科林研發 (LRCX) 2002 Q4 法說會逐字稿

Good afternoon. I'm Kathleen Bela, Director of Investor Relations & Corporate Communications for Lam Research. Welcome to our 2002 analyst meeting.

We will begin this meeting with a review and discussion of the financial results for the quarter ended June 28, 2002. Following the financials, management will present an update on our outsourcing products and a product and market positioning update. We issued our June quarter financial release over business wire at approximately 1:00 p.m. Pacific Daylight Time. If you do not have a copy of that release, you can access one on our Web site at www.lamrc.com, or if you are here with us at the Westin in San Francisco, just raise your hand we will bring one right over to you.

Here today are Jim Bagley, our Chairman and Chief Executive Officer, and Steve Newberry, our President and Chief Operating Officer. Mercedes Johnson, our Chief Financial Officer, is at our headquarters in Fremont, California. Steve Newberry will review the June financial results and provide guidance for the September quarter. Following the financial review and corporate presentations, there will be a Q&A session. Please use a microphone to ask your question. Not only will we be able to hear you, but those who are listening on the conference call and the Web will be able to hear you as well. Please state your name and company name before you begin.

This meeting is scheduled to last for one hour - one-and-a-half hours. Excuse me. And as a reminder, today's conference call and the proceedings of this meeting are being recorded Tuesday, July 23, 2002.

Now, we will be talking about the future during this meeting. And as you know, the future has a habit of being unpredictable. So, please note that except for historical information, the information Lam is about to provide and the questions Lam answers during this meeting may contain certain forward-looking statements, including, but not limited to, statements that relate to the company's future revenue and operating expenses, particularly as they relate to our outsourcing strategy, management's plans and objectives for future operations and product development, our market share expectations, the demand, acceptance and competitiveness of the company's products. These statements are subject to various risks, uncertainties and changes in conditions, significance, value and effect that could cause results to differ materially and in ways not readily foreseeable. These risks are detailed in our filings with the SEC. We encourage you to read these reports in their entirety. Lam would also like to disclaim any obligation to correct or update any of the information we are about to provide.

And now, it is my pleasure to introduce our President and Chief Operating Officer, Steve Newberry.

Thank you, Kathleen. I guess in listening to the Safe Harbor activities everybody pays a lot more attention to, especially in today's environment.

Good afternoon. I'd like to thank you all for inviting us here today. This afternoon I will cover the highlights of our new orders and shipments momentum and briefly analyze our P&L and our asset management.

New orders for the quarter of $234 million represented an increase of 41 percent when compared to March 2002. Incoming orders for the three-month period met our expectations of last April and were driven by increased capital investments for leading edge capacity. Orders growth was above average in Asia Pacific and Europe and partially offset by declines in North America. Following the new order trend, shipments of $198 million rose by 62 percent past mark, resulting in a book to bill ration of 1.2 to one and a financial backlog approximately $287 million.

Revenues for the quarter exceeded $180 million, slightly above our management guidance and 10 percent higher last March. Geographically, revenues rose significantly in North America to 38 percent of the total. We're flat in Asia Pacific at 33 percent. The decline in Europe and Japan at 15 and 14 percent, respectively. Excluding the benefit of a $1.7 million restructuring reserve reversal, gross margins as a percent of revenues improved to 35.9 percent, 2.4 percent higher than last quarter's and a result of better capacity utilization, both our factory and in our field service organization. Reported gross margins 66.3, or 36.8 percent, reflect a partial recovery of discontinued product lines where inventory was previously written off.

Recurring operating expenses were $76 million, declining by $1.4 million in total. Continued investments in product development and product improvements drove R&D expenditures up in absolute dollars, but declined as a percent of revenue to 23 percent of sales. SG&A expenses fell $2.5 million to $34.4 million, reported by cost reduction programs, delivering our intended results. Reporting operating expenses of $73.7 million include the release of $2.4 million of restructuring reserves as actual expenses to complete the June and September 2001 restructurings were lower than the original estimate.

Other income and expense was impacted by a $26 million charge. We revalue our cashless collar to its market value the end of June based on a stock price of $17.98 per share on June 28th versus a stock price of $29.32 per share last March. Excluding the benefit of the one-time credits and the derivative instruments revaluation charge, losses after tax were $7.9 million, or six cents per share. Including the two items, reported losses after tax were $31 million, or 24 cents per share.

To conclude, let's review some key balance sheet items and statistics. Action short term investments increased by $43.7 million from last March, generated primarily by the acceleration of accounts receivables turnover to 67 days of the sales outstanding. Inventories rose slightly in preparation for increased shipment volumes in the coming quarter, while fixed asset retirements drove overall net plant and equipment down to $67.5 million. Capital purchases amount to $2.1 million. Appreciation was $10.5 million. Deferred revenues were $121 million and our worldwide workforce was approximately 2,500 full-time employees, down from 2,700 prior quarter. Cash, short term investments and restricted cash stand at $945 million at fiscal year end, providing ample liquidity to report our expected growth needs and to meet our short term obligations.

