Good morning. My name is Zekenia, and I will be your conference operator today. At this time, I would like to welcome everyone to the Teradyne Q1 2018 earnings conference call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question-and-answer session. Thank you. I would now like to turn the call over to your host, Mr. Andrew Blanchard. Sir, please go ahead.

Thank you, Zekenia. Good morning, everyone, and welcome to our discussion of Teradyne's most recent financial results and the details of our acquisition of MiR, the leader in autonomous mobile industrial robots. I'm joined by our CEO, Mark Jagiela, and our Chief Financial Officer, Greg Beecher. The press release containing our first quarter results was issued last evening, and the release describing the purchase of MiR was issued earlier this morning. We're providing slides on the Investor page of the website that may be helpful to you in following the discussion. Replay of this call will be available via the same page after the call ends. The matters that we discuss today will include forward-looking statements that involve risk factors that could cause Teradyne's results to differ materially from management's current expectations. We encourage you to review the Safe Harbor statement contained in the earnings release as well as our most recent SEC filings. Additionally, those forward-looking statements are made as of today, and we take no obligation to update them as a result of developments occurring after this call. During today's call, we'll make reference to non-GAAP financial measures. We've posted additional information concerning these non-GAAP financial measures, including reconciliation to the most directly comparable GAAP financial measure, are available on the Investor Page of our website. Also, between now and our next earnings call, Teradyne will be participating in Investor Conferences hosted by Cowen, Bernstein, BofA, Baird, Stifel and Credit Suisse. We'll also be meeting with investors in Munich at the automatica trade show in June and at SEMICON West in early July. Now let's get on with the rest of the agenda. First, Mark will comment on our recent results, market conditions and the acquisition of MiR. Greg will then offer more details on our quarterly financial results and the acquisition, followed by our guidance of the second quarter. We'll then answer your questions and this call is scheduled for one hour. Mark?

Good morning, and thanks for joining us a bit earlier than usual this morning. In my remarks, I'll cover three main topics: our Q1 results, our latest view of this year's Semi Test market and our Industrial Automation investments, including the purchase of MiR, which was announced in a press release earlier this morning. Our first quarter financial results were at the high end of our guidance as conditions in most end markets, other than mobility, remained quite strong. In Semi Test, our Memory Test shipments reached a level of $73 million, a new record. Flash demand continued to grow, driven by the increasing adoption and increasing densities of high-performance NAND devices for mobile and data storage applications. We also saw record shipments into the DRAM wafer segment. In analog test, our Eagle product line shipment soared in the quarter on demand from industrial and automotive end markets. The one soft spot in Semi Test, and it was especially soft, was mobility test, where the expected demand did not materialize and the full year outlook declined. More on that in a few moments. Universal Robots growth in the quarter was on target, with sales up 34% from last year's atypically high first quarter. UR remains on track to increase sales at 50% or more for the full year. Sales growth at UR comes from both increasing our penetration of existing markets and establishing beachheads in new markets. For example, in automotive assembly, which we've served for many years, our new application uses UR cobots to vacuum the inside of cars on the production line before their interiors are installed. A new example is in the entirely new market of banking. UR cobots are now operating banknote sorting and counting machines, eliminating adult task for human operators while improving accuracy and…

Thanks, Mark, and good morning, everyone. I'll provide some brief comments on 2018, cover the financial aspects of our recent Industrial Automation M&A moves, and then cover the first quarter results and second quarter outlook. As you can see, 2018 has opened with first quarter sales of $487 million and non-GAAP EPS of $0.45, followed with softer second quarter guidance than expected. I'd like to first flesh out the short-term picture of both in terms of our updated 2018 SOC Test market estimate and the Universal Robots first quarter sales comparisons. Starting off with semi test, as Mark noted, we now estimate the 2018 SOC Test market to run between $2.2 billion and $2.4 billion, down about 12% at the midpoint from our prior estimate. What appeared earlier this year to be a delay in mobility demand may now shape up as more of an off year, with recent reports of lower complexity growth for a significant portion of the smartphone market and a near-term focus on lower cost. This pause in complexity growth is reminiscent of the tick-tock pattern we had a few years back, where every other year the degree of performance jumps as measured by transistor count growth oscillated quite significantly, which in turn moved our semi test sales up or down about $200 million a year. While we can't be certain, we may be witnessing this prior pattern reemerge, at least in the short term. I should also add that last year, our initial SOC Test market size range was $2.2 billion to $2.5 billion, and the market ended the year at $2.7 billion. As we have previously noted in our quarterly calls, tooling cycles will continue to experience quarterly and annual volatility. While these annual ebbs and flows of market size will continue, they don't…

Thanks, Greg. Zekenia, we'd now like to take some questions, and as a reminder, please limit yourself to one question and a follow-up.

