FormFactor, Inc. (FORM) reported strong monetary outcomes for the second quarter of 2024, with vital year-over-year and sequential progress. The corporate’s income, non-GAAP gross margin, and non-GAAP EPS all exceeded the anticipated outlook vary. Progress within the probe card phase, significantly within the DRAM market, drove the corporate’s success, with document income ranges achieved.
FormFactor’s diversified product portfolio and strategic positioning in superior packaging applied sciences like high-bandwidth reminiscence and silicon photonics have poised the corporate for continued optimistic efficiency. The outlook for the third quarter stays optimistic, with steady demand and anticipated income of $200 million.
Key Takeaways
- FormFactor’s Q2 income elevated by 17% sequentially and 26.7% year-over-year.
- Non-GAAP gross margin reached 45.3%, exceeding the goal monetary mannequin.
- The probe card enterprise, particularly DRAM, contributed considerably to the expansion.
- Q3 income is predicted to be round $200 million, with non-GAAP gross margin round 43%.
- The corporate is well-positioned for the adoption of superior packaging applied sciences.
Firm Outlook
- Q3 income forecast at $200 million, with a non-GAAP gross margin of 43% plus or minus 150 foundation factors.
- Non-GAAP earnings per absolutely diluted share for Q3 are anticipated to be $0.31 plus or minus $0.04.
- Expectations of great adoption and quantity manufacturing of silicon photonics by late 2025 or early 2026.
Bearish Highlights
- Uncertainty within the DRAM market’s shift from HBM to DDR5 persists into This fall, with restricted visibility.
- Geopolitical headwinds might constrain vital market contributions from China.
Bullish Highlights
- Report revenues in DRAM probe-card market, with DDR4 and DDR5 markets strengthening.
- Constructive buyer suggestions and spot pricing point out an upturn within the non-HBM DRAM market.
- Profitable positioning within the HBM market, with progress alternatives as HBM3 and HBM4 are launched.
Misses
- No vital misses had been reported within the earnings name.
Q&A Highlights
- CEO Michael Slessor mentioned the potential progress within the DRAM probe-card market and the corporate’s readiness for shifts in foundry logic probe-card calls for.
- The corporate is specializing in qualifying mems-based probe playing cards for GPU checks and acknowledges the rising significance of HBM in AI merchandise.
FormFactor’s earnings name revealed a powerful efficiency in Q2 2024, with the corporate benefiting from its numerous product choices and strategic concentrate on rising applied sciences. The corporate’s monetary well being is underscored by elevated revenues, significantly within the probe card phase, and strong gross margins.
Whereas the way forward for the DRAM market stays considerably unsure, FormFactor’s administration is optimistic concerning the firm’s place and the potential for progress in varied high-tech markets. The corporate’s progress and strategic initiatives, such because the qualification of mems-based probe playing cards and the anticipated adoption of silicon photonics, point out a forward-looking method to addressing the wants of the evolving compute trade. The upcoming conferences in August and September are anticipated to offer additional perception into the corporate’s route and trade developments.
InvestingPro Insights
FormFactor, Inc. (FORM) continues to exhibit robust monetary well being and strategic market positioning, as mirrored of their current quarterly outcomes. To additional perceive the corporate’s current efficiency and future potential, let’s delve into some key metrics and insights from InvestingPro:
InvestingPro Information:
- Market Cap: $4.13 billion, indicating a considerable presence within the trade.
- P/E Ratio: At present at 34.97, suggesting that traders are keen to pay a better worth for earnings, probably resulting from expectations of future progress.
- Income Progress (Quarterly): A rise of 0.76% in Q1 2024, displaying constant progress in gross sales.
InvestingPro Suggestions:
- Analyst Confidence: Six analysts have revised their earnings estimates upwards for the upcoming interval, signaling confidence within the firm’s monetary trajectory.
- Valuation Metrics: Regardless of a excessive earnings a number of, the corporate is buying and selling at a low P/E ratio relative to near-term earnings progress, which might point out a horny funding alternative for these wanting on the firm’s progress prospects.
These insights, mixed with the corporate’s strong Q2 efficiency and optimistic outlook for Q3, underscore FormFactor’s potential for sustained progress. For these fascinated with a deeper evaluation, InvestingPro affords extra recommendations on FormFactor, which may be discovered at https://www.investing.com/professional/FORM.
Full transcript – FormFactor Inc (FORM) Q2 2024:
Operator: Thanks, and welcome, everybody, to FormFactor’s second-quarter 2024 earnings convention name. On right now’s name are Chief Government Officer Mike Slessor, and Chief Monetary Officer Shai Shahar. Earlier than we start Stan Finkelstein, the corporate’s VP of Investor Relations, will remind you of some vital info.
