芯源系統 (MPWR) 2018 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to Monolithic Power Systems, Inc.'s Third Quarter 2018 Earnings Conference Call. (Operator Instructions) And as a reminder, this conference is being recorded. I'd now like to turn the conference over to Bernie Blegen, Vice President and Chief Financial Officer. Please go ahead.

Thank you. Good afternoon, and welcome to the Third Quarter 2018 Monolithic Power Systems Conference Call.

In the course of today's conference call, we will make forward-looking statements and projections that involve risk and uncertainty, 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 the actual results to differ are identified in the safe harbor statements contained in the Q3 earnings release and in our SEC filings included in our Form 10-K filed on March 1, 2018, and Form 10-Q filed on August 2, 2018, both of which are accessible through our website, www.monolithicpower.com. MPS assumes no obligation to update the information provided on today's call.

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 on 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 2017, Q2 2018 and Q3 2018 earnings releases as well as to the reconciling tables that are posted on our website.

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.

Let me start by saying -- by telling everyone that our Q3 2018 quarterly revenue of $160.0 million established another record for MPS, our fifth high-water mark for quarterly revenue in the last 6 quarters. Likewise, non-GAAP gross margin grew 10 basis points sequentially, representing the 12th consecutive quarter MPS' non-GAAP gross margin has either expanded or matched the prior quarter's performance. As expected, we reached $160 million revenue milestone, reflecting strength in each of our targeted market segments: Q3 revenue for computing and storage up 64% year-over-year; automotive up 54%; and industrial up 52%.

During the quarter, MPS did experience some unexpected softness in high-volume consumer-related businesses, especially in the Greater China region. However, we still see high demand for these products and remain optimistic in our prospects for high-end consumer products. We also gained market share in low-end communications segment. We will stay opportunistic in these high-volume businesses and focus on growing higher-margin products.

Looking at our revenue by end market. In our computing and storage market, revenue of $47.7 million increased $18.6 million or 64.2% year-over-year. Growth in the market was broad-based when compared to the year-ago quarter with all applications, high-end notebooks, cloud computing and storage increasing at rates well above the market average. Computing and storage revenue represented 30% of MPS' third quarter 2018 revenue compared with 23% in Q3 2017.

Third quarter 2018 industrial revenue of $24.9 million increased $8.5 million or 52.1% from the third quarter of 2017, primarily due to increased sales for applications in power sources, meters and security. This market represented 16% of our total third quarter revenue versus 13% in the prior year.

Third quarter automotive revenue of $19.8 million grew $6.9 million or 53.9% over the same period of 2017 as a result of increased sales of infotainment, safety and connectivity application products. Automotive is MPS' largest TAM opportunity at $7 billion, and we're in early stages of penetrating this market. In the years ahead, we plan to offer a number of new products for applications in body control, lighting, infotainment, ADAS and battery management. Automotive revenue was 12% of MPS' total Q3 2018 revenue compared with 12% for Q3 2017.

Third quarter communications revenue of $19.2 million increased $3.8 million or 24.6% over the same period in 2017. This represents a combination of share gains in our legacy markets and initial ramping in products for the 3G -- for the 5G network. Revenue from consumer markets of $48.5 million decreased $6.8 million or 12.4% from the third quarter of 2017. Consumer revenue accounted for 30% of our total Q3 revenue compared with 43% in the prior year. While loss of revenue in these high-volume consumer markets is likely a reflection of geopolitical or trade policy changes, we did not lose projects and continued to gain market share.

GAAP gross margin was 55.6%, 10 basis points higher than the second quarter of 2018 and 60 basis points higher than the third quarter of 2017. Our GAAP operating income was $33.5 million compared to $24.9 million reported in the second quarter of 2018 and $23.8 million reported in the third quarter of 2017. For the third quarter of 2018, non-GAAP gross margin was 56.1%, 10 basis points higher than second quarter of 2018 and 40 basis points higher than the third quarter of 2017. Our non-GAAP operating income was $49.2 million compared to $41.4 million reported in the prior quarter and $38.9 million reported in the third quarter of 2017.

