FormFactor, inc (FORM) Q2 2021 Earnings Call Transcript

Image source: The Motley Fool.

FormFactor, inc (FORM 0.75%)
Q2 2021 Earnings Call
Jul 28, 2021, 4:25 p.m. ET

Contents:

• Prepared Remarks
• Questions and Answers
• Call Participants

Prepared Remarks:

Operator

Thank you, and welcome, everyone, to FormFactor's Second Quarter 2021 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.

Jason Cohen -- Vice President and General Counsel

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 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 laws.

Examples of such forward-looking statements include those regarding projections of financial and business performance, future macroeconomic conditions, the benefit of acquisitions and investments in capacity and in new technologies, the impacts of COVID-19 pandemic, the impacts of regulatory changes; the anticipated demand for products; 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 2020 and in 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, July 28, 2021, and we assume no obligation to update them.

With that, we will now turn the call over to FormFactor's CEO, Mike Slessor.

Mike Slessor -- President and Chief Executive Officer

Thanks, Jason, and thank you, everyone, for joining us today. FormFactor delivered solid results in the second quarter, achieving the second highest revenue in company history. We delivered gross margins comparable to the first quarter, exceeding our outlook range through a combination of factors that were more positive than anticipated on our April 28 earnings call. These, together with continued operating expense control and share repurchases resulted in non-GAAP earnings per share at the high end of our outlook range. Solid demand for FormFactor's diversified set of market-leading semiconductor tested measurement products is continuing in the current quarter, and we are executing on our planned investments to increase capacity for our products. Several factors have our second gross margins. First, overall product mix was more favorable than originally anticipated, with Systems segment revenues reaching all-time record levels at near 50% gross margins. In probe cards, DRAM was, as expected, a significant portion of revenue nearly matching the decade high level of fourth quarter 2019. DRAM probe parts are generally lower-margin products and a substantial portion of our second quarter DRAM revenue was concentrated in one specific design with an unfavorable cost profile.

In response, our operations and engineering team focused on yield improvement for this design and delivered better-than-expected manufacturing cost, which improved gross margins significantly above the initial shipments for this design. Our second quarter results provide a reminder that shifts in FormFactor's tech mix can have a significant impact on gross margins, both quarter-to-quarter and with lead times typically less than a quarter, even inter-quarter. Shai will provide more details on our gross margin results and outlook later. We continue to advance our environmental, social and governance or ESG initiatives during the quarter. Our successful efforts to aggressively reclaim and recycle our rhodium waste, reduce our environmental footprint while also producing a financial benefit as our higher recycling credits favorably impacted gross margins by partially offsetting higher input costs. On the governance front, we welcome to our Board of Directors, adding a director with significant industry and executive experience who's also well recognized for his hash commitment to diversity and inclusion. is the most recent of three new directors who have joined our Board over the last two years.

FormFactor's Board of Directors has become an ethnic and gender diverse team of accomplished individuals with world-class expertise in areas from cybersecurity to organizational development in addition to the requisite experience in semiconductors. Our ESG initiatives, including our progress against quantitative goals are discussed in the Corporate Citizenship section of our website. Turning now to segment and market level detail. Foundry and Logic probe cards, the expected sequential decline in revenues materialize. Even so, the top two foundries and the largest IDM remain customers and client PC, data center and mobile designs each contributed significantly to the top line. In the third quarter, we expect to see the topical seasonal reduction in mobile application processor demand partially offset by an increase in microprocessor demand and continued strong by the RF applications. Foundry and Logic customers, including all three of our second quarter 10% customers are making significant wafer fab equipment investments in 2021. As capacity comes online later this year and into early next year, we expect to see increased demand for leading-edge Foundry and Logic probe cards. Along with these capacity additions, our customers are adding innovative advanced packaging architectures like EMIB,

Foveros and 3D fabric to the road maps to help offset the slowing of front-end-driven Moore's Law. As we've discussed in the past, these chiplet or tile-based integration schemes drive both higher test intensity, which expands the number of probe cards required per wafer out, and test complexity, which widens FormFactor's competitive advantage. With lead times of less than a quarter, short-term visibility remains challenged as always, that the combination of significant customer capacity investments paired with the adoption of advanced packaging provide longer-term secular demand drivers for FormFactor's probe card products. To ensure we're prepared to meet this increased probe card demand, as I mentioned earlier, we're continuing to execute our planned capacity increases with our new 100,000 square-foot manufacturing center in Livermore on track to produce initial customer shipments in this year's fourth quarter. Consistent with the flexible manufacturing strategy that has allowed us to quickly and efficiently capitalize on customer demand across our served markets.

