Aehr Test Systems (AEHR) 2015 Q1 法說會逐字稿

Good day, everyone, and welcome to the Aehr Test Systems first-quarter fiscal 2015 financial results call. Today's call is being recorded.

At this time, I would like to turn the conference over to Jim Byers of MKR Group. Please go ahead.

Thank you, operator. Good afternoon, and thank you for joining us today to discuss Aehr Test Systems' fiscal 2015 first-quarter financial results. By now. you should have received a copy of today's press release. If not, you may call the office of MKR Group, Investor Relations for Aehr Test, at 323-468-2300, and we will get one to you right away.

With us today from Aehr Test Systems are Gayn Erickson, President and Chief Executive Officer; and Gary Larson, Vice President of Finance and Chief Financial Officer. Management will review the Company's operating performance for the quarter before opening the call to your questions.

Before turning the call over to management, I'd like to make a few comments about forward-looking statements. We 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.

Factors that may cause results to differ materially from those in the forward-looking statements are discussed in our 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 introduce Gayn Erickson, Chief Executive Officer.

Thank you, Jim, and good afternoon to those joining us on the conference call, and also listening in online. Gary will go over our first-quarter fiscal year 2015 financial results in a few minutes, but I want to first start by discussing our overall business and product highlights. As we discussed on our last earnings call, we anticipated and experienced a softer quarter in fiscal Q1, with revenue seasonally down, similar to the year-ago quarter.

Revenue for the first quarter was $3.6 million compared to $3.8 million in the same quarter last year, and we've reported non-GAAP net loss for the first quarter of approximately $700,000, or $0.06 per share. While last year we had major expense reductions in place, this quarter our expenses were higher by approximately $600,000 as we have increased expenses in R&D, manufacturing, sales, and application resources for the development and sales activities associated with our new FOX products. In addition, unlike our first fiscal quarter of last year, we did not have any system sales with a significant amount of previously written-down material during the quarter.

We continue to build momentum for our two key new products under development that are focused on wafer level test, cycling, and burn-in. These include the FOX-1P, our next-generation, single-wafer test system that is a follow-on product to our FOX-1 test systems in production today; and the FOX-XP, our next-generation, multiple-wafer test and burn-in solution that is the follow-on product to our FOX-15 multi-wafer test system that is also in production today.

We're excited about the substantial emerging test market opportunities we validated for these products, and we believe that they will significantly expand our served available market in calendar 2015 and beyond.

As we discussed last quarter, one of the key emerging markets that we believe will be a substantial market for our new FOX-1P is for automotive devices such as microcontrollers, safety, and mission-critical components that are currently tested today as packaged parts for extended early failure rate or burn-in testing. There is a value and demand for moving these extended test from packaged part test to wafer level, and we believe our new FOX-1P solution is very well positioned to meet this demand.

Our new FOX-1P features a massively parallel test interface, which we believe will provide more test resources than any other automated test equipment system on the market. For example, the FOX-1P is capable of testing over 1,000 microcontrollers in a single touchdown, which is a significant differentiator to competitors that cannot deliver the power, number of tester resources, or handle the full wafer probe cards necessary for this application. We see this system as highly differentiated from competitive alternatives for flash wafer test, and also for embedded memories and microcontroller test, particularly in automotive applications.

The number of resources we are able to provide with our system is critical to cost-effectively moving the long test times associated with burn-in and cycling from packaged part to wafer level. We estimate the market for test systems with this capability will add more than $50 million annually to our served available market, beginning as early as calendar 2015.

We will be introducing our FOX-1P at the International Test Conference, held this year in Seattle, this month, October 21 -- or, I guess, next month. We believe that this is the ideal forum to make this introduction, as ITC has long been the key semiconductor test conference for customers worldwide that focus on key technologies and challenges in design for test, testability, and built-in self-test functions. Our new FOX-1P is very well positioned to address these test methodology challenges.

Building on our current FOX-15 multi-wafer test and burn-in solution experience, our next generation FOX-XP adds the test capabilities of our new FOX-1P to a multi-wafer test platform to allow us to serve new markets, such as the significant emerging market for flash wafer level test and cycling. We continue to be excited about the prospects for this new multi-wafer test system and are seeing increasing levels of interest in the FOX-XP, particularly for test, cycling, and burn-in applications for high-density memory devices, including those targeted for high-density solid-state drives.

With the assistance of independent semiconductor research analyst firm, VLSI Research, we've been able to validate that there is an emerging demand for a cost-effective solution to do wafer level testing and cycling of flash memories of that are destined for applications driving multiple die in a single package.

