泰瑞達 (TER) 2019 Q1 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Teradyne Q1 2019 Earnings Conference Call. (Operator Instructions) As a reminder, this conference call is being recorded.

各位女士、各位先生，大家好，歡迎參加 Teradyne 2019 年第一季財報電話會議。(接線員指示) 提醒各位，本次電話會議將被錄音。

I would now like to introduce your host for today's conference, Mr. Andy Blanchard, Vice President of Corporate Communications. Sir, you may begin.

現在我想介紹今天會議的主持人：企業傳播副總裁 Andy Blanchard 先生。先生，您可以開始了。

Thank you, Dan. Good morning, everyone, and welcome to our discussion of Teradyne's most recent financial results. I'm joined this morning by our CEO, Mark Jagiela; and CFO, Greg Beecher. Following our opening remarks, we'll provide details of our performance for 2019's first quarter with our outlook for the second quarter of 2019 as well.

謝謝你，Dan。各位早安，歡迎參加我們對 Teradyne 最新財務結果的討論。今天早上與我一同出席的有執行長 Mark Jagiela，以及財務長 Greg Beecher。在開場致詞之後，我們將說明 2019 年第一季的表現，並同時提供 2019 年第二季的展望。

The press release containing our first quarter results was issued last evening. We're providing slides on the investor page of the website that may be helpful to you in following this discussion. Replays of this call will be available via the same page after the call ends.

包含第一季結果的新聞稿已於昨晚發布。我們在網站的投資人頁面提供投影片，可能有助於各位跟進本次討論。本次電話會議結束後，也可透過同一頁面收聽重播。

The matters that we discuss today will include forward-looking statements that involve risk factors that could cause Teradyne's results to differ materially from management's current expectations. We encourage you to review the safe harbor statement contained in our earnings release as well as our most recent SEC filings. Additionally, those forward-looking statements are made as of today, and we take no obligation to update them as a result of developments occurring after this call.

我們今天討論的事項將包含前瞻性陳述，其中涉及風險因素，可能導致 Teradyne 的結果與管理層目前的預期出現重大差異。我們鼓勵各位查閱財報新聞稿中的安全港聲明，以及我們最近向 SEC 提交的文件。此外，這些前瞻性陳述係以今日為準，我們不承擔因本次電話會議後發生之發展而更新其內容的義務。

During today's call, we'll make reference to non-GAAP financial measures. We've posted additional information concerning these non-GAAP financial measures, including reconciliation to the most directly comparable GAAP financial measure where available, on the investor page of the website.

在今天的電話會議中，我們將提及非 GAAP 財務衡量指標。我們已在網站投資人頁面張貼關於這些非 GAAP 財務衡量指標的補充資訊，包括在可取得的情況下，與最直接可比的 GAAP 財務衡量指標之調節表。

Also, between now and our next earnings call, Teradyne will be participating in investor conferences hosted by Baird, BofA, Bernstein, Cowen, Stifel and UBS.

另外，在下一次財報電話會議之前，Teradyne 將參加由 Baird、BofA、Bernstein、Cowen、Stifel 及 UBS 主辦的投資人會議。

Now let's get on to the rest of the agenda. First, Mark will comment on our recent results and the market conditions as we enter the second quarter. Greg will then offer more details on our quarterly results along with our guidance for the second quarter. We'll then answer your questions. And this call is scheduled for 1 hour.

現在我們進入其餘議程。首先，Mark 將評論我們近期的結果，以及在進入第二季時的市場狀況。接著 Greg 將就本季結果提供更多細節，並說明第二季的財測指引。之後我們將回答各位的問題。本次電話會議預計為 1 小時。

Mark?

Mark？

Hello, everyone, and thanks for joining us this morning. My prepared remarks will provide a summary of our recent results, describe current conditions in the markets we serve and provide an update on how we're looking at the full year. Greg will then take you through more details on the quarter.

各位好，感謝各位今天早上加入我們。我準備的發言將概述我們近期的結果，說明我們所服務市場的現況，並更新我們對全年展望的看法。接著 Greg 會帶各位深入了解本季的更多細節。

Our first quarter sales of $494 million and non-GAAP profits of $0.54 per share came in ahead of our January guidance. This was due to a few strong pockets of notable demand. First, image sensor tester sales were the highest on record as the proliferation of cameras and smartphones, automotive, industrial and security applications continues to expand rapidly. Second, semiconductor test capacity for devices used in 5G base stations and related infrastructure began to ramp. This was primarily for high-performance digital infrastructure processors and represents the very early phase of what is expected to be a multiyear ramp-up capacity for a variety of 5G technologies. And third, in Industrial Automation, we saw the beginning of some large deployments of UR cobots at enterprise-level accounts in automated assembly applications.

我們第一季銷售額為 4.94 億美元，非 GAAP 每股盈餘為 0.54 美元，均優於我們 1 月的指引。這主要是因為幾個需求特別強勁的領域。第一，影像感測器測試機的銷售創下歷史新高，因為相機的普及以及智慧型手機、車用、工業與安防等應用持續快速擴張。第二，用於 5G 基地台及相關基礎設施之裝置的半導體測試產能開始爬坡。這主要針對高效能數位基礎設施處理器，並代表預期將持續多年的各類 5G 技術產能擴張之非常早期階段。第三，在工業自動化方面，我們看到 UR 協作機器人在企業級客戶的自動化組裝應用中，開始出現一些大型部署。

Elsewhere in Semi Test, aside from our one large account, demand for mobile device testing remains high and trending ahead of 2018 levels. At our large account, we expect another downtick in demand this year but overall in line with our annual guidance.

在半導體測試的其他領域，除了我們的一家大型客戶之外，行動裝置測試需求仍然強勁，且趨勢高於 2018 年水準。在該大型客戶方面，我們預期今年需求將再小幅下滑，但整體仍符合我們的年度指引。

Demand for RF test capacity for legacy standards is low, but interest in RF test capacity for next-generation ultra-wideband, WiFi and 5G standards is growing and should expand throughout the year.

傳統標準的 RF 測試產能需求偏低，但對下一代超寬頻（UWB）、WiFi 與 5G 標準之 RF 測試產能的興趣正在升溫，並應在全年持續擴大。

In analog test, after a record 2018 for our Eagle product line, shipments for the automotive, industrial and consumer markets cooled in the quarter and came in about as expected. We expanded our footprint in the high-powered discrete module test with the acquisition of Lemsys early in the quarter, which strengthens our position in the electrification trends of vehicle, solar, wind and industrial applications.

在類比測試方面，繼我們 Eagle 產品線於 2018 年創下歷史新高之後，本季車用、工業與消費市場的出貨降溫，結果大致符合預期。我們在本季初收購 Lemsys，擴大了在高功率分離式模組測試領域的布局，強化我們在車輛電動化、太陽能、風電及工業應用等趨勢中的地位。

In the memory test side of the market, our sales were strong and level with the fourth quarter at $48 million. Demand was well distributed across flash package test and flash and DRAM wafer test, and we expect a slight uptick in Q2 driven by high-speed protocol test of flash devices.

在記憶體測試市場方面，我們的銷售表現強勁，與第四季持平，為 4,800 萬美元。需求分布於快閃記憶體封裝測試，以及快閃與 DRAM 晶圓測試；我們預期第二季將小幅成長，主要由快閃裝置的高速協定測試所帶動。

In Wireless Test at LitePoint, sales were down sequentially but up 29% compared with Q1 of 2018 as we're seeing early step-ups in buying for ultra-wideband, WiFi 6 and 5G.

在 LitePoint 的無線測試方面，銷售較前一季下滑，但較 2018 年第一季成長 29%，因為我們看到超寬頻（UWB）、WiFi 6 與 5G 的採購開始出現早期增溫。

In System Test, sales were up both sequentially and year-on-year. The sequential increase was due primarily to growth in our hard disk drive test demand as test times for high-capacity drives continue to expand with density.

在系統測試方面，銷售無論較前一季或較去年同期皆成長。其中，季增主要來自硬碟測試需求的成長，因為隨著儲存密度提升，高容量硬碟的測試時間持續拉長。

In Industrial Automation, sales grew 35% from Q1 '18 driven by both UR robot growth of 16% in the quarter and the addition of MiR and Energid. Three points are noteworthy in these results. First, as noted earlier, we are beginning to see a mix of larger deployments of UR cobots. As we described in past calls, we've been increasing our sales investment at enterprise-level accounts, and it's encouraging to see early results of this investment paying off. Following a 6- to 9-month period of evaluation, 2 of these large customers are now moving forward with deployments at a rate of dozens per month that should extend through the rest of the year. One of these customers is concentrated in China, and the other is geographically distributed.

在工業自動化方面，銷售較 2018 年第一季成長 35%，主要由 UR 機器人在本季成長 16% 以及併入 MiR 與 Energid 所帶動。這些結果有三點值得注意。第一，如先前所述，我們開始看到 UR 協作機器人出現更多大型部署的組合。如我們在過去的電話會議中所描述，我們一直在企業級客戶加大銷售投資，而看到這項投資開始產生早期成果，令人鼓舞。在經過 6 到 9 個月的評估期後，其中 2 家大型客戶目前正以每月數十台的速度推進部署，且應會延續至今年其餘時間。其中一家客戶主要集中在中國，另一家則在地理上分布更為分散。

Second, the arsenal of certified products in our UR+ program continues to grow and uniquely reduces the application's development time. We began the UR+ program in 2016 and by the end of 2017 had about 60 solutions available. 1 year later, it was over 130 solutions, and by the end of this year, we expect over 200. These products range from cameras to grippers to welders and screwdrivers and much more. A key part of our strategy is to push the competitive boundary in our price and hardware performance. Our UR+ playbook is very clear, provide a software and hardware platform to allow the development of a broad range of peripherals that are easy to deploy, reliable and have a short ROI. In doing so, we are able to leverage the talent of over 400 global UR+ partners to address more of the market than we could alone. This provides our customers with a range of solutions that gives them the flexibility to meet their immediate requirements and have the flexibility to redeploy our cobots to handle other tasks they may encounter. This is a significant distinction between UR and other emerging cobot competitors, large and small.

第二，我們 UR+ 計畫中通過認證的產品組合持續擴大，並能以獨特方式縮短應用開發時間。我們於 2016 年啟動 UR+ 計畫，至 2017 年底約有 60 種解決方案可用。一年後已超過 130 種解決方案，而到今年年底，我們預期將超過 200 種。這些產品涵蓋從相機、夾爪到焊接設備、螺絲起子等，以及更多其他品項。我們策略的一個關鍵部分，是在價格與硬體效能上推進競爭邊界。我們的 UR+ 作戰手冊非常清楚：提供一個軟硬體平台，使各式周邊設備得以開發，並且易於部署、可靠且投資回收期（ROI）短。藉此，我們能運用全球超過 400 家 UR+ 合作夥伴的人才，觸及比我們單獨能做到更大的市場。這為客戶提供一系列解決方案，使其能彈性滿足即時需求，並可彈性重新部署我們的協作機器人，以處理其可能遇到的其他任務。這是 UR 與其他新興協作機器人競爭者（無論規模大小）之間的一項重要差異。

Third, a year ago, MiR joined Teradyne, and we're delighted with the results. After growing 163% last year, we are planning to about double again this year. Earlier this month, we introduced 2 significant new products to fuel this growth. One is a 1,000-kilogram payload, pallet-ready mobile robot, and the other is an AI-enabled camera that feeds into our fleet management software to provide better navigation. Think of it as a Waze-type augmented information system for mobile cobots where cameras monitor congestion points in a factory and use machine learning to classify and inform our fleet-wide navigation engine. This advanced warning system provides for increased navigation efficiency and safety.