This facility will be capable of producing probe cards for Foundry and Logic, DRAM and Flash. The actual initial capacity addition will be modest as we plan to progressively add tools and labor to match customer demand. We're excited about the capability and flexibility this new footprint will provide, removing a growth constraint that we have operated under for the past 18 months, and helping us to achieve the $850 million revenue of our target financial model. Third quarter demand for DRAM cards continues to be strong, although we do expect a slight sequential reduction from high second quarter levels. New design activity continues at high levels across all DRAM customers with remarkable breadth across multiple nodes for DDR4 and DDR5 designs in both mobile and PC server application. As a reminder, probe cards are a consumable that is specific to each new chip design, and so demand is generated from not just node migrations, but also the release of chip designs on existing nodes and architectures.

As also noted earlier, our Engineering Systems business delivered record revenues with contributions from the acquired FRT and HPD businesses augment a steady legacy station business. FRT tools provide unique metrology capabilities that enable advanced packaging applications, and we're currently capitalizing on strong demand for FRT's high-performance multisensor 300-millimeter optical metrology tool, especially among the leading foundry customers. Advanced packaging is a key driver for all of our businesses as our customers begin to adopt chiplet-style integration strategies that generate new test and measurement requirements. Our capability to quickly, accurately and non-destructively measure and control critical dimensions, film thicknesses and surface topologies on chiplet-to-chiplet interfaces is key to improving and maintaining assembly yields. HPD's business targeted longer-term growth by enabling quantum computing applications also had good momentum in the quarter.

Together with Northrop Grumman, a key customer partner in the quantum computing field, we announced a fully automated cryogenic wafer probe system, which will accelerate the development of superconducting compute applications. Let me close by noting that with a continued solid demand outlook and good execution on both short- and long-term gross margin and capacity expansion on the path toward the target financial model, we unveiled last year that delivers $2 of non-GAAP earnings per share on $850 million of revenue. Test and measurement is becoming a more important and strategic place in the semiconductor industry, driven by powerful trend, including 5G, advanced packaging and memory content growth. Our leadership position in these attractive markets, paired with our differentiated strategy and disciplined execution to drive continued growth and share gains as we progress toward our target model.

Shai, over to you.

Shai Shahar -- Chief Financial Officer

Thank you, Mike, and good afternoon. As you saw in our press release, second quarter revenues and non-GAAP EPS were at the high end of our outlook range in revenue reaching the second highest level in company history and non-GAAP gross margin exceeded the high end of our outlook. FormFactor's second quarter revenues were $188 million, a 1% sequential increase in Q1 and an increase of 19% year-over-year. Probe card segment revenues were $154 million in the second quarter, a decrease of $5.2 million or 3.3% from Q1. The decrease was driven by lower Foundry and Logic and Flash revenues, partially offset by an increase in DRAM revenues. Systems segment revenues were $35 million in Q2, an increase of $6.7 million or 24% from the first quarter, mainly as a result of higher sales of optical metrology and thermal systems, driven by advanced packaging and automotive applications. Within the probe card segment, Foundry and Logic revenues decreased by $10 million from Q1 to $104 million in the second quarter, comprising 55% of total company revenues as compared to 61% in the first quarter. DRAM revenues were $42 million in Q2, a significant increase of $8 million or 24% from the first quarter and were 22% of total quarterly revenues as compared to 18% of revenue in the first quarter.

Building on the strength demonstrated over the past year, this was our second highest quarterly revenue from DRAM since Q1 2008. Flash revenues of $8 million in Q2 were $3.7 million lower than in the first quarter and were 4% of total revenues in Q2, down from 6% in Q1. As we said in the past, we expect Flash revenues to be lumpy from quarter-to-quarter. GAAP gross margin for the second quarter was 40.6% of revenues as compared to 41.1% in Q1. Cost of revenues included $7.2 million of GAAP to non-GAAP reconciling, 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 second quarter was 44.4%, slightly above the high end of our outlook range and 60 basis points lower than the 45% non-GAAP gross margin in Q1. The decrease from the first quarter was mainly due to less favorable mix with lower Foundry and Logic and higher DRAM revenues. Mike briefly discussed some of the reasons margin in the Q2 was above the high end of our outlook rate. They include a more favorable for mix, better yields on a new ramping design and improved utilization and absorption.

In addition, precious metal usage was more efficient as we took measures to reduce the impact of higher material prices and were able to reclaim and recycle more road units. Our probe card segment gross margin was 43.3% in the second quarter, a decrease of 90 basis points compared to 44.2% in Q1. The decrease is mainly due to the change in mix between Foundry and Logic and DRAM revenues I just mentioned. Our Q2 segment gross margin was 49.2%, essentially flat with Q1. As we've said previously, we expect our Systems segment gross margin to range between high 40s to low 50s. As a reminder, as we continue to make progress toward achieving our target financial model gross margin of 47%, we expect that margins will fluctuate from quarter-to-quarter, mainly as a result of changes in product mix. Our GAAP operating expenses were $56 million for the second quarter, $2 million higher than in the first quarter. Non-GAAP operating expenses for the second quarter were $48 million or 25.7% of revenues compared to $46 million or 24.9% of revenue in Q1. The increase of $2 million quarter-over-quarter is mainly due to increase in labor costs related to higher headcount and annual salary raises, partially offset by lower performance-based compensation.