With these remarks, the information included in the press release, all the significant financial data for the quarter has been covered. So, now I would like to provide guidance for the September 2002 quarter.

New orders will be flat to approximately 10 percent down. Revenue will be $190 million to $200 million. Production output will be $210 million to $220 million. Gross margins for the quarter will be 40 percent. And EPS will be one to three cents per share. And we'll base that one to three cents per share on 134.5 million shares.

So, that covers the financials and the guidance. So, at this time, I want to go in and talk to you about our outsourcing program, which we have mentioned on a number of occasions during our conference calls, and provide you with a little bit of background on those programs.

Now, when we look at our business we can define aspects of our business as core or critical capabilities, those that are in support. What we have defined as critical for success in the marketplace our ability to develop leading-edge products, our ability to market and sell what we develop, our ability to service and support directly with the customer what we sell. All the other activities in the company are in support of product development, marketing and selling, customer service.

When we set out and looked at our outsourcing goals a year ago, we said that we wanted to create a transition that would enable the company to have a more variable cost structure. We wanted to have secure access to the expertise - that we wanted to secure access to the expertise of world-class partners. We wanted to redistribute our headcount from about 68 percent involved in core activities, 32 percent in support - we wanted to transition that to 80 percent of our employment headcount in core activities and 20 percent in support. We wanted to provide a more stable employee environment. Over the last eight years or so, we've hired 6,000 employees and we've let go

employees. And with this outsourcing strategy, our goal was for those employees involved in our core areas and for those employees involved in support we would provide a much more stable environment for their career development. We wanted to reduce our space requirements from about 1.1 million square feet, which we had at the peak of the last upturn, to 790,000 square feet by September 2002. And we wanted to improve our margin performance over the next cycle.

Specifically, we've had significant outsourcing programs in the core areas you see here. The manufacturing production we have been actively engaged over the last six to nine months in taking major assemblies and modules, primarily initially targeted at our 2,300 family of products as well as our 2,300 CMP products and moving them in to world-class contract manufacturing providers. Those programs have made excellent progress. They're well along. We are significantly complete relative to the 2,300, and we are now looking to expand our outsourcing activities to certain components of our alliance-based products as - we will see the alliance as a product that will ship in fairly significant volumes over the next two or three years. We want to transfer the benefits that we're seeing in 2,300

in cycle time to fill up in the alliance product lines as well.

We have recently completed outsourcing our manufacturing warehouse. We have gone to a major third party logistics provider. The benefits of this program will be significant, as we have locked in a cost that is transaction based. It's variable as our transactions decline. And we - as our transactions grow, our cost per transaction has built in and guaranteed cost reductions that will result in approximately a 15 to 20 percent reduction in warehouse transaction related activities versus when we did this ourselves.

Facilities, we have recently completed an activity that started July 1st where all of our major facilities activities are in the hands of one of the world's leading outsource providers of facility capability. We expect over the next three to five years that we will also see 15 to 20 percent reduction in our cost of facility support maintenance that we were doing ourselves.

In the IT arena, we have outsourced our IT activities in two ways. One, our SAP posting and our applications that run on our server activities has been outsourced. That has been successfully completed and is up and on line and operating well. In addition, we have taken the application support. And while a portion of it we still do in-house, we have outsourced the vast majority of applications development built around our SAP architecture, but also integrating other third party solutions. We have outsourced that and we have been successfully operating with both those outsource providers in the new IT world.

As a function of that, we have created the ability to change our financial model. And the effects of outsourcing on our business model can be seen in this slide. If you look at the numbers that we just reported today at $180 million of revenue and a gross margin of 35.9 percent, these numbers all reflect the numbers I just gave you. And our model says that if we double the business, when that doubling occurs, we would expect that our gross margin will grow to 47.5 percent, that our operating expenses will drop dramatically to 26.5 percent, our operating profit will be at 21 percent, which is a very significant increase on this level of business. And the comparison is that if you go back to the nearest quarter, which happened to be June 2001, where we did the closest amount of revenue to this doubling model, you can see that our gross margins will be six percentage points higher at a very similar revenue level, that our operating expenses will be down about 1.4 percent, but operating profit as a function of that gross margin level will be up dramatically and our earnings per share will be up 14 cents greater where we were in that similar revenue level. Our headcount will be down almost 750 people, or 20 percent, and we will have successfully completed the shift where you see this 68 percent and 32 percent in core and support to 81 and 19.