第三，一年前 MiR 加入 Teradyne，我們對成果感到非常滿意。在去年成長 163% 之後，我們計劃今年再約翻倍成長。本月稍早，我們推出 2 項重大新產品以推動這項成長。其中一項是可承載 1,000 公斤、可直接處理棧板的移動式機器人；另一項則是具備 AI 的相機，可與我們的車隊管理軟體整合，以提供更佳的導航能力。你可以把它想成是用於移動協作機器人的類 Waze 增強資訊系統：相機監控工廠中的壅塞點，並運用機器學習進行分類，將資訊提供給我們的全車隊導航引擎。這套先進的預警系統可提升導航效率與安全性。

Shifting to our outlook. The market conditions we're seeing in both our Test and IA businesses are essentially unchanged from what we saw at the start of the year. In Semi Test, our full year outlook for the SOC market remains in the $2.3 billion to $2.7 billion range, down about 17% from the midpoint of last year.

接著談我們的展望。我們在測試（Test）與工業自動化（IA）業務所看到的市場狀況，基本上與年初所見並無二致。在半導體測試（Semi Test）方面，我們對 SOC 市場的全年展望仍維持在 23 億至 27 億美元區間，較去年中位數約下滑 17%。

In Memory, given the well-reported slowdown across the market, we are reducing our full year market estimate by about $50 million to $600 million to $700 million, down about 30% at the midpoint from 2018. Offsetting this, we expect LitePoint and our System Test markets to be up about 10% from last year based on early-stage adoption of new wireless standards and a continuation of System Test trends seen in Q1.

在記憶體（Memory）方面，鑑於市場普遍報導的放緩，我們將全年市場規模估計下調約 5,000 萬美元，至 6 億至 7 億美元區間；以中位數計，較 2018 年約下滑 30%。作為抵銷因素，我們預期 LitePoint 與系統測試（System Test）市場將較去年成長約 10%，主要受惠於新無線標準的早期採用，以及第一季所見系統測試趨勢的延續。

In Industrial Automation, we continue to see some of the same headwinds that began to emerge last summer in China and in the automotive supply chain in general. Balancing those headwinds is an internal data showing accelerating monthly growth in IA shipments through March. For the full year, we're keeping our IA growth projection at 35% to 40% and the longer-term rate at 30% to 40%.

在工業自動化方面，我們仍看到一些自去年夏天開始在中國以及整體汽車供應鏈浮現的逆風。與這些逆風相平衡的是，內部數據顯示 IA 出貨量至 3 月為止呈現每月加速成長。就全年而言，我們維持 IA 成長預測在 35% 至 40%，長期成長率則維持在 30% 至 40%。

Summing things up. In Test, the demand drivers of unit growth and device complexity remain in place across all of our test markets. For example, we're beginning to see 5G-related test demand in both Semi Test and at LitePoint. Recall, when 5G-millimeter wave is adopted in mainstream handsets, we expect the test -- related test volume to represent about $300 million to $400 million of the Semi Test TAM and about

總結來說。在測試業務方面，單位出貨成長與裝置複雜度提升等需求驅動因素，仍在我們所有測試市場中持續存在。例如，我們開始在半導體測試與 LitePoint 兩端看到與 5G 相關的測試需求。請記得，當 5G 毫米波被主流手機採用時，我們預期相關測試量將約占半導體測試 TAM 的 3 億至 4 億美元，以及約

(technical difficulty)

（技術問題）

We will be ramping to those levels over the next 3 years or so as deployments of 5G roll out globally. So despite year-to-year volatility, we remain very optimistic about the systemic sector growth in Test.

隨著 5G 在全球部署推進，我們將在未來約 3 年逐步提升至上述水準。因此，儘管年度間可能波動，我們對測試業務的結構性產業成長仍然非常樂觀。

In IA, we've got a great product lineup that's well aligned to global demographic, quality and economic trends, which we expect to drive high growth for years to come. Our next-generation intelligent automation capabilities are reshaping the nature of industrial activity worldwide. We are investing in these software-rich, technology-driven building blocks of the future because we believe we are very early in the transition from centralized, complex and costly automation, usable by only a small percentage of global manufacturers, toward a distributed, easy-to-implement, low-cost automation that's available to nearly all industrial companies regardless of their size or automation proficiency.

在 IA 方面，我們擁有優異的產品組合，與全球人口結構、品質與經濟趨勢高度契合，我們預期這些趨勢將在未來多年推動高速成長。我們的下一代智慧自動化能力正在重塑全球工業活動的樣貌。我們投資於這些以軟體為核心、技術驅動的未來基礎模組，因為我們相信：從僅少數全球製造商可用、集中式、複雜且昂貴的自動化，轉向分散式、易於導入、低成本、幾乎所有工業公司不論規模或自動化能力皆可使用的自動化——這場轉型仍處於非常早期的階段。

In closing, I'll note that as previously announced, Sanjay Mehta will join Teradyne as our next CFO effective tomorrow. Sanjay brings a wealth of experience in finance, the semiconductor industry and China operations to Teradyne. At the same time, Greg has served 18 years as Teradyne's CFO and has been instrumental to our evolution as a company. This will be Greg's last earnings call, so I'll turn it over to him for details and final thoughts.

最後我補充說明，如先前所宣布，Sanjay Mehta 將自明日起加入 Teradyne，擔任我們下一任 CFO。Sanjay 在財務、半導體產業以及中國營運方面擁有豐富經驗，將為 Teradyne 帶來助益。同時，Greg 擔任 Teradyne CFO 已 18 年，對公司演進至關重要。這將是 Greg 最後一次財報電話會議，因此我把時間交給他，請他說明細節並做最後的分享。

Thanks, Mark, and good morning, everyone. I'll provide some key highlights, cover UR's market advantage and longer-term outlook, then I'll move to the first quarter results and second quarter outlook.

謝謝你，Mark，各位早安。我將提供幾項重點摘要，說明 UR 的市場優勢與長期展望，接著談第一季業績與第二季展望。

First, though, our first quarter sales of $494 million and non-GAAP EPS of $0.54 came in slightly above the top end of our guidance, and EPS was 20% above our year-ago start. The EPS year-over-year improvement was principally due to favorable product mix and share buybacks, offset somewhat by higher Industrial Automation OpEx, including having MiR and Energid in our 2019 first quarter results.

不過首先，我們第一季銷售額 4.94 億美元與非 GAAP 每股盈餘（EPS）0.54 美元，均略高於財測區間上緣，而 EPS 較去年同期起點高出 20%。EPS 年增主要來自有利的產品組合與庫藏股回購，但部分被較高的工業自動化營運費用（OpEx）所抵銷，其中包括 MiR 與 Energid 納入我們 2019 年第一季業績。

Our second quarter guidance for sales of $520 million to $550 million with non-GAAP EPS of $0.56 to $0.65 shows sequential 8% revenue and 12% non-GAAP EPS growth at the midpoint. That has us tracking to similar first half sales with non-GAAP EPS up about 10% compared with a year ago.

我們對第二季的財測為：銷售額 5.20 億至 5.50 億美元，非 GAAP EPS 為 0.56 至 0.65 美元；以中位數計，營收季增 8%，非 GAAP EPS 季增 12%。這使我們上半年銷售額走勢與去年同期相近，而非 GAAP EPS 較去年同期約成長 10%。

While there are pockets of Semi Test buying in support of 5G infrastructure and image sensor for handsets and security, as Mark noted, annual semiconductor units are forecasted to be well under the 10%-plus growth rate of the prior 2 years. So we're conservatively planning for a flat to slightly softer second half in Semi Test albeit with considerable uncertainty.

雖然在半導體測試方面，仍有部分採購需求用於支援 5G 基礎建設，以及手機與安防用的影像感測器；但如 Mark 所提，年度半導體出貨量預測將遠低於前兩年 10% 以上的成長率。因此，我們保守規劃半導體測試在下半年持平至略為走弱，儘管不確定性仍相當高。

Second half growth Industrial Automation should fill the gap so that the first and second halves should be pretty symmetrical from a revenue and EPS perspective.

工業自動化在下半年的成長應可補上缺口，使上、下半年在營收與 EPS 的表現上大致對稱。

Semi Test volatility has been the norm for many years, so we long ago structured our operations with these swings in demand. For example, in the current decade, our non-GAAP operating profit rate has averaged 23% over the last 9 years and swung from a high of 28% in 2010 to a low of 18% in 2013. In the last 3 years, with Industrial Automation in the fold, we've averaged a 24% non-GAAP operating profit rate and $420 million in annual free cash flow. As a result, we don't need to spend resources on judging where we are in a particular cycle; rather, we keep our focus on our midterm growth plan.

半導體測試的波動多年來一直是常態，因此我們早已依據這些需求起伏來建構營運模式。例如在本十年，我們過去 9 年的非 GAAP 營業利益率平均為 23%，並在 2010 年高達 28%、2013 年低至 18% 之間擺動。在過去 3 年，隨著工業自動化納入版圖，我們的非 GAAP 營業利益率平均為 24%，年度自由現金流約 4.2 億美元。因此，我們不需要耗費資源去判斷目前處於哪一個景氣循環階段；相反地，我們持續專注於中期成長計畫。

As outlined last quarter, we set our sights on reaching $3.50 to $4 in EPS by 2022, which requires Test sales growth of 3% to 5% off of 2018 and Industrial Automation growth of 30% to 40% off 2018. We're confident in both of these assumptions and are executing our plans accordingly despite the bumpy ride.

如上季所述，我們目標是在 2022 年達到每股盈餘 3.50 至 4.00 美元，這需要測試業務銷售額相較 2018 年成長 3% 至 5%，以及工業自動化相較 2018 年成長 30% 至 40%。我們對這兩項假設都很有信心，並在顛簸的環境下持續按計畫執行。

Turning to Industrial Automation highlights. We had first quarter sales of $66 million or 35% growth over the first quarter of a year ago but down sequentially due to normal seasonality as distributors often buy a bit more in the fourth quarter to max out on available discounts. This typically does not affect second quarter sales buying as fourth quarter buying is burned off by mid-first quarter. So although there were cyclical headwinds across global industrial markets, we see strong secular growth looking ahead.