Your first question comes from the line of Vivek Arya with Bank of America Merrill Lynch.

Thank you for taking my question. If I heard you correctly, I think for 2018, you're suggesting roughly $2 billion plus/minus in terms of the top line outlook. Is it fair to say a lot of the delta versus consensus expectations has been captured in Q2, or should we anticipate any other trends in your mobility business in the second half?

I think that if you look at what we said, meaning that the first half of this year and the second half of this year are roughly going to be 50/50 in our annual revenue, that you can do the math there and get close to what you just cited. We don't expect any further weakness in mobility. I think this is a pretty isolated case, and the rest of the year and most all other segments should be strong.

Thanks. And as a follow-up, as your business mix continues to shift more towards robotics, how should we think about just the overall growth contribution from robotics right with these new acquisitions that you have made? They seem to be obviously of a smaller base, but growing faster than the UR business, so growth outlook there as well as what it means in terms of operating profitability level?

In terms of the Industrial Automation segment, we've actually updated our model slide that's in the deck. Previously, we just had Universal Robots and we had 45% to 50% growth through 2021. Now we've upped that to 50% to 55% growth with these other new additions. MiR is certainly growing at a higher rate, as it's early in its life cycle. So we see the Industrial Automation business in 2021 getting to $900 million to $1 billion, so it should be a significant part of Teradyne's business. The profitability rate last year was 19%. This year, we're making aggressive investments, so the profit rate could be a little bit less. It may still end up close to 19%. But over time, we would expect the profit rate to move up into the low to mid-20s.

Okay, thank you.

Your next question comes from Atif Malik with Citigroup.

Thank you for taking my question. In your prepared remarks, you commented mobile is lower complexity growth this year. Application processor opportunity is a complex function of optimization of die size through shrink and increasing transistor count. As you look at the ramp down of 10-nanometer this year and ramp up of 7-nanometer and then 7-nanometer plus 5-nanometer the next couple of years, do you expect complexity for your test business to return for mobility testing? And then I have a follow-up.

Yes, I do think that what we see in the aggressive move to finer geometries and the transistor counts will keep propelling complexity in mobility and test time. So everything we see from today would say that that general trend is intact. And as we said in our prepared remarks, for most of this decade, what we had seen customers doing is have a fairly aggressive year of complexity growth, followed by a more modest year of complexity growth. 2017 was a bit of an exception to that rule. And by all measures, as far as we can see, we may be just reverting back to that trend.

Okay. And then, Greg, I don't know if I heard a number for the wireless test market this year. In January, you're expecting that market to be flat year over year. Does this change your plans to potentially divest this business?

We don't have plans to divest the business, the Wireless Test business. Let me get that one first. The market in Wireless Test is likely to be a little bit smaller this year, slightly smaller. Some of the 802.11ax introductions have slipped out in time, the new wireless standard. But Wireless Test will be profitable. And it's all about positioning for the buying waves that occur now and next year for 802.11ax and, longer term, in 2020 for 5G.

Thank you.

Your next question comes from Toshiya Hari with Goldman Sachs. Toshiya Hari - Goldman Sachs & Co. LLC: Great, thank you so much. My first question is on the Semi Test business. Mark, you talked about lowering you SOC Test TAM by $300 million. I just wanted to confirm that that was all mobile, and your view on Analog Test and MCUs and so on and so forth haven't changed over the past three months.

That's correct. All of that is mobility, and the other segments are quite strong. Toshiya Hari - Goldman Sachs & Co. LLC: Okay. And sticking to Semi Test, you talked about share gains on the wafer test side and memory. Can you give us an idea where market share is today, where you see that going over the next couple of years? And curious, have you seen any kind of retaliation from your nearest competitor given that it's a pretty important business for them?

