Thank you, and welcome everyone to FormFactor's Third Quarter 2019 Earnings Conference Call. On today's call are Chief Executive Officer, Mike Slessor; and Chief Financial Officer, Shai Shahar. Before we begin, Jason Cohen, the company's General Counsel, will remind you of some important information..

Thank you. Today, the company will be discussing GAAP P&L results and some important non-GAAP results intended to supplement your understanding of the company's financials.

Reconciliations of GAAP to non-GAAP measures and other financial information are available in the press release issued today by the company and on the Investor Relations section of our website. Today's discussion contains forward-looking statements within the meaning of the federal securities laws.

Examples of such forward-looking statements include those with respect to the projections of financial and business performance, future macroeconomic conditions, foreign exchange rates, business momentum, business seasonality, the anticipated demand for products, customer requirements, our future ability to produce and sell products, the development of future products and technologies and the assumptions upon which such statements are based.

These statements are subject to known and unknown risks and uncertainties that could cause actual results to differ materially from those expressed during this call.

Information on risk factors and uncertainties is contained in our most recent filing on Form 10-K with the SEC for the fiscal year ended 2018 and our other SEC filings, which are available on the SEC's website at www.sec.gov and in our press release issued today.

Forward-looking statements are made as of today, October 30, 2019, and we assume no obligation to update them. With that, we will now turn the call over to FormFactor's CEO, Mike Slessor..

one, the increased emphasis on optical technologies such as silicon photonics, VCSEL, MicroLED and image sensors; and two, the secular growth opportunities driven by advanced packaging.

Shai will provide some more financial details in a moment, but we're excited about the potential to inject FRT's leading technologies and products into FormFactor's longstanding partnerships with the top semiconductor customers worldwide. Finally, with average lead times of less than a quarter, our visibility remains limited as always.

But as is reflected in our revenue outlook range, we're encouraged by the broad-based strength in our diversified consumables and R&D-driven demand profile.

This strength is being demonstrated in 2019, where we are experiencing strong year-end demand and expect to deliver annual growth in a year where equipment spending is forecasted to be down double digits.

By capitalizing on structural opportunities like advanced packaging and 5G, we are making progress towards our target financial model, growing the top line to $650 million while delivering $1.25 of non-GAAP EPS and $110 million of free cash flow. One more note before I turn the call over to Shai.

During the quarter, we welcomed Rebecca Obregon-Jimenez to our Board of Directors. With 25 years of engineering and executive experience in packaging and test, Rebecca brings a wealth of relevant technology and market knowledge to our boardroom that will help us navigate to our target financial model and beyond. Shai, over to you..

Thank you, Mike, and good afternoon. As you saw in our press release and as Mike noted, our third quarter revenue and gross margin were at a midpoint of our outlook ranges, and our EPS exceeded the midpoint of our outlook range.

These results again show the benefits of our diversified leadership in probe cards and engineering systems which dump in cyclicality and enables us to participate in industry spending on both node transitions and new design releases.

FormFactor's revenues for the third quarter of 2019 were $140.6 million, a 1.9% sequential increase; and a 4.2% increase over Q3 2018. Probe card segment revenues were $116.4 million in the third quarter, an increase of $2.8 million, or 2.5% from Q2 2019. System segment revenues of $24.2 million in Q3 were basically flat with the second quarter.

Within the probe card segment, Foundry and Logic revenue decreased 6.8% from Q2 to $68.4 million and was 49% of total company revenue in Q3, down from 53% in the second quarter. DRAM revenues were $39.4 million in Q3, an increase of $3.4 million from the second quarter, and were 28% of total quarterly revenue as compared to 26% in the second quarter.

Continuing the strength demonstrated over the past 2 years and in contrast to generally weak industry results, this was our highest quarterly revenue from DRAM since Q1 2008. Flash revenues of $8.3 million in Q3 were $4.1 million higher than in the second quarter and were 6% of total revenue in Q3, up from 3% in Q2.

