Seeking Alpha
We cover over 5K calls/quarter
Profile| Send Message|
( followers)  

Photronics, Inc. (PLAB)

April 03, 2013 2:30 pm ET

Executives

Peter C. Broadbent - Vice President of Investor Relations and Marketing

Constantine S. Macricostas - Chairman, Chief Executive Officer and President

Frank Lee

Sean T. Smith - Chief Financial Officer, Principal Accounting Officer and Senior Vice President

Christopher J. Progler - Chief Technology Officer and Vice President

Peter S. Kirlin - Senior Vice President of U S & Europe

Soo Hong Jeong - President of Asia Operations and Chief Operating Officer of Asia

Peter C. Broadbent

Welcome, everyone, to the 2013, Photronics Analyst/Investor day. I'm Pete Broadbent, Vice President, Investor Relations and Marketing of Photronics. Thank you all for joining us.

Slides for today's presentation are available on our Investor Relations website at Photronics.com. Before we begin, I'd like to draw your attention to Slide 1, which contains our Safe Harbor statement and also to Slide 2, which discloses that some of our comments today will reference non-GAAP results.

Our agenda today is as follows: Constantine "Deno" Macricostas, Chairman and CEO, will begin the day with introductions and opening remarks; Sean Smith, Senior Vice President and CFO, will provide investment highlights; Dr. Chris Progler, Vice President and Chief Technology Officer and Strategic Planning, will offer insights on technology trends; Dr. Peter Kirlin, Senior Vice President, U.S. and Europe, will then cover high-end IC operation strategies; Dr. S.H. Jeong, COO and President, Asia, will provide an FPD market update; and Sean will address our financials and outlook.

At the end of these presentations, there will be a question-and-answer period moderated by Sean, and we will be taking questions from those assembled here with us in New York City.

It is now my pleasure to turn the meeting over to Chairman and CEO of Photronics, Deno Macricostas. Deno?

Constantine S. Macricostas

Thank you, Pete. Welcome, everyone, to our 2013 Analyst/Investor Day. It is our pleasure to be hosting today. We will continue to make excellent progress in our strategy in our business.

I'd like to introduce you to our management team, [indiscernible] successful. I'm proud of my team. They're all very focused, very intense, very passionate and they like to win. S.H. Jeong, the President and COO of Asia for 15 years. I will not elaborate his past experience [indiscernible] in the semiconductor business for a long, long time.

Peter Kirlin, Senior Vice President of U.S. and Europe. He's been with us for 6 years.

Frank Lee is the President in Taiwan, [indiscernible] semiconductor mask corporation in Taiwan and has around 7 years of experience. I mean, he's been 7 years with the company. How many years? He has experience as much as I do.

Chris Progler, Vice President and CTO, Strategic Planning. He's been with the company for 12 years.

Sean Smith is Senior Vice President, Chief Financial Officer; and Peter Broadbent is the VP, Marketing and Investor Relations. He's a newbie. He's been with the company for 2 years.

And a couple of other ones in the audience today. I appreciate the support. There's PJ, [indiscernible] and Richelle Burr, [indiscernible], General Counsel. [indiscernible] I cannot thank you enough for the support.

The company strategy overall, really, the company started like the American dream in a garage. But it was always behind 2, 3 generations. There was no easy money. Mortgage your home, you start. Over the years, really, for me, this generation has become a bench model. A few years ago, we decided to pursue continue growing and become, a force in the industry we had to have high end. And to do that, we had this JV with Micron. Micron is a memory company, a memory drives technology. And that was a big deal.

Of course, under the JV, we invest over $300 million in Boise, [indiscernible] some difficulties. We overcame the difficulties and very quickly the market share [indiscernible] market share and basically became a force in the industry.

And I believe today, we're very successful in the high end, and we continue to gain much.

At the same time, on the flatiron display in the high end. And today, we'll do all the high end as they will explain later, the AMOLED [indiscernible] the size, the technology and display. And at the same time, basically, we've got this through the memory, we've got some foundry logic. So the size worked well if you compare the situation there overall, today, we are a force in the industry.

Meantime, to accomplish our goal, we have to reduce cost into our markets to invest our money, definitely the company to announce. By reducing cost and continue reducing cost, we've been able to improve our balance sheet. And please, of course, cash flow, gave us the opportunity to invest in additional equipment and technology.

Even with small problem, global, CapEx basically, U.S., Europe and Asia, we only have the company at the same type of footprint is global. We have the right locations to service our customers. And at the present time focused mostly in -- mostly U.S., focused in the U.S. and Asia.

Our continued goal is to reduce cost and improve our balance sheet and continue to gain market share is working very well [indiscernible].

This slide shows you really, the only area have [indiscernible] U.S. and Asia, that's why we decided to invest safely on the high end in U.S., Korea and Taiwan. At the present time, really, I believe, Taiwan got opportunities for us. I think the epicenter of the semiconductor industry, and I believe there are more companies in Hsinchu Park making chips for United States. [indiscernible] So I think it really gives us opportunities that's why really at the start of PSMC later, I'd be able to eventually compete on price but definitely have the opportunity out there because we see that's what it looks like.

Talking about Taiwan, nobody knows better than Dr. Frank Lee. Frank has been with the company, how many years?

Frank Lee

Seven.

Constantine S. Macricostas

Seven. Frank has worked in the United States, educated in the United States, [indiscernible] or more importantly I believe he knows Taiwan and China, Taiwan customers in Asia. And I'd like to thank [indiscernible] in Taiwan.

Frank Lee

I don't need that. Deno, Thank you for the very good comments. Thank you. Welcome, everybody. Good day. As Deno mentioned that there's a lot of opportunity coming in Taiwan. So to begin with, I'd like to share with you our status in Taiwan.

I believe most of you are aware, Taiwan is a very, very competitive market because all our competitors are in Taiwan. At the same time, the market has a lot of opportunities, and that's also why all the competitors have operation in Taiwan.

However, we are a strong player and we have a lot of production capacity, and we have very good capability in high-end, and most importantly, we have been profitable all years.

We have a very solid market share in Taiwan, especially in the foundry fab, including UMC, [indiscernible], power chip, and many artifact. We know UMC is a very strong foundry player. They are our top 20 customer. Their CapEx this year is over $1.5 billion, many to expand their capacity in 28-nanometer capacity. Photronics, we are a close partner with UMC, and this year we will be their 28-nanometer photomask supplier.

In Taiwan, there are many, many IC design house, and they are all our key customers, including MediaTek, Novatek, RealTek, Aurotek company, you name it. So our market share with all these customers are very high because of our service, our local presence and our support for this customers.

In addition, we provide a very strong local support to our Micron affiliates in Taiwan. We know Micron has a joint venture, Inotera, and technology partner, Nanya, both in Taiwan.

Our local support covers customer service, and we also have local manufacturing and also local repair service. So the 3 service combined together are critical to our customers.

We move on to Singapore and Malaysia. Singapore and Malaysia, especially in Singapore, they are Micron IM Flash, and also they are Microtek semiconductor in Singapore. We also support these 2 companies.

Other foundry house like UMC, GLOBALFOUNDRIES, SSMC, et cetera are also our major customers in this area. So from Taiwan, we do support Singapore with all our good performance in terms of delivery and also sales service support. And that's why our revenue from the region increase every year even after we shut down our Singapore fab operation in year 2011.

Talking about China. In 2000 -- from 2011 to 2012, our China revenue grew 70%. This market place continue to grow in terms of capacity and capability, and we spent a lot of efforts in China market, and we have deepened our close relationship with some of the major customers, including SMIC, Huawei NEC, GSMC and UMC.

If you look at the map, this is in Asia, and we have customer in Taiwan, China, Singapore and Malay. And you will look beyond outside of Asia, we have customers in North America, Europe. So to serve all the customers globally, we have been working very closely with PKL in Korea, nanoFab in Boise, other high-end product, and also with Rufier [ph] and Aron [ph] on a mainstream product.

So Photronics, we have this integration of global manufacturing network. We serve all the customers. And this has been proven very successful, and yet it helps Photronics in our both bottom line and top line. So in summary, we have a lot of opportunities in front of us in Taiwan and Asia, especially the Micron pen with the Cherry Inotera [ph] and expected new members, the Opita and grass strip [ph], hopefully it happens soon this year. All this always combined, we have very strong demand for photomask in this year and next year. So this is a real golden opportunity. We will work from Taiwan with nanoFab, closely to support all these customers and to capture the billings.

As Deno mentioned in the beginning, we are very excited with all the opportunity in Taiwan and Asia, and I believe Deno has very strong expectations for us in Taiwan, and we are very confident this year or next year, we will achieve great success in Taiwan. Thank you. Thank you to everyone.