Let's review our operating expenses. Our GAAP operating expenses were $55.5 million in the third quarter compared with $52.7 million in the second quarter of 2018 and $47.0 million in the third quarter of 2017. Our non-GAAP third quarter 2018 operating expenses were $40.5 million, up from the $36.9 million we set in the second quarter of 2018 and up from $32.9 million reported in the third quarter of 2017. On both a GAAP and a non-GAAP basis, third quarter litigation expenses were $343,000 compared with a $639,000 expense in Q2 of 2018 and a $327,000 expense in Q3 2017.

The difference between non-GAAP operating expenses and GAAP operating expenses for the quarters discussed here are stock compensation expense and income or loss from an unfunded deferred compensation plan. Total stock compensation, including $471,000 charged to cost of goods sold, for the third quarter of 2018 was $14.8 million compared with $15.9 million reported in the second quarter of 2018.

Switching to the bottom line. Third quarter 2018 GAAP net income was $31.6 million or $0.71 per fully diluted share compared with $0.55 per share in the second quarter of 2018 and $0.54 per share in the third quarter of 2017. Q3 non-GAAP net income was $47.3 million or $1.06 per fully diluted share compared with $0.90 per share in the second quarter of 2018 and $0.84 per share in the third quarter of 2017. Fully diluted shares outstanding at the end of Q3 2018 were 44.7 million.

Now let's look at the balance sheet. Cash, cash equivalents and investments were $353.1 million at the end of the third quarter of 2018 compared to $318.7 million at the end of the second quarter of 2018. For the quarter, MPS generated operating cash flow of about $52.2 million compared with Q2 2018 operating cash flow of $25.4 million. Third quarter 2018 capital spending totaled $5.1 million. Accounts receivable ended the third quarter of 2018 at $59.9 million or 34 days of sales outstanding compared with the $53.5 million or 35 days reported at the end of the second quarter of 2018 and the $50.8 million or 36 days reported in the third quarter of 2017.

Our internal inventories at the end of the third quarter of 2018 were $136.8 million, up from the $128.9 million at the end of the second quarter of 2018. Days of inventory decreased to 175 days at the end of Q3 2018 from the 189 days at the end of the second quarter of 2018. Days of inventory are in our range -- days of inventory are in our new range reflecting changing customer requirements, particularly in automotive and computing and our new product introductions.

Turning now to our outlook for the fourth quarter of 2018. We are forecasting Q4 revenue in the range of $151 million to $157 million. We also expect the following: GAAP gross margin in the range of 55.2% to 56.2%; non-GAAP gross margin in the range of 55.6% to 56.6%; totals of stock-based compensation expense of $13.5 million to $15.5 million, including approximately $500,000 that would be charged to cost of goods sold; GAAP R&D and SG&A expenses between $50.6 million and $55.6 million; non-GAAP R&D and SG&A expenses to be in the range of $37.6 million to $40.6 million. This estimate excludes stock compensation and litigation expenses. Other income is expected to range from $1 million to $1.2 million before foreign exchange gains or losses. Fully diluted shares to be in the range of 44.6 million to 45.6 million shares.

In conclusion, as expected, we continue to execute according to our plan of diversification in both products and geographical markets. We grew in greenfield segments while gaining share in high-value products in consumer communications amongst the uncertainty in the market and geopolitical environment.

I will now open the phone lines up for questions.

(Operator Instructions) Our first question comes from Matt Ramsay with Cowen.

Congratulations on a strong set of results with a lot of uncertainty out there. The question that we've been getting the most often is, there's a lot of new content growth and market share gains that are embedded into the long-term forecast that investors have for your company across a number of businesses. Maybe, Michael, you can talk about at a high level how you're progressing there and then just juxtapose that against a lot of concerns about a weaker macro environment. Just I know that folks have an expectation of 20% growth for your company over the next couple of years, and we're just trying to understand the risks associated with that versus the opportunities for share gains.