Yes, so right now we're in the mid-20s, just 27% – 28% market share coming out of last year. And as we open up the wafer test portion of the market, our target by 2021 in our midterm plan is to get close to 40% share roughly. In terms of retaliation, we've been very careful on how we've gone after market share gains, both in SOC and in memory, so that we don't precipitate that kind of retaliatory move. The current environment in memory test, being in a very high-growth phase and investment phase, has given a rising tide to lift all boats, let's say, despite some of the share gains that we're achieving here. So I don't see anything happening, that silliness in the environment of the Memory market, but we are making good progress on wafer test this year, and our target is 40%. Toshiya Hari - Goldman Sachs & Co. LLC: Okay, and then I had a quick follow-up on MiR. Congrats on the acquisition, by the way. I was just curious, if you can talk a little bit about the competitive landscape there. I know on the UR side, you've talked about having 60%-plus market share. I'm curious what the positioning there is for MiR and how you're different from both the hardware and software perspective relative to your peers. That would be great. Thanks so much.

I'll take a crack at that. MiR is even a little bit murkier than Universal Robots. There's less third-party data out there. But our research has indicated that MiR is the leader in the industrial segment of mobile robots. And they probably – it's hard to put a number on the market share, but they're number one, and it's a number we can't be precise with, but we don't see a close second at the moment. So we have an early lead, but it's very early. And what MiR has done, similar to what Universal Robots did, it's the right price point. It's very easy to use and navigate, and it maps its own environment, so you don't need to be an engineer to deploy it. And we have a whole bevy of distributors who signed up, trained, out in the field providing these cobots. So it does have a lot of the Universal Robots first-mover advantage, the first mover with a reliable product. So that's how I'd ask you to try to think about it. We hope to get more market data over time, but it's sketchy at this stage. Toshiya Hari - Goldman Sachs & Co. LLC: Thank you.

Your next question comes from Mehdi Hosseini with Susquehanna International Group.

Yes, thank you. Just a follow-up to the acquisition of MiR, it seems to me that you're paying about five times forward-looking sales ramp, given you a very generous assumption with the forward-looking sales, and the five times seems to be at a much higher valuation premium compared to when you acquired or announced acquisition of cobot. And I want to follow up with you. What's the rationale for paying the premium? And also, whether you can use the existing infrastructure in terms of manufacturing and your channel to scale MiR. I think, Greg mentioned that the longer term operating margin could be adversely impacted. And is that really driven by the changes that you need to make to the infrastructure to be able to scale MiR? So the question has to do with the premium applied and how you're going to be able to scale that.

Okay, Mehdi, maybe something was lost in translation. I'll get to the operating margin thing last, but there's no issue there. But starting with the premium, we did the traditional valuation analysis with a football chart. We looked at it multiple ways, and the multiple is somewhat similar to Universal Robots, but they're growing at a much higher rate than what Universal Robots was growing at, significantly higher. So it's earlier, growth rates are higher. If you look at Universal Robots as a case study, what it's done to Teradyne's value versus what we paid for, that's been a spectacular return. We expect MiR will do the same thing as well. And I think it's hard sometimes with valuation models when businesses are growing more than 50% a year, it's hard sometimes to get traditional valuation metrics and see how they fit because there aren't many businesses that grow at these super high rates. And they grow into these valuations very quickly and then they are substantially higher than what we paid for them, which is what happened with Universal Robots. The next thing on the operating margin, there's no deterioration in operating margins. The model we put out, we have the same operating profit rate in the future at 25% to 27%, with higher sales given the addition of MiR and Energid, so I apologize if something got lost in communication. I think the only thing that's happening is in the near term, we're investing aggressively in MiR and Energid, and they won't have the same profit rate as Universal Robots because they're early, they don't have the same scale; but over time, they will, and they'll contribute spectacularly to our model, which will have about 30% coming from Industrial Automation with very good margins on profit rate. In…

Actually, it was one question, three parts. May I ask one follow-up, quick follow-on SOC Test?

Make it quick. Yeah.

Yes.

Okay. Back in the January, you revised SOC upward. Now we're going backward. Isn't the problem nothing to do with the off-year, on-year? I thought we took that off table. Maybe perhaps the problem has to do with just the end market and until 5G is ready to go, the SOC Test market is going to be adversely impacted and the growth rates are not going to be as much as we thought it would be?

I don't see it that way. The market unit growth in smartphones has not – it's kind of plateaued three or four years ago. And so what's been propelling the complexity growth since then isn't new wireless standards and such, it's more compute power behind more features in the phone. Now 5G will certainly kick in another round of bandwidth growth in the phones, which is additive on top of what we see in the past few years. So the lack of new wireless standards for the past three or four years, I don't think has impacted the Semi Test market. It certainly impacted LifePoint but I think 5G will be additive. And the feature growth and the complexity growth of the applications processors, power management ICs and such around that will continue to grow on average.

