Nova Ltd (NVMI) 2013 Q1 法說會逐字稿

Good day, and welcome to today's Nova Measuring Instruments first quarter 2013 results conference call.

For your information, today's call is being recorded.

And at this time, I'd like to turn the call over to your host today Mr. Kenny Green, CCG Investor Relations.

Thank you, operator, and good afternoon to everyone.

I would like to welcome all of you to Nova Measuring Instruments first quarter 2013 results conference call and presentation.

And I'd like to thank management for hosting this call.

With us on the line today are Mr. Gabi Seligsohn, President and CEO, and Mr. Dror David, CFO.

I'd like to draw your attention to the presentation that accompanies today's call.

The presentation can be accessed and downloaded from a link on Nova's Website at www.novameasuring.com.

Before we begin, may I remind our listeners that certain information provided on this call may contain forward-looking statements?

And the Safe Harbor statement outlined in today's earnings release also pertains to this call.

If you have not received a copy of the release, please view it in the Investor Relations or News section of the Company's Website.

Gabi will begin the call with a business update, followed by Dror with an overview of the financials.

We'll then follow with a question-and-answer session.

I will now hand over the call to Mr. Gabi Seligsohn, Nova's President and CEO.

Gabi, go ahead, please.

Thank you, Kenny.

And hello, everyone, and welcome to our first quarter of 2013 earnings conference call.

I will begin today's call by describing the main aspects of our performance during the first quarter of 2013.

I will then provide some color on the current state of the industry as it relates to us and, finally, state our guidance for the second quarter of 2013.

As evidenced in today's published results, we are clearly off to a very good start for the year.

During the first quarter, we exceeded the high end of our revenue guidance with $27.4 million in revenues and a non-GAAP net income of $3.6 million or $0.13 per diluted share.

As I indicated in our previous quarter's conference call, gross margin during the first quarter returned to our normal levels of 52% to 55% with actual results coming in at 54%.

The 28-nanometer ramp up in foundries, which we have been discussing for quite some time, is currently in full swing.

And our service and operations teams have been busy with manufacturing and installing a very large fleet of both standalone and integrated metrology tools.

In light of the delayed ramp up during the second half of last year, customers have been working at a faster pace to bring installed tools into production.

As this ramp up continues, in our largest account, we see two other very positive trends.

The first which relates to significant 28-nanometer demand, has led our other foundry customers to initialize a significant 28-nanometer ramp up of their own.

Although these projects are smaller in size than the current one, they are still meaningful.

It is our understanding that these customers have by now crossed a relevant yield threshold, allowing them to release budgets and move forward.

Orders for these ramp ups are starting to come in.

And we expect most of the order and delivery action to take place during the second half of the year, thus extending the 28-nanometer rollout.

The second very significant trend is the 20-nanometer ramp up in foundries.

Our target customer has been gearing up for pilot production in a current existing fab as well as in a new fab.

This level of pilot production capacity should carry the customer into the second half of this year.

Based on the plans that we have received, significant orders are expected sometime during the second half with most of the 20-nanometer ramp up planned for 2014.

At the same time, another one of our largest accounts is moving forward with a 20-nanometer ramp up as well.

We have already supplied multiple tools to support this plan and, again, expect the pace of that ramp up to increase during the second half of this year and into next year as well.

The acceleration of the 20-nanometer ramp up plans also requires a similar acceleration in our software development efforts to deliver extended capabilities in conjunction with the tools we are planning to deliver.

As a strategic supplier to leading foundries, our customers are expecting us to accommodate their plans by speeding up roadmap deliverables in order to keep pace with their requirements over the next 12 months.

Overall, in the foundry sector, there continues to be a positive atmosphere.

Listening to recent announcements by Samsung, ARM, Qualcomm, and Broadcom, we understand where this vibe is coming from.

We were very pleased to hear recent announcements by two of our largest customers, TSMC and GlobalFoundries, that they plan to increase CapEx further beyond their previously announced plans.

We see this as clear indication that the current ramp ups still have several quarters of activities ahead of them.

And given our strong position with all leading foundries, we are confident in our ability to reap the benefits.

On the memory front, we also see some positive indications.

The unprecedented decision of Samsung to approach archrival Hynix for DRAM capacity paints an interesting picture.

We believe it relates to supply insufficiency on the one hand as well as a corporate focus of Samsung on NAND and logic on the other hand.

DRAM prices, especially mobile DRAM, have held up well, and supply has been limited at the high end, which means that capacity investments will soon need to take place.