As we have discussed, there are applications such as solid-state drives, or SSDs, where there's a real push for a large number of die within a single package. These stacked die packages are needed in order to cost effectively address the demand for high-density SSDs, such as those greater than 1 terabyte.

Cost-effective, highly reliable SSDs are key catalysts to the growth of the SSD market, particularly in enterprise applications such as massive data centers. While today this market is being served primarily with sub-terabyte drives, SSDs with 4, 8, and 16 terabyte capacity are projected to become available over the next few years.

In order to ensure quality and reliability of these devices, there's not only extensive testing of the SSD drive itself, but also extended testing of the flash ICs that go into these drives to confirm their reliability.

It is just not cost effective to use today's available semiconductor ATE systems for testing these devices in wafer form for the long test times required for this application. The additional cost of testing, using test systems available today, exceeds the cost of the significant yield loss at package. We feel the yield loss of large die stacks, such as 8 die or more, in a package is prohibiting companies from supplying low-cost or ultrahigh density flash memory packages comprised of 8, 12, or 16 die stacks.

The FOX-XP will be capable of functionally testing these Flash memories at wafer form before they are assembled into multi-die stacked packages. We're able to cost effectively do this through a unique wafer level test and burn-in solution that can test up to 25 wafers per system at a time. The stacked die manufacturer can save significant manufacturing cost due to substantial yield improvements of the stacked die package. By moving the cycling/burn-in test from package level to wafer level, we expect our customers to be able to provide cost-effective stacked die of 8 die or more, and enable cost-effective, high-yielding 16 die stacks and beyond.

We see this as a key enabler to the long-term growth of the SSD market for higher-capacity drives. The market for test equipment and consumables for this emerging flash market is estimated to be $200 million to $300 million annually, starting in calendar years 2015 or 2016. And we're targeting our new FOX-XP, with our proprietary WaferPak contactors, directly at this market.

We believe there are currently no available competitive solutions in the market comparable to our FOX-XP, which delivers a multi-wafer solution of up to 25 wafers tested in parallel in a single system. We also believe we have a very low-cost solution and a roadmap for full wafer contactors for our FOX-XP [test out].

As I know we'll get questions in the Q&A section on orders and deliveries of these new systems, I'd like to cover that topic directly here. We continue to expect shipments of our first FOX-1P system within the next several months. As I mentioned, we will formally introduce this system at International Test Conference in Seattle on October 21. We've timed our annual shareholders' meeting to coincide with this introduction on October 21, here at our headquarters in Fremont, California.

In addition to our shareholder voting and Company presentation, we will be doing a facility tour where shareholders can see our new FOX-1P system in a production configuration. We will also be showing our manufacturing facility, with our ABTS-P systems on the floor, as well as highlight some facility improvements we're making in preparation for the new FOX products rollout.

For our FOX-XP multi-wafer test, cycling, and burn-in system, we're actively engaged with multiple customers for production test, cycling, and burn-in of flash devices. We currently believe that we will receive an order/commitment from one or more of these customers within the next few months.

We are expecting the first shipments of the FOX-XP in the second half of calendar 2015. I'm personally very excited about this level of customer interest, and the opportunities that Aehr Test has for these new products. The new FOX products really have the opportunity to set a new standard for how production, reliability, cycling, and burn-in testing can be done for high-volume manufacturing for semiconductor devices.

Having said that, we anticipate that we could see as many as one or two additional soft quarters as we roll out these new FOX products. However, we do see our installed base for both our current ABTS packaged part and FOX wafer-level test systems at or near maximum capacity. This is a positive sign, as increased capacity needs could drive upgrades in new system orders and revenue.

We believe having material and systems available with short lead times is prudent to be able to meet short-term demand from our customers, and to allow us to ship systems quickly against purchase orders received in the same quarter.

Gary, our CFO, will now review the quarter's numbers before we open up the lines for questions. But I'd like to conclude by saying we continue to be very optimistic about Aehr Test's ability to grow our revenue and profitability from our packaged part test business with our ABTS platform, and particularly as we add the significant new market opportunities available with the introduction of our new FOX-1P and FOX-XP wafer level test products.

Gary?

Thank you, Gayn. As we reported in today's press release, our net sales in the first quarter of fiscal 2015 were approximately $3.6 million, down 34% sequentially from $5.4 million in the fourth quarter of fiscal 2014, and down 5% year-over-year from $3.8 million in the first quarter of the previous year.

Non-GAAP net loss for the first quarter of fiscal 2015 was $710,000 or $0.06 per diluted share, compared to non-GAAP net income of $443,000 or $0.04 per diluted share in the preceding quarter, and a non-GAAP net loss of $23,000 or $0.00 per diluted share in the first quarter of the previous year.

