

Yes, hi. Thanks for taking my follow-up questions. First of all, on CapEx, Joseph, I know that you're not guiding for next year. But looking at what Dr. Wu mentioned, it definitely feels like CapEx is still likely to keep rising into 2027. Is that a fair statement to make, given the gap between supply and demand that you're currently facing, especially for LEAP?

是的，你好。謝謝讓我追問幾個問題。首先關於資本支出（CapEx），Joseph，我知道你們不會對明年提供指引。但從吳博士提到的內容來看，感覺 CapEx 很可能仍會一路增加到 2027 年。考量你們目前面臨的供需缺口，尤其是針對 LEAP，這樣的說法是否合理？

Gokul, you're looking for some hint in terms of 2027?

Gokul，你是在尋求一些關於 2027 年的提示嗎？

CapEx, yes.

是的，CapEx。

Well, I think it's better if we -- I know it's going to be big, but how big? I think we want to wait for another quarter or so to see -- to have a better clarity on how much we would need to spend next year.

嗯，我想我們最好——我知道會很大，但到底有多大？我想我們希望再等一個季度左右，看看——以便更清楚明年需要花多少。

Okay, understood. Could you also talk a little bit about anything that you're seeing on the CoWoS or the panel-level packaging development based on your current assessment? When do you expect this to potentially start entering production, given that there are so many different views out there in the market?

好的，了解。你也能談一下，依你們目前的評估，在 CoWoS 或面板級封裝（panel-level packaging）的發展上，你們觀察到什麼嗎？考量市場上有很多不同看法，你們預期這可能何時開始進入量產？

Gokul, you're looking for an update on our panel processes?

Gokul，你是在詢問我們面板製程的最新進展嗎？

Yes, panel process.

是的，面板製程。

The different panel process for ASE's panel process, our fully automated line will start production by Q1 of next year. And that is in the form factor of 310 by 310. In terms of the CoWoS, I know there are many alternative materials that people are evaluating. Our R&D people are working with substrate supplier, foundry, as well as customer to evaluate the feasibility and also the economics of it. But right now, we do not have the glass substrate in production, not for the next 12 months if that's the question you're asking, thank you.

關於不同的面板製程，就 ASE 的面板製程而言，我們的全自動產線將在明年第一季開始量產。其外形尺寸（form factor）為 310 乘 310。至於 CoWoS，我知道目前大家在評估許多替代材料。我們的研發團隊正與載板供應商、晶圓代工廠以及客戶合作，評估其可行性以及經濟性。但就目前而言，我們沒有玻璃載板進入量產；如果你問的是接下來 12 個月內是否會量產，答案是不會，謝謝。

Got it. And how does the ASE's 310 by 310 solution differ? Is it for a completely different kind of market or customer compared to the foundry's CoWoS kind of solution? Or is it quite complimentary compared to what they are trying to offer?

了解。那 ASE 的 310 乘 310 解決方案有何不同？它是面向與晶圓代工廠 CoWoS 類解決方案完全不同的市場或客戶嗎？還是說它與他們想提供的方案其實是相當互補的？

They're quite complimentary. It's actually the same customer set; similar, the reticle size. In terms of the pitch size, line width, they're all identical, covered a similar range. But in terms of which customer will adopt what, not only depends on the capacity and the performance, and also the speed of execution.

它們相當互補。其實客戶群是相同的；光罩（reticle）尺寸也類似。就間距（pitch）尺寸、線寬而言，都一致，涵蓋的範圍也相近。但哪個客戶會採用哪一種，不僅取決於產能與效能，也取決於執行速度。

Okay. That's clear, yes. Thank you very much. Thanks.

好的。很清楚，是的。非常感謝。謝謝。

On CapEx, I want to give you guys a bit of a clarification. I think for this year, out of the total CapEx, like I said, $4 billion is for factory and facilities, $6.5 billion for equipment. And for equipment, I think it's about 56% is for assembly, 40% for test, and the remaining for EMS and some for material in terms of assembly and test breakdown. In terms of the leading edge, I think 70% of the equipment CapEx is for leading edge for this year.

關於 CapEx，我想再給各位一點釐清。我想今年在總 CapEx 裡，如我所說，40 億美元用於廠房與設施，65 億美元用於設備。而在設備方面，我想約 56% 用於封裝（assembly）、40% 用於測試（test），其餘用於 EMS，以及在封裝與測試拆分中也有一部分用於材料。就先進製程（leading edge）而言，我想今年設備 CapEx 的 70% 是用在先進端。

Gokul, we got you covered there? Yes, he's on. Okay. Another online question?

Gokul，這樣有回答到你的問題嗎？是的，他在線上。好的。還有下一個線上問題嗎？

Sunny Lin, UBS.