While the Micron-Elpida deal is being delayed further, we believe it will ultimately be completed and result in a ramp up of 2X-nanometer capacity in Japan and Taiwan as well.

And again, we are very well positioned for that activity as well.

Now, let me provide some more color on activities during the first quarter.

As we have noted several times in the past, collaboration with leading-edge customers on novel technologies is important to enable us to further solidify our market position.

During the first quarter, we were happy to take part in the industry's leading technology conference SPIE.

This year's conference was especially significant as we published several papers with two leading customers, Toshiba and GlobalFoundries.

The publication with Toshiba was focused on using optical CD to control nano imprint lithography for a manufacturing technique called direct self assembly, which is being evaluated by Toshiba for future generations of NAND flash memory.

After several years of repeated efforts to collaborate with Toshiba, we were very pleased and honored to be included in a joint effort looking far into future manufacturing techniques by such an advanced customer.

We believe it is testament to the reputation our technology and scientists have built in recent years.

The paper with GlobalFoundries was the first demonstration of the use of hybrid metrology in production.

Hybrid metrology is a concept that we have been developing for the last few years, whereby information from various metrology tools is fed into our MARS modeling station and thereby improving metrology sensitivity to very subtle process artifacts and changes.

GlobalFoundries shares our view that combining data from multiple sources, each of which is sensitive to different process elements, will become an enabler for process control of future generations of semiconductors.

We have developed a software package, which we plan to further develop as part of our strategic software development effort.

We believe this product will both strengthen our position in the OCD market as well as offer increased revenue opportunities in the future.

During the quarter, we were all -- we also recognized revenues for the second out of four installations of our V2600 product for 3D interconnects.

We expect the third and fourth tool to be recognized during the second quarter and third quarters of 2013.

Usage of all of the installed tools is increasing, and they are now part of the manufacturing flow.

Once TSV-based wafer capacity output increases at these customer sites, we expect they will require additional tools.

We are also continuing several offline evaluations of the tool with other leading memory and foundry customers and expect to ship more evaluation tools to these customers pending successful demo results later this year.

Turning to the current overall environment, let me now provide some color as to how we see the year progressing.

We expect orders from foundry to reduce somewhat in the very near term and strengthen again during the second half of the year with overall foundry spending continuing well into 2014.

During the second half of the year, we expect a significant ramp up in foundries of 28-nanometer capacity by two of our customers as well as a significant 20-nanometer ramp up by two customers.

The additional 28-nanometer capacity will come from existing fabs, while most of the 20-nanometer capacity increase will come from new fabs.

Recent discussions with our leading customers indicate that spending will continue well into 2014 to support an expected ramp up of 20-nanometer capacity.

In fact, we expect most of the 20-nanometer ramp up to take place during 2014.

On the memory front, we are seeing signs of possible pickup in demand for some near-term specific reasons, yet it is difficult to anticipate when a broad sustained pickup in memory demand will occur.

Capacity insufficiency of high-end mobile DRAM is creating some order traction in the near term to increase capacity in an existing fab at one of our customer sites.

One of our other customers is gearing up to modify an existing fab to the 2X-nanometer technology node.

This fab modification will require incremental additions of tools.

On the NAND flash side, we expect a new fab buildout to start taking equipment during the second half of the year.

Another existing NAND fab is expected to require some incremental metrology equipment as well.

The much-expected industry transition to VNAND or 3D NAND will take at least two to three more years until it is ready for high-volume manufacturing.

In the meantime, customers will be making serious efforts to innovate methods to improve device capabilities, and in doing so will perform design rule shrinks, rely more heavily on double and quadruple patterning, requiring more etch and litho steps, both of which will rely strongly on more OCD measurements.

Although these projects are not large in size for the next few quarters, we see these developments on both the DRAM and NAND flash sides as positive after a lengthy period of almost two years with very minimal investments.

In the last few years, we were able to show exceptional business results, which were based mainly on the foundry ramp up.

Hence, we believe that a possible return of the memory segment to a sustained investment cycle will further fuel our growth.

As the year progresses, we will continue to add color for these projects.

Now, let me turn to our guidance.

For the second quarter of 2013, we expect revenues of $26 million to $28.5 million with GAAP diluted earnings per share of $0.03 to $0.07.

On a non-GAAP basis, which excludes adjustments of deferred income tax assets and stock-based compensation expenses, we expect diluted earnings per share of $0.06 to $0.11 for the second quarter of 2013.

And with that, operator, let me now turn it over to Dror for a closer view on the numbers.

Dror?