On a GAAP basis, net loss for the first quarter was $907,000 or $0.08 per diluted share. This compares to GAAP net income of $239,000 or $0.02 per diluted share in the prior quarter, and a GAAP net loss of $166,000 or $0.02 per diluted share in the first quarter of the previous year.

Gross profit in the first quarter was $1.6 million or 45% of net sales. This compares to gross profit of $2.9 million or 54% of net sales in the preceding fourth quarter, and gross profit of $1.9 million or 52% of net sales in the first quarter of the previous year. Gross margin in this latest quarter was lower than the preceding quarter because labor and overhead spending increased while the net sales amount was lower. The prior-year first quarter included shipments that incorporated more previously written-down material than this latest quarter.

Operating expenses of $2.6 million for the quarter were down slightly compared to $2.7 million in the previous quarter, and up from $2.1 million in the first quarter of the previous year. SG&A was $1.6 million compared to $1.7 million in the preceding fourth quarter, and $1.4 million in the prior-year quarter. Spending increased from the prior year quarter due to restoration of previously reduced pay, and sales activities associated with our new FOX products.

R&D expenses were approximately $1.0 million for the quarter, about flat with approximately $1.0 million in the prior quarter, and up from approximately $680,000 in the prior-year quarter. Spending increased from the prior-year quarter due to development activities related to our new FOX products, and also to restoration of previously reduced pay. As stated previously, R&D spending can fluctuate from quarter to quarter, depending on the development of our new products.

Turning to the balance sheet and changes during the first quarter, our cash and cash equivalents were $2.1 million at August 31, 2014, up from $1.8 million at the end of the preceding fourth quarter. Net cash -- or cash, less the line of credit -- of $1.1 million was up from $1.0 million at the prior quarter-end. Accounts receivable at quarter-end was $2.1 million, down from $3.4 million for the preceding quarter. Our inventories increased sequentially by approximately $0.2 million. Accounts payable decreased by $0.3 million from the prior quarter-end.

This concludes our prepared remarks, and we're now ready to take your questions.

Operator, please go ahead.

(Operator Instructions). Colin Denman, Westerly Capital.

So, first of all, I wanted to talk about -- in the press release, you said that you had orders for engineering and production versions of the FOX-1P. And I was wondering, can you say how many customers you currently have orders from?

Hey, Colin. Thanks for asking. We've been deliberately not talking about numbers or quantities at this time, nor have we even given details around ASPs; a lot of that having to do with not wanting to identify who are our lead customers are, and on signal pricing. So we have not.

And then when I think about the commentary, or the comments you made around potential for one or two more soft quarters before these new products really start to ramp, how should I compare that with the comments that shipments of the FOX-1P could start in the next several months? Should I be thinking your commentary around softness, meaning this mid-$3 million level of the base business? Or how would the initial shipments of FOX-1P layer in there?

Sure, Colin, let me take a stab at that. So, what I want to try and balance here is the, if you will, not necessarily lack of visibility, but the uncertainty of what the outcomes could be. And I think it's a conservative stance to try and set expectations that we could see a couple more soft quarters. At the same time, I'm sitting on a significant amount of inventory of both full systems and material that's available for short lead times. I think we even talked about this last quarter, where there's a double-edged sword with that.

On the good side, we have the ability to ship very quickly against orders. On the downside, you don't have the visibility if you said you had a 6- to 12-month -- or 6- or 9-month lead time. So I have significant opportunity to ship within -- have terms business within the quarter.

And so the reality is is there is still a lot of dynamic range within what our quarters could look like, just from what I would call base business. This is currently installed base customers of ABTS and FOX products and WaferPak. Add to that the timing of not only the FOX shipments, but revenue recognition of the lead customers; as you knew, we do have rev rec requirements with respect to acceptance of the first systems. And then as a policy, our follow-on systems basically have revenue recognition as they ship. And so we're trying to get our arms exactly around that.

At the same time, we're trying not to -- I don't want to say, we don't want to focus on that. But we want to make sure that we're, A, focusing on our current ABTS customers. We have really important customers with some really important projects. They are all, right now, at maximum capacity, both the engineering and production accounts, and we want to be able to serve their needs.

But also with respect to our FOX products, a big or primary focus of the R&D organization in the whole Company is trying to accelerate and get these new FOX products out. Because we have customers for both -- we have customer requests to pull in as soon as we can, both FOX-1s and XPs at this time.

Right, okay. So, with something like a FOX-1P, or just any of your products in general, is there any way for us to think about an average amount of time it takes to recognize revenue on an initial shipment? Is this like a few-month-long process for --?