Stan Finkelstein: Thanks. Immediately the corporate will probably be discussing GAAP P&L outcomes and a few vital non-GAAP outcomes meant to complement your understanding of the corporate’s financials. Reconciliations of GAAP to non-GAAP measures and different monetary info can be found within the press launch issued right now by the corporate and on the Investor Relations part of our web site. Immediately’s dialogue incorporates forward-looking statements throughout the which means of the federal securities legal guidelines. Examples of such forward-looking statements embrace these with respect to the projections of monetary and enterprise efficiency, future macroeconomic and geopolitical circumstances, the advantages of acquisitions and investments in capability and new applied sciences. Th influence of world, regional and nationwide well being disaster, together with the COVID-19 pandemic, anticipated trade developments, potential disruption in our provide chain, the impacts of regulatory modifications, together with the current US China commerce restrictions, the anticipated demand for product, our potential to develop produce and promote merchandise and the assumptions upon which such statements are primarily based. These statements are topic to identified and unknown dangers and uncertainties that would trigger precise outcomes to vary materially from these expressed throughout this name. Data on threat elements and uncertainties is contained in our most up-to-date submitting on Kind 10-Ok with the SEC for the fiscal 12 months ended December 30, 2023 and in our different SEC filings, which can be found on the SEC’s web site at www.sec.gov and in our press launch issued right now, forward-looking statements are made as of right now, July 31, 2024, and we assume no obligation to replace them with out. We are going to now flip the decision over to FormFactor’s CEO Mike Slessor.
Michael Slessor: Thanks, everybody, for becoming a member of us right now. FormFactor’s second-quarter income, non-GAAP gross margin and non-GAAP EPS all exceeded the midpoint of the outlook vary we offered in Could. Anticipated power in our probe card enterprise drove second quarter outcomes as we recorded sequential will increase in all three probe card markets. DRAM probe-card income reached document ranges with income from high-bandwidth reminiscence or HBM doubling for the third consecutive quarter to just about 75% of our DRAM income. Within the present third quarter. We’re experiencing regular general demand for FormFactor’s merchandise as we proceed to make progress in the direction of our goal monetary mannequin. Our second quarter outcomes and third quarter outlook illustrate the distinctive worth of FormFactor’s diversification technique, which differentiates us from our direct rivals with a broad lab to fab product portfolio throughout foundry and logic, DRAM and Flash probe-cards, along with our Techniques phase merchandise. This portfolio permits us to compete for enterprise throughout numerous demand swimming pools in any respect main prospects producing comparatively steady outcomes throughout final 12 months’s downturn and the highest line progress we’re delivering this 12 months. Our income progress is being pushed by publicity to increasing areas like high-bandwidth reminiscence and DRAM probe-cards and co-packaged silicon photonics and techniques enabling FormFactor to develop at the same time as we await refresh cycles in vital excessive unit quantity finish markets that drive Foundry and Logic probe-card spending like cellular handsets, shopper PCs. FormFactor’s third-quarter outlook additionally demonstrates a number of distinctive options of probe-card demand. As we regularly be aware, probe playing cards are a consumable that is particular to every new chip design, which means we profit from each expertise node transitions and from the discharge of recent designs on current nodes, along with our broad market and buyer publicity. This produces a extra steady demand profile than the extra risky demand cycles that characterize capital gear. This distinction is especially evident within the present atmosphere as FormFactor’s steady third quarter outlook contrasts sharply with the sequentially weaker third quarter outlooks provided by a number of corporations within the check and meeting capital gear. Turning now to market and phase stage particulars as I famous we set a document for DRAM probe-card income within the second quarter, pushed by the sequential doubling of high-bandwidth reminiscence income layered on prime of regular DDR5, new design exercise to place HBM towards progress into perspective. Within the first half of 2024 alone, we doubled the HBM income we delivered in all of 2023. As well as, our second quarter HBM income is bigger than FormFactor’s whole quarterly DRAM income in every quarter of 2023. This exercise is a direct results of the big hyperscaler investments in generative-AI infrastructure and our prospects’ corresponding ramps of HBM capability and output. HBM, which is a stack of 8, 12 and even 16 particular person DRAM die assembled with superior packaging processes like by Silicon Through’s and Thermo-Compression Bonding continues to supply a robust instance. Superior packaging is driving our present outcomes. Additionally foreshadows our long run alternative as we profit from the elevated check depth and elevated check complexity that is inherent in superior packaging structure. Within the third quarter, demand for DRAM probe-cards continues to be strong, and we anticipate a slight enhance in income from the second quarter document. We do, nevertheless, anticipate a shift in our DRAM product combine with sequential enhance in DDR5 shipments and a discount in HBM shipments. we talked about previously that our prospects typically have one to 2 quarter intervals of digestion following a powerful