All right, okay. As you know, yes, we're not in a position to answer what the macro position is, and that we only heard from the -- from you guys. And we have a very small percentage of a market share, particularly in those greenfield market segments such as auto and data centers and as well as the industrial side. And we have a very little market segment. Those segments should be relatively immune to what the market condition is, and we expect it to grow in -- according to our plans. And in the next few years, we are all set.

I think one point to add to that is that we're continuing to be very aggressive as far as securing new design wins and that as we've said in our prepared comments, that while there has been a downturn that's reflected most in our consumer business, we didn't lose market share. And in fact, we continued net gains.

Now that commentary is really, really helpful. And I guess as a follow-up, I've spent a decent amount of time with my team digging into the programmable aspects of many of your solutions as you sort of bring programmability into the portfolio. Michael, maybe you can talk a little bit about at a high level what percentage of the products that you're shipping now and winning designs with and then -- I guess second winning designs with are programmable in nature today and how you might see that trend going forward and what that might mean for market share.

All right, okay. At this point, still probably single-digit percentage of our total revenues and we expect quadruple in the next couple of years. And probably most likely in the 4, 5 years, 100% of our products would be programmable.

Our next question comes from Rick Schafer with Oppenheimer.

So maybe -- obviously, auto and industrial up 50 -- each up 50%-plus in the quarter. I think the auto business is tracking to something north of $80 million this year. Mostly, it's been infotainment. And I think now you're starting to see lighting and motor control start to ramp. Maybe you could walk us through what that next leg of growth there looks like. I know, Michael, you talked a lot about having 2, 3 years of really solid visibility there. I'm curious, as ADAS and BMS wins ramp, kind of, a, what the time line looks there and then what we could see for a margin impact. I guess when would we start to see sort of a noticeable margin impact within that auto business?

Even if I talk about infotainment, we still scratch the surface. We just started periphery. And as we just went through the auto business, the deep dive in the auto segments, and we're just starting it. Even in those -- in the infotainment. And for next couple of years, we have all these products designing such as the lighting, including headlights, and also, as Bernie said earlier, the body controls, like in the models and any kind of moving pieces in a car. We have electronic control. And we have those products designed in ADAS. I see it in 3 years out. And all these activities in a week, we have to pick and choose, which product we want to do. And we just don't have enough people. And our name is out. Our product reputation is out. And we are proven to be a quality supply, and so the margin will be -- stay the same as now or even a bit higher.

Got it. And maybe switching gears to e-commerce. I know it's something that a lot of people care about with you guys. Maybe start with any customer feedback or what the feedback has been from customers so far. I know you've talked about adding hundreds of customers there. I know you've also talked about, I believe, first revenues or material revenues sort of in 2020. I guess I'm curious, to smaller customers that e-commerce would be targeting, I mean, do they need to go through a full call like some of your more established, larger customers? And I guess what I'm getting at is, could we see revenues in e-commerce kind of pull forward? Could we see those start to hit before 2020?

Very possible. Particularly those smaller customers in the industrial shops, they buy 10,000 to 20,000 units a year. And those ones are -- those are the customers. And if we have -- if we proved to be a quality supply, they will stay with us for years. And so those ones are just starting now, okay. So in terms of how their feedback is, so far, our website is still -- it's up. And it's not -- we still have a lot more to do, and particularly under the e-commerce market segments, and how do we do digital marketing. And we start to do this it year and early next year. You will see some significant changes. In terms of the current feedback, so far, we hand them a floppy disk and they download it. Almost every one of them, they're very happy. They want to see more of these types of products. And what can I say, every one of them, we have -- in recent months, we gained a few -- more than a few hundred customers. And most of them said they wanted these kind of products.

And have you seen any competitive response from some of the larger HPA guys?

I think these are -- we address the very segment of the market, and we don't see any other players that do a similar thing as we do.

