芯源系統 (MPWR) 2019 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by, and welcome to Monolithic Power Systems Q3 2019 Earnings Conference Call. (Operator Instructions) Please be advised that today's conference is being recorded. (Operator Instructions)

各位女士、先生，感謝您稍候，歡迎參加 Monolithic Power Systems 2019 年第三季財報電話會議。(接線員指示) 請注意，今天的會議將被錄音。(接線員指示)

I would now like to hand the conference over to your speaker today, Chief Executive Officer of Monolithic Power Systems, Michael Hsing. Sir, please go ahead.

現在我想把會議交給今天的講者——Monolithic Power Systems 執行長 Michael Hsing。先生，請開始。

Thank you. Hello, everyone. Good morning from Shanghai. We held our third quarter board meeting in a newly established customer support and marketing center for China. We had our board members for the facilities and overseas operations here.

謝謝。各位好。我在上海向各位道早安。我們在中國新成立的客戶支援與行銷中心召開了第三季董事會會議。我們的董事會成員以及負責設施與海外營運的同仁都在這裡。

In the last few days, we received many messages that have showed overwhelming support in response to a recent publication. I'm representing MPS management to express our deep appreciation for your trust and support. Our reaction to this publication is that it is malicious and manipulative. We will not bore your intelligence to go through the details. Lastly, I assure you we will take appropriate actions, including legal actions, against any malicious and unlawful activities damaging MPS' reputation and our shareholders' interest.

過去幾天，我們收到許多訊息，對近期一篇刊物的內容表達了壓倒性的支持。我代表 MPS 管理團隊，對各位的信任與支持致上深切的感謝。我們對該刊物的回應是：其內容帶有惡意且具操縱性。我們不會浪費各位的時間逐一說明細節。最後，我向各位保證，我們將採取適當行動，包括法律行動，以對抗任何損害 MPS 聲譽及股東利益的惡意與違法活動。

Thank you again. Now I [will pass it back to our CFO], Bernie.

再次感謝。現在我[把時間交回給我們的財務長] Bernie。

Thank you for that. Good afternoon, and welcome to the Third Quarter 2019 Monolithic Power Systems Conference Call.

謝謝。各位午安，歡迎參加 Monolithic Power Systems 2019 年第三季電話會議。

In the course of today's conference call, we will make forward-looking statements and projections that involve risks and uncertainties, which could cause results to differ materially from management's current views and expectations. Please refer to the safe harbor statement contained in the earnings release published today. Risks, uncertainties and other factors that could cause actual results to differ are identified in the safe harbor statements contained in the Q3 release and in our SEC filings, including our Form 10-K filed on March 1, 2019, and Q2 '19 Form 10-Q filed on August 2, 2019, both of which are accessible through our website at www.monolithicpower.com. MPS assumes no obligation to update the information provided on today's call.

在今天的電話會議中，我們將提出前瞻性陳述與預測，這些內容涉及風險與不確定性，可能導致實際結果與管理層目前的觀點與預期出現重大差異。請參閱今日發布之財報新聞稿中的「安全港」聲明。可能導致實際結果不同的風險、不確定性及其他因素，已在第三季新聞稿及我們向美國證券交易委員會（SEC）提交的文件中的安全港聲明中加以說明，包括我們於 2019 年 3 月 1 日提交的 Form 10-K，以及於 2019 年 8 月 2 日提交的 2019 年第二季 Form 10-Q；上述文件可透過我們的網站 www.monolithicpower.com 取得。MPS 不承擔更新今日電話會議所提供資訊的任何義務。

We will be discussing gross margin, operating expense, R&D and SG&A expense, operating income, interest and other income, net income and earnings on both a GAAP and a non-GAAP basis. These non-GAAP financial measures are not prepared in accordance with GAAP and should not be considered as a substitute for, or superior to, measures of financial performance prepared in accordance with GAAP. A table that outlines the reconciliation between the non-GAAP financial measures to GAAP financial measures is included in our earnings release, which we have filed with the SEC. I would refer investors to the Q3 2018, Q2 2019 and Q3 2019 releases as well as the reconciling tables that are posted on our website.

我們將同時以 GAAP 與非 GAAP 基礎討論毛利率、營業費用、研發與銷售及管理（SG&A）費用、營業利益、利息及其他收益、淨利與每股盈餘。這些非 GAAP 財務衡量指標並非依 GAAP 編製，不應被視為 GAAP 財務績效衡量指標的替代或更優指標。我們已在向 SEC 提交的財報新聞稿中，附上非 GAAP 財務衡量指標與 GAAP 財務衡量指標之間的調節表。我也建議投資人參考 2018 年第三季、2019 年第二季與 2019 年第三季的新聞稿，以及我們網站上所公布的調節表。

I'd also like to remind you that today's conference call is being webcast live over the internet and will be available for replay on our website for 1 year, along with the earnings release filed with the SEC earlier today.

同時提醒各位，今天的電話會議將透過網際網路進行即時網路直播，並將在我們的網站上提供為期 1 年的重播，並附上今日稍早向 SEC 提交的財報新聞稿。

MPS achieved record third quarter revenue of $168.8 million, 11.8% higher than revenue in the second quarter of 2019 and 5.5% higher than the comparable quarter in 2018.

MPS 第三季營收創下歷史新高，達 1.688 億美元，較 2019 年第二季高 11.8%，較 2018 年同期高 5.5%。

Looking at our revenue by market. In our computing and storage market, third quarter revenue of $52.8 million increased $11.2 million or 26.9% from the second quarter of 2019. Computing and storage revenue represented 31.3% of MPS' third quarter 2019 revenue. The sequential revenue -- the sequential quarterly revenue growth was broad-based with sales gains reported in power management for CPUs, high-end notebooks, servers and storage. Coupled with our recent design wins using QSMod, Quantum State Modulation, in server and AI applications, MPS is at the early stage of this important revenue ramp.

接著看我們依市場別的營收表現。在運算與儲存市場，第三季營收為 5,280 萬美元，較 2019 年第二季增加 1,120 萬美元或 26.9%。運算與儲存營收占 MPS 2019 年第三季總營收的 31.3%。就季對季而言——本季營收成長相當廣泛，我們在 CPU 電源管理、高階筆電、伺服器與儲存等領域的銷售均有提升。再加上我們近期在伺服器與 AI 應用中，採用 QSMod（Quantum State Modulation，量子狀態調變）所取得的設計導入（design win），MPS 正處於這項重要營收放量的早期階段。

I also would like to introduce a killer product, a next-generation 48-volt power solution for artificial intelligence, server and supercomputing applications. This solution offers one of the highest levels of energy efficiency with the smallest form factor of its class. Our customers have validated our solutions, which we believe will win market share in this important segment.

我也想介紹一款殺手級產品：用於人工智慧、伺服器與超級運算應用的下一代 48 伏特電源解決方案。此解決方案在同級產品中提供最高等級之一的能源效率，且具備最小的外形尺寸。我們的客戶已驗證我們的解決方案，我們相信這將在此重要細分市場中贏得市占率。

Third quarter automotive revenue of $24.4 million grew 15.1% over the second quarter of 2019, reflecting initial sales of product for infotainment, smart lighting, ADAS and autonomous driving. In addition, we were awarded a relatively large contract at a major Tier 1 automobile supplier. We expect to begin generating revenue from this relationship in the next 2 to 3 years. Automobile revenue is 14.5% of MPS' total third quarter 2019 revenue.

第三季汽車市場營收為 2,440 萬美元，較 2019 年第二季成長 15.1%，反映出資訊娛樂系統、智慧照明、ADAS 與自動駕駛相關產品的初期銷售。此外，我們也在一家主要的一級（Tier 1）汽車供應商取得一份相對大型的合約。我們預期將在未來 2 至 3 年開始自此合作關係產生營收。汽車營收占 MPS 2019 年第三季總營收的 14.5%。

Third quarter 2019 industrial revenue of $28.9 million increased 28.6% from the second quarter of 2019 due primarily to increased revenue for security systems, power sources and industrial meters. Much of this growth is projected to be project-based, which can vary significantly by quarter. Industrial revenue represented 17.1% of our total third quarter 2019 revenue.

2019 年第三季工業市場營收為 2,890 萬美元，較 2019 年第二季增加 28.6%，主要由於安防系統、電源供應與工業儀表的營收增加。其中相當一部分成長預期為專案型（project-based），因此各季之間可能有顯著波動。工業營收占我們 2019 年第三季總營收的 17.1%。

In our consumer market, revenue of $43.9 million was essentially flat with the second quarter of 2019 and represented 26.0% of our second quarter 2019 revenue. Sequentially, third quarter revenue reflected a seasonal increase in certain legacy consumer markets offset by decreased sales of products for wearable applications and set-top boxes.