接著談工業自動化的重點。我們第一季銷售額為 6,600 萬美元，較去年同期第一季成長 35%，但因正常季節性因素而較前一季下滑，因為經銷商常在第四季多買一些，以最大化可取得的折扣。這通常不會影響第二季的採購，因為第四季的採購量會在第一季中期前消化完畢。因此，儘管全球工業市場存在循環性逆風，我們仍看好未來的強勁結構性成長。

Universal Robots had a 60% gross margin this quarter, up from 51% in 2015 principally due to synergies from Teradyne, which I'll touch on later. At UR, we continue to see very high interest from SMEs for existing applications such as machine tending and packaging, but we're also encouraged by the large order sizes that Mark noted. This wave of demand is driven by high operator turnover rates along with ongoing pressure on labor costs.

Universal Robots 本季毛利率為 60%，高於 2015 年的 51%，主要歸因於來自 Teradyne 的綜效，我稍後會再談到。在 UR，我們持續看到中小企業（SMEs）對既有應用（例如機台上下料與包裝）保持高度興趣，同時我們也對 Mark 提到的大型訂單規模感到鼓舞。這波需求主要由操作員高流動率，以及持續的勞動成本壓力所驅動。

Another wave that we expect is bin picking, which has been referred to as the Holy Grail for manufacturers. Later this year, we plan to release a bin picking solution that combines UR's easy-to-train model with Energid's pathfinding and motion-control software allowing for precise part placement after the pick. More on this in future calls.

我們預期的另一波需求是料箱取料（bin picking），這一直被稱為製造商的「聖杯」。今年稍晚，我們計畫推出一套料箱取料解決方案，結合 UR 易於訓練的模型與 Energid 的路徑規劃與運動控制軟體，使取料後能精準放置零件。未來的電話會議將再分享更多內容。

Turning now to MiR, which continues to be a standout performer. After the very successful launch of the MiR500 last year, MiR has just delivered an encore with its new MiR 1,000-kilo product introduced this month. These 2 heavier payload products can do much of the transport work of a traditional forklift but more safely and at a lower cost. They self-navigate and learn their environment, can be called or dispatched from a smartphone, tablet or fully automated through an ERP system. And they work seamlessly with their sister, AMRs and a fleet. MiR is targeting about $60 million of sales in 2019, up from $31 million for the full stand-alone year in 2018.

接著談 MiR，其持續是表現突出的業務。在去年非常成功推出 MiR500 之後，MiR 本月再度帶來新作，推出新的 MiR 1,000 公斤產品。這兩款較高載重的產品可承擔傳統堆高機的大部分運輸工作，但更安全且成本更低。它們可自主導航並學習環境，可透過智慧型手機、平板電腦呼叫或派遣，或透過 ERP 系統實現全自動化。並且可與其姊妹機型、AMR 以及車隊無縫協作。MiR 目標在 2019 年達成約 6,000 萬美元銷售額，高於 2018 年獨立完整年度的 3,100 萬美元。

Mark noted the key highlights in Test, so I'll note a few things that may help you in your full year modeling. In Semi Test, 2019 will likely be a year of digestion, with sales declining 5% to 10%. Highlighting the strength of our operating model, we expect to deliver 20% plus operating margins. For the other test businesses, at LitePoint and System Test combined, we expect annual revenue to grow 10% or more with profit growth of 20% plus from last year.

Mark 已說明測試（Test）業務的重點亮點，因此我補充幾點，或許能協助各位進行全年模型推估。在半導體測試（Semi Test）方面，2019 年很可能是消化調整的一年，銷售額將下滑 5% 至 10%。凸顯我們營運模式的強勁，我們預期可達成 20% 以上的營業利益率。至於其他測試業務，LitePoint 與 System Test 合計，我們預期年營收成長 10% 或以上，且相較去年獲利成長 20% 以上。

Before I get to the financial details of the quarter, let me address some frequently asked questions on the cobot market size and Universal Robots market advantage. First, UR has just scratched the surface of the opportunity for cobots. We proved the industry has cumulatively sold about 60,000 cobots to date against an estimated opportunity of a few million given today's cobot capabilities. New capabilities like bin picking expand the market by 50% or more, and there's a nice pipeline of new solutions planned that will continue to drive the TAM higher in the years ahead.

在我進入本季財務細節之前，先回應一些關於協作機器人（cobot）市場規模與 Universal Robots 市場優勢的常見問題。首先，UR 在協作機器人的機會上才剛觸及表面。我們證實，迄今全產業累計約售出 60,000 台協作機器人；以今日協作機器人的能力推估，潛在機會約為數百萬台。像料箱取料等新能力可將市場擴大 50% 或以上，而且我們規劃中的新解決方案管線相當可觀，未來幾年將持續推升 TAM（總可服務市場）。

So the next question is why isn't it easy for competitors to gain traction in this growing market, too. First, as you might expect, the traditional robot companies seem to be focused on their large installed base customers who value compatibility with their high-payload complex robots. It's a very big challenge to leverage their platforms designed for automation experts into an easy-to-use or flexible cobot designed for shop floor-level skill sets. UR, on the other hand, encapsulates that complexity and designs their cobots specifically for easy training and flexibility. That's a hard thing to do. Further, traditional robot companies' distribution doesn't easily reach to SMEs and does not include an open ecosystem, which for us attracts many third-party developers and new applications.

接下來的問題是：在這個成長中的市場，為什麼競爭對手不容易也取得進展？首先，如各位所預期，傳統機器人公司似乎更聚焦於其龐大的既有客戶群，這些客戶重視與其高載重、複雜機器人的相容性。要把他們為自動化專家設計的平台，轉化為一款易用或具彈性的協作機器人、並符合產線現場技能水準，是非常大的挑戰。相較之下，UR 將複雜性封裝起來，並專為易於訓練與彈性而設計其協作機器人。這是一件很難做到的事。此外，傳統機器人公司的通路不容易觸及中小企業，也不包含開放式生態系；而對我們而言，開放生態系能吸引許多第三方開發者與新應用。

New entrants for cobot start-ups face some of the same design challenges in getting the product right and safety, repeatability, ease of use as well as building effective global distribution and partner ecosystem networks to meet the broad scope of customer applications. We're not confused that a potential market this big will attract competition, though. There are dozens of competitors today, and more will likely enter. And their capabilities will improve over time. The point is, today, they're not there yet, yet the path to commercial success in industrial markets is a difficult one, and significantly, we're not standing still. So in line with the strategy Mark described, our UR focus remains on expanding our competitive modes and product flexibility, distribution to SMEs and large accounts and attracting even more third-party developers to our open platform. There will no doubt be waves of adoption over the next several years, including large companies deploying automation beyond individual plant manager investments as workforce demographics, quality requirements and cost pressures will be relentless forces driving more widespread cobot deployment.

協作機器人新創的新進者，在把產品做到位方面也面臨類似的設計挑戰，包括安全性、重複精度、易用性；同時還必須建立有效的全球通路與夥伴生態系網路，以滿足客戶應用的廣泛範疇。不過，我們並不意外這麼大的潛在市場會吸引競爭。目前已有數十家競爭者，未來很可能還會有更多加入。而他們的能力也會隨時間提升。重點是，就今天而言，他們還沒到位；而且在工業市場走向商業成功的道路本就艱難；更重要的是，我們也不會停下腳步。因此，依照 Mark 所描述的策略，我們 UR 的重點仍是擴大我們的競爭護城河與產品彈性、強化對中小企業與大型客戶的通路覆蓋，並吸引更多第三方開發者加入我們的開放平台。未來幾年無疑會出現一波波採用潮，包括大型企業將自動化部署從單一廠長的投資決策擴展到更廣泛的層級，因為勞動人口結構、品質要求與成本壓力將持續成為推動更廣泛協作機器人部署的強大力量。

Regarding margins, our pricing remains very attractive for our customer ROI calculations, even against lower-price competitors. Safety, repeatability, accuracy and reliability required to support industrial operations 24/7 remains a major competitive mode, especially for the newer entrants. We also have significant cost advantages from volume buying and innovative flexion techniques. For example, since we acquired UR in 2015, we have taken down the bill time from 10 days to 1.5 days. This includes using our cobots in numerous assembly and calibration tasks. These production advantages and material cost savings have brought our UR gross margins up to 60%, about 9 points higher than when we acquired UR in 2015. An additional benefit of the shorter manufacturing cycle time is its significant production space freed up, allowing us to operate with our current footprint through 2023.

就毛利率而言，我們的定價對客戶的投資報酬（ROI）計算仍非常具吸引力，即使面對更低價的競爭者亦然。支援 24/7 工業營運所需的安全性、重複精度、準確性與可靠性，仍是重要的競爭護城河，尤其對較新的進入者更是如此。我們也透過大量採購與創新的柔性製造技術取得顯著成本優勢。例如，自 2015 年收購 UR 以來，我們已將製造週期時間從 10 天縮短至 1.5 天。其中包括在多項組裝與校準作業中使用我們的協作機器人。這些生產優勢與材料成本節省，使 UR 毛利率提升至 60%，較我們 2015 年收購時高約 9 個百分點。製造週期縮短的另一項額外好處，是釋放出大量生產空間，使我們能以目前的廠房規模運作至 2023 年。

We also continue to build competitive modes with new software features, developing a vibrant ecosystem of UR+ accessories, advancing our distribution excellent with SMEs and large accounts and application breadth and expertise.

我們也持續透過新的軟體功能建立競爭護城河，打造蓬勃的 UR+ 配件生態系，並在中小企業與大型客戶的通路卓越性，以及應用廣度與專業能力方面持續精進。

Now moving to the details of the first quarter. Our sales were $494 million. The non-GAAP operating profit rate was 22%, and non-GAAP EPS was $0.54. We had no 10% customer in the quarter, and gross margins were 58%.

接著進入第一季的細節。我們的銷售額為 4.94 億美元。非 GAAP 營業利益率為 22%，非 GAAP 每股盈餘（EPS）為 0.54 美元。本季沒有任何單一客戶占比達 10%，毛利率為 58%。

You'll see our non-GAAP operating expenses were $179 million, up $4 million from the fourth quarter but less than the forecast primarily due to pushouts of Semi Test and OE expenses. Comparing our first quarter OpEx of $179 million versus the year-ago period, it's up by $14 million primarily due to the addition of MiR and further distribution and product development investments at UR.

各位會看到我們的非 GAAP 營業費用為 1.79 億美元，較第四季增加 400 萬美元，但低於預測，主要因半導體測試（Semi Test）與 OE 費用遞延。將第一季營業費用 1.79 億美元與去年同期相比，增加 1,400 萬美元，主要因納入 MiR，以及 UR 在通路與產品開發上的進一步投資。

Semi Test sales were $341 million in the first quarter, with SOC making up $293 million and Memory Test sales of $48 million in the quarter. Semi Test service revenue totaled $78 million in the quarter.

第一季半導體測試（Semi Test）銷售額為 3.41 億美元，其中 SOC 為 2.93 億美元，記憶體測試（Memory Test）本季銷售額為 4,800 萬美元。本季半導體測試服務營收合計為 7,800 萬美元。

In Industrial Automation, sales were $66 million, a new first quarter record. Regionally, IA's first quarter sales broke down 44% in Europe, 28% North America, 23% in Asia and 5% rest of world.