quarter of shipments with lead instances of lower than 1 / 4 for brief time period visibility stays difficult as all the time, however given our prospects’ continued robust funding in HBM capability, we anticipate the third quarter pause in HBM progress will signify a brief digestion section and the expansion will resume in a single to 2 quarters. Shifting to the Foundry and Logic probe-card market, we delivered the anticipated second quarter sequential progress pushed by the seasonal ramp of recent cellular utility processor designs and stronger probe card demand for shopper PC and server microprocessor designs. Within the present third quarter, we anticipate foundry and logic demand and product combine to be just like that achieved within the second quarter. Superior packaging processes like foveros and are being more and more used to architect foundry and logic chip designs. As with HBM and DRAM, this disaggregation of a chip into subcomponent chiplets or tiles will increase each check depth and check complexity in comparison with an equal mannequin of the chip. The rise in check depth is pushed by the necessity for our prospects to show and check every part chiplet previous to stacking after which to show and check the multi-chip flip stack at varied factors throughout the meeting course of. The ensuing enhance in probe card use per good die out is identical dynamic that’s driving the robust probe card spending by our prospects for HBM, DRAM and because the supply of comparatively robust midyear Foundry and Logic demand, regardless of the dearth of restoration in vital finish markets like cellular handsets, and PCs. On the similar time, the technical necessities for probe-cards for Foundry and Logic designs constructed utilizing superior packaging processes are considerably extra demanding than for normal unstack merchandise involving greater check speeds and more difficult thermal and energy specs as we have seen with HBM and DRAM, our differentiated potential to fulfill these efficiency necessities to drive each market share and profitability positive factors because the adoption of superior packaging in foundry and logic continues. Within the Techniques phase, we anticipate a slight sequential enhance in Q3 income as prospects proceed to interact us to resolve probably the most advanced electro optical check and measurement challenges in areas like quantum computing and co-packaged silicon photonics. In co-packaged optics, which is poised to revolutionize chip-to-chip communication within the knowledge middle by considerably lowering energy consumption at excessive knowledge charges. We’re collaborating with key prospects within the early levels of the lab to fab transition from R&D to low-volume manufacturing. In these collaborations, prospects are deploying FormFactor’s turnkey electro-optical measurement techniques constructed on our CM300 and SUMMIT200 engineering probers, along with our proprietary optical probes. The extremely differentiated optical probes that allow both floor or edge coupling to the photonic dye with unmatched coupling effectivity, offering our prospects with greater yields and shorter check instances. As silicon photonics matures and strikes to high-volume manufacturing within the coming years. We anticipate that our management positions in mixed electrical and optical check will present a brand new progress vector for each our techniques and probe-card companies. In closing, we’re enthusiastic about each our robust second-quarter outcomes and our strong third quarter outlook because the accelerating adoption of superior packaging drives elevated demand throughout FormFactor’s lab to fab product portfolio. Long term, we’re assured within the progress prospects for FormFactor within the trade general, pushed by the elemental developments of semiconductor content material progress in superior packaging improvements like HBM, chiplets and co-packaged silicon photonics. It’s evident from our current outcomes and outlook. These are developments had been FormFactor’s nicely positioned, and we’re assured that our investments in R&D capability and expertise will additional improve FormFactor’s market management. This can allow us to attain after which surpass our goal mannequin that delivers $2 of non-GAAP earnings per share on $850 million of income. Shai, over to you.
Shai Shahar: Thanks, Mike, and good afternoon. As you noticed in our press launch, Q2 revenues had been $197.5 million, $2.5 million above the midpoint of our outlook vary. And non-GAAP gross margin of 45.3% was 0.3-percentage-points above the midpoint of the vary. These, along with OpEx barely decrease than the midpoint of the outlook, resulted in non-GAAP EPS on the prime finish of the vary. Second quarter revenues elevated 17% sequentially from the primary quarter and elevated 26.7% year-over-year from Q2 ’23 revenues. The upside versus the midpoint of the outlook vary was resulting from greater revenues in our probe-card phase. Probe-card phase revenues had been $167 million within the second quarter, a rise of $30 million or 22% from the primary quarter. The rise was pushed primarily by greater Foundry and Logic and DRAM income Techniques phase revenues had been $30.7 million in Q2, the $1.3 million lower from the primary quarter and comprised 15.5% of whole firm revenues, down from 19% within the first quarter. Inside the probe-card phase, Q2 foundry and logic revenues had been $104 million, a 19.5% enhance from the primary quarter. Foundry and Logic revenues elevated to 52.5% of whole firm revenues in comparison with 51.4% within the first quarter. DRAM revenues had been a document $58 million in Q2, $12.1 million or 26.5% greater than within the first quarter and rose to 29.4% of whole quarterly revenues as in comparison with 27.2% within the first quarter. Inside HBM income virtually doubled from $22 million in Q1 to