Approximately $5.6 million of the flash revenues in Q3 were from NAND flash applications. Consistent with our opportunistic approach to this market, we continue to expect flash revenues to be lumpy. GAAP gross margin for the third quarter of 2019 was $55.3 million, or 39.3% of revenues, 80 basis points lower than the 40.1% GAAP gross margin in Q2.

Cost of revenues included $5.8 million of GAAP to non-GAAP reconciling items, which we outlined in our press release issued today and in the reconciliation table available on the Investor Relations section of our website.

On a non-GAAP basis, gross margin for the third quarter was $61.1 million, or 43.5% of revenue, 80 basis points lower than the 44.3% non-GAAP gross margin in Q2 and at the midpoint of our outlook range. The decrease from Q2 was a result of less favorable product mix, partially offset by better factory utilization.

Our probe card segment gross margin was 41.3% in the third quarter, a decrease of 140 basis points compared to 42.7% in Q2. The decrease is due to the factors I described earlier. Our Q3 systems segment gross margin was strong at 53.9% as compared to 52% in the second quarter. The increase of 190 basis points was driven by more favorable product mix.

As mentioned in prior earnings releases, we expect our systems segment gross margin to be in the high-40s to low-50s range. Our GAAP operating expenses were $46 million for the third quarter, $0.4 million lower than in the second quarter.

Third quarter operating expenses included $7 million of GAAP to non-GAAP reconciling items, similar to the $6.8 million of reconciling items in Q2. Non-GAAP operating expenses for the third quarter were $39 million, or 27.7% of revenues, compared to $39.6 million, or 28.7% of revenue in Q2.

The decrease of $0.6 million is mainly due to lower R&D spend related to timing of projects and lower sales and marketing expenses during the quarter, partially offset by higher performance-based compensation.

Company noncash expenses for the third quarter included $6.1 million for the amortization of intangible assets, $6.5 million for stock risk compensation and depreciation of $4.4 million. Amortization of intangible assets was $1 million lower than in Q2 due to certain intangible assets reaching full amortization during the second quarter.

Stock-based compensation was $1.2 million higher than in Q2 due to the timing of annual grants. GAAP net income for the third quarter was $8.3 million, or $0.11 per fully diluted share, compared to GAAP net income of $6.9 million, or $0.09 per fully diluted share in Q2.

The non-GAAP effective tax rate for the third quarter of 2019 was 23.8%, 250 basis points lower than the 26.3% in Q2 due to certain discrete items recorded in the quarter and in line with our previously communicated estimate of approximately 25% for the year.

As a reminder, our cash tax rate is expected to remain at 5% to 8% of non-GAAP pretax income until we fully utilize our remaining $180 million of U.S.-based NOLs and R&D credits. Third quarter non-GAAP net income was $17.3 million, or $0.22 per fully diluted share, compared to $16.1 million, or $0.21 per fully diluted share in Q2.

Moving on to the balance sheet and cash flows. We generated $25.6 million of free cash flow in the third quarter, compared to $29.8 million in Q2, taking our total cash and investments to $202 million at the end of the quarter.

The decreases in free cash flow in the third quarter as compared to the second quarter was mainly a result of timing of payments and collections. We did not have scheduled principal payments on our term loan during the quarter, and the loan balance remained $46.2 million at quarter end.

A scheduled payment of $11.3 million was made on September 30 at the beginning of the fourth quarter. Our total cash balance exceeded the debt balance by $155.6 million, an increase of $22.6 million.

While paying down the term loan remains our first priority for using cash, M&A continues to be an important part of the strategy, and we intend to continue to deploy capital to acquire leadership positions and to expand our served markets as we did with the acquisition of FRT earlier this month.

We funded the FRT acquisition with a 3 years, EUR 21 million denominated loan, utilizing the low Euro-based interest rates to optimize our cost of capital. We invested $2.8 million in capital expenditures during the third quarter of 2019, bringing our year-to-date investments to $14.2 million.