Our next question comes from Quinn Bolton with Needham.

Congratulations again on the very steady execution. Michael, Bernie, just wanted to sort of address, I guess, one of the biggest investor fears I hear is kind of the order rates declining in the China region. Can you give us any sense whether those order rates -- it sounds like they started to decline late summer, whether that stabilized or are they still sort of on a downward trend? And maybe just address the -- sort of where you are in terms of starting backlog looking into the fourth quarter of 2018. I think the last few quarters, you'd gone into the quarter with nearly 100% plan in backlog. And then maybe a couple of product follow-ups.

Sure. I think that you can see from the Q3 numbers that -- and again, the comments that we referenced that in the Greater China market, that we did see a downturn in demand, particularly for consumer. And so as a result of that, we are sort of monitoring to see if that branches out into our other end markets or if there are any changes in ordering pattern. You referenced also our last 3 quarters as far as what our experience has been with backlog going into the quarter. And in fact, what we've seen in this quarter is that we've sort of returned to a more normalized level where we're not at the accelerated pattern that we've been experiencing previously. But the -- it's really, again, returned to a more normalized backlog for this time in the quarter. So it's something that we have to continue to monitor. And certainly, as there are different developments, we have demonstrated the ability to respond to that. But I think the thing that is most encouraging for us is that the targeted areas that we focused on, again, the automotive, the computing and storage and the industrial, where they have long ordering visibility that those remain very solid and that we're just seeing gains that are significantly better than market.

Great. And then maybe, Michael, could you just address, as we look into the end of 2018 and in 2019, Intel will be launching their Cascade Lake platform by the end of the year and then the new Whitley or Cooper Lake platform maybe second half of '19. I know Purley was a big uptick for you in terms of the server content. How do you feel you're positioned with Cascade Lake and Whitley or Cooper Lake over the next 12 months?

I think of this as we -- as I expected it. I think that we grew just -- we don't expect it to have anything different, again, in all the designing activities and all the new product we release. And we expect that it's the same as the last couple of quarters, and I think we'll even grow faster than -- most of the revenue is going to grow in the next -- 2019 all the way to 2022.

So it's sort of a steady ramp in server power management over that '19 to '22 time frame?

Yes, yes. And they are in a similar rate of the growth.

Got it. Go ahead, Bernie.

No. I was going to say that what you were referencing is that Intel has been adjusting their product release schedule around a couple different issues. And obviously, that's something we need to adapt to. We don't see it as a negative to or an impediment to the growth that Michael is referring to.

I care less about couple of -- plus/minus a couple quarters. And that's out of our controls. But the direction of the growth and we cover not only from traditional servers, all this new AR systems and the new type of servers, we cover all of them.

Our next question comes from Ross Seymore with Deutsche Bank.

Just wanted to get into the inventory in the channel side and dovetail that into the backlog question that was asked a bit ago here. Well, Bernie, you said that the backlog coverage has returned to normal. Over the course of this year, correct me if I'm wrong, but have you guys have been shipping to the elevated backlog? Have you been controlling it? And where does channel inventory stand today?

Sure. So in -- the reference that you're talking to is between Q1 and Q2 that we'd actually had a decline in terms of channel inventory, in terms of days. And as we look ahead here, both as far as how we finished Q3 and in Q4, a lot of the sales commitments that we have for both quarters are related to commitments that were made upwards of 4 to 6 months ago. And so we are continuing to deliver against that. But with the amount of uncertainty that is in the market, the timing of when that gets drawn from the channel is a little bit changed. And so we went up a little bit in Q3. And I don't know -- I don't have a forecast for how Q4 is looking, whether it will go up or down a little bit. But again, it's something that we have to continue to monitor.

I guess similarly to that, as the backlog falls, the fear people have for the whole market and companies like yourself is the what goes from 100% coverage plus to something that's normal, then the next step is a further step down. I know you guys have a great secular trend over the course of the number of years. But how do you guys mitigate that risk as we look into, say, the first half of next year?