在消費性市場，營收為 4,390 萬美元，與 2019 年第二季大致持平，並占我們 2019 年第二季營收的 26.0%。就季對季而言，第三季營收反映部分既有消費性市場的季節性成長，但被穿戴式應用與機上盒產品銷售下滑所抵銷。

Third quarter 2019 communications revenue of $18.8 million was down 14.5% from the second quarter of 2019. Infrastructure, 5G sales, along with sales for legacy router and wireless applications, decreased sequentially. Despite the sequential downturn in 5G revenue, we believe MPS is well positioned to benefit as existing design wins move to revenue. Communications sales represented 11.1% of our total third quarter 2019 revenue.

2019 年第三季通訊市場營收為 1,880 萬美元，較 2019 年第二季下滑 14.5%。基礎建設與 5G 銷售，以及既有路由器與無線應用的銷售，均呈現季對季下滑。儘管 5G 營收季對季下滑，我們仍相信，隨著既有設計導入逐步轉為營收，MPS 已具備良好定位以受惠。通訊銷售占我們 2019 年第三季總營收的 11.1%。

GAAP gross margin was 55.2%, 10 basis points higher than the second quarter of 2019 and 40 basis points lower than the third quarter of 2018. Our GAAP operating income was $30.0 million compared to $20.1 million reported in second quarter of 2019 and $33.5 million reported in the third quarter of 2018. Non-GAAP gross margin for the second quarter of 2019 was 55.6%, matching the gross margin reported for the second quarter of 2019 but 50 basis points lower than the third quarter from a year ago. Our non-GAAP operating income was $51.4 million compared to $43.7 million reported in the prior quarter and $49.2 million reported in the third quarter of 2018.

GAAP 毛利率為 55.2%，較 2019 年第二季高 10 個基點，較 2018 年第三季低 40 個基點。GAAP 營業利益為 3,000 萬美元，相較於 2019 年第二季的 2,010 萬美元，以及 2018 年第三季的 3,350 萬美元。2019 年第二季的非 GAAP 毛利率為 55.6%，與 2019 年第二季所報告的毛利率相同，但較去年同期第三季低 50 個基點。非 GAAP 營業利益為 5,140 萬美元，相較於前一季的 4,370 萬美元，以及 2018 年第三季的 4,920 萬美元。

Let's review our operating expenses. Our GAAP operating expenses were $63.1 million in the third quarter of 2019 compared with $63.1 million in the second quarter of 2019 and $55.5 million in the third quarter of 2018. Our non-GAAP third quarter 2019 operating expenses were $42.5 million, up from the $40.3 million we spent in the second quarter of 2019 and up from the $40.5 million reported in the third quarter of 2018. The difference between non-GAAP operating expenses and GAAP operating expenses for the quarters discussed here are stock compensation expense and income or loss on unfunded deferred compensation plan. For the third quarter of 2019, total stock compensation expense including approximately $641,000 charged to cost of goods sold was $21.3 million compared to $22.7 million reported in the second quarter of 2019.

讓我們回顧一下我們的營運費用。我們2019年第三季的GAAP營運費用為6,310萬美元，與2019年第二季的6,310萬美元相同，並高於2018年第三季的5,550萬美元。我們2019年第三季的非GAAP營運費用為4,250萬美元，高於2019年第二季的4,030萬美元，也高於2018年第三季所報告的4,050萬美元。此處所討論各季度中，非GAAP營運費用與GAAP營運費用之差異，主要來自股票薪酬費用，以及未注資遞延薪酬計畫的收益或損失。2019年第三季的股票薪酬費用總額（包含約64.1萬美元計入銷貨成本）為2,130萬美元，較2019年第二季所報告的2,270萬美元為低。

Switching to the bottom line. Third quarter 2019 GAAP net income was $29.5 million or $0.64 per fully diluted share compared with $20.7 million or $0.45 per share in the second quarter of 2019 and $31.6 million or $0.71 per share in the third quarter of 2018. Q3 non-GAAP net income was $49.5 million or $1.08 per fully diluted share compared with $41.9 million or $0.92 per share in the second quarter of 2019 and $47.3 million or $1.06 per share in the third quarter of 2018. Fully diluted shares outstanding at the end of Q3 2019 were 45.8 million.

接著看最終獲利。2019年第三季GAAP淨利為2,950萬美元，或每股完全稀釋後0.64美元；相較之下，2019年第二季為2,070萬美元或每股0.45美元，2018年第三季為3,160萬美元或每股0.71美元。第三季非GAAP淨利為4,950萬美元，或每股完全稀釋後1.08美元；相較之下，2019年第二季為4,190萬美元或每股0.92美元，2018年第三季為4,730萬美元或每股1.06美元。2019年第三季末的完全稀釋後流通股數為4,580萬股。

Now let's look at the balance sheet. Cash, cash equivalents and investments were $422.0 million at the end of the third quarter of 2019 compared to $369.7 million at the end of the second quarter of 2019. For the quarter, MPS generated operating cash flow of about $72.0 million compared with Q2 2019 operating cash flow of $44.1 million. Third quarter 2019 capital spending totaled $9.1 million.

現在我們來看資產負債表。截至2019年第三季末，現金、約當現金及投資為4.220億美元，相較於2019年第二季末的3.697億美元。本季MPS產生的營運現金流約為7,200萬美元，相較於2019年第二季的4,410萬美元。2019年第三季資本支出合計為910萬美元。

Accounts receivable ended the third quarter of 2019 at $58.3 million, representing 31 days sales outstanding, which was 2 days lower than the 33 days reported at the end of the second quarter of 2019 and 3 days lower than the 34 days at the third quarter of 2018. Our internal inventories at the end of the third quarter of 2019 were $135.6 million, down from the $143.6 million at the end of the second quarter of 2019. Days of inventory of 163 days at the end of the third quarter of 2019 were 30 days lower than at the end of the second quarter of 2019. The drop in inventory, both in terms of dollars and days, represent management's decision to slow wafer starts in response to the market downturn. Under normal circumstances, we are comfortable carrying a higher level of inventory to support our accelerated rate of revenue growth given that most of our products are not customer- or application-specific and carry minimal obsolescence risk.

2019年第三季末應收帳款為5,830萬美元，代表應收帳款週轉天數（DSO）為31天，較2019年第二季末所報告的33天少2天，也較2018年第三季的34天少3天。截至2019年第三季末，我們的內部存貨為1.356億美元，低於2019年第二季末的1.436億美元。2019年第三季末的存貨天數為163天，較2019年第二季末少30天。存貨在金額與天數上的下降，反映管理層因應市場下行而決定放緩晶圓投片（wafer starts）。在正常情況下，鑑於我們多數產品並非客戶或應用特定，且淘汰風險極低，為支撐我們加速的營收成長速度，我們可接受持有較高水位的存貨。

I would like to turn to our outlook for the fourth quarter of 2019. We are forecasting Q4 revenue in the range of $160 million to $166 million. We also expect the following: GAAP gross margin in the range of 54.8% to 55.4%; non-GAAP gross margin in the range of 55.2% to 55.8%; total stock-based compensation expense of $18.3 million to $20.3 million, including approximately $600,000 that would be charged to cost of goods sold; GAAP R&D and SG&A expenses between $56.2 million and $60.2 million; non-GAAP R&D and SG&A expenses to be in the range of $38.5 million to $40.5 million. Litigation expense should range between $800,000 to $1.2 million. Interest income is expected to range from $1.4 million to $1.6 million. Fully diluted shares to be in the range of 45.8 million to 46.8 million shares.

接下來我想談談我們對2019年第四季的展望。我們預估第四季營收介於1.60億至1.66億美元。我們也預期如下：GAAP毛利率介於54.8%至55.4%；非GAAP毛利率介於55.2%至55.8%；股票基礎薪酬費用總額介於1,830萬至2,030萬美元，其中約60萬美元將計入銷貨成本；GAAP研發與銷售、一般及行政（SG&A）費用介於5,620萬至6,020萬美元；非GAAP研發與SG&A費用介於3,850萬至4,050萬美元。訴訟費用預計介於80萬至120萬美元。利息收入預計介於140萬至160萬美元。完全稀釋後股數預計介於4,580萬至4,680萬股。

In conclusion, we continued to execute and deliver the results that speak for themselves. I will now open the phone lines for questions.

總結來說，我們持續執行並交出不言自明的成果。我現在開放電話線路供各位提問。

(Operator Instructions) Our first question comes from the line of Rick Schafer of Oppenheimer.

(接線員指示) 我們的第一個問題來自Oppenheimer的Rick Schafer。

Guys, congratulations on a nice quarter and another -- another tough setup and a nice job on the quarter. I guess my first question is really compute, industrial, auto, all these sort of high-margin kind of new verticals for you, are all up nicely, I think double digits year-over-year. Maybe could you talk a little bit about relative strength by region or any color you could give there and if you had any 10% customers or anybody close to 10%, any particular customer strength in the quarter?