At the current rate of investments, we expect to be at the high end of our annual capital spending plan of $16 million to $20 million for fiscal 2019. Turning to the fourth quarter non-GAAP outlook.

As Mike mentioned, we expect a strong year-end demand for Foundry and Logic probe cards layered on top of continued solid demand for our other products, which result in our Q4 revenue outlook to be in the range of $145 million to $155 million.

This includes a contribution of $2 million to $3 million from FRT, which was acquired at the beginning of the quarter. Although revenue is expected to be higher than in Q3 primarily from an increase in Foundry and Logic revenue, the specific design mix is expected to be less favorable.

These factors, partially offset by continuous expense control and good operational execution, lead us to estimate the non-GAAP gross margin for Q4 in the range of 42% to 46%. Non-GAAP earnings per fully diluted share for Q4 is expected to be in the range of $0.22 to $0.30.

Although accretive, the addition of FRT is not expected to significantly change our Q4 gross margins or EPS. A reconciliation of our GAAP to non-GAAP Q4 outlook is available on the Investor Relations section of our website and in our press release issued today. With that, let's open the call to questions.

Operator?.

[Operator Instructions] Your first question comes from the line of Craig Ellis of B. Riley..

Mike and Shai, congratulations on the strong financial performance and the outlook. Mike, I couldn't help but noticing that with the December quarter guide, it's the fourth consecutive quarter of both sequential and year-on-year growth for the business. So I wanted to use that as a takeoff point for a longer-term question.

As you look out to calendar '20, can you just help us frame what you think are some of the more powerful growth drivers for the business and the extent that you see any headwinds? Can you help us understand what those would be as we think about the coming year? Thanks..

Sure. Thanks, Craig. I think Q4 is an interesting data point for us. As you note, it represents sequential growth, annual growth, and a step up to a new operating level for us. In that context, we're hopeful but cautious. In a business that's a terms business with less than one quarter lead times, our visibility is limited.

On the other hand, as we mentioned in the prepared remarks, we are adding capacity in a flexible way. So maybe one way to approach this is to talk about the components of this step up.

Really, as we conveyed in the prepared remarks, it's an increased Foundry and Logic demand coming from both an overlap in the 10 nanometer and 14 nanometer nodes with our largest customer. That strength we probably think is reasonable to think about extending into the first half of 2020.

Beyond that, again, given the visibility limitations we have, very difficult to make any statements. But we would expect 14 nanometer to roll off sometime in 2020 and return that customer to kind of the $100 million run rate level. The second component present at work in the fourth quarter is the 7 nanometer foundry activity.

And I want to emphasize that this is at multiple foundries with multiple fabless customers on designs for both mobile application processors and high performance compute. So it represents, again, an increase in FormFactor's footprint in serving the foundry space at these advanced nodes, 7 nanometer in particular.

I think one of the things we're seeing there is kind of a construction of multiple designs at the same time which we might consider to be a little bit of a one-time pop.

The other thing that we're keeping our eye on is early in the 7 nanometer node, we would expect that test intensities are fairly high, which will drive an outsized growth toward spend.

Having said all of that, I think we're cautiously optimistic about 2020, retaining the flexibility to adapt our capacity and meet the demand levels that our customers are asking from us..

That's really helpful. And if I could as a follow up, I noticed in the geographic revenue breakdown that we had a particularly strong quarter in China in the third quarter. It seems that there's an upward trajectory to the longer-term trends, although there's quite a bit of movement quarter-to-quarter.

But can you give us some sense for how you see that business coming off of what was a very strong calendar third quarter?.

Yes. Craig, as you noted, China was a little bit over 20% of our revenue in Q3, higher than it has been. It's typically been in the high-teens for the past couple of quarters. As a reminder, that's composed of both the multinationals that operate in the region and the domestic Chinese semiconductor manufacturers.

If I characterize the here and now, I think FormFactor's winning on technology. Obviously, a lot of these customers need our MEMS technology to ramp, need our engineering systems technology to improve their yields.