Can we have our next question, please?

Thanks so much.

Your next question comes from Timothy Arcuri with UBS.

Thank you very much. So I have to say, I'm a bit confused on two fronts. First of all, on the SOC weakness, it seems like something else is happening because if you assume that most of the Wireless Test business is with your big customer, that would leave, based upon what was in the K last year, that would leave maybe $375 million to $400 million for that customer for your Semi Test business. And you're cutting about $300 million from the SOC TAM, so that's about 75% to 80% of that customer's test revenue. So I guess, like that's a much bigger number than you'd think given what's going on in the supply chain, so are there other customers cutting as well, number one? And second part of that question is, does that come back in 2019? And then I had another point that I'm also confused on. Thanks.

So one point is that $300 million is not – although a lot of it, most of it is Teradyne, it's not all Teradyne. So that's one piece of the math. And then the average buying that occurs, the concentration of buying that occurs in our mobility segment has a lot of variability in it year to year. And if you do the math, coming off of 2017, at least, the numbers that you cited, are slightly higher in terms of – if you were to take the full $300 million out of Teradyne, you would still end up with a higher number than the bottom number that you suggested. So the fact that it's not $300 million for us, it's less, and we're coming off of a high 2017, I think neutralizes your math. And longer term, I'll go back to say that I think we've seen a pattern this decade of aggressive complexity growth, followed by something more moderate. This doesn't look much different than that so – and we know what is in development in terms of lithography nodes and such. So I don't think this is a reset or a change to the trend line, it's more a reversion back to what we've seen for most of this decade.

Okay. Then I guess the second point is just on the model, so you guys just are spending $124 million to acquire MiR, and you're spending another $25 million. So you're spending like $175 million to acquire these two companies and possibly $300 million if that earn-out comes through. You did drop another $150 million to the op model in revenue for 2021. But the EPS number didn't change, and it's still the range. I mean, maybe this is all within the EPS range. But it's sort of is a bit of an odd message that you'd be spending up to $300 million to buy two companies and you wouldn't change your 2021 earnings model? Thanks.

Yes, Tim, we looked at that, and we thought about changing it, but we just put instead, addition of MiR and Energid, helps accelerate mid-term EPS target. We thought that was simpler that perhaps we could pull it in with these moves, but we could have changed it a bit, but we just opted not to, but it should improve, there's no doubt. There's no message we're trying to send to you that we didn't improve it, but we just chose to go about it differently.

Okay guys. Thanks.

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

Yeah. Good morning. Thank you for taking my question. I guess a follow-up question on the mobility side. And so as you think about 2018, would love to hear your thoughts in terms of the pullback in spending. Was it primarily isolated to the AP arena only or did you also see a falloff in image sensor, power management, et cetera? And then I guess more importantly, as you look to 2019 and you think about the world transitioning or continuing to transition 10 to 7 and 7 to 7 plus in adopting EUV, and I know it's extraordinarily early today, but how are you thinking about a tick up in SOC into next year?

Yes. So I think around the components that make up the mobility space, the predominant issue this year is AP-related. Interestingly, things like image sensors that you mentioned, PMIC, in fact, are stronger this year, so that's slightly offsetting some of the weakness in AP. So it's not broad-based mobility, it's a pretty focused issue. And as again, we look forward into next year with some of the aggressive designs and aggressive lithography nodes moving forward, people are doing this to get more complexity. They're not doing it to get lower cost. And we expect, therefore, this trend will come back, and we will see decent complexity growth from the future. It's just a very difficult thing. It's been true for decades that in any given short year, the predictability of the SOC market is very difficult. So we tend to look at the longer-term trend lines. This certainly surprised us, but it's not outside the bounds of what's normal in the SOC Test market.

That's helpful. And as my follow-up, as you think about the inclusion of these two acquisitions into your Industrial Automation bucket, what kind of growth should we expect for that business here in calendar 2018? And what would the linearity look like through the year?

The MiR acquisition, we expect, will grow from – last year, it was $12 million, but obviously it's not in our numbers. It was $5 million in the first quarter. We think for the calendar year, it'll be about $30 million. So $30 million minus $5 million is what you put in for the last three quarters skewed towards the fourth quarter. So you've got about $25 million in our numbers. Energid will be small, maybe $6 million, of which maybe $5 million is in our period going forward. But MiR is the high grower. Energid is more of a company that has capability that can help us sell more cobots long term.