各位，恭喜本季表現不錯，而且在另一個——另一個艱難的環境下仍交出漂亮成績。我想我的第一個問題是，運算、工業、汽車，這些對你們而言屬於高毛利的新垂直領域，看起來都成長得很好，我認為年增都是雙位數。能否談談各地區的相對強弱，或提供一些相關背景？另外，本季是否有任何占比10%的客戶，或接近10%的客戶？是否有任何特定客戶在本季特別強勁？

Sure. Let me sort of scope that out. It wasn't quite -- it wasn't very clear. It's a [little bit of a nugget]. I think to restate Rick's question, if that will help, is that we are observing that there is good relative strength in the quarter for automotive, computing and industrial. And so he was looking for a little bit of color as far as the geographic distribution and if there are any 10% customers or what the level of concentration is like. So yes, I got it. So Rick, you put your finger on the exact right mix of what drove our Q3 performance, so if I can comment first on the computing and storage.

當然。我先把這個問題的範圍稍微界定一下。剛才不太——不太清楚。這算是[一點小重點]。我想如果我重新表述Rick的問題，可能會有幫助：我們觀察到本季在汽車、運算與工業方面有不錯的相對強勢。因此他想了解一些地理分布方面的資訊，以及是否有任何占比10%的客戶，或客戶集中度大概如何。所以，是的，我理解了。Rick，你點出了驅動我們第三季表現的正確組合，所以我先從運算與儲存談起。

Compute and storage, for us, in the first half exhibited some weakness over what it had done in 2018. And we saw in Q3 strong momentum across all of our lines of business, and there was no individual customer concentration that drove that, nor geographic. So that was nice to see, just because in the first half of this year, there had been a weak demand.

對我們而言，運算與儲存在上半年相較於2018年的表現顯得有些疲弱。而在第三季，我們看到各條業務線都有強勁動能，這並非由任何單一客戶集中所帶動，也不是由特定地理區域所驅動。這點很令人欣慰，因為今年上半年需求一直偏弱。

In industrial, that has been an area that has been benefiting or maybe I should say affected by ordering ahead of demand today. And some of that is occurring in China, anticipating any further trade actions. So that was pretty concentrated in power sources and surveillance.

在工業方面，這一直是受惠於——或者我應該說——受到目前超前於需求下單所影響的領域。其中一部分發生在中國，因為市場在預期可能會有進一步的貿易行動。因此這部分相當集中在電源供應與監控（surveillance）領域。

Automotive was very broad-based as far as seeing a beginning of sales in the new model years to Europe and Korea as well as a return of some ordering levels in China. And while I don't have it exact at my fingertips, I do not believe we have any 10% customers that we'll be expecting this quarter.

汽車方面則相當廣泛，我們看到新車款年度在歐洲與韓國開始有銷售起步，同時中國的部分下單水準也有所回升。雖然我手邊沒有精確數字，但我不認為本季會有任何占比10%的客戶。

Okay. I apologize if it's difficult to hear me. My second question and my follow-up is just on 5G. It's obviously becoming a bigger contributor to the MPS growth story. And I know you talked about it being softer in the recent quarter, but obviously, the order book looks good. So what's your visibility on reacceleration in your 5G business? And maybe if you could provide a little bit of color about what that dollar content opportunity is for you guys, some color on your competition there and maybe share assumptions or expectations. Just some general color on 5G, please.

好的。如果我這邊聽起來不太清楚，我先致歉。我的第二個問題以及追問是關於 5G。很明顯它正成為 MPS 成長故事中更大的貢獻者。我知道你提到最近一季表現較疲弱，但顯然訂單能見度看起來不錯。所以你們對 5G 業務重新加速的能見度如何？另外，能否請你們多補充一些你們在其中的單機美元含量（dollar content）機會、競爭態勢，以及可能的假設或預期？請就 5G 整體給一些概括性的說明。

Okay. General color on 5G?

好的。關於 5G 的概括性說明？

Yes.

是的。

Okay. On the 5G, what we see is that our customers telling us that there's more to ramp, okay, as they said, this current -- this quarter than that. And I see this is very initial ramp. So it is somewhat lumpy. And for the next few quarters, our customers, they will follow -- our forecast is that they start to pull in materials. And so whether from the last few quarters to this quarter to the next few quarters, okay, our guess is that still the initial ramp, that kind of looks -- the ramp is not very smooth, kind of lumpy.

好的。就 5G 而言，我們看到客戶告訴我們，還有更多要爬坡的量，好嗎，就像他們說的，這一段——這一季相較於之前是這樣。而我認為這是非常初期的爬坡。所以會有些起伏、不均勻。接下來幾季，我們的客戶——依我們的預測——會開始拉料。因此，不論是從前幾季到這一季再到接下來幾季，好嗎，我們的判斷仍然是在初期爬坡階段；看起來——爬坡不會很平順，會有點忽高忽低。

Okay. Michael, any comment on sort of dollar opportunity, dollar content opportunity when you think about a 5G base station for MPS?

好的。Michael，針對美元機會、也就是談到 MPS 在 5G 基地台的單機美元含量機會，你有什麼看法嗎？

For us, it is difficult to capture because we provide the very basic building block. From the base station to signal chain and to central stations, we provide just the basic building block. We actually don't know where the product ended up.

對我們來說，這很難去量化掌握，因為我們提供的是非常基礎的構建模組。從基地台到訊號鏈，再到中心站，我們提供的只是基本的構建模組。我們其實不知道產品最後會用到哪裡。

Our next question comes from the line of William Stein of SunTrust.

下一個問題來自 SunTrust 的 William Stein。

Great. I think it's -- we're all having trouble hearing each other perhaps. So I'll speak really slowly. I'm hoping you can comment on channel inventory levels, perhaps remind us what the target level is, what that level of inventory was earlier in the year, where it was in Q3 and perhaps where you expect it to go in Q4?

很好。我想——我們可能彼此都聽不太清楚。所以我會講慢一點。希望你能評論一下通路庫存水位，或許也請你提醒我們目標水位大概是多少、今年稍早的庫存水位如何、Q3 時在哪個位置，以及你預期 Q4 會走向哪裡？

Sure. So at the beginning of this year, and let me open up with, is that we haven't disclosed the range that we manage to. We talk in more qualitative terms. But certainly, as we looked at the end of Q1, we finished above our normal range. And a lot of that had to do with the fact that in the quarter, we had higher sales in January and March with the March sales being disproportional to an even spread. The distributors did not have an opportunity to sell through to end customers. So that's why that was up.

當然。所以在今年年初，我先說明一下，我們並沒有揭露我們管理的區間範圍。我們比較多用定性方式來談。但可以確定的是，當我們看 Q1 結束時，我們的水位高於正常區間。這很大一部分是因為該季我們在 1 月與 3 月的銷售較高，且 3 月的銷售相對於平均分布而言偏高。經銷商沒有機會把貨順利銷售到終端客戶。所以庫存才會上升。

In Q2, we had a more even distribution, and days in the channel returned to normal -- at the upper end but normal range. And then in the current quarter, we actually saw terrific sell-through. And again, this was broad-based in all of our geographies. And what we observed is that we're actually at the low end of our range as we finished Q3. So I'd make that similar comment as I did on the inventories on the books, that, that's probably not a level that we care to sustain at, but it certainly was positive to show that, the strong sell-through.

在 Q2，我們的出貨分布更平均，通路天數回到正常——在區間的上緣，但仍屬正常範圍。而在本季，我們其實看到非常強勁的終端拉貨（sell-through）。同樣地，這在我們所有地區都很普遍。我們觀察到 Q3 結束時，我們其實落在區間的下緣。所以我會做出和我剛才談帳上庫存時類似的評論：那大概不是我們希望長期維持的水位，但能顯示強勁的 sell-through，確實是正面的。

Yes. Let me remind everybody. I think that in the early -- regarding the inventory, in early this year, in one of -- I think it was in February earnings call, we commented that the outlook was uncertain and our inventory buildup, that was due to last December quarters that we anticipated the normal growth for MPS is over 20% growth for the entire 2019. And in February, we started to reduce our wafer starts in reaction to more criticism. And also, we don't see a clear futures. And now we ended up at the very low inventory days and low inventory. We have to build it up again.

是的。我提醒大家一下。我想在今年年初——關於庫存——在其中一次（我記得是 2 月的法說會），我們提到當時展望不確定，而我們的庫存增加，是因為在去年 12 月那一季，我們原本預期 MPS 在 2019 全年會有超過 20% 的正常成長。而在 2 月，我們開始降低晶圓投片（wafer starts），以回應更多的批評。同時，我們也看不到明確的未來走向。而現在我們最後落在非常低的庫存天數與低庫存水位。我們必須再把庫存建立回來。

And this is referring to the inventory that's on our books, not in the channel.

這裡指的是我們帳上的庫存，不是通路端的庫存。

Great. If I can have one follow-up, please. I think we're all very interested in your expectations for growth in the coming year. I'm not expecting you to guide necessarily for next year, but I think we have an expectation that through the cycle, you post something on the order of up 20 per year. Do you think next year is going to be typical or reflect a recovery in the broader market so you could at least get close to that number? Or do you think that the current trade situation or anything else would hinder that and keep you materially below it?