Yes. Okay. The inventories, they're my favorite questions now, Ross. We have lot of growth in the new market segments. And it changes the pattern of how we do inventory. In auto, industrial and telecom, now the 5G network, although the revenue is small, it started to ramping, and also as well as for data centers and cloud computing. We really need a lot of inventory to cover the ramp. And regarding all these traditional high-volume business, and those businesses we can go in and out within the half a year. And so we're not worried about the inventory.

So we're basically prepared to be able to service the targeted markets. And we believe that, as you look to the first half of '19, that they are less impacted by the price sensitivities that might be -- occur as a result of the tariffs or geopolitical trade policies. And then as Michael just said, that on the areas that are more affected, we have the ability to course correct, and as we demonstrated in Q3, even within the quarter, to be able to achieve our revenue goals.

The high inventory -- again, it's my favorite now, okay. High inventory, in a company like MPS, we have to build a reputation. With regard to traditional, we'll achieve it and we will ship our products. And where we're the newcomers, we have to have a record of reliability as well as a continuation of our delivery. We'll never have that kind of problems. Any one glitch that cause a lot more problems, much bigger market cap hit, then a few million dollars and a few days of inventory.

I guess as my last question, as you guys look back to the last cycle in kind of late '15 and early '16, it was barely a blip for MPS. You still in '16 grew, I think, 16%, 17% year-over-year, way above the analog market as a whole. If you just did compare and contrast the positioning of MPS today versus what it was looking like back 3 years ago, can you just walk through some of the puts and takes as you see it?

Sure. I think that obviously, with all of the new product releases that we've had that really started to ramp beginning in the second half of 2017 and now we're in full swing, we have a much different portfolio that we're able to manage. And so from that regard, I think that the diversification, both in terms of products, end markets and geographies, allows us an awful lot of flexibility that we didn't have back at the second half of 2015. Having said that, I would hate to try and draw a parallel between 2015, which ended up being a rather short-term and pretty much an issue specific to the semiconductor industry, and what we're observing today. I don't think that people have enough visibility or confidence as to how long this current environment might last.

Our next question comes from William Stein with SunTrust.

I just want to make sure I understand the sort of the narrative around end markets and demand trends. So consumer came in light as customers backed off orders in that end market that you think is attributable to tariffs. But you made up more than the difference in the comms end market. I want to make sure I have that right. And also understand better about what drove that upside in comms. Is that something that perhaps, if the consumer end market would have been fine, that you would have posted meaningful upside to the quarter? Or was this something that sort of came and surprised you to the upside? Or was there a lot of extra work that it took to drive this to allow you to deliver good results as you usually do? Helping me understand that would be great.

Thanks, Will. So there's a couple of observations. If you look at consumer sequentially from Q2 to Q3, this was a lower increase than we historically experienced. And it was fairly broad-based as far as the number of areas that were impacted. Interestingly, that's in the traditional consumer -- or high-value consumer end market, we actually performed pretty well. And then as far as the ability to course correct within a quarter, even with relatively short lead time, we play opportunistically in a lot of different markets. And historically, we have not taken on the comms business, this is referring to our legacy gateway and router business, because it tends to be lower margin. And in this quarter, we were able to really just pivot. So it was not a lot of effort in order to be able to accept opportunities that we have an opportunity to quote on and bring that in within the quarter. So as I look ahead, I think the most encouraging thing is that we can have that opportunity. We see the comms business is actually lining up in very good shape for Q4, but I think one of the most encouraging things that might have been a slight upside -- surprise is that we started to see some very initial sales on the 5G side.

The comm business is a -- in the last few quarters, we are flat or slightly down. And so they have a refresh, okay. And we gained some market shares. And -- but there's a significant portion of it. We see the 5G network start to ramp. And those product, we designed it back in -- a couple years ago.