And just to be clear, those revenues are embedded in your first half, second half similar revenues?

Yes, it's in that because it's not that large in the grand scheme of things. It's not going to move that conclusion that the first half and second half are about similar, because you're adding under $30 million, certainly under $30 million in the second half because you've got a second quarter, too.

Great, thank you.

Your next question comes from the line of Farhan Ahmad with Credit Suisse. Farhan Ahmad - Credit Suisse Securities (USA) LLC: Hi, thanks for taking my question. My first question is related to margins. Historically, whenever you have a lower mix of mobility, in general, what I have noticed is that your margins – on the gross margin tend to be a little bit higher. I'm just a little bit surprised that we're getting around 55% this year versus more like 57% – 58% that we see in low mobility years. So I just wanted to understand that piece a little bit better.

There were mobility shipments in the first quarter, so that perhaps explains that. But there won't be as many shipments in the second quarter as a percent of the mix, and that's why the second quarter moves up to 57%. But you are correct that there's heavy mobility buying. Because there's large volumes, there's better pricing. Farhan Ahmad - Credit Suisse Securities (USA) LLC: Got it. And then as we think about 2019, you mentioned that the pattern of tick-tock might be applicable to smartphones. But I'm just wondering given that we aren't having a big change at TSMC in terms of 7-nanometer to 7-nanometer plus next year and in this year, which is going from 10-nanometers to 7-nanometers, which is the same lithography node. So basically for three years, we are at the same lithography node. So why should we expect that there should be a dramatic increase in complexity next year?

I do think next year you will see a much larger proportion of 7-nanometer utilization, and that's what will drive that change. Farhan Ahmad - Credit Suisse Securities (USA) LLC: Got it. And just one clarification, can just you talk about if the change in mobility was pretty much driven by one customer or if it was coming from diverse end markets?

I think all I will say on that is it was primarily driven by AP, and it's pretty concentrated. Farhan Ahmad - Credit Suisse Securities (USA) LLC: Got it, thank you. That's all I have.

Your next question comes from the line of Edwin Mok with Needham. Edwin Mok - Needham & Company, LLC: Great, thanks for taking my question. First, on Semi Test. So tests are typically reusable (50:18) with weakness in mobility. Do you see any customer start to probably use all that capacity for other test space and that might have an impact on your demand on also mobility or mobile SOC?

No, and it turns out that the capacity utilization for the current fleet of mobility testers is quite high and it's being added to. So it's not like there's nothing happening in mobility. It's just down quite a bit, the capacity adds. So it's highly utilized. Additional systems are being put in place this year, just fewer than we had expected. Edwin Mok - Needham & Company, LLC: Okay, great. Thanks for clarifying that. And then on MiR, I guess I have a two-part question. First is, you said the market is around $50 million, right? Does that include warehousing and other things that people use those kind of robots, or is it just for manufacturing automation? I remember, I think Amazon bought a company around five years ago and they paid, what, $700 million for that. And my understanding at that point in time they were talking about a market much bigger than $50 million. So I'm just trying to understand if that includes the warehousing part which I think is what Amazon is using it for. And then longer term, do you see that, frankly, do you see Amazon potentially become a competitor for that business long term?

So first of all, the market that we are describing for MiR is not the warehouse logistics market that would be something like Amazon Robotics would serve. It's industrial manufacturing applications primarily. There are also some applications in the healthcare segment for the mobile platform. But it's not logistics and big warehousing markets. On the other hand, the market is quite new and early because the sweet spot of this market is for the guy who needs to incrementally automate material movement in his factory. They're not required to build a new factory or retool the layout of the factory. They can incrementally go in and change the way they move material around through automation. So I think it's a very different market. It's very similar to what we saw with UR. UR wasn't going after the large established robotics applications. It was more the small to medium-sized enterprises. In the case of MiR, it's probably more medium to large-sized enterprises because those tend to move more material around in their factories. Edwin Mok - Needham & Company, LLC: Okay. But do you think there's opportunity for you to go after warehousing market long term?

I think rather than warehousing, it will be more, let's say, distribution centers for companies that may need to aggregate and disaggregate shipments to their local stores or suppliers. So rather than, let's say, an Amazon or a DHL or those kind of areas, it might be more a regional distribution center for a consumer goods product or for a foodservice product that would use the product. Edwin Mok - Needham & Company, LLC: Okay, actually that's very helpful. Thank you. I appreciate it.