Constantine S. Macricostas

I think [indiscernible] very, very well the opportunities in Taiwan. And that's why a couple of years ago, slowly buying the outstanding shares in the market and we increased our position of 37% [ph] to 75% so in many, 25%, [indiscernible]. I think it's more appropriate in a few minutes. So that's one of the reasons we made a commitment to Taiwan, because we do believe it is very important to us to be at the service -- I mean, China, Japan, Malaysia, Singapore and of course, Taiwan. And the remaining 25% now remain. In the past, [indiscernible] so 100% or 100% PSMC become part of the plans, 100%. Overall, I think, we're financially -- we should continues strengthening our balance sheet, and at the same time, investments in technology. And I believe that would give us the opportunity really to continue growing and outperform the competition. I believe that today, we're in excellent position because our competition is not investing. The only company in photomask [indiscernible] investing. I believe, what we hear and what we know from our suppliers, we invested more than all our competition towards [indiscernible]. Now they go and invest it one more previously, because of one investment. So it comes down to 20-nanometer, 14-nanometer. And of course, the photomask. We're the only [indiscernible] but we will be the only supplier they got. So we feel very well with the longer-term about the company. In time, we feel we will continue to improve our balance sheet and continue to make sure we make a profit everybody with good disposition. With that, I'd like to turn over to Sean. Sean?

Sean T. Smith

Thanks, Deno. Good afternoon, everyone. Deno said something in his remarks when he was talking about the management team over here. But I think throughout this presentation -- these presentations, you'll see the optimism we have for the future, the market share gains that we earned over the last few years. This team was together during the financial crisis in the last downturn, and they worked hard together. We worked hard together. And that's why we believe we're more flexible than our competition and we're winning. Hopefully you'll see that throughout this presentation.

Taking a look at some of investment highlights. Last -- at the end of 2011, the semi industry had a bit of a slowdown. 2012 was a down year for semi, but Photronics remain profitable. It was able to improve its balance sheet year-over-year. We continue to strengthen it, as you'll see in the slides later on when we get back up from 4 or 5 years ago, to tremendous turnaround.

We continue to execute on our high-end strategy, while servicing our mainstream customers. On the IC side, our mainstream last year was about 70%, 65%, it is -- it generates a lot of cash for us. It enables us to invest in high-end technology.

We still believe, and you'll hear from Peter and Chris and S.H., that there's opportunities that exists for additional high-end share gains. And as we stated at the end of Q1 on the conference call, Deno stated and I reaffirmed it, we do expect, on a sequential basis, Q2 to be stronger than Q3 -- Q1, I should say, and we should see that each quarter there out. And we have a lean and flexible business model we talked about. We have the same disciplined approach to cost controls that we instituted or increased 4, 5 years ago. Deno, what are the 3 words you always tell us when you're talking to us?

Constantine S. Macricostas

Cost. Cost. Cost.

Sean T. Smith

So our model on incremental revenue generates about a 50% drop through to gross and operating margin line. Now we sell 2 types of photomasks, IC photomask, which on average are about 75% of our top line on a quarterly basis. We sell those worldwide. And FPD photomask, which is about 25% principally in Korea and in Taiwan, and S.H. will be going through how well we're doing there a short time.

Our revenue is driven by new designs coming to the marketplace and consumer applications are accelerating the number of new designs and new designs drive both IC and FPD proliferation. We're seeing a greater number of product types with shorter product life-cycles. Every new design that comes to market or an enhancement to an existing product requires a new photomask set.

On the flat-panel side, we're seeing FPDs appearing in more devices, more sizes in a variety of definitions. And S.H. will talk about the proliferation of these tablet shipment share. You see that's going up from fewer tablet sizes to many more in the future. And certainly the AMOLED-based product is driving a lot of that. S.H. will talk about that in a few minutes.

Looking at an overview of our IC mainstream high-end, Peter's going to talk about the significant growth in our high-end business over the last 5 years. It's really been outstanding. We expect that to continue. On a down year, our high-end business last year was up 16% and FPD is down from a record in 2011, but still, we saw good high-end penetration.

When you look at our worldwide customers, if you take a look at that list with Samsung, which has been a customer for an extended period of time, so our #1 customer, about over 20% of our top line. However we sell both ICs and FPDs to them. When you take a look at Micron and affiliates, that's greater than a 10% customer for us. But some customers on this list weren't there 4, 5 years ago. And these customers would be -- we count on being the net winners in the future.

Pete Broadbent is going to come up and talk a bit about market share. He spent a lot of time working with SEMI studying the report.

Peter C. Broadbent

Sure, Sean. Thanks. For the past several years, on the FPD side, due to the lack of the supplier for data in the market, but also due to our strong customer relations among the leading manufacturers in FPD, and also our strong presence in the high manufacturing areas of Korea and Taiwan, we provided our own estimates. And in a down year, last year, as S.H. will provide some color on, we believe we grew share. And over the last 3 years, we've grown share from 2010 to 2012 from 20% to 25%.

In the past, for IC, we cited SEMI data. They've just released last week. But for the same reasons on FPD, we've decided to provide our own estimates, which we think are consistent over the last 3 years in terms of methodology. And for the same reasons, now we have strong key relationships with the major providers. We are in a strong presence in the key manufacturing areas, U.S., Korea and Taiwan. And with our daily contact, weekly contact, with approximately 600 customers worldwide, we have a keen perspective on the market. And we believe, over the last 3 years, we've grown share from 19% to 22% to 23% of the merchant sector. And even -- despite compression in the merchant space last year, we grew share by 1 point to 23%.

SEMI did release. We've analyzed their data, and their data and their trends support these figures. And so we have further color and commentary coming forward as a result of further analyzing this. And we'll continue to provide our own estimates in this area, but we like a consistent approach to this and continue with these efforts in the future.

For now, we take questions on this for Q&A? Back to Sean.

Sean T. Smith

Thanks, Pete. Just a brief summary. I'll be back up to give more granularity to the financials and where we expect to go. New designs drive IC and FPD sales. We have a diversified blue-chip client base and broad diversification. We're not segmented in one specific area. We firmly believe we're well positioned for additional growth and market share gains. Deno said, the team wants to win, we all want to win. No question about it, and we expect to win. And we expect increased market share in the high-end segment of all, high-end IC and FPD, and we expect that to continue to 2014.

And with that, I'll turn it over to Chris Progler -- Dr. Progler, I'm sorry.

Christopher J. Progler

Thank you very much. I'm going to give a little perspective on the technology side of photomask, what some of the trends are, kind of what drives us and the new technologies maybe in the next 3 to 5 years. What's going on? First, how do we think about our technology development priorities? Definitely, we listen to the customers. We always have. We drive our customers or already talked about [ph]. We're locked into them. We use their road maps directly in our development agenda. We communicate with them often. So a lot of our even advanced R&Ds is very much driven by customer needs. It's that kind of business.

For example, something like double patterning, multi-patterning, that we'll talk a little bit more about in a couple of slides, even as back as 2007, 2008, we were already in production on this technology. It came from flash memory. It's the first chip technology to go into multiple patterning. So we've done mass production of double patterning for 4 or 5 years now. At the same time, we stay very flexible. We customize our processes for regional needs, Korea, Taiwan, foundry, memory, microprocessor. We're very application-specific. We have a strong technology platform that allows us customizations.

On the display side, of course, we stay very focused on the trends. We anticipate the greater complexity and displays of AMOLED, higher definition TVs, that drove a lot of our early roadmap activities. And we have good alliances with manufacturers and other suppliers.

So in terms of our pipeline, just on the IC side, this gives you a flavor on some of the early view we get at some of these advance nodes because we're working with these customers. For example, the memory side, Micron, Samsung, back in 2010, 2011, we're already at 2X node on memory. Intel, of course, 22-nanometer, Q1 2012. So if we have any access or any view into these technologies, it pushes us early and it allows us to get a strong technology platform that we can deploy with foundries and other segments, which usually come online with these kinds of nodes a few years later.

In terms of what driving IC mass technology today, it's really this multi-patterning era. EUV, which I'll talk a little bit about, is a single masked solution. So 1 mask, 1 device layer in current model.[ph] We believe, and I think industry consensus is starting to come around, that, that will be somewhat delayed, maybe 1 year, maybe 2 years, et cetera. So almost every customer we spoke to now tell us, we're going to need multiple patterning or multiple masks for single device layer for a longer period of time. You are going to have to rely on that 20-nanometer, 14-nanometer, perhaps even a node after that. This is generally good news for the mask company. We expect to see more masks from device layer required to deliver those nodes, and that a number of masks it takes to deliver a single device layer could go up substantially. For example, in the bottom here, you see, this is a slide from an Intel presentation, summer 2012. This shows 215 -- 2015 high-volume logic node, 5-mask composition to do a single-chip device layer. So there's one layer that does kind of grid of lines. Then there's 4 masks that cut these lines to make a functional device. So in that case, this is 5 masks needed to deliver a single device layer of a functioning circuit.