A follow-up, if I can. When people ask about what's perhaps something that investors don't understand about Monolithic, one of the answers I've heard you give is that, well, consensus for next year is up 16% and we think long-term growth is at 20%. Would you still endorse a 20% growth number for 2019? Or would you think that's too optimistic given the geopolitical and other risks?

I think it's a -- if you -- we will clearly beat whatever your forecast is for the next year, industrial growth, okay, we will beat it about 15 points or higher.

I think that when we look at next year, as far as our confidence in 20%, that remains very consistent and solid. Again, we -- one of the things that we want to be known for is our consistency of execution and that we can support that, again, with the amount of visibility we have, particularly in these targeted areas that have the longer design cycles and longer ordering patterns. Now having said that, we're not immune to the macro. And certainly in the first half of the year, we feel reasonably confident that we have the shots on goal to be successful. We don't have clear visibility on how that's going to play out in the early stages of the year.

Our next question comes from Tore Svanberg with Stifel.

Congratulations on the record operating margin. A few questions here. So first of all, I know you have some operations in China. And certainly, your supply chain is there. Any changes to that strategy at all? I mean, I know it's really early days, right? But just given some of the macro political turmoil, any changes to the thinking at all about your operations in China at this point?

We need to diversify it. And we have a lot of resources from -- in China. So we started 2 years ago. We started to diversify it in the different political regimes. And in terms of our R&D, most of the orders are developed in China. And all these -- other than the geopolitical issues and now it's out of our controls, if we see the one region is down the other region is up, okay, that's the area we really want to focus and diversify our presence.

And Tore -- go ahead.

No, no. Go ahead. Finish the answer, yes.

Yes. I was just going to say that on the supply chain issues right now, Michael is exactly right, we don't have control or visibility as far as what the next steps of the process. But again, to the extent that we diversify in the end markets that we sell into and the different customers we have, we would probably adopt a similar profile longer term for our supply chain as well.

That's very fair. And a question on 5G. So you're starting to see some revenue contribution there. Is that power management that's based on Intelli-Phase or QSMod just like in the data center market? Or are these different types of power products?

They are other types of products in a different network segment. As well as the way you said it, it is based on the Intelli-Phase. And those products are in the high-power computing segment, segment of 5G.

Okay, very good. And I think you may have already answered this question, Michael, but as far as your lead times, since you obviously are so focused on making sure the deliveries are there, I assume your lead times are still very stable. They were stable first half. They're stable now?

Yes. The lead times are very, very stable. Bernie can add on, okay.

Yes. We've had no change in the lead times in all phases of production, whether it's wafer packaging tests. Everything is going very consistently.

Okay, very good. Just one last question. You generated almost $50 million in free cash flow this quarter. I mean, I know sort of your approach to capital management, but just kind of based on what the stock price is, any thought about how to put that free cash flow to use?

We're either going to increase the dividend or looking for some tuck-in technology company we can buy. So -- and we are not going to buy another company for revenue increase.

Our next question comes from Alessandra Vecchi with William Blair.

Just to go in a little bit more on the consumer side and the weakness you've been seeing there. I think historically, you guys have said that of your consumer revenue, about 50% of it's traditional. I think 30% of it's high value and 20% of it's gaming. Are you seeing the weakness in gaming as well? Or is it only in that 50% traditional bucket? And if it's in the traditional bucket, is it -- should we consider the whole 50% weak? Or it's only parts of that? Just sort of trying to quantify where consumer could go from here given that seems to be -- in one bucket, that's the most at risk in the short term.

Yes. I hate to say that in the last couple of times that we talked about gaming, our customers not happy, okay. We can't do that. And the one time I said that it wasn't my favorite topic, okay, our customer heard that. Okay, so gaming is my favorite. There's a lot of money to be made with 100% of the components, okay. But here is what have we got, and again, we see high volumes. And we're not talking about what kind of gaming, okay, we see a bit of softness. And these are high-volume ones. And other than that, we see connected device or the IoT, if you will, okay, and a variety of other gauges. And we see -- in our designing, we see the market demand is still very similar.

