Novellus Q2 2007 Earnings Call Transcript

TRANSCRIPT SPONSOR

Novellus Systems, Inc. (NVLS)
Q2 2007 Earnings Call
July 16, 2007 2006 6:00 pm ET

Executives

Rick Hill - Chairman of the Board, Chief Executive Officer
Tom Caulfield - Executive Vice President, Sales and Marketing
Tim Archer - Senior Vice President, Dielectrics Business Group
David Smith - Senior Vice President, General Manager of Metal Interconnect Business Group
Kevin Jennings - Vice President and General Manager, Surface Integrity Group
Bill Kurtz - CFO

Analysts

Brett Hodess – Merrill Lynch
Harlan Sur – Morgan Stanley
Unidentified Participant

Presentation

Operator

Ladies and gentlemen, good afternoon and welcome to Novellus Systems Semicon West Analyst and press event and Q2 conference call. Ladies and gentlemen, would you please welcome Novellus' Chairman of the Board and Chief Executive Officer, Mr. Rick Hill.

Rick Hill

Thank you, everybody. Welcome to Novellus' annual Semicon event. Today we have a unique event. It’s actually simulcasted. We are actually having our conference call along with our Semicon show, so we have you listening to it live providing the laugh track for everybody on the phone. You unfortunately won’t be able to hear them but they will be able to hear us.

We have a full show today. We are going to talk about what the market outlook is. We are going to talk about some key product introductions we have and how we are doing in the marketplace with some really exciting new products.

In addition to that, we’ll actually have our earnings announcement, the conference call where we’ll take questions live both from the audience as well as from those listening on the phone. So I really do appreciate you taking the time to come here early and hopefully you’ll enjoy with us later this afternoon a little bit of our hors d'oeuvres and drinks after the presentation.

Before we get started, what I would like to do is introduce to you today John Hertz, our new Controller, if he would stand up. He just joined us. Welcome him to Novellus. He’s been with us now about a month -- a baptism of fire as always with Novellus. Always exciting.

With us, all the executives of Novellus who are in the stage. They are planted with easy softball questions for me. If you would all stand up, I would appreciate it. So everybody, if you want to ask them questions after the event is over, you are free to do so.

The second quarter, obviously we came in with an earnings number that was very, very good. We’re going to talk about that in some detail and we are going to talk about the overall market condition and how we see it going forward and what the drivers are. With that, what I would like to do is turn it over to Tom Caulfield, our Executive Vice President of Marketing and Sales.

TRANSCRIPT SPONSOR

Do you get frustrated during earnings season? Have you had trades go south because of bad earnings dates? We know what it's like. We’ve been there. We’re Wall Street Horizon and we work with some of the largest firms on Wall Street. Founded by former Fidelity Investments executives, we understand the power of trading on good information and the pain and suffering of trading otherwise. We obsess about earnings and economic events calendars so you don’t have to. Accurate. On time. Guaranteed. Let us help. Get Smart Get Wall Street Horizon. View our Free 30-day trial for investment professionals To sponsor a Seeking Alpha transcript click here.

Tom Caulfield

Thank you, Rick. Welcome to our annual Semicon event. Today I would like to cover some key themes. We’ll talk about the semiconductor capital equipment industry and outlook. It’s important for us to address the industry dynamics that are taking place in our sphere of influence here because it’s key to what our product announcements and what we’re trying to do to the company to grow our position. And then we’ll have a selection of our product families and products which come up and go into great detail about the products we’re announcing improvements on here at Semicon.

Without any further ado, let me go on to the market outlook. So let’s start with the semiconductor industry, this food chain. It starts with a macro economy, from interest rates to the feds, to government spending, interest rates and inflation and it drives what consumers and end users do in the electronics industry from a purchasing perspective. We’re talking about a $1.3 trillion industry -- what a locomotive that drives this. It’s growing at 5.5%. It represents 3% to 4% of the world economy.

By the way, there’s about -- whoops. Let’s go back one, please. This is about 1,000 to 2,000 brands out there, or companies producing these electronics and drives the semiconductor industry -- that’s $250 billion -- with 100 to 200 manufacturers. This semiconductor industry of $250 billion relies on a capital equipment industry that we play in of roughly $40 billion to $45 billion where 10 companies share in 75% of that growth, so it’s a very steep, inverted pyramid. If you were trying -- if you were in an industry that you are hooked up to a locomotive that’s running at 5.5% at $1.3 trillion that drives your business at $45 billion, it’s a pretty good place to be.

So the electronics industry is where it all happens. Let’s take a quick look at what’s happening on this historically. You can see from 1990 to 2012 that this is always a growth industry. In fact, only three times since 1970 has the electronic industry headed down here, and two of those were 2000 and 2001 as part of the dot.com. So this is an industry that delivers every year growth. If you look at what the consensus view of USI is, it’s growing at 5.4% in 2007 and over the strategic horizon, 5.6%, so let’s just call it the 5.5% growing industry that’s driving what the semiconductor manufacturers are doing and what the equipment guys are doing in response to that.

So if it’s so important, let’s take a look and break it down a little bit. I’m going to concentrate on the right side of the pie here, where we have computers at 35%; communications at 20%; and consumer at 11%. Now, let’s aggregate those three together because each of those have an element of consumerism in there. So even though we talk about computers, there’s the personal PC that’s a consumer-driven product that’s in there. Even though we talk communications, there’s cell phones in there that’s driving growth in the communications segment and obviously the 11% growth of the consumers, or the 11% market share of the consumer is there all along. And you can see computers is growing at 6.6%, communications at 7.7 and consumer at 8.7, the highest growth rate.

But make no doubt about it, the lion’s share of this industry is computers and for this industry to grow and to be solid, computers needs to grow. And that’s where we are today, by the way, with the replacement cycle for corporations. This replacement cycle got all synced up since the year 2K came in 1999 with the advent of refreshing the computers in response to 2K. Every three to four years, that refresh cycle or replacement cycle kicks in and in 2003, we went through a replacement cycle and right now in 2007, we are at the beginning of the replacement cycle. And the impetus for that is the Vista.

So we need the electronics industry to keep going at 5.5%. For that to happen, the computer segment has to be strong and part of that strength comes in the replenishment cycle for corporate PCs that we are about to enter.

The other two areas of communications and consumer electronics look like the following. So here’s electronics, the consumer electronics piece -- the fastest growing segment at 8.6%. What drives it is the digital revolution -- everything is digital -- our HD TV, our recordings. Everything that we own in electronics today we save in mass storage. I went on vacation recently. My daughter took 1,000 pictures, didn’t erase a single one of them and it didn’t matter because it was on a $30 flash drive.

The second element I think is really impressive. It’s the subset within communications called cell phone. In 2007, 1.1 billion phones will be shipped. One-sixth of the world’s population is getting a phone. On top of that, 50% of those will have the latest technology in 3G technology. The ultimate disposable electronic is our cell phones.

So tremendous growth in this, or tremendous opportunity in this big engine that drives the semiconductor industry, this $1.3 trillion business, growing at 5.5% and it will be $1.9 trillion in 2012.

So what does it mean for semiconductor revenues? Let’s take a look at the historical trends there. In the early 90s when this industry was growing at a much faster rate -- this industry being the electronics -- semiconductors were growing at 15%. Some time between the mid-90s and 2002/2003, that rate slowed down to 5.5%, mostly due to the fact that foundries became a real player, semiconductor foundries became a real player in the industry and the promise and the capability to leverage the large-scale manufacturing brought average selling prices down, which by the way allowed for this industry to continue to grow this electronics industry. Now we are starting from about ’03 to ’04, you can see we’re back on a growth rate in this, semiconductor 7.1%. Again, this is a composite or consensus view of SIA, Dataquest, and BLSI.

So while 2007 represents 1.8% growth for semiconductor, the tactical or strategic horizon growth of 7.1% predicated on the electronics industry growing at it’s 5.5%. The difference between these two is that there is more semiconductor content in electronics as we move forward.

So if that’s what semiconductors are doing, how does it drive where we play? Well, it isn’t so much how many dollars our customers ship that drives how many tools they buy. It’s the units of semiconductors. It’s the amount of silicon they put out. So if we take an historical view back to 1984 and plot through today, this is actually IC unit demand in the industry. You can see a couple of facts about it.

One, it is cyclical. It does not follow a straight line. It goes up and down. But more importantly, we can fit it with a line -- try hitting that slide one more time -- can we go forward a chart, please? There it is. We can fit it with a line that we’ll call the IC diffusion ray. And you can see it increases on a compounded annual growth rate exponentially of 10.3%, and that’s the rate at which unit demand is driving with this electronics industry that’s growing at 5.5%.

We could superimpose on top of that what the capacity looks like. You can see it by the blue line with the capacity. It has a harmonic and a modulation to it. You even see it going down in certain years when older capability or technology is retired and then increase in new capability, similar to what’s going to happen in 200-millimeter DRAM where it’s becoming economically not feasible to produce cost-competitive DRAM in 200-millimeter. Capacity is going to need to be converted over to 300-millimeter. Some of that capacity will get redeployed in the industry, some of it will just be retired.

But in any case, this blue line of capacity will always be ahead of the demand line. In fact, an ideal fab will run at 15% or 85% utilization rate because they need to respond to unit demand.

What happens in these cycles that we’ve been through is when we have this unit demand fall away from the capacity timeline it causes a pause in the add of equipment or capacity in the industry. As we get steeper up this curve, you can see an increment of capacity represents that much less to the total base and the steeper we get, we get up to where we are today in 2007.

2007 started, we exited 2006 where capacity was being added, unit demand was up. The first quarter capacity kept coming on. Unit demand came down. And in the first two months of the second quarter, you started to see unit demand start to bounce back up. Now we believe the historical cyclicality of this will continue to go on. We also feel that the fact that we are on a steeper part of the curve that the time for this imbalance between capacity and demand to close back to a point where fabs need to reinvest or continue to invest for capacity is getting shorter, so the cycles are less in amplitude and shorter in time and we are in that such mode right now.

So if this is what unit demand looks like, let’s take a moment to take a look at what cap spending is going to be. In here we have Dataquest, BLSI, Consensus View. Semi came out today and said 2007 is going to be 1% growth. The consensus view here says 2007 is going to be flat, with a strategic five-year outlook compounded out a growth rate of 4.8%. So if 2007 is flat, give or take, what does it mean to 2008?