Company noncash expenses for the second quarter included $6.6 million for stock-based compensation, $7.1 million for the amortization of acquisition-related amounts and depreciation of $6.6 million. Second quarter stock-based compensation was $0.5 million lower than in Q1 due to timing of grants and depreciation was $0.4 million higher due to additional assets placed in service. Non-GAAP operating income for the second quarter was $35.2 million, $2.4 million lower than the first quarter. GAAP net income for the second quarter was $17.9 million or $0.23 per fully diluted share compared to $19.6 million or $0.25 per fully diluted share in Q1. The non-GAAP effective tax rate for the second quarter was 18.7%, practically the same as in Q1 and within our anticipated non-GAAP effective tax rate for fiscal 2021 of 15% to 20%. As a reminder, our annual cash tax rate is expected to remain at 6% to 8% of non-GAAP pre-tax income until we fully utilize our remaining U.S.-based R&D credit. Second quarter non-GAAP net income was $28.4 million or $0.36 per fully diluted share compared to $30.8 million or $0.38 per fully diluted share in Q1.

Moving to the balance sheet and cash flows. We generated $16 million of free cash flow in the second quarter compared to $19 million in Q1, and we had total cash and investments of $260 million at the end of the quarter. The second quarter sequential decrease in free cash flow reflects an increase in capital expenditures and changes in working capital, of which the largest portion is higher inventories related to timing of manufacturing and product shipments. As of the end of the second quarter, we had two term loans remaining on our balance sheet, totaling $29.5 million. We invested $18 million in capital expenditures during the second quarter compared to $13.5 million in Q1. This brings our year-to-date capex to $31.5 million and chiefly relates to the capacity expansion in our Livermore manufacturing center. We continue to expect a significant investment in capacity in 2021. And with half of the year behind us, we are lowering the range for forecasted cash capex for the year to be $70 million to $90 million from the $80 million to $100 million previously communicated. As a reminder, we expect capex to return to 3.5% to 4% of revenues in our target financial model after we conclude these capacity expansions. At quarter end, our total cash and investments balance exceeded the debt balance by $230 million, a decrease of $10 million from Q1 quarter end.

We had strong cash flow from operations in Q2 offset by cash usage for capital expenditures and $18.2 million used for share repurchases. During the second quarter, we purchased 484,000 shares. This brings our year-to-date share repurchases total grew by 0.8% of our outstanding shares, offsetting approximately one year worth of dilution from stock-based compensation. At quarter end, $26 million remained available for future repurchases under our $50 million share repurchase plan. Turning to third quarter non-GAAP outlook. As Mike mentioned, we expect generally strong demand to continue with sequentially higher Systems demand, offset by moderate decreases in probe cards, resulting in overall similar revenues to Q2 at the midpoint of our outlook range. This factor resulted in a Q3 revenue outlook in the range of $182 million to $194 million. Non-GAAP gross margin for the third quarter is expected to be in the range of 43% to 46%. At the midpoint of these outlook ranges, we expect Q3 operating expenses to be higher than Q2 by $1 million to $2 million, mainly due to the increased investment in R&D and higher travel expenses as things start to get back to normal. Accordingly, non-GAAP earnings per fully diluted share for Q3 is expected to be between $0.31 and $0.39. A reconciliation of our GAAP to non-GAAP Q3 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 for questions. Operator?

Questions and Answers:

Operator

Thank you sir. [Operator Instructions] Your first question is from the line of Craig Ellis from B. Riley Securities. Your line is now open.

Craig Ellis -- B. Riley Securities -- Analyst

Thanks for taking the question. And Congratulations on the good gross margin execution in the quarter. Mike, and Shai, I just wanted to start there and understand what's embedded in the outlook for the third quarter? Because I think three months ago, we expected that in 2Q, we could see around 150 basis points of mix-related headwinds, maybe 100 basis points of rhodium-related headwinds. Clearly, the team executed very well on the utilization side and you got back some of the rhodium as well. But as we look to the third quarter, which looks flattish, what's inside that flattish guide with gives and takes on the mix and the rhodium issues? And for that efficiency gain that was achieved in manufacturing, is that something that rolls forward? Or is that more of a onetime benefit?

Shai Shahar -- Chief Financial Officer

Yes. Great. So when we look at Q3, the midpoint of our outlook range, gross margin is expected to be similar to Q2. We see higher Systems revenue in Q3, which, as you know, revenue -- gross margin for Systems revenue is higher. So this will help. But we also see some offsets in the probe card business, which impacted the gross margin. We still have the design that we're talking about. We still have it in Q3, the one with unfavorable costs. And although we are getting better at it, with improved yields and more certainty, it still has some impact on as well. So if you take all of these together, these are the main factors. We saw rhodium prices ease up starting in May, although they still fluctuate. So taking all of this together, basically, I'm talking about flat gross margin quarter-over-quarter with flat revenue at the midpoint of the ranges.