Your next question comes from Patrick Ho from Stifel. Brian Edward Chin - Stifel, Nicolaus & Co., Inc.: Hi, this is Brian Chin on for Patrick. Thanks for letting us ask a few questions. First, in 2018, the memory market is poised to be roughly double where it was pre-2017. Do you view the current market size a sustainable; i.e., a higher new norm, or is it still prone to cyclicality? And can you also clarify what SOC market size is embedded in your 2021 target model and whether this has changed any since your January call? And I had one follow-up on the acquisition.

Okay. So on the memory market size, yes, there's been a significant increase in the memory market these past few years. I do expect it to remain volatile. So I don't think we're at a new plateau that's going to sit up at this $800 million to $1 billion level without sort of a dip. But I do think that there's, on average, if we look forward, let's say, for the next five years in memory, it will certainly operate above that sort of $500 million level it operated at earlier in the decade. It may be operating on an average closer to the $700 million to $800 million. But that doesn't mean that there couldn't be a year in here where it drops back to $500 million. That's been the nature of the memory market forever, and I would expect we'll see that again. And then in terms of what's embedded in our market size?

The 2021 model, we have in the range for ATE market size the $3.3 billion to $3.5 billion. The SOC is $2.7 billion at the low-end, $2.9 billion at the high-end. By comparison, 2017, the SOC Test market was $2.7 billion. So there's – at low-end, there's no growth for 2017 or very low growth from 2017 if you go out to the high-end, $2.9 billion. Brian Edward Chin - Stifel, Nicolaus & Co., Inc.: Okay. Great. And then for my follow-up on the MiR acquisition. Can you maybe want to describe any existing integrated workflows at customers with the UR robots? And then also, how overlapping are the existing distribution channels between UR and MiR?

So there are some applications today where the UR cobot is mounted on the MiR mobile robot. It is by far the exception, not the rule. And there's opportunities to expand that where you need to have a mobile platform dispatched to other location and perhaps be able to use the cobots to load or offload material under the mobile robot. But there's usually much different conveyor belts or even humans at the various loading and unloading stations to do that. So I don't think that's going to be a huge trend, but it'll be a growing sub-segment of the market. In terms of the distribution overlap, today, there is a significant Venn diagram overlap between the distributors for UR and MiR so – and there's a lot of synergies in the selling story because the value proposition is very similar. So we do see common customers, for sure. I would say that in general though, MiR's customers tend to skew toward medium to larger enterprises, and Universal Robots is a little small to medium. Brian Edward Chin - Stifel, Nicolaus & Co., Inc.: Okay, great. Very helpful. Thank you.

And operator, we have time for just one more question please.

Your final question comes from David Duley with Steelhead.

Yes. Thanks for taking my question. I guess first question I have is, you mentioned in the memory market it being an $800 million or $900 million size now. Could you help me understand what part of the market is wafer level versus final tests? I'm trying to figure out how much your SAM increased by moving into wafer level test?

It's roughly – if you look at the market, it's 50/50. Between NAND final test, DRAM final test and then the wafer test of those two parts is 50/50.

Okay. So this is essentially adding half of the market to your served available market by moving into the wafer test?

Correct, correct. So roughly said, each market this year is $200 million, we've been very, very strong in NAND final test and have a high share there. We're opening up another $400 million market with the move to wafer test.

Will that be mainly in NAND or DRAM?

It will be – in terms of what we participate and I think it will mainly be in NAND, flash or wafer test. There is some business we're getting in DRAM wafer test, but I think the bigger piece of that will be NAND.

Okay. A final thing from me is, could you just help me understand with the size of the APU market was in 2017 and in 2018? I guess it's a follow-on to an earlier question because it seems like the cut in that segment is pretty big.

Yeah. I don't know that we have it broken down by APU.

Right. We have the mobility in total. Perhaps offline, Andy, can speak to you, maybe figure that one out.

Yes, the mobility market came from about $1 billion.

He was trying to get APU.

Yeah, I know. We don't break out AP specifically, but the mobility in total was about $1 billion and this year we think it's going to be down $300 million or so, lower than that.

Yeah. I think just – I've got some numbers in front of me. Roughly, for mobility, from $1 billion to $750 million. So obviously, in that, AP is a big piece of it, but that's mobility.

All right, thank you.

All right. Folks, we're out of time. Thanks for joining us a little bit earlier today. I look forward to talking to you in the days and weeks ahead. And for those in the queue, I'll get back to you this morning. Thanks so much.

This concludes today's conference call. You may now disconnect.