Well, we have our own view of 2008 and what’s going to happen from a 300-millimeter perspective. We see 17 new 300-millimeter fabs coming on line compared to nine fabs in 2007. So let me take a moment to explain what that means. It doesn’t mean 17 new fabs get completely outfitted. It means 17 new buildings -- they may be part of a campus where other fabs, they may be greenfield fabs -- will start accepting equipment and begin the different phases of their ramp of capacity.

We also see 55 total fabs, ones already in existence today plus the ones we talked about at a line above of 17 new fabs, adding capacity. All said and done, 2008 we see about 700,000 wafer starts a month of capacity at 300-millimeter capacity being added to the industry, bringing the total to 2.8 million wafer starts per month. That would represent about a 30% increase in 300-millimeter capability in the build-out for 2008 and would be in line with the fact that this is a flat year to continue this growth at roughly 5%, the kind of capacity build-out we would need to see.

But having said that, that’s kind of our view of where the market is right now. We’re starting to see demand and capacity come back in line as this downturn starts to reverse itself. We see ’08 being a relatively strong year for 300-millimeter fab additions.

With that, I would like to segue it now to what’s going on in the industry dynamics because this is what drives what Novellus needs to do to grow its place in this industry, to grow its business.

We’ve already covered the fact there’s no end in sight for the digital revolution. In fact, this is what’s driving the consumerism. This is what’s driving the $1.3 billion train that’s driving this industry.

It’s come to a point though that the consumption is more and more for consumer electronics. So while it’s always been a paradigm for this industry that it had to address with cost, cost of ownership and to grow its market they had to take cost out of their system, it’s coming at a premium right now because of the pervasiveness of the consumer dimension.

But it’s happening at a time when complexity is creeping into the system. It is coming at a time where to keep up with Moore’s Law, we need to add complexity and there’s only one word for complexity -- it means cost. You see it in metal gate, you see it in the changes for dielectrics, you see it in stress films on devices to get performance. You see it in the advent of immersion lithography and the costs of doing critical levels of technology generation and generating. You see it in the transition of memory from aluminum to copper to take cost out of the system.

So we’re at a point in this industry where consumers are putting a tremendous pressure on our customers for costs and Moore’s Law is putting a tremendous pressure on adding complexity and costs. And the winners in this game are the companies that can produce the best capital productivity cost of ownership to their customers. Because let’s face it -- if you don’t have enabling technology, you are not in the door anyhow. Not that it’s not important but it’s no longer just enough. It’s more and more you have to deliver the value proposition of cost of ownership. This is an industry that’s playing to the strength of where Novellus was born and created this industry. This is where we play the best.

You’ll see later today when our vice presidents and general managers of a subset of our business units come up here, you’ll see how we are responding with our new product announcements right into the throat of this dynamic.

With that, let me summarize. Again, we saw where we are in this downturn, that the first two months of second quarter 2007, IC unit demand started to rebound and close the gap with capacity. If we take the historic rates, it’s the steep part of the curve we’re on. We believe the time for these imbalances to close is getting smaller.

On the upside end of this, we have 2008 that looks like it will be a pretty healthy year for 300-millimeter capacity adds at 700,000 wafer starts per month being added. That’s kind of an overview of where the market is going.

We have an industry dynamic that requires an acceleration in the cost of ownership improvements our customers need, capital productivity and enabling technologies. There’s no letting up on that.

What Novellus is doing is we’re responding. We’re responding across all our product lines but today we are going to focus on our Dielectric Business Group where our VECTOR platform is now leapfrogging to VECTOR Express, VECTOR Express to Ashable Hard Mask and VECTOR Extreme. We’re going after 30% productivity in films at less than 1,000 angstrom that represents a segment of the business that’s growing that fastest. We do an Ashable Hard Mask with a single application that’s growing faster than all others because of the number of critical levels and devices that require advanced lithography.

And we’re going with Extreme after the mega-fab trend for memory manufacturers who are trying to drive cost out of the system, putting large mega-fabs in place, trying to take the overall cost out of the equation with the VECTOR Extreme that’s going to produce for Dielectrics 250 wafers an hour type of productivity.

We were here last year. We introduced a tool called GAMMA Express in our surface integrity business group, or resist strip. This year, we’re here to talk to you about how well that’s done in the marketplace and the next enhancement with a high task force that’s going to take that tool model and with the least amount of silicon resist from a technology perspective but productivity up to 300 wafers per hour. And we’re here today to talk about our integrated metals business unit where we are complementing our already leadership in best-of-breed cost of ownership, capital productivity, with technology enablement by adding to our Hollow Cathode Magnetron technology something called IONX. This will allow us to do next generation technologies without sacrificing any of the cost benefit we’ve been able to supply to our customers. It will allow us to regain our position in this marketplace.

So without any further ado, I think it’s fair for me to say the Novellus product engine is in high gear. We have products coming out across our product line and with that, I will leave the floor to my colleague, Tim Archer, Senior Vice President and General Manager for the Dielectric Business Group. He’ll talk about the exciting things we’re doing with our VECTOR platform. Thank you very much.

Tim Archer

Thanks very much, Tom. My name is Tim Archer. I am the Senior Vice President and General Manager of Novellus' Dielectric Business Group, and it’s my pleasure today to be able to talk about three exciting new plasma-enhanced chemical vapor deposition systems that Novellus has launched into the marketplace over just the last several months.

We started in March of this year with the introduction of the VECTOR Express as the industry’s thin film productivity leader. We followed this up a few months ago, last month, a few weeks ago, with the VECTOR Express AHM tool as the best-in-class Ashable Hard Mask deposition system. And here at Semicon West, we’re launching the VECTOR Extreme as the industry’s fasted PECVD system.

Now you might wonder, why is it that Novellus, over such a short period of time, is introducing three brand new PECVD systems? Well, first, this is a large market opportunity for Novellus. In 2007, the PECVD market is $1.3 billion and because of new applications like thin films, Ashable Hard Mask, and the ongoing conversion of memory to copper, this market will grow by about 40% over the next several years to reach nearly $1.8 billion by 2011.

With the right products delivering the best capital productivity and the highest film quality, Novellus can grow faster than the market. And the market reception of the VECTOR Express since its launch in March shows that we are on the right track. We launched the VECTOR Express at the end of Q1 and at the end of this quarter, we will have grown the installed base of VECTOR Express five-fold and we can announce that we have received orders now for more than 50 VECTOR Express systems.

So why is it that the VECTOR Express has been so well-received by the marketplace? Well, first, the VECTOR Express was specifically designed to target a certain trend in the PECVD marketplace. As our customers race ahead to 65-, 45-, and even 32-nanometer production, there’s one thing that’s certain -- the films are getting thinner and the requirements for the quality of those thin films is increasing.

What you are looking at here is a typical 45-nanometer logic device, and what you see is that as we’ve moved to 45-nanometer, greater than 70% of the PECVD layers are now less than 1,000 angstroms thick. In fact, on several occasions here, you can see in the case of silicon carbide, an anti-reflective layer, some of these films are used in excess of 10 times to manufacture a single device. So the VECTOR Express was specifically designed to target capital productivity leadership for these thin films.

What you are looking at here is a chart that shows the VECTOR Express throughput improvement relative to our previous generation system. You can see that for those thin films like anti-reflective layer, silicon carbide and TEOS, we are delivering to the customer throughput improvement of approximately 40% compared to our previous generation systems.

And perhaps more importantly, this gives VECTOR Express a capital productivity advantage ranging from about 20% to as much as 35% relative to our competition’s latest three-chambered system. So today, we can confidently state that the VECTOR Express delivers the industry’s best thin film capital productivity. We think that this capital productivity advantage is sustainable, because the throughput improvement that we gained with VECTOR Express is founded on the unique architecture of Novellus' multi-station sequential deposition tool.

What I’m showing here is the graphical representation of the VECTOR Express system and a typical single-wafer processing system. What I would like to do is just take a moment to walk you through how wafers are processed in each of these respective systems, so that you can get some feel for why the VECTOR Express with “smart-soak” now has an inherent advantage in terms of thin film throughput.

So first, we start by loading wafers into both systems and that’s represented here by the blue circles. The first step in any PECVD process is to heat the wafer to the processing temperature and then prepare the surface for deposition. You can see that since this process is driven by integration requirements and film quality requirements, it’s typically identical regardless of the system that’s used.

But what’s unique about the VECTOR Express is that for the first time, Novellus has enabled the capability to deposit film on the three other wafers that are in the chamber in parallel to the wafer heat-up and surface preparation being done on station one. This gives us a tremendous advantage in terms of throughput for thin films and films where the wafer heat-up and surface preparation time is relatively long. In fact, for certain applications, we are already transferring our next wafer into the system where the competition’s tool would just start depositing film.

This ability to do parallel processing on the VECTOR Express is only going to become more important in the future. That’s because, as I mentioned, one thing is certain, is every generation the customer requires better and better film quality, not just higher throughput. So what we are seeing is a transition from what we call conventional thin film PECVD processing, where the wafer goes through a wafer heat-up step film deposition and wafer transfer, to now what we would call advanced thin film PECVD processing for 45-nanometer and beyond.

In this case, now what you can see is that film quality and integration requirements are driving the wafer heat-up and surface preparation time to be a larger fraction over the overall wafer processing time.

Just a couple of examples of that thin TEOS I showed in the 45-nanometer logic device, this film can be used up to 10 times. Defectivity is of course important and what you see is that we can improve the defectivity by about 4X simply by extending the time that we prepare the wafer surface for processing.

In this case now, you are looking at line-to-line leakage for advanced films, in this case, PECVD silicon carbide deposited on top of a low K copper stack. Now, by increasing the wafer surface preparation time prior to silicon carbide deposition, you see we can drop the line-to-line leakage by almost two orders of magnitude. Again, in both cases, the choice of the wafer heat-up and surface preparation time will be driven by film quality and integration requirements, not just by the customer’s desire to have higher throughput.

The fact that we can put this process in parallel with film deposition gives us an inherent advantage, an architectural advantage that is sustainable. The VECTOR Express with “smart-soak” was specifically designed to enable the meeting of these advanced process requirements without having to compromise productivity.

Now, it was on this high productivity VECTOR express platform that we then launched several weeks ago another significant advancement in terms of film quality and productivity, this time in the area of amorphous carbon deposition for Ashable Hard Mask. We call this tool the VECTOR Express AHM.