UBS 的 Sunny Lin。

Thank you very much for taking my follow ups. So my first follow-up will be on CPO. And so maybe good time if you could share with us, given the complexity of the technologies, what type of services ASE as a group will be able to offer. And then based on the current development, when do you think the revenue contribution for ASE could become more meaningful? And then given USI, they also acquired an optical module company as well earlier. And so what type of synergies that you think you could drive since you have IC ATM and also EMS capabilities?

非常感謝讓我追問。我第一個追問是關於 CPO。鑑於技術的複雜性，也許是個好時機請你們分享，ASE 集團能提供哪些類型的服務。另外，基於目前的發展，你們認為 ASE 的營收貢獻何時會變得更有意義？再者，USI 之前也收購了一家光模組公司。因此，考量你們同時具備 IC ATM 與 EMS 能力，你們認為可以帶來哪些綜效？

Sunny, you're looking for an update on what particular we would be doing in regards to CPO and also potential linkages with our EMS business. Is that correct?

Sunny，你是在詢問我們在 CPO 方面具體會做什麼，以及與我們 EMS 業務之間可能的連結，是嗎？

Yes, thank you.

是的，謝謝。

On the CPO, I would like to wait for two quarters before I give you more detail. The CPO service, as was the revenue, also the 2027 outlook, I prefer to have a little bit more time to talk about that. In terms of the optical hierarchy, the optical -- the hybrid between the electrical signal and also the optical signal, I think that direction is definitive.

關於 CPO，我希望再等兩個季度再給你更多細節。CPO 的服務、營收，以及 2027 年展望，我希望有多一點時間再來談。就光學層級而言，也就是電訊號與光訊號之間的混合（hybrid），我認為那個方向是明確的。

The question now is when and how could we execute that at the system architecture from near field to outer field? Some optical devices are quite mature. It has been used for many, many years. And some near field, at a chip level, at a substrate level, that needs to be created, which is very, very difficult, which is why the industry has taken so much time trying to develop.

現在的問題是，我們何時、以及如何在系統架構上，從近場到遠場把它執行出來？有些光學元件相當成熟。它已經使用很多很多年了。但有些近場的、在晶片層級、在載板層級的部分，需要被創造出來，這非常、非常困難，也因此產業花了這麼長時間在開發。

I think by the end of the year, we'll have some database in terms of how that is behaving, how that is working. And whether -- how much benefit? How much ramp up? I think in two quarters' time. By the way, we have been working on this for about 20 years. So two more quarters, I think we can wait.

我想在今年年底前，我們會有一些資料庫（database），用來了解它的表現如何、運作如何。以及——能帶來多少效益？能拉升多少量？我想兩個季度後會更清楚。順帶一提，我們在這方面已經投入大約 20 年。所以再等兩個季度，我想我們可以等。

Sure, looking forward to that.

好的，期待。

Thank you.

謝謝。

Yes, looking forward to that update. And my second question will be very quick. So for your full-process CoWoS, given you have better visibility now for 2027, so want to understand your progress in terms of diversifying for your client, also product mix. Should we assume for 2027 CPU should be very major, or should we assume good volume coming from the other applications like accelerators as well?

是的，也期待那個更新。我的第二個問題會很快。關於你們的全流程 CoWoS，既然你們現在對 2027 年有更好的能見度，我想了解你們在客戶多元化以及產品組合方面的進展。我們是否應該假設 2027 年 CPU 會占非常主要的比重，還是也應該假設其他應用（例如加速器）也會帶來不錯的量？

Sunny, you're looking for probably something that can't be answered. But in terms of you're looking for what our product set would be in terms of our full process services, is that --

Sunny，你問的可能是個無法回答的問題。但就你想了解我們在全流程服務方面的產品組合而言，--

Yes, basically your progress in terms of product-based diversification for full process.

是的，基本上是你們在全流程、以產品為基礎的多元化方面的進展。

I think for the full process, we should be able to give you a better visibility in a quarter or two. Right now, we do have line of sight in terms of the full process. We're tracking nicely. And also for the next year, we do have line of sight. We have full visibility in terms of capacity that we're developing.

我認為就全流程而言，我們應該能在一到兩個季度內給你更清楚的能見度。目前我們對全流程已有掌握。我們追蹤進度良好。另外，對於明年我們也有掌握。我們對正在建置的產能有完整的可視性。

Now in terms of the customers, that is the -- we need a little bit more time to digest how much information can we share I think in two quarters' time, we should be able to give you the following. Our leading edge, our lead services revenue for next year, and also the OS, full process and others, the assembly and test, I think that we can share.

至於客戶方面，這部分——我們需要多一點時間消化哪些資訊可以分享；我想在兩個季度後，我們應該能提供你以下內容。我們明年的先進製程、先進服務營收，以及作業系統（OS）、全流程與其他項目，包括組裝與測試，我想這些我們都可以分享。

But in terms of GPU, CPU, ASIC, we need a little bit more time to digest how much information we can share. But right now, our clientele portfolio covers all of them, which is good. And we're just waiting for a clear execution such that we know our yield and also how well the system performs in the marketplace, and that we're waiting anxiously, thank you.