Thanks, Gabi.

Good afternoon to everyone, and welcome to Nova's quarterly conference call.

Total revenues in the quarter were $27.4 million, slightly higher than the upper end of the first quarter guidance, representing 24% increase quarter over quarter.

All of the increase was in the integrated metrology front, which is usually linked to capacity expansions.

Service revenues remain stable at the $5 million level.

Product bookings distribution in the quarter was approximately 80% from the foundry segment and approximately 20% from the memory segment.

On a regional basis, approximately 85% of the bookings in the first quarter came from Asia-Pacific and the rest from US and Japan.

Blended gross margins during the quarter came in at 54%.

It is important to note that blended gross margin tends to fluctuate based on few factors, mainly related to the proportion of new products during a given quarter and the amount and percentage of service revenues.

As you know, new products are shipping at relatively low volumes, which initially does not enable us to take full advantage of cost-reduction measures.

Thus, a larger portion of new products in the mix mean a lower blended gross margin, and this can vary from quarter to quarter.

These fluctuations are an indication that our new products are being gradually accepted by the customers, which in turn leads to improvement in our market position.

During the first quarter of 2013, we have seen a product mix which included a lower portion of the new products.

This was accompanied by general increasing revenues utilizing the same infrastructure, as well as the improvement in service gross margin due to lower repair costs.

All these elements together accounted for the improvement in blended gross margin during the first quarter of 2013.

Looking forward into the rest of 2013, we expect to continue to see fluctuations in gross margins based on these parameters.

Specifically, in the second quarter of 2013, we expect to see a modest sequential reduction in gross margin due to a higher proportion of new products in the mix.

Over time, we continue to believe that our gross margin will stay in the 52% to 55% range.

Operating expenses increased to $11.8 million in the first quarter, mainly as a result of an increase in R&D expenses.

We see a high level of operating expenses over the next several quarters as we accelerate our planned spending.

As Gabi mentioned, this is to ensure that we deliver on our roadmap according to the pace by which our customers are implementing advanced technology node.

We believe that the timing of these investments is important to maintaining our role as a key technology partner to our customers.

Tax expenses in the first quarter of 2013 were $0.7 million and included mainly conversion of deferred tax assets created in 2012.

GAAP net income in the quarter was $2.4 million or $0.09 per diluted share based on a share count of 27.3 million shares.

Non-GAAP net income in the quarter was $3.6 million or $0.13 per diluted share, at the high end of our guidance for the quarter.

Operating cash flow in the quarter was $2.8 million, increasing the Company's overall cash reserves to $94 million.

Moving into balance sheet key metrics, accounts receivable increased to $19 million with DSOs decreasing to 60 days.

Inventories decreased by $1 million during the quarter with inventory turns increasing to 2.9 times a year.

Gross capital investments came in at approximately $2 million, reflecting continued investments in infrastructure related to growth and new products rollout.

Gabi?

Thank you, Dror.

And with that, operator, we'd be happy to take some questions.

Certainly.

Thank you, sir.

(Operator Instructions).

And we'll now move to our first question, which comes from Patrick Ho from Stifel Nicolaus.

Please go ahead.

Thank you very much.

Gabi, maybe just a little color on the commentary about the incremental pickup you're seeing on the memory side.

I'm assuming you're talking about conversions and upgrades.

How do you see that business trending I guess over the next quarter or two, and maybe a little bit of commentary on the margin impact, given that upgrades typically have pretty decent margins as well?

Yes, well, as I mentioned, we do see some action right now on the memory front.

It's not yet at a level that I would call it a sustained recovery yet.

As far as what I was speaking about, indeed, there are some fab conversions.

What the customers are trying to do is use as much existing capacity as possible to make modifications in order in DRAM to allow them to do 2X-nanometer, which will be somewhere around 25-nanometer in most cases.

And in NAND flash, it's still around a 19-nanometer design rule.

So, first of all, what we're seeing mostly is modifications.

At the same time, the big project in China that everyone's been discussing we believe is going to happen during the second half of the year.

That's a NAND flash project.

We do see also that the demand that I mentioned -- it's public so I can speak about it.

With Samsung coming to Hynix for DRAM demand I think that there will be some moves made by Hynix in order to increase incrementally their capacity and capability.

So, first of all, I think, indeed, this is something positive.

I don't see it as a trend as of yet.

As far as upgrade opportunities, these will vary between what I call a fab upgrade in the sense that, okay, they need more equipment, I would say process control equipment because they're doing a design rule shrink.