Now, this is a particular market segment where our customers are pulling for a higher productivity alternative to what they’ve had available in the past. The reason for that is if you look at this chart, every generation, the most cost-sensitive producers, DRAM and Flash makers, are being forced to use more and more layers of this advanced film to pattern their devices. You can see that in the instance of a 60-nanometer DRAM, they’ll use this film as many as six times. That will continue to grow generation by generation. That makes this right now the fastest-growing PECVD market segment with a 2008 market size estimated to be in excess of $300 million.

So it was into this market that we’ve jumped with the VECTOR Express AHM system and the value that we are going to deliver to our customers here is significant. In terms of capital investment, we estimate that we can face a 75,000 wafer start per month DRAM fab, $19 million in initial capital investment by selecting VECTOR Express AHM. This is simply because of the extremely high throughput of this tool.

The customer also building one of these large fabs gets an advantage in terms of footprints or fab space, and by choosing the VECTOR Express because of its high throughput, they can save as much as 30% in fab space.

Now, in selecting an Ashable Hard Mask tool, footprint and throughput are very important. But again, film quality is also what essentially drives customers going forward and we’ve develop specialized hardware for the VECTOR Express, which is why this tool was now known as the VECTOR Express AHM system, to enable and to meet the film quality requirements of the Ashable Hard Mask application.

Remember, this is the layer that is being used to transfer the pattern into the device and so therefore, its film quality is critical to the quality of the pattern transfer. So this film has to have high etch selectivity to enable the creation of high aspect ratio features used in memory devices. And it must also have very high lateral etch selectivity because we are asking this film to hold very tight CD tolerances throughout the etch. It must also, to be usable in a manufacturing line, be transparent to the wavelengths that the stepper uses to align to underlying layers, and finally and perhaps most importantly, it has to have low defectivity because any defect on this film can end up causing a defect in the pattern transfer.

So as I said, the VECTOR Express hardware was specifically modified to give us the best-in-class Ashable Hard Mask deposition tool in terms of film quality and productivity. The first change that we made was within the process chamber itself.

What you are looking at here is a picture of the plasma distribution inside a PECVD reactor when depositing amorphous carbon for Ashable Hard Mask with standard PECVD hardware. Our engineers have worked over the last six to nine months on new hardware that allows us to confine the plasma in the vicinity between the showerhead and the wafer, significantly increasing the density of the plasma used for depositions, which gives us better film etch selectivity, one of the key quality criteria our customers value, and this hardware drives the deposition rate up on our tool 5X compared to depositing the same film in a standard VECTOR Express, which is one of the reasons now we have such a fast, highly productive Ashable Hard Mask system.

But we didn’t stop there. We also incorporated into this tool very unique hardware to meet a specific Ashable Hard Mask integration requirement. And that requirement is that the film after Ashable Hard Mask must be passed to the next step without film on the edge of the wafer. The reason for this is film on the edge of the wafer can easily flake off, causing problems downstream, especially in a very expensive immersion lithography tool. So every Ashable Hard Mask tool today is generally asked to remove the film from the edge of the wafer in some way.

We took advantage of the unique VECTOR Express architecture to build into the outgoing load lock an edge bevel removal module. So as the wafer comes out, after having film deposited over the entire surface of the wafer, it is moved in close proximity to a protective shield and etching gas is then flowed over the edge of the wafer to remove the film. What this gives us is a very well-controlled, very well-defined region at the edge of the wafer where there is no Ashable Hard Mask film, exactly what the customer wants.

Now, this is also in stark contrast to the way in which our competitor has tried to meet this customer requirement. In the competitor’s tool, they have generally chosen to include a deposition exclusion shield, a shield to prevent deposition at the edge of the wafer. Now, the problem is you are putting something inside an already complex plasma deposition system, so you are just adding further chamber complexity and cost. It also limits the process engineer in terms of the types of films they can develop because when you have an exclusion shield inside your plasma system, they are prone to arcing at low pressure. But perhaps most importantly, because when you are depositing on the wafer you are also depositing on the shield, these shields must be cleaned very frequently in order to prevent flaking of the film off of the shield and on to the wafer, creating the defects that I specifically told you the customer needs to avoid.

And so with our Novellus EVR solution done after the deposition of the film across the entire surface of the wafer, we have solved the problems that have been plaguing our competitor’s tool.

So that brings us to the product announcement that we are making here at Semicon West, which is the VECTOR Extreme. This is a tool that we can probably claim is the industry’s fastest PECVD system. This tool is being specifically introduced targeting a trend towards larger and larger fabs as our memory customers try to gain economies of scale. What you are looking at here is the largest fab capacity announced for DRAM and NAND Flash fabs over the last several years. You can see that it was only a few years ago when the largest fabs for DRAM were in the range of about 50,000 wafer starts per month. Today, in 2007, it’s very common to have NAND Flash production fabs being announced and built with planned capacities greater than 100,000 wafer starts per month. So this has created a unique requirement for PECVD tools from these customers. They want PECVD tools that have the highest system throughput in order to minimize the number of tools they have to install in one of these mega-fabs.

So to meet that requirement, we have introduced today the VECTOR Extreme -- again, the industry’s fastest PECVD system. It’s a multi-module system, can be configured with up to three VECTOR Express chambers, so it incorporates all of those benefits that I just talked about in terms of the process chamber itself. But now, around one single handler, you can have up to 12 PECVD processing stations. That is 2X what our competition can offer.

And the benefits for the customer who needs this kind of capacity -- industry-leading throughput. Thin-film throughput on the VECTOR Extreme will reach 250 wafers per hour for thin film, and for thick films, because of the fact that we have two times the number of processing stations active at any one time, we will exceed 2X the throughput of the competition on thick films.

This leads for our customers to improved cycle time and fewer tools being required when building one of these mega-fabs. Just to highlight what that means in terms of numbers, for a customer wanting to build a 100,000 wafer start per month 60-nanometer DRAM fab, he can reduce the number of tools that he has to install for PECVD applications by 40% by choosing the VECTOR Extreme over the competition.

So with that, I’d like to summarize the three products that I just talked about. First, the VECTOR Express that we launched in March, 2007, the thin film productivity leader in the market with an architectural advantage for throughput that is sustainable. The VECTOR Express Ashable Hard Mask launched a few weeks ago into again the fastest growing market segment in PECVD -- best-in-class Ashable Hard Mask deposition incorporating very unique chamber modification, very unique edge engineering capabilities, with the edge engineering capabilities done outside of the process chamber but integrated on to the same system. And then, obviously, here at Semicon West, the VECTOR Extreme, the industry’s fastest PECVD system, up to 250 wafers per hour system throughput targeting the memory mega-fab trend. Thank you.

With that, I would like to turn it over to Dr. David Smith.

David Smith

Thank you, Tim. I’m David Smith. I’m the Senior Vice President, General Manager for the Metals Interconnect Business Group. Today I wanted to come and present to you our newest technology in the area of Hollow Cathode Magnetron and that is HCM IONX.

We introduce INOVA in 2000 and then in 2005, we introduced INOVA NExT. This built upon the INOVA platform, providing new capabilities, both from the standpoint of technology and productivity. Today I am going to announce INOVA NExT HCM IONX. This is the latest in our HCM technologies. What this does is provide extendible technologies beyond 45-nanometers -- I’ll show you some examples of that today -- as well as maintaining the production capability of the INOVA platform.

The market space that we are addressing is that of both the copper and the aluminum market space. We see significant growth in this market going forward. Today the market is approximately $700 million. We see that growing to slightly below $1 billion, almost a 30% growth in that market going forward toward 2011. And the growth is coming about really in two areas, the first area being the dramatic growth in copper, one in the area of the logic device -- that is to say, the amount of logic that’s being produced, as well as the number of metal layers within the logic device. In addition, the transition of memory from aluminum to copper is going to drive additional copper demand.

In addition, there are new market opportunities or market opportunities that we currently serve, one being aluminum. Aluminum will remain as the top layer within memory devices. There are also new fill requirements, new fill capabilities that we need to introduce into the market space and the HCM IONX technology provides this capability, specifically in the aluminum space.

In addition, there are new materials that are of great interest for the Interconnect. As we talk to our Japanese customers, they are interested in titanium technology for barrier applications and again, the HCM IONX provides this capability to them, so we see quite a bit of opportunity for us within this market space, a significant amount of growth for us.

What is HCM IONX? What it is it builds upon the core technology that makes up the INOVA platform -- that is to say, our HCM technology. HCM is a three-dimensional source, shown here in cartoon form. Really, the key is this high-density plasma that we can generate with this three-dimensional source.

This technology was introduced in 2000, as I said, with the INOVA platform and it served the industry quite well now for over five generations, starting at 180-nanometers. IONX is a technology that we developed over the last 18 months that brings this high-density plasma to the wafer surface. That is the core of the technology and the key points is we do this such as to achieve the type of film characteristics that we want but also doing it with a soft etch.

So HCM IONX enables a high-density plasma at the wafer surface while maintaining the manufacturability production worthiness of the HCM technology.

What does this mean for the actual deposition? I’m going to show you an example for copper seed. So on the left here is a micrograph of the as-deposited copper seed. We then turn on the re-sputter capability. This is now the first example of a copper deposition and re-sputter technology, and so what you can see here is after the re-sputter, what we are able to achieve is in essence a conformal PVD copper film.

From a plating standpoint, electrical standpoint, this is an ideal structure for film. You have a very thick copper film on the side and you’ve eliminated the overhang in this feature. This provides extensibility of the technology beyond where we are today.

For the barrier, the key is with logic, the introduction of ultra low-K materials, so here I show you an example of baseline technology where we have damage to the low-K material. What you see are fangs, roughness at the bottom of the trench. Now, using the HCM IONX technology, what we are able to achieve is very uniform trench bottom, very little damage, if any, of the ultra low-K, while providing a very deep anchor which is required for device reliability.

Also, since we are building upon our HCM technology, we haven’t given up anything as far as the manufacturability productivity of the technology. Here I show a comparison of the cost of ownership measured by cost of consumable and throughput for the INOVA HCM IONX technology measured against the competition. We have an over 30% advantage in cost of consumable relative to the competition. This is due to our long lifetime of our HCM targets, as well as the overall architecture of the system.

In addition, we have industry-leading defectivity performance, 30% lower on wafer defect performance for HCM. We’ve extended that with the HCM IONX technology, and it comes about because of our very simply single target source design. We have not compromised at all in this source design going to the IONX technology.

From an installed base standpoint, here I show the installed base for the INOVA product, starting in 2001, with a year-over-year growth, going now into 2007, both for copper and aluminum. The key message here is that we’ve seen continuous year-over-year growth for us in this market space and we continue to see growth for us next year going into 2008 and on into 2011.