Just another interesting point in terms of where our model takes us is at the peak of the last cycle we generated 48 cents a share on revenues, about $465 million. We expect that we will be able to generate 48 cents per at a revenue level that's 15 to 20 percent below what we achieved during the last cycle, even equivalency of 48-cent earnings per share.

So, I think that you can see that outsourcing is a very powerful model, giving us significant level on gross margin, giving us significant leverage in terms of how we utilize our employee base. We expect also to achieve significant benefits by tapping into the best management practices, our outsourced providers, and we also expect to be able to ramp up and ramp down our transaction needs much more efficient and effective way.

So, in summary, Lam support organizations are partnering as we speak very effectively with outsource providers. The program is in fact creating a more variable cost structure. Our outsourced partners are delivering the anticipated benefits that we have negotiated with them and they are contractually committing to provide. And we are beginning to see and have visibility to the improved margin performance. We have very high expectations for this being a very successful

as we go forward.

So, with that, I'd like to introduce Jim Bagley, our Chairman and Chief Executive Officer, where he will talk about the marketplace and our products.

Thank you, Steve and Kathleen. And I'd like to add my welcome to those of Steve's and Kathleen's. We do appreciate you taking time out of your busy schedule because I know that my schedule is difficult to take time out to be here, so yours must have been even more difficult than that, because I'm supposed to be here.

What I'd like to do is talk about our products and talk about a little bit of change in shifts in the market. The products that we are focusing on today are basically etch and CMP as well as clean. And if you start at the bottom of this chart, the Versys 2300, which is basically our conductor product, and in this particular case we're focused more on silicon applications than the metal applications, you can see that they're largely involved in the front end of the wafer production process. This is a model of what a complex logic microprocessor DSP chip would look like with several levels of metal. The test, in addition to performing STI polishing, is largely involved in the interconnect activities within the wafer processing. And the Versys, our dielectric etcher, is - operates both in the front end as well as well as in the interconnect processes. So, our products are quite pervasive through the production of wafers because they are involved in most of the process steps.

Now, kind of as an overview of where we are, our revenue base market share is continuing to change. And over the past 18 months, we believe we've gained share in Taiwan and Asia Pacific, that we've seen significant share gains in 300 millimeter, particularly driven by the 2300 Teres as well as the Exelan. And we believe we've maintained share in the market in CMP, where our focus is on STI and copper, although we have sold CMP systems for other applications beyond STI and copper.

I'd like to spend a little bit of time on the product portfolio in total. We'll start with CMP. You see here the CMP system. And what we believe is the heart of our CMP system is the polishing system. And most of you have heard of linear planarization technology, which is the technology that we use in our CMP system, along with a proprietary air bearing, and I'll spend a little bit of time on that.

Our focus in CMP has changed over time and it's largely the result of what's happening in the marketplace. Today, we work on polishing copper that is being integrated with fluorine

silicon dioxide, FSG, and we work on copper integrated with

dielectrics. And it really doesn't make any difference whether it is films from -

films from Applied and films from other suppliers, as a matter of fact. And we are focused on shallow trench isolation.

Our focus on shallow trench isolation is relatively straightforward. While it's a dielectric polish as opposed to a metal polish, as in the first two examples, we - because of the global planarization capability of the Teres tool, it gives us an advantage in STI that we've been able to exploit in selling the product.

Now, the reason for focusing on copper and STI, if you go back to 2000 when the business as at its most robust level, copper and STI represented about 40 percent of the market, where STI was about a third of that 40 percent. As we look forward to 2005, our expectation is that copper and STI will be 50 percent of the market. So, our choice to focus on copper was that is an area where the install base is minimum, the opportunities are greater and the business would be growing. We are not focusing on silicon polishing, tungsten polishing and dielectric, although we do sell into other dielectric applications, largely because of the install base that currently exists as well as the opportunities are going to be smaller as we go forward over the next three or so years.

When we look at focusing on copper and particularly low k, there are several challenges. And one of the largest challenges we have is the fact that most of the low k materials that we are looking at that are being put - integrated into devices today have less mechanical integrity than FSG. The result is that the sheer forces tend to cause the films to delaminate. We're addressing that through our ability to have very low pressures and the design of the machine, which incorporates a closed loop ultra low down force capability. So, it puts less sheer on the wafer as the poly sheen head is rotating and the belt is moving against the wafer.

We also have an advantage because we have a very simple polishing head. The wafer is retained in a flat position, where our competitors in order to address profile control have to bend the wafer. And when they bend the wafer, it increases the sheer stress on the low k films. We've had in the past, and we continue to have today, the leading position in metal loss. We have better dishing and better erosion performance in process today relative to our key competitor.