但就GPU、CPU、ASIC而言，我們還需要多一點時間消化哪些資訊可以分享。不過目前我們的客戶組合涵蓋上述所有類別，這是好事。我們正等待明確的執行結果，讓我們了解良率，以及系統在市場上的表現如何；我們也正焦急地等待，謝謝。

Thank you very much.

非常感謝。

Our next online question is from [Michael Rasnikas].

我們下一個線上提問來自[Michael Rasnikas]。

Hi, how are you? Thank you for the call here. I just want to clarify, when you gave Q3 guidance, did you say that EMS will grow 40% quarter-on-quarter revenues?

嗨，你好嗎？謝謝這通電話。我想確認一下，你們給出第三季指引時，是不是說EMS營收將季增40%？

Michael, you're looking for a little bit of explanation behind the EMS growth?

Michael，你是想了解EMS成長背後的一些說明嗎？

Yes, so it's a pretty huge quarter-on-quarter growth. So yes, I just want to clarify what's going on there.

是的，因為這是相當大的季增幅。所以我想確認一下那邊的情況。

I think the 40% growth, it is a little bit abnormal seasonality. I think this is largely because of the component price hikes, particularly in the memory sector. So if we take that part of the number out, I think the third-quarter EMS should see a typical seasonality kind of movement.

我認為這個40%的成長，帶有一些不太尋常的季節性因素。主要是因為零組件價格上漲，特別是在記憶體領域。所以如果把那部分數字扣除，我認為第三季的EMS會呈現較典型的季節性變動。

Makes sense. And how about Q4 in terms of quarter on quarter? Is it still growing over Q3 or is there a slow -- will it shrink? Will revenues -- EMS shrink -- revenue shrink in Q4 versus Q3?

了解。那第四季相較第三季呢？還會繼續成長，還是會放緩——會下滑嗎？第四季相較第三季，EMS營收會不會縮減？

At this point, we're seeing a pretty similar level of revenue in the fourth quarter for EMS.

就目前來看，我們看到第四季EMS的營收水準會相當接近。

Okay, that's very helpful. So overall for the year, did you say in your overall comments that you're looking for 25% plus full-year growth, revenue growth?

好的，這很有幫助。那整體來看，你們在總體評論中是否提到全年營收成長會是25%以上？

Michael, you're looking for full-year guidance or full-year outlook for EMS and ATM?

Michael，你是想問EMS和ATM的全年指引或全年展望嗎？

Yes, overall revenue growth. I thought there was a comment about sort of 25%-plus type of revenue growth. Is that right?

是的，整體營收成長。我以為有提到大概25%以上的營收成長。是這樣嗎？

EMS at this point, I think during the full year is a sub 20% kind of growth for the year.

就EMS而言，目前我認為全年會是低於20%的成長。

Sub 20%?

低於20%？

Correct.

沒錯。

Okay, so that's sub 20%, and the ATM business will be 35% plus type of growth.

好的，所以EMS是低於20%，而ATM業務會是35%以上的成長。

Also correct.

也正確。

Okay, got it. And what is the impact on EMS margins with the big memory component and larger revenues? Is it flat year over year or is there an increase or decrease?

好的，了解。那記憶體零組件占比很大、營收也更高，對EMS毛利率的影響是什麼？年對年持平，還是上升或下降？

I can talk about third quarter. I think if we take out the component price hikes, I think the operating margin will be very similar to a typical seasonality of around 3.7%, 3.8% level.

我可以談第三季。我認為如果把零組件漲價因素扣除，營業利益率會非常接近典型季節性水準，大約在3.7%、3.8%左右。

Okay. And so the memory is just sort of a pass-through type of cost for you guys?

好的。所以記憶體基本上對你們來說只是類似轉嫁（pass-through）的成本？

That's correct.

沒錯。

Got it. Okay. Thank you for that clarification, and congratulations on a very strong performance.

了解。好的。謝謝你幫忙釐清，也恭喜你們表現非常強勁。

Thank you.

謝謝。

Do we have more questions online? More questions on the floor? No? Okay, very good. Thank you very much. I would like to thank everyone for attending our conference call today. Joseph, do you want to close up?

線上還有更多問題嗎？現場還有更多問題嗎？沒有？好的，非常好。非常感謝。我要感謝各位今天參加我們的電話會議。Joseph，你要做結語嗎？

Well, I'm sure we're going to have another good quarter in third quarter, and we'll bring you some more new good news next quarter. We'll see you next quarter.

我相信第三季我們還會有另一個不錯的季度，我們下個季度會再帶給各位更多新的好消息。我們下個季度見。