The other thing we notice is even in the fabless space, for example -- this is a quote from a paper given by a gentleman from NVIDIA. While they don't have obviously their own fabs, they're starting to understand and realize the criticality of photomask for them to deliver their processes. And the title of this paper was Mask Making: The Most Critical Step in New Product Introduction Cycle. That's pretty unusual for a fabless company to call out the strength of the mask technology and how important it is to it. So we believe multiple patterning is here. We think even fabless companies understand the importance of mask technology. And this multiple masks to deliver a single device layer will be a trend we're prepared for and will be a good opportunity for us in the future.

In terms of that, we have technology as I said, since 2007, 2008, connected with flash memory that evolved into DRAM and logic that have used these multiple masks. And there's 2 things really that make it unique what we've developed. The first thing is if you can imagine using multiple masks to make a single device, they have to align very precisely. You have to register them and align them extremely well because they will have to compose to form a functional device layer.

That alignment technology and the methods we used to do that, how we used the equipment actually is a very strong piece of our capability refined over 4 or 5 years. As this moves more pervasively into logic, that will be a very powerful technique that we have that can be used in that space.

The other thing that's very critical for multiple patterning is the process flow. So the wafer fab now is waiting on 3 masks to arrive before they can start a single layer of their device, such as a contact layer or gate layer. So how we synchronize and coordinate those 3 layers to get into the wafer fabs that doesn't disrupt the wafer flow, a very critical technology we've had to develop over the last 4 or 5 years, connected with multiple patterning work. So we really do think this is differentiating for us. Since 2007, we've been working in this area. It would allow us this kind of coordinated entry into the wafer fab means very, very high mask yield. This is the only way to do it. If one of these does not show up, the wafers do not start. We need all 3 of them to be there. So we have very strong yield in our high-end fabs, mostly driven by the memory mask platform. And we have ways where we can understand that these masks will work well together in advance them to getting into the wafer fabs. So these 3 things are quite powerful pieces, part of our multi-patterning portfolio.

In terms of what we're doing to actually put this kind of technologies into mass production, as Deno and Sean mentioned, we did make investments in incremental tools, particularly state-of-the-art writers, EBM 8000. We're having installs in 2013 in U.S. and Asia. These 2 will be fully extendible from 14-nanometer node, and they really do have capabilities that address and allow us to deploy that multi-patterning technology. And particularly, the placement of the mask features takes a big step-up on these tools. So it's compatible to this multi-mask composition. And the second is they're faster tools. So when they try to deliver multiple masks through one device layer, we have much higher speed coming off of these.

The final benefit we have, since we do have a joint venture that's had this piece of equipment in place for quite some time, to some extent, for certain of our products, it's essentially prequalified. We already have process flows running on this that are qualified in our JV. So once we get one of these in our own facilities and we can start running, it's prequalified for certain technologies.

So how do we convert this kind of capability, both the equipment and the process know-how into business? Some of the other speakers, Peter and S.H., will talk more about this, but I'll just give you a view from the node perspective and how we look at it. In terms of mass production ramp, 28-nanometer node, for sure, this will be a strong year for that. We think this node will be a strong node. We'll have pretty long tails in foundry. We're entering mass production ramp for this node now. For 20-nanometer flash and so-called 3X DRAM, also strong nodes, we think, this year. After memory consolidation, which we believe will happen this year, we'll have access to more of that production capacity that we didn't have previously.

In terms of leading-edge qualifications that are closing or getting very close to closing, that could result in material business benefits. This 22-nanometer microprocessor, the next node of DRAM, so-called 2X node, [indiscernible], so new architectures, all of the companies is looking at for flash memory, we expect this year those qualifications finalizing.

In terms of new qualifications, it's really 20 and 14-nanometer logic. Another turn of the crank on DRAM, low 2x node and also particularly extensions to DRAM for mobile applications. There are some differences in the DRAM architecture used in mobile applications are initiating some of these qualifications now to address those differences.

And finally, in the R&D phase, next node of DRAM is so-called 1X. The variety of new memory cell types that are going to come in at that time, particularly 3D NAND and a variety of things like that, 11-nanometer logic. And then multiple patterning extension, we think, will hit at least the 11-nanometer node. Also, we have to talk about EUV as well on that timeframe.

So how will this -- will it be wafer fab capacity that will use this kind of technology is an important question because that's how you convert technology into business. This is a slide that came out of the recent SEMI report just released in March, last month. And I'll draw your attention to this red bar, 18- to 37-nanometer. This is really 28, 32-nanometer. At least according to this, you'll see a somewhat of a transition expected in this node where there will be some foundry work that will continue. And we do think this will be a strong node for particularly the memory and microprocessor stuff and the most leading-edge foundry are going to transition out of this pretty quickly. We think 20-nanometer from that view will be a pretty short node, and a lot of the foundries are going to try to push the 14 or a high teen kind of node pretty quickly. And you'll see that reflected here in this light blue lines, starting in the middle of this year, a fairly rapid transition, 14-nanometer node.

So if you look at Photronics' positioning to service this kind of wafer fab capacity, we have good install base and equipment in this timeframe addressing these nodes. We have multi-patterning technology with differentiation. That's the only way to do these. There is no other way to deliver those kind of wafer ground rules without multi-patterning. And we see a reasonably healthy ramp of those advanced technologies. We think it comes together quite well for us in terms of the growth opportunities.

One of the things we are also looking at down the road and in the future, our new architectures to help us be even stronger. And as Deno mentioned, for many years, we were a little more of a fast follower, different kind of business model. Now even some of our customers look to us to lead new technology, new initiatives, what are you bringing next, how you can help us from a technology perspective. One of the things we are working on, on mask writing, particularly because of multi-patterning and having to deliver all of these masks at once, we have to get write times faster. We have to get them written faster. So we made an investment in 2011, alongside Intel, and since then, other companies have joined in the so-called multi-beam mask writing platform. And to give you a flavor of how that's different than how we do it today, the way masks are written today, single pencil, you write it basically, 1 beam, 1 line and you go through the entire mask. The so-called multi-beam architecture has 256,000 simultaneous beams writing the mask in kind of a sweeping fashion. So it's difficult technology to put into production but the progress is going well. We're an early mover on this, and we do think, if it comes to realization, will really give us a cycle time advantage in multi-patterning and for the foundries.

So beyond 14-nanometer node, which I've spent most of the time on so far, we've always talked about complex masks and the so-called idea of gridded design. This was the picture I showed that had the lines and spaces that were cut out. We also see gridded design, of course, in flash and microprocessor technologies. This is a FinFET picture that's been around so much. That's a so-called gridded design. It's worked its way into memory and MPU today. We think that will eventually move into the foundry space because it's really the only way to do some of these advanced ground rules with multi-patterning.

Beyond that EUV, 2 things that are getting on to the technology agenda we're working harder on. EUV, there's a lot of discussion in the industry on that. The second is so-called directed self-assembly. This I point out because there's a misconception among some that this is a mask-less technology. Directed self-assembly is really a combination of a mask getting printed to create a base layer, and then from that base layer, there's a process that happens that assembles the lines and connections based on chemical process. But it's a mask process. Masks make the base layers, and masks are heavily integrated into that. So we do not view this as a threat at all. We view this also as a roadmap enabler, which masks can make a big -- have a large participation.

In terms of EUV, I'd like to talk a little bit about what we're doing in that area. While we do think it's been pushed out 1 to 2 years, based on our assessments now, that generally is -- from a merchant mask perspective, it's a fairly positive thing if EUV gets delayed. It really needs longer reliance on multiple patterning, and I already talked a lot about that. It means continued push to our so-called gridded design. These are easier layouts and multiple masks to compose chips. As I said, microprocessor and memory are already doing this. This is the only way they could deliver those nodes. This will work its way into foundry. It has to because it's going to be required to deliver these ground rules. That plays to our strength in building array masks, which is really a core competency for us. Of course, if EUV pushes out, this capital associated with EUV, that would get pushed out as well, and we believe node extensions and more sustained node usage will go on longer, which is a good thing, generally, for masks.

On the other hand, if EUV happens on a timeframe that people predict or a little bit later, not necessarily a bad thing for mask making either. There are about 10 ASML systems shipping starting this year, 2013. Full spec systems may be late to intercept this 10-nanometer node, but nevertheless, broad industry adoption of EUV at any time in the future has 2 benefits for mask making. One is it really does simplify the design flow. By that, I mean, if -- for every chip you design, it's very nice you need 5 masks to do 1 device layer. It's good for us. It's a multiplier effect. On the other hand, particularly in foundries, if the overall design flow could be simplified with EUV, single mask, single device layer, less RET, we do believe that will be a stimulus for generally more design activity in industry. So we're not afraid of the EUV, preparing for it, but nevertheless, as it gets pushed out, it creates opportunities for us in multiple patterning.

The other comment on the EUV, the masks do have a certain level of complexity to them, so the selling price or the cost of EUV masks will not be that much different than the multiple 193 mask. So the industry does not shrink either but you see EUV come online.