And in some cases, it's upgrade for existing tools.

So, it's either additional tools, which they use as part of an overall fab upgrade, or it is a move from an existing version to a more advanced version of the tool.

In both cases, this is good business and good opportunity for us.

I think what's also important -- and I've indicated this in the past -- is that we're seeing a transition, especially in the integrated metrology front.

In standalone, it always comes with the OCD modeling software.

Now, integrated -- and a lot of this activity for us is integrated metrology -- is shipping when these people are upgrading with the modeling software, which incrementally improves average selling price.

So, that's kind of the way you should look at that.

Great.

That's helpful.

Maybe a bigger-picture question in terms of the opportunity on the NAND flash side.

As you're seeing the industry go from planar to 3D, can you give a little commentary about I guess what's the incremental increase, whether by percentage or the total available market for OCD metrology as you go from planar to 3D, or what would you see on that front?

Yes, I think it's an excellent question.

And as I mentioned in my prepared commentary, the move to 3D NAND is going to take some time.

We even heard SanDisk -- you may have heard them also saying it's two, three years out.

They're not the only ones talking that way.

At the same time, what we're hearing from other customers is that the fabs that they are planning in the near term are fabs in which they do want to be able to finally come out with 3D NAND.

So, the work that we've been doing with them, and by no means is it now ready for high-volume manufacturing, has added several more steps of measurement because of the complexity of the etching especially that's associated with this process.

So, I think, in general, you are going to see an improvement there.

I will be looking probably maybe by the next conference call to providing some more color on how much more business I believe that creates for OCD, just like we did for the foundry.

I think that the reason why I want to do that probably in a few months is because we're studying exactly that now with customers.

And so, right now, I will say yes, it does offer more opportunities for OCD definitely.

The extent of it I think we should be able to communicate sometime in the next few months once we study in more detail how customers think they're going to do the initial move to high volume.

Great.

And final question from me.

In terms of the margin profile and a lot of the moving pieces you mentioned, whether it's OpEx, R&D, and as well some of the pressure on the gross margin front, on the R&D side, can you give a little bit of color in terms of the mix?

Are you investing in certain areas more?

Is it 450 related, or is it some of these next-generation technologies that are on the horizon that's causing the increase in R&D?

Well, first of all, I think that, to say because this is really implied in the question, we expect operating expenses to hover around the level that they were in the first quarter.

Okay?

So, that's, first of all, an indication right there.

As far as where we're spending, if you remember in my prepared comments, I spoke about the fact that the acceleration of the 20-nanometer ramp up has made these specific customers come to us and ask us to accelerate the work that we're doing to develop novel software capabilities.

So, a lot of the investment and focus is going to what I've coined our strategic software development effort, so a lot of activity there.

We're also, at the same time, working on our N plus 1 and N plus 2 systems because it's very important.

And actually, the depth of collaboration that we now have with these key customers has created a situation in which they want to see at early stages what it is that we're developing for them even two or three years down the line.

So, I would say, if I would try to break it down not in an accurate way but just to give you a feel, a lot of the investment is happening on the algorithm and software side in the short term at a faster pace than before, again, specifically to cater for customer demand, which is where -- exactly where you want to be, obviously.

And quite a bit, obviously, is also geared towards next-generation tools, which are necessary in order to solidify our position in next-generation technologies.

And finally, I will say, when we speak about this 20-nanometer ramp up, as many people know already, it's actually a 20- and 16-nanometer or 20- and 14-nanometer ramp up, depending on which customer you're talking about.

And so, the capabilities that we have had to demonstrate have been very high end because that's what they want.

They want to be able to qualify these -- both capabilities for their ramp up.

So, that's hopefully a good summary to give you a feel as to where the investment is coming.

Great.

Thanks again, Gabi.

Nice quarter.

Thank you very much, Patrick.

Thank you, sir.

We'll now move to our next question, which comes from Edwin Mok of Needham & Company.

Please go ahead.

Your line is open.

Great.

Thanks for taking my question.

So, first question I guess follows Patrick.

If you guys can give some commentary about your gross margin -- based on your guidance, it implies gross margin should come in at the lower end of your guidance range or your target range.

Is that just the timing of revenue?

I know, Dror, you gave some color there already.

But, is that due to the timing of revenue?

And do we expect a quick recovery as those new products start to get recognized in the back half of this year?

Well, as I mentioned in the prepared remarks, the main item currently at this level of revenues impacting the gross margin, the blended gross margin, is the revenues from new products.

And these were at a relatively low level at the first quarter.