So briefly, what I’ve hopefully shown you today is that we have taken the INOVA NExT platform, production proven, very high throughput, low cost of ownership, cost of consumables, industry-leading defectivity performance, combined this now with our HCM IONX technology to provide an extensible technology not only to 32-nanometers but we believe beyond 32-nanometers from a PVD standpoint. In fact, today we have eight customers that are running HCM IONX at multiple technology nodes.

At this point, I would like to transition and hand it over to Kevin Jennings.

Kevin Jennings

Thank you, David. Good afternoon, everyone. As David said, my name is Kevin Jennings, I am the Vice President and General Manager of the Surface Integrity Group at Novellus. The Surface Integrity Group is a family of products for photo resist removal, dry strip tools, sometimes referred to in Asia as etching tools.

What I'd like to do is give you an update today of how our Gamma Express product is doing, where we see the market and the technology heading, to give you a better feel for the overall dry strip business.

First let's start with the product platform. In 2001, Novellus introduced the Gamma 2130 platform. This was the first high through-put, multi-station, sequential processing dry strip tool; really changed the marketplace for dry strip tools.

In 2005, we built upon that. We added a very high reliability wafer handling system, changed the architectural framework a little bit for better defectivity performance and much higher reliability.

Last year, for those of you that were in the room, you know that last year at Semicon West we introduced the Gamma Express platform, a new high ash rate source for the Gamma product line, along with a more open architecture for improved serviceability. I'm going to talk extensively about this product, about our traction in the market, give you a little bit of a feel for where the technology is headed as well.

First, let's talk about the dry strip market. What is the market? How big is it? Well this year we think that the market opportunity for dry strip tools is about $330 million to $340 million. We see significant growth in that through 2011, to the point where this market will be about $510 million to $520 million in 2011, primarily driven by front end of line applications. So remember that as we move forward and we start talking about the challenges in this marketplace.

Our customers – remember, that for photo resist removal there's a lot of processing steps that occur and there are various steps in the manufacture of a device where the actual silicon and oxide of the transistor are exposed to the photo resist chemistries. One of the technological challenges that has emerged at 90 nanometer and beyond is really controlling the loss of the silicon or the oxide for high dose implant strip. You can imagine you have photo resist covering part of the device, and another part of the device that's left open to react with the ash and chemistries.

So what we're seeing is an increased trend towards concern about silicon loss and oxide loss. Tim talked about the importance of defectivity. Defectivity here in this case is not only important for yield, but also important from the perspective of minimizing the wet chemistries that typically follow a dry strip process. These chemistries are very expensive, there's a lot of effluent or waste that's generated, so our customers increasingly want us to try and manage the dry strip process so that they can minimize their wet cleaning.

This market, perhaps more than any other market segment, requires a very high reliability tool, so we routinely see requests for quotes coming in for tools with greater than 95% availability, sometimes 98% availability. So a dry strip tool needs to run very, very reliably.

In addition, no tool in the fab is perhaps as productive as a dry strip tool. It is not unheard of for our tools to run greater than 300 wafers per hour; very, very high productivity, high throughputs for our tools.

So what's happening is this market is changing a little bit. As we've gone below 90 nanometer it's become a critical application. There are some key technical challenges that are emerging in the marketplace that our customers are fighting with.

I want to talk to you a little bit about where we're headed. First let's talk about productivity. The Gamma XPR platform with the iSquared CP source, this is actual customer data, I haven't cherry picked anything. This is actual performance at a customer. Bulk throughputs of about 175 wafers an hour and high dose implant at about 35. Now why the big delta between those two things? Well, remember that in a high dose implant strip you have a highly polymerized crust that builds up on the top of the photo resist. If you heat that wafer too quickly, embedded solvent essentially boils and pops out and you end up with defectivity problems.

So you can see when it comes to high dose implant throughputs, the throughputs are much, much lower than the bulk throughputs, so you have to be careful when you're thinking about this market.

We introduced the product with about 175 wafers an hour and you can see the competition came out and introduced a new product with 250 wafers an hour and 50 wafers per hour in high dose implant strip. We then turned around and raised the bar again last year with the introduction of the Gamma Express. This is actual customer data at a very large flash memory fab, 304 wafers per hour on bulk and almost 100 wafers per hour on high dose implant. This tool is without a doubt the most productive dry strip tool on the market.

It's not enough in this market, as I talked about earlier, to have just productivity. They expect these tools to run, and run reliably. So what I've done here is I've plotted for you the first seven months of a large order we had to a Korean memory manufacturer for Gamma Express systems. You can see here that these tools went in and from the first month they've been running greater than 95% availability. Not only that, the volume of wafers that have moved through this tool is pretty staggering. They are running the heck out of these tools. In the first seven months of production nearly 2 million wafer passes have been run through these tools.

So the tool is productive, it's reliable; how about technology? Well, let's talk about the technology of the Gamma Express for a second. Remember that it's based on what Novellus made famous, which was the multi-station sequential processing, right? So I want you to keep in your mind two things. One, the in plant dosage levels are increasing and that's creating a very thick polymerized crust at the top of the photo resist. Two, customers are increasingly concerned about oxide loss and silicon loss.

Well our architecture enables us to do some things that a single wafer tool can't do. As the wafer moves through the system, we can change the temperature and chemistry, first to remove that highly polymerized crust and second to get through the bulk photo resist; and then finally, to clean up any residues that might be left afterwards.

Let me show you the proof in the pudding here. This is a very, very high dose implant strip, a boron implant level of 1.5e to the 16 from a North American memory manufacturer. What you can see is we're able to clean this with this architecture, changing the temperature and the chemistry as it goes through the tool with no residue, with very low silicon loss – less than 5 A -- very low oxide loss, less than one A and more importantly, very high throughputs, greater than 80 wafers an hour.

A single wafer tool couldn’t do that because they'd be changing the temperature in each station as they went through the system and it would take too long. It'd be a very slow process. This particular architecture, this multi-station sequential architecture really lends itself to the future of where we see strip heading.

As a result, I'm here to tell you that the product we introduced last year at Semicon West has had a great deal of market traction. We are winning head-to-head competitions today. In 2007 more than 40% of the 300 millimeter systems I ship will be Gamma Express systems. They are deployed in every region of the world.

We have a state of the art wafer handling system with very high reliability, greater than 95% up time. The tool is exceptionally productive, greater than 300 wafers per hour in bulk and depending on your implant levels, we can get up to 150 wafers per hour for high dose implant strip. We have shown technology extendibility to 32 nanometer and beyond, very low silicon loss, very low oxide loss. Not only low silicon and oxide loss, but the ability to do that in very, very high dosage levels.

We leveraged the Novellus' global infrastructure. This is something that my competition in this business unit can't do. Most of the competition in the dry strip business is a lot of small competitors that don't have the global infrastructure that we have. When customers are deciding to deploy tools in a fab in the middle of China, this is one of the things that they're concerned about. Novellus has that capability.

So thank you for your time. I'd like to introduce Bill Kurtz, our CFO, to talk about the Q2 earnings.

Bill Kurtz

Thank you, Kevin and good afternoon, everyone. It's good to see a lot of familiar faces in the audience today. Now we're going to shift the call to a discussion of our Q2 results and our outlook for Q3, so this is the financial portion of the call. I'll start by summarizing our Q2 results, provide the outlook for Q3, and then once I'm finished, I will then turn it over to Rick who will open up the discussion for Q&A and provide some closing remarks.

Now as you know, we issued a press release on June 28 that provided an update of our expected results for the quarter in which we indicated we would come in at the low end of guidance issued on May 31. Well, we ended the quarter with bookings of $332 million, which as expected, were down 19% from the first quarter and at the low end of our revised guidance range.

By wafer size, Q2 bookings were 75% 300 millimeter, 25% 200 millimeter, bringing our year-to-date bookings to 79% 300 millimeter and 21% 200 millimeter.

Second quarter shipments of $436 million were up 12% from the first quarter, which was approximately $4 million below the low end of our guidance range as we experienced some push out caused by absorption and excess capacity.

Revenues of $416 million were up 5% over the first quarter, which was in line with the mid range of our guidance. Second quarter revenues by geographic region were as follows: The United States was 29%, Greater China and Southeast Asia was 37%, Korea was 17%, Japan 11%, and Europe was 6%.

Now turning to gross margin, we increased gross margin to 50% in the June quarter, up from 49% in the first quarter due to more favorable product and customer mix.

Total operating expenses for the second quarter were $135 million, which is up $8 million from the prior quarter due to annual merit increases, higher profit sharing resulting from higher shipments, increased R&D spending tied to our key product releases, many of which you've heard today, and implementation costs associated with our global business structure.

Now our effective tax rate in the second quarter was up to 34.3%; that's up 3 points from the first quarter. The sequential increase is due to a change in the mix of international versus domestic income resulting from a change in our forecast for the year, which had a negative impact in the second quarter.

Now, comparing to the prior quarter of 2006, our effective tax rate has come down 2 points, mainly as a result of the implementation of our global business structure. We now expect our effective tax rate for the balance of the year to be somewhere between 33% and 34%.

Second quarter net income was $57 million, or $0.45 per share, which was at the high end of our guidance range due to revenues coming in basically in the middle and slightly higher other income. Excluding the impact of FAS 123 R, EPS was $0.48.

Now, turning to the balance sheet, we ended the quarter with a very strong balance sheet with cash just over $1 billion, including $148 million of restricted cash. Now while cash levels remained about constant, we did repurchase 1.8 million shares of our common stock for approximately $57 million in total at an average price of $31. That was offset by positive cash flow from operations of approximately $69 million.

Net accounts receivable of $422 million grew by $50 million in the quarter and DSOs increased to 90 days. The increase was driven by higher shipments and a slight increase in extended payment terms. Inventory levels, however, remained flat due to good controls.

Now with that as a summary of Q2, I'd like to turn my attention to our outlook for Q3. We expect bookings in Q3 to be flat, plus or minus 5%, compared to the June quarter. Shipments are forecasted to be in the range of $370 million to $385 million. Revenue is forecasted to be in the range of $380 million to $395 million. Gross margin is expected to be somewhere between 49% and 50%.

Now operating expenses are expected to decline on a net basis by about approximately $3 million as we've taken actions to reduce our spending through shut downs and other belt-tightening measures. That equates to about a $4 million savings. That's offset by some restructuring of about $1 million to bring a net decrease of $3 million in the quarter.