And we're benefiting from a more than decade presence in the region with a strong team of local support, local sales and local applications that we've invested in over the years. So in the here and now, I think a pretty solid contribution. I have a hard time believing that this doesn't turn into a headwind at some point. As a U.S.

supplier, we are subject to tariffs. And the trade war is something that customers bring up in conversations as not an advantage for FormFactor, let's put it that way.

So I view this longer term as something that we're going to have to look at in terms of our strategy, but for now doing pretty well in growing the business as one of the few alternatives they have if they want to ramp this advanced semiconductor industry..

Your next question comes from the line of Brian Chin of Stifel..

Congratulations on the strong Q3 and a very healthy outlook. Maybe the first question, going back, I think we just talked a little bit about China and that it sounds like, Mike, that you're saying that, maybe, $30 million run rate is a hard place to stay at consistently as above where the business has been operating.

But I'm also curious on the DRAM side of the business, which isn't necessarily a complete discrete event from China, but obviously a very high intake. I have a question also about the capacity that you reference as well.

But kind of curious what trend you expect into Q4, and to the extent that business levels seem like they're still going to be pretty healthy even moving to next year, how you can characterize what visibility do you have maybe even beyond Q4?.

So I think maybe trying in particular with DRAM and then transitioning to DRAM more broadly, if I got the question right. There are components as some of the domestic China DRAM producers and the multinationals ramp more activity in producing DRAM chips in China.

There is a pretty healthy component of domestic China DRAM in our fourth quarter, and an encouraging sign where we are making inroads as that customer starts to ramp. I think DRAM more broadly, as we said in the prepared remarks, we see maintaining at the third quarter levels, which was a decade high for us in terms of DRAM revenue.

Some interesting dynamics at work where maybe customers aren't investing in capacity but are continuing to do node transitions and release new designs. And as we've seen in a whole bunch of different examples in our business in the past, that drives probe card demand.

And with the high test intensities at these new nodes and these complex new devices like DDR5 that operate at higher speeds, we're seeing an increased spend on FormFactor's probe technologies for DRAM not just in China, but more broadly overall globally.

As those nodes ramp and as those designs ramp, as we've said in the past and you've seen from our results, customers will work hard to get their test intensities down as their yields improve.

And so I don't want to create an expectation that we'll be continuing to ratchet up all time DRAM revenue records, but we do see this being a robust business driven by some pretty healthy fundamental drivers..

Okay. Great. And in terms of what you referenced in terms of maybe adding capacity, sounds like it will still be a heavy materials labor component, not just fixed cost. But CapEx has been running 3%, 5% of sales of late.

Can you give us a sense of where CapEx could go in the very near term, how quickly you can add some revenue capacity?.

Sure. So when we're talking about adding capacity in Q4, at this point we are mainly talking about direct label. And we did add some tools to support the increased level of revenue notable in Q4, but at this point mainly direct label, we're very flexible.

We continue to be agile with these resources, and we're not expecting to increase our capital plan over the $20 million or so that we talked about, $16 million to $20 million a year. In 2019 this is still our range, no change there..

Okay. Great. Maybe just one last question to circle back now on FRT.

Mike, can you give us a sense of the relative positioning historically of FRT just relative to lab configurations, higher volume fab configuration tools maybe in this past year? And then you can, moving forward, what you think that mix could more or less look like, and also what you think the steady state profitability on the business will be? And also maybe, just to throw one more question within that one question would be what you view the key differentiation of the product and technology?.

Yes. So on FRT, I think when you look at some of the attributes of this business, a lot of commonalities and resonances with our existing engineering systems business.

A couple of them, you've heard us talk in the engineering systems segment about increasing optical content in both the applications and the sensors and instruments that we're putting on the tools, either driven by things like silicon photonics applications, MicroLED, stuff like that.

Obviously, those are optical applications, but they require optical instruments as well. So we saw nice resonance with FRT there.