over $43 million within the second quarter. Flash revenues of $5.1 million in Q2 had been $1.1 million greater than within the first quarter, and we’re at 2.6% of whole revenues in Q2 as in comparison with 2.4% in Q1. GAAP gross margin for the second quarter was 44% as in comparison with 37.2% in Q1. Value of revenues included $2.5 million of GAAP to non-GAAP reconciling objects, which we outlined in our press launch issued right now and within the reconciliation desk obtainable within the Investor Relations part of our web site. On a non-GAAP foundation, gross margin for the second quarter was 45.3%, 6.6-percentage-points greater than the 38.7% non-GAAP gross margin in Q1 and 0.3-percentage-points above the midpoint of our outlook vary. The will increase in comparison with Q1 was principally the results of greater gross margin within the probe-card phase. Whereas we anticipate fluctuations quarter over quarter, primarily resulting from product combine modifications. Attaining these 45.2% gross margin within the second quarter validates our progress in the direction of our 47% goal monetary mannequin non-GAAP gross margin at annual income of $850 million. Our probe-card phase gross margin was 45.1% within the second quarter, a rise of 8-percentage-points in comparison with 37.2% in Q1. Our Q2 system phase gross margin was 46.2%, a rise of 0.9-percentage-points in comparison with 45.2% gross margin within the first quarter. As in comparison with Q1, roughly two-thirds of the rise in non-GAAP gross margin is attributable to the upper quantity and the remaining third pertains to a extra favorable merchandise. Our GAAP working bills had been $69.4 million for the second-quarter as in comparison with $61.7 million within the first-quarter. Non-GAAP working bills for the second quarter was $60.9 million or 30.8% of revenues, as in contrast with $52.3 million or 31% of revenues in Q1. The $8.5 million enhance relates primarily to spotlight two greater performance-based compensation. Firm non-cash bills for the second quarter included $10.2 million for stock-based compensation, $0.6 million for amortization of acquisition associated intangibles and depreciation of $7.4 million, all just like the primary quarter. GAAP working revenue was $17.8 million for Q2 in contrast with GAAP working revenue of $21.3 million in Q1. Non-GAAP working revenue for the second quarter greater than doubled to $28.5 million in contrast with $13 million within the first quarter, a rise of $15.6 million or 120%, demonstrating the leverage in our working mannequin. GAAP web revenue for the second quarter was $19.4 million, or $0.25 per absolutely diluted share in contrast with a GAAP web revenue of $21.8 million or $0.28 per absolutely diluted share within the earlier quarter. GAAP web revenue in Q1 included a achieve of $20 million from the divesture of our Chinese language subsidiaries. The non-GAAP efficient tax fee for the second quarter was 15.4%, 1.7-percentage-points greater than the 13.7% within the first quarter, and we proceed to anticipate our annual non-GAAP efficient tax fee to be between 14% and 18%. Second quarter non-GAAP web revenue virtually doubled to $27.3 million or $0.35 per fully-diluted share in comparison with $14.3 million or $0.18 per absolutely diluted share in Q1. Q2 EPS was $0.17 greater resulting from considerably greater revenues and better gross margins with comparatively flat OpEx as a share of income. Shifting to the steadiness sheet and money circulate, we generated free money circulate of $14.2 million within the second quarter in comparison with $19.7 million in Q1. The principle causes for the lower in free money flows was elevated working capital attributable to the income progress, partially offset by decrease capital expenditure spending of $5 million. We invested $8.4 million in capital expenditures throughout the second quarter in comparison with $13.4 million in Q1. There is no change in our beforehand communicated anticipated CapEx vary for 2024 of $35 million to $45 million. At quarter finish, whole money and investments had been $366 million, a rise of $8 million from Q1. On the finish of the second quarter, we had one time period mortgage remaining with the steadiness totaling $14 million. Concerning inventory buyback, throughout the second quarter, we used $2.9 million to purchase again shares underneath the $75 million two-year buyback program that was permitted in This fall 2023. At quarter finish, $53.5 million stay obtainable underneath that authorization. As a reminder, the principle objective of the share repurchase program is to offset dilution from stock-based compensation. Turning to the third-quarter non-GAAP outlook, we anticipate Q3 income of $200 million plus or minus $5 million with a slight enhance over Q2 coming from DRAM techniques, inside DRAM and we additionally anticipate a mixture shift with a better share of DDR5 versus HBM income. Third quarter non-GAAP gross margin is predicted to be 43% plus or minus 150 foundation factors, anticipate a lower in non-GAAP gross margins within the third quarter on the midpoint of the outlook vary on barely greater revenues pertains to a much less favorable product combine with DRAM revenues as a share of whole income is predicted to extend and HBM income throughout the route is predicted to lower On the midpoint of those outlook ranges. We anticipate Q3 working bills to be $61 million plus or minus $2 million just like Q2. Non-GAAP earnings per absolutely diluted share for Q3 is predicted to be $0.31 plus or minus $0.04 and a reconciliation of our GAAP to non-GAAP Q3 outlook is accessible on the Investor Relations part of our web site and in our press launch issued right now. With that, let’s open the decision for questions. Operator.
Operator: [Operator Instructions]. Our first query comes from the road of Craig Ellis of B. Riley Securities.