Craig Ellis -- B. Riley Securities -- Analyst

Okay. And then with the point on revenue there, Shai, all you said is a segue into a revenue-related question. So nice to see 3%, 10% customers in the quarter. with healthy activity from TSMC and Sensing. The question is more on Intel. It's been a while since Intel was down closer to the mid-teens percentage of mix, of course, at these revenue levels, that's still a very material amount of sales. But the question is, as you look at that opportunity and with the backdrop of the recent packaging and process webcast showing annual product cycles over the next four or five years. How do we think about the path back to kind of a normalized mid-20s to mid-30s percent of sales for that customer? How quickly could that happen? And what are the risks that we would actually stay here closer to the mid-teens for some protracted period?

Mike Slessor -- President and Chief Executive Officer

Yes, Craig, it's Mike. I'll take that. So prior to the second quarter here, we had several quarters where that customer was around and above $50 million or a $200 million annual run rate which we said we expected was kind of an elevated run rate due to a variety of factors. So in Q2, the pendulum probably swung a little bit hard in the other direction. We do embedded in our Q3 outlook, some sequential strengthening with that customer might comments about microprocessor strengthening in the quarter. But as you point out, longer term, there are some really exciting things going on with that customer. Long-term relationship, decades long where we've been a key supplier to them with strong market share. And both their strategic initiatives in entering the foundry business as well as their adoption of advanced packaging technologies that they highlighted earlier this week are really exciting opportunities for us that I do think allow us a path back to kind of this $200 million run rate. Now it's going to take a while to get there, right? They're not entering the Foundry business in a material way tomorrow. But we've got such a solid foundation with this customer and a nice competitive position with the technologies that we have that their strategic initiatives do offer us a longer-term path back to that $200 million annual run rate level, even as we work our way a little bit closer to there in the back half of the year.

Craig Ellis -- B. Riley Securities -- Analyst

That's really helpful. And maybe just a follow-up on that, Mike, it's been a couple of quarters since we had three customers that were 10% customers. Do you expect that as we look ahead to the third quarter that you would have an equal number of 10% customers? Or do you think that it looks ultimately more like Q1 where there was a couple?

Mike Slessor -- President and Chief Executive Officer

Well, Craig, as you know, our strategy really is to be partnered with the leaders in the industry and try and drive significant market share with those leaders. So we over the long term tend to have 5%, 10% customers. And by and large, we expect each of those customers to kind of move off and on the 10% list in any given quarter. Some of them are quite close to the 10% threshold. In the third quarter, as I noted, we do expect some seasonal weakness that we've seen in the past few years associated with the mobile application processor business. Just the ramp of those designs and the concentration of those designs at the world-leading foundry. We'd expect that to soften in Q3. But at the same time, we're making good progress in expanding that business to more fabless customers, more design, moving into high-performance compute, but mobile is such a big part of that, you could see that customer not be a 10% customer in Q3.

Craig Ellis -- B. Riley Securities -- Analyst

Yes. Yes. That's helpful guys. Really appreciate the help. Thank you.

Mike Slessor -- President and Chief Executive Officer

Okay. Thanks Craig.

Operator

We have our next question from the line of Brian Chin from Stifel. And kindly limit yourself with one question and one follow up. Your line is open.

Brian Chin -- Stifel -- Analyst

Okay. Hi. Thanks for taking the question. We'll do. Maybe first, just to follow up on that last one. Mike, I think you're talking about sort of the traditionally softer 3Q from Foundry. I guess applying that similar logic when 4Q typically be seasonally stronger for you? And I guess, do you think that -- is that part of the rationale in terms of having some of the new capacity online?

Mike Slessor -- President and Chief Executive Officer

So we have seen for the past couple of years, the fourth quarter will be very strong from a seasonal perspective in the Foundry and Logic business. If you go back and look at both '19 and '20, our fourth quarters very strong Foundry and Logic business, driven by the foundries and mobile applications. I have to temper that with the usual caveats that we're sitting here at the end of July, and that's a long way out compared to our lead times. So the historical pattern, certainly, we would expect to see that strength. But we don't have anything like a forecast or a backlog yet emerging that would support that. As you note, though, each of those customers is building significant capacity through their WFE spend. And so whether it's a 4Q event or a first quarter, second quarter 2022 event, this is part of the reason why we're continuing to invest pretty heavily in our Livermore manufacturing center and in other areas around the world to increase our capacity.