很好。如果可以的話我再追問一個。我想大家都很關心你們對明年成長的預期。我不期待你們一定要對明年給指引，但我想我們的預期是，你們在景氣循環中大致能做到每年約 20% 的成長。你認為明年會是典型的一年，或反映更廣泛市場的復甦，讓你們至少能接近那個數字嗎？或者你認為目前的貿易情勢或其他因素會抑制成長，使你們明顯低於那個水準？

Yes. Good question. And although this quarter, we delivered -- remember, we don't want to forecast the next quarter -- no, we want to focus on next year. And throughout this period and I think last year, we said that it was not very clear. Now -- and -- but we did say that all the greenfield products for MPS, our greenfield market segments rather, they will start to grow. And the initial growth, as we just said, even with the 5G ramp, it's not very smooth. It's not very -- it's kind of lumpy. And we -- this quarter, we see 5G is not as good as expected. And now the next few quarters start to pull in. And so we do see all these -- we see this very similar to other segments in auto and in industrial. And so next year we can't forecast, so -- what's the growth rate. I think we'll be better than the industry by 10% to 15%.

是的。好問題。雖然本季我們交出——請記得，我們不想去預測下一季——不，我們想聚焦在明年。在這段期間以及我想去年，我們都說過情況不是很明朗。現在——而且——但我們確實說過，MPS 所有的綠地產品（greenfield products），更精確地說是我們的綠地市場區隔，將會開始成長。而初期成長，如同我們剛才所說，即便是 5G 的爬坡，也不會很平順。不會很——會有點忽高忽低。而我們——本季我們看到 5G 沒有如預期那麼好。而接下來幾季會開始拉料。因此我們確實看到這些——我們看到這與汽車與工業等其他區隔非常類似。所以明年我們無法預測——成長率會是多少。我想我們會比產業平均好 10% 到 15%。

Our next question comes from the line of Tore Svanberg of Stifel.

下一個問題來自 Stifel 的 Tore Svanberg。

Yes. Let me echo the congratulations and especially on the record cash flows. So first question, and maybe just to get a clarification in here. So right now, your inventory days are at 160. So what you're saying is that they will probably go back up again in Q4? That -- will they sort of go back to the level you were in Q1 of this year?

是的。讓我也呼應一下恭喜，尤其是創紀錄的現金流。第一個問題，可能先請你澄清一下。目前你們的庫存天數是 160 天。所以你的意思是 Q4 可能會再回升？——會回到今年 Q1 的水位嗎？

Yes. The range that we're trying to manage between is 160 to 100 days for inventory on our balance sheet. And that may be a little bit higher than what industry and peer norms would be. But again, just going back to Michael's point, is that we're growing at 10 to 15 percentage points ahead of the market. And as a result, this growth is coming from greenfield opportunities with new products and new markets. And so what we do is we invest above the initial demand so that we have positive experiences with our customers.

是的。我們在資產負債表上要管理的庫存天數區間是 160 到 100 天。這可能會比產業與同業的常態略高一些。但再次回到 Michael 的觀點，我們的成長率比市場高出 10 到 15 個百分點。因此，這些成長來自綠地機會（greenfield opportunities），也就是新產品與新市場。所以我們會在初期需求之上先行投資，確保客戶有正面的體驗。

Very good. And yes, I mean that's obviously why you had record cash flows this quarter, right, because that inventory started to selling through, so that's great.

非常好。而且是的，我的意思是，這顯然也是你們本季創下現金流量新高的原因，對吧，因為那些庫存開始出貨並被消化了，所以這很棒。

Second question is on automotive. So it's about a $100 million a year business now. And I think there's no secret that you're starting to gain a lot of traction with -- directly with some of the OEMs. And you even mentioned one big contract. So as you think about the $100 million, that does not include any other sort of direct OEM relationships that you've started to have, correct?

第二個問題是關於車用。所以現在大約是一年 1 億美元的業務。而且我想大家都知道，你們開始直接與一些 OEM 建立很強的動能。你們甚至提到了一個大型合約。所以當你們談到這 1 億美元時，這不包含你們已開始建立的其他任何直接 OEM 關係，對嗎？

Sorry. I apologize. I didn't know which end market you were talking about.

抱歉。我道歉。我不確定你在談的是哪一個終端市場。

Automotive. Automotive.

車用。車用。

Automotive?

車用？

Yes. $100 million business now, yes.

是的。現在是 1 億美元的業務，是的。

Okay. And so the question then is how much exposure do we have to the OEMs?

好的。那麼問題是，我們對 OEM 的曝險有多少？

No, no, no. The question is, you've started to establish some deeper relationships with the OEMs. I'm wondering if that's sort of already contributing to that $100 million run rate. Or is that all new business that's still on the cusp?

不，不，不。問題是，你們已開始與 OEM 建立更深的關係。我想知道這是否已經在貢獻那個 1 億美元的年化營收（run rate）。還是說那全部都是仍在臨門一腳、尚未完全落地的新業務？

So let me play back the question because we're having a bit of difficulty hearing you. And I apologize about that. So you're saying that on the $100 million run rate, as far as what is our exposure to the OEMs and how much of that is still -- that's roughly the question, yes.

我先把問題複述一下，因為我們聽你那邊有點困難。也為此向你致歉。所以你是說，關於 1 億美元的年化營收，就我們對 OEM 的曝險而言，其中有多少仍然是——大概是這個問題，對嗎。

Yes, I mean that's fine, yes. I'm just trying to understand how much of that business is coming from your more recent wins with bigger OEMs.

是的，這樣問可以，是的。我只是想了解其中有多少業務是來自你們近期在較大型 OEM 的新案子／新勝出。

Oh, yes, yes. Okay. Recent win and the recent award, which is a relatively large project from a Tier 1 automotive supplier, it's a relatively large project. These will be in the next 2 or 3 years. And so far, the existing revenues all ramp up from the Tier 2, some of it, very little, from Tier 1. And as we will see in the next 6 to 9 months, we will win a lot more this type of a large project. So that will take the next 2 to 3, 3 to even 4 years away.

喔，是的，是的。好的。近期的勝出與近期拿到的案子，是來自一家一級（Tier 1）車用供應商、規模相對大的專案。這些會在未來 2 到 3 年內發生。而到目前為止，既有營收主要是由二級（Tier 2）客戶逐步放量而來，其中一小部分、非常少的一部分來自一級（Tier 1）。而在接下來 6 到 9 個月，我們會再贏得更多這類大型專案。因此，那會是接下來 2 到 3 年、甚至 3 到 4 年後的事情。

That's exactly what I was asking. Just one last question, and I'll ask a more simple question. When will you be in production with 48-volt?

這正是我想問的。最後一個問題，我會問得更簡單一點。你們什麼時候會進入 48 伏特的量產？

It's starting now.

現在就開始了。

Our next question comes from Alessandra Vecchi of William Blair.

下一個問題來自 William Blair 的 Alessandra Vecchi。

Congratulations also on an amazing quarter in this environment. Just on the OpEx side, you guys have done a phenomenal job sort of holding that steady and managing it. How do we sort of think about balancing your operating expenses going forward with future revenue growth? And then additionally, sort of similar question on the gross margin front, as you guys similarly have done an outstanding job holding those steady. What do we think -- how do we think about the puts and takes of potential margin improvement next year or 2021?

也恭喜你們在這樣的環境下交出非常亮眼的一季。先談營業費用（OpEx）方面，你們在維持穩定並加以管理上做得非常出色。我們該如何看待未來在營業費用與營收成長之間的平衡？另外，毛利率方面也有類似的問題，你們同樣在維持穩定上表現非常好。我們該如何看待明年或 2021 年毛利率改善的利多與利空因素（puts and takes）？

I think as we stick to our model, which I wrestle with our large shareholders, so I want to spend more money and they want to see the EPS growth as well. And -- but here, we stick to our models, which is now, I think that this year, we're slightly above, right?

我想只要我們堅持我們的模型——我也常跟我們的大股東拉扯，我想多花點錢，他們也希望看到 EPS 成長。但是——在這裡，我們會堅持我們的模型；而我想今年我們是略高於模型，對吧？

Yes, yes. It's -- again, Alex, it's a fair question. We have a model which I believe our investors are familiar with, where we want to be able to provide operating leverage along with revenue growth. However, in the current market conditions, what we've done is we've tried to hold our gross margin steady while continuing to invest in certain of the, what we consider long-term strategic investments that will position MPS for the long term.

是的，是的。再說一次，Alex，這是個合理的問題。我們有一個模型，我相信投資人都很熟悉：我們希望在營收成長的同時，也能提供營運槓桿。然而，在目前的市場狀況下，我們所做的是在持續投資某些我們視為長期策略性投資的同時，盡量把毛利率維持穩定，讓 MPS 能在長期更有利的定位。

Okay. That's very helpful. And then similarly, I apologize if I didn't hear it, but did you give any color on sort of what end markets you expect to be strongest or weakest within the Q4 guide?