Going into the second quarter, we again see a higher level there of new products.

Hence, there is some pressure on the gross margin.

And yes, we do expect the blended gross margin to be more leaning in towards the low end of our model.

Okay.

I think that ties to my next question.

You guys talked about this new software product for the hybrid metrology.

I think you guys have previously discussed that.

Was wondering when do you expect that to start contributing to top line?

And I suspect when that happens, that should have a positive benefit on gross margin as well.

Is that correct?

Yes, this is primarily a software product.

It does have some computation capabilities.

But, it's primarily a software product.

The nice thing about it is that we have built a business model that allows us to charge for the intake, obviously, of information coming from several sources, several possible other metrology tool sources.

This -- I would say that the remainder of the year, probably towards the end of the year, is when we should be completing a very, very thorough work and testing and high-volume manufacturing with the customer.

I mentioned GlobalFoundries because there's been a publication.

So, I believe that initial revenues should probably start coming in towards the end of this year, beginning of next year.

The extent of it depends on the extent of deployment.

Where it has demonstrated significant value is in complicated etching applications, where sensitivity to certain parameters is limited in some tools and stronger in other tools.

And so, the move, for instance, to FinFET, to the 3D transistors, is something that would be extremely helpful to proliferate this product.

The move to VNAND is also very helpful.

So, I would say, Edwin, on the top line influence, probably more pronounced in 2014, perhaps some already during the end of 2013.

Great.

That's very helpful color.

First -- second question I have is related to your largest competitor KLA, right?

Last week, they gave some color on their view of the market.

And they sounded a lot more cautious on the foundry side.

Any way you can help us reconcile that?

Is it just market share issue, that maybe they're starting to lose some share in the area that you guys compete with them?

Or do you think it's more related to inspection area, that they're seeing less spending there?

Any way you can help us reconcile that?

Well, I will not attempt to understand exactly how their numbers are broken down.

I'll try to maybe address your question through indications on what we're seeing from the customers.

Because of the fact that we have attained such a strategic position with key customers, like TSMC, we have gotten I think relatively early visibility of their plans relative to what we used to get in the past.

Same thing applies to other customers, such as Samsung, GlobalFoundries, etc.

I tried to articulate in great detail what we see.

And just to kind of summarize that again, what we see is very positive.

If you remember, probably in the last conference call, I spoke about the fact that we expected the 28-nanometer ramp up to be completed probably by September.

A very positive indication that we have and we already see this starting is that, actually, there's two other customers here, without mentioning, that are planning to ramp up 28 nanometers in the second half of the year.

The previous expectation we had was that it was questionable as to whether they're going to go after that business in the short term or whether it would take them more time to gain some market share in order to justify a ramp up.

So, that's one place where we gain confidence because we know we're well positioned for those two ramp ups.

I did say that they are smaller ramp ups than the one that occurred in TSMC, obviously.

But, they are meaningful for us.

And then the 20-nanometer, as I said, I think pilot production should carry the two customers that are actively engaged in that until the end of the year.

I think what's very important at this point is to see what will be the order timing for that significant ramp up that I spoke about.

When during the second half should those orders start coming in?

Is it the beginning of the second half?

Is it towards the end of the second half?

I think that is still not clear and very much depends on new product announcements that their customers will come with.

So, that's kind of what we're seeing.

As far as market share is concerned, we are staying our course but becoming more aggressive in achieving market share on the etch front.

This is a key focus for the company.

I've indicated in the past that it is already on the standalone front.

The revenues have exceeded 50% of our overall standalone revenue coming from etch.

That continues to be a focus.

And there I think is probably where we are gaining market share.

Does that influence numbers of a big company like KLA or not I think is difficult for me to indicate to you.

I think that's fair.

A quick question on the TSV product.

You guys recognized revenue this quarter on another tool, right?

And I believe previously you guys have said two other customer.

I missed that comment that you said regarding when you expect to get revenue at those two other customer.

And I think you made some comment about potentially more evals going out.

Do you expect those to start contributing to the revenue, or is it more like 2014?

So, the second -- sorry, the third and fourth tools are expected to be recognized in the second and third quarters.

As far as the other activity and as it relates to actual revenue, there will be two possibilities here.

First of all, we are hopeful to start seeing some repeat orders from some of these customers in the second half of the year.

So, that's one possible growth trajectory that we're hopeful to see, even in 2013.

The other one is the early engagements that we've had with customers that are still offline.

We are hoping that at least one or two of those will turn into actual tool evaluations on site.