Now, if I move to etch technology, I want to remind people of something that I'm not sure is always in the forefront of your minds. We are the leader in etch technology. And I'll explain how I arrive at that conclusion. We pioneered inductively coupled plasma. We called it TCP technology. And we introduced that in 1992. It is today the leading technology for silicon as well as for etch - metal etch processes. And we believe that it will continue to be the leading technology for silicon processes as we move to 90 nanometers and below. In 1995, Lam introduced dual frequency confined plasma. We call it DFC. And today, it is the leading technology for critical dielectric etch applications. This is whether it is DRAM applications for a

ratio contact in

or whether it is a very complex integrated etch processes for dual damascene on FSG or dual damascene on low k materials.

The reason we believe that we are the leader in both of these technologies is we are being copied by - and as they say, flattery is - I guess copying is the sincerest form of flattery. And we see our two major competitors in the dielectric etch arena moving to dual frequency. In some ways, they're trying to reach a quasi confined plasma. So, we know that our competitors are recognizing the value of our technology, and they are trying to create a technology that is competitive as we sit here today. But since we have - in dielectric etch particularly, we have a couple hundred dual frequency confined plasma chambers operating in production today. We are well ahead in the practical implementation of this new technology or this leading technology.

If we look at the products that we have that serve the dielectric market, we have the Exelan, which is on the alliance platform 200 millimeter. And we have a version of it that we use for non-critical applications. And the non-critical applications as a percent of the total etch market are continuing to decline because as we shrink feature size almost all the applications are becoming critical. We have a critical dielectric etch version of the Exelan. It's basically a high performance turbo-assisted process chamber, which has a very high commonality in parts count with the Exelan chamber itself. And then we have the 2300 Exelan, which is 200- and 300 millimeter compatible. It's our advanced etcher. This is where we will put our energy relative to 90 nanometer and below activities.

If you look at some of the applications for the Exelan and 2300 Exelan etchers, critical front end, this is high spec ratio contacts. This is a

etch. This is ultra low k dual damascene. We do this either in an ex situ, like our competitors, or complete in situ process, both dual damascene as well as ultra low k as well as low k. And, of course, we have been running in production for some time with dual frequency combined plasma on copper integrated with FSG.

Now, there's been some discussion about what does the move to copper and the elimination of most of the metal etch do to the etch market and, in particular, what's the impact on Lam Research. What we are showing here is a model. This is not what is done. We just model this for understanding. If you take a quarter micron aluminum device, five metal layers of aluminum interconnect, and you compare the number of etch chambers with that application for our 5000 wafer start per month facility or in increments of capacity, you compare that with what would happen if you had four levels of copper and one level of aluminum, again five levels of interconnects, what would happen to the number of etch chambers? The number of etch chambers goes up by about 30 percent. Metal goes down. Dielectric goes up. It's obvious that you have to be successful in dielectric etch, and that's why we're emphasizing dielectric etch so heavily within our product development activities.

Now, if you do a comparison of a 130-nanometer device, one level of aluminum, seven levels of copper, you can see that the number of process chambers goes up from the quarter micron all-aluminum system by about 70 percent. As a result, both of the increased process steps as a result of adding more interconnects as well as the number of processes that are going to be required in order to make a 130 nanometer device.

So, the net net of all of this is the conversion to copper has a beneficial effect on the etch market. It increases the etch market. The combination of transition to copper and 130 nanometer has a significant impact on the size of the etch market. The principle beneficiary is dielectric etch.

Now, I want to make sure you understand this chart, so don't write anything until I explain it because I don't want you to walk out of here completely mislead. We know and are forecasting that the dielectric etch market will grow. The entire dielectric etch market is going to grow with time. I just showed you that in the previous chart. We believe that this year dielectric etch represents about 57 percent of the total market and that in a couple years it's going to be well over 60 percent, closer to 65 percent, of the total etch market.

Now, within the total dielectric market, there is a market that we are targeting, which is sub 180 nanometer. So, that's what you see up here. The sub 180 nanometer market has grown - use this side of the chart - from about 15 or 16 percent of the total dielectric market, sub 180 as a percent of the total dielectric market, not etch just dielectric. So, 15 percent. In 2001 it will grow to almost 60 and in 2002 it will exceed 65 percent. So, it's obvious that if you're going to win market share in dielectric etch and the portion of dielectric etch that's going to grow the fastest is the sub 180 nanometer. That's where you have to focus your attention. That's where your products ought to be targeted and that's where ours are targeted.

Now, what we have done back when it was a small percent of the total dielectric etch market, we booked about 18 percent market share. So, if we look at the total bookings for the sub 180, we booked about 18 percent. We did not have a 300-millimeter product introduced at this time. As we introduced our 200-millimeter product and made it available in 2001, you can see the dramatic change that has occurred in our booked market share for sub 100 nanometers. But we have moved from about 18 percent to 27 percent market share. Now, we expect to continue to grow our market share in this very important category. So, as it's continues to grow in importance, with time, we expect to continue to grow our market share with time in the fastest growing portion of the largest part of the etch market.