Earnings per share, including FAS 123 R, is expected to be in the range of $0.36 to $0.39. Excluding 123 R, we expect earnings to be between $0.39 and $0.42.

Now with that as a summary of our guidance for Q3, I'd like to invite Rick to come on board and provide opening comments and open it up for Q&A.

Rick Hill

Thank you. That concludes the formal remarks of the afternoon. I'm just going to make a couple of observations. I think some of the things you should see here today is (1) Novellus has been known for a long time for productivity. Productivity is the operative word today. When we look at a semiconductor industry that's continuing to try to drive down the curve of Moore’s Law, one of the levers to be utilized is certainly cost of ownership and the cost and productivity of the fab. I think we're well positioned in order to be able to provide that. You saw that today with the products that we presented to you.

But in addition, the thing you should notice is that what you also see is a steady flow of technology solving the most critical elements that are going to enable the semiconductor industry to continue down Moore’s Law geometrically.

So I think those are major things you should take away today. When we talk about the outlook for the market going forward, we made an announcement because we took preemptive actions. We wanted to give that information to the market at the same time. We had to inform our employees of our intended shutdowns during the third quarter and the fourth quarter, and we also had to inform some of our executives of our decision to cut pay for a period of time. Because we would normally announce this with an internal memo, we felt it only fair to give everybody the information simultaneously and not let that information leak out, sort of haphazardly, into the marketplace. Hence we put out the press release reiterating what we thought our guidance would be for the year.

As we closed the books, as expected bookings fell a little bit short of what we wanted to see or we anticipated seeing, as Bill spoke to you earlier about the forecast going forward, we're tentative about the outlook in the third quarter. The expectations are slightly below what we would have predicted earlier in the second quarter. So we wanted to give everybody that information at the same time.

In particular, it's driven by the memory, the pause in memory and NAND flash production capacity and demand mismatch. We don't expect that surplus to last long. Our best estimate is somewhere between a quarter to maybe two quarters that that over-capacity could still plague the industry. But we see, as Tom presented earlier in the presentation, strong underlying fundamental demand in this industry.

Novellus, as you've seen from these presentations, is really positioned to attack the high growth elements within the semiconductor industry. That interconnect is going to continue to grow. The demands on thin films, better particle performance, overall quality of the film is going to be greater and greater. We have the technology and we deliver it with productivity that's unprecedented in the industry.

So with that as always, I want to open it up for questions and answers. After the question-and-answer period, everyone who is here is invited over into the forum for appetizers and some drinks. We look forward to the casual conversation there. It's unfortunate for all of you on the phone, you can’t join us, but we know you will be here with us in spirit.

Speaking of which, I'm going to take the center seat. On my right is going to be Tom Caulfield, because if he doesn't get the bookings, we take him out on Friday afternoon and we string him up. And after he's dead, the holy ghost runs away on my lap. So we'll take Q&As and we'll take them both from the floor and from the phone. Thank you, first question.

Question-and-Answer Session

Operator

(Operator Instructions)

Unidentified Participant

… being up a few percent. Talking about the memory oversupply lasting no more than a couple of quarters. They are guiding September bookings flat, plus or minus. Should we take this as a sign that you see a positive momentum for your bookings in the out quarters, beyond September?

Rick Hill

Well, we never discuss beyond this quarter. You know, our ability to forecast the future is always limited. We try to do the best, we’ve tried to present to you some of the data we look at – not all of the data, there is additional data. There are external forces going on that we think have a positive impact in ’08. We’ve alluded to those from the standpoint of it is an election year, so you want to make sure the economy is going, no matter what party you are in.

Also, from a standpoint of the Olympics. Certainly China wants to display a lot of high tech capability at the Olympics in ’08, so there is still a substantial amount of spending there that is going to be driven; flat panel technology which is all going to be driven by semiconductor technology.

So I think ’08 stands to firm up. It also stands to firm up. It also stands to firm up because there's a lot of new product cycles coming out. We've got the iPhone, there's mixed results in the beginning but you also have coming out new computer technology with instant on, more solid state flash drives come out, which could lead to a very, very significant PC cycle, replenishment cycle.

It's also, while Vista has taken time to be absorbed in the marketplace, it's a freight train. You can't stop it. We still have not seen that impact on the DRAM market. So that's what leads us to the 2008 optimism.

Unidentified Participant

A real quick follow-up. You talked about a lot of new products here with higher productivity. Is your outlook for Q3 bookings, how much of that is embedding any significant share gains that might be uniquely helping your bookings in Q3? Thanks.

Rick Hill

Well, we'd like to get 100% share. It's just unfortunately not really likely because we just can't deliver it all in the third quarter.

I think that one of the things when you see this productivity -- and I think maybe everybody doesn't appreciate it -- but when you look at the productivity, Novellus has brought to the market, we shrunk a lot of markets. You look at the PVD market, we didn't have any PVD business before and it used to be close to a $5 billion business. Today it's a $1.5 billion business and we’ve shrunk it because we’ve taken a process that was doing 20 wafers an hour and a $7.5 million machine, and we've taken it to a $5.5 million machine and 75 to 80 wafers per hour; so it shrinks the market.

14 years ago when I joined Novellus we were doing roughly 30 wafers an hour with a system -- 30 wafers an hour. Today we introduced for you a system upwards of 150 wafers an hour with PDCVD with higher quality levels, better uniformity, better particle performance.

So from a standpoint of giving value, we are the value vendor for the customer. We believe we're going to win market share. But in that, we're shrinking the market simultaneously. But as you can see from our numbers, our relative position relative to the competition, we do continue to gain market share. There is some market share gain within the third quarter and will be beyond the third quarter. But it's hard to separate which is marked share gains and which is the turnout on the market.

Next question, please.

Robin Yim

Ladies and gentlemen, if you could wait to be recognized and a question microphone will be brought to you so our friends out on the phone can hear your question as well. Please go ahead.

Brett Hodess – Merrill Lynch

You talked specifically about some of the market sizes next year, but I was wondering if you could talk specifically about the adoption rate of copper in the memory market, NAND and DRAM, and how much that figures into the growth next year?

Also considering how cost conscious memory is, what predictions for how much that will allow you to increase your market share, specifically in metal interconnect given that your productivity is so high?

Rick Hill

Well, I'll let Tom take a swing at that one and then I'll comment.

Tom Caulfield

The question is on memory and the transition. 2007 is a key year because this is where the memory manufacturers are putting their pilot production in and doing their best of breed tool selection. So really, towards the end of '08 is where you'll start to see production buys in copper implementation of memory in a large scale. So I wouldn't say 2008 is a big year for that. It's the transition year where 2009 leads to that. There will be, obviously in our electric fill where we dominate the copper fill business, there's no reason to believe we will not continue our leadership position there.

We hope to pick up, as Tim pointed out, the dielectric business that comes along with that because it goes from a high density gap fill type of technology to a film PECVD film. Also, our PVD business for copper. So we think it's a big opportunity for us. '09 is the year where the conversion is fully after the transition. '08 is the beginning of that. This is the year of pilot line production and technology feasibility.

Rick Hill

Let me give a little bit more color too from the standpoint of when you look at memory, we've got to look at two pieces. We've got to look at NAND flash, we've got to look at DRAM. The reality is in NAND flash, the best NAND flash is already in copper, so that horse is out of the barn already. If you're going to compete in NAND flash technology and continue down the technology curve, you got to move to copper. I think we're well positioned to dominate in that particular area.

When you look at NAND flash, then certain companies have strategies where they want to basically be able to have fungible capacity. That says, if I'm making both DRAM and I'm making NAND flash and I've got to go to NAND flash in order to be competitive and stay on the technology curve, I've got to bring DRAM along faster in order to be able to have that fungible capacity. We think that bodes well for how we're positioned and we'll actually see that accelerate rather than slow down. I think that's our general consensus on that. Thanks, Brett.

Go ahead.

Unidentified Participant

Your older guidance appears much better than what the expectation was. It was probably partially held by our market share gain. In the last conference call, you commented on Peter Walters being better. Did that help your order guidance?

Rick Hill

Well, Peter Walters does help our order guidance. Peter Walters is the leader in providing polishing equipment for the raw wafer business. So we have a unique position of really understanding what's going on in the wafer polishing business and who is demanding wafers. So that also gives us information that helps tell us where the market is going and they're doing exceptionally well. So yes, that's a positive input. But no less positive than the things we've seen from PECVD, the things we're seeing from PVD, as well.

Unidentified Participant

So the Q3 Peter Walters order will be in line with what you had at the normalized level?

Rick Hill

We don't break out the orders when we give the forecast, but the Peter Walters orders are a definite contributor.

Unidentified Participant

Thanks.

Rick Hill

Mark, go ahead.

Unidentified Participant

Your shipments are holding up here, so can you help us understand how your shipments and bookings are translating into revenues? With two bad quarters here, two low quarters of bookings, does that mean still the revenue number has to drop further into the back half of the year?

Rick Hill

I'll let Bill handle that question.

Bill Kurtz

Well, as you know, we have different types of customers that accept revenue, in some cases within the same month as shipment, on a bifurcation basis and some customers that accept revenue on final acceptance; for example, in Japan. There are some dynamics that you'll always have to understand in the mix.

But as a general rule, I would suggest for you to go back and track the bookings for the last six quarters, subtract the shipments and the revenues and what you'll see is the delay between bookings to shipments and then shipments to revenue. Then you'll be able to see the fall off. That ultimately recaptures all of the cycle.

So right now, if you summarize where we're at, we've had a strong bookings momentum from September of 2006 all the way through to the current period where it slowed down. Shipments are now delivering against that and revenues are following the shipments. You can see it in the way the quarter-to-quarter trends line up.

Unidentified Participant

Just to follow-up, is backlog coming down at this point? I know you don’t give backlog numbers for the end of the year, but can you give us a direction?

Bill Kurtz

I think we commented last time that we're operating with what we consider to be a normal backlog at this time.

Rick Hill

I think that we're not trying to sidestep any question here. Revenue, given revenue recognition policy – and I've tried on the telephone conversation so many times to explain the multiple variations of how we recognize revenue. It's often difficult for us to predict.

If I had to take a guess, we're in a better position today relative to revenue recognition because most of the systems that we've shipped we've shipped multiple units to the customer and therefore we usually have a 90% recognition on shipment, 10% in 30 days. That's a lot different.

But then I've got to have the exception of if we get a big shipment in the quarter within Japan, that's all recognized on acceptance. So that might not be for three months to four months out.