在工業自動化（Industrial Automation）方面，銷售額為 6,600 萬美元，創下第一季新高紀錄。以區域來看，IA 第一季銷售占比為：歐洲 44%、北美 28%、亞洲 23%、世界其他地區 5%。

System Test sales were $58 million, and Wireless Test sales were $29 million in the first quarter.

系統測試（System Test）第一季銷售額為5,800萬美元，無線測試（Wireless Test）銷售額為2,900萬美元。

Let me move to a few GAAP items. First, we had a tax benefit of $15 million in the quarter due to a $26 million release of tax reserves based upon the successful completion of an IRS examination. We also adopted a new lease accounting standard, which added $51 million to our assets and a similar amount to our liabilities. This applies to operating leases greater than a year, which were previously disclosed in our footnotes.

我來談幾個GAAP項目。首先，本季我們認列1,500萬美元的稅務利益，係因IRS稅務查核成功結案，釋放2,600萬美元的稅務準備金所致。我們也採用新的租賃會計準則，使資產增加5,100萬美元，負債亦增加相近金額。此準則適用於一年以上的營業租賃，先前係於附註中揭露。

Regarding capital allocation, our balanced approach for 2019 includes $500 million targeted for buybacks, about $63 million for dividends and, of course, maintaining dry powder for selected Industrial Automation M&A.

就資本配置而言，我們2019年的均衡策略包括：規劃5億美元用於庫藏股回購、約6,300萬美元用於股利，當然也會保留資金彈藥以進行精選的工業自動化併購（M&A）。

Cash and marketable securities declined in the quarter by $208 million to $997 million driven by that balanced capital allocation strategy, which included $156 million of share repurchases, $35 million for earnout payments tied to Universal Robots and MiR and $16 million of dividend payments.

受該均衡資本配置策略影響，本季現金及有價證券減少2.08億美元至9.97億美元；其中包括1.56億美元的股票回購、3,500萬美元用於與Universal Robots及MiR相關的或有對價（earnout）付款，以及1,600萬美元的股利支付。

Turning now to the guidance for the second quarter. Sales are expected to be between $520 million and $550 million, and the non-GAAP EPS range is $0.56 to $0.65 on 173 million diluted shares. Q2 guidance excludes the amortization of acquired intangibles and the noncash convertible debt interest. The second quarter gross margin should run at 58%, about flat with the first quarter. And total OpEx should run from 34% to 36%. The operating profit rate at the midpoint of our second quarter guidance is about 23%.

接著談第二季展望。預期銷售額介於5.2億至5.5億美元；以1.73億股稀釋股數計算，非GAAP每股盈餘（EPS）區間為0.56至0.65美元。第二季指引不包含收購無形資產攤銷及可轉換公司債的非現金利息。第二季毛利率應為58%，與第一季大致持平。總營業費用（OpEx）應介於34%至36%。以第二季指引中點計算，營業利益率約為23%。

Let me quickly cover OpEx plans for 2019, which were essentially unchanged from our January call. We expect full year Test business OpEx spending to be about flat year-over-year, apart from normal variable compensation changes. In Industrial Automation, we expect to grow quarterly OpEx to the mid- to high-40s towards the end of the year, up from about $33 million in the fourth quarter of 2018. Notwithstanding our planned Industrial Automation investments, we expect us, I think, to achieve mid-teens operating profit rates for the full year.

我快速說明2019年的營業費用（OpEx）規劃，基本上與我們1月電話會議時所述一致。我們預期全年測試（Test）事業的OpEx支出年對年大致持平，僅有正常的變動薪酬調整。在工業自動化方面，我們預期到年底單季OpEx將成長至4,000萬美元中段至高段，較2018年第四季約3,300萬美元增加。儘管我們規劃投入工業自動化投資，我們仍預期全年可達到約中十位數（mid-teens）的營業利益率。

Shifting to taxes. Our full year tax rate is expected to be about 16%, unchanged from our January estimate.

接著談稅務。我們預期全年有效稅率約為16%，與1月的估計不變。

So now for my parting remarks. After 75 mostly enjoyable conference calls, I can safely say that Teradyne is the strongest it's been in my tenure. Teradyne has Test businesses expected to grow 3% to 5% annually on a trend line with strong growth and operating margins and the leading cobot arm in autonomous mobile robot platforms, which we expect to grow at a 30% to 40% rate annually through 2022. Teradyne generates significant free cash flow, has a strong balance sheet and has the capacity and capability to successfully make more Industrial Automation chess moves. Teradyne will no doubt continue to be admired as both a leader in test equipment and one of the few companies who has successfully added a new growth platform with next-generation automation.

接下來是我的結語。在經歷了75次大多相當愉快的電話會議後，我可以很有把握地說，Teradyne在我任內從未如此強健。Teradyne的測試事業預期沿著趨勢線每年成長3%至5%，並具備強勁的成長與營業利潤率；同時在自主移動機器人平台方面擁有領先的協作機器人手臂（cobot arm），我們預期其將在2022年前以每年30%至40%的速度成長。Teradyne能產生可觀的自由現金流，資產負債表穩健，並具備能力與實力，成功進行更多工業自動化的「棋步」布局。Teradyne無疑將持續受到讚譽，既是測試設備領導者，也是少數成功新增新成長平台、導入次世代自動化的公司之一。

With that, I'll turn the call back to Andy, and shortly, the reins over to Sanjay.

接下來我把電話交回給Andy，並在不久後把接棒交給Sanjay。

Thanks, Greg. Danny, we'd now like to take some questions.

謝謝你，Greg。Danny，我們現在想開始提問。

(Operator Instructions) Our first question comes from Mehdi Hosseini with SIG.

(接線員指示) 我們的第一個問題來自SIG的Mehdi Hosseini。

First one has to do with the 5G. Mark, can you help us understand what are the specific opportunities like market segment? You referenced base station, but I want to better understand a specific content exposure that you have there. And I have a follow-up.

第一個問題與5G有關。Mark，你能幫我們了解具體的機會，例如市場區隔嗎？你提到基地台，但我想更清楚了解你們在那裡的具體內容曝險（content exposure）。我還有一個追問。

Sure. So as you can imagine, right now, infrastructure is starting to roll out for 5G. And 5G is a collection of technology, some of it is sub-6G, some of it is millimeter away. But predominantly at the moment, it's China deployment of sub-6G infrastructure base stations. And inside those base stations, typically, there's a digital-rich network processing chips that we test, and that's been predominantly what's driving this year's strong growth in that segment for Semi Test.

當然。如你所想，目前5G的基礎建設正開始部署。而5G是一組技術的集合，其中一部分是sub-6G，一部分是毫米波。但就目前而言，主要是中國在部署sub-6G的基礎建設基地台。在這些基地台內，通常有我們會測試的、數位含量很高的網路處理晶片，而這一直是今年半導體測試（Semi Test）該區隔強勁成長的主要驅動因素。

In conjunction with that, in the 5G world, there's going to be more small cells, more small base stations, so there's also infrastructure around the RF content on the towers. That drives a little bit of RF demand as well. But the big, big segment that really propels both Semi Test and LitePoint is when handsets and mainstream handsets adopt 5G millimeter wave. That's still a couple of years away, but what I referred to earlier today is more this infrastructure.

此外，在5G世界裡，將會有更多小型基地台（small cells）、更多小型基地站，因此在塔台上的射頻（RF）內容周邊也會有相關基礎建設。這也會帶動一些RF需求。但真正、真正能同時推動半導體測試（Semi Test）與LitePoint的大區塊，是當手機以及主流手機採用5G毫米波時。那還要再過幾年，但我今天稍早提到的主要是這些基礎建設。

Got it. And one follow-up for Greg, and we'll surely miss you in the next conference calls. And please feel free to join us if you have a spare time. But the days of inventory is almost a near 2-year high. Can you help us understand what's driving this? It's been going up on a sequential basis for almost 8 quarters.

了解。另外一個追問給Greg，我們接下來的電話會議肯定會想念你。如果你有空，也歡迎隨時加入我們。不過存貨天數幾乎來到近兩年高點。你能幫我們了解是什麼在驅動這件事嗎？它已經連續大約8個季度逐季上升。

Was it inventory you said, Mehdi?

Mehdi，你是說存貨嗎？

Days of inventory.

存貨天數。

Oh, days of inventory. Okay. We are investing more inventory in our industrial automation businesses, keeping the high growth. That inventory is on our balance sheet. In many of our other businesses, the inventory is off the balance sheet because it's the ship from our contract manufacturers so you don't see it. So I would expect that inventory probably continue to increase a bit over time as those businesses grow and it remains on our balance sheet.

喔，存貨天數。好的。我們在工業自動化事業上投入更多存貨，以支撐高成長。這些存貨列在我們的資產負債表上。在我們其他許多事業中，存貨不在資產負債表上，因為是由我們的代工廠出貨，所以你看不到。因此我預期，隨著這些事業成長且存貨仍列在我們資產負債表上，存貨可能會隨時間再增加一些。

Could this be an overhang or impact your future free cash flows?

這會不會形成壓力（overhang）或影響你們未來的自由現金流？

No, I don't think so. No. I think, if anything, it helps our future free cash flows because it'll be a source of cash when we liquidate it. But we do need to have inventory for these short-cycle times Industrial Automation customers. The model has been 4 days, 3 days, 5 days, so we ship very quickly. And we expect the demand to increase second and third quarter quite considerably.

不，我不這麼認為。不會。我認為如果有什麼影響，反而會有助於未來的自由現金流，因為當我們把它去化（liquidate）時，會成為現金來源。但我們確實需要存貨，以因應工業自動化客戶的短週期交期。我們的模式一直是4天、3天、5天，所以出貨非常快。而我們預期第二季與第三季需求會相當明顯地增加。

So is there a target of days of inventory that we could use to build a model, cash from operation and free cash flows?

那麼存貨天數是否有一個目標值，讓我們可以用來建立模型、推估營運現金流與自由現金流？

We could try to do that offline. The tricky thing is there's different businesses who have inventory on the balance sheet and others don't. And sometimes new products are on our balance sheet and other times they're not. So it's a hodgepodge. We tend to look at inventory on and off the balance sheet. You can't see that. You only look at inventory on the balance sheet, so you kind of have part of the puzzle. But maybe offline, Andy and I can try to do that with you.

我們可以私下再試著處理。棘手之處在於，不同事業有些存貨在資產負債表上、有些則沒有。而且有時新產品的存貨在我們資產負債表上，有時又不在。所以情況很雜。我們傾向同時看資產負債表內與表外的存貨。你看不到那些。你只能看到資產負債表上的存貨，所以你其實只看到拼圖的一部分。但也許私下Andy和我可以和你一起試著做。

And our next question comes from Vivek Arya with Bank of America Merrill Lynch.