So specifically, what are we doing in EUV? Even back as 2009, we had a joint program with the captive. We led a commercialization effort there. We actually started commercially selling EUV masks in the industry back -- as far back as that. 2010, we actually had fairly healthy business in EUV, mostly the OEMs, these are equipment makers, because there were no chip makers doing it at the time. We established some pretty good relationships with OEM and equipment makers, you know their names, back as far as 2010. 2011, we expanded our program. Again, we're always watching the roadmaps of customers. We don't want to get ahead of where they need to be, obviously, we don't want to get behind either. We expanded a little bit in 2011, initiated more of an R&D effort, did more work inside our joint venture with Micron, and we started doing more of our own wafer print feedback through consortia. And then last year, 2012, we established our full baseline EUV process in our facility in Boise. Over 175 mask blank writes in the nanoFab. It seems like a small number, but for this technology and where it is in its evolution, that's actually a fairly robust learning cycle for EUV. So it's getting pushed out. We believe that's good for us. But on the other hand, it's not a technology we fear, and it will be ready to service this business when the time comes.

One advantage we have with our 193 platform and the way we develop it is we make sure as we're doing things, we also think in advance about the EUV need coming in the future, so we don't have to do things all over again. So particularly, I mentioned the way we validate mask from multi-patterning. Actually, much of that technology is applicable to EUV as well. We have 6 different inspection machines that are required to deliver full 193 multi-patterning set. We combine those sensors together to make sure EUV mask is good, and this is a very cost-effective way to do inspection for EUV. We have modeling tools that tell us what's happening in EUV printing as well. The first pass yield, I mentioned it's important for cycle time and multi-patterning, very important for EUV because the blanks and materials are so expensive. And finally, the lifetime and durability of the masks are very critical in EUV as well. And we'll work on that in 193.

I have a couple of slides on technology trends for flat panel display. It kind of covered IC masks, multiple patterning, EUV and how that creates growth opportunity. Through FPD, the technology drivers actually are somewhat similar, but they're just rewound about 4 or 5 years in terms of roadmaps where IC is today. So 2 things drive more complex masks for FPD. One is the pixel density. If you have the same size of display and you have more pixels, the resolution of mask has to be small to deliver those pixels. Just basic scaling. So as you go from 720p to, say, 2000p, same substrate size, the pixels get smaller and mask resolution has to get smaller. So that is much like an IC roadmap. The other thing that happens is individual pixels, in order to switch them, turn them on and off to make a display function, particularly for AMOLED, more complicated process requires more transistors.

So these are 2 things that drive mask technology, fundamentally, flat panel display. And this just shows a simple picture. If you look at old DVD resolution going all the way up to 2K and then 4K, as that size goes up, for the same size display, these pixels have to get smaller, so the mask feature has to get smaller. Then inside each one of these sub-pixels, the transistor, to switch it on and off from the transistors' switching mechanism, the backplane gets more complicated. So if you combine that resolution into more complicated backplane, you get much more complicated photomask, and hence, higher value brought to the mask-making step, which drives ASP and stronger value propositions.

What does that look like if you parallel it to the IC roadmap? This really means what we're seeing in FPD now is the need for so-called resolution enhancement technology. Roughly, believe it or not, where IC masks were about in 1998, just started, so standard mask making, you have a design pattern, print the mask, print the wafer. Now even for FPD masks, it would seem customers are needing to make proximity corrections, corrections to those to make the print well. So we are at the early phases of the resolution enhancement roadmap for FPD that for optical masks, for IC, it's really kind of in its final stages, multi-patterning and EUV. So we do believe there's a long -- a fairly long lifetime of increased value going into the mask for FPD as the displays get more complicated and the transistors also become a little more complicated to drive the pixel switching on and off.

So in summary, at least this technology section, the multi-patterning is at the early stage of at least a 3- to 5-year cycle, mostly driven by delays in EUV. This gridded design methods may push towards 3, 4, 5 mask compositions depending on single device layers. I think that's -- it's not a wishful thinking. I think if you look at the roadmap and extending 193 through the 14-nanometer node, there's just not many other ways to do it, if you're doing full strengths of devices. We have very good capability in this area, refined over 5 years of work, starting in flash memory back in 2008. I talked about registration and other things. We've invested in equipment that takes advantage of that capability and allows us to employ it in mass production.

And finally, FPD, at least from a technology perspective, mask technology is at the early stage of a value trend. Pixel size and the transistor complexity will drive mask technology, and we have to start doing things we did 5, 7 years ago on IC mask patterning, proximity correction and mask complexity, to deliver those ground rules. So it's a good story for FPD, the early phase of that trend, and then for multi-patterning, Photronics is really in a differentiated position to deliver technology that's going to be needed over the next 2 or 3 nodes, for sure. So that's my summary.

So with that, I'll introduce Peter Kirlin. Dr. Peter Kirlin.

Peter S. Kirlin

Okay. I'll begin with a summary of our high-end IC strategy. So the beginning of the strategy has its roots in the past, and Photronics started out in 1969, a small mill shop in Danbury, and we started out as a low-cost producer and a service leader. Over the next 30 years or so, we leveraged that strategy to become the #3 mask maker in the world with a large global footprint.

Around 2005, we made a decision that we wanted to migrate upstream, not just the leader in the mainstream but inspired to be the leader in the high end. So in order to do that, we formed a partnership with Micron. The JV was called MP Mask, was formed in 2006. 2008, we opened our own mask-making facility in Boise and really worked to leverage that technology into a leadership position in memory. Once -- whereas, as we worked to achieve that goal, as you heard Deno and Chris remark already, we migrated the technology out of Boise to, first, Korea and then Taiwan. And today, we're using the combination of the customer footprint developed in the early days coupled with a technology and manufacturing capabilities developed with Micron in memory and a local presence into the logic market. We're really chasing a profitable -- chasing the IDM's first, they're the most profitable logic mask, and then moving into the foundry segment.

So now if you look at this high-level strategy through different lenses, the first being time, as I've said earlier, in 2007, before the nanoFab was opened, we really only had a presence in the mainstream market segment. A year later -- or 2 years later, 1 year after we opened the nanoFab, we were supplying high-end flash and high-end DRAM photomasks to the market. Move forward to 2011, and we were actively qualifying MPU and IDM logic and starting our presence in the foundry space. And now as we stand here in 2013, we are supplying customers up and down the market here, whether it be logic, memory, high end, low end, analog, power, we're present in every market segment.

If we look at geography rather than time as a lens to view our high-end strategy, as I mentioned earlier, we started out in Boise. We moved then to Korea, and then finally, Taiwan. And I think as Chris said, we're in the process of installing 2 14-nanometer lines, 1 in the U.S. and 1 in Korea. They should be up and running by midyear. And by the end of the year, we'll have a very similar capability in Taiwan.

So of course, at the end of the day, we're in business to make money and the buck stops with the revenue profile. And as I said, 2008, we were just starting, less than 1% of our revenues were high end, that being 45-nanometer and below. We've grown that up to $37 million in 2010. This is largely all memory. And now 2012, we have a book of business, both memory and logic, $110 million, about 1/3 of our revenues, significant traction in the high-end business.

So beyond that momentum, what else gives us optimism that we're going to be successful looking into the future? This is a snapshot today. The first thing that really gives us optimism is the industry, not by our own doing, but by virtue of how the technologies evolving, is headed right in our direction. As Chris mentioned, multiple patterning was really first developed for flash. This is a cross-section of the flash array. It's now showing up in high-end logic, first MPU, but shortly, foundry logic. And of course, this particular process flow is periphery with the multiple patterning technology, which means multiple mask locals. It's really our core capability.

So the entire industry is headed our way, from a technology point of view, has good reason in order to generally be optimistic. The other reason to be very optimistic, we believe about our prospects, really is looking at where the money is being spent because CapEx spending doesn't completely correlate with high-end capability because there's some additional capacity in these numbers at lesser nodes but it's a very -- generally, a very good proxy, perhaps the best we have for where the industry is headed. So if we look at 2005, you see that between 80% and 85% of the CapEx was split more or less evenly between 4 regions, North America, Korea, Taiwan and Japan, a little more in North America, a little less in Korea. If we look forward to today, and this is IC Insights projections, that same 80% to 85% is being spent in 3 regions, the 3 regions where we're investing in leading-edge capacity, North America, Korea and Taiwan. And of course, Japan is really fading away as a high-end player. We just note that our 2 largest competitors are both Japanese, and historically, this has been their cash cow, their profit pool. And again, this has nothing to do with us, but it is definitely fate.

So if you peel this onion, these geographic onions, back one level further, what you see is there's only 6 companies that are projected to invest more than $2 billion in CapEx in the coming year, only 6. And of those 6, guess what, 3 of them is dark blue, our top 10 customers of Photronics, and 2 others, we have significant business with. There's only 1 out of the 6, and that 1 is TSMC, and they are fully captive where we don't have significant business. So these are the industry winners, clearly. And these are our primary customers.