So it's a complex equation. I think that if our bookings to shipment ratio remains in a negative, what I would call book to ship, less than 1 book to ship range, then ultimately revenues have to track down. They do. You just can't manufacture unless of course you were in the Internet business in the year 2000. Since we've never been in that business and we're not worried about that happening, our anticipation is as long as the industry stays solid, given our position with our products both from a productivity standpoint and a technology standpoint, we stand to win market share. We stand to basically grow at a faster rate than the market. We run a pretty efficient operation because we respond quickly to changes in the environment. Thank you.

Unidentified Participant

Bill, I heard an admission that results are much better than the press announcement that you gave on June 28, but I didn't hear an explanation why? So I'm just wondering what changed from June 28 to the actual results you reported today and why there's such a big variance or difference in revenue and earnings? I would have thought that was a little bit more predictable than bookings and shipments.

Bill Kurtz

Sure, let me just summarize again where we were on June 28 and the final results. As we commented, bookings were right at the low end of the guidance range, where we pointed them. The shipments fell slightly below and revenue fell into the middle of the range. The middle of the range on revenue is about two systems that were accepted earlier than we expected. It wasn't, I would say, a significant difference in terms of timing but two systems for us do make a quantifiable difference as it flows through the earnings.

The other aspect of the earnings results, I mentioned there was slightly higher other income. We had some non-recurring insurance settlements that came through at the end of the quarter that we didn't forecast so we're just being conservative on that.

So the real difference on revenue was two systems that got accepted that were planned to be accepted in July and came through in the end of June at customers' request.

Unidentified Participant

Thanks. Rick, how good can we feel about your order guidance for September flat, plus or minus 5? What kind of assurances can you give us that it's not just a forecast parking lot in Q3 that everything got pushed out to? How am I supposed to feel good about flat, plus or minus 5 in this environment?

Rick Hill

I think your question is really begging what's the industry going to do?

Unidentified Participant

I guess that, and maybe you could help us out on specific end markets and what's happening quarter to quarter in Q3 there.

Rick Hill

I think right now you've got some of the bigger memory manufacturers really analyzing capacity and demand curves. We anticipate those decisions to be made during the early parts of August on what they're going to do. That'll determine whether or not the third quarter comes in on the plus side or on the minus side, realistically.

I think from a standpoint of foundry demand, we see an increasing demand in foundries as the trend is inescapable; more and more people are going away from having their own fabs, IDMs, outsourcing it to foundries. We don't see that trend changing materially. I think those are the fundamental drivers.

One of the things when you look at the graph that Tom presented, we track pretty closely what that difference is between capacity and demand. We also track, given the absorption rate of ICs, how many quarters does it take until you get in a situation to where the available capacity isn't enough to meet the demand in the timeframe it takes you to bring the capacity back on? We actually take a deep look at how much capacity is already being installed and what's that impact on that line versus how fast that unit volume is going up?

We try to give you our best guess at it. It's not an exact science. I think we've been fairly good at predicting what's going to happen, but we've had some quarters where we missed it. But we're trying to give you the best information we possibly can.

Unidentified Participant

Would it be reasonable to model OpEx down another $4 million or $5 million in Q4 based on shutdown days and executive pay cuts in the back half of the year?

Bill Kurtz

Well, we're providing you with a forecast for Q3 which does include the effect of the shutdown and the salary cuts that we expect in Q3 that would continue in Q4. I would not expect an incremental impact unless we take additional action.

Unidentified Participant

Flat. Okay, great. Thanks.

Harlan Sur – Morgan Stanley

Maybe if we can step back, take a little bit longer term view, if I look at your customers that are making the technology transition from 65 nanometer to 45 nanometer and we think about the incremental capital spend as it relates to that technology transition, I'm just curious; what are you seeing or are you seeing any difference across the three areas -- logic, flash, and DRAM -- in terms of the incremental spend?

I can tell you guys real quickly that our analysis shows that the incremental spend seems to be a lot higher for the memory technologies given the more aggressive design rules, the more complex cell structures, probably less equipment reuse and a bunch of other things. So just your views on the incremental CapEx spend.

Bill Kurtz

I think one of the issues that the industry is facing is this capital intensity. How do you keep it in the range of 17% to 20% to make the economic model close, but at the same time there is complexity that keeps seeping into the equation and is driving it to higher levels?

But I don't think it has to be a constant. I think there's a period now where that's taking place and as people learn how to deal with that complexity, as they round out their capacity and keep adding on, they'll have to bring that level down.

Fundamentally, this equation has to close somewhere. If they're going to spend more on capital intensity and drive the depreciation costs up they're going to have to get the dollars from somewhere else. Is it going to be selling prices? Is it going to be reduced R&D? The equation needs to close.

I think you're seeing some of the harmonics that will come with capital intensity.

Harlan Sur – Morgan Stanley

Great. One more question for the team. One of the key themes that I think came out of here as it relates to both your existing products and your new products is a lot of differentiation, right? Throughput, productivity, defect density reduction. I think this is what the team has been talking about in terms of differentiation and value capture for your customers as it relates to at some point, the team being able to extract better margins from their customers.

So with all of these improvements, I guess the question is for you, Bill. When should we expect to see the team making progress back towards its target gross margin structure?

Bill Kurtz

Well, as we've said when we've discussed our road map for gross margin, it's clearly dependent on product differentiation. As you can see today, we outlined quite a bit of differentiation, so we remain optimistic that as we see the results of that differentiation in terms of future orders, bookings and then shipments that we'll see our target margin.

In terms of the specific timeframe, we haven't put a specific timeframe on it, but I think you can tell from the releases that we have some really good momentum underway.

Rick Hill

One of the things you have to remember and we've articulated 52% to 54% gross margin. When you look at a lot of our products, we enjoy the level of margins that we've historically enjoyed because of their productivity and because of their differentiation.

You also note that some of the product lines that we have, we published the gross margin of Peter Walters, which are somewhat less, so there's some depression from that level. You notice the SIG market. I think with some of the innovation that Kevin spoke about today, there's the ability to push that upward. But does that ever hit our corporate margin at that level? No, but we get the same returns because we don't require to deploy as many assets with that particular product line.

Our key is it's been fundamental to the company. You've got to have more cash coming in the door than going out. We were pretty excited this quarter by buying back stock and still having $1 billion in the bank. We're still committed to finishing the buy back that we have, which is now down to around $540 million left and get that done. We also have the financial wherewithal to make a bigger company.

One of the things that excites me the most on where we are today is I can say that our acquisition of a strip product line has been a success. We've clearly taken a business which was floundering, it's taken us probably a year longer than we anticipated, but that's hitting on all cylinders, we're winning market share and I think we can take a dominant market share in that particular business.

PVD, which has been a long slug in the teeth of a dragon has unparalleled performance with the technology and extendibility. I think one of the key elements to take away is with our IONX capability, we have extendibility down to 32 nanometers. That's something that the industry is really reaching for.

There are problems associated with atomic layer deposition, adhesion issues, reliability issues. I think what we've demonstrated is without giving up productivity, we can deliver the technological solution the industry needs. So we feel pretty good about what we have and where we are.

Tim?

Unidentified Participant

You put a slide up there that I think showed the capacity adds next year. If I take that 700,000 wafer starts a month of 12-inch capacity and convert that back to 8-inch capacity and I compare that relative to what I think is 6 million wafer starts a month of global capacity, that's an increase of 23% next year year-over-year; total capacity, year over year.

So even if orders were to get better in Q4, given that sort of capacity add, how long can that last unless you have a really bullish view on demand next year?

Rick Hill

Well, I think there's no question you've got to have a bullish view of demand. One of the things Tom pointed out in his presentation and I don't want you to be confused, the driver in this industry is going to be computer technology. If there's anything that everybody in this room is tired of, is how long it takes your laptop to boot up.

Now this year you are going to see a recycling of computers. The one thing you're going to get is you are going to get fast boot up. Flash is going to be an enabler of that technology. Lower power consumption is going to be another advantage with new laptops that will be cycling out here at the beginning of the school year and at the Christmas season.

Now if we're right, that's going to lead to growth in the computer business that's going to be greater than 10%. I've historically said a key to health in this business is a 10% growth in the computer business. When you've got that double-digit growth, you've got a great foundation for the semiconductor industry to play into.

Now, you throw on top of that the communications, the phone business and some of the changes that are going on there where you're seeing the convergence of various technologies. Whether it be the cell phone with an MP3 player, whether it be the cell phone with your PDA, the reality is there's lots of semiconductor consumption there that's going to go on. As I've said before, NAND flash is an application generator and you're going to see more and more applications. That's why I'm bullish long term on demand for semiconductors. It's insatiable.

Another factor here that you have to think about that in the short term -- and for those of you that read the paper yesterday and the Mercury News made the comment -- that in the short term there are some fundamental road blocks that are basically inhibiting everybody to move to 65 and 45 as quickly as we have moved from other generation technology.

But nothing has slowed down the creativity of the design engineer to slow their desire to use more and more transistors. To the extent you don't shrink the device, you make the die bigger. To the extent you make the die bigger, you need more wafers. To the extent you need more wafers, you have got to buy more spray painters, more polishers, more sanding equipment from Novellus.

Bill Kurtz

Let me just add something on that too. There is another dynamic on the 200 millimeter versus 300 millimeter. I was in Asia recently right at the low port of memory pricing and at least 16 customers were all emphatic that this was the last hurrah for 200 millimeter for DRAM, so there is some of that going on too. There is the retirement of DRAM 200 millimeter capacity and adding to the 300 millimeter capability.

I'll give you one other tidbit, I think we did the calculation this year, 300 millimeter capacity was north of 600,000 wafers per month added. So the 700 isn't that different from this year.

Unidentified Participant

Agreed. I guess when you look back in your history in this business, can you remember a year where you are going to add 23% wafer capacity and there has not been a problem?

Rick Hill

Well, I would say in '96 there wasn't a problem adding that level of capacity. Obviously the year where everybody had a problem was in 2000-2001, but you have to remember that the fundamental underlying demand was flawed. You were basically taking dollars from the venture capital community and from the public market and using it to buy computer equipment. You don't have that phenomena going on today.

You have the proliferation of product, real products that people want to buy. Is the market going to continue on upward forever? The answer's clearly not. But does this business, the semiconductor business, have growth rate that outpaces any other industry for the next ten years? I think it does. Can I predict it year by year, quarter by quarter? No. That's why you guys are here. You usually come and tell me what it's going to be and then I just tell you back what it is.