我們的下一個問題來自美國銀行美林（Bank of America Merrill Lynch）的 Vivek Arya。

Good luck to Greg on his next adventure. First question on UR or actually on IA overall. For it to grow 35% to 40% this year implies a very steep ramp in the back half. I'm estimating over 50% half-over-half ramp versus what's been more like 25% to 30% growth in the second versus first half. So I'm curious how's your visibility into that second half? What are the risks and opportunities to hit your target growth this year?

祝 Greg 在下一段旅程一切順利。第一個問題關於 UR，或者其實是關於 IA 整體。若今年要成長 35% 到 40%，代表下半年需要非常陡峭的拉升。我估算下半年相較上半年需要超過 50% 的環比拉升，而過去更像是下半年相較上半年成長 25% 到 30%。所以我想了解你們對下半年的能見度如何？達成今年目標成長的風險與機會是什麼？

Okay. I think we outlined some of this last quarter that we do expect MiR to grow at about 100%. So that obviously gives -- MiR is much smaller than Universal Robots, but that gives us a lift. And for us to hit the 35% to 40% -- higher end of that 34% range we outlined, Universal Robots, we're looking at growing around 28%. So nothing has really changed there. So the things that we are focused on is we see -- we could be at the early stages of some larger enterprises doing deployment that would affect both MiR and Universal Robots. We see there's demographic trends in our favor. We see there's a new product, MiR1000. So there's a number of items that we think will help us as the year unfolds. But we think the target we have is very credible and we're often raising.

好的。我想我們在上一季已概述過其中一些內容，我們確實預期 MiR 約成長 100%。所以這顯然會帶來——MiR 的規模比 Universal Robots 小得多，但這會給我們帶來提振。而要達到 35% 到 40%——也就是我們所提出約 34% 區間的較高端，對於 Universal Robots，我們預期成長約 28%。所以這方面其實沒有太大改變。我們目前聚焦的重點是：我們看到——我們可能正處於一些大型企業開始部署的早期階段，這將同時影響 MiR 與 Universal Robots。我們看到人口結構趨勢對我們有利。我們看到有一個新產品 MiR1000。因此，隨著今年推進，我們認為有多項因素會幫助我們。但我們認為我們的目標非常可信，而且我們往往還會上調。

Got it. And as a follow-up on 5G, what specifically about 5G creates the demand for additional testers? Like if you look at the 4G transition, what was that ramp like? And let's assume next year, only 5% of the smartphone market transitions to 5G. What does that imply just conceptually? I know it's early to guide next year, but what does that imply conceptually for the benefit to you from a test perspective?

了解。另外追問 5G，5G 的哪些具體因素會帶動對額外測試機的需求？例如回顧 4G 的轉換，當時的拉升曲線大概是怎樣？假設明年只有 5% 的智慧型手機市場轉向 5G。這在概念上意味著什麼？我知道現在要對明年給指引還太早，但從測試的角度，概念上這對你們的受益意味著什麼？

Okay. So we've said that 5G, when it's fully up and deploying globally, which is probably around 2021 or 2022, will add somewhere around $400 million to $500 million of buying to the market, our markets, test markets. Teradyne's roughly 50% share in those segments, let's say, so we would expect, once it's up and peaking, that's the benefit we would see. But this penetration, the cellphones of millimeter wave is the key, the largest single portion of that $400 million to $500 million bump. That's not likely to start to be meaningful until 2021 and beyond. So next year's 5G millimeter wave handset test's additional revenue is handful tens of millions of dollars of impact probably out of that total.

好的。我們曾說過，當 5G 完全建置並在全球部署時——大概在 2021 或 2022 年左右——將為市場、也就是我們的市場（測試市場）新增約 4 億到 5 億美元的採購需求。以 Teradyne 在這些細分領域約 50% 的市占來說，我們預期一旦 5G 建置完成並達到高峰，那就是我們會看到的受益。但其中的關鍵、也是這 4 億到 5 億美元增量中最大的一塊，是手機端對毫米波的滲透。這在 2021 年及之後才可能開始變得有意義。因此，明年 5G 毫米波手機測試帶來的額外營收，可能只是其中的一小部分，影響大概是數千萬美元等級。

And our next question comes from Brian Chin with Stifel.

我們的下一個問題來自 Stifel 的 Brian Chin。

Best wishes to you, Greg. First, just to confirm, did you say monthly UR sales from China has been picking up through March? And also, should Q1 represent the trough then in terms of year-over-year growth momentum for UR? When do you expect larger cobot deployments to really hit their stride in terms of shipments? And what does your current pipeline look like in terms of these larger-scale enterprises? And sorry for all the questions bunched up.

祝你一切順利，Greg。首先確認一下，你是說來自中國的 UR 月度銷售一路到 3 月都在回升嗎？另外，就 UR 的年對年成長動能而言，Q1 是否應該就是谷底？你們預期較大型的協作機器人（cobot）部署，何時會在出貨量上真正進入加速期？就這些較大規模的企業客戶而言，你們目前的管線看起來如何？抱歉一次問了這麼多問題。

Okay. Let me give you a few data points around that. So we -- coming into 2019, we expected that Q1's growth would be about where it turned out for UR, 16% or so. We saw throughout the quarter, January, February, March, month-over-month growth rate compared to '18 accelerate, started at about 4% and ended March at about 20%. That's globally. One piece of encouraging news is that China has returned to growth in that mix. The second half of last year, China was pretty flat. But as I mentioned, one of these large enterprise deployments started late in the quarter, in Q1, in China. We expect that to run through the rest of the year. And there's other larger enterprise businesses in the mix in China that we expect to also mature. So it's -- the conviction is sort of based around the plan we had, we're kind of on the plan and the trend lines we see in China. If there's anything that's still, I would say, somewhat soft, it's automotive-related cobot deployments. The global automotive market is down a bit. And we ship probably somewhere -- I would say somewhere in the automotive supply chain, 30-some percent of our cobot sales are somehow related to automotive. So that has been -- impacted us in U.S. and Europe and in China a bit. But we expect that to slowly unwind as we get toward the back half of the year as well.

好的。我提供幾個相關數據點。所以——進入 2019 年時，我們預期 UR 在 Q1 的成長大致會如同實際結果，大約 16%。我們在整個季度看到，相較於 2018 年，1 月、2 月、3 月的月對月成長率在加速：一開始約 4%，到 3 月底約 20%。這是全球層面。一個令人鼓舞的消息是，中國在這個組合中已恢復成長。去年下半年，中國基本上相當持平。但如我提到的，其中一個大型企業部署在本季末、也就是 Q1，在中國啟動。我們預期它會延續到今年剩餘時間。而且在中國的組合中，還有其他較大型的企業業務，我們也預期會逐步成熟。所以——我們的信心某種程度上是基於原先的計畫：我們大致按計畫進行，並且看到中國的趨勢線。如果還有什麼我會說相對偏軟的部分，那就是與汽車相關的協作機器人部署。全球汽車市場有些下滑。而我們的協作機器人銷售中，大概有——我會說在汽車供應鏈的某個環節，約 30% 多與汽車相關。因此這在美國、歐洲以及中國都對我們造成了一些影響。但我們預期隨著接近今年下半年，這種影響也會逐步消退。

Okay. That's really helpful. Maybe second, switching over to Semis. In the past, I was wondering if you could take a stab maybe at the expected distribution of Semi Test revenue first half versus second half, understanding that individual test markets are likely to vary quite a bit.

好的。這非常有幫助。第二個問題，切換到半導體（Semis）。我想請問你是否可以嘗試說明一下，半導體測試（Semi Test）營收在上半年與下半年的預期分布，當然我理解各個細分測試市場可能差異很大。

Yes. I think as Greg said in his remarks, the Semi Test itself in the second half will probably be similar to where it was in the first, perhaps down a bit. But our System Test business, on the other hand, and LitePoint business has been growing. So if you look at Test in the aggregate, the second half of the year is probably awash, maybe slightly up, could be slightly up at the first half. But as always, these test markets have been, especially Semi Test, very volatile and difficult to predict. We do not focus a lot of our marketing energy in trying to predict quarterly or even annual market sizes because it doesn't affect our operational strategy. So it's more less a science and more judgment at this point.

是的。我想如 Greg 在談話中所說，Semi Test 本身在下半年可能與上半年相近，或許會略微下滑。但另一方面，我們的系統測試（System Test）業務以及 LitePoint 業務一直在成長。所以如果把測試業務整體來看，今年下半年大概會與上半年持平，可能略增，也可能比上半年略高一些。但一如既往，這些測試市場——尤其是 Semi Test——波動非常大，也很難預測。我們不會投入太多行銷精力去嘗試預測季度甚至年度的市場規模，因為那不會影響我們的營運策略。所以此時與其說是科學，更像是判斷。

I think the key thing is the Semi Test decline, which might be $50 million or $60 million or some number like that, we think would be offset with Industrial Automation growth. That was what I was trying to say in my prepared remarks. And the second half will look similar.

我認為關鍵在於，Semi Test 的下滑可能是 5,000 萬或 6,000 萬美元，或類似的數字，我們認為可以由工業自動化（Industrial Automation）的成長來抵消。這就是我在事先準備的談話中想表達的重點。而下半年看起來會相似。

Our next question comes from Toshiya Hari with Goldman Sachs.

我們的下一個問題來自高盛（Goldman Sachs）的 Toshiya Hari。

Mark, you guys maintained your 2019 SOC TAM number, $2.5 billion at the midpoint. I just wanted to confirm what the puts and takes were within. I think you called out image sensors and 5G infrastructure on the positive side. It sounded like mobility or mobile hasn't changed all that much since 3 months ago and then maybe analog a little bit weaker. So if you can talk about the pluses and minuses there, that would be helpful. And then related to that, can you remind us why you expect your business to be down about 10 percentage points less than the market?

Mark，你們維持了2019年SOC TAM的數字，中點為25億美元。我只是想確認其中的增減因素是什麼。我想你提到影像感測器和5G基礎建設是正面因素。聽起來行動/手機端自3個月前以來沒有太大變化，而類比可能稍微弱一些。所以如果你能談談其中的利多與利空，會很有幫助。另外相關地，能否提醒我們為什麼你預期你們的業務下滑幅度會比市場少大約10個百分點？

Yes. So again, things in the Semi Test SOC market really haven't changed much. I would say that on the mobility side, it strengthened a bit. A lot of that benefit is coming to Teradyne, fortunately. So that's one reason why the declines we expect in the market were a bit -- performing a bit better than that. The automotive market, in particular, I would say, if there's a pocket, it softened more than we might have thought at the beginning of the year, it would be automotive. But that's been offset, as you mentioned, by things like image sensor and 5G infrastructure.