Brian Chin -- Stifel -- Analyst

Okay. Thanks. And maybe for my follow-up, maybe to Shai. Just kind of going back to the gross margins. Can you just roughly quantify what the drag in 3Q is from first, the DRAM mix issue piece of business? Is it something like 100 basis points still? Because I would imagine that if it could probably be nonrepeating when you get into 4Q. So maybe it's still there in 3Q, maybe not repeating in 4Q. And then also rhodium prices, who knows where it goes from here, but if it kind of stays at these more lower levels. Do you think you could get additional benefit in 4Q relative to what you're seeing in 3Q?

Shai Shahar -- Chief Financial Officer

On rhodium prices, I think it's very hard, of course, to model or predict the prices of precious metals. We took benefit in Q2 in purchasing inventory that will be used in Q3. And it's too early to say what's going to be the impact on Q4. But we have the materials that we need, and we're keeping an eye on it, hopefully, it will stay this lower levels than the peak. They are still elevated if you compare it to previous last year. But still not at the peak that we saw during the first quarter. In terms of mix, I don't want to go into all these details and drag you to the red here, but we see Systems going up and we have a better margin. Foundry and Logic expect to go down a little bit in Q3. DRAM, as Mike mentioned, expect to go down a little bit, but we see improved margins there, because of the things we learned and experienced during this quarter's manufacturing. And Flash is expected to go up a little bit as well. So all these moving parts, each one of them has kind of a small impact. But overall, that's where we end up at the same gross margin as in Q2 on similar revenue. On Q4, I think it's too early. As I said, volume volume prices and all the other factors are still a lot depend on mix, which we still too early for us.

Brian Chin -- Stifel -- Analyst

Okay. Got it. Thanks.

Operator

Next is Krish Sankar from Cowen & Company. Your line is open.

Stephen Chin -- Cowen & Company -- Analyst

This is Stephen Chin calling on behalf of Krish. Thanks for taking my question. First one, if I may, on your revised capex range for the full year. I think you mentioned it's currently lower to $70 million to $90 million for full year. Is that a reflection of ll efficiency? And how you're able to utilize the spending for the capacity of this year? Or is there any push out in spending first of all?

Shai Shahar -- Chief Financial Officer

So we've made significant progress in completing the manufacturing center year-to-date. And we expect to start shipping, as Mike mentioned, the shipping product in. So we have six months of actual behind us. And since we are coming in a little bit below budget. That's why we decided to update the estimate, but there is no significant change to our plans or to our target financial model gross margin of 47%. Some of the update is just simply relates to timing as things push a little bit to early 2022 or to the timing of cash payments for these fixed assets. But no significant changes to our plans or to the gross margin target.

Stephen Chin -- Cowen & Company -- Analyst

Okay. Understood. And one part of related question. Just given how DRAM probe card demand has been so strong, and it seems like it's going to gradually ebbing a little bit into back half, but still fairly healthy. A question on the DDR4 versus DDR5 technology. As mentioned earlier, every design had required a different probe cards as far as the DDR4 and five are any of those products that are given customers? Are those shared probe card designs? Or would the different technologies require different cards there?

Mike Slessor -- President and Chief Executive Officer

Yes. Well, DDR4 or DDR5, mobile server, any change in the chip design drives a change in the probe card. And so something as major as being going from DDR4 to DDR5 almost always changes the layout of the chip and how the customer is going to test it, which means new probe cards. That's why we try and emphasize so people understand some of the demand drivers of our business that this diversity of designs, not just in DRAM, but in Foundry and Logic as well being accelerated by advanced packaging, it really is an interesting growth driver for us. So when I talk about design activity, we've got a lot of first articles for across one alpha, one beta, 1Z nodes in DRAM, DDR4, DDR5 in both mobile and server. And each of those is unique and drives a refresh of the probe card fleet.

Stephen Chin -- Cowen & Company -- Analyst

Great. Thank you so much.

Operator

Your next question is from the line of Tom Diffely from D.A. Davidson. Your line is now open.

Tom Diffely -- D.A. Davidson -- Analyst

Yes, Good afternoon. First question on the Systems business that was strong in the quarter and growing in the next quarter. Is that strictly just the addition of some of the acquisitions? Or is there something in the core business that is growing faster than seasonal this year?

Mike Slessor -- President and Chief Executive Officer

Yes. I think when you look at the Systems results, you can think about it as a couple of different components as you broke them down. One is the legacy systems business we acquired as part of the Microtech acquisition in 2016. That business has been pretty steady. If you look at our target financial model and the underlying assumptions, low single-digit percent grower and has been executing to that. So the step up here to the mid- to high 30s in revenue, really associated with the acquisitions we've made. First, FRT to get us into metrology for advanced packaging back in 2019. That business beginning to accelerate nicely as especially the foundry ecosystem adopts our tools to do process control for their advanced packaging lines. And the HPD business showing some nice momentum in the early part of characterization, nowhere near production for quantum computing. And so probably the way to think about the systems growth is that the acquisitions are delivering growth on top of the stable legacy systems business.