好的。這很有幫助。另外同樣地，如果我漏聽了先說抱歉，但你們是否有提供一些說明：在第四季指引中，你們預期哪些終端市場會最強或最弱？

When you look at Q4, there's actually a continuation of some of the positive trends that we saw in Q3 with respect to compute and storage and also with automotive. And as I said before, we think that some of industrial had to do with pull-ins. And I look at the 5G, just to echo Michael's point there, is that this is a lumpy business. And while it had a down Q3 we still believe that, that end market has a lot of traction.

看第四季，其實會延續我們在第三季看到的一些正向趨勢，包含運算與儲存，以及車用。如我先前所說，我們認為工業的部分有一些是提前拉貨（pull-ins）所致。至於 5G，呼應 Michael 的觀點，這是一個波動較大的（lumpy）業務。雖然第三季下滑，但我們仍相信那個終端市場有很強的動能。

So Q4 will be better. And the consumer, like normal, would be slightly lower. And the other ones will keep a very similar level.

所以第四季會更好。而消費性如同以往會略低一些。其他的則會維持非常相近的水準。

Very similar. Okay.

非常相近。好的。

Our next question comes from David Williams of Loop Capital.

下一個問題來自 Loop Capital 的 David Williams。

Let me echo the congratulations, guys. It's an impressive quarter and guide, and happy to see that. But I was hoping you might be able to give maybe a little more information on your 48-volt architecture, maybe the benefits there and what that opportunity could be for you and maybe what traction that you've already begun to see there.

我也想再次恭喜各位。這是一個令人印象深刻的季度與指引，很高興看到這樣的表現。不過我希望你們能多提供一些關於 48 伏特架構的資訊，例如它的優勢、對你們而言可能帶來的機會，以及你們目前已經看到的進展。

Okay. The 48-volt system, we started 4, 5 years ago, that's where we see the trend. And we -- in the 48-volt system, there's many components that we ship, and we started shipment about a year ago for some power components. And for the -- for converting 48 volts to the GPU, those are the products and we started to win.

好的。48 伏特系統我們在 4、5 年前就開始投入，因為我們看到這個趨勢。在 48 伏特系統中，我們出貨的零組件很多；大約一年前我們已開始出貨一些電源元件。而在——把 48 伏特轉換到 GPU 的部分，這些就是相關產品，我們也已開始拿到案子（win）。

What do you think that total opportunity could be if we look a year or 2 out? What do you think the revenue potential is there?

如果我們把時間拉到未來 1 到 2 年，你認為整體機會規模可能會是多少？你覺得那裡的營收潛力有多大？

We have actually -- we don't have any idea like, what's the IRR and what's the artificial intelligence market and what's the market size. And for us it's very big, and that will certainly move MPS' revenue needle. And we've done some efforts to find out how big it can be. Certainly hundreds of million dollars. And in the data centers, the adoption rate in the 48-volt as the income piece, we don't see many data centers going to 48 volts other than one, okay, I can think of. They're advocating 48 volts, but that's still -- their majority of servers still running at 12 volts.

其實我們——我們也沒有任何概念，例如 IRR 是多少、人工智慧市場是什麼、以及市場規模有多大。但對我們來說那非常大，而且肯定會推動 MPS 的營收成長。我們也做了一些努力去了解它可能有多大。確定是數億美元的規模。至於資料中心，以 48 伏特的採用率作為收入來源來看，我們看不到很多資料中心轉向 48 伏特，除了有一家，我想得到的就那一家。他們在倡導 48 伏特，但他們大多數的伺服器仍然是在 12 伏特下運行。

David, if I can add one more comment is, I think that the 48-volt is the direction that nearly all data centers, whether it's for supercomputing or for e-commerce or artificial intelligence, will be migrating to.

David，如果我可以再補充一點，我認為 48 伏特是幾乎所有資料中心的方向，不論是用於超級運算、電子商務或人工智慧，未來都會往這個方向遷移。

That's what we believe. And we believe that those market segments, sooner or later, give us 2 or 3 years, they will merge to 48-volts.

這是我們的看法。我們也相信這些市場區隔，遲早——給我們 2 到 3 年——它們會整合到 48 伏特。

Okay. Excellent. And then just lastly for me. Can you comment just maybe on the data center inventory? It's been obviously an issue. But what are you seeing now? And maybe from your perspective, are you seeing inventory issues there? Or are you maybe benefiting from just the design wins and company-specific drivers versus the macro?

好的。非常好。最後一個問題。你們能否談談資料中心的庫存狀況？這顯然一直是個問題。但你們現在看到的是什麼？從你們的角度來看，那邊還有庫存問題嗎？或者你們是否更多是受益於設計導入與公司特定的驅動因素，而不是宏觀因素？

Well, if you're looking at it now, all the new product wins for us, including the -- you just mentioned the 48-volt, these are new products. In the past, I've said again and again, all these new products, our customers take a risk on new products and we're taking a risk. And we don't want to have any problem in the -- during -- in the shipping to the supply market. We don't want to have any interruption with the qualities or with all these annual issues. We want to build inventory. So we have more material we can ship. Now we are at the stage, it's -- we have to put a lot of effort to which -- who do we ship, and we have to build more inventory.

嗯，如果你現在來看，我們所有新的產品得標，包括——你剛提到的 48 伏特——這些都是新產品。過去我一再說，對於這些新產品，我們的客戶在承擔風險，我們也在承擔風險。我們不希望在——在出貨到供應市場的過程中——出任何問題。我們不希望在品質或各種年度性問題上出現任何中斷。我們希望建立庫存。這樣我們就有更多材料可以出貨。現在我們所處的階段是——我們必須投入很多精力去決定——要出貨給誰，而且我們必須建立更多庫存。

David, were you also talking about the data centers as far as whether that was a penetration in new markets?

David，你剛剛也在談資料中心，是否是指在新市場的滲透情況嗎？

Yes. I was just curious if there was still excess inventories within the data centers that had caused some softness prior, if that had been mopped up yet.

是的。我只是好奇，資料中心內部是否仍有過剩庫存，先前造成了一些疲弱；那部分是否已經被消化掉了。

Oh, the data center. Okay. I didn't get that. Yes.

喔，資料中心。好的。我剛剛沒聽懂那個。是的。

So I think that the way to characterize this is that -- and what's interesting is that the downturn that we experienced as a result of overcapacity in the data center was broad-based. And likewise, with just our one Q3 experience, that also was an individual customer, individual geography. It was an example of very diversified growth, both in terms of our products as well as customers.

所以我認為可以這樣描述——有趣的是，我們因資料中心產能過剩而經歷的下行是全面性的。同樣地，就我們第三季那一次的經驗而言，那也是單一客戶、單一地區的情況。這是一個高度分散成長的例子，無論是在我們的產品面，還是在客戶面。

Our next question comes from Quinn Bolton of Needham & Company.

下一個問題來自 Needham & Company 的 Quinn Bolton。

I wanted to just start, and I know you probably haven't had a lot of time to go through Texas Instruments' results. But they had a pretty poor outlook, talked about, at the midpoint of their December guide, revenue being down 14%. And that's, I think, the eighth quarter in a row where their revenue has decelerated on a year-over-year basis. So if I use Texas Instruments as a proxy for the analog market, it certainly looks like Texas Instruments might be looking to a mid-single-digit revenue decline in 2020. And so I guess can you talk about, if the analog market is down again next year, how comfortable are you with kind of The Street consensus for 18%, 19% growth? And to the extent you're comfortable, can you give us some points where you think you're seeing growth or where you're differentiated versus some of your broader analog peers? I know that's a big, big question, but let's start there.

我想先從這裡開始，我知道你們可能還沒有太多時間仔細看德州儀器（Texas Instruments）的財報結果。但他們的展望相當疲弱，並提到以 12 月指引的中位數來看，營收將下滑 14%。而且我認為這已經是他們連續第八個季度在年對年基礎上營收增速放緩。所以如果我把德州儀器當作類比市場的代理指標，看起來德州儀器可能預期 2020 年營收會出現中個位數的下滑。因此我想問的是，如果明年類比市場再次下滑，你們對於華爾街共識預期的 18%、19% 成長有多大把握？如果你們有把握，能否給我們一些你們認為會看到成長的點，或是相較於更廣泛的類比同業，你們的差異化在哪裡？我知道這是個很大、很大的問題，但我們先從這裡開始。

Sure. At this point, we've only had the opportunity to review the flash report on TI. So I am at a deficit to understand how much of the business was affected by downturn in analog versus other elements of their business. But certainly, the significance of the decrease that they projected, both from Q3 to Q4, along with the difference between them and consensus, I think they're a market driver. We found in 2019 that we're not -- MPS is not immune to what the market movements are like. We continue to focus on the elements that are within our control, which is to introduce the new products. And as Michael said earlier on this call, we still continue to believe that we can outperform the market at 10 to 15 percentage points, but that the market hasn't found its equilibrium yet.