是的。所以再說一次，半導體測試SOC市場的情況其實沒有太大改變。我會說在行動端方面，稍微轉強了一點。幸運的是，其中很大一部分的好處落在Teradyne身上。所以這也是我們預期的市場下滑幅度會比市場表現好一些的原因之一。汽車市場方面，特別是如果要說有哪一塊更疲弱、比我們年初想的還要軟，那就是汽車。但如你所提，這已被影像感測器和5G基礎建設等因素所抵消。

The other thing about 2018 compared to '19, the market size, there was -- we, in the last conference call, described some one-off events that occurred in 2018 that sort of drove the market to an unsustainable high. There were some large semiconductor companies that shifted from foundry A to foundry B that required a onetime retooling of test capacity of foundry B that doesn't -- it's not an annuity that then continues on next year. Turns out that wasn't -- that segment has shifted, wasn't a place we had a high concentration in, so that change affects our competitor more than us.

另外，關於2018年相較於2019年的市場規模，我們在上次電話會議中描述過一些2018年發生的一次性事件，某種程度上把市場推升到不可持續的高點。有一些大型半導體公司從晶圓代工廠A轉到晶圓代工廠B，這需要代工廠B一次性地重新配置測試產能；這不是那種會在隔年持續發生的經常性收入。結果那並不是我們高度集中的領域，所以這個變動對競爭對手的影響比對我們更大。

Got it. And another quick follow-up on the UR side of the business. It was encouraging to learn about some of the growth you're seeing from the large enterprises. But the 2 customers you mentioned, Mark, specifically, what sort of applications are the cobots used for at those 2 customers and I guess the implications for profitability for UR as demand from large enterprises continue to grow? Is it dilutive to gross margins? Is it pretty much the same? Any color there would be helpful.

了解。另外針對UR業務再追問一個問題。聽到你們在大型企業客戶那邊看到一些成長，令人鼓舞。但你提到的那兩個客戶，Mark，具體來說，協作機器人在那兩個客戶那裡是用於哪些應用？以及隨著大型企業需求持續成長，對UR獲利能力的意涵是什麼？會稀釋毛利率嗎？還是大致相同？如果能提供一些補充說明會很有幫助。

It's Greg. I'll take that. The profitability should be the same with these large customers. And so these are both lighting companies, which is an interesting vertical. And there's 4 different applications that our cobots will be doing. And there are the standard applications we do at other locations. And they're having -- these companies are having difficulty with turnover as well as rising labor costs. So they've strategically made the decision that they need to automate to have a better cost structure and responsiveness.

我是Greg。我來回答。這些大型客戶的獲利能力應該會相同。而且這兩家都是照明公司，這是一個很有意思的垂直領域。我們的協作機器人將在那裡執行4種不同的應用。這些都是我們在其他據點也會做的標準應用。而這些公司同樣面臨人員流動以及勞動成本上升的困難。因此他們策略性地做出決定：需要自動化，以取得更好的成本結構與反應速度。

And one thing we have talked about over a period of time is that there's waves of adoption that will not be smooth. So it's encouraging we're starting to see some very early signs of some large enterprise deployments. So this is something we've had a keen eye towards. The other thing we've talked about is bin picking, which, well, we expect to have some deployments later this year. That may be more of a volume next year story, too. So there are some waves from some of these new adoptions in addition to more ecosystem accessories into new verticals as well.

另外我們長期以來一直談到，採用的浪潮不會是平滑的。所以令人鼓舞的是，我們開始看到一些大型企業部署的非常早期跡象。這一直是我們密切關注的方向。我們也談過料箱取放（bin picking），我們預期今年稍晚會有一些部署。那也可能更像是明年才會放量的故事。因此，除了更多生態系配件導入新的垂直領域之外，這些新採用也會帶來一些波段式的成長。

Our next question comes from John Pitzer with Crédit Suisse.

下一個問題來自瑞士信貸（Crédit Suisse）的John Pitzer。

This is Ada calling in for John. You saw excellent gross margins in the quarter despite it being a lower revenue quarter. Can you maybe go over some of the puts and takes around gross margin through the remainder of the year?

我是Ada，代替John提問。儘管本季營收較低，你們仍看到非常出色的毛利率。你能否說明一下今年剩餘期間毛利率的增減因素？

Our gross margins over the last several years, as you might have noticed, have moved up 2, 3 points. We've done a very good job overall in our supply line group with material costs down and getting second sources where necessary. So that effort continues. And then with customers, they are getting advantages through the architectural advantages versus, versus lowering the price of a product. So I think the last year or 2, we've done much better across the company in creating and capturing value at the gross margin line. So I think gross margins are going to operate at this higher level. Last year, we were at about 58%. This year, we'll probably be at 58% this year as well.

過去幾年我們的毛利率，如你可能注意到的，上升了2到3個點。我們的供應鏈團隊整體做得非常好，把材料成本降下來，並在需要時取得第二來源。所以這項努力會持續。另外在客戶端，他們獲得的優勢更多來自架構上的優勢，而不是、不是透過降低產品價格。所以我認為過去一兩年，我們在全公司在毛利率層面更擅長創造並掌握價值。因此我認為毛利率會維持在較高水準運作。去年大約是58%。今年我們大概也會是58%。

Great. And can you maybe just provide some additional color on the trends you're seeing in automotive and industrial in the test side of the business?

很好。另外你能否就測試業務中汽車與工業端的趨勢，再提供一些補充說明？

Yes. So I think for trends, obviously, the aggregate volume of test capacity is down this year. In terms of what's going on and looking forward, there's a lot of activity around chip design for more autonomous driving and the electrification of vehicles. And so there's a lot of activity around new instrumentation and new test methodologies for this emerging technology. And that extends all the way up into the sort of heavier power electronic side of the automobile. So IGBT, silicon carbide, those kind of technologies are also coming online. So there's a lot of design activity -- a lot of design and activity, but the aggregate volume is down. And with many of the major automotive manufacturers planning to produce EV vehicles starting with next year's model year, there'll be quite a few introductions, we expect the automotive market for test will return to growth next year.

好的。所以就趨勢而言，很明顯今年整體測試產能的需求量是下降的。就目前狀況與未來展望來看，圍繞更高階自動駕駛與車輛電動化的晶片設計活動非常多。因此針對這些新興技術，也有很多新的儀器與新的測試方法學的活動。而且這一路延伸到汽車較重的功率電子領域。所以IGBT、碳化矽等這類技術也正在上線。因此設計活動很多——設計與活動都很多，但整體量是下降的。而隨著許多主要車廠規劃從明年車型年度開始生產電動車，將會有相當多的新產品導入；我們預期汽車測試市場明年會恢復成長。

Our next question comes from C.J. Muse with Evercore.

下一個問題來自Evercore的C.J. Muse。

Let me echo the thoughts, Greg. Congrats on all your time at Teradyne, and I wish you the best of luck as you move on to a professional ping-pong player. I think that's a great move for you. First question on OpEx. You implied a midpoint of the guidance roughly $187 million, yet I believe you guided for full year OpEx flat. So trying to understand how we should model OpEx off of that elevated Q2 going to the second half of the year.

我也呼應一下大家的想法，Greg。恭喜你在Teradyne的這段時間，也祝你接下來一切順利，轉行成為職業乒乓球選手。我覺得這對你來說是個很棒的決定。第一個問題關於營業費用（OpEx）。你暗示指引中點大約是1.87億美元，但我記得你們指引全年OpEx持平。所以我想了解，我們應該如何從第二季偏高的水準去建模下半年OpEx？

Got it, C.J. What we've said, and maybe it wasn't clear enough, is that Test will be flat, other than the Test OpEx can move around based upon variable compensation changes. If it's a high sales or high profit year, there's more OpEx. But what drives our OpEx up is in Industrial Automation, we continue to invest in those high-growth businesses. We talked about getting OpEx up to the mid-upper $40 million by the end of the year, so that is where the OpEx growth will be in 2019. Now in any one quarter, I should add, sometimes you can also have NROE from Semi Test that can increase it, but then it goes away the next quarter. But stepping back, it's similar to last year. Our growth is Industrial Automation, and Test is flat other than variable comp.

了解，C.J. 我們的說法是——也許之前不夠清楚——測試（Test）部門會持平，除了測試部門的OpEx會因變動薪酬的變化而上下波動。如果是高銷售或高獲利的一年，OpEx就會更高。但推升我們OpEx的因素是在工業自動化（Industrial Automation），我們持續投資那些高成長業務。我們談到要在年底前把OpEx提高到4,000多萬美元中段到高段的水準，因此2019年的OpEx成長會來自這裡。另外我補充一下，在任何單一季度，有時半導體測試（Semi Test）也可能出現NROE而使OpEx增加，但下一季就會消失。但拉遠來看，情況和去年類似。我們的成長來自工業自動化，而測試部門持平，除了變動薪酬之外。

So just to follow up on that. So despite implied top line guide of, I think, flat to down 2% based on your first half, second half commentary year-over-year, you're assuming variable comp will move higher because that's associated with IA.

所以只是想就此追問一下。因此，儘管根據你們對上半年、下半年年對年評論所隱含的營收（top line）指引，我想是持平到下滑2%，你們仍假設變動薪酬會提高，因為那與 IA 相關。

No. Variable comp for the year will be lower, but we are investing in Industrial Automation, MiR and Universal Robots and Energid and this bin picking solution. So there is more OpEx going into those 3 businesses, and that is the increase that we're going to have this year. And it was similar to the increase we had last year and the year before if you go further back, which is Universal Robots. I would say for about 5 years in a row now, give or take a handful of million, Test OpEx has been flat, but it's moved around with the variable comp. But we've been growing in Industrial Automation. So that's the story. And we're very disappointed in Test, and we have the infrastructure that can scale. We generally don't need to add new locations or new engineering projects, generally speaking.

不是。全年的變動薪酬會更低，但我們正在投資於工業自動化（Industrial Automation）、MiR、Universal Robots、Energid，以及這個料箱拣選（bin picking）解決方案。因此，會有更多營運費用（OpEx）投入到這3個業務中，而這就是我們今年將會增加的部分。而且如果你往回看，這與我們去年以及前年（再往前就是 Universal Robots）所做的增加幅度相似。我會說大約連續5年左右，測試（Test）的營運費用大致持平，前後差異大概就是隨著變動薪酬而上下波動。但我們一直在工業自動化方面成長。所以故事就是這樣。而我們對測試業務非常失望，但我們有可以擴張的基礎設施。一般來說，我們不需要新增據點或新的工程專案，概括而言是如此。

Excellent. And I guess as my follow-up, on the UR side, it's a business that grew 14% year-over-year and I think you said growing 28% for the full year. Can you kind of walk through what rank order the key drivers of that uplift would be? Is it bin picking? Is it enterprise scale business? Other? How should we think about the moving parts that get that run rate up exiting the year?