And as I have indicated in the past, once those tool evaluations start on site, then there is a strong likelihood that you will turn those into revenue.

So, yes, there is a chance that you will see several more tools in the second half of the year.

I think it's mostly a timing issue as far as revenue recognition capability.

Great.

That's all I have.

Thank you.

Thank you very much.

Thank you.

We'll now move to our next question from Keith Maher from Singular Research.

Please go ahead.

Your line is open.

Hey, thanks a lot.

I was wondering.

You're talking about these -- the products, the new products that are going to depress gross margins for the next few quarters.

Is that mainly related to the 20-nanometer ramp?

And could you just elaborate as to what those products are?

Yes, and again, I want to -- since we're spending time on this, I want to make sure that everyone is crystal clear on this thing.

As Dror had indicated, during the penetration process with these new products -- and when I say penetration, it means that we're still shipping low volumes -- we don't yet have the ability to take full advantage of cost measures associated with larger volumes.

So, that's the way that you should read that, meaning I think we've done a good job price positioning these tools.

And I think that the influence here is mostly related to the extent of these tools which vary from one quarter to the next.

What tools are these, and where are they going?

So, on the standalone front, this is the latest model, the T600.

And on the integrated metrology front, it's the i500.

These are the two latest products that we have.

And as far as where they are headed, yes, indeed, they are focused on the 20- and 16-nanometer ramp ups that I had mentioned.

Okay.

And I think just a general question, I know you mentioned about software.

And I know you mentioned that some of the effort this year is focused on that.

And I'm wondering -- and this is just kind of a general question, not kind of getting to the specific products that Edwin was asking about.

But, when would you start seeing a benefit from that investment?

And is there any chance, longer term, that the mix of your solutions may include a higher software component?

Well, first of all, yes, this is a very strategic effort for us.

What we are doing, clearly, and this is why we're investing in it because we see exactly the path to getting there, that Nova in the long term should be a company that has much more significant software components in its revenue.

We believe that that is where the industry is heading, that the complexity associated with the type of measurements that we do requires very, very innovative software solutions.

For that reason, we put together a pretty significant group of people.

We put that whole group under a VP.

And therefore, from a strategic standpoint, indeed, I expect that, for the long term, software will become a larger and larger portion of our revenues, which obviously, as any hardware company expects, offers a very nice potential from a gross margin standpoint.

As far as the extent of it, well, I think that, if you look at Nova's results in the last few years and you see other companies out there, without mentioning names, with different gross margin numbers, I think we've done well on the gross margin side because you already have significant portions of software in our product offering at this point in time.

What you will see, though, going further down the line, is that those elements increase, that perhaps some of these products will actually have their own separate existence.

In most cases, they are incidental to the hardware.

But, in some cases, these will be software solutions that have an existence of their own.

And at that point, it will become more and more meaningful.

When do I expect that starts happening?

I think significantly starts happening in 2014, already happening in the last few years for us.

And I think the way to look at that is if you analyze our revenues over the last eight to 10 quarters that it has oscillated up and down many times with the industry.

You see the gross margin holding up nicely because of those software components that we've been developing all along.

Okay.

Great.

That is helpful.

And one final question for Dror.

Inventories were obviously down sequentially about $1 million.

And I'm just trying to understand, though, on the cash flow statement why there was an expense for an increase in inventories.

Yes, that's an accounting issue actually because some of these inventory changes in the quarter were move of some of the tools to fixed assets because they will be used for application and demo.

Then this is -- this move, which was significant in the quarter, is neutralized in -- within the cash flow.

And this is why you see a different number there.

Okay.

Thanks.

That's all I had.

Thank you, Keith.

Thanks.

Thank you.

(Operator Instructions).

And we'll now move to our next question, which comes from Krishna Shankar from ROTH Capital.

Please go ahead.

Your line is open.

Yes, for the quarter, what was the mix on revenues between your foundry and memory business?

And are there any industry data points that you see out there which would suggest an acceleration in conventional 2D-related NAND capital spending going into the second half of this year, especially since NAND flash capacity seems to be pretty tight?

Yes, so, on the mix of revenues, Dror, you indicated it was --?

80% foundry and 20% memory.

And as far as your question, Krishna, on the investments in NAND flash, I think, as I had mentioned earlier on, because of the fact that 3D NAND high-volume manufacturing is going to take two to three more years, you're going to see a lot of innovation taking place with NAND flash manufacturers to try to go below 19 nanometers down to 16-nanometer design rules and still make some changes, but still, I would say in planar fashion rather than three-dimensional fashion.