當然。截至目前，我們只有機會看過 TI 的快報。因此我還無法充分理解，他們的業務中有多少是受到類比市場下行影響，或是受到其業務其他部分的影響。但他們所預期的下滑幅度之大——不論是從第三季到第四季，或是他們與市場共識之間的差距——他們確實是市場的驅動者。我們在 2019 年發現，我們——MPS 並非對市場波動免疫。我們會持續專注在我們可控的要素，也就是推出新產品。而且如同 Michael 在本次電話會議稍早所說，我們仍然相信我們可以比市場高出 10 到 15 個百分點，但市場尚未找到其均衡點。

Yes. There is no comparison between MPS and TI because they are so big, they are the market. We are very small. We can't -- but we do have a lot of opportunities. And so there's no comparison. But for us, our -- we start to building 5% to 10% of these market shares in the different segments. And -- but at the same time, we're also expanding our opportunities so that growth for us in this market segment is enormous. And we -- in the -- we should be much less affected by the macroeconomic conditions.

是的。MPS 和 TI 之間沒有可比性，因為他們太大了，他們就是市場。我們非常小。我們不能——但我們確實有很多機會。所以沒有可比性。但對我們而言，我們——開始在不同的區隔建立 5% 到 10% 的市占率。同時——我們也在擴大我們的機會，因此我們在這些市場區隔的成長空間非常巨大。而且我們——應該會較不受宏觀經濟狀況影響。

The second question I had is, I think in the June quarter, you guys saw one customer that was over 10% of sales, and you thought that some of that activity was build ahead. You were able to provide and hit your guidance for Q3. You mentioned some additional build ahead activity in the industrial segment here in Q3 in power sources and in surveillance equipment. Have you sort of factored that build ahead in Q3 into your fourth quarter guidance, meaning that if you did see some unexpected strength in build ahead demand in Q3, that you've sort of tempered those segments in Q4 and that's still reflected in your 160 to 166 guidance?

我的第二個問題是，我想在 6 月那一季，你們看到有一個客戶占銷售額超過 10%，而你們認為其中一部分活動是提前拉貨（build ahead）。你們也能夠提供並達成第三季的指引。你們提到在第三季的工業領域，在電源供應以及監控設備方面也有一些額外的提前拉貨活動。你們是否已經把第三季的這些提前拉貨因素納入第四季指引的考量？也就是說，如果第三季確實出現一些超預期的提前拉貨需求，你們已在第四季對那些區隔做了相對保守的調整，而這也仍反映在你們 160 到 166 的指引之中？

Yes. Well obviously, we're aiming for that in Q4. We don't know, but we have to be in it already for Q4.

是的。嗯，很明顯地，我們的目標是在第四季達成。我們不知道，但我們必須在第四季之前就已經在進行中。

Yes, he said it all.

是的，他都說完了。

Okay?

好嗎？

Okay. So the answer is yes?

好。所以答案是是嗎？

Oh, yes, yes, absolutely. Otherwise, we will miss quarter by quarter.

喔，是的，是的，絕對是。不然我們就會一季一季地落後。

Got it. And then lastly, Michael, I just want to come back to the 48-volt. You had mentioned that you've been sampling product, I think, for about a year. I think in the third quarter, GPUs was one of the areas where you saw strength. And I just wanted to clarify, was that strength in GPU more from your existing 12-volt design wins? Or have you seen -- was that GPU strength in Q3 from 48-volt product?

了解。最後，Michael，我想再回到 48 伏這個話題。你提到你們一直在送樣產品，我想大概有一年了。我記得在第三季，GPU 是你們看到強勁表現的領域之一。我想釐清一下，GPU 的強勁表現更多是來自你們既有的 12 伏設計導入（design wins）嗎？還是你們看到的——第三季 GPU 的強勁表現是來自 48 伏產品？

No. It's a -- GPU is a more traditional. It [follows]. The 48-volt is not moving the needle yet, but we do see a healthy growth of revenue stream come in.

不是。GPU 是比較傳統的。它[遵循]。48 伏目前還沒有帶動明顯變化，但我們確實看到有一股健康的營收成長動能進來。

Our next question comes from Matt Ramsay of Cowen.

我們的下一個問題來自 Cowen 的 Matt Ramsay。

Michael, one relationship that my team has been watching closely that MPS has continued to grow is the relationship with Microsoft across the business. Maybe you could -- to the extent that you can, could you give us a little bit of color as to the nature of that relationship strategically and how it might benefit the different segments of the business as you move forward?

Michael，我的團隊一直密切關注的一個關係是，MPS 與微軟在各項業務上的合作關係持續成長。也許你可以——在你能分享的範圍內——給我們一些背景說明，這段關係在策略上是什麼性質，以及未來推進時可能如何讓不同業務部門受益？

Oh, sure. Yes. We don't comment on customers. But with all our customers, I -- in the past, I've said this, and that doesn't mean I don't like it, very cyclical business, which move our revenue up or down. And it's not as smooth as we can manage it. In that regard, I made a statement that is not my favorite, for gaming. And we want to -- for our shareholders, we want to build a model that's very predictable and very predictable growth and for top line growth, margin growth. And so in that regard, in gaming, you have every year or every 2 years in that you have extra revenue, and then the next 2 or 3 quarters it starts to get lower.

喔，當然。是的。我們不評論客戶。但對於我們所有客戶，我——過去我說過這點，這並不代表我不喜歡——這是一個非常具週期性的業務，會讓我們的營收上下波動。而且它不像我們能管理得那麼平滑。在那方面，我曾做過一個不是我最喜歡的陳述，針對遊戲（gaming）。而我們希望——為了股東——建立一個非常可預測的模型，具備非常可預測的成長，包括營收成長、毛利成長。因此在遊戲領域，你每年或每兩年會有額外的營收，然後接下來兩到三季就開始下降。

So that give us a difficulty. And then our customers said that this is not your favorite, and they are also listening. And in the following conference call, I clarified. We love gaming because that is still money for our shareholders, still value. And -- but that's all in models. That's all in business models. And as MPS grows, well, revenue grows in roughly over $600 million range. And we -- in the last couple of years, we see that accelerating growth. And that effect is much, much less. So I love the gaming. I love the gaming.

所以這會帶來一些困難。然後我們的客戶說，這不是你最喜歡的，而他們也在聽。在下一次法說會上，我做了澄清。我們喜歡遊戲，因為那仍然能為股東帶來收益、帶來價值。而——但那都在模型裡。那都屬於商業模式。隨著 MPS 成長，營收大致來到超過 6 億美元的區間。而我們——在過去幾年看到成長加速。那種影響就小很多很多。所以我喜歡遊戲。我喜歡遊戲。

If I can add one additional comment. While we do not talk to individual customers, one thing that I believe separates MPS from our peers and competition is that we're very responsive to our customers' requirements. So if they have an engineering challenge, more often than not they're turning to us in order to find a solution. And as a result, all of our customers benefit from the innovation and the level of technology that we're bringing forward, it's differentiated.

如果我可以再補充一點。雖然我們不談個別客戶，但我認為讓 MPS 與同業及競爭對手不同的一點是，我們對客戶需求的回應非常迅速。所以如果他們有工程上的挑戰，往往會轉向我們來尋找解決方案。因此，我們所有客戶都能受益於我們帶來的創新與技術水準，這是有差異化的。

By the way, what we said, this is not corporate BS. Our customers are telling us we provide the best support, we provide the best service.

順帶一提，我們說的這些不是公司式的官腔。我們的客戶告訴我們，我們提供最好的支援、最好的服務。

Got it. No. I appreciate the sensitivity to the individual customer. As a follow-up question from my side, there's obviously been -- not on the 5G side but in just general data center spending over the last 12 to 15 months, there's been some challenges. And it seems like the market now is anticipating a reacceleration in cloud spending in the server market. You guys have some great content with certain Intel-based servers, but my understanding is there's some diversity of that customer base into other computing providers. Maybe you could talk a little bit about the data center customer base overall and how you're expecting that business to grow in the next few quarters.

了解。不。我理解你們對個別客戶的敏感性。我這邊追問一下，顯然——不是在 5G 端，而是在過去 12 到 15 個月整體資料中心支出方面——有一些挑戰。而市場現在似乎預期伺服器市場的雲端支出會再度加速。你們在某些以 Intel 為基礎的伺服器上有很好的內容（content），但我的理解是，該客戶基礎也有多元化到其他運算供應商。也許你可以談談整體資料中心客戶基礎，以及你們預期未來幾季該業務如何成長。

Sure. I think the strength of MPS is diversity. We have all kinds of different technologies that are suitable for different solutions. So right now, we tend to think very much in terms of the e-commerce data center model. And while that model is very solid and is going to continue for several generations to come, we're seeing, as Michael mentioned earlier, the initial growth through artificial intelligence. And it's hard to really size what those market opportunities look like, but here again, we're positioned to take advantage of it.