很好。我想作為追問，在 UR 這邊，這是一個年對年成長14%的業務，而且我記得你說全年成長28%。你能否大致說明一下，推動這個提升的關鍵驅動因素按重要性排序會是哪些？是料箱拣選（bin picking）嗎？是企業級規模的業務嗎？還是其他？我們應該如何看待讓年末退出時的運行率（run rate）提升的各個變動因素？

Okay. Well, UR grew -- on a quarter basis, Q1 over Q1, grew about 16%. But why will it grow more going forward? Let me try to explain that. First, some of the cyclical headwinds we don't think persist indefinitely that we're seeing in auto, for example. Two, we see larger enterprises. This is the wave we talked about many times. At some point, larger enterprises will be forced to adopt cobots versus individual plant managers making small purchases. So we're starting to see a handful of accounts doing enterprise purchases, and those orders will start shipping later in the year. They haven't shipped yet. Bin picking will start shipping later in the year, will be a bigger driver next year. Demographics is a big issue around the world. There aren't enough workers. Inflation, quality, that's also driving it. And then we have this ecosystem that keeps on finding new verticals to use an arm to solve a problem. Now we can't forecast all those, but we see each quarter there's new applications coming from our ecosystem developers.

好的。UR 成長——以季度來看，Q1 相較 Q1，大約成長了16%。但為什麼未來會成長更多？我試著解釋一下。第一，我們看到的一些景氣循環逆風（例如汽車產業）我們不認為會無限期持續。第二，我們看到更大型的企業。這就是我們多次談到的那一波趨勢。在某個時間點，大型企業將被迫採用協作機器人（cobots），而不再只是由單一工廠經理做小額採購。所以我們開始看到少數客戶以企業層級進行採購，而這些訂單會在今年稍晚開始出貨。目前還沒出貨。料箱拣選（bin picking）也會在今年稍晚開始出貨，明年會是更大的驅動因素。人口結構（demographics）是全球性的重大議題。勞動力不足。通膨、品質等因素也在推動。此外，我們的生態系持續在發掘新的垂直領域，使用機械手臂來解決問題。我們無法預測所有這些，但我們看到每一季都有新的應用從我們的生態系開發者那邊出現。

And our next question comes from Krish Sankar with Cowen and Company.

下一個問題來自 Cowen and Company 的 Krish Sankar。

I have 2 of them. First one is on the LitePoint side. Would you say that WiFi 6 -- the vertical change from WiFi 5 to WiFi 6 is a bigger opportunity than 5G for LitePoint? Or is it vice versa? And along the same path, there are already 3 well-entrenched competitors on the 5G side, so do you think there's still room for a fourth player like LitePoint to get in and gain some share in 5G? And then I have a follow-up.

我有兩個問題。第一個是關於 LitePoint。你會說 WiFi 6——也就是從 WiFi 5 到 WiFi 6 的垂直轉換——對 LitePoint 來說是比 5G 更大的機會嗎？還是反過來？同樣沿著這個思路，在 5G 這邊已經有3家根基很深的競爭對手，所以你認為像 LitePoint 這樣的第四個玩家仍有空間切入並在 5G 取得一些市占嗎？然後我還有一個追問。

Okay. Good questions. So first of all, WiFi 6 is a modest test intensity evolution of the WiFi standard. 5G, especially 5G millimeter wave, is dramatically more test intensive. So 5G will by far be more impactful to LitePoint than WiFi 6. WiFi 6 is beginning to happen now, so it'll be sooner. But the big payoff is still 5G. And then with respect to the competitive environment, it's been the case there's been 4 principal competitors in this market for a decade. LitePoint's true value proposition is production test, efficient, focused production test equipment, not R&D. The other providers tend to focus more on R&D and then try to fit their product into the production environment. So I do think that same differentiation that LitePoint had that's given them a 70% share of WiFi test, they're now poised to increase their market penetration in cellular test in the 5G era. And it will be through that same mechanism whereas the lab equipment that's out there today for development is coming from the other principal providers. As we get close to production, all the major chipset companies in the 5G space have adopted LitePoint as one of their potential and qualified production rollout providers. Again, when does that happen? Probably it's 2021 is the mainstream 5G millimeter wave penetration into handsets.

好的。好問題。首先，WiFi 6 是 WiFi 標準的一次溫和的測試強度（test intensity）演進。5G，尤其是 5G 毫米波，測試強度則大幅提高。因此，5G 對 LitePoint 的影響將遠大於 WiFi 6。WiFi 6 現在開始發生，所以會更早帶來效益。但最大的回報仍然是 5G。至於競爭環境，過去十年這個市場一直都有4家主要競爭者。LitePoint 真正的價值主張在於量產測試（production test）：高效率、聚焦於量產測試的設備，而不是研發（R&D）。其他供應商往往更聚焦於研發，然後再嘗試把他們的產品套用到量產環境。所以我確實認為，LitePoint 既有的差異化——讓他們在 WiFi 測試拿到70%市占——如今也使他們在 5G 時代有望提高在蜂巢式（cellular）測試的市場滲透率。而且會透過同樣的機制：目前用於開發的實驗室設備（lab equipment）主要來自其他主要供應商。當接近量產時，所有 5G 領域的主要晶片組公司都已將 LitePoint 納入其潛在且合格的量產導入供應商之一。再說一次，這何時會發生？大概在2021年，5G 毫米波在手機中的滲透會進入主流。

Got it. Got it. That's very helpful. And then as a follow-up on the automation side, these enterprise customer wins you got on cobots, are they typically sole sourced? In other words, now that you won them, is this a good like locked and loaded business for you for the next few years? And also, is there any cross-selling opportunity at the enterprise between MiR and cobots? And also, congrats to Greg on a great career at Teradyne.

了解。了解。這非常有幫助。然後作為自動化方面的追問，你們在協作機器人（cobots）拿下的這些企業客戶勝利，通常是單一來源（sole sourced）嗎？換句話說，一旦你們贏下來，這是否就像是未來幾年都已經鎖定、隨時可上線（locked and loaded）的業務？另外，在企業客戶層級，MiR 與 cobots 之間是否有任何交叉銷售機會？也恭喜 Greg 在 Teradyne 的精彩職涯。

Okay. I'll take that one. Thank you very much. We believe we will be the sole source for a number of years, or at least for this round of deployment, and we expect to stay in there thereafter because it's more of an enterprise deployment. And in one of these situations, they did look at a lower-cost indigenous company and concluded they were not reliable enough. But obviously, we have to deliver and perform, but that looks good on these key wins. What was the...

好的。我來回答這題。非常感謝。我們相信在若干年內我們會是單一來源，或者至少在這一輪部署中是如此；而且我們預期之後也能持續留在其中，因為這更像是企業級部署。在其中一個案例裡，他們確實看過一家成本較低的本土公司，最後結論是對方不夠可靠。但顯然我們必須交付並且表現到位，不過就這些關鍵勝利來看，情況看起來不錯。剛剛那個是……

Cross-selling between MiR and UR.

MiR 和 UR 之間的交叉銷售。

Oh, cross-selling. There is a little bit more of that being discussed. I think in the future, you might see trade shows where there -- they might periodically be in the same trade shows. You might see an arm on a mobile robot a bit more. So there are some opportunities. But at the moment, we don't want to combine the businesses in any formal way. MiR is operating with an earnout, so we don't want to do anything that could disrupt that earnout and keep their focus away from achieving the maximum sales. So I think longer term, there might be greater opportunities in that regard.

喔，交叉銷售。這方面確實有更多討論。我想未來你可能會看到一些展會——他們可能會不定期出現在同樣的展會上。你可能會更常看到把機械手臂裝在移動機器人上。所以確實有一些機會。但目前，我們不想以任何正式方式把這些業務合併。MiR 是以分期對賭（earnout）的方式在運作，所以我們不想做任何可能干擾該 earnout 的事情，或讓他們的重心偏離達成最大銷售。所以我認為長期來看，在這方面可能會有更大的機會。

And I just want to add one comment, a general comment about the cobot market. I think people -- there's been a perception that the hardware around cobot is a commodity and the differentiation over time is going to be software-centric. But in fact, what we're seeing today is building a reliable, industrial, 24/7 cobot to work in a high mixed [torque] moving mass around environment is very difficult. In fact, companies are failing because of the hardware. We've talked in the past about Rethink having an issue there. But some of our competitors have been binned out of these larger deployments because they have shown reliability problems after a couple of months of stress test. So that, it turns out, reliable hardware is still a significant differentiation for UR. And for the foreseeable future, we see that will be equally important in these large deployments as compared to some of the other benefits we've talked more about like UR+ and the ease of use.

我只想再補充一點，針對協作機器人（cobot）市場的一般性評論。我認為外界一直有種看法——覺得協作機器人周邊的硬體是商品化的，長期的差異化會以軟體為核心。但事實上，我們今天看到的是：要打造一台可靠、工業等級、可 24/7 運作的協作機器人，能在高混合（high mix）、[扭矩]、移動質量（moving mass）環境中工作，是非常困難的。實際上，有些公司失敗就是因為硬體。我們過去談過 Rethink 在這方面遇到的問題。但我們的一些競爭對手也在這些較大型的部署中被淘汰，因為在經過幾個月的壓力測試後，他們顯示出可靠性問題。所以結果證明，可靠的硬體對 UR 來說仍然是一項重要的差異化因素。而在可預見的未來，我們認為在這些大型部署中，這一點將與我們更常談到的其他優勢（例如 UR+ 與易用性）同樣重要。

And our next question comes from Sidney Ho with Deutsche Bank.

下一個問題來自德意志銀行（Deutsche Bank）的 Sidney Ho。

I'll add my congratulations to your retirement Greg, and you'll be missed. So my first question is on the wireless. You talked about expecting a downtick from your largest customer. Historically, is there still a time for an upsized surprise if they were to meet the same product launch window in September? And kind of related to that, would that customer no longer buy modems from Intel and now working with other suppliers, would that be a net positive for you in the long term? And do you expect that a onetime benefit because of that change in supplier?

Greg，也讓我恭喜你退休，你會被想念的。我第一個問題是關於無線業務。你提到預期來自你們最大客戶的需求會下滑。歷史上，如果他們仍然在 9 月達到相同的產品發表時程，是否還有可能出現超預期上修的驚喜？另外相關的一點是，如果該客戶不再向英特爾購買數據機（modem），而改與其他供應商合作，長期而言這對你們會是淨正面嗎？你是否預期這會因供應商更換而帶來一次性的好處？

So at this point in the year, there could be a little bit of upside, but it's not likely. I think that's the case on that subject. The advertised shift in buying dynamics in recent weeks is certainly, potentially a benefit. We do not -- certain pockets -- or certain suppliers, we've not had a historical position at in terms of test. But the way the winds are starting to blow now, that should accrue as an incremental benefit to Teradyne.

就今年這個時間點來看，可能會有一點上行空間，但不太可能。我認為在這個議題上大致就是如此。近幾週所宣稱的採購動態轉變，確實可能帶來一些好處。我們並沒有——在某些區塊——或某些供應商上，在測試方面過去並沒有既有的切入位置。但就目前風向的變化來看，這應該會成為 Teradyne 的增量利多。

Okay. That's helpful. Maybe moving on to sourced test. Other than the part that's related to longer test times, as you guys alluded to, are you starting to see some unit demand recovery in that area as well? And if I recall correctly, about 1 year and 1.5 years ago, this business had recorded that revenue doubled quarter-over-quarter. Can you remind us what that was related to? And is that something that will repeat in the future?