So in summary, we were, we are and we fully expect that we will be industry's low-cost producer and service leader in the mainstream segment. This is our DNA. Secondly, we now are, without a doubt, a market leader in the high-end memory photomask market and we're gaining traction in logic. And finally, if you look forward, industry is headed right in our direction, either in the direction of our technology or in the direction of our key customers. They are, without a doubt, the industry's best.

So with that, I'll turn it over to S.H. S.H.?

Soo Hong Jeong

All right, thank you, Peter. Good afternoon, everyone. Let me give you a quick overview of our flat panel display photomask side. I will -- our FPD operation are located in Korea and Taiwan. These 2 region alone account for 80% of worldwide LCD panel capacity. We are located in close to our leading customer in this area. And this is strategically important because of the size of substrate maybe this is cost of shipping and the close engineering relationship at the high-end economics. Our leading edge capacity for FPD photomask was developed [indiscernible] in Korea and transition to the Taiwan. By working together with our leading-edge customer like Samsung, LG Display in Korea, AUO in Taiwan, we will maintain our technology leadership position in this market.

In terms of the overall FPD photomask market, last year was a really tough year. Industry was overworking [indiscernible] while -- and tablet sales were very strong. LCD shipment fell in year 2012. As a result, photomask market was down and we estimated by more than 25%. Our business was softer last year however, decline was not steep as the industry. As our high-end sales outperformed, we believe we increased market share to 25% in this year.

Overall, display market in year 2013 is improving. Our customer are focusing on premium product and volume are also increased. In the TV segment, the market is growing because the demand growth for the Smart TV and our customer also focused on large sized panel, over 60-inch, and high definition TV. Overall, market is very stable. In PC and the Tablet market, tablet demand is continuously growing. PC panel demand is stable.

Mobile area. Demand for the smartphone including AMOLED is growing. We are dominant position for AMOLED product because we developed technology together with our customers. Additionally, developing Flexible display is very much expected to continue to the benefit to us. Overall, Display market is much improved from year 2012. This was reflected in our first quarter revenue, which increased 25% over the Q4 last year and high-end revenue increased significantly by more than 40%.

Most specifically to the TV market, market demand become more stable. In addition to the success of Smart TV, High Definition TV and bigger screen TV are key driver of this segment. Both Samsung and LG continue to grow share in this TV market. Together they now have over 40% of market share. We are clearly at a beneficial position from this, our customer success.

AMOLED is one of the high-growth market segment as you can see here. AMOLED Mobile Displays shipment are actually growing to reach more than 300 million units in year 2015. Late this year or next year, new AMOLED TV will be developed. AMOLED TV set will be forecasted more than high-density LCD TV demand in year 2016. Our #1 customer, Samsung, has over 95% market share in this area. We are dominant position on AMOLED photomask supply. Recently, we are fully supporting AMOLED display for Samsung Galaxy 4 smartphone. The pixel density in GALAXY 4 is 441 dpi versus 306 dpi in Galaxy 3. Photronics was chosen as technology partner by Samsung AMOLED TV. Both Samsung and LG are investing heavily in AMOLED market. We are surely to be a -- their technology partnership for this market.

In summary, global display environment is improving in year 2013 and we are standing at very beneficial position on this improving market. Our prior investment continue to enable us to build more revenue. We will continuously develop technology together with our customer. We will be chosen for preferred FPD photomasks supplier by our key customers. Again, AMOLED, Flexible display and big size high-density TV will be growing and we are in a really beneficial position in this growing market.

Thank you, again, and I'd like to turn over to Sean.

Sean T. Smith

Thanks, S.H. I'm going to review our financials. Giving a little look recent historical perspective. Deno spoke about where we were with it and Peter talked about the opening of the nanoFab in '08, it's a significant investment, and take a look briefly at 2012, our fiscal year, quick look at 2013, give everyone an update -- business update on Q2 2013 and talk about some cash priorities, financial forecasts and key stats.

If you take a look at 2008, we are in net debt position of $140 million. We essentially had completed our significant investments spend at Deno as alluded to at Boise and our working capital, our EBITDA, was about $112 million, $66 million of working capital. And then as we open the nanoFab, financial crisis occurred, business slowed down and we were at downturn, took us a while to get out of it. And as you can see, each year, there was improvement. There's $181 million improvement from $140 million net debt in the end of '08, to $41 million in net cash in 2012. Working capital improved to $234 million from $66 million in 2008, as a result of all the hard work of the entire team, each site and location put into it is that -- the result of this proliferation and increased market share in the high-end. And looking at our intangible book value, it's going up each and every year including last year which "was a down year", as we said earlier.

This slide talks about our strong free cash flow. If you look to 2008, that's Q1, it's a trailing 12 months capital spend. That's the blue bar. So our cash CapEx, it was our cash CapEx that significantly exceeded our EBITDA back debt and as you can see, that's come down quite a bit. And as we got into 2009, those metrics turned around to prove since 2009 that we have generated significant EBITDA well in excess of our cash CapEx. And we expect that trend to continue in the future. That has also led to the improvement, obviously, of the balance sheet that Deno was speaking to. Taking a look at our 2012 results, again, I said earlier that it was a down year in the semi industry. We had revenues of $450 million down sequentially from a record of $512 million in 2011. We did see, as Peter talked to, yet again a high-end IC growth, 16% improvement. FPD was $100 million, which I believe S.H. you said it was a tough year, but that was the second highest amount of the revenue in our history in FPD. Despite the down year, we strengthened the balance sheet. Working capital improved by $25 million and net cash increased as well.

Last fiscal year, our EBITDA was $135 million and cash CapEx, $72 million. That still enables us to improve our balance sheet, that differential. Q1, we announced mid-February, it was -- we had forecasted it to be down, seasonally down especially with the IC a little bit slow. We achieved sales of $100 million. IC revenue was $74 million and FPD was $25 million. We were down sequentially on high-end IC, which we had expected and as S.H. spoke to, we had a significant improvement in FPD. Our gross margin is at 21.1% and operating margins at 5.2. And our net cash actually grew and again, sequentially down quarter from $43 million. Once again, our EBITDA well in excess of our cash CapEx on trailing 12-month basis.

Looking at our capitalization. At the end of Q1, we had $218 million of cash. We had $175 million of debt. We do not have any current maturities other than term debt, normal maturities big balloons coming due in fiscal 2013. Our largest payment is $22 million in October of 2014. So we expect these metrics to continue to improve. We do get asked when we meet with shareholders and investors, "What are you going to do with all that cash that you have, $218 million?" It's certainly a capital intensive business but this -- the cash that we have in our balance sheet and our financial flexibility we have provides us with the flexibility to pursue strategic opportunities in the marketplace to advance our leading-edge capability in IC and FPD. And these opportunities can come in many forms. We -- after reviewing these opportunities and ensuring that we can, as we talked about Peter and Chris brought in 2 tools, we looked to reduce our outstanding debt and build eventually to a net cash position of $75 million.

To give you a quick 2013 Q2 update, we're about 2/3 the way through Q2. We did project revenues to be up about 6% to 10% this quarter with EPS at $0.07 to $0.10 per share. During the quarter, as Deno talked to, we initiated the intent to acquire all outstanding shares of PSMC. We talked about -- we accelerate the delivery of 2 advanced e-beam tools from North America and Asia, which is going to significantly help us in the high end in 2014. We financed one of those tools to return capital lease, 5-year term. Business, generally has been tracking to our expectations through the quarter. However, we are continuing to monitor some softness in one of our key foundry customers in Asia. However, we remain extremely confident about our business model, that are sequentially quarter improvement for the remainder of 2013.

Just briefly to go through some of the Taiwan investment strategy or key statistics here. Over the last 2 years, we've increased our share from 57% to 75%. And Taiwan is one of the 3 critical areas in which we have been investing the last 3 years. nanoFab, Boise, Idaho, Taiwan, Frank talked about Hsinchu being in the epicenter, so did Deno, and in Korea. Financially, once we acquire, it's expected to result in, among other things, the potential elimination of minority interest expense. Minority interest expense that we have on our income statement consists of substantially all of the minority interest of the PSMC and that has been $7.5 million since 2010.

For 2013, we expect our cash CapEx to be in the range of $70 million to $90 million. In 2014, I'm sorry, it's 2013 -- in 2014, we will have 3 leading-edge tools strategically located in U.S., Korea and Taiwan to serve the high-end memory, logic and foundry customers. We don't believe our competitors have those strategic tools located in that geographic, same geographic footprint. We're supporting an enhanced demand for a high-end IC and FPD photomasks. As I mentioned earlier, we do expect and are confident that our 2013 EBITDA will be in excess of our cash CapEx spend. We'll generate strong free cash flow, yet again. And we're going to focus on 28 nanometers and below technology -- are focused on 28 nanometers and below technology that Chris talked to, mitigates our downside risk.

Some key stats that we have. Our net cash per share is $0.56. Our book value per share is $7.76, we're certainly trading at less than that and our price to sales is about 1x and fluctuates a bit. And our price to book is about 0.78x.