當然。我認為 MPS 的優勢在於多元化。我們擁有各式各樣的技術，適用於不同的解決方案。所以目前我們往往會以電商資料中心的模型來思考。而這個模型非常穩固，並且會在未來好幾代持續下去；同時我們也看到，如 Michael 先前提到的，人工智慧帶來的初期成長。而這些市場機會的規模很難精確估算，但同樣地，我們已經就位，可以把握它。

When we look at some of these applications, data center and support of 5G, we're also very well positioned. And it's also at the very initial stage of growth. So I think that when you look at the customer relationships we've developed and how adaptive our technology is for different power solutions, I think we're as well positioned as anybody to take advantage of the new developments of an evolving market.

當我們看一些應用，包括資料中心以及對 5G 的支援，我們也處於非常有利的位置。而且這也還在非常初期的成長階段。所以我認為，當你看我們建立的客戶關係，以及我們的技術對不同電源解決方案的適應性，我們和任何人一樣都很有利於掌握演進市場的新發展。

Our next question comes from the line of Ross Seymore of Deutsche Bank.

我們的下一個問題來自德意志銀行（Deutsche Bank）的 Ross Seymore。

First on the 5G side. Just want to understand a little bit better where you guys are playing there. Is the power management that you're doing largely for one form of processor versus another? Are you more FPGA related? ASIC related? ASSP related? Is there any kind of evolution of how you see that progress, as people are a little bit worried that the FPGA side might come down and then other sorts of processors would backfill? And trying to figure out how, if that does occur, it would impact or benefit Monolithic?

先談 5G 這一端。我想更清楚了解你們在那裡的切入點。你們做的電源管理主要是針對某一種處理器形式相對於另一種嗎？你們比較偏 FPGA 相關？ASIC 相關？ASSP 相關？你們如何看待這個進展的演變？因為有些人有點擔心 FPGA 端可能會下滑，然後由其他類型的處理器來補上。我想了解如果真的發生，會如何影響或有利於 Monolithic？

The answer is all of the above. I don't want to be vague, but we try to figure out where our product goes. We -- just as a reminder to everybody, our products are fundamental blocks. And that's the power supply block. All the items that you mentioned, they all need the power. And our customers typically design several size of these power blocks with our product. And so when each segment from transmitters to the processing -- the data processing to the communication to the fiber line, to optical line, they all need to build the power. And we -- they don't -- within their company, we have become a standard product, standard component that their design engineers design to be in the same systems. So we don't -- we actually really don't know where our product ended up.

答案是以上皆是。我不想講得含糊，但我們會試著弄清楚我們的產品會用到哪裡。我們——提醒大家一下——我們的產品是基礎模組。也就是電源供應模組。你提到的所有項目都需要電力。而我們的客戶通常會用我們的產品設計出多種尺寸的這些電源模組。因此，從發射器到處理——資料處理到通訊到光纖線路到光學線路，每一段都需要建立電源。而我們——在他們公司內部，我們已經成為標準產品、標準元件，他們的設計工程師會把它設計進同一套系統中。所以我們不——我們其實真的不知道我們的產品最後用在了哪裡。

I think you have 2 parts to your question here, Ross. The first is sort of we have an initial ramp and how defensible or predictable is that. And I think we've been very clear that we're in the early stages of 5G, and it's really hard to foresee how that's going to develop. But to Michael's point, as far as the long-term, when this ramp does occur, I think that we get to take advantage of all aspects of 5G. And that's what we're looking forward to.

Ross，我想你的問題可以分成兩個部分。第一個是，我們有一個初期的爬坡期，而這個爬坡有多可防守或多可預測。我想我們一直很清楚地表示，我們仍處於 5G 的早期階段，要預見它將如何發展其實非常困難。但呼應 Michael 的觀點，就長期而言，當這個爬坡真正發生時，我認為我們能夠受惠於 5G 的各個面向。這也是我們所期待的。

Got it. My next question is -- it wouldn't be an earnings call at Monolithic if I didn't ask about inventory. So congratulations on that one as well.

了解。我的下一個問題是——如果我不問庫存，那就不像 Monolithic 的財報電話會議了。所以這點也恭喜你們。

I lost my focus in the February call after Ross asked it. That's a fair question. Actually, we view it as a little too high still because we don't know what the market actually will do.

在二月那通電話會議上，Ross 問完之後我就失去專注了。這是個合理的問題。其實，我們認為目前還是有點偏高，因為我們不知道市場實際上會怎麼走。

Well, if I'm going to give you grief when it goes up, I have to congratulate you when it comes down. But not to blow too much sunshine at you guys, when you talked about it needing to go back up, how do I -- one, is that a dollars or a days comment? And secondarily, how do I reconcile kind of controlling utilization and lowering utilization with inventory rising?

好吧，如果它上升時我要給你們壓力，那它下降時我也得恭喜你們。但也不要把你們捧得太高：當你們談到庫存需要再回升時，我該怎麼理解——第一，那是以金額還是以天數來說？第二，我該如何把「控制產能利用率、降低利用率」與「庫存上升」這兩件事協調起來？

Sure. So if you think about what you do to manage inventory levels and it starts with the wafer starts, so back in February, we really started to ratchet down the wafer starts. And we did that for an extended period of time, it wasn't just one month, but over the last 6 months. Well, those wafers are now coming into our balance sheet. We're receiving them. And it's at a much lower level than we had historically seen and trend lines had been. And yet we had at the same time okay demand and there was a lot of good sell-through as well.

當然。如果你想想我們如何管理庫存水位，起點是晶圓投片（wafer starts），所以回到二月，我們確實開始大幅下調投片。而且我們維持了一段較長的時間，不是一個月而已，而是過去 6 個月。現在，這些晶圓正開始進入我們的資產負債表。我們正在收貨。其水位比我們過去歷史所見、以及趨勢線所呈現的都低得多。但同一時間，我們的需求還可以，而且也有不少良好的終端銷售（sell-through）。

So that means that our finished goods started to come back into aligning as well. So when we look at sort of composition, we're a little bit light on the raw materials, work in process, and we're probably at about the right level of finished goods. When you look at the ramp that we expect for Q2

這表示我們的成品庫存也開始重新對齊。所以從組成來看，我們的原物料與在製品（WIP）稍微偏低，而成品大概在合適的水位。當你看我們對第二季（Q2）預期的爬坡時

(technical difficulty)

(技術問題)

(Operator Instructions) Our next question comes from the line of Chris Caso of Raymond James.

(接線員指示) 我們的下一個問題來自 Raymond James 的 Chris Caso。

Yes, hello. Can you guys hear us now? Operator, we may have lost the management team.

是的，你好。你們現在聽得到我們嗎？接線員，我們可能跟管理團隊斷線了。

I believe we did. Ladies and gentlemen, please stand by. Apologize. There may be a slight delay in today's conference. Please hold. The conference will resume shortly. Thank you for your patience.

我相信是的。各位女士先生，請稍候。抱歉。今天的會議可能會有些微延遲。請稍等。會議將很快恢復。感謝您的耐心。

(technical difficulty)

(技術問題)

(Operator Instructions)

(接線員指示)

Bernie and Michael back online.

Bernie 和 Michael 已重新上線。

We'll go ahead and go to the next question. Ross Seymore from Deutsche Bank.

我們繼續進行下一個問題。德意志銀行的 Ross Seymore。

Yes. So I guess you were halfway through, and you probably continued talking even though we couldn't hear you. The inventory side of things, Bernie, you were walking through the utilization and the WIP versus the raw materials versus all of that. I think the punchline to it was what you didn't say. Does that mean that dollars of inventory are going to start going back up? Days of inventory? Both? And any sort of target range that you expect to get to.

是的。所以我想你們剛講到一半，而且你們可能在我們聽不到的時候還繼續講了。在庫存這部分，Bernie，你剛在說明產能利用率，以及在製品（WIP）相對於原物料等等。我覺得重點反而是你沒說的部分。這是否代表庫存金額會開始回升？庫存天數？兩者都會？以及你們預期會到達的目標區間是什麼？

Again, while we quote our range in days between 160 and 180, the fact is that we manage internally to dollars of inventory anticipating wafer starts about 6 to 8 months in advance of when the revenues are expected to ramp. So right now, we're starting to put wafer starts in for -- in support of business for Q2 and Q3 of 2020. And so in that regard, we'd expect to sort of see a replenishment on a dollar basis. And then obviously, as you look at the cyclicality of our revenues where normally Q1 is about 2 to 4 percentage points lower than Q4, that with a lower denominator, the days would also go up because of that. So my comment was really intended to talk in terms of dollars. And then I think we would have to reflect on the impact on the days given our seasonality.