Tom Diffely -- D.A. Davidson -- Analyst

Okay. That's helpful. And then as a follow-up, I'd like to go back to Craig's question. You talked about how yields were better on the DRAM business, and that helped you get back to a higher margin level. Are those improvements, yield improvements items that can progress on in the future? Or are those onetime items? A little more around that would be great.

Mike Slessor -- President and Chief Executive Officer

Yes. So remember, our DRAM revenue is pretty concentrated in a single design. It's not often we have a high runner that is this larger portion of the design -- And as we told you in the last earnings call, we had a pretty unfavorable cost structure on the initial shipments of those designs. So want to give kudos to our operations and engineering team who really went to work on improving this yield. Given that these are design-specific parts, this probe card and its layout and the yield DCs are really associated with this one design. It is kind of isolated to the one design. However, we have learned some things about designs of this level of complexity and some of the things here that could translate into other designs. So probably think about it as being pretty design specific. That one design in DRAM is still a large portion of our Q3 DRAM revenue. So helpful there. But I wouldn't sort of say it's a general uplift across all our DRAM designs.

Tom Diffely -- D.A. Davidson -- Analyst

Okay. And maybe it's a nice learning in case you have another volume orders.

Mike Slessor -- President and Chief Executive Officer

That's right. That's right.

Tom Diffely -- D.A. Davidson -- Analyst

Great. Thank you

Mike Slessor -- President and Chief Executive Officer

Thanks Tom.

Operator

Next is Charles Shi from Needham & Company. Your line is now open.

Charles Shi -- Needham and Company -- Analyst

Thank you for taking my question. Mike and Shai, congrats on the strong recovery of your gross margin. I want to ask a couple of questions around the Foundry logic probe card. I think you talked about design, you're talking about packaging, but no conversion, no upgrade is still a powerful demand driver. So I want to ask something around 3-nanometer really outside of Intel. I wonder what's -- I mean without getting into anything too customer-specific, what does your outlet for 3-nanometer probe card, I mean, were in July 2021, it's probably a year away from production. But are you seeing any leading indicator like design activities at the 3-nanometer? And is it the ramp by the end of this year? Or do you see more of the ramp at 3-nanometer on the mobile side or on HPD side? So that's a multi-facet question, but I hope you can give us some color about that.

Mike Slessor -- President and Chief Executive Officer

Sure, sure. Well, I think you're absolutely right, right? One of the demand drivers is node transitions for us, because a node transition automatically results in new designs, right? If you're going to convert the node, they become new chip designs that both fabless customers and IDMs take advantage of that new node capability to drive new products and new designs and that means new probe cards. On 3-nanometer, in particular, as you also indicated, it's pretty early. Some activity, but I think probably the exciting part of it is it's very likely to be one of the nodes where Advanced packaging is a really key part of how the node gets deployed. Probably high-performance compute tiles will be built on the 3-nanometer node. But then stacked together with other parts of the eventual product that maybe are built on 7-nanometer or even 10-nanometer. So as with leading customers, we're engaged in R&D, but it's pretty early, and I certainly wouldn't expect to see 3-nanometer be a significant part of our near-term results.

Charles Shi -- Needham and Company -- Analyst

Got it. Got it. So I mean, still around 3-nanometer, I understand you said it's still pretty early, but your leading foundry customer did say that they are seeing the 3-nanometer tape out being a lot higher than the number of tape-out at 5-nanometer when it was about two years ago. I wonder what does that mean to your business? Do you sort of expect that you can gain more of the revenue from the 3-nanometer maybe next year I mean, compared to what you got from 5-nanometer in 2020 or maybe 2019. Not sure whether that's a fair comparison or not, but any good color will be great.

Mike Slessor -- President and Chief Executive Officer

Yes. Well, and I think generally, if you look at some of these node progressions, test intensity is definitely going up. So there's lots of different dynamics at work. But these new nodes, especially when they intersect with the advanced packaging technologies are requiring higher test intensity, which means more testers and more probe cards per wafer. And I think you're seeing that from some of the manufacturers as well. So those ATE systems requires multiple probe cards over its life. So a pretty good indicator there. So I do think that the progression for five to three and at other customers from 10 to seven or whatever we're calling it now, our tailwinds for FormFactor's business because test intensity and test complexity are going up.

Charles Shi -- Needham and Company -- Analyst

Great. Thanks Mike. I am limited to two questions. Thank you very much. Hold and back to you.

Mike Slessor -- President and Chief Executive Officer

Thanks Charles.

Operator

We have our next question from the line of Christian Schwab from Craig-Hallum. Your line is now open.

Christian Schwab -- Craig-Hallum -- Analyst

Good quarter, guys. What should we be thinking about the range of quarterly revenue rate for the DRAM business now?