Our financial focus is to maximize our free cash flow; to reduce our cost of capital and debt; to continuously evaluate all costs for reductions and avoidance, including our manufacturing network and our support functions; of course, to invest prudently; and certainly to maintain our financial flexibility.

A few year -- 2 years ago, we established these financial goals and targets. We wanted an operating revenue -- I'm sorry, operating revenue growth, we wanted to exceed 10% per year. Obviously, last year was a down year. We started the year with a trough quarter, but we expect debt to improve sequentially. These are longer-term targets. We want to generate an operating margin of 17.5%. Q3 of 2011, we came very close to that. Incremental contribution margin and incremental revenue, we want that 50% target, that's a gross and operating line. We did achieve that in 2010 and 2011. Free cash flow, greater than $75 million. I believe in 2011, it was not over $90 million. And as I talked to you, net cash eventually to get it to $75 million.

So on our summary, again, despite 2012 being a down year, we remained profitable and improved the balance sheet. We're going to continue to execute on our high-end strategy, while servicing our mainstream customers. We do expect sequential growth, moving forward and bottom line improvement. And we have a lean and flexible business model with a target of 50% drop through.

And before we turn it over to Q&A, let me turn it over to Deno, he had some few final comments.

Constantine S. Macricostas

I would like to acknowledge my Board of Directors and the Director, George Macricostas, Joe Fiorita, Walter Fiederowicz, Mitch Tyson and LC Hsia, thank you. LC Hsia will give us direction and we'll be responsible to implement and execute. And for the team, you've done a great job explaining it better than I do and before finishing this conference or meeting, okay, not impress enough, you guys, probably the most cost-efficient and definitely provide the best service for the customers. That's been the DNA, the culture of the company. Now we're doing investing. We have the technology. We're investing very strategically in Boise, Korea and Taiwan and our competition is not investing. We're poised, not only to become the largest, next to the employers, but also to become a major force in the photomask industry. I believe that, and I can look you in the eyes, it's going to happen, not only the largest but a major force. We have the same goals for the FPD but we'll take it to 3 years together. On the IC, we're very close. We're investing for 20 nanometer and 14 nanometer and then up. But to make it happen, we need a team and I'm blessed to have the best team available of there is in photomask industry. They know the industry, they're passionate, they have enthusiasm and they like to win, and very cohesive so they we work together very, very well. They make my life very easy. I cannot tell you how that supports and indefinitely, we're together from [indiscernible]. That's my final remarks, and thank you for attending. Sean?

Question-and-Answer Session

Sean T. Smith

Yes, we're just going to turn it over for questions and answers and then bring in a team member.

Unknown Attendee

Maybe a question for Chris in terms of some of the opportunities particularly, multiple patterning and the eventual transition to EUV and you've highlighted some of the opportunities you guys have. In terms of the customer traction and what they're looking for, what are some of the key metrics that you guys are trying to supply for them? What are some of the areas that you're helping to accelerate their processes particularly on the double and multiple patterning front. What are some of the -- when you said listen to the customers, what are they telling you to do?

Sean T. Smith

Chris?

Christopher J. Progler

We polish [indiscernible].

Constantine S. Macricostas

Probably we should turn this. When you answer questions, it's got to be at the end of the mic. You can come up.

Christopher J. Progler

So I'll -- thanks for the question, Patrick. So I think what our customers tell us on multiple patterning is most of them have used 2 mask approaches over the past year to do cutting lines and that sort of thing. And the second mask, relatively simple mask. So one difficult mask, one relatively simple mask that can pose. Because of some of the delays in EUV in particular, now they're having to use multiple, much more complicated masks and they -- 2 things they asked us about there, given the masks are more complicated, how do we ensure, as I said during my presentation, we can synchronously get them into the retro fab at the same time and they work well together. We have to be sure that happens because as they come in and they're not meshed together well, we have a basic problem and then to reproduce that cycle is difficult. So doing as much as we can on our side to make sure before those masks leave, they're synchronized and they will work as a pair or a triplet or 4, is really what they're asking for. We don't want to find out, when we put them on a wafer production line that they're not married together well, if you will. The second thing is the prototyping process, the R&D process. Of course it gets very costly for the customer now because a number -- even to re-spin a single layer, redesign a poly layer or a metal layer, now 3 to 4, 5 masks. So since we're trying to test the strength and the efficiency in cost management, they really look to us to say, "How do we make that process more efficient?" So if we're doing prototypes, we have to re-spin the metal layer or contact layer. This is 4 or 5 masks. Help us make that more efficient, faster and more cost effective. So those are the 2 main areas that they have us focusing on.

Unknown Attendee

Great. Maybe a question for both Sean and Peter. In terms of the high-end IC strategy you guys discussed and some of the capital allocations and the investments that are acquired, it seems like you've done a good job of broadening your footprint in several of these areas yet at the same time, I also see that your customer base continues to expand particularly on the logic and foundry side. How do you manage against that continued customer penetration versus the investments that are needed particularly at the leading-edge, given that these are demanding customers?

Sean T. Smith

Peter?

Peter S. Kirlin

Yes, the way we manage that, Patrick, is really look down 2 trajectories. One is, in the last time we were together, you guys heard a lot about what we described as mix and match, and Frank mentioned it today. So one way we manage it is we try to localize the manufacturer of the critical layers in 1 or 2 locations. And then build the semi-critical and our non-critical layers in a close proximity to customer. That's what we described as our mix and match and it operated primarily between Taiwan and the nanoFab. But there was sort of a hidden agenda there that we didn't speak broadly about in the last time that we were trying to do, and that is the mix and match was really between Boise and Taiwan, focused on our memory customers there, the Micron and their affiliates. But magically, we were raising the capability of Taiwan in the logic space by doing that. So we're really, by using the mix and match, we paved the way for PSMC to accelerate the manufacturing technology so that they are now very close to being fully capable, staying on their own 2 legs in the foundry space. And unlike the memory space, cycle time -- cycle time is important across the industry at the leading edge but in memory and IBM logic, you get some latitude from the customers because as Chris said, they want the best foot [ph] on this as possible. The foundry is still -- it's cut throat, cycle time is usually important. So at the end of the day, to add that capability locally, allows us to separate ourselves from our competitors. So it's -- in sum, that's been our approach and our strategy and it's a work in progress but it's a tree that's clearly growing and putting out new limbs.

Sean T. Smith

And if I could, Patrick, we talked about on our conference call in Q1, those 2 advanced e-beam tools were originally earmarked to come in later. The team made a decision based upon what they see or what we see coming down the road. Chris talked about it, Peter talked about it, in that opportunity. So we have to pull those in. And a lot of time and effort goes in amongst the group on the capital allocation and what the returns are, but we're all confident that we will generate cash off those tools next year.

Unknown Attendee

In case of EUV [indiscernible] Photronics [indiscernible].

Unknown Executive

I'm sorry.

Sean T. Smith

S.H.?

Soo Hong Jeong

In case of EUV of 8,000. Photronics is the only company who has EUV of 8,000 especially in Asia, our competitor, nobody has EUV [indiscernible] so KLA will have EUV at 8,000 and that's why our previous customers, Samsung, asked us to sub with POR of 14 nanometer technologies. So in this is a [indiscernible], I believe.

Sean T. Smith

Come on out here. I'm sorry, we can't hear you. That only goes to the audio.

Unknown Attendee

I was wondering if you could talk more about the Intel arrangement about that 256,000 beam tool that you're working on. I was just wondering if you could give us what that [indiscernible] is and would that longer-term if you save on the CapEx side? Is that a possibility? And then second question is just about, I mean, actually, excitement, are you, in a more broader term, can you think about this as perhaps a point where you think you can reverse the trend of captive share and perhaps go into other direction where you start to gain some more share from the capture side?

Peter S. Kirlin

Let me clear the question. So the first part of it on that multi-beam program. It is at a fairly early stage although there is a equipment built and things are functioning well, it's hitting all its milestones. The technology does have potential to really change the way masks are written to go from a single beam to a multi-beam -- much, much faster, much higher throughput. And because of that, system is more complicated but not more complicated in a way that drives up its manufacturing cost tremendously. So if that platform comes to realization, we would see a dramatic improvement in what we call cost of ownership of the mask writing set. Very similar to draw a parallel with the lithography scanners. ASML, for many, many years had a scanner with a single stage and system had to wait for the single stage to move around of the wafer, wafer came out, they went to so-called twin stage that essentially allows the system to do multiple things at one time. So out of that 1 tool, much, much more throughput and the cost of the tool did not go up double, but the throughput almost doubles. Similar concept for multi-view. The second nice thing about it is the writing times are very predictable. One of the problems we have now in today's writing strategy is the writing time it takes to get the mask out depends on the complexity of the design. And sometimes we don't know exactly for sure until the mask is written. How long it's going to take. The multi-beam platform by design, it's a very predictable writing time so for queuing, multiple patterning, all of these things, a very, very powerful technology. But on the other hand, we don't want to overstate where the situation is now. It's encouraging, meeting all the milestones but it is still a developing program. In terms of a technology like that giving us leverage, have stronger relationships with captives, I think it's a piece of the puzzle. The major piece is as Photronics continues to gain market share and strengthen our technology and manufacturing network, naturally captives will look more and more to us as a partner. I think that it's naturally what happens. Second is, as we do bring unique technologies, that even though captives are very, very deep in that area, they do look at partnering for the purposes of getting access to that technology and that could be one lynchpin we use in that area. So I do think there is a trend when we can make a stronger value proposition for the larger captives in the future, as well to be stronger partner, pull back some of that business as they build confidence in us and we'll be there for the long term.