And as far as investment, well, it has been announced that Samsung will be opening a fab in China.

We expect those investments from what has been communicated publicly to start towards the end of the year.

And in other places in NAND flash, we think that there will be incremental investments in existing fabs is what we see right now.

Great.

Thank you.

Thank you.

Thank you.

Thank you.

(Operator Instructions).

And we'll now move to our next question from David Wu from Indaba Global Research.

Please go ahead.

Yes, thank you.

Let me ask two questions.

First thing is, if you go back in history and look at memory as a percentage of total revenue, if you look at the last five or six years, if we have a normal quote-unquote memory market, what's the percentage of your business that comes from memory?

It used to be 50-50.

For instance, in the years of 2010 and 2011, it was 50-50 between the two, between foundry and memory.

And so, that was -- in the previous up cycle where memory was doing well at the same time as foundry, we were at about 50-50.

Okay.

In terms of software, I guess it's hard to break out your software contribution right now, right, because it's bundled in your hardware.

Correct.

In a couple years' time, once you get 3D NAND and all the more sophisticated stuff coming down the pike, would it be 10% of your revenue potentially as a standalone business?

You know what?

I think it's too early for me to indicate about that.

I would like to see it as a significant part of our revenues.

What does that imply is a little bit too early for me to say.

And the reason I'm being cautious here is not just because we still need to see a significant adoption of these products, but it's also the case, if -- and you seem to know the industry well -- that software which is incidental to equipment sells well.

Software which is separate is another thing.

And so, it's still early to say how this thing will play itself out.

I think the good news is that this is an opportunity to grow top line.

It is an opportunity to improve margins.

So, this is what I answer to investors.

To customers, it's all about expanding the solutions that we're providing.

And I think that's the most critical thing.

So, I think it will, as time passes, become a more significant part of our business.

As I said, there's a lot of strategic thinking that has gone into this process in the last few years.

We have an extremely strong team here in Israel that I think is going to come up with novel solutions and directions.

And so, that's the way you should look at that.

I was thinking, going on the same line, that ASML bought this company Brion, which is a pure software company, as an adjunct to their business.

I was wondering, with so much cash on your balance sheet relative to the size of the Company, would tuck-in acquisition in software make sense for you?

It could make sense.

As I've said in the past, we're actively looking for things that are synergetic to our business, both from a technology and business standpoint.

So, that would not be out of the question.

We are looking for innovative ways to satisfy customer needs.

And if there are opportunities to shorten that cycle, if it makes sense, we'll take very serious looks at it.

And indeed, I think the example of Brion is -- it was a very good example in the case of ASML.

Last question I have is, can you remind us of the metrology intensity going up between the 20-nanometer planar/FinFET versus 20-nanometer planar?

Is there much of a kick up?

Well, in foundry, I think that I would say the following.

What we have indicated to the market is we've been talking about how the capital intensity grows it when we go from one node to the next.

So, we have indicated that, in a 100,000-wafer-start foundry with 28-nanometer technology, the addressable market for us is $150 million to $200 million.

Below 20-nanometer, that number goes to above $200 million for a single 100,000-wafer-start fab.

So, that's the way to look at that.

Okay.

Thank you.

Thank you.

Thank you.

We'll now move to our next question from Rob Ammann from RK Capital.

Please go ahead.

Your line is open.

Yes, nice quarter.

Congratulations.

Thank you.

Great.

Can you talk about hybrid metrology and the opportunity for data intake from other sources?

It sounds like that could be a recurring source of revenue that would be high margin over time.

Is that the right way to think about that?

Yes, well, let me use by way of example what we've been presenting with GlobalFoundries.

And we've been showing this development effort with GlobalFoundries in the last three years already.

And this year was the first time we showed it in production actually.

The initial usage that we're seeing is an interesting one in the sense that we are taking CD-SEM information and bringing it into our modeling capability.

As you know, we perceive CD-SEM as competing technology, what I call indirectly competing technology.

And that is a classic case of SEM having sensitivity, CD-SEM tool having sensitivity to very accurately give you information about the width of the critical dimension and us injecting that information into our model in order to increase our sensitivity to artifacts that they cannot even see, such as side wall angle of trenches, etc.

So, first of all, that's the technical usage example.

Now, there are more metrology instruments that the customers are asking us to look at in order to what I would say hybridize.

And I think that will grow over time as well.

As far as recurring revenue is concerned, the business model will be built such that there will be licenses that are associated with being able to do this.

You will find a server, and you will find run-time licenses associated with these particular tools, both of those being revenue opportunities.