Mike Slessor -- President and Chief Executive Officer

Yes. I think, obviously, we've put together a few quarters here in the mid-30s to high 30s and up around so I'm getting a lot of feedback. Can you mute, Chris?

Christian Schwab -- Craig-Hallum -- Analyst

Yes. Thanks

Mike Slessor -- President and Chief Executive Officer

And obviously, this past second quarter came awfully close to a 10-year high that we posted in 2019. So I think there's an implicit question is DRAM probe card -- should our DRAM probe card revenues be going up. What we talked about previously is sort of an average, although it's a cyclical business, an average of kind of the mid-$30 million of revenue per quarter. If we continue to see this level of design activity, things, again, the advanced packaging theme at work in DRAM with things like HBM3. I think it's possible that we could see that average move up to the high 30s, the mid- to high 30s. But historically, if you look at DRAM customers, they've done a pretty job of constraining their overall spend, whether it be on probe cards or equipment and then driving efficiency to drive their bit growth. So I wouldn't expect a spectacular increase in the spend, but you may see the mean tick up a little bit into the high 30s.

Christian Schwab -- Craig-Hallum -- Analyst

Great. And then the recovery within Intel get this back over time back into the $200 million range plus and DRAM is kind of in the mid-30s. Is there an opportunity for TSM to become a much more material customer on a go-forward basis than they have been kind of over the last six quarters.

Mike Slessor -- President and Chief Executive Officer

Yes. I think we've talked about it as being something like a $100 million annual run rate opportunity. I think the past couple of quarters, you've seen them be a little bit short of that, but in that neighborhood. And I think go back to Charles' question, as they drive their node progressions as more wafer starts occur on the advanced nodes and in particular, the really advanced nodes like five and 3, our addressable market there grows. Reminding you that we're really only relevant for 10-nanometer and below. And although that's a big chunk of that foundry revenue, it's still a pretty moderate chunk of their overall wafer starts, unit wafer starts. And that's one of the things that drive probe card spend. So I think as their business grows as their node migrations more wafer starts to these advanced nodes. And we continue to work with them on some of the innovation and enabling advanced packaging, I think there is the opportunity for that to grow significantly

Christian Schwab -- Craig-Hallum -- Analyst

Great. Thank you

Operator

Next is Amanda Scarnati from Citi. Your line is now open.

Amanda Scarnati -- Citi -- Analyst

Hi. Good afternoon. Question on implications from the DDR5 pushout, right? Does that impact how you're looking at overall growth in that DRAM business rate? You mentioned that it could potentially grow to the mid- to high 30s on an average basis. But does that have any sort of near-term implications?

Mike Slessor -- President and Chief Executive Officer

It doesn't seem to have had much of an implication. I think, again, because of the the broad diverse designs that our customers are running. Even if certain parts of DDR5 are going to be a little bit slower than anticipated, there's a big chunk of DDR4 designs that are taking up the slack. And obviously, still some pretty aggressive bit growth forecast being driven by the combination of the DRAM manufacturers. So I don't think a significant impact associated with DDR5 timing for us.

Amanda Scarnati -- Citi -- Analyst

And then the other question is a clarification question on the capex coming down a little bit. It seems like it's more of a function of better expense management, better cost dynamics in that rather than a reduction in overall capex. Are we still at a point where you are capacity constrained versus what the demand environment is? Or are you in a better place now? And do you have enough room to expand capacity even further that TSMC business continues to grow?

Shai Shahar -- Chief Financial Officer

I think in some areas, we are still capacity constrained, right? It really depends on the product. We have multiple factories with different dynamics around the world. But in large, at these levels of operation, we are still capacity constrained and that's why we are putting this capacity expansion in place to answer the growing demand.

Amanda Scarnati -- Citi -- Analyst

And so the reduction in capex is not a function of reducing your overall capacity plans is just better cost?

Shai Shahar -- Chief Financial Officer

No, no, at all. It's only really about fine-tuning the budget. We put that in place at the end of last year, and now we have six months behind us. We have some actually in place know better what to expect, what's coming in. Some of it is just pushed out to the second -- to 2022, the beginning of 2022. Some of it is just when the cash payout is going to fall, is it going to fall in 2021 or 2022, but no changes in our line.

Operator

Great. Thank you.