The business right now, primarily in the lab in R&D applications in a very similar way to our existing engineering systems business, and serving some interesting new advanced packaging and MEMS applications that are probably not all that well served by the big providers of metrology and inspection equipment.

So we feel like there's a nice differentiated niche position in FRT's ability to bring multiple different sensors in a flexible way primarily geared for R&D and pilot production to this environment of new applications being driven by advanced packaging..

Another question from the line of Quinn Bolton of Needham..

Nice job on the results and the strong fourth quarter outlook. I wanted just to start with the 7 nanometer demand. Obviously, you've seen that drive your fourth quarter outlook. But I guess if I look a lot at the press and talk to folks in the industry, it sounds like it's the 5G base bands that are driving a lot of that 7 nanometer demand.

And so I guess I'm curious if you're seeing any pull through for the 5G RF probe cards. It doesn't sound like it's coincident that it might be a quarter or 2 further out, but it certainly would sound that if they're building up a bunch of 5G base bands that the 5G filter opportunity for you shouldn't be far behind it.

So I was just wondering if you could comment on that opportunity. And then I've got a follow up question..

Yes. I think it's a very good observation, Quinn. Even if you look at, regardless of whether it's 5G or previous 4G handset releases, we do see this sequencing as we go through time where FormFactor has a pretty unique position in being able to serve both the apps processors, the base band modems and the memory that end up in these handsets.

That tends to be earlier in the cycle, call it 3 quarters ahead of the handset launch, where the RF front ends, the BAW/SAW filters, and for 5G the additional content, things like beam steering and phase arrays that are going to drive our business probably the middle part of 2020 if the historical cadence is about right.

So I think you've identified the correct sequencing or what we view is the historically consistent sequencing of RF probably being the last piece of our demand that comes to the show for a 5G handset.

And if you assume that there are going to be Q3, Q4 2020 handset releases, that should be a nice business for us in the middle part of 2020 from an RF perspective when we're beginning to work on the apps processors and the memory pieces right now..

Got it. Then I guess just a longer-term outlook. With DRAM now at a decade year high and running almost about $10 million above the quarterly level that I think you guys have embedded into your $650 million revenue plan, I guess I'm just trying to wonder.

Do you think that long-term DRAM outlook has improved and we should be thinking now about a long-term model that might not have DRAM at about $30 million a quarter but could be something in the mid-$30 millions or upper-$30 millions? Or is there something that you see happening in DRAM that will over time ever get back down to closer to that $30 million level implied by your long-term target?.

Yes, yes. You've hit on a topic of current debate internally. I think we have seen several quarters here where DRAM is at an elevated level compared to what we baked into the long-term model, and some of the maybe more recent results, call it 2018, 2017.

There's a couple of things going on, some of them that probably are persistent but maybe don't hold us at this level. The first one is the additional test intensity and complexity associated with advanced packaging. We've talked about HBM in the past being an adder to our fundamental DRAM business.

These are difficult chips to test, both electrically and mechanically, and play pretty well to FormFactor's competitive advantages. And so I think as long as HBM continues to accelerate and be a big part of our customers' wafer start mix, I think there's probably some potential upside there.

Having said that, DRAM historically over a multidecade period as been a bit of a fickle mistress. If you think about customers and their test intensities and how they really try and ratchet down their test costs, there have been times where we've seen them pull back on wafer tests significantly.

So some things that certainly are positives at the present that we see continuing with things like HBM, but some things to keep our eye on as well as our customers focus on cost optimization..

Your next question comes from the line of David Duley of Steelhead Securities..

I guess our first question has to do with RF test and probe intensity in a 5G phone handset versus a 4G handset. I guess there's a lot of increased content on the RF front end here in places that you've a lot of exposure.

So could you just talk about how much probe or test intensity there is in a 5G handset versus a 4G handset?.

Yes. An interesting question, David, and one that we don't have a crisp, quantitative answer to, but certainly directionally, as implicit in your question. There's more wafer test content at RF frequencies in a 5G handset for a couple of reasons.