Obviously, there will be a lot of work going on in order to develop applications for this usage.

And indeed, as you had mentioned, from a margin standpoint as a hardware vendor, that does offer an interesting opportunity.

The reason why the customers are so actively engaged and willing to spend money on that is because, for them, this is an excellent situation to be in, where they know that they must invest still in CD-SEM.

It's still a very significant market.

They know that OCD is an enabling technology.

And they say, is there the possibility of us combining the best of all worlds into a solution?

We chose to go after that three years ago before anyone else, even though it was not obvious because it implied cooperating with competitors.

Gotcha.

That's helpful.

Thank you.

Thank you.

We'll now move to our follow-up question from Edwin Mok from Needham & Company.

Please go ahead.

Your line is open.

Hi.

Thanks for taking the follow up.

Just one quick question.

So, if I look at your operating expense, right, we understand that you need to invest in and you're developing some of these new technology rules in TSV as well as in the software product.

But, it looks like OpEx has increased maybe around 40% from the 2011 run rate.

Was wondering, Gabi, if you think about your size of your operation right now, do you -- what kind of revenue run rate do you think your operation can sustain, or do you see that you might need to increase that OpEx a little bit more to capitalize on all these opportunity that you guys are targeting?

Well, I think the way -- as a CEO, the way to look at OpEx, it's a means to an end, right?

I think that what people should be expecting us to do is to utilize to the best of our ability the opportunities that we see.

At the same time, what I have said repeatedly and I think we've done a good job, even in the times where we have increased our investments, we have held profitable I think in a nice sustainable way.

So, first of all, I would say, from a philosophical standpoint, the increase in OpEx that we're monitoring very carefully is a means to an end.

And we should continue to focus on that.

As far as how far it can carry us, I think that, again, given the product offering that we have right now, we have what we need probably for the short term.

But, you must understand also that, as these opportunities come and customers are eager and interested in receiving more product from us, we go after those.

So, as I had mentioned, for the near term, we expect to stay at the current level of OpEx.

In Q1, I believe it was $11.8 million.

You should expect us to hover around that level for the foreseeable future.

As far as how far that OpEx carries us, it's tough for me to name a number for that, right?

I don't know, Dror, if you want to take a crack at it and make comments here.

Well, I think only one elaboration on that is the fact that the connection between operating expenses and the changes in the industry is quite strong.

And currently, we have three transitions going on.

One is 450-millimeter.

The second is going down to 10 nanometers.

And the third is Through Silicon Via.

So, we need to prepare to all these opportunities.

And this is why you see in general operating expenses increasing.

I see.

Maybe a different way of thinking about that, how much of your OpEx is sales-related expense, meaning that, let's say you are successful in some of these new adventure and you can drive revenue to be quite meaningfully higher than your current level.

How much should we think about your OpEx will have to increase just because you have a higher sales level?

Well, most of the expenses are not strongly correlated to the revenues.

So, I believe, in general, we can increase revenues significantly without significantly changing the position of operating expenses.

As of today, this, of course, depends on more of the development cycles and the new-generation products, as Gabi had mentioned.

Some of the sales and marketing expenses are related to the level of revenues.

But, that's not at the level which dictates a significant increase of operating expenses.

I would say the following, Edwin, just to add one more comment here is that significant market share and increases for us that we're working very aggressively on can be executed with the current OpEx if we make those significant increases in market share based on the current product offering.

I think that there's a strong likelihood that we could do that.

So, that could solidify the ability to take up revenues and stay with the same OpEx.

I think, still, the way you should look at it is we always look back and see, okay, what else needs to be developed here?

Are there enough support people to support the growth effort that's taking place at the customer site?

These are the things that come to mind.

I see.

One last question, and I'll go away.

I noticed that the guidance range or the revenue guidance range is a bit wider this quarter.

Is it just because of timing, or why was that a little bit wider?

Well, the main reason, and this was the case also in Q4, the main reason for that is, again, the level of new product.

As you can imagine, these new products involve customer acceptances, which can roll between quarters.

And this is why the range in the second quarter is a bit larger than the first one.

Great.

That was very helpful.

Thank you.

Thank you, Edwin.

Thank you.

And it appears we have no further questions at this time, gentlemen.

Okay.

With that, I'd like to wrap up the call, thank everyone for attending.

Look forward to meeting you in the near future.

Thank you very much.

Thank you, sir.

That will conclude today's conference call.

Thank you for your participation, ladies and gentlemen.

You may now disconnect.