Yes, I think that's a -- we've got detailed schedules with customers about how long it takes for them to get it first in, for them to be able to roll it into production. And a pretty typical number is within, say, three months.

Okay. And then the FOX-XP sounds like really exciting product, with the big TAM. I'm just curious on the order/commitment that you expect to see in the next few months, I'm assuming that something you'll probably able to press release. Is that correct?

Yes, that would be our expectation. I think that would be considered material by everybody's opinion.

Yes, of course. And then maybe you can just help me understand a little bit what the dynamics are, in terms of the sales cycle of selling a FOX-XP to a lead customer. I know that you've probably been in conversations with them previously, and then next up is the order and commitment. And it sounds like you're setting expectations for first shipments not until second half of 2015. So, talk about maybe that nine-month window between now and when you expect first shipments. What are things that could either pull that forward or push that out?

And then is this, just to level-set expectations, should we be expecting very first small-volume engineering orders to start shipping in the first half of 2015? Or is it something where they're already evaluating the product, and you could go right to more of the production-type order?

All right, I'm actually going take a moment for a second, Colin, to make sure I get this. I'm writing down bullets of the sales cycle -- why or what about nine months; and then what maybe ramp looks like. Is that okay? Did I oversimplify that?

No, that's fine.

And you're going to cover everybody else's questions, you know, so it will seem lonely after this. Okay, let me try on the sales cycle. One thing of note is that today with our ABTS products, we have a mixture of production and engineering customers. In engineering -- or in production, which is simplest, they will test their devices for so much time in, say, production burn-in, for so many hours, and then they will ship it and it shows up in your automobile in your dashboard.

In engineering, we'll do applications where we're proving that the device will last for 1,000 hours. And when that device lasts 1,000 hours, it's pretty much kaput. It's proven it can last, but you don't ship it to a customer at that point. In the engineering applications, customers can have a few of these, and every customer has them. In production, it scales with volume per capacity. And so an engineering customer might have a few, and my production customers might have tens or hundreds of systems.

In the FOX products, all of our FOX-1 and FOX-XP customers -- FOX-1P and XP customers -- they are all production accounts. What I want to talk about that is, a production sale in the semiconductor ATE space is a longer sale process. It is very typical for it to be a minimum of six months, and can sometimes last over a year or more in the sales cycle.

And so, for example, in the -- I guess I got to be careful of the numbers -- in the multiple customers I'm engaged with with the XP today, each of those we've been engaged with now for at least a year. And so the sales cycle feels about right. We would expect them to be at a stage of being in decision-making, and doing ROIs, et cetera, at this point.

Now, what is the schedule, with respect to nine months? As we've said before, we had already -- the FOX-XP core technology of the electronics and software is directly leveraged from our FOX-1P. The FOX-1P has our first customers, our first orders, and we'll ship it first. And then we said we'd probably take about 2 to 3 quarters before the XP would come out. All of those schedules are firming up now. So that's really very consistent with what we've said all along. It just seems like, okay, second half of next year, at nine months to a year away, seems like a long time. (multiple speakers) this timeline.

So, a customer that would make a commitment, we would be making whatever slight modifications are needed specifically for their needs, and be able to ship them basically a standard FOX-XP product in that timeline.

Now, what does a ramp look like? In my experience in semiconductor production ATE test, is that the customers along that time were building up a first production system will be developing engineering, prototypes; will be doing test programs; will be doing a lot of things to -- in parallel with the development of the final production system.

Then we need to prove out the production system. So they would take one whole test cell, release it into production, do correlation, and ship products off of it. Then it can be slow or a step function. In my experience at HP, Agilent, and Verigy, I've seen customers see the first production system and buy 40 systems immediately; and I've seen them do one at a time, one at a time, on a quarterly basis.

These systems, particularly the XP, have a significant amount of capacity out of them, so we weren't expecting 40 system orders. But we're ensuring that we have the manufacturing capacity here to be able to ship multiple systems to a customer per quarter, if needed. And as I alluded to, we're doing some changes to our facility here to be able to facilitate that, as needed.

I really appreciate that. I'll let others ask questions. Thanks a lot.

(Operator Instructions). Tom Diffely, D.A. Davidson.

I guess Colin left me a few questions. (laughter) So first for Gary, what was the cash burn during the quarter? And what are the factors you're looking at to manage your cash over the next couple of quarters when business remains soft? And I'm curious what type of inventory build and for the new systems you might have to add.