And then similarly, just in terms of the Q4, you guys traditionally don't give us any sort of granularity on the directional segment -- or the direction of the different segments. But if you had to say on the segment from Q4 what's the strongest to the weakest, do you still think about computing and storage is the strongest and then maybe industrial and optical? Or how has that shifted?

Yes. Again, I think if you look at the results, clearly, what we saw in computing and storage, automotive and industrial, those weren't onetime drivers. This is all about the secular expansion that we've been discussing now for several years, and we're now seeing the results come in full term. And so I don't see anything in my -- in the outlook that would indicate a diminution of those growth rates. And then as far as how we manage the margin with the consumer, that's something that we're adept at being able to do. So I think that as we look at Q4, we feel very confident not just in the number in total, but also the individual markets as far as they're going to source that growth.

Our next question comes from Chris Caso with Raymond James.

Just first question with regard to inventory levels. Can you talk about what your visibility is both to the distribution channel and to OEM inventories? And I know that through the year, you guys have been taking steps to try to discourage customers, prevent customers from building excess inventories. Could you speak to some of those actions and your level of visibility and confidence in the inventory levels?

Yes, that's my favorite question, Chris. And inventory -- okay, I try to convince all of you guys, okay, it's good for us to increase it. And now I think that we're pretty normal.

And then as far as inventory in the channels, again, I'd just refer to an earlier response, is that in the prior 3 quarters, we'd experienced a pattern of over-ordering. And this is reflected in higher-than-normal historic backlog levels at certain points in the quarter. And while we've seen a moderation of that, that is changed because there is a certain level of uncertainty. And we want to make sure that we're really satisfying real demand as opposed to creating a problem for us down the road. Now having said that, with the channel -- we did a very thorough assessment of it at the end of Q3 as well as what we expect to be the sell-through in Q4, and that all got taken into account as we gave our guidance for Q4 in total. So right now, we are feeling that we're getting timely feedback, and we're managing it accordingly.

All right, great. And I guess just following on from that, could you talk about what you consider to be normal seasonal patterns in Q1? And I know there's not a lot of visibility right now, and you're probably doing a lot of work on that. But are there any aspects that we should take into consideration with regard to Q1 based on what you're seeing right now?

Yes. We only provide guidance 1 quarter ahead, and so I don't want to overreach with any comment. But if you traditionally look at -- with the exception of last year, which was an unusual set of circumstances because we had greenfield opportunities, particularly in the compute and storage that we introduced, that we've traditionally had a step down from Q4 to Q1 of between 3% to 4%.

(Operator Instructions) And we do have a follow-up from Quinn Bolton with Needham.

Bernie, you kind of addressed it in answering Chris' question, but I guess I just want to ask more specifically. As you look into your distribution channels, I think there's, especially in uncertain environments, a tendency for the distributors to reduce their inventories in the calendar year-end. And so I guess, when you look at your fourth quarter forecast, you said you took into account sort of the supply chain and the disti channel. Are you expecting them to reduce their days of inventory in the fourth quarter? Or do you expect them to sort of keep a constant days inventory on hand? Any guidance you could provide us would be helpful.

Yes. The formula sort of works that the channel tries to reduce the dollar value of inventory in Q4. But the denominator as far as the quarterly revenue in Q4 is less than Q3. So my expectation is that the dollars will be at or below the Q3 level and the days may be at or maybe even a little above in Q4. And it's really an arithmetic exercise rather than anything that we're concerned about.

And I'm showing no further questions in queue. So I'd like to turn the conference back over to Bernie Blegen for closing remarks.

Thank you. I'd like to thank you all for joining us for this conference call, and I look forward to talking to you again in our fourth quarter conference call, which would likely be in February. Thank you, and have a nice day.

Thank you. Ladies and gentlemen, that does conclude today's conference. Thank you very much for your participation. You may all disconnect. Have a wonderful day.