First of all, operating at the higher frequencies of 5G, maybe not all the way to the millimeter wave frequencies, but things sub-10 gigahertz are a pretty demanding application. And so for things like the BAW and SAW filters that have been the meat of our RF business at 4G, we see those being a more complex test and therefore driving more intensity.

If you put on top of that some of the increased functional content, multiple input, multiple output, phased array beam steering to deal with the higher frequencies, there's some fundamentally new types of chips in an RF frontend. These are some of the things where I've talked about as working closely with the leading customers in the world.

We see that driving some increased content as well. It's so early in the cycle that, again, we don't have a good quantitative view of it. But there's going to be more chips, and they're going to be harder to test. So we do view this as a favorable trend for our RF test business..

Okay. Great. And one of your largest customers, the foundry in Taiwan, had a pretty interesting disclosure on their conference call about suggesting next year now that 15% to 17% of handsets are going to be 5G capable versus a quarter ago only 5% or 7%.

And I guess that that would suggest that we should see a really strong uptick in the business there on the handset side. And I'm just wondering what sort of visibility you have into that besides just them declaring that on the conference call.

Is there any other evidence that some of the large customers there are gearing up for that kind of rollout in 5G, which seems to be better than expected?.

Yes. I think in terms of tangible visibility, I'd have to answer no. Right, again, with lead times much less than a quarter, we're not going to see those demand signals until pretty close to when they need to actually be starting the wafers and shipping the chips to meet those expectations.

I will say the activity that we're going through in both engineering and pilot production with the customers that are going to drive the component growth in those handsets are indicative of 2020 production activity. Whether it's 5% to 7% of handsets or 15% to 17% of handsets, we don't really have a clear path to.

And our forecasts that we get from customers wouldn't really be showing that yet, but obviously something we look forward to try and serve..

Okay. Final question from me. You mentioned a bunch of detail about your foundry logic business at 7 nanometers, and that you had multiple application processors and customers. And could you just elaborate a little bit more about your previous comments there? It sounds like you've made progress signing up new customers.

So if you could just give us some greater detail, that would be great..

Sure. Well, I'll restate it and then try and layer on a bit of detail. What we said was the 7 nanometer fourth quarter demand was being driven in the foundry space by both leading edge foundries at 7 nanometer as well as multiple designs from their customers, the fabless semiconductor manufacturers.

Historically, at least over the last couple of years, our business with the world's largest foundry has been dominated by a single mobile application processor design.

And one of the things that we're seeing and we wanted to share was the notion that we're seeing that business at 7 nanometer broaden out more than just a single application processor from a single fabless customer. So in the fourth quarter, we're shipping multiple designs to multiple fabless customers at both foundries.

And we view that as a nice proof point of further diversifying that leading edge foundry business as we continue to grow our relationship with the 2 leading foundries in the world..

Your next question comes from the line of Tyler Burmeister of Craig-Hallum..

So you already touched on a lot of things, but kind of a bigger picture, longer-term term question, I guess. It sounds like the DRAM business is robust and expected to maintain that robustness, but probably not drive a lot of the delta growth between here and your long-term model.

So it sounds like on the Foundry and Logic side is where that growth's coming from. And with your largest customer, 14 nanometer rolling off, or expected to roll off sometime next year, I guess, if you could highlight what you're maybe most excited about as far as being the pieces for the top line delta between today and your long-term model..

Yes. Tyler, good question. I think we've talked about a couple of them in the questions we've had so far. But I want to highlight the general theme of advanced packaging as being one of them.

I think one of the reasons for the higher test intensity at the 7 nanometer node, at least for some foundry projects, is advanced packaging and fan-out in particular. To test those devices requires a more complex probe card and a more capable probe card, which drives the spend and how our customers compensate us for that tooling up a little bit.

So advanced packaging continues to be a key component of the bridge from here to the long-term target financial model. I think the other piece inside of Foundry and Logic is one we just touched on with David, and that's RF and 5G.

The content increases and the complexity increases are something that we're not seeing in our business right now in a major way.