再說一次，雖然我們對外引用的區間是 160 到 180 天，但事實上我們內部是以庫存金額來管理，並且會在預期營收爬坡前約 6 到 8 個月就先規劃投片。所以目前我們正在安排投片，以支援 2020 年第二季與第三季（Q2、Q3）的業務。因此在這個意義上，我們預期會看到以金額計的補庫（replenishment）。然後很明顯地，考量我們營收的循環性，通常第一季（Q1）會比第四季（Q4）低約 2 到 4 個百分點，在分母較低的情況下，庫存天數也會因此上升。所以我剛剛的評論主要是以金額的角度來談。至於天數的影響，我們還必須考量季節性的因素。

Got it. Two relatively quick ones. I just wanted to understand. One, the litigation expense seems like it's guided to be about double of what it usually is. Is that something -- a court case coming due? Or is it something that is onetime in nature? Or is that a new base we should be looking at?

了解。兩個相對快的問題。我只是想確認一下。第一，訴訟費用看起來指引大約是平常的兩倍。這是因為有案件即將開庭？還是一次性的因素？或者這是我們應該視為新的基準水位？

Yes. We have a one...

是的。我們有一個一次……

Definitely not a long time -- long term.

絕對不是長期——長期的。

It's a onetime case, and it's -- we have a trial date coming up here in the next quarter or so.

這是一個一次性的案件，而且——我們在接下來一季左右會有一個審判日期。

Got you. And then the last quick one is IT security camera side of things. I just wanted to -- I know you commented on that earlier. But when you said that it's customers who were building inventory, is that something that was meant to be a retrospective comment for the first half of this year or second quarter, whatever have you? Or did you see that as an ongoing benefit to your revenues in the third quarter? And if so, what is it -- what do you believe that will do in the fourth quarter?

了解。最後一個快問是關於 IT 安防攝影機這一塊。我只是想——我知道你們先前有提到。但當你說是客戶在建立庫存時，這是指對今年上半年或第二季等已發生情況的回顧性評論？還是你認為這會在第三季持續成為你們營收的助力？如果是，那你認為第四季會有什麼影響？

Well, again, it just speaks to sort of the market uncertainty that we're all experiencing. And I think that if I look at the trend lines for industrial where we report this revenue stream, it is then significantly higher than its normal run rate for us. And so I think you can infer that there were some pull-ins and that could have some impact on how this particular segment and market performs in the next quarter or 2.

嗯，再次強調，這反映的是我們大家都正在經歷的市場不確定性。而且我認為，如果我看我們在工業（industrial）類別中所揭露的這條營收線，其水準明顯高於我們正常的常態水位。因此你可以推論有一些提前拉貨（pull-ins），而這可能會影響這個特定區隔與市場在接下來一到兩季的表現。

We heard different messages from our customers and from our end customers. And so the good news is that they have a few new large projects that are start to ramp. And at the same time, we also heard they're pulling in because of the tariffs. And so again, we're building very basic building blocks. And we're not Intel. So we're building them a building processor, and you can see how many servers or how many PCs they build. We can help, are still very small cost of the entire bills of materials.

我們從客戶以及他們的終端客戶那邊聽到不同的訊息。所以好消息是，他們有幾個新的大型專案開始爬坡。同時，我們也聽到他們因為關稅而提前拉貨。所以再說一次，我們提供的是非常基礎的構建模組。而我們不是英特爾。所以我們不是在替他們做處理器，讓你可以看到他們做了多少伺服器或多少 PC。我們能提供協助，但我們在整體物料清單（BOM）中的成本占比仍然非常小。

Our next question comes from Chris Caso of Raymond James.

我們的下一個問題來自 Raymond James 的 Chris Caso。

First question is with regard to the 10% plus customer that you had in the second quarter. It sounds like that customer wasn't a 10% customer in the third quarter. Could you give some color about what was going on there? And has the revenue level at that customer now normalized as you look forward into the Q4 guidance?

第一個問題是關於你們在第二季有一位占比超過 10% 的客戶。聽起來該客戶在第三季就不是占比 10% 的客戶了。你能否補充一些背景，說明當時發生了什麼？以及展望第四季指引時，該客戶的營收水準現在是否已回到常態？

Yes. I think that we don't talk about individual customers, but since they get reported in our 10-Q for the Q2 we'll say that we have a number of relationships beyond even this particular customer that are very broad-based and where they can have multiple product ramps that are occurring at the same time. And that was clearly what transpired here.

是的。我想我們不會談論個別客戶，但既然他們在我們第二季的 10-Q 中有揭露，我們可以說，除了這位特定客戶之外，我們還有多個關係非常廣泛的合作對象，而且他們可能在同一時間有多個產品放量（ramp）正在發生。而這裡顯然就是這種情況。

Yes. And we don't have consistent more than 10% customers. This quarter, we don't have it, okay? And we don't have it. And in the past, I don't remember whether we reported it or not. There's one for a certain week and a certain period of time where they swapped all their simple things. But it can be different customers. And overall, we have a very long tail of a number of customers. That's due to our business. We build the building block. And many customers, they can use that.

是的。而且我們並沒有持續性地出現占比超過 10% 的客戶。本季我們沒有，好的？我們沒有。至於過去，我不記得我們是否有揭露過。曾經在某一週、某一段期間，有客戶把他們所有簡單的東西都切換了。但可能是不同的客戶。整體而言，我們有非常長尾的眾多客戶。這是由於我們的商業模式。我們提供的是基礎構件（building block）。而許多客戶都可以使用它。

Got it. I guess the last question is with respect to your goal of growing 10% to 15% above the industry, it's sort of an abstract goal that -- the way I'm tending to think about it, and tell me if you agree, is that you need to generate somewhere between $75 million and $90 million of incremental revenue. Perhaps your run rate business grows at the market rate, and this would be on top of that. And I guess if that's the right framework to think about it, what gets you to that? What sort of things can you point to as you look into next year that drive that visibility for that level of design win as you look into next year?

了解。我想最後一個問題是關於你們「比產業成長率高出 10% 到 15%」的目標；這算是一個比較抽象的目標——我目前的理解方式是（也請你確認是否同意），你們需要創造大約 7,500 萬到 9,000 萬美元的增量營收。也許你們的既有常態（run rate）業務以市場增速成長，而這部分是額外加在其上。如果這樣的框架是對的，那麼是什麼因素能讓你們達成？展望明年，哪些事情能讓你們對達到那樣水準的設計導入（design win）有可見度、並驅動這樣的成長？

Yes. It's a very simple business process. And we have a product line, we have a sales -- a geographic sales group and we have a product line review there for projecting for the next couple of years and more in the longer term. And the sales is more immediate in the next few quarters. And we gather all these numbers. And the numbers, we're checking against the product line and also sales group. And also -- and we see some wide swings in the last 9 months due to the tariffs, due to what is modeled in the last year or so.

是的。這是一個非常簡單的業務流程。我們有產品線，也有銷售——一個按地理區域劃分的銷售團隊；我們會做產品線檢討，用來預測未來幾年以及更長期的情況。而銷售端則更聚焦在接下來幾個季度的短期。我們把所有這些數字彙整起來。然後用這些數字去對照產品線以及銷售團隊的看法來檢核。另外——我們也看到過去 9 個月因關稅、以及過去一年左右的模型假設而出現很大的波動。

And it's much more rougher than the previous 2 or 3 years. That's because due to the tariffs, due to government policies. And those, we cannot predict. We cannot predict. And now we're looking for next year why we can say we can do that. One is based on historical numbers and that we can always -- we can do that. And what is that real theory behind it is that this is more empirical than really a theory.

而且相較於前 2 到 3 年，現在粗略得多。這是因為關稅、政府政策等因素。而這些我們無法預測。我們無法預測。至於我們為什麼現在看明年可以這麼說，其一是基於歷史數據，我們一向——我們可以做到。而其背後真正的「理論」其實更偏向經驗實證，而不是真正的理論。

And on the other hand, we're looking at all these sales numbers. These sales numbers are sometimes much bigger than we expected. And so we believe that, and we can grow a lot more than that. Even at the beginning of the year, in January, we forecast to grow more than 20%. So the reality is very different. So we -- to our shareholders, we say that, that was based on the historical numbers, we can grow this much.

另一方面，我們也在看所有這些銷售數字。這些銷售數字有時候比我們預期大得多。因此我們相信我們可以成長得更多。甚至在年初、一月的時候，我們預測成長會超過 20%。所以現實情況差異很大。因此——我們對股東所說的是：這是基於歷史數據，我們可以成長到這個幅度。

At this time, I'd like to turn the call back over to Bernie Blegen for any closing remarks. Sir?

此刻，我想把電話交回給 Bernie Blegen，請他做任何結語。先生？

Great. I'd like to thank you all for joining us in this conference call and look forward to talking to you again during our fourth quarter conference call, which will likely be in early February. Thank you, and have a great day. Thank you very much.

很好。感謝各位參與本次電話會議，也期待在我們第四季電話會議時再與各位交流，時間很可能在二月初。謝謝，祝各位有美好的一天。非常感謝。

Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.

各位女士、先生，今天的電話會議到此結束。感謝各位參與。您現在可以掛線。