Unidentified Participant

Hi Rick, a couple of quick questions. One for Bill too. What was the percentage of memory as a percentage of shipments and bookings and then NAND verses DRAM within that? I’ll just give you all three and then I will go away. Second --

Rick Hill

You've already seen I don't have a memory that long, so let's go one at a time. The memory as percent of bookings. Do you have that number handy?

Bill Kurtz

We don't actually quote the specific percentage. I think we've said over time that we're fairly well balanced between logic and memory. It will, in the more recent periods it's been higher than 50%. But we mirror the industry in overall weighting between memory and logic.

Unidentified Participant

On the model, if revenues came down to where bookings are today and shipments stayed where bookings are today -- so kind of $330 million in revenue, $330 million in shipments in terms of your factory loadings -- what would your EPS be, roughly?

Bill Kurtz

What would our EPS be roughly? Well, it would depend on a few things. I'd have to really model the mix of the business and I'd have to model the expense stream. But if you look at where we are today, and I'll have you do the calculation and you can assume relatively constant gross margin rate, then you can have an estimation of what, if we didn't change expenses, what it would look like. Then the question is, at what rate would we adjust expenses further?

Unidentified Participant

Well when you said constant gross margin, if your loadings go down, does that change your gross margin or no?

Bill Kurtz

It has a small impact on us. But given that only 5% of revenue represents our manufacturing overhead, it's not a significant part of the calculation. So I'd suggest that you could model it off of the current revenues in the period and adjust it for that change in the revenue and keep the expenses where we forecast to Q3 and then you have a starting point. Then it's up. Then it's a judgment call as to what you would do on variable expenses from that point forward.

Unidentified Participant

The dynamics of the memory market that you're talking about are a little bit different from what Micron was suggesting last week at their analyst meeting in terms of CapEx. I don't want to put words in their mouth, but they kind of said, look. It is going to get soft and it is going to stay soft for a little while. We don't need that as much as we've been adding. That's a little different from what you're saying.

Rick Hill

Well, it all depends upon the seat you're sitting in. I think that it depends upon the uptake of NAND flash in the computer business; that's a big driver. You'll get a little preview of that in the coming six weeks. I like to be optimistic about it.

I think that from a DRAM perspective we've got about another quarter to a quarter-and-a-half and we have to see Vista have a material impact on the DRAM demand, or it's a longer-term issue.

But the other phenomena you have to remember is if there is an uptick in demand for memory and you don't have the capacity, you lose market share. Market share is a substantial power factor within the memory business.

You also have to remember that as we articulated, we do see a transition to copper in memory, and that bodes well for us. We are, I would think, fairly well-positioned and we continue to try to improve our position by improving the quality and the technological performance of our products to attack that market. So that's what sort of shades our thought process.

Unidentified Participant

This is mostly for Bill, I think it may be a little redundant with what Jim asked, so I apologize if it is. This last quarter, because your shipments were up so much and you also built inventory, is it fair to say that your manufacturing operations were running at pretty much full capacity?

Bill Kurtz

Well, I just want to clarify, inventories were flat quarter to quarter. Shipments were up, shipments were up but we certainly had additional capacity within manufacturing to serve more demand. We were not flat out at full capacity.

Unidentified Participant

There wasn't an underutilization charge or anything like that in the gross margin?

Bill Kurtz

The absorption line was very small as it would impact the gross margin line, given where we were on shipments.

Unidentified Participant

The next quarter, since your shipments are going to fall something like 15%, roughly, you probably will have some of those underutilization charges, I would think?

Bill Kurtz

We would have some, but as I pointed out, manufacturing has, over the years, implemented a strategy of relying very heavily on outsourcing modules and components such that we're doing final assembly and test. Now we feel it's important to remain in control of the final assembly and test, but that reduces our cost to about 5% of revenue for the final assembly and test.

Manufacturing overhead, if you model that out on a percentage of revenue, will not have a significant impact on gross margin; it'll have some, but it won't have a significant impact on the variation on gross margin. The more significant impact comes from product mix, customer mix and then growing differentiation.

Rick Hill

We also use contract labor within manufacturing. Some flexibility, to make it variable.

Bill Kurtz

To keep it variable, that's right.

Unidentified Participant

Receivables are up, I'm comparing December to now, something like between 30% and 40%. Can you just interpret that for us?

Bill Kurtz

Well, we have higher shipments that does drive the receivables. I also commented that our extended payment terms have increased as some customers have requested a little longer to pay and we've accepted that.

Unidentified Participant

One more if I can slip it in. Can you just describe how the profit sharing works? I think you mentioned that it was keyed somehow to your shipment levels.

Bill Kurtz

Yes, it is. We have a profit-sharing plan that is based on our target earnings model of 15% net profit after tax. It is geared towards our shipments P&L because that's the P&L that we believe the operation has the most control over and should be measured again. So as shipments go up, the profit sharing expense rises; as shipments fall, it goes in the other direction. So you will see a variation on profit share as shipments vary.

Unidentified Participant

Two questions, the first for Rick. Can you just give us an update on the CMP business and where that stands?

Bill, it was good to see you guys getting back in the share buyback this past quarter. What are your plans going forward? After three quarters of not doing anything, what made you decide to do it and what are your plans going forward?

Bill Kurtz

Well, we never disclose that. It's a good question. We have a strategy relative to buyback. All of you long-term investors, we think just like you do. That's what I want you to know. With that piece of information, you ought to always be able to anticipate what we're doing.

I think the CMP business is making some progress. Is it exactly where we want it to be? No. But I would have to say it's on an upward vector rather than a downward vector.

Bob.

Unidentified Participant

You put out a press release about Gamma Express and as part of your presentation made some bullish comments about it being a successful acquisition. When I look at the VLSI data last year you were three out of three or four serious competitors and the competitor who took the most share, most recently in the market seems to be on a cost based system as compared to a throughput based system; I'm referring to PSK.

Can you put some meat behind the statements you made here and put out? Do you believe you're gaining share in the market? Are you booking at a higher share? Are you booking up from where they were last year? Do you think you're in a number one or number two position in terms of bookings currently?

Rick Hill

I think you can imply from what we say that we feel we're on an upward trend. We stumbled when we came out initially with our XPR, had a reliability issue with one customer; we lost an order we shouldn't have lost because of that reliability issue.

But aside from that, absolutely no question, hands down winning and a majority of the customers, in a competitive situation. Even against the cost based PSK competitor, in their home turf.

Unidentified Participant

Given that they've been gaining share on a cost-based system versus a throughput system, you believe the market will be more accepting of a throughput based system?

Rick Hill

No question in our minds whatsoever.

Unidentified Participant

The 2.8 million wafers per month, does that include 200 millimeter?

Kevin Jennings

No, the 2.8 was 300 millimeter total capacity at the end of '08.

Unidentified Participant

Only 300 millimeter?

Kevin Jennings

Only 300 millimeter.

Unidentified Participant

How much north of 600K this year?

Kevin Jennings

Correct me if I'm wrong, 645K.

Unidentified Participant

How much was the number in 2006?

Kevin Jennings

You got me. I don't know offhand. We can get that to you.

Steve O'Rourke - Deutsche Bank

Rick, your largest competitor has been very vocal about some other growth avenues, like solar, less vocal about some other ones like maybe LEDs and solid state lighting. How do you view these as growth opportunities (1) for this industry and (2) for Novellus?

Rick Hill

Well, again, I am often wrong, never in doubt. The solar business, I always go back to fundamental economics. It's one thing, you guys can make a lot of money investing in stocks as long as they're going up and you get out before they go down or you short them before they go down and then when they go down you make money on both sides of that equation.

When you're running a business, it's a little bit different. You have to try to build economic value. One of the things we believe at the company is that in order to build economic value you've got to target markets that are economically sound.

Now, the solar business I think everybody accepts the fact that it's not economically sound. A consumer cannot go out and buy a solar panel and expect any type of reasonable payback. Tom was commenting, there's a whole bunch of people that they're going out and buying these things because it becomes a status symbol, if you got them on your roof at home; frankly, I think they're a little ugly myself.

The reality is maybe there's some people in Beverly Hills. I know there's a few politicians that think it's pretty cool, Al Gore, have them on your roof, have them on the head of your Prius, whatever, have your kid break them when he's drunk driving into the tree.

The reality is fundamental economics is what you have to be looking at when you're going into a business. When you look at solar, it wasn't fundamentally economical. Back in the 1974-1976 time period -- I'm old enough to remember that -- it didn't pan out. A lot of people lost a lot of money on it; a lot of people made a lot of money. They went out, a lot of hype, people took the companies public and somebody got left holding the bag. But I can't build a business on that.

If I had equipment that I could just sell without making any modification or investment to somebody that wanted to make solar panels, would I do it? Sure. We don't oppose it. We'll sell them CVD systems, we’ll sell them PVD systems, anything they need. But it wouldn't be something, because of the fundamental economics of the business, that I see going into.

Now you talk about LED lights as an example; energy conservation. Now that's a big boon. When you look at an incandescent bulb and you realize that 95% or 99% of the energy that goes in that bulb is dissipated in heat and you look at the fact you can put it into an LED and get the same lumens out without getting the heat so therefore you are reducing the power, those are significant things in the marketplace that can be game changers.

It isn't as though we don't look at those. We look at them all the time. Again, the economics are such that you got to wait a long time to get a pay back and consumers are pretty fickle. There's a few of us that will go out and buy iPods and play around with them and then take them back. There's a lot of consumers go and look at the price of a bulb and that's what they see. Until you can get that technology down a little bit lower, I don't think it's going to take off.

I think you're right to be looking at that market because I think it's getting close to being a market that could be very, very significant and one that we like. We think it's pretty good.

What were the other markets you want me to comment on?

Unidentified Participant

It was just those two.

Rick Hill

Just those two? I have a better memory than I thought. Okay. Thanks.

Unidentified Participant

With the PVD technology you're talking about, you mentioned extending PVD to 32 nanometers and possibly below. The other place where people have been worried about extendibility is high density plasma CVD. What do you think about the opportunity for that? Can that be extended to finer line widths? Are there some other opportunities for you there?

Rick Hill

Good question. HDP is a gap-fill technology. It was created initially to go after filling the bioelectric in between tight aluminum lines. It migrated itself down into shallow isolation because as the interconnect changed to copper, you don't need a gap-fill technology and so HDP started to be segmented into the memory market and into the FTI market.