好的。這很有幫助。接著談到外包測試（sourced test）。除了你們提到、與測試時間拉長相關的那部分之外，你們是否也開始看到該領域的單位需求有所回升？如果我沒記錯，大約在 1 年到 1.5 年前，這項業務曾出現營收季對季翻倍。你能提醒我們那是因為什麼因素嗎？未來是否可能再度重演？

Sidney, the business has been lumpy. There are 2 customers we do cloud, 3.5-inch magnetic, and we do a system-level test for a semiconductor company. So one or the other typically is buying, and they buy in lumps and consolidated purchases. And that can move us way up or below the $65 million a year revenue model that we said we established to hit model profit. More recently, the business looks much stronger. The hard disk drive customer is coming back, buying in reasonable volumes. And then there's opportunities for the other customer as well. So I think for the next year or 2, that business looks like it has tailwinds.

Sidney，這項業務一直比較波動。我們有 2 個客戶做雲端、3.5 吋磁性（硬碟）相關的業務，另外也為一家半導體公司做系統層級測試（system-level test）。因此通常是其中一方在採購，而且他們會以批量、集中式的方式下單。這會讓我們的營收大幅高於或低於我們所說、為了達到模型獲利而建立的每年 6,500 萬美元營收模型。最近這項業務看起來強勁得多。硬碟客戶正在回來，以合理的量在採購。另外那位客戶也有機會。所以我認為未來 1 到 2 年，這項業務看起來有順風。

Okay. Great. And maybe if I can squeeze in one, how much do you expect the Lemsys acquisition to contribute to Semi Test revenue in Q2?

好的。很好。如果我可以再塞一個問題，Lemsys 併購案預期在第二季對半導體測試（Semi Test）營收的貢獻會有多少？

Yes. Lemsys is a small revenue company at this point. There's more of a longer-term play here. So revenues for the year -- last year, Lemsys' revenues were below $10 million. So it's a smaller company, but there's a longer-term upside.

是的。Lemsys 目前是一家營收規模很小的公司。這裡更偏向長期布局。所以以全年來看——去年 Lemsys 的營收低於 1,000 萬美元。因此它是一家較小的公司，但長期有上行空間。

(Operator Instructions) Our next question comes from Timothy Arcuri with UBS.

(接線員指示) 下一個問題來自瑞銀（UBS）的 Timothy Arcuri。

Well, Greg, I recall you were -- when you were named CFO, so I guess that makes us both pretty old. But I just wanted to say that you've done a great job. So first thing. Great. And then I wanted to ask you about Industrial Automation. How much of the revenue today is in China in aggregate? And how much do you think will be in China, let's say, like a year from now?

Greg，我記得你被任命為 CFO 的時候——所以我想這表示我們兩個都挺老的。但我只是想說你做得非常好。先說這個。很好。接著我想問工業自動化（Industrial Automation）。目前整體營收中，有多少比例來自中國？那你認為一年後，中國大概會占多少？

We're looking that up. It's historically been just over 10%. But long term, we expect it to be in increasing percentage of Universal Robots sales long term. I think last year may have been 12% or something like that?

我們正在查這個數字。歷來一直是略高於 10%。但長期來看，我們預期中國在 Universal Robots 銷售中的占比會逐步提高。我想去年可能是 12% 左右？

Yes, it was just under 15% last year, 12%, 13%, yes.

是的，去年略低於 15%，大概 12%、13% 左右，是的。

Yes. So 12% to 13%, but long term, it could certainly be 20%.

是的。所以是 12% 到 13%，但長期來看，當然有可能到 20%。

Okay. And then with respect to -- no one's really asked about your largest customer on the Semi Test side. And there's a big -- obviously, they're a big -- probably the biggest factor in your outlook as you look into next year. So now the time has gone on a bit, have you had any more information that would lead you to conclude whether there's something structural that's going on there or whether it's just cyclical and that once all that test capacity is basically digested this year and if units are a bit better next year that you could see a snap back in that business back to what it used to be?

好的。接著關於——其實沒有人真的問到你們半導體測試（Semi Test）端的最大客戶。而且很明顯，他們是你們展望明年時的一個很大——可能是最大的影響因素。所以現在時間又過了一段，你們是否有更多資訊，讓你們判斷那邊到底是結構性問題，還是只是景氣循環；也就是說，一旦今年把所有測試產能基本消化完，且明年出貨量稍微好一些，你們是否可能看到該業務回彈，回到過去的水準？

Yes. I think there's so many moving parts in that area that it's very hard to forecast. I would expect that as the phones start to move into the 5G world, that's going to be a balloon. But there's other offsetting issues around unit volumes and other things that are hard for us to predict. I do think utilization is very high. New features are in the pipeline beyond 5G, and that this year, I would expect, should be a minimum that we get to. But in terms of any kind of quantification going forward, we're really not in a position yet to know.

是的。我認為那個領域有太多變動因素，所以非常難以預測。我預期當手機開始進入 5G 世界時，那會是一個放大效應。但也有其他抵銷因素，像是出貨量與其他我們難以預測的事情。我確實認為產能利用率非常高。除了 5G 之外，還有新功能在研發管線中，而今年我預期應該會是我們能達到的低點。但就未來任何量化預估而言，我們目前還沒有足夠條件去判斷。

And our final question comes from Richard Eastman with Baird.

最後一個問題來自 Baird 的 Richard Eastman。

Congrats again to Greg. Great career at Teradyne. Just -- my first question is just around -- there's a commentary in the press release and I think it was -- Greg, you actually referenced it as well. But you talked about the mix favorable, the favorable mix and the lower operating expense. And I'm curious, the favorable mix doesn't really show up in the gross margin. I mean I think you had guided towards a 58% gross margin for the quarter. Is -- what is the mix there, benefit that you referenced? Does it show up in the operating expense?

再次恭喜 Greg。在 Teradyne 的職涯非常精彩。我第一個問題是關於——新聞稿裡有一段評論，我想——Greg，你也有提到。你談到產品組合有利（mix favorable）、以及較低的營運費用。我很好奇，這個有利的產品組合似乎沒有反映在毛利率上。我的意思是，我記得你們對本季的毛利率指引大約是 58%。你所提到的那個組合效益是什麼？它是反映在營運費用上嗎？

No. The mix, I think, in my comments was comparing Q1 of last year with Q1 of this year.

不是。我想，我在評論中提到的組合，是在比較去年第一季與今年第一季。

Okay. Not to the guide, okay.

好的。不是對指引，好。

Correct. So I think that's where you got off base with -- and the improvement year-over-year is about 3 points.

沒錯。所以我想你是在這裡理解偏了——而年對年改善大約是3個百分點。

Okay. And then just around UR, just a couple of things in UR. One, at the revenue run rate that had delivered in the first quarter, what was the op profit margin in the first quarter for UR or just IA in general? And then also around bin picking, the application that you're speaking to later this year and into next year. Is that bin picking application in -- kind of targeted at the industrial SME market? Or is it targeted at the warehouse fulfillment center level?

好的。接著關於UR，我有幾點想問。第一，以第一季所達成的營收運行水準來看，UR在第一季的營業利潤率是多少，或更一般地說IA的營業利潤率是多少？另外，關於你提到今年稍晚到明年會推出的料箱取放（bin picking）應用。這個料箱取放應用是——比較針對工業中小企業（SME）市場嗎？還是針對倉庫履約中心層級？

Okay. I'll -- we don't normally break this out, but UR was -- this is usually lower -- UR's lowest profitable quarter, but it was about 12%. We expect it to improve as the year unfolds. But in terms of the industrial bin picking application, we're not thinking about it as a e-commerce type thing, picking up different objects from trays or bins. We're thinking about it picking up the identical jumbled up pile of items in the bin that traditionally are used in manufacturing, where an operator takes things out of a bin and puts it on a conveyor belt or puts it in a machine for some processing. So it's an industrial application. And the key thing is to pick it up at the right post, to make it easy to develop a program, which we have this auto pilot technology that makes it very easy to develop the program and to place it at the right post so you don't have to put it down again and pick it up and that wastes time, and therefore, you won't be able to do as many applications. So we think we've got something that is a step-function change but it will be in beta shortly and then hopefully with some customers later in the year.

好的。我會——我們通常不會把這個拆開來講，但UR——這通常是較低的——UR最低的獲利季度，但大約是12%。我們預期隨著年度推進會改善。至於工業料箱取放應用，我們並不是把它想成電商那種，從托盤或料箱中抓取不同物件。我們想的是從料箱裡抓取同一種、雜亂堆放的一堆物品——這些在製造業中傳統上會用到，由操作員從料箱取出並放到輸送帶上，或放入機台進行某些加工。所以這是一個工業應用。關鍵在於要在正確的姿態（post）把它抓起來，讓程式開發更容易；我們有這個自動駕駛（auto pilot）技術，能讓程式開發非常容易，並且把它放到正確的姿態，這樣你就不必再放下來重新抓取——那會浪費時間，因此你也就無法做那麼多應用。所以我們認為我們有一個階躍式（step-function）的改變；它很快會進入beta，然後希望在今年稍晚能與一些客戶一起導入。

And these 2 enterprise agreements, the direct sale agreements, could you just tell us -- to the lighting companies, but what was the competitive advantage relative to -- I think you referenced an indigenous cobot manufacturer, but what was the competitive advantage to pick UR?

那這兩個企業協議、直接銷售協議——你能否跟我們說一下——是給照明公司，但相較於——我想你提到一家本土的協作機器人製造商——選擇UR的競爭優勢是什麼？

It was similar to what Mark said, the incumbents often might have lower capital costs, but the product isn't ready for 24/7 operations. And there's a variety of things that go wrong. Sometimes, the product stops. Sometimes, it doesn't repeat to the same location. Sometimes, it doesn't have the right safety features. So they're not really ready for 24/7 industrial, safe, high repeatability, easy-to-program applications. We know they're coming, but they're not quite there yet is what we're trying to describe.

這跟Mark說的類似：既有廠商往往可能有較低的資本成本，但產品還沒準備好支援24/7運作。而且會出各種問題。有時候產品會停機。有時候無法重複回到同一個位置。有時候沒有適當的安全功能。所以它們其實還沒準備好用於24/7的工業環境——安全、高重複精度、且易於編程的應用。我們知道它們會追上來，但我們想表達的是，它們目前還差一點。

Yes. So it is hardware oriented. Okay.

是。所以是偏硬體導向的。好的。

Yes.

是。

Yes. We are out of time. Thanks for joining us today. This concludes the call, and it also concludes Greg's CFO career. So Greg, hang in there, buddy.

是。我們時間到了。感謝各位今天加入。本次電話會議到此結束，也代表Greg的CFO職涯告一段落。所以Greg，加油，老兄。

Thank you.

謝謝。

Thank you all. Bye-bye.

謝謝各位。再見。

Ladies and gentlemen, thank you for participating in today's conference. This does conclude today's program, and you may all disconnect. Everyone, have a wonderful day.

各位女士先生，感謝您參與今天的會議。今天的議程到此結束，您可以掛線離開。祝各位有美好的一天。