Now, we still have products that serve other parts of the market. And I want to talk about those because they are obviously important to us because they still will represent 35 percent of the total market, even a couple years from now. These are conductor etch products. That's both silicon and metal. We have a 200-millimeter version that's on the alliance platform. The 2300 Versys is 200 millimeter, 300 millimeter and the same machine with a nine or so part change - nine parts change, goes from silicon to metal. And this product has been responsible for us being able to capture market share in 200 millimeter as well as 300-millimeter applications for advanced devices.

If you look at some of the applications - I'll cover these very quickly - this is a data etch for high dielectric. This is shallow trench isolation. Again, we do this with an in situ hard mask open process that we believe differentiates us from our competitors. A metal gate process - and then I'll go through this one just very briefly. This is an all in one gate etch where you can see this is the developed photo resist in gray. We do have photo resist trim, which means we make the photo resist smaller. We then do a hard mask open, which is represented by the red and green. Once that is completed, we then remove the remaining photo resist, then we do the gate etch itself, as you can see here. And that's an example of gate etch that's in - that's 70 to 80 nanometers in length, which is kind of the current state of the technology for gate etch. This is the key feature that determines the speed of a device, the length of the gate. And we have a technology to allow an all in one gate etch, which improves the total cost as well as achieves a CD control that we believe puts us in a leading position in the industry.

So, in summary, we believe the CMP systems that we are currently delivering as well as what we are currently developing are positioning us to meet the industry's demand - copper, polishing - irrespective of the film as well as SCI. Our view is we're the leading company in etch technology for applications, particularly at 130 nanometer and below. We believe that because of our in situ processing capability, which differentiates us significantly from our competitors, we provide a manufacturing cost advantage. And we know that the 2300 Etch Series is driving market share gains today.

So, with that, we'll open the floor for questions as well as the telephone lines. And Kathleen will moderate for us.

OK. Once again, please use a microphone before asking your question. Take first question from the floor. Yes,

.

: Jim, on the CMP front we've been hearing there's been tremendous improvement in uptime in Teres dual. Would you - can you talk about that and whether that changes the

scenario?

Well, I think - the answer is I think that the fact that our hardware performance is substantially better - this is 300 millimeter now that we're talking about. Our hardware performance is substantially better in copper than it was when we first introduced the product. I believe that relative to our key competitors we were behind in hardware performance even though - our view, looking at the data, say that we were leading in process performance. We believe that we have substantially closed the gap, if not completely, in hardware performance and we have retained our position as providing better process performance. I think that will allow us to be far more competitive as we go forward on 300 millimeter.

The 200-millimeter product is actually following the 300-millimeter product in improvement in performance because we have been focusing our attention on 300-millimeter copper and then forwarding the improvements back to our 200-millimeter machine.

Ladies and gentlemen, at this time we will begin the question-and-answer session. If you have a question, please press the star followed by the one on your push button phone. If you would like to decline from the polling process, please press the star followed by the two. You will hear a three-tone prompt acknowledging your selection. Your questions will be polled in the order that they are received. As a reminder, if you are using speakerphone equipment, you will need to lift the handset before making your selection.

I'm not sure I count eight, but it's - I can think of six.

I think roughly eight have been completed or roughly, give or take,

their R&D lines.

I think that we have achieved revenue, I believe, in at least six 300 millimeter

.

take place in the second quarter in May and June. I was wondering if you have any update on what transpired.

relative

.

I think in most of the 300-millimeter purchases that have been made recently, we have done very well with a couple of notable exceptions. So, I think we are doing as well as we expected to do so far. We'll see how things go for the rest of the year.

Of the backlog, could you break out how much of that is 300 millimeter versus ...

For the calendar year, we expect that when the year is over 30 or 35

will be 300 millimeter. Probably more importantly, about 80 percent of the orders are

.

OK, you guys, it also looks like I'm going to have to state your name and company name. You're not going to do that. I apologize in advance if I get your names wrong.

Next question.

Kathleen,

Morgan Stanley. Questions about the outlook.

broad based or

.

When we talk about guidance in the past, we really

. But I think that given the

quite strong.

pretty volatile market

. When you look at the size

, given that most of who's doing the orders

. And if any one of them chooses not to go forward with their order,

. So, I think that given what the semiconductor

be cautious

.

really aggressive.

pricing environment.