Unknown Attendee

Sean, I have a couple of questions. Number one, the market share number that you gave that's very helpful, on the 23% IC market share, how would you split between memory, logic and foundry?

Sean T. Smith

We don't, for competitive reasons, split that out. As Peter alluded too, our logic business did grow in 2012, but we don't split that up. That market share data includes mainstream as well, which we have good shares, significant share and Peter will be the one answer to that.

Peter S. Kirlin

The only addition I would make to that is we see the opportunity to grow dramatically at the high-end in front of us. And we're already a leader in the memory space in the high-end. So the bulk of the opportunity for growth for us is in high-end logic, moving forward.

Unknown Attendee

Got you. And then another question is that, your partner/customer, Micron, seems to be moving, [indiscernible] away from DRAM to NAND, what does it mean for the photo masking...

Peter S. Kirlin

I think as we expected, I guess that's right where [indiscernible], as the expected acquisition of Elpida closes, Micron will, and I think that, really, for them it's a point in time decision will, I think, manage their manufacturing capacity to be in best lockstep with their technology road map with one hand in market, market demand on the other. Because they can shuffle, they can shuffle the cards in more than one way. Regardless of how they shuffle the cards, that's the combination of the Elpida or the addition of the Elpida business to the Micron portfolio is, that's light field for us. That was Japanese business that is going to reappear as Micron, an affiliate business. So that's almost, there's no such thing as a golden ticket but that's about the closest thing you can have in this business regarding a golden ticket for additional revenue growth. Whatever it is, we'll get it. I can't say exactly what the mix would be but over time, we will most certainly get that market share.

Christopher J. Progler

Yes. I guess, one other comment I could make on, anytime there's a conversion of technology is that at any fab, it's good for masks because it means the set has to be retooled, so if any customers switches a DRAM fab to nan, that in addition to what may be volume, there's always an increase in demand because retooling any fab requires a lot of work so it's almost like a known transition, but actually more extreme. So we wish our partner the best in whatever manufacturing network strategy they might have but as fabs get retooled, it definitely drives multiple photomask demands.

Unknown Attendee

So then, for the whole team here, I would like to bring up some perennial industry questions. One relates to the share of the industry, which is captive versus merchant. And if you look over the past 5 years, has it been a trend in captive versus merchant in the industry? And within the 6 major customers, has there been any specific trend within those customers as part of a growth metric in the industry? So that's 1 perennial question. And then another is, particularly, now that we are going to double patterning our mask? This investment in mask is taking a larger share of total cost of building an IC than historically, is there a trend there? Is there an outlook there? So particularly for Peter and for Chris, those who don't often have a chance to talk with, that would be great to hear from you on those topics.

Sean T. Smith

Well, I think Chris already alluded to the fact that our strategy in the near-term is really to become the trust partner of the industry winners. I said it when Chris answered a question earlier, I think he alluded to that. And one of the reasons that, that's a key to our strategy is if you look at that CapEx slide I put up, everyone of those 6 industry leaders has their own captive capability. So there's no doubt that the industry winners are in the mask making business. And our strategy is to drive as much value to them as we possibly can. And that will, as Chris said, expand our share of that merchant market and maybe over the long run, if we're really effective, we can start bringing some of that captive business back to the merchant. So you're right, there is a shift occurring. We see that shift, our strategy has that shift embedded in it and we like what we see. So Chris, you?

Christopher J. Progler

Yes. No, not much more to add. When there are certainly fewer captives than there have been, I don't think that, that number will shrink much anymore though. I think that's been wrung out, but the fewer captives are larger and stronger and more heavily invested by their parent. So I think that's clear. However, we haven't seen any change in the inclination of captives to really want to work cooperatively with us merchants particularly in the case of Samsung, as S.H. mentioned. Always very, very strong interest and desire to have a growing and powerful relationship with the merchants so we learn from each other, it's harmonious and it's a partner arrangement so I think that's the key. But I think that the few left very powerful captives are not going anywhere. We have to figure how to partner with them to continue to bring value, continue to build confidence in them that they focus on their core, the technology is very, very early things, very, very enabling things and they rely on us more and more to manufacture things that -- for them. They've become, frankly, more routine but are still very, very leading edge. That's something to think about. On the double, triple patterning, that one, for sure, I mean, I think more of the value, and I see this -- at least the lithography flow, continue to flow into masks. And I think that is inevitable but that happens and it comes along with multiple patterning. As far as delivering an entire IC, still mask is relatively small, a part of that in total of a long running devices. [indiscernible] It's the cost to deliver a chip, the mask piece in that, for the memory chips, microprocessors still relatively small because of the number of wafer units we can build from that. There's a trend towards more of the share going into the masterpiece, and from the foundry side for sure.

Sean T. Smith

Yes, and the sort of -- at historical perspective, photomasks have floated for about 1% to 1.25% of IC revenues for the last 30 years or so, but there was a time when that curve diverged where mask niche-d up above 2%. And that was during the period when RET was implemented. So there was a major change in how chips were designed where the additional shrink in the line width was driven by the photomask not a step down in lithography wavelength. So in many ways, this multiple patterning, I see it as being similar to what happened with RET and that this all of a sudden, you need 2, 3, 4, 5 masks to do the same job that 1 used to do. So there's more value in that to the customer. Nothing else is happening, same litho tool. With lithography, as Chris said, NA is not going up, it's the same stepper or scanner, it's just now got a different set of masks built into it and we're only, we're just on the uptake, that's happened, as Chris said earlier is, it's been in flash but we're just on the uptake of that in logic. So that is a trend that I think we will see emerge rather than one that already has emerged and that is, as Chris said, good for us.

Unknown Attendee

If you really [indiscernible] question here, when you look at all the drivers you talked about, all the increase in flexibility, so you somehow quantify a little what that means, with [indiscernible] asset just importantly [indiscernible] partners [indiscernible]

Sean T. Smith

Chris, you may.

Christopher J. Progler

We -- depending on the -- you can talk about the number of mask layers and how that's going on for the device. I mean, it varies for a heavily double patterning sets you could see 30%, 40%, 50% more masks per set to deliver the same IC. For a more modest device or particularly a foundry part where maybe only the back end of the process shrinks and it's just a few contact layers, may only increase the layer count, 8% to 10%. So it varies from that point of view. The question is -- it's actually quite a complicated one because it varies by application and segment. To get a generic number of what the impact could be for double patterning, if we assume the layer counts go up on an average, 20% for mask set, then you can draw your -- draw some conclusions on the benefit that could have to the overall high-end segment on mask industry. But we would have to fill in, I think, more details before we give a really crisp answers.

Unknown Attendee

[indiscernible] a blended average [indiscernible] should go up [indiscernible] years.

Christopher J. Progler

Well, we do expect our high-end share, our percentage to continue to increase as a percentage of our high-end, of our total IC business. The reason that the 2 litho tools are coming in, and we said earlier, we expect to monetize those tools beginning in 2014 plus another high-end tool in Taiwan should provide growth for us in the future.

Peter S. Kirlin

Yes, and I think, and I would just leverage, I think the way Chris described it was I think the right way to think about it. And we're really explode right at the 14- to 16-nanometer node, at the 14- to 16-nanometer node, in a relatively complicated device, you see a number of layers expand by about 30%. And first order just scaled the price because if you look historically, and I go back to RET, we looked historically, what happened before, the prices effectively scaled. So without a lot of knowledge and yet at the 16-nanometer node, first, blush, I think that's a -- the best estimate we could give you.

Constantine S. Macricostas

To get the total number for the mask industry, I would have to say, that's a subset of total of tape outs of a certain node, the captives do some, the merchants do some, those are increasing in layer count and you get a, you have to get a blended number, which we can talk about that perhaps next time. But anyway, I think these are rough kinds of estimates. Edwin?

Unknown Attendee

Great. [indiscernible] any [indiscernible] why is that [indiscernible]

Peter S. Kirlin

I believe we said it was a agent foundry customer. We didn't get into further detail with 1 customer. We did say the business is generally performing well in other geographic areas so we're just monitoring it. We're not done with the quarter. So it's in -- we continue to -- we are turned just-in-time business, limited backlog. So I just want to let people know, we're look at something but generally, the move is upbeat for the quarter.