But if I look at our R&D budget and our engagement with leading customers and the topics of our engineering review meetings, it's very focused on high frequency RF test and their need to increase that capability as it comes to production in 2020 and beyond.

And so I think those are probably the 2 pieces we're most excited about from where we sit now on the way to the target financial model..

Your next question comes from the line of Tom Diffely of D.A. Davidson..

I guess getting back on the DRAM side, I just want to address it from a different angle.

Now, how much of the slugged business you're getting right now is due to the fact that we added a lot of capacity a year ago and the timing of that versus just the natural design cycles?.

Yes. I think, so, the capacity that was added a year, maybe 18 months ago, I think the tooling associated with that, the probe cards associated with that, I think was probably part of what you saw in our midyear DRAM results.

The designs we're shipping now and plan to ship in the fourth quarter are really more associated with new designs, new architectures on that existing capacity that our customers have already put in place.

I think there's a very good analogy here to what's going on in DRAM now as to what went on at 14 nanometer with our largest customer maybe a year, year and a half ago, where they began to shift designs onto an existing capacity footprint at 14 nanometer. And you saw our business step up significantly.

Whether that continues or not is certainly an open question, but things like DDR5 and mobile DDR5 are nice pieces of design innovation from our customers that continue to drive new probe card demand.

So we're hopeful, as I said before, hopeful that this continues, but a little bit cautious in the capacity we're adding since we don't have a lot of visibility into those longer-term design roadmaps past the fourth quarter..

Okay. And I also noticed that over the last few quarters there's been a new 10% customer on the memory side, so really nice diversification there.

Do you think that makes that market less volatile over time?.

Yes. I think just first maybe on the 10% customer front, we've got 5 customers that typically are either 10% customers are very close to 10% customers, at least in 2019. As you note, we've seen now all 3 major DRAM manufacturers be 10% customers in recent quarters.

And one of the things we're seeing is their roadmaps and design release cadences are a little bit off cycle from each other. So it used to be they all shrank at once and released designs at once into a commodity market.

I think with some of the diversification of the DRAM end markets that's no longer the case, and so we're seeing a little bit of smoothing across each of the customers.

And as you note, if you look at our 10% customer list as it's wandered through 2019, you see each of the DRAM manufacturers popping on and popping off as their investments change quarter-to-quarter..

Okay. That's helpful.

And then on the foundry side, when the designs go from 10 nanometers to 7 to 5 nanometers, does that give you -- do you have a technology advantage that's greater at the lower nodes in that spectrum that might give you a larger share over time?.

I think there's a couple of factors. A lot of them really are dependent on the details of the design. But in general, I think you can say as a probe card goes from 14, to 10, to 7, to 5, it becomes a more complex thing.

Typically, our customers are trying to drive higher densities and more functionalities into their chips, which means higher test complexity in a more complicated probe card. So generally, we do see an ASP uplift.

Whether we're competing in Foundry and Logic or in memory against our major competitors, each of us have relative advantages compared to the individual design.

But I think as you look to the consolidation of market share towards the leaders who can actually serve these requirements, there's no question there's a trend where share is shifting to the, call it, the top 3 probe card manufacturers as things get more complicated with node transitions..

Okay. Great.

And then, Shai, on the model, when you add the extra capacity, what kind of impact will it have on the operating expenses on a go forward basis?.

The capacity we added is mainly in a cost of sales line, directly related to the increased revenue in Q4. And so no significant impact, or not impacted at all on the OpEx. And if you think about our Q4 OpEx, the range that we used to model the EPS, in between $39 million to $41 million, and that includes FRT. And so that's the run rate for Q4..

Okay. Great. And then finally, I missed the -- you mentioned the NOLs, but I didn't catch the number..

$180 million. That's both NOLs and R&D credits..

There are no further questions at this time. I would now like to turn the call back over to Mike Slessor for closing remarks..

All right. Thank you, everybody, for joining us today, and we'll talk to you in a quarter. Bye-bye..

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