I don't think that the pricing environment is - it hasn't deteriorated. It certainly it's gotten better than what the situation was kind of in the September - the June-September time period of last year '01. It was - it was deRAMish at that point, where people were pricing, I guess, on variable cost or selling inventory they had or whatever. T Eh pricing environment is much better than that. There are still cases where people either want to penetrate markets or where they are trying to defend their markets and they lack maybe some advantages in their tool and they're under competitive pressure and they resort to price in order to stave off a competitor's market share gains. But that's been - that's the world and that's been going on since I've been at Lam. And it was like that before I joined Lam, that the etch market was volatile relative to pricing. But as far as I'm concerned, it's improved and it's kind of in the - in its normal mode.

Let's take a question from the phone. We have

on the line.

Thanks. Can you hear me OK?

Yes, we can hear you fine. Go ahead ...

Just so you know, I can barely hear you on the phone. I think you guys are going in and out.

OK.

A question, Jim, on the U.S. market in particular. Orders were down pretty significantly there. I'm calculating about 32 percent in the quarter. Wanted to know if you can give us a sense as to what's going on there.

And then additionally in Europe where orders look pretty strong, sounds like one of your major customers just cut their cap ex number a little bit. Wanted to know if that's reflected in your guidance and what you're seeing out of Europe going forward.

I'll answer - well, why don't you go ahead and answer that?

bookings. They're all about bookings. I think when we look at the bookings trend relative to North America, I think that what you're

based on the fact that the foundries have been

in place 130 nanometer leading etch capacity. I think that that's the largest part of

slightly down as a function of that. I think also when you look at what's going on in the bookings environment, you have a situation where again it's the major players. If any one of those players is in an ordering environment and maybe they're not where we have a strong market share position, then they may not participate as much as what the North American market may do as a whole. '

I think in Europe we've had many discussions with our customers. And capital in the Europe environments seems to go on hold, off hold, on hold, off hold. I can tell you that in the recent discussions that we've had with the major European players recently as a couple of days ago is that as it relates to the significant investment

of the leading edge environment

would be released on a going forward basis. I think that as a function of what's happening in the

markets, there's

of being exhibited by our customers. That's why

bookings forecast

on a cautionary note. But

believe that the intent on the customer relative to leading

.

, J.P. Morgan.

. They were talking about one three micron copper yield

. They

that

. So, I'm just curious

. Is that congruent

?

I'm not sure,

that it's having a significant impact in the hear term.

. I would say that the range of yields is much, much broader than that. It's

well up into, I think, the 90 percent area. When you look at the copper related yield defects as opposed to other defects that may occur in wafer processing. You have companies that are on 200-millimeter copper with FSG whose yields are very high. You have companies that are on 200-millimeter low k. Their yields may be moderately high, or moderate. You have companies that are on 300 millimeter FSG whose yields may not be quite to the 200-millimeter level, but it may not necessarily be copper related. You have companies on 300-millimeter low k whose yields are quite low, which, again, there's a sudden impact on 300 millimeter, but it may be across all of the processing. So, it's very difficult to differentiate.

One of the customers that I talked to about their copper yields said that there's a significant issue with copper which is very different from what they have seen in the past. In the past, when you started a foundry, whether it was on 200 millimeter, 300 millimeter - of course, there was no examples of that for the

. But if you were starting a foundry up and you wanted to determine how the foundry was loading, you could it with DRAM. You could load it with static RAMs. You could run those. It's a commodity part. You can sell all you can make, maybe at low prices, but you can still sell all you can make. And it's a good product for determining defect levels in your processing. And with aluminum you had two levels of metal. It was on SI02 and you could shake down your multi-level interconnects using two levels of metal for either a DRAM or a static. Today, there are no parts that have copper interconnect that run at high volumes except microprocessors. And I suspect that, too, microprocessor suppliers have high yields as a result of that. I suspect.

But there is nothing that you can load into a 300-millimeter facility that has eight levels of copper interconnect so that you can debug the process. And I think this is a difficulty that the industry is facing. The FSG yields are better than low k, but we've got to find a vehicle for improving the yields at 300 millimeter with copper interconnect and substitute something other than what has historically been an easy thing to load in wafer fabs than determine the defect levels of the wafer fab. So, this is a problem.

Now, whether that's had an impact on people's plans, I would say

. But I think most people are planning to have 130 nanometer copper either on FSG or on low k and they've got to have it by next year and they're going to go ahead and make their investments. They may moderate their investments some, but I don't believe they're going to stop investing because they have yield issues.

.

I would guess that that's probably the major impact on change in outlook because the customers don't know what's going to happen. They don't know - they can't read the market any better - maybe they can read it better than you can today because they -

they've got their customers to talk to. But it's not clear that their customers know what the market's going to do. So, I think the volatility in the market is a - certainly a cause because there are just no clear as to what's happening in the overall end market, which will affect semiconductors, which will affect us.