The reality is HDP has some limitations. The good news is from a Novellus perspective as we see memory go to copper, and given the productivity of our PECVD films, we feel very well-positioned. So the losses we'll see in HDP we'll pick up within PECVD, plus there's additional layers required given the multiple etch steps that have to be done, dielectric type etches. So we think that's a good one.

Extendibility within FTI is a tough one. 65 nanometer is sort of okay, 45 nanometers, it's a real stretch and you've seen technologies like FACVD come into play and flow fill, but all of these are junk films. I'm not trying to trash the competitor. I'm just saying the reality. The quality of the film isn't what HDP is.

So if we can come up with a breakthrough to be able to extend that down to 45 nanometers, 32 nanometers, we're talking home run. Are we focused on that? You bet. It's got my personal attention on it, much to some people's chagrin.

Unidentified Participant

One additional question. You've been very vocal in the past about the opportunity for NAND flash spending in the second half of the year and NAND flash growth and opportunities. How do you see that today? Do you still feel confident about the NAND flash opportunities? Has the timeframe changed at all?

Rick Hill

I'm a NAND flash junkie and I want more people like this buying their kids $30 flashes to take pictures on. That’s what we need. I think NAND flash is, as I said, an application generator. I don't see it even close to saturation. I think we're going to go to another level of density. I think you're going to get chips stacking, which is really going to make it a powerful device. I think as it permeates the laptop, you won't buy one without it. I think the outlook is good. I haven't changed my outlook on NAND flash. A big market.

Unidentified Participant

Rick, good afternoon. Two questions. The first one is have there been any revisions to your internal forecast over the last couple of weeks or customer inputs that created your flat plus or minus 5% guidance for the third quarter today to be potentially better than it might have been two or three weeks ago?

Rick Hill

That's information we wouldn't disclose no matter what. It is what it is. You hear what we have today and that's what it is.

Unidentified Participant

The other one is the Dataquest market share for CVD had you down in '06. Is the expected share gain that you discussed today over the next year-and-a-half enough to offset or more than offset what appears to be a down trend in HDP CVD market share?

Rick Hill

We tend to think so. We've highlighted the fact that the one area where we have lost some market share that I articulated to Ed about, which he honed in on which is the FCI area with the HDP capability. As we see memory going to PVCVD we think that plays more to our strength. We haven't necessarily thrown away the HDP business either. It's still a big player.

Unidentified Participant

If you look at the bookings trend here, you pick up in the middle of last year; it weakens into the middle of this year. Is there a seasonality that you think is building into the business here in forecasting two quarters of downturn?

Rick Hill

There's a little bit of seasonality as you look at the frosting on the cake, when you look at the consumer business, providing ebbs and flow of demand around back to school and also Christmas. I think you have to back off from there to calculate when the capacity has to come online, when they start ordering again and do that analysis.

But I think there's an overlay of seasonality driven by an ever-shifting trend towards consumer. I understand the whole world, including China, doesn't necessarily celebrate Christmas but if you've ever been in China, you sure think they do. Everybody who's been in Japan knows they do. I think that is seasonality.

Unidentified Participant

I just wanted to follow-up on the comment about the memory customers being on their last leg of spending on the 200 millimeter factories. When do you think we can get that step function increase in the 300 millimeter factories? Or is that in your 2008 numbers? Do you think there's 17 new factories coming up?

Rick Hill

I'll let Tom comment.

Tom Caulfield

It's not going to happen overnight. It's one of the drivers that's going to accelerate 300 millimeter. Certainly that's one of the dimensions that feeds into next year's 300 millimeter build out and that 700,000 wafer starts per month.

Rick Hill

I'll be bolder. You cannot be economically viable with an 8-inch memory fab, period. DRAM. Just won't happen.

Unidentified Participant

Just a quick follow-up for Bill. The tax rate is a little bit higher in the second half the year, is that simply a function of more sales coming to the North American region? Are there high regions or is there some other reason behind that?

Bill Kurtz

Yeah, it's a comment that I made during my talk. We had a forecast during the year between the mix of domestic versus international. That has changed from our earlier forecast, which as a result we have a slightly higher proportion of income being domestic versus international where we have a lower rate. So yes, it has gone up, as a result of changing that mix. We do expect that we will see a reduction next year consistent with our expectation as we outlined the change in our domestic and international structure.

Unidentified Participant

Rick, if you look at the foundries revenue and you weight it by technology node, if you just take the top four foundries and you weight what their sales are by node, there was a pretty continual downward trend in that, as you would expect. The last five quarters, it's dead flat. Something seems to be going on. Maybe it's temporary, maybe it's a structural change, but it seems like the incremental unit being demanded is a lower-end device and so that's what's driving a slowdown in their shrink rate.

Do you think that's the right conclusion? Do you think it's structural or just cyclical?

Rick Hill

I think that there's a fundamental issue on how many different designs can afford 45 nanometers as an example, 65, 45 and beyond. There becomes a cost element. There's no question in my mind that fewer and fewer devices are going to be able to support the non-recurring charges, design costs to design at those nodes.

From a standpoint of as you go farther down the node, how many different designs are going to agglomerate at those levels? I think you're going to see a growth in programmable logic arrays, your Xilinx, your Altera’s are going to become a more and more viable entity for two reasons. One is the nonrecurring charges they can amortize over a larger volume and with the higher density on the devices, they can put more functionality in and make them more applicable to more design engineers and still get the performance out. So that may be a timing difference that you see. You haven't seen some of the more advance designs in those particular areas being picked up.

You have to remember there's several phases of this. They have to create it, they have to make it available. Designers have to pick it up and use it. So those things take a longer period of time. Hence the reason I think we're slowing in some of these transition zones.

But I think clearly at 90 nanometers and even at 65 nanometers, 65 is the sweet spot, we think. We think there will be a lot of demand move at 65 nanometers because it's a big economic gain for most people with any kind of volume. There are going to be multiple dies on a wafer that the foundries are beginning to agglomerate so they can give their customers some value in doing that.

But as we go down to 45 and beyond, I think the phenomena will only lengthen itself to what you see, smaller and smaller. It'll take longer and longer to absorb those capacities. It gets harder and harder to design at those levels there’s a fundamental issue, but it's reality.

Unidentified Participant

I just needed clarification on your EPS number with and without FAS 123 R. Most companies when they talk about with and without stock option expense, I think you are not excluding the restricted stock. I have difficulty reconciling the numbers you gave out. Is it true that you're not excluding the restricted stock expense?

Bill Kurtz

Yes, what we've done is we've shown you the impact of implementing FAS 123 R which required the expensing of stock options. We had been expensing restricted stock prior to that. So the total impact of all stock comp expense is greater and you can actually see the impact of that in a schedule that we put out in our press release, if you'd like to see how the total stock comp expense gets captured which includes restricted stock and the expensing of stock options. But the answer to your specific question is when we show excluding FAS 123, it does still include restricted stock in there. We don't take the restricted stock out.

Unidentified Participant

Right. Your number appears worse than what other –

Rick Hill

That's correct. That would be potentially correct. We don't know how they do it.

Bill Kurtz

But you can get to a total exclusion of stock comp expense by taking what we detail on our press release, which includes all forms of stock comp expense.

Unidentified Participant

Could you provide us with a qualitative as to how the Peter Walters business is doing and how you expect it to do going into 2008?

Rick Hill

Peter Walters, again an acquisition that we're very, very happy with. The business is growing nicely, both from a standpoint of the industrial segment as well as the wafer polishing segment. We articulated they were very, very strong from a bookings standpoint in the first quarter. Our outlook for them is pretty positive.

But again, they're not a major part of the company. We break it out in our 10-Q for you so you can get all of the detailed information. I don't have it at my fingertips to divulge to you, but overall the business is very, very good. We're very, very happy with that acquisition. It's one that we interestingly enough, use totally with debt. So when you look at the return we're getting on it, it's actually pretty good.

Unidentified Participant

Given that it's margin dilutive, looking at your guidance for the next quarter, I was frankly surprised a little bit by the strength in the gross margins. Could you give us some sense of how the other products are doing in terms of cost reduction and a couple of specifics perhaps of where you are seeing the improvement in the cost structure and where that can go in the second half?

Rick Hill

Well, when we look at operations, we're constantly focused on the cost structure. We took preemptive actions with what would be a relatively mild change in the business in order to make sure we can maintain the type of model we want to maintain on the financials.

You can see from the products that we've introduced there's a tremendous amount of value in the products to the customer. We're on an upswing relative to what we can offer to the customer and the value we can deliver. So that also helps us from a pricing standpoint. Product mix, to the extent we get favorable product mix -- bookings, shipments and revenues -- that number jettisons upward from where we are.

But we're constantly focused on costs, material cost reduction. We articulated we don't have a lot of labor content other than the labor from our outsourcing that we're continually working on, but it's unrelenting.

The same thing is true with operating expenses. Because the operating expenses, we say they're flat, that doesn't mean every day somebody isn't in my office or Bill's office having to justify why they're spending another nickel. We're pretty hard on operating expenses. We're constantly looking at R&D programs and looking at programs that don't look like they're going to pan out and get the kind of return that we want. We're cutting those to make sure that we're anticipating where the market is going.

Frankly, from a standpoint of both SG&A and R&D, they're too high for my satisfaction. There's constant pressure to justify why those numbers are where they are and if we're not going to get a pay off on the R&D, why are we doing the programs? We're continually going through that process. That's a way of life in our business.

What we try to do with these sessions is interact with you to give you, in the best way we possibly can, how we think about the business, where we think we're going to go with the business, what's driving the business. The thing you can be sure of is we believe in one thing, and that is we have got to have more cash coming in the door than going out the door. If we start not being able to do that, then we're not creating economic value, then we're not doing what we need to do. We're focused on creating economic value, we've got an active buyback program, we're actively searching for new businesses where we can add value as well.

With that, I would like to welcome all of you over to the forum for some drinks and some hors d’oeuvres. Thank you very much for your attention here today and we look forward to seeing you at the next quarter. Thanks.

TRANSCRIPT SPONSOR

Do you get frustrated during earnings season? Have you had trades go south because of bad earnings dates? We know what it's like. We’ve been there. We’re Wall Street Horizon and we work with some of the largest firms on Wall Street. Founded by former Fidelity Investments executives, we understand the power of trading on good information and the pain and suffering of trading otherwise. We obsess about earnings and economic events calendars so you don’t have to. Accurate. On time. Guaranteed. Let us help. Get Smart Get Wall Street Horizon. View our Free 30-day trial for investment professionals To sponsor a Seeking Alpha transcript click here.

Copyright policy: All transcripts on this site are copyright 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.