Unknown Attendee

[indiscernible]

Peter S. Kirlin

I think it would be fair to say that we're keeping an eye on the situation and we'll have better information as we close out the quarter.

Unknown Attendee

That's fair and then on long term, you mentioned that [indiscernible], what it's your advice to anybody [indiscernible] and what revenue level did you get that and when these happened, [indiscernible] growth driver [indiscernible].

Sean T. Smith

That's a fair question, Edwin. If we exit in Q1, let's say, to get to 17.5% operating margin, we would have to do in the range of revenue, $132 million to $135 million. If we -- remember back to third quarter of 2011, we have an operating margin at $136 million at 16.7%. And I believe at the time I said, to get to the 17.5% operating margin, we will need revenues of $140 million. So since that time, we pulled down some costs, it's certainly mixed dependent but the key drivers have been high-end logic, high-end memory, high-end FPD to get us there.

Unknown Attendee

[indiscernible] high-end margins [indiscernible] products just.

Sean T. Smith

It's fewer orders, higher ASP to drive the top line. And yes, once we get past, we'll talk about in the past, once we get it past our breakeven point, which is last quarter, $95 million to $96 million. It's target 50% to drop through and that drives gross margin up as we move forward.

Unknown Attendee

Great. And do you want [indiscernible] let him answer 1 question. What was [indiscernible] accounts versus the [indiscernible] performance of [indiscernible] one to [indiscernible] yourself and supply of [indiscernible] but what if you have a customer that says, "Hey, you know what it's [indiscernible] and it makes sense when you [indiscernible]. Does that makes sense [indiscernible]

Soo Hong Jeong

I don't foresee it in the near future like that [indiscernible]. So we're flexible depending with the situation. But you can't assume like you know who they are [indiscernible] if you're [indiscernible] and you invested me [indiscernible] really.

Unknown Attendee

So basically, you don't know the [indiscernible] you need to [indiscernible].

Soo Hong Jeong

Well, yes, we do believe both companies, especially Samsung, announced our [indiscernible]. The business model is to so-called invested enough towards [indiscernible] so called proprietary or anything that will boost our power. It happens, competition stealing and but that was [indiscernible]. And that's what PKL in Korea, it is our biggest -- by far, biggest customer. [indiscernible] but a few companies. We want captive, it's very big, it's the PSMC, are the biggest captive in the world and not outsourcing in 99% of the time. And that [indiscernible]. They keep everything inside.

Unknown Attendee

All right. You spoke about DRAM, is multi-DRAM a bigger opportunity than DRAM for your -- more masks involved in that? This is just the first question.

Constantine S. Macricostas

I'll let Chris come here and answer that.

Christopher J. Progler

Not necessarily in the flow there are more masks but it is more customized application compared to the server or commodity here and there more varieties and flavors of it. So that's kind of a little bit of an FPD style model in that. It's still DRAM but it gets customized by application. The reason that particularly is opportunity we're excited about partially that more design activity but the second is, it's not a space right now that our main DRAM partners got exposed to. This is one of the reasons they're engaged in acquiring, I think, Elpida. So to that extent, it's an opportunity for us because that's a new sub-segment of DRAM that we have not serviced to some extent but it could be a much, much stronger part of [indiscernible].

Unknown Executive

Sure. But [indiscernible] with another customer reform.

Unknown Attendee

Secondly, you mentioned GALAXY. Do you guys get any revenue-based [indiscernible] unit production successful product or is it more entertaining for the [indiscernible] steps in? How should we think about that, it seems like an analysis of a very good product?

Constantine S. Macricostas

Actually, they -- just to clarify to say royalty stream or are we somehow in the direct? Yes. I think...

Sean T. Smith

I think you're asking, Chris, the more units that they ship, do we get more revenue? Generally, no. It's one mask set. If they need a second set, yes. But it's still more smart design iterations or enhancements to existing designs coming out to just drive the top line.

Unknown Attendee

Just a follow-up question. The [indiscernible] merchants can be [indiscernible] installed it than the operating.

Sean T. Smith

I'll let Peter discuss that.

Peter S. Kirlin

Yes, the delivery of those 2 new tools is really imminent and it's -- it depends on the installation of them. it's 60 to 90 days. And in the case of the tool that's being sold in U.S., as Chris mentioned, within our JV, we already have one. We've had one for a very significant period of time, more than a year. So the manufacturing process behind it is already fully elaborated. So the qualification of that tool is almost instantaneous. The second tool we're installing is the first-of-its-kind, so the qualification we take incrementally longer. So one generates revenue almost immediately, the other would be a slight phase length.

Unknown Attendee

You mentioned that [indiscernible] at some level Taiwan.

Peter S. Kirlin

That's right. What I said, without being completely clear, was we'll have a tool of similar capability in Taiwan at the end of the year -- by the end of the year.

Unknown Attendee

How much does the yen have to fall before cost of your competitor is able to see [indiscernible]?

Soo Hong Jeong

It goes both ways. We're buying tools from Japan. We kind of -- to leave a quick customers this $40 because if you kind of benefit now being yet for paying yen, you get about 20% discount. It's not going to affect a little bit, still it helps the CapEx. If you make them compare and now the business is in Japan, part of the business is Japan, part is export to Taiwan. Maybe some in Korea. So some of the business that benefit because of the yen, but still not competitive.

Unknown Attendee

Put another way, how much cheaper are you relative to the [indiscernible].

Constantine S. Macricostas

Substantially, because we get money and that was...

Unknown Attendee

You said more than 10%, more than, 20%?

Constantine S. Macricostas

It cannot the exact estimate…

Unknown Attendee

And then you talk about that current yen levels…

Constantine S. Macricostas

And not all the business will get the benefit so any of these opportunities is 100% because it's the business [indiscernible] versus the yen, 94%, 78%, 105% [indiscernible].

Peter S. Kirlin

In the mainstream segment, our competitors are doing business locally. So the yen has, on the mainstream market, effectively no impact. It's only at the high-end it and as Deno said, even the high-end are not all of our competitors photomasks are manufactured in Japan. Some are made in Europe and others are -- and some are made in U.S. and some are made in Taiwan.

Unknown Attendee

When you look at the PSMC acquisition, buying out the other 25% and you compare that with buying your own stock, I guess, your conclusion was that, that's a better valued than buying back your own stock. I'm curious as to what your rationale is, is that correct? Why do that and then why not use the stock buyback given to you?

Constantine S. Macricostas

It's a good question but Mike, [indiscernible] stock [indiscernible] used PSMC's money to buy back the stocks. We expected that the money is usually is it's going to cost you money. If you take money, bring money to the U.S., you have to pay taxes. So in using the Taiwanese government's money to buy their own stock, they increase it from 57% to 75% and the additional 25% of course, you have to pay for it and then shuffle [indiscernible] for, but I believe it's going to cost about $25 million to $30 million. Buy $25 million to $30 million of Photronic stock, you're not going to get the same kind of impact as much as you make buy 100% [indiscernible] -- 100% subsidiary of Photronic if you do buy the remaining stock with Taiwan.

Unknown Attendee

Numerically, why is that the case? I mean, for $30 million you could buy, say, 8% of Photronic stock, which I believe your long-term earnings potential, that's us a pretty good value.

Constantine S. Macricostas

But if you believe that's what the epicenter of the account [ph] is. Also plan to heavily invest and capitalize opportunities in Japan, Malaysia, Singapore and China. So we believe it's more important this time Taiwan. Invest in Europe is not as important.

Peter S. Kirlin

Deno, if I may. It's not dissimilar back in 2001, I believe, Photronics acquired the majority control, I believe it was 50.1 at the time of the PKL, publicly traded company in Korea. And you saw the vision of PKL going up and we took them private, I believe in 2005 or 2006, back in 2001 when we made that acquisition, they were just starting on FPD. Now they're dominating Korea. So that's why we're talking about Taiwan, really, believe as a team is a great opportunity there and it's the epicenter, as Frank was talking about, for those customers.

Constantine S. Macricostas

If we do expressed it by Taiwan, you start 5 years ago because it's going to go way up.

Unknown Attendee

Just aside from that, I understand that rationale now but why is the stock buyback, just outright at PLAB just not make sense.

Peter S. Kirlin

I think that [indiscernible] is talking, we do have net cash of $43 million. We do expect to grow some cash but we see strategic opportunities in the marketplace that are going to put us in this position that Deno alluded to. It's that we're very confident, he's very confident about in the next 1, 1.5 years, the position we're going to be in. And hopefully that becomes more apparent in that point in time.

Constantine S. Macricostas

The consolidation, I don't -- I see a lot of things happening in the consolidation because it really [indiscernible], they lose money [indiscernible].

Peter C. Broadbent

Thank you, everyone. That concludes the Q&A.

Sean T. Smith

Thank you, all.

Peter C. Broadbent

Appreciate you all for joining us today. [indiscernible].

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.

THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.

If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!

Source: Photronics' CEO Hosts 2013 Analyst and Investor Day (Transcript)
This Transcript
All Transcripts