By the way, before I leave that

it's bad. We don't know. So, before everybody walks out and says it's a disaster because we have no visibility, we had no visibility that the market was going to be anything like as good as it was in the middle of '99 looking into 2000. So, the fact that we have no visibility does not indicate things are bad. It means things are potentially changing and we're not able to discern what the changes are and their level.

Jim, a couple questions. First ...

Your name and your ...

, First Boston.

Thank you.

Can you just characterize the cancellation environment right now? Has that slowed down, picked up in recent weeks? What's your expectations as you move into the third calendar quarter? And then secondly, a follow-up on the pricing issue. You said that the pricing environment had gotten better. Does that mean you've actually been able to come back and raise some of the pricings on the etch tools? And then when you look from the move to 200 to 300, what are ASPs doing? And then a follow-up for Steve as well.

(inaudibles) and adjustments to backlog are very minimal. I think there's a couple million dollars in cancellations and $3 million or $4 million in backlog adjustments that reflect our normal process of taking a look at what orders do we have in the books and what's the customer's situation. So, the overall adjustment factor this quarter was very small, and I think it reflects the fact that all of this major change in terms of all the stuff that was on order got pushed out. Is it going to be delivered or not delivered? All of that has been worked through. So, I think where we are is a very clean booking sheets in terms of good backlog that's on the books and then where we go from here I think is ready to go.

On the pricing environment, we have not raised prices on anything. What we have done is in some cases we have been less aggressive on discounts and less aggressive on other things that we would provide as an inducement to purchase when our competitors have come in and dramatically lowered pricing. We're trying to elevate ourselves above this

. We're trying to be the class act in the business by not changing pricing. We've had some competitors who have attempted to raise pricing with very embarrassing results. And we priced for the long term. We haven't changed that. Average selling price changes on all of our products with time as we add features and enhance our capabilities or add options to the systems, but we haven't changed our pricing.

Steve, going back to your outsourcing strategy, when you look at the target model it appears like you're getting a lot more leverage on the gross margin line than on the operating expense line, despite the fact that headcount's coming down fairly dramatically. I'm wondering if you can just help me understand better why you can't get better leverage off of operating expenses. Is that just a higher investment rate on the R&D line, or is there something else going on?

Well, there is an aspect of a higher investment in R&D that we're taking advantage of because I think that as we get more efficient one of the things we want to do in this relatively low revenue environment is to maintain a very strong investment in R&D. So, that's part of it. The other is that there clearly are significant both headcount reductions, but as well cost reductions by going to our contract manufacturing suppliers. We've actually been pleasantly surprised that their ability, their pricing and procurement leverage with some of the types of components that we have has been very strong. So, we're seeing their ability to deliver us parts where we're seeing 15 to 20 percent reductions in what it was costing us to do ourselves. And with that, we're also able to take manufacturing buildings out of the equation. So, we've gone from a situation where four years ago we had five manufacturing buildings. We're going to have - and from that we can probably do $2 billion worth of revenue. We've reduced that to three buildings and have the ability to do $3.5 billion worth of revenue in the last cycle. We'll now be able to put out $4 billion plus out of one building with our outsourcing strategy. So, we've really taken a fixed cost structure of the factory down dramatically.

I think where you'll see additional leverage in the outsourcing model is as we grow beyond these 360 type of numbers that we've talked about. I think we'll then be able to leverage beneficially in the op ex arena as well.

This is

Partners. Could you put any detail on how concentrated the bookings were this quarter among your customers and what you expect that to be like next quarter?

The question was what was the concentration level?

Yes, like what percent of the top were or something like that.

Well, I think that what you're seeing in the past quarter, as you have in the past couple quarters, is that really the change in demand is occurring in the foundries, in the memory arena and with certain microprocessor and very specific targeted logic players. But for the most part, it's all the top tier big name companies who have the financial resources to invest. They're the ones that are on the leading edge. They're the ones that are having to add capacity at 130 nanometers. What we haven't seen is a lot of the second tier or middle tier players, primarily in the logic arena, come and add capacity because the reality was that where this wafer start decline occurred was very significantly in that middle tier logic arena. Wafer starts have begun to recover quite significantly. And I think that if we get to macroeconomic growth that the

says is there, if we can continue to have the consumers spend and then get some more IT spending from a corporate standpoint, then I think that we're going to see the middle tier logic players who have really not participated much in terms of ordering equipment they'll come into the play. But whether that's going to be in this next six month timeframe or whether it's going to be after the first of the year we're going to have to wait and see.

Just one other on pricing.

couple other equipment commented that in Asia and China the pricing environment is a lot more competitive and the discounts are more severe there than they are in the more developed areas. Given that those

is that going to have a longer-term implications on margins for the overall equipment industry? Are you seeing anything like? And is your pricing lower in those regions?

I think the characterization is time dependent. The two major Chinese foundries based orders and quasi orders in the June, September and December time period of '01, which was the lowest point in bookings in this current cycle. And I don't think that their pricing - I know from Lam the pricing in China was no different than it was in other regions at that same point in time. So, if you look - if you characterize China where most of the bookings were placed back at that time and

, I don't think that is an accurate depiction of what the future will be because the overall net realizable pricing in other regions has come up as the business has improved. And were China placing large orders in this environment, I think that you would see that the pricing would be improved in China also.

.

Back to the outsourcing for a second. When you look at the model that you showed us earlier where you double revenues and where the margins can go, how linear do you think it is between where you're at now and when you get to that double revenue and that better margin model - in terms of the outsourcing program that you've laid out for us, how far into that are you at this point? And therefore, would it be fairly linear - are we going to look at sort of like break points as you move up where you'll get larger leverage?

I think there's a fair amount of linearity to it. We talked about that as we move up our revenue to $190 million to $200 million, which really is maybe 10 percent, we're going to go up four gross margin points or so from where we were in the prior quarter. I think that there's a situation where we're now down. We're into one building. And we've accomplished that. So, we'll get the benefit in the September quarter of those kind of things. We're significantly outsourced and

.

Once again, ladies and gentlemen, if you would like to take - if you would like to ask a question, please press the star followed by the one on your push button phone at this time.

...

revenues rising. But it also benefits that even if the revenue kind of flattens out a little bit we'll still get margin improvement as a function of those new suppliers we consume inventory and then start getting in

supplies from those suppliers. So, it is relatively linear, but it's not a perfect linear

.

A follow-on. By the way, it's

, Merrill Lynch. In the past, a lot of the etch products were pretty customized towards the end. And I'm wondering is it becoming more of a standardized market at this point? Does that help the margins at all?

I think - I mean, there are still certain places in the world where the customization aspect is still present. But for the most part, the equipment companies, and certainly it's true in our case, have created such a capability that customers are buying whether it's your standard product and your standard set of option suites that go with it. Because what's been proven is that the ability to optimize and create the reliability to get the process window width that you want and get that repeatability and performance that's so critical really is in fact enhanced by standardizing as opposed to in the specials arena you actually can get degradation. Customers have come to realize that, so the environment has changed to a more standard approach.

We'll take a question from the phone. We have Byron Walker on the line. Byron, can you hear us?

Yes, I can. It's Byron Walker, UBS Warburg.

OK. Go ahead.

Financial clarification. In your 10-K when you reported you'll list a backlog of deferred revenue balance and deferred profits. Can you give us those numbers now?

We said that the backlog is $287 million. The deferred profit for the quarter was $63.4 million. And the current deferred revenue for the June quarter was $121 million.

And what's the deferred revenue balance, Steve?

Did you have another question, Byron?

Yes. I was inquiring what the balance is on deferred revenues. What's left on deferred revenues?

A hundred and twenty-one is the balance at the end of June ...

That's the balance. OK.

Yes.

Thank you.

Thank you. A question up front here. Theodore.

Thanks.

,

Tobin. I just want to follow-up on Jay's question earlier. Your customers are slowing down their order growth rate. Is that because of as they look at the utilization of your tools out there they're doing fewer wafer starts and it's a direct result, or are they concerned that there's a wafer start slowdown?

not a wafer start slowdown. In fact, you can talk to a number of customers. And depending upon the success of where they've been in the design cycle, many customers today are at higher wafer starts than where they were at the peak of the previous cycle in, let's say, end of 2000 or the first quarter of 2001. When we look at total wafer starts throughout the industry, it's likely that on an industry total basis wafer starts by the end of the calendar year will be equal to the peak. And so, I think that when you look at wafer starts at the leading edge they're continuing to grow. Most of the equipment that's been shipped in in the last six to nine months has all gone into driving up that wafer start environment. I think that where the excess capacity was in terms of quarter micron capability, .18 micron capability that's where you see the wafer start environment growing and beginning to create a situation where if that situation sustains itself, then there will have to be an expansion in wafer start capability. But at the present time, they're still starting into that capacity that they've bought back in the prior cycle.

. The order number of $234 million, that's a net number?

. What we talked about was the adjustments in the previous question where I think it was $7 million.

It was six.

Six?

We have a question over here.

Hi.

, a manager at Morgan Stanley. Just a question. If you look at your backlog at the end of the March quarter and if you compare that with your backlog at the end of the June quarter, what proportion would you say is for shipment over the subsequent two quarters?

What percentage of that backlog will ship in the next two quarters?

Yes. If you look at the backlog, what is due for shipment over the ...

I really haven't looked at the detail of that, so I can't answer it specifically. But I can tell you that typically most customