Aehr Test Systems, Inc. (AEHR) Earnings Call Transcript & Summary

Good day, and welcome to the Aehr Test Systems Fiscal 2021 Fourth Quarter and Full Year Financial Results Call. Today's conference is being recorded. At this time, I would like to turn the conference over to Mr. Jim Byers of MKR Investor Relations. Please go ahead.

Thank you, operator. Good afternoon, and welcome to Aehr Test Systems Fiscal 2021 Fourth Quarter and Full Year Financial Results Conference Call. With me on today's call are Aehr Test Systems' President and Chief Executive Officer, Gayn Erickson; and Chief Financial Officer, Ken Spink. Before I turn the call over to Gayn and Ken, I'd like to cover a few quick items. This afternoon, right after market close, Aehr Test issued a press release announcing its fiscal 2021 fourth quarter and full year results. That release is available on the company's website at aehr.com. This call is being broadcast live over the Internet for all interested parties, and the webcast will be archived on the Investor Relations page of Aehr Test's website. I'd like to remind everyone that on today's call, management will be making forward-looking statements today that are based on current information and estimates, and are subject to a number of risks and uncertainties that could cause actual results to differ materially from those in the forward-looking statements. These factors that may cause results to differ materially from those in the forward-looking statements are discussed in the company's most recent periodic and current reports filed with the SEC. These forward-looking statements, including guidance provided during today's call, are only valid as of this date, and Aehr Test Systems undertakes no obligation to update the forward-looking statements. Now with that said, I'd like to turn the conference call over to Gayn Erickson, President and CEO.

Thanks, Jim, and good afternoon, and welcome to our fiscal 2021 fourth quarter and full year's earnings conference call. Thank you for joining us today. Let's start with a quick summary of the highlights of the quarter and the improved business momentum we're experiencing. And then I'll dig deeper into our expectations for increased revenue growth in our new fiscal year, which is already underway. We saw continued signs of recovery and a strong increase in customer demand during the fourth quarter, which is a positive turnaround from the customer production ramp delays and pushouts we experienced this past year related to COVID-19. I'm pleased to report improved revenue and operating profit for the fourth quarter that reflect a return to above pre-pandemic levels. Fourth quarter revenue was up 45% sequentially quarter-on-quarter and up over 100% from Q4 last year, and we're profitable for the quarter on a GAAP basis. Additionally, we're off to a strong start for fiscal 2022, with $5.4 million in bookings and an effective backlog of $7 million as of today. We're seeing improvement in multiple test and burn-in segments, including silicon carbide, silicon photonics and mobile sensors, each of which we expect will contribute to our expectations for significant revenue growth year-over-year in our new fiscal year. Now let me dig a little deeper into each of these opportunities, starting with silicon carbide. This past fiscal year, we made significant inroads into the emerging silicon carbide device market, which continues to be a very promising key growth driver for Aehr and will be a major focus in the coming fiscal year. Silicon carbide power semiconductors have emerged as the preferred technology for battery electric vehicle power conversion in onboard and off-board electric vehicle battery chargers and the electric power conversion and control of the electric engine. Our FOX-P family of products are very cost-effective solutions for ensuring the critical quality and reliability of devices in this market, where performance and reliability cannot only mean increased battery life but also whether you have to walk home from a vehicle whose power semiconductor fails in the powertrain. I once heard from a very wise general manager of automotive semiconductor components supplier that the quality and reliability of devices such as these power semis in the engine drivetrain of vehicles should not be measured by how many failures per million they have, but how many walk-homes per million they have as a failure in this component results in the driver and passengers walking home after it fails in the vehicle. During this past fiscal year, our lead silicon carbide customer qualified Aehr's FOX-XP system for high-volume production burn-in and infant mortality screening of silicon carbide power devices at wafer level for electric vehicles. This customer is a leading Fortune 500 supplier of semiconductor devices with a significant customer base in the automotive semiconductor market. They have now qualified several devices for automotive applications on our solution, ordered multiple FOX-XP systems and have purchased multiple new WaferPak contactor designs that are expected to be qualified and moved to production during this new fiscal year. During our fiscal fourth quarter, we received and shipped a follow-on order from this customer for an additional FOX-XP system for higher volume production test and burn-in of those devices. In this past week, we announced another follow-on from them for an additional FOX-XP system and multiple WaferPaks to meet their increased production capacity needs. These follow-on orders for additional FOX-XP systems and WaferPaks are the result of our working closely with this lead customer to achieve their test requirements and validation of our FOX-XP platform and WaferPak full wafer contactors as their production-qualified solution. This customer is forecasting orders for multiple additional FOX-XP systems and WaferPaks this year and a significant number of systems and WaferPaks over the next several years due to the electric vehicle semiconductor test and burn-in demand. Each of these silicon-carbide-focused FOX-XP systems are configured to test 18 silicon carbide wafers in parallel in the footprint of a typical single wafer test solution while contacting and testing 100% of the devices in parallel on each wafer. Our solution cannot only test 4-inch and 6-inch silicon carbide wafers but can test the future 200-millimeter 8-inch wafers planned to be introduced over the next several years. Aehr provides a unique, fully integrated solution that includes the test systems, full wafer, WaferPak contactors and the WaferPak aligners. In addition to the very large opportunity for silicon carbide with our lead customer, we're currently engaged in detailed and very promising discussions with several other major suppliers of silicon carbide, some of which have also publicly indicated plans for significant capacity increases. We expect to move to on-wafer evaluations with multiple potential new customers this fiscal year. As I mentioned on our last call, we're very excited that a potential new customer -- I'm sorry, that a new potential customer that produces silicon carbide power devices has asked us to demonstrate our full wafer level burn-in solution on their silicon carbide wafers, including putting the system on their manufacturing floor to demonstrate our capabilities. While this is a new customer for us for silicon carbide, they're actually currently a customer for us in other application and already have several FOX-XP systems in production for testing and burning in mobile sensors. They are a significant player in silicon carbide right now, and we're confident that we can prove to them that our solution will catch their infant mortality failures that otherwise show up in customer devices. As I've discussed many times, silicon carbide is an optimal material for high-power and particularly high-voltage devices for applications such as electric and hybrid electric vehicle powertrain and electric vehicle charging infrastructure. These devices reduce power loss by as much as greater than 75% over power silicon alternatives like IGBT devices, which has essentially changed the entire market dynamic. With this development, we've seen most, if not every, automotive company that's working on electric vehicles moving to silicon carbide-based powertrain and charging systems in the near future. The challenge with silicon carbide is that it's known to have high infant mortality rate. However, with the reliability burn-in and screening that Aehr is able to offer with our FOX product solutions, these defects can be removed to provide extremely reliable devices for these mission-critical applications. Aehr's FOX-XP solution allows for one of the key reliability screening tests to be completed on an entire wafer full of devices, basically testing all of them at one time, while also testing and monitoring every device for failures during the burn-in process to provide critical information on those devices. This is an enormously valuable capability as it allows our customers to screen devices that would otherwise fail after they are packaged into multi-die modules, where the yield impact is 10x or even 100x costly. Okay. Sorry. We had some background noise going up. Silicon carbide appears to be one of the hottest potential application spaces Aehr Test had seen in many years. And we're extremely excited about our ability to service this emerging market. We anticipate that wafer level test and burn-in will become the industry standard for quality and reliability screening for silicon carbide devices for the automotive market. And with the most cost-effective solution on the market to address this opportunity, we believe that Aehr has a chance to build a dominant market share. Power semiconductor market for electric vehicles is expected to triple between 2020 and 2026, growing at nearly 26% CAGR to $5.6 billion, according to Yole Research. And a report with Deloitte forecasts total electric vehicle sales will grow at a CAGR of 29% from 2020 to 2025 before reaching 31.1 million units by 2030 and securing approximately 32% of the total market share for new car sales. These stats highlight the tremendous opportunity Aehr Test has in front of it, with its wafer level test and burn-in solution for electric vehicle semiconductors. Now turning to silicon photonics. We're seeing an improvement in the silicon photonics market, which has -- was significantly impacted by the pandemic this past year. During the fourth quarter -- or fiscal fourth quarter, we shipped FOX-XP system and multiple WaferPak contactors to an existing customer that's transitioning from our FOX-NP system for initial production burn-in to our production FOX-XP system to begin volume production of their high-performance silicon photonics devices. This is a -- this customer is a major supplier of fiber optic transceivers in the data center interconnect market today. Silicon photonics fiber optic transceivers, which are used in data storage and 5G infrastructure, require a process step in manufacturing called stabilization, where the devices are subjected to high temperatures and power to stabilize their output power. Our customers are using our FOX wafer level test and burn-in solution for production test and burn-in of their integrated silicon photonics devices, and we currently have 5 silicon photonics customers that are shipping products to their end customers using our FOX solutions. We see a significant opportunity for growth as we expand within these customers and add additional new silicon photonics customers in this fiscal year. Now let me touch on the mobile sensor market. This past year, we successfully implemented our FOX systems and DiePak carriers for production test and burn-in of 2 new applications for 2D/3D sensors for mobile devices. Aehr has now successfully executed on a number of programs for highly custom and unique sensors in packaging and configurations unlike any other devices on the market. Aehr's engineering team has been able to design and develop custom DiePak carriers and contactors to address the unique electrical, mechanical, optical and thermal needs of these devices with our FOX-XP systems and proprietary DiePak carriers. These solutions are turnkey with development of the proprietary carriers and test schematic, custom DiePak singulated die and module socket and carriers, highly proprietary thermal conductivity and transfer solutions, custom application test plans and automated handling systems to load and unload our DiePaks with 100% traceability of test results and burn-in. We feel we continue to meet and exceed the customers' expectations and are happy to continue to meet their needs on these extremely challenging test and burn-in applications. We expect to see follow-on orders for system capacity in DiePaks this year and continue to be optimistic about this market space. Now let me talk a little bit about our consumables and contactor business. As I mentioned on past calls, our FOX family of test systems include our customized WaferPaks and DiePaks that are proprietary, full wafer singulated die and module contactors and are needed not only for full -- for new system orders but also for each new design win or each new device added to production test. During the fourth quarter, we launched our newest DiePak solution, which is capable of handling extremely small and complex devices and very high power density devices with higher parallels than ever before. This new class of DiePak can handle devices small enough to rest on the tip of a pen or a pencil. Devices this small are extremely hard to handle and particularly in any kind of parallel. Often, a discrete device this small is handled with special handling equipment and a tester that can only test one device at a time. This new FOX-XP system and DiePak solution is capable of testing very complex die modules in addition to them being tiny. This solution is a great addition to our product family, and we believe it further sets us apart from any other company in the industry. As we increase our installed base of FOX systems with current and new customers, particularly with our FOX-MP and XP multi-wafer and singulated die module test and burn-in systems, we expect this consumables business will continue to grow in absolute value and also as a percentage of our total sales. Over the long term, we expect these recurring consumable sales to account for up to a half or even more of our total annual revenues. And lastly, touching on our packaged part business, we continue to see indications of renewed demand for packaged part burn-in applications, particularly from customers in automotive applications and those seeking high-voltage capability. Our new packaged part burn-in product with very high voltage test capabilities continues under development, and we expect to generate additional new opportunities with our planned shipments to begin later this fiscal year. So let me go ahead and touch on supply chain. I received many questions from customers and shareholders about our supply chain and ability to meet capacity demands for systems and contactors. Given all the issues with semiconductor shortages and rising cost in raw materials and lengthened lead times across many industries, I quickly had to acknowledge this -- the reasonable concern and understand where the questions are coming from. Aehr has a very robust supply chain with world-class subcontract manufacturers on subsystems of our test systems, contactors, WaferPak aligners and DiePak handlers. These subcontractors have successfully supplied these subsystems for years to Aehr and are very mature. In all cases, these suppliers have capacity well in excess of Aehr historical shipment and the ability to ramp significantly higher as well. We're very confident in our ability to meet the customer-forecasted demand plus considerable upside. The one area that we do want to highlight is the risk associated with semiconductor component lead times that have been very irrational over the last 6 months, which is causing us to jump through many hoops to ensure we have near- and long-term volumes to meet both our forecast and considerable upside to this forecast. I don't want to overstate the risk as we're confident in meeting our projected revenues that we're guiding for this fiscal year and we believe we can meet a considerable upside to this forecast. However, we know there's a risk in the near term in particular with some semiconductor components but also have seen logistics issues with shipping lead times that continue to make our team keep on their toes and may add to the lumpiness of our quarterly shipment and revenues. So let me conclude my remarks. We're now 1.5 months into fiscal 2022, and we're confident in our revenue growth projections for this new fiscal year. We're seeing a recovery across our customer base, along with significant demand for wafer level test and burn-in of silicon carbide devices for electric vehicles, silicon photonics devices for data center and 5G infrastructure, and 2D and 3D sensors for mobile devices. For the fiscal year ending May 31, 2022, Aehr expects full year total revenue to be greater than $28 million, which would represent growth of approximately 70% year-over-year and to be profitable for the fiscal year. Now before I turn over the call to Ken to go over our financials, I want to note that during the quarter, we announced the appointment of Fariba Danesh to our Board of Directors. Fariba is a technology industry veteran with special emphasis on semiconductor, photonics, telecommunications and data storage, and she brings incredible knowledge, experience and contacts in the compound semiconductor and optical semiconductor spaces. We, and certainly I, are really excited to have her on our Board. With that, let me turn it over to Ken to review our financial results in more detail before we open up the line for questions.

Thank you, Gayn, and good afternoon, everyone. As Gayn noted, we saw continued signs of recovery and a strong increase in customer demand during the fourth quarter, resulting in improved revenue and operating profit for the fourth quarter that reflect a return to above pre-pandemic levels. To add some perspective on this, our fourth quarter revenue of $7.6 million is our highest reported quarterly revenue since Q2 of fiscal 2018. Looking at our financial results in more detail, starting with the fourth quarter. Fourth quarter net sales of $7.6 million are up 45% sequentially from $5.3 million in the third quarter and up 102% year-over-year from $3.8 million in the fourth quarter last year. The sequential increase in net sales from the preceding Q3 reflects an increase of $2 million in wafer level burn-in revenue and $386,000 in customer service revenues. The increase in wafer level burn-in revenues is primarily due to an increase in system revenue of $1.4 million and an increase in WaferPak, DiePak revenues of $577,000. The increase from Q4 last year includes an increase in wafer level burn-in revenue of $3.4 million and customer services revenue of $509,000. The increase in wafer level burn-in revenue is primarily due to an increase in system revenue of $3.8 million, partially offset by a decrease in WaferPak DiePak revenue of $487,000. It is important to note that there were no system revenues in Q4 of fiscal 2020. Non-GAAP net income for the fourth quarter was $870,000 or $0.04 per diluted share compared to a non-GAAP net loss of $464,000 or $0.02 per diluted share in the preceding third quarter, and a non-GAAP net loss of $720,000 or $0.03 per diluted share in the fourth quarter of the previous year. The non-GAAP results exclude the impact of stock-based compensation and, in the fourth quarter of fiscal 2020, included a $1.6 million excess and obsolescence provision and $220,000 in restructuring charges. On a GAAP basis, net income for the fourth quarter was $567,000 or $0.02 per diluted share. This compares to GAAP net loss of $735,000 or $0.03 per diluted share, which included a $337,000 warranty provision in the preceding third quarter and GAAP net loss of $2.9 million or $0.13 per diluted share, which included the impact of $1.9 million or $0.08 per share in inventory write-down and restructuring charges taken in the fourth quarter of the previous year. Gross profit in the fourth quarter was $3.5 million or 46% of sales, up from gross profit of $1.9 million or 36% of sales in the preceding third quarter and a gross loss of $93,000 or 2% of sales in the fourth quarter of the previous year. The increase in gross margin from the preceding Q3 is primarily due to a decrease in unabsorbed overhead cost as percent of sales due to higher revenue levels in Q4 '21, accounting for a 5.5 percentage point improvement in gross margin and a decrease in other cost of goods sold, as Q3 '21 included a warranty provision accounting for 4.8 percentage point improvement in gross margin. Because our manufacturing overhead costs are relatively fixed, we scaled very well. As our revenues grow, the increases flow to the bottom line and our margin percentage are favorably impacted, which is reflected in our Q4 '21 results. Product mix also impacts our gross margin percentage. The increase in gross margin from the fourth quarter last year is primarily due to a decrease in inventory reserves, as Q4 last year included a $1.6 million charge related to the write-down of excess and obsolete inventory, accounting for a 43.2 percentage point improvement in gross margin. Unabsorbed overhead cost decreased as a percentage of sales, resulting in a 7.2 percentage point improvement in gross margin due to higher revenue levels in Q4 '21 compared to Q4 '20. Operating expenses in the fourth quarter were $2.9 million, up $387,000 or 15% from $2.5 million in the preceding third quarter and up $185,000 or 7% from $2.7 million in the fourth quarter of last year. The sequential and year-over-year increase in operating expenses is primarily due to an increase in employment-related expenses for bonuses in Q4 '21 and annual pay increases to employees effective Q4 '21 and an increase in R&D project materials. This was partially offset by a reduction in restructuring charges, as Q4 '20 included $220,000 in costs related to the closure of our subsidiary in Japan and reduction of head count in our Germany subsidiary. With customer activity and business improving, we eliminated the 30% pay reductions of our -- for our executive staff at the start of June 2021, the beginning of our current fiscal year. SG&A was $1.9 million for the fourth quarter, up $261,000 from the preceding third quarter and up $230,000 from the prior year fourth quarter. R&D expenses were $1 million for the fourth quarter, up $126,000 from the preceding third quarter and up $175,000 from the prior year fourth quarter. Now turning to the results for the full fiscal year. Net sales for fiscal 2021 were $16.6 million, down 26% from net sales of $22.3 million in fiscal 2020. The decrease includes a decrease in wafer level burn-in system revenues of $5.8 million. Customer service revenues were relatively flat. While year-over-year net sales decreased, second half fiscal 2021 revenues were $12.9 million compared to $9.9 million in the second half of fiscal 2020, an increase of 31% over prior year. Fiscal 2021 net sales were -- comprised of $13.1 million in wafer level burn-in revenues and $3.5 million in customer service revenue. For the full year 2021, system revenues accounted for 44% of revenues compared to 36% in the -- in prior 2020. WaferPak and DiePak consumable revenues accounted for 35% of total revenue in 2021, compared to 48% of revenues in fiscal 2020. Customer service revenues accounted for 21% of revenues in fiscal 2021, compared to 15% of revenues in fiscal 2020. Non-GAAP net loss for fiscal 2021 was $3.3 million or $0.14 per diluted share, which exclude the impact of stock-based compensation and a noncash net gain of $2.2 million and a tax benefit of $215,000 related to the closure of Aehr's Japan subsidiary in the first quarter. This compares to a non-GAAP net loss of $27,000 or $0.00 per diluted share, which excludes the impact of stock-based compensation expense, inventory reserves of $1.6 million and restructuring charges of $220,000 in fiscal 2020. On a GAAP basis, net loss for the fiscal year was $2 million or $0.09 per diluted share. This compares to a GAAP net loss of $2.8 million or $0.12 per diluted share, which included the impact of approximately $1.9 million or $0.08 per share in inventory write-down and restructuring charges taken in fiscal 2020. Gross profit for fiscal 2021 was $6 million or 36% of net sales compared to a gross profit of $8.4 million or 38% of net sales in fiscal 2020. The decrease in gross margin percentage in FY 2021 compared to the prior year is primarily due to a decrease in unabsorbed overhead costs to cost of goods sold related to higher revenue levels in the prior year, a change in product mix and an increase in warranty cost as a percent of sales. This was partially offset by a reduction in inventory reserves as FY '20 included the $1.6 million provision for excess and obsolete inventory. Operating expenses for fiscal 2021 were $10.2 million, a decrease of $922,000 or 8% from $11.1 million in fiscal 2020. The decrease is primarily due to a decrease in SG&A of $968,000 and restructuring charges of $220,000, partially offset by an increase in R&D of $266,000. SG&A was $6.6 million in fiscal 2021, down from $7.5 million in fiscal 2020. The decrease includes a decrease in labor-related costs resulting from cost reduction initiatives implemented in fiscal 2021, lower commissions related to a decrease in bookings and revenues and lower travel and sales expenses due to restrictions in place from the pandemic. R&D expenses were $3.7 million in fiscal 2021, up from $3.4 million in fiscal 2020 due to higher R&D project materials and employment costs due to head count increases. Turning to the balance sheet for the fourth quarter. Our cash and cash equivalents were $4.6 million at May 31, 2021, down $156,000 from $4.7 million at the end of the preceding quarter. Included in the cash balance at Q4 '21 and Q3 '21 quarter ends were $1.4 million in borrowings under our line of credit. Accounts receivable at quarter end was $5.2 million, an increase of $2.5 million compared to $2.7 million at the preceding quarter end, related to the increase in revenue in Q4 compared to Q3 and an increase of $1.5 million from Q4 last year. Inventories at May 31 were $8.8 million, an increase of $510,000 from $8.3 million at the preceding quarter end. The increase in inventories at May 31 is to support forecasted revenues, obtain long-lead-time materials and to ensure an adequate supply of critical components. Property and equipment was $677,000 compared to $617,000 to preceding quarter end. Customer deposits and deferred revenue, short and long term, were $288,000, a decrease of $379,000 at the preceding quarter end related primarily to the decrease in backlog from the prior quarter. Our current and long-term debt of $1.7 million is related to funds we received during the fourth quarter of the last fiscal year under the Paycheck Protection Program through the Small Business Administration. Last month, we received notice from Silicon Valley Bank that the Small Business Administration has forgiven the loan and accrued interest. We will be posting a benefit of $1.7 million in our first quarter fiscal 2022 results related to the loan forgiveness. Bookings in the fourth quarter totaled $5.5 million. Backlog at May 31 was $1.6 million compared to $3.7 million at the preceding quarter end. Effective backlog, which includes backlog at the end of the fiscal fourth quarter plus orders since the end of the fourth quarter, is $7 million. Now turning to our outlook for the coming fiscal year. As Gayn noted, we're 1.5 months into our fiscal 2022 and off to a strong start, with $5.4 million in bookings and effective backlog of $7 million as of today. We are seeing a recovery across our customer base, along with significant demand for wafer level test and burn-in across our markets, which is giving us confidence in our revenue growth projections for this new fiscal year. For our fiscal 2022 year ended May 31, 2022, we expect full year total revenue to be greater than $28 million, which would represent growth of approximately 70% year-over-year and to be profitable for the fiscal year. This concludes our prepared remarks. We're now ready to take your questions. Operator, please go ahead.