As far as cash burn, we actually generated cash. Our net cash went from $1 million in the prior quarter to $1.1 million this quarter, so that was favorable for us. As far as cash management, you know we always run a tight ship here, but we do have a number of opportunities to get cash when we need it. We've got a $2.5 million line with Silicon Valley Bank. As Gayn mentioned, we've got some inventory, both systems and major subassemblies, that are available. So that's if a customer comes in and orders quickly, we would both be able to ship quickly and obviously collect on that quickly.

The new program that Gayn put into place when he joined, of requesting down payments for our customers, has been successful. So we get significant down payments along with the order. And also if push came to shove, and we were threatened with large orders, we've already had some discussions with our bank. We believe it would be possible to increase our line of credit as necessary, to support any significant orders that were coming in. So we do see a reasonable comfort level, even though we don't have all the cash reserve that we would like to at any point in time.

Okay, that sounds good. So what drove the cash generation in the quarter? Was it collections?

Yes, it was the receivables going down.

Okay. All right. And then, Gayn, looking at your two future markets here, the $50 million and the $200 million to $300 million markets, are those the size of the markets when your tools are in them? Or is that the size of the markets you are replacing with your tools? Because I assume that you're a little more effective than your competitors, if there really are competitors, and that you really bring down the size of the market.

Well, those are really good questions. And I know that people say that a lot. It's a little -- they are different. And we actually -- we talk about three different new, emerging markets, although today I only talked about two of them. If I get a chance, I will talk to the third, as well. But let's talk about the two -- the $50 million and the $200 million to $300 million.

The $50 million one today we believe is customers are spending that money today in wafer sort and/or packaged part burn-in on an annual basis. But they do that in two steps: they buy ATE systems for wafer-level microcontroller for automotive, and they buy it in packaged part burn-in systems. We believe those dollars can move to wafer level test and burn-in.

And so the dollars -- it's not necessarily lower dollar amount. It is just a shift to the front end, which has several quality and process step minimization steps. So from a TAM perspective, people are already spending that money; they would just hopefully be spending it with us for on tools like this. And I say that broadly, because right now we believe we're unique in being able to provide a solution that can do that type of parallelism.

So to the extent that when we introduce the FOX-1P, people are able to do that, that market is emerging with the entrance of the FOX-1P.

On the $200 million to $300 million range -- and this is where, in particular, we got VLSI Research involved to make sure that they could help us with the customer feedback and turning that into real numbers, because those are big numbers -- is the customers today are doing packaged part burn-in. And as they move from packaged part burn-in to wafer level burn-in, it creates this new opportunity.

Now, we sell packaged part burn-in systems; we think we make a great one. And if you'd like to leave it at packaged part, we'll be happy to sell you a bunch of ABTSes. However, in fact, as we were talking to one of the many customers about our packaged part burn-in system, we uncovered this opportunity that if you move that testing to wafer level, you get all this yield back, and the yield dollars dwarf the size of the tester spend.

And so, in reality, to do this, you spend more dollars in test. It is more expensive to test these devices at wafer level than at packaged part. We had that conversation. But it is significantly cheaper manufacturing cost, because of the yield improvements. And that's the big ah-ha. And so, if someone is spending $200 million to $300 million, or the market were to spend that, the yield improvements on the multi-billion-dollar SSD market is in excess of that. So it's -- you pay for these tools through the yield improvements.

I just want to add one more thing. As soon as we talked to our customers about this, they also asked us to extend our capability to do more testing. In that case, we're also significantly cheaper testing than traditionally available ATE platforms. So, between the two of those, we can save the customers directly in what their budget would have been for ATE spend, and we can save them in yield.

Okay. So when you look at the SSD market or other flash markets, are you still going to have to do a final test in the package, or a burn-in in the package, even though you have a higher confidence that the chips inside are better?

The data that we've seen and the correlation efforts would imply no. And the reason is in the burn-in -- and burn-in is overstated. These are actually cycled, and that's a subtle difference. The parts themselves -- when you cycle them so many numbers of times, and we don't want to talk about that, okay? -- they will fail, and they are failed in a way that cannot be repaired. What's failing is inside the core. It is nothing to do with the packaging step, the assembly, or the processing.

There are cases where there's an argument you may want to burn-in a part because you may be testing the bonding of the packaging. But that's less and less or negligible at all, and certainly not in these temperature ranges of these devices.

Okay. That makes sense. And then when you look at the different products here, the FOX-1P engineering versus production versus the XP, how much commonality is there in the actual tools themselves, or in the parts of the tools?

We have these great slides that talk about our leverage between all of our products. But specifically on that one, we introduced in the last two years this ABTS-P. We have that shipping today in production, as we talked about. If you'd come and ever visit us, we'll always have systems out there on the floor, and production on it. That ABTS-P architecture uses subcomponents that we've extended over into our new FOX products. That's why that carryover of the FOX-P, for pin.

So we already have the leverage there. Interestingly, our ABTS and our FOX products share the same software. In fact, it is sort of subtle. Our ABTS systems will test maybe 36 or 72 burn-in boards. And we get to 36 burn-in boards completely asynchronous from each other. And when you test the wafers, you can test them completely asynchronous from each other. They can be running all different test programs.

So our FOX-XP looks like an ABTS system. It uses our ABTS chamber. You walk up to it and the door is closed, you'd have to open it to tell that it's the difference between them. We have mechanical parts; we share the electrical parts between the FOX-1 and the FOX-XP. And the FOX-EWS, as we call it, the engineering system, is the same between the FOX-1P and the FOX-XP. So, this explains how we're able to leverage from arguably a very small company such an enormous product line and portfolio that can address a lot of different markets.

Okay, that sounds good. And then finally, when you look at the -- obviously you don't want to talk about the sales, the revenues, the cost of these tools, the price of the tools. But from a margin point of view, would you expect the margins to be above corporate averages, in general, for the new products?

Yes, we will.

Okay, good.

They are higher than our corporate averages.

Okay.

And the ASPs of our FOX-1Ps and our FOX-XPs are higher, to significantly higher, than our ABTS products.

Okay. Thank you.

Dominik Schmidt, QVT Financial.

Thanks for taking the question. It follows up on the question that was just asked. You had mentioned that you were doing both burn-in and test on these new devices, or they are capable of both. Do you see them being able to charge a premium on test, versus burn-in? Or is it the same in both usages?

In fact, that's a good question. Thanks for the question, Dominik. So, let's see here, the easiest way. So, I don't know that everyone on the phone would know, necessarily, the difference between test and burn-in. But a simple way of thinking about it is when you test a part, you are testing it good from bad, or how good it is. And that might take seconds or minutes, or tens of minutes, in typical memory applications.

But in the cycling application, in order to cycle these devices many, many times, typical test times might be from three hours to 30 hours. They are 2 to 3 orders of magnitude longer. And so, in our -- when we designed the FOX-XP, for example, we designed it to be very cost-effective and, in fact, to ensure it's much cheaper than the yield loss or gains that it would provide.

And so our FOX-XP is more in the lines of the price or the cost of what is a burn-in system like our ABTS, which is substantially lower than a typical test system. So, we go in and we show a customer that we're able to do their wafer level burn-in almost as cheaply as packaged part burn-in, but they get all this yield improvement. But those burn-in systems are much cheaper than the ATE systems that are being used today for wafer-level sort.

So, arguably, they are much more valuable if they can test. And therein, not surprisingly, each of our customers has said, hey, while you are doing the burn-in, can you add a few more test functions? And some of those functions, in fact, have driven up our ASPs as we've added more capabilities and features, and it's been part of our defining of it. And we will have options to the XP that will add -- have pricing options for different features that are harder for us to build, or add hardware or software features. So, our ASPs in test, if it were used only as test, would definitely be higher for wafer than just for traditional burn-in.

So, it sounds like an upside on the margin side, if customers really get this usage, that sounds like a rather large upside to the margin.

It could be. We're trying to focus on ensuring that we can make a solid profit, and generate the R&D, and everything that we need to do. And I think one of the reasons we have customers that are very interested in this product.

Thanks, Gayn. That's good.

(Operator Instructions). Geoffrey Scott, Scott Asset Management.

Quick question on 2015 total. Last year, you said you'd be profitable for 2014. Any comments for the total of 2015?

Let me try that here, Geoff. I want to make it harder on me. Last time, I had talked about 2015, our belief that we could, I think, grow or maintain our revenue and break even, as well. We're looking at that again right now. And I think the over/under continues to be exactly the shipment of the new products, and how these next couple of quarters shape up. We're still very optimistic. We're focusing on the right things. We'll have the manufacturing capacity, and are just watching our customer orders and shipments of this.

One of the keys ends up being, in any of these scenarios, is we do believe that as we ship the -- start seeing volume productions of the FOX-1s and particularly the XPs, it has a bit of a step function to it. So, we're still very optimistic. I'm not necessarily reiterating guidance, but I also just -- I think we're focused on the right thing to get these products out.

Okay, fair enough. The first two shipments of the FOX-1P will go to the development customer, correct?

I will say that our first -- right now, our plans are first engineering and our first production system will go to the same customer. And since they are usually listening, they would be really mad if I said anything else.

Okay. My real question is, how broad is the market for the 1P? If you take out your development customer, is it your expectation that you will sell 1Ps over the next couple of years to an additional three customers, six customers, 12 customers? Just how broad is that market?

I think we're still getting our arms around that. But it doesn't feel like it's a 10-customer type of number though, Geoff. I think that tends to be in the few, to half a dozen, or maybe a little more, but with much deeper density, if you will. No customer would take this system if they weren't going to buy 10 [test test], or more. So, each of those customers can be a significant number to us. So, we don't need a lot of customers to make this worthwhile.

The reality of the semiconductor test business is, with the consolidation in the industry, there are fewer and fewer companies that are out there. And if you can go penetrate 10 customers with a production ATE system, you are doing pretty darn well.

Okay. I want to try to parse one of your -- one of Colin's questions, and one of your answers to him. In terms of the shipments of the XP being basically a year out, I gather from that, that that -- there's nothing you really could do to accelerate that from 12 months down into nine months or less. Is that a fair statement?

Did some of my customers put you up to that, Geoff? All right, let me -- I say that sort of tongue-in-cheek. The reality is, there is some that we can do. Our focus is really -- because of the leverage -- is getting the FOX-1 key material that's out. There was a part of this, which is the customer commitments, that has a front-end aspect of it. Each of the customers we've talked to about the XP today has some slight variations, which is not atypical in any semiconductor test space, that range from the applications to some small software enhancements to maybe some small tweaks to the product line hardware.

All of these are fine with us, and within the platform, and do not create challenges for us from a manufacturing or support perspective, but they tend to add to the development cycle. So, right now, as we line up those first customers and there's a little bit of game of chicken of who buys first, has partly to do with the schedule.

For example, I will say if we focus simply on pure wafer level burn-in, each of the customers is simpler. As they pull us towards more and more test, that would add some time to the schedule.

And I'm quite aware with each of the customers that they are aware of that, and they are having those conversations. So, message to customers: if you could focus on the wafer level burn-in sooner, we could get the product out faster.

Okay. The SSD potential customers, are they currently evaluating the technology, or are they getting into the -- evaluating the economics of it?

Yes, I would say a little of both, and it depends on the customer. There are some that are purely on the economic side of it, and some that are doing both. In some cases, the economics can be done on a napkin.

I'll give you one example on the moving packaged part to wafer level. And I think people on the phone can understand, that is not being done today because it's too expensive. Now, having said that, you can do exactly what we're doing with our new FOX-XP in production today, one wafer at a time.

So, we have customers that have done the experiment. They've taken a wafer; they test it for 20 hours on a production ATE system; and then they can correlate that to the yield improvement at the back end. However, a production ATE system for -- might cost $100 an hour at, say, a contract manufacturing typical hourly rate. So 20 hours adds $2000 worth of test cost to a wafer that the industry standard of a 300-millimeter NAND wafer is about $1600.

So, you can't get enough yield improvement to make that makes sense. We are substantially lower than that. Think more than an order of magnitude. So we are able to bring it down to be more like the cost of a normal test insertion, even for a 10-hour cycling. So, when they're faced with the knowledge that we know this is a good thing for yield, and then we give them a quote, it's more of, okay, can you do this, and how fast, and let's talk. So, it's a pretty exciting time for us here.

How broad is that XP market? If the FOX-1 is half a dozen, how broad is the XP?

Probably similar.

Similar?

Although you could argue a few customers, or even one, could be substantial.

Yes. Last question. In terms of the decision-making at your customer, you explained eloquently before that the test was more expensive, but the total manufacturing cost was lower because the yield. Is the decision to purchase being made by the test person, where it's only the test cost? Or are you able to get to the person who has the total manufacturing cost (multiple speakers)?

Yes, I think it's a balance of both. Upper management can convince themselves that there's going to be straight yield improvement, but the test guys have to make it work. So it is a strategic sale that involves multiple layers of management.

And you think you have that relationship depth to get that decision?

I think in some customers more than others, and some of that is it takes time.

Okay. I'll pass it on to somebody else. Thank you.

(Operator Instructions).

See, I told you Colin had answered all the questions -- asked all the questions, I should say.

Marty Cawthon, ChipChat.

What I'm interested to hear is -- well, it's, as I recall, about 8 to 10 years since the original FOX full wafer tester was introduced by Aehr Test. And now we're introducing, or you're introducing, the FOX-1P and the FOX-XP. And I'm interested in two things: one is a comparison and contrast with the FOX-1P and XP today with the original FOX and FOX-15 that were introduced roughly 8 to 10 years ago; and also with the environment in which you are selling these testers today, relative to back then.

And what motivates my question is, when I first learned of the FOX-1 full wafer tester, I thought it would make a big change and a very dramatic effect on the testing world and also sales for Aehr Test. And for a whole bunch of reasons, that didn't have the spectacular success that I initially expected.

And so I'm interested to hear on how the product has changed from that time, and also how the testing environment has changed from 8 to 10 years ago. And I realize, Gayn, you are new to the Company since that original introduction, but I'd like to hear your thoughts anyway.

Well, as you said, this is [a quote on my at like this place]. I'm 2 1/2 years into the Company, but I've spent the last 20 years in this industry. And most of those we were at places like ITC; and between HP, Agilent, and Verigy we would be looking across the hall at Aehr Test. Aehr Test has been around a very long time, as well. So I've been in this entire space the entire time. And I certainly remember intimately what's gone on in the industry over this last, as you said, eight years or so.

And we've studied -- we're very clear what is different between the new products. So, we had a FOX-1, and now it's a 1P. And then we'll also talk a little bit about the FOX-15 and now XP. The key differences that I think span across both of those products and the changes is the FOX-1 and the FOX-15 were architecturally similar in terms of test electronics; in some ways, how they are today between the 1P and the XP.

Those products were leveraged from our burn-in platforms, and added some very unique capabilities in terms of technology of contactors, power supplies, software, to be able to enable of couple of key markets for us.

However, I think as we look back on it, in particular with respect to the new markets we're looking at, the big change is the new XP and 1P have significantly more tester functionality than we had in the older products.

These are test systems with much more sophisticated, if you will, pattern generation, pin electronics, I/O selectability, clocks, PMUs, and other functions that make it a more general-purpose test system. And we understood that with the FOX-1 and the FOX-15 when they came out. And we started developing a new system that is much more capable.

So, an example of the FOX-15, which can test 15 wafers in parallel -- functionally, that is not capable of testing those flash wafers. It doesn't have enough resources or capability. So that particular system, by going to the XP, we've opened up a market for it. So, mostly the FOX-1P and XP are more general-purpose, look more like an ATE system; and, as such, have more flexibility to go after a wider range of devices. Number one.

And the other one is -- I think you said the environment. And I'm going to put that in two categories. One is when Aehr Test introduced the FOX-1 and started to ramp, which was in 2007 and 2008, I was sitting on the other side. I was on the ATE side. And I watched the impact of what Aehr Test technology was able to do by introducing a system that was capable of doing full wafer test when our production ATE systems took 6 or 10 touchdowns, or 10 times as long to test the same thing. It was devastating to that market at the time. There was a significant value to the customers. And we were also assuming that this was going to take off substantially.

But now you have to remember where we were in 2008. The world kind of imploded. And there was significant contraction within the semiconductor suppliers, within the semiconductor capital equipment suppliers; and the greatest -- Great Recession of all time had an enormous impact. I think Aehr had the wisdom that, during that same downturn, we started to retool to build these new products like the ABTS-P. And then I was able to, as I came, help to focus the Company, taking that technology to build these more general-purpose testers, with the XP and the 1P.

This is a much better environment where our customers are healthier, our customers are at full capacity. And in 2008, 2009, our own customers were sitting with 30%, 40% excess tools, and it was pretty obvious it was going to be a long haul until we could get through that. Across the board, all my customers are full, and they are a lot healthier than they were before. And so that's encouraging.

I'll do one more. The other part of the environment is we've got these new opportunities that we see with SSDs that didn't even -- we weren't even thinking about those back then, where flash memory and reliability is important -- that there's new capabilities like 3D memories, and TSBs, and technology that are driving the need for reliability or a need for what is traditional burn-in to move to wafer sort that create opportunities for Aehr Test, not just immediately but over the foreseeable future.

And this is the first call I think I've not talked about automotive. And we -- a lot of what drives our business and our customers is the automotive semiconductor devices, and what we're seeing with the dashboards and the infotainment hubs, and how that's changing the semiconductor market, which was consumer-based, to need the reliability associated with automotive. And so this is just a -- we're in a good place right now in the environment, both our customers and the overall industry.

Okay. Thank you very much.

And with no further questions in the queue, I'd like to turn the call back over to management for any additional or closing remarks.

All right. Well, thank you, and I just appreciate everybody for joining in. We covered a lot of ground here today after about a 10-, 15-minute intro, so a lot of Q&A. I appreciate it. As always, we invite you to call in. We'll be happy to take your call or have you visit. We do our shareholders' meeting on the 21st. If you have a chance to come in to the Silicon Valley Bay Area, we are located nearby in Fremont, and we're happy to have you here in attendance. Or if you get a chance, come see us at ITC up in Seattle. Thank you very much. Bye-bye.

This does conclude today's conference. We thank you for your participation.