David Silver -- CLK -- Analyst

Yes. Good morning or good afternoon, sorry. In the VLSI research reports that I scan, your company was mentioned a number of times in their customer satisfaction surveys. And I'd like to maybe just focus on the broadest one, I think the semiequipment group category. But you have a very high score overall. And in particular, your scores increased kind of across the board versus just one year ago, which I thought was interesting. But could you maybe talk about the strategies or the process improvements, Mike, that, that you would mainly credit for the broad-based improvement in your scores in that survey. And then maybe to take it one step further. I guess in your experience, how does that customer satisfaction survey result translate into either new customer acquisition or maybe increased business with your existing customer base? Thank you

Mike Slessor -- President and Chief Executive Officer

Yes. Yes, an interesting point. We are -- we did make a good showing in this year's VLSI customer satisfaction survey, which is a poll of all of our key customers and are proud of a higher ranking and higher raw scores. Some of the details are interesting. If I recall, we scored very high in technical leadership and customer support. And I think those are interesting validation points of our stated customer strategy. We're trying to be very close to the leaders in the industry as you see from our 10% customers, both in the second quarter and over the last couple of years. We want to be partnered with those customers driving their road map and our road map forward and deploying what's by far the biggest R&D spend in the space on solving their problems and then generating competitive advantage from it. I think rather than maybe being a driver of market share, I think it probably ends up being a reflection of market share. if those customers -- and all the customers that answer the survey are not happy with you, they're not going to order from you. And so I think the two go hand in hand. And I think both having compelling differentiated products, but also having strong customer satisfaction is central to our strategy and central to the history of market share gains we've executed to.

David Silver -- CLK -- Analyst

Okay. Great. And then maybe just one last question. This is kind of maybe perspective over a longer period of time. So when I look at the breakdown of your revenues by Logic and Foundry versus DRAM versus Flash, etc. I mean there's an element there where you can only sell probe cards for things that your customers desire them for. In other words, you track, I guess, the product development cycles. But I think on the other hand, over a longer period of time, there are different profit or competitive, sustainable competitive dynamics in the individual product lines. And I'm just wondering, over a longer period of time, do you have kind of a normalized, I don't know, share breakdown between, let's say, Logic and Foundry on the one hand and the memory categories on the other? Or is it literally the case where you want to align yourselves with your customers and just track the development cycles in the industry. Thank you.

Mike Slessor -- President and Chief Executive Officer

Yes. Well, if you back up over the very long time frame, we have some customers where we really didn't have a significant market share position. If I back up 10 years, two of the 10% customers in the second quarter were not really FormFactor customers. And so part of -- And this, again, consistent with the strategy I talked about. We're working very hard to make sure we have strong incumbent market share positions at all the leaders in the industry. And so regardless of those development cycles regardless of where large fabless customers, what foundry they decided to go to, we want to make sure we're in a position that's essentially neutral to that. And as long as we have strong market share with all the leaders in the industry and have good customer SAT scores with them, we think we'll do OK in the long run.

David Silver -- CLK -- Analyst

Okay. Thank you very much.

Mike Slessor -- President and Chief Executive Officer

Thanks.

Operator

No further questions at this time. I turn the call back over to Mr. Mike Slessor.

Mike Slessor -- President and Chief Executive Officer

All right. Thanks very much again for joining us today, and we hope to see you at some of the investor conferences in the second half and maybe even see you in person. Take care.

Operator

I'm sorry. We have our next question from Charles Shi from Needham & Company. Your line is now open.

Charles Shi -- Needham and Company -- Analyst

Really, my apologies. Thanks for squeezing me in. So Mike, really, just a question around DRAM. I think you might have alluded to this, but I really want to get a quick comment on this. So your DRAM strength going into second half of this year. Are you sort of seeing a broadening out in terms of the by customer? Is that because my understanding is your Q2, possibly a little bit at the first part of the Q3, your demand is I mean, primarily driven by one of your DRAM customers. That's my last question.

Mike Slessor -- President and Chief Executive Officer

Yes, that's correct. Although that is not unusual historically. We've seen our DRAM customers be -- each have different intensities in different quarters. And so it's not unusual for one of them to be very strong in a given quarter. and then another one be strong in a subsequent quarter. Again, go back to the customer strategy. That's one of the reasons why it's important to be a key supplier to all the leaders so that we're able to run our factories at high utilization and be able to make sure that we're on the leading edge of the industry, no matter which one of these DRAM manufacturers is winning market share with their big customers. So not unusual to see customer concentration in any given quarter.

Charles Shi -- Needham and Company -- Analyst

Thank you Mike. Once again my apologies for squeezing me in the question in the last minute.

Mike Slessor -- President and Chief Executive Officer

No problem. Alright. Operator?

Operator

[Operator Closing Remarks]

Duration: 55 minutes

Call participants:

Jason Cohen -- Vice President and General Counsel

Mike Slessor -- President and Chief Executive Officer

Shai Shahar -- Chief Financial Officer

Craig Ellis -- B. Riley Securities -- Analyst

Brian Chin -- Stifel -- Analyst

Stephen Chin -- Cowen & Company -- Analyst

Tom Diffely -- D.A. Davidson -- Analyst

Charles Shi -- Needham and Company -- Analyst

Christian Schwab -- Craig-Hallum -- Analyst

Amanda Scarnati -- Citi -- Analyst

David Silver -- CLK -- Analyst

More FORM analysis

All earnings call transcripts