Craig Ellis: Thanks for taking the query and congratulations on the execution, guys. And Mike, I needed to begin simply by following up on a degree you made with regard to high-bandwidth reminiscence, I feel you indicated the corporate anticipated that after some digestion and the third quarter progress might reaccelerate there. I hoped you would simply share a few of that both buyer interplay or issues that you simply’re seeing that lend confidence to the reacceleration? And any shade on timing can be useful. Thanks.
Michael Slessor: Yeah. So Craig, good query. Simply to stage set, everyone. We once more, within the second quarter, doubled HBM income to 75% of our DRAM income. So actually vital contribution from that, as I feel most of you understand, is fairly concentrated with a single buyer, though every of the three main DRAM producers, you may have contributions in there and it is common after we ship at that scale for a buyer to have a interval of digestion the place they then use the probe-cards for 1 / 4. You’ll be able to see this traditionally by our main prospects as they’re of their cadence quarter to quarter that we report for a ten% buyer. And so what provides us confidence that that is that digestion, I feel a few issues. If I take a look at it from a macro stage, continued hyperscaler funding in generative AI, which in flip is driving our prospects, our DRAM prospects to extend their capability and output of HBM3 and three at the moment and starting to do improvement on HBM4 and HBM5 manufacturing. And actually in conversations with these prospects. There is no there isn’t any pause in that. They are surely persevering with to speculate very closely on this fairly profitable a part of the DRAM market, once more, pushed by hyperscaler investments. In our direct visibility, as most of you understand, is fairly restricted. We function with lead instances of lower than 1 / 4. And so when it comes to POs and backlog, we will not see a lot previous the top of the third quarter right here. However after we take a look at the conversations we’re having with these prospects, after we take a look at the investments, the hyperscalers are making it does appear to be that is fairly cheap to attribute this Q3 are nonetheless HBM nonetheless at fairly wholesome ranges. It is a digestion interval.
Craig Ellis: That is actually useful, Mike, thanks. After which the follow-up query is considerably comparable three months in the past. You famous {that a} foundry logic buyer, previously the most important buyer for this final quarter the place it is now DRAM and would at instances previously are undergo a interval of digestion after a extremely robust quarter? Are you seeing indicators? And is that baked within the steering that we’ll see that digestion within the third quarter? Or how are you desirous about the chance that that would occur at this level?
Michael Slessor: Yeah, foundry and logic general, we see within the third quarter fairly corresponding to the second quarter, each from general ranges and buyer and product combine. And I feel that is actually one other proof factors related to how superior packaging is driving our enterprise. You see main prospects, actually in compute, but in addition to some extent in cellular starting to undertake superior packaging tech extra starting to undertake superior packaging. And we see that with throughout the compute sector applied sciences like Foveros and co-los, even because the PC market stays, PC finish market stays considerably muted. We’re seeing fairly strong demand for probe playing cards due to the rise in check depth and complexity pushed by superior packaging within the sector. So Q3, fairly just like second quarter in foundry and logic from an general buyer and product combine perspective, once more, pushed by superior packaging.
Craig Ellis: That is actually useful. After which simply the commentary on that, Mike, can be sense. These are heterogeneous die primarily based merchandise are such a small a part of combine now however over time will turn out to be a majority of combine that might appear to be an actual good long-term secular tailwind for the enterprise?
Michael Slessor: Yeah, I feel throughout the trade, it is a actually attention-grabbing and compelling secular alternative for these of us uncovered to superior packaging.
Craig Ellis: Thanks, Mike. Good luck in 3Q.
Michael Slessor: Thanks.
Operator: Our subsequent query comes from the road of Charles Shi of Needham & Firm
Charles Shi: Hello, Mike, so good afternoon and thanks for letting me ask two questions. The primary one, it is about DDR5 versus HBM. I feel previous to this quarter, the best way you characterize the DRAM probe-card market was in case you again out HBN., the usual DRAM or DDR probe-card had been nonetheless run ranking round that. I might say $20 million-ish, that downturn stage. Principally, it appears to be like like subsequent quarter, the September quarter, you are anticipating a considerably comparable, perhaps a DRAM whole income, however up because you’re speaking about combine shift from away from HBM extra to DDR5 Would you characterize perhaps that is the start of the usual 12 months finish restoration or that is perhaps only a one-off quarter? I simply needed to know the sustainability of the DDR5 probe-card progress from right here?
Michael Slessor: Yeah. It is an attention-grabbing query is, as you may think, we’re having comparable discussions internally and with our prospects. I feel in case you take a look at the general non-HBM DRAM market, we’re characterizing it as DDR5, however that is a fairly broad brush that low-power server and compute, and also you’re seeing a major step-up right here within the third quarter to ranges which might be corresponding to the earlier cyclical highs. Now after we see and the place HBM is we mentioned there is a little bit of a one quarter digestion interval, and that is common. Keep in mind, probe-cards are particular to every buyer chip design. And in order prospects transfer their wafer begins in forwards and backwards between applied sciences and totally different merchandise for them to optimize their output for his or her market. We are able to see these fairly vital swings between designs and market segments and submarkets and whether or not that is the beginning of a DRAM upturn, I feel remains to be to be seen, however it’s encouraging in case you’ve seen DRAM spot pricing, in case you’ve heard because the narrative from our prospects on the general DRAM market bettering actually would be the starting of an upturn for now. We simply do not have the visibility to say so whether or not it’s or not, however good to be once more working our DRAM enterprise, a vital enterprise for us up at document ranges.
Charles Shi: Thanks, Mike. I’ve a follow-up on the foundry logic facet of the enterprise, it is up in Q2 was up foundry logic probe card was up in Q2 by quite a bit about, however your largest microprocessor prospects, the income you do disclose on a quarterly foundation typically. So up imply as a lot I imply, it would not clarify all of the incremental you are seeing within the foundry logic facet and so this type of a mid-year stress cash. Should you make clear a bit of bit of what is driving that outdoors of this microprocessor firm and extra importantly, going into subsequent quarter, who’s driving that incremental progress in foundry logic probe-card?
Michael Slessor: Yeah, we regularly see this midyear power in cellular utility processors. Should you take a look at cellphone launch, cadences and work again to the timing the place prospects are going to want probe playing cards for these. That is precisely what we’re seeing proper now. And we talked about it on the final name with the anticipated second quarter power, which materialized being related each with microprocessors, but in addition at midyear, our releases of those cellular utility processors. That is a theme that we see carrying by the second quarter. Once more, we see foundry and logic having an identical stage and profile within the third-quarter as within the second-quarter and so past that, once more, visibility fairly restricted, however we’re fairly pleased with how we’re executing within the broader foundry and logic market. All the time some enhancements, all the time some share to achieve, however that is each of these elements driving our Foundry Logic power within the center a part of the 12 months.
Charles Shi: And perhaps a fast follow-up to that. Do you see a bit of little bit of a mixture shift in foundry logic probe-cards from Q2 to Q3? Thanks.
Michael Slessor: No, it’s totally comparable Charles.
Charles Shi: Thanks.
Operator: Our subsequent query comes from the road of Tom Diffely of D.A. Davidson.
Tom Diffely: Sure, good afternoon, thanks, Mike. I used to be curious once you take a look at the high-bandwidth reminiscence ramp that you have had this 12 months, how a lot of that was pushed by simply rising volumes of high-bandwidth reminiscence versus a design change? After which if there’s a large shift to the following era design, what does that do to what you are promoting?
Michael Slessor: Yeah, as we have mentioned, probe playing cards are a design particular consumable and so there’s these two items. There’s the discharge of a brand new design, however there’s additionally the variety of wafer begins at variety of check cells and subsequently, probe playing cards that have to run on this design. I might characterize the expansion within the second quarter of the HBM enterprise being concentrated throughout, let’s name it in spherical numbers, 10 designs which might be actually the excessive runners driving the expansion. And in order that may very well be one ship keep in mind, HBM and a stack of principally eight excessive DRAM, however there is a base die. There’s check insertions for the eight core die after which there’s additionally a check insertion when the shoppers carried out stacking. Every of these three units of probe-cards is a singular fleet of probe playing cards and in order that’s a part of the check depth enhance related to HBM specifically and superior packaging basically. However it greater than a handful of designs actually driving the power.
Tom Diffely: After which once you take a look at the enterprise subsequent 12 months in ’25, do you assume it is fairly nicely break up between the three fundamental suppliers? Are you continue to going to be very leveraged to one of many gamers.
Michael Slessor: No, I feel you understand, as they mentioned, we at the moment see contributions from all three producers, though we’re over-indexed to the chief in HBM market share and a elementary tenet of our technique is to be sure that we’re a key provider at the vanguard to all main buyer. And so qualification work is ongoing and each of them had been ready positioned to compete for that enterprise as HBM3 and HBM4 out. These are alternatives for our buyer share improvement and for us to go compete for brand spanking new enterprise as nicely.
Tom Diffely: After which ultimate query, Shai. While you take a look at the margin steering for the following quarter, is it merely the lower of high-bandwidth reminiscence as a share of product combine challenge?
Shai Shahar: Whereas, it is a mixture of DRAM income general as a better share income and inside DRAM, the shift from HBM to DDR5, so that is each of them.
Tom Diffely: Okay, nice. Thanks.
Shai Shahar: Thanks, Tom.
Operator: Our subsequent query comes from the road of Christian Schwab of Craig-Hallum.
Christian Schwab: Nice. Thanks for taking my questions. So simply once more on high-bandwidth reminiscence, we are able to sort of again into the mathematics, fairly modest income from the opposite two large guys as we get into 2025, do you assume you haven’t any three significant prospects in that house, or do you assume that there is the potential for as the opposite two guys rise up to hurry they usually start to take share. I am simply making an attempt to dimension what your expectations and ’25 and ’26 are for high-bandwidth reminiscence as the opposite two guys come to market that I’ve talked about being bought out for subsequent 12 months.
Michael Slessor: Yeah, we’ll keep in mind if we again out a bit of bit, we’re a key provider to all three main DRAM producers in DDR4, DDR5 and we hope to be for HBM as nicely. At present, one in every of them owns the lion’s share of HBM shipments. And in order that’s why we’re over-indexed to that buyer. However once more, as Tom requested, we do anticipate and as obscure as the opposite two DRAM producers achieve share within the transition from HBM3 to HBM4 work their method by some fairly vital technical and yield challenges in ramping this up. We do anticipate that share to steadiness out a bit of bit. So, you understand, as we glance into ’25, the transition to HBM4 actually thrilling for us. It raises the velocity necessities for the probe-card. So once more, a notch up in complexity and as prospects go to stacking 12 and even 16 excessive die. That clearly goes to extend the general check depth and complexity and testing that general stack. So we really feel like we’re in a fairly robust aggressive place even because the share begins to steadiness amongst all three DRAM producers.
Christian Schwab: Within the follow-up to that, on the high-bandwidth reminiscence for transition, you understand, given as you mentioned, the velocity and enhance stacking capabilities, I might assume that there ought to be some pricing energy on these probe-cards, not all the best way, clearly to foundry and logic ranges, however probably greater than the place we sit right now. Is that truthful?
Michael Slessor: Yeah, that is truthful. And one of many causes proper, we touched on it in a query a few minutes in the past, however the combine shift away from HBM in the direction of DDR5 inside DRAM at these excessive ranges, actually as a gross margin hit. And so you possibly can join the dots there, our potential to construct probe playing cards which might be extremely differentiated and supply loads of worth to our prospects in testing it velocity and throughout a number of temperatures in a really advanced mechanical state of affairs. You are testing a complete 300 millimeters wafer of the stack die throughout a temperature vary of over 100 levels Celsius. And you’ll think about the thermal scaling challenges related to which might be vital. And that is why HBM probe-cards value extra.
Christian Schwab: Nice. After which my second query is now we have two quarters of name it at roughly $200 million, which is sort of the goal mannequin. Do you guys have a set expectations once you plan on updating your goal mannequin?
Shai Shahar: Positive. I am going to take this one, Christian. So we’re two quarters at $200 million, however the goal mannequin is a bit of greater than that, proper. Should you take the $850 million, it is about . So what we need to do is what we have now carried out previously as soon as we have now 1 / 4 or two at this mannequin run fee. And I agree we’re getting nearer to it, however we’re not there but. Then we’ll arrange a date schedule, a brand new analyst day after which work on publishing the brand new goal mannequin. So we’re not there but however keep tuned.
Christian Schwab: Obtained it, sounds nice. Thanks guys.
Shai Shahar: Thanks, Christian.
Operator: [Operator Instructions]. Our subsequent query comes from the road of Brian Chin of Stifel.
Brian Chin: Thanks for letting us ask just a few questions on Mike, are you able to remark whether or not you anticipate form of that the combination shift you are seeing within the DRAM enterprise to persist into This fall? And I suppose that is sort of the primary query, as a result of I suppose perhaps it is like I heard 50 foundation factors to 80 foundation factors headwind or one thing relative to what it was in 2Q. And so and that is, I suppose the primary query.
Michael Slessor: Yeah, Brian, it is Mike. I am going to take that. I feel when it comes to timing, I am going to once more caveat this with our visibility, our direct visibility, it would not actually go a lot previous the top of 3Q right here, we’re working with brief lead instances. We actually have forecasts from our prospects. However the dramatic shift in DRAM in the direction of manufacturing capability for DDR5 is one thing that is a comparatively dynamic state of affairs, whether or not it persists or not is one thing that in all probability we’d like just a few extra weeks or months to essentially get a learn on from our direct backlog, having mentioned that, you understand, in case you do take a look at the and you understand that the underlying DRAM market DDR4 and DDR5, low-power server and PC, they do appear to be strengthening and so it would not be all that shocking if there was a shift of some capability and in the direction of these markets and we noticed a bit progress within the non HBM elements of the market a bit of bit early to inform. And once more, our visibility would not go that far into the longer term, however among the elementary elements are in place for a DRAM restoration agenda.
Brian Chin: Yeah, I used to be even pondering, sure, Samsung (KS:) appear fairly assured final night time that their HBM3 shipments, if they’re the dominant participant, however they might speed up considerably within the again half of the 12 months. So perhaps that 3Q, however I think about that would think about your outlook for HBM in some unspecified time in the future within the second half?
Michael Slessor: That is proper.
Brian Chin: Okay, and I do not know if anybody requested, however when it comes to the I missed this half, however the China elevated Q-on-Q, was that primarily simply multinational of any sort of extra shade on, what drove that?
Michael Slessor: You understand that there was introduced matter to touched, reminder. So in Q2, a bit of over 10% China as a share of income and contained in the Q3 look, we’re forecasting a average enhance. It is going to nonetheless be within the 10s, nowhere close to the 40% to 50%. A number of the WFE suppliers have for China. You keep in mind, we have been fairly proactive in managing the state of affairs. We divested our China operations. After which I fashioned an unique distributor ship with the with the purchaser of that enterprise. And we’re opportunistically serving the home China enterprise as finest we are able to within the face of some fairly vital geopolitical headwinds each from direct export controls and the China trade semiconductor response to face up its personal home semiconductor trade. So we see a average enhance. I do not see it as a theme the place we’ll get vital contribution from the China enterprise, once more due to these virtually structural geopolitical headwinds.
Brian Chin: And let me simply sort of get off matter right here for one final query. In your power effectivity utilization bandwidth. These are clearly key themes in knowledge middle. I perceive that silicon photonics is an space the corporate has invested and the place are we on the adoption curve for co-packaged optics and the way vital of a market alternative might that be in perhaps a few years?
Michael Slessor: Yeah, we’re very early, as I mentioned, simply inching out of the lab the place we have been engaged with prospects for years and co-developing the elemental measurement applied sciences like our Pharos optical probe, that permits prospects quicker check instances and higher general yield due to its higher signal-to-noise and coupling effectivity. However we have a handful of instruments inside manufacturing web site which might be within the very early levels of pilot manufacturing as you understand. So this is likely one of the methods to assist resolve. It isn’t been totally solved, but it surely helps resolve a fairly vital downside related to knowledge facilities. And that is power consumption going to silicon photonics on chip-to-chip communication and the info middle has the potential to influence the power utilization power budgets for these knowledge facilities. And that is one thing the broader compute trade goes to should confront as knowledge facilities turn out to be greater and larger customers of electrical energy worldwide. So we view silicon photonics as one thing that is a I can’t and sure, I feel the timing in all probability on the earliest, late ’25 into ’26, the place we begin to see shifting into the 12 months, the curve of adoption and working vital volumes contained in the foundries worldwide.
Brian Chin: Thanks, Mike.
Operator: Our subsequent query comes from the road of Robert Mertens of TD Cowen.
Robert Mertens: Hello, that is Rob Mertens for Chris. Thanks for taking my questions. You talked about the higher than anticipated gross margins within the probe-card enterprise, largely resulting from greater volumes and could also be round 200 foundation factors or so attributed to the combination. Was the optimistic combine, largely the high-bandwidth reminiscence enhance or extra of a normal combine all through each the Foundry Logic and reminiscence finish market?
Shai Shahar: So in case you check with Q2 we had been proper on spot, proper. Gross margin was 45.3% versus midpoint of the outlook of 45%. I am undecided what to.
Robert Mertens: Simply when it comes to the gross margins, I believed I heard two-thirds pushed by volumes within the third pushed by combine?
Shai Shahar: That is been elevated from Q1, so Q1 gross margin, in case you recall, was within the excessive 30s, 38.7%. So the rise from Q1 to Q2 from the 38.7% to the 45.3%, about two-thirds of it pertains to quantity and a couple of third to combine. And that change in combine is extra HBM, extra Foundry Logic, resulting from
Robert Mertens: After which might you simply present a bit of extra shade on the progress with {qualifications} of males primarily based probe-cards for GPU checks and perhaps what the dimensions of the market alternative might seem like?
Michael Slessor: Yeah, I feel in sizing it’s attention-grabbing perhaps to get everyone on the identical web page as the most important GPU producer transitions to superior packaging, you are primarily co-los on the world’s largest foundry that requires mems brokers, proper? The legacy probe playing cards, they use for monolithic GPUs, not work. And so and we’re in a qualification course of there. ‘d say our main competitor is has been forward of us and has acquired the preliminary enterprise sort of what we have been extra targeted on HBM and that chance. However it’s a strategic crucial us for to qualify there right here in 2024 and make some progress. And also you assume, once more, comparatively troublesome to grab the chance. However as you heard from and as you heard from among the reminiscence producers, there’s an attention-grabbing asymmetry related to a few of these AI merchandise the place the quantity of reminiscence within the package deal is nearly 10 instances that of the GPUs. And so for unit-driven companies like ours and HBM turns into a way more compelling alternative. That is to not diminish the significance of qualifying throughout this vital transition of GPUs to needing mems primarily based applied sciences for probe. However we like clearly the publicity to HBM.
Robert Mertens: Obtained it. Thanks.
Operator: I might now like to show the convention again to Mike Slessor for closing remarks. Sir.
Michael Slessor: All proper. Thanks, everyone, for becoming a member of us right now. We have got a few conferences in late August and in early September. That we hope to see you at, till then take care.
Operator: So this concludes right now’s convention name and thanks for taking part. Chances are you’ll now disconnect.
This text was generated with the assist of AI and reviewed by an editor. For extra info see our T&C.