[Operator Instructions] Our first question comes from Christian Schwab of Craig-Hallum Capital Group.

Congratulations, guys, on a great quarter and a solid outlook. Gayn, in your greater than $28 million outlook, can you tell us your expectations for how many 10% customers you'll have in that?

Let me think. My guess is we'll have 2 or 3. 3 -- I'm getting the signal 3 from Ken off to the left, about 3 in our current estimates.

Okay. Fantastic. And then by application, can you walk us through between silicon carbide, silicon photonics and just maybe recovery orders from existing customers that were kind of pushed out last year? Can you give us an idea of the mix by application at all?

Yes. I think I want to be a little careful, not for any proprietary reasons, but just always difficulty in understanding how it looks like. But let me just give you a little bit of windage and elevation. I think we're quite confident that silicon carbide will be our largest segment this year. And I think as it plays out, then it would be followed by silicon photonics and then the 2D/3D sensor market, each of those being substantial, 15%, 20% or more. So you kind of do your own math there. I think we do plan to get some -- see some packaged part burn-in business, but it will still be relatively small. I think we'd probably do about 10% of service and support. But clearly, we do see strength in silicon carbide. And I think pretty confidently, it will be the largest segment this year.

Fantastic. And then my last question has to -- we talked a lot about electric vehicles. And have you guys done the math? As penetration rates look to improve meaningfully over the next 5 to 10 years for sure, do you have any idea of the potential market TAM that your products could address over kind of a mid- to long-term basis?

We do. And I think what I do is I may defer and take that to -- as we go forward with a little bit more clarity. But we -- as we are engaging with multiple customers and we start to understand their capacities and their anticipated test and burn-in times and the level of quality, expectations versus those times, it continues to reinforce a model that I think I shared in our last call, which is if you just start to take a look at the number of systems that are required for every, call it, 1 million cars that are out there, it's substantial. I think we estimated about 8 systems or so for every 1 million incremental cars that are shipped per year. And so you start looking out at 30 million cars in 2030, which -- fair enough, that's 8, 9 years away. The world needs to purchase a lot of our full-complement FOX-XP class systems. And currently, we have not only an enormous cost advantage and -- but footprint advantage if a large silicon carbide customer is likely to have tens of systems for certain on their floor, keeping in mind that each system has 18 wafers in it. And another way of thinking about it for every 10 systems, they can basically process 180 wafers per day at 1-day test times, for example. And you start looking at capacity forecasts that are out there, that are well in excess of that in order to meet the silicon carbide requirements of the battery electric vehicle market and the on- and off-board chargers, and you can see that it's a very significant number. So I -- maybe I'll give you a little bit more clarity. I'd like to make sure I have all our math done. And as we capture and understand kind of consistently -- by the way, we're doing a lot of our own research in testability and quality, reliability. So we're able to actually offer to customers unique proprietary test solutions. We are not going into a customer and they tell us exactly how to test it, and then we're running around the world and telling them. We actually got proprietary test schematics that we don't even share with our customers on how we're doing this, having to do it -- to be able to take advantage of the FOX-XP system and full wafer contacting on wafers that range from 4-, 6- and going to 8-inch and have device counts from 500 to 3,000 devices on a wafer that can all be tested on a -- one of our FOX-XP blades. So as we do that and we understand how quality and reliability correlate to their test times, I think we'll be able to speak more confidently as a generic stance in the market on what it looks like, but data continues to suggest that you'll need to do 24-hour-plus burn-in times on any wafer to try and remove the infant mortality issues before they sneak into even discrete and certainly modules. So again, it is certainly the biggest market we're addressing right now. And there is -- there are some estimates that suggest it's as large as the memory burn-in market.

[Operator Instructions] Our next question comes from Jon Gruber of Gruber & McBaine.

I have a -- I guess it's a nitpick here. You said your effective backlog is the backlog plus the orders, but then it should be minus the shipments since the quarter is 60% over. I take that you did not take the shipments out of it. Is that correct?

Correct.

So it's really not backlog, correct?

That is correct, yes.

That's accurate, yes.

Then it's not backlog then, okay. So don't call it backlog. Okay. Second of all, in excess of $28 million, I mean if things go well, are we talking $28 million to $30 million? Or are we talking $28 million to $36 million? I mean give us sort of some range here that -- because $28 million -- you were looking for $28 million 3 years ago. And so $28 million in a good year is sort of [ eon ]. So how do you -- if things go well, what's the upper end here on the shipment side?

Yes. I appreciate the question and I fully understand that in fact, last year, we were talking about doing $28 million, and we clearly were much less than that. And I do -- I sit here with certainly more confidence because that would -- at least certainly in the U.S. and U.S.-based customers and the customers that are talking about ramping, they seem to have gotten through the COVID aspects and are letting us get on their floor and do the installations and all at will. I -- we chose the words pretty specifically not to be coy. I actually don't want to put a high end on it. There is significant upside to the $28 million. And as we get the bookings and print them and give us confidence, we'll go from there. But candidly, Jon, we've been talking about how great it's going to be for a while. And we'll -- we continue to believe we're taking a conservative stance in the forecast we gave you, I believe, stronger this year than ever before, but I will -- we'll give an answer...

I take if you give an answer that you're talking $28 million to $36 million, more likely than $28 million to $30 million, correct?

If you had to pick a number and I'm not going to have you do another one, yes.

[Operator Instructions] We'll take our next question from Larry Chlebina with Chlebina Capital.

Congratulations on the highest EPS quarter, I think, in 6 years, right? Is that correct?

It sounds about right. Thank you.

On the next silicon carbide hopeful customer that you discussed, that's a current customer on your 3D sensing, have you started the testing of those wafers yet?

I'm going to be...

Or not quite yet?

Yes, not quite yet, soon.

That should start any time now? Is there any -- like within days?

I'll tell you what, Larry. I'm not trying to be coy to you folks. Competitive knowledge is -- there's lots of ears out there. I'm less concerned about being clear with my shareholders than I am with being -- letting everybody know what we're up to exactly. But we will be on wafer with test results this quarter and we're already halfway through it.

The $4.3 million system that you delivered in the May quarter -- and obviously, something slipped on that application. Is that -- the application, is there a problem with it where it may actually not show up in the next product release? Or is it just kind of delayed, I'll say, a month or 2?

Larry, actually, I'm sorry. I'm not really following you. The $4.3 million in what segment?

For the pick and place application on the sensor.

Okay, okay, okay. Yes. I mean actually, the revenue for that, we think, was in Q3 or something. Okay. Fair enough. What was the question again?

At any rate, there apparently was a problem, and that product launch, the automated product launch may have slipped. But is it just a slippage of the launch of that product and why you haven't had a follow-on order?

Yes. Let me -- well, you guys keep -- these are really good questions again. That particular customer, in particular, is unbelievably secure and understanding. What I will share is the following. We continue to believe that, that program is going well. We continue to believe we have a plan of record and we do have forecast for incremental systems and DiePaks in our fiscal year. And I think that's the most I can say right now. But generally speaking, at least I personally believe that, that was going to happen a little sooner. And I'll leave it at that rather than talking about whether there was actually a product slip or not. I think that's best but they're very happy with us.

All right. Then lastly, any opportunity on the big OSAT that you're currently engaged with or other...

You know what, the OSAT and their kind of front-end customer, each of which have systems now, there is active engagement with customers beyond the first lead customer for those guys. And Vernon has been giving you updates on where they are. I still think -- at some point, we want to work this in. I still think there's still some front-end new customer engagement impacts related to COVID. I plan to fly to Europe next Sunday, and our reps there said, "Get on Teams. The customers aren't going to see you." That -- we're still doing everything phased, over the phone, et cetera. So while the U.S. -- I mean at least in California here for the most part, it's just wide open. Ken and I are in the office now, and we actually are not masked in the facility for everybody who's been vaccinated. So -- but I'll tell you what. Customers still -- there's still a little bit of slowing on that. But that's okay. I don't really need that to hit my plan or the upside. But I'm still hopeful. We know that they're engaged. They're serious about it. But we've heard some things too from the big OSAT, but they're pretty booked out that -- a lot of the interesting things with the whole semiconductor supply chain as they talk about, "Oh, we're booked all the way for the next 6 or 9 months." So I'm not currently all that, I'll call it, hopeful. And I'm certainly not forecasting a lot of volume in sort of new customers and that -- through them this year, but maybe late in the year and definitely into the following year, if that helps.

Yes. And you said you're traveling to Europe next week, did you say?

I didn't say where. I don't think I did.

All right. You just said...

But I'll tell you what. It was Europe. It was where I was headed. And I was probably to go Sunday, and I'm not now. So...

So you're not going?

A week from Sunday. No, I'm going to sit here and do some more Teams meetings and things like that for right now. So even though I'm the CEO, I think they -- I'm a thinly disguised sales guy too sometimes. So I think right now, they'll let applications engineers and support people to go in and install things, but there -- a lot of the restrictions in Europe, up and down, are still with salespeople. So I would just -- and that's true also in Asia.

One last question then on the CP opportunity for the data center. Is that still ongoing? And do you expect that to get kicked off anytime soon?

It is still ongoing. They use it every day. In fact, I think last quarter or the quarter before, they were buying some parts and spares, and that system continues to have incredible uptime. I was curious why they were buying some parts since it hadn't gone down or anything. But my understanding is they're using it like 24/7 right now. It's my understanding -- and since we've been so elusive about who it is, I'll be a little bit more forward because actually, no one's guessed yet who it is. My understanding is that program has been pushed out in time again. And I think we have very little to -- if any, forecasted in this fiscal year again. But they're continuing to operate with full intent to go down that path, but it's -- that project is kind of continuing to push out. I'd say the total is, what, 2 years different than what we originally heard so far.

All right. Is that a silicon photonics application? Would you...

I haven't said what it is. I called it a data storage application and I've been wonderfully elusive that nobody's guessed who it is because once you find out, you'll know why. Sorry.

We'll take our next question from Matt Winthrop of Aegis Capital.

Congratulations. It sounds like things are doing much better. And I'm not going to beat you up like these other guys. I'm excited to the story.

You're just going to ask me if it's greater than $36 million. I'm waiting for each one of you to ask just one question. I know. All right. Go ahead, Matt.

No problem. I have 2 quickies. Can you sort of, in layman's terms -- because I talk to clients all the time. I'm a retail guy. I'm not an analyst. On the electric vehicle, how can I explain, in an elementary level, what your product addresses in terms of setting the average [ efficiency ]? Is that okay to ask?

Yes, yes, yes. I'll try to. All right. So on -- you walk out to a Tesla, okay? And the Tesla basically has an onboard charger internally, okay? So what it means is if you apply electric power to it from your grid in your garage, it actually converts that AC electricity to DC and charges the onboard batteries. That AC-DC charger has these power semiconductor FETs inside that are made out of silicon carbide, which are way better than IGBTs. And in fact, Tesla and Elon Musk are credited with the people that basically, I guess, had the nerve to try silicon carbide without going through the tens of years of other things and have identified it as a superior product, put it in their Model 3 and immediately had like a 10%, 20% extended battery range. Since then, they have put it in all of their cars. And as we understand, all suppliers are following suit. And I think that's well publicized by multiple people out there. So in that charger, it converts AC to DC and it's extremely efficient. And what that means is you can do it faster and you can do it with more efficiency. Then when the battery is charged, the funny thing is the battery, which -- you can only charge battery with DC, direct current, just like a regular battery in your flashlight, right? But then the motor in the Tesla and everyone else is actually an AC motor. So it's not a DC motor. So now you have to convert it back to AC in order to work, and that's called an inverter and it uses the same MOSFETs, right? So it's pretty interesting. And so their MOSFETs are there to efficiently charge your car and there to efficiently power the car, and they're directly in line with the power to the electric engine. So if they fail, yes, your car is dead, right?

Okay.

So there's one other place and that is -- more and more, we're seeing them, where -- like the electric gas stations, where you go and there's an electric charge pump. They -- those are big converters as well. And they also have silicon carbide electric MOSFETs in them. And some of these modules, if you look at them -- I've got one in my hand. That doesn't do you any good, but you look at this module and it's -- some of our -- about the size of your hand or a small hand. And in there are 8 or 10 different semiconductors. In this case, they're the silicon carbide FETs, switches. And they -- the reason they have 8 or 10 of them in there is because they're all in parallel so they can have 400 amps of current. So maybe each one of them can only do 50 amps. But if you put 8 of them in parallel, you can do 400 amps. So what -- the first thing when -- the first customer that was using us was realizing that they put 8 of these silicon carbide devices into this module, right? And then they had to burn it in. But every one of the devices had some material failure rate. Let's say 1%. So they had -- on 8 devices, you have an 8% failure rate in 24 hours of that device, and you have to throw away the module. Those modules, if you go right now and click on DigiKey or Mouser or eBay, they're $600 okay? And they're coming down quickly. But imagine that you have one of the devices fail and you have to throw away a $300 module or something. So our value proposition was we test the devices before they're put into the module. And not only is it cheaper per device, but now all of that -- all those failures are removed so you don't throw away the $100 package.

But this technology...

The cost effect -- yes.

Is it fair to assume this technology also would be applicable to a Ford or a GM or an Audi? It's not just a Tesla thing. This is basically...

100% of them. Every...

Got it.

Every electric vehicle is going to have silicon carbide in it, every one of them.

On the sales side, because last couple of quarters -- you were a little better last time. A couple of quarters ago, you were pretty upset about how things are progressing, but it sounds like you're more excited. You had built this room in your facility where people would come and corona -- is that starting to open up? Are you doing actual displays in this room now that...

Yes. Displays -- we're testing wafers. How's that? Customers still aren't coming, but that's okay. It actually works out pretty nice. We kind of market it as, "Hey, look, it's a touch-free environment, but we're kind of lonely. It'd be nice to have people come here." But we're getting really good at Teams and Zoom and WebEx. But, for example, the customer wafers that are going to be tested with silicon carbide will be in that room.

So you can do it and just virtually show it to them. They don't have to physically be there. They give you the product in advance to test or something like that?

Absolutely. So the -- a silicon wafer has, say, 1,000 die on it, just to make it easy, right? Those devices already have some level of actual failure to them. So they don't get 100% yield. So let's say -- whatever, it doesn't matter how many, 10%, 20%, 30% of the die are failed, they'll give us the wafer map or not. When we test it, I can tell you immediately which devices have failed. And over a 24-, 48-, 96-hour period, I can tell you exactly when it failed on every single device with 100% traceability. So if they're wondering if we're serious, they can give us the wafer without the wafer map. I'll tell them which die had failed and it always correlates.

All right. And I'll leave you with this on an -- that's great. I appreciate it. I'll leave you with this on an [ upside barrel ]. Let's say you've scored up, your new sales manager scored a bunch of really nice contracts, what kind of run rate can you guys -- what is your capacity if like everything hit in the next 3 or 4 quarters? I'm not saying you're saying that's sales, but what could you do out of California without having to expand on your factory, let's say?

So there's a few ways to look at it. We have subcontractors that...

Is that a fair question?

Yes. It's a fair question. I mean the simple answer is there is nothing in our supply chain that couldn't, within 6 to 9 months, get us to even 10 systems a month, right? You just -- some of it, you -- some of it would include some of the people growing and adding some people. But even...

So even on that, what's 10 systems retail out to your -- or wholesale out to your customer?

If you include an ASP of, say, a silicon carbide configuration, let's say $2.5 million or so and the complemented wafer pack set $1.5 million, it's $4 million apiece. So you...

That's a lot.

1 And I -- that's correct. And that seems -- I'm sure people are like, "Okay, okay," but that's real. Even -- I mean that seems like big numbers to us here, but that is not big. I mean 10 systems a month is -- although they have 18 wafers of capacity in each one of them. So if you were comparing it to, say, a J750 from Teradyne, our -- each of our blades has more pins than a J750 and more electrical channels. So each testing...

I'll tell that to my girlfriend. She'll be excited. Listen, so I would suggest -- because you guys are on a roll, get off this Q&A as soon as you can, get on the phone and close a couple more because it sounds like you really have some -- and I know that takes...

Vernon's off doing that right now.

I know they're doing [ it years ago ] but some of these guys -- I'm tired of these guys whining and crying. We're finally here. Let's go out and close it. And I think you've got the opportunity.

Thank you, Matt.

We'll take our next question from Frank Barresi with Ameriprise.

The people with all this COVID, interfering with new installations and sales, a lot of these electric vehicles that are being sold now aren't using your -- I mean they're not using your system in production, correct, as of yet?

Correct, correct.

Okay. And so Tesla, of course, is using it for everything. Are there many other -- of the other...

Frank, let me clarify this. Okay. There's a couple of ways to look at it. So of the electric vehicles, okay, all of them have this inverter and battery charger, et cetera, okay? Until 2018, none of them used silicon carbide. And keep in mind, that's like Toyota Prius, all the hybrid electric vehicles, et cetera, and then Tesla, 0 market share of silicon carbide, okay?

Okay.

Then the Model 3 did it. Then in 2019, Tesla did it. I believe there's an Audi out there right now. I believe that -- my understanding is that Toyota, Taycan is still IGBT. So even some of the new models that are coming out do not use silicon carbide yet. But projections from folks like Yole, which is a big one, Exawatt, which is another big forecaster, they are projecting pretty dominant share to almost 100% share of all of those power modules will -- all of a sudden can -- the power FETs in these devices will go to silicon carbide over the next like 3 to 8 years or something. And now -- so that's the devices. For us, today, we currently have one lead customer, right, who is currently moving everything to wafer level on our system. So you would have to say, well, how many of their customers? How many design wins? How many cars have they gotten into? Who do they have? And we know more than we'll ever admit to, but they obviously don't have 100% market share yet. So that -- we're also capturing other customers. At some point, there's an opportunity for us to compete with every single customer, and potentially everyone could use our tools. Then it doesn't matter who wins. But for now, we -- this is a pretty hot wave for us. And our systems work. It's right in the sweet spot of the capability of this machine, fully loaded, 18 wafers. This is a -- there's a great opportunity for this platform.

And that's just in the -- well, it's charging the battery and discharging the battery needed on both sides and then the chargers that...

Yes.

Okay, okay. And...

Yes.

So basically, like in a lot of the tests, let's say then, they don't have 100 -- I don't know if it's a Tesla necessarily, but the companies that are using silicon carbide, a lot of them, they're not able to do -- they're not able to use your system. So they must have a lot of these failures you're talking about?

No, no. That's not fair. So what they'll do is, just like our lead customer, they actually test them in a packaged form after they've already been packaged up, in many cases with multiple devices per package. So it's -- they basically can do -- we sell packaged part burn-in systems, too. We currently are forecasting 0 for silicon carbide because why would you do that when you could do it at wafer level? So...

Right.

Yes. So everyone is testing silicon carbide and doing some sort of a burn-in of it. There's no way they're shipping those to automotive suppliers without it. And they're either doing it a packaged part, okay; or they're doing it with our system, in this case we've already talked about the one customer; or they're evaluating, trying to figure out how to get to wafer level. We have heard -- every customer we talk to is planning to move to wafer level. Now the question is, well, what is the competitive situation? What's the alternative? The most obvious alternative that we know works, right, we know is in production and is capable is you take an ATE system, automated test equipment system, from multiple suppliers that are out there. You design it. You put lots of power supplies on it and you put it on a prober. And if you take an ATE system that ranges from a few hundred thousand to $1 million and you put it on a prober that's $350,000, right, and you take a probe card that's $50,000 or $100,000, you've got a system. And I'm -- generically, it's a $1 million test cell. And you can test it. And in 24 to 48 hours, you can do theoretically everything we can do. We think we have some competitive advantages with the way we do it, okay? But it's a $1 million test cell for 24 hours. Our system -- remember, I just did the math with you. It's about $4 million divided by 18. You're like a $200,000 or less, right? So you're at 1/5 of the cost, and so you could test it 5x longer for the same price point from a capital depreciation perspective. But a wafer prober, if you've seen one, if you stood in front of it, it's like a -- has a footprint of a lot of Prius, right, and maybe so is our system. It's actually a little smaller than that, so is our system, but it has one wafer. And in that same footprint, we have 18. So now go test 180 wafers, and it's 10 of our systems, 10 Prius next to each other -- actually, Prius is big, but I never used that analogy before but I'll go with it. And with them, it's 180, 180 in a clean room space. So you want to put in place 1,000 wafer starts. You have 1,000 probers, okay, or hundreds of them. So the cost effectiveness is 1/5 from a cost of ownership and capital depreciation on the capital cost, and it's 1/20 of the footprint and we make good margins.

So somebody -- they're going to have to design a system to compete with you? I mean...

And they're going to be stomping all over our patents and IP if they try.

Well, that's good.

Yes.

And I guess is the situation analogous in the silicon photonics? Is it -- does the same -- well that's silicon carbide.

Right.

I'm sorry. What?

Very similar, very similar.

Oh, it is similar. Okay.

Yes. Silicon photonics and silicon -- I mean it's a different application. In this case, it's a fiber optic transceiver fully integrated under a piece of silicon. And the huge advantage of it is they can actually manufacture like 1,000 transceivers at a time on a single wafer. The cost effectiveness is so dramatic that folks like Intels of the world have already put out -- multiple companies have gone under because they simply can't sell them at the cost that the folks like Luxtera or Intel, Inphi or Sequoia, which are known names, are building them for. And so the beauty of it is that with our FOX systems that can either test them in wafer form or singulated die form, we're able to actually do this so cost effectively on the integration that we've heard from our customers that it actually enables them to, in fact, do whole wafer silicon photonics manufacturing. And they -- then they test them and they burn them in. In this case, they stabilize them before they're put into the packages where they're extremely expensive to burn in, and there's yield loss associated with it, same kind of value proposition. The only difference is that on a -- the fiber optic transceiver market is measured in millions of units a year, like 10 million units a year, total. That's it, right? The -- I think a Tesla has 48 MOSFETs in each car. And so if you're going to do 31 million of them, it's literally 2 orders of -- an order to 2 orders of magnitude higher size. Given the same test time, similar die per wafer, the silicon carbide market is absolutely more than 10x larger than a silicon photonics one. So that's to follow all that math, but I was trying to do it quickly.

Sure, sure. And then on the sensors, you've talked about 2 and 3D sensors. I mean is that a...

Historically and it has proven to be -- has not been a very big market. We make good margins. The customer is happy with us. They're very unique but has traditionally been -- and we've talked about it before, is sampling market. What it means is that the devices do not inherently have either a high-enough infant mortality or need the structural stabilization to test every one of them. So the customer has proven to themselves that by sampling some percentage -- we don't get into it but imagine small, they're able to -- by buying just a few systems from us, they can actually sample hundreds of millions of parts a year, right? And we continue to execute for them. We keep our fingers crossed. We have put in place enough infrastructure that if we want to do 100% sampling, we're at -- so far, they have not chosen to do that. Now we do have a couple of devices that we have done that are -- we know we're are 100% -- sampling sounds funny, 100% burnt in, but the volumes are much smaller because of the target device it's going after. And we don't go into it, but most people have guessed that when you think about mobile, certain mobile devices are in hundreds of millions and others are done in millions. And so if you get 100% of the millions, it's about the same size as a few percent of the $100 million. So as soon as we get 100% of the hundreds of millions, then we'll let you know. But I'm sick and tired of saying how great that one is going to be because we continue to get forecast that seem bigger than they -- actually plays out. But we love them, they love us and we continue to execute.

Okay. Good deal. Well, it sounds exciting. So...

I'm so glad 2020 is behind us and all our employees made it through safe. We actually -- as we look back in our records, we had only one employee that had mild symptoms in the whole company, one employee that tested positive but was convinced later that, that was a false test. And we've gotten through this. So we're pretty happy about it. And I can tell you the manufacturing folks are thrilled to death not to have to wear masks anymore out on our floor. So that -- because they've been here through thick and thin the entire time. So...

And I can see that...

And I'm glad.

Yes. Me too. I'm glad you're back. So -- and even in Europe, aren't these the people in -- that work in -- I mean aren't most of them essential? I mean I don't know if they have that category in Europe or an essential employee or...

Most of them.

Right, right.

And most of our customers' fabs have continued to operate. We've heard this time and time again that what they did is they were not ramping new devices. They were just building what they had before. I mean silicon photonics is a perfect example. It turns out there absolutely were less fiber optic transceivers purchased last year, but the biggest difference is people didn't ship silicon photonics away from the standard fiber optics. So our customers had pretty bad years. And if they're not growing, they're sure as heck not buying equipment from us. So now we've seen that they're publicly strengthening, projecting things. We watch the testers. They get filled up. And so we can -- you can kind of see it's like, "Okay, they're about full. They're going to need another system," and that's why we can confidently project volumes from them this year.

And then depending on how quickly EV sales grow, they just are going to need more because once they're installing new systems, they're not going to go back to this old way. They're going to buy...

That's correct. Automotive is even more sticky than that. Once you get qualified, it's actually quite hard to change. So there's a mad land grab right now. But one of the things that we're actually hoping to see is when our customer qualifies a part, the way it works is they not only qualify us, but their customer qualifies the process. So we know for a fact that there are some of the big suppliers out there, think automotive guys, that have teams or individuals that have to sign off on a FOX-XP system for wafer level burn-in. And they have done that. What we're actually hoping is that, that rubs off, that, that same company can turn to another supplier and say, "Hey, how come you're not doing this?" And so we think we're just continuing to execute. We have a lot of domain knowledge, too. We're not just picking it up from any one customer. So we believe ourselves to be expert enough to help people with these challenges with burn-in because we've been doing it forever. I mean I have people here on my staff but I've only been here 10 years. I have people on the staff here that have been doing this for 40 years and can really understand the challenges. And we've -- I mean without being offensive, we've corrected people. They said, "We're doing it this way," and we pushed back and said that doesn't make sense. And they come back and said, "You're right." So we hope to be a partner and help them and hopefully be an advocate and they enable it to a rising tide of silicon carbide. And we think all the customers that use us are really smart. But I think if everybody used us, the whole industry would be better off because there would be more demand for silicon carbide because it would be more quality.

And the silicon -- there's plenty of silicon carbide, I guess? I mean I saw -- or some...

Yes. I think so. It's been debatable how easy it is to get it. I mean there's been a mad rush for people to get it. I feel like there's still some constraints but -- and there's still a lot of people getting into it. I think it will be more available as time goes on than it is now. How constrained it is now? I'm not sure.

Well, that one company in Cree, I think they're building a fab, another fab. I don't remember though, but I'm pretty sure they're building a fab in the U.S., aren't they?

100%, 100%.

Yes.

Yes. 100%. And they supply substrates under contract to other -- to their competitors. So that's always interesting.

Okay. Yes. And they're also making the -- I guess they're MOSFETs.

Yes.

And -- okay, okay. Well, Gayn, thanks for the information. I appreciate it. It sounds like you should have a really good year.

Thank you, Frank.

Thank you. At this time, we'll turn it back to management for closing remarks.

Okay. I -- there were a couple of things that we had somebody call in and I wanted to just make sure they kind of hear this. One was -- actually, there were some questions related -- loading our systems. And we -- sometimes on purpose, sometimes on accident, we confuse people. And I say on purpose because there's competitors listening and things like that. But our FOX-XP systems, in particular, can test up to 9 high-power wafers or 18 lower-power wafers. And sometimes, we ship them full. And sometimes, we ship them partial. That tends to be at the discretion of the customer and being able -- and to pay what their capacity is. And I want to make it absolutely clear that these products are fully released, that our lead customers in 2D/3D sensors are doing [ full mine ] blade systems up to 2,000 watts per blade. Our silicon photonics customers are also 2,000 watts per blade, 9 blade systems, and we've shipped multiple of them. They're fully loaded. They're fully operational. And our silicon photonics -- or silicon carbide customers because the lower-power wafer is actually 18 wafers and the systems we're shipping are fully loaded 18-wafer system. So software is released. They're fully functional. They have great MTBF in terms of reliability themselves. And I just want to dispel any rumors that there's any concerns related to -- if they work full because sometimes, we ship them partially populated because that's what the customer asks for. So it didn't come out. I want to get that out there somewhere. And then I'll just take this one. Ken was going to do it, too. There was some discussion related to quarterly revenues. And Christian, I think we covered some of that, too, but I just want to get it out there. There have been years -- and last year was one of them, where our forecast was heavily weighted in the back end. I hate the term hockey stick but that comes to mind because it was, well, when things recover, it's going to get better. And of course, it didn't. This year, well, it's not that way. We don't promise everything smooth, but it does feel more up and to the right as we do that, and there's no hockey stick nor is there any miracles that need to happen, a customer that isn't forecasting need to start forecasting, et cetera. So our confidence level in this is certainly as high as it's been in years. And while we may start out slower than we ended the fourth quarter, it's not going to be a hockey stick. Can we just -- somehow, I want to get it out there. We're not giving quarterly guidance, but it's important for people to know that as they're thinking about what our business is going to look like over the next 4 quarters. And stay tuned. This is going to be a very fun year for us. It's going to be exciting. We're going to be dealing with components and things like that, but it's a great place to be in. And we're excited and we appreciate everybody joining us on the call. And as always, if you have follow-on questions or want to follow up, we'll be happy to get on a one-on-one with you individually. Just reach out to our IR folks or us directly, and we'll set that up. So thank you very much, and we will look forward to talking to you next time. Bye-bye.

This concludes today's call. Thank you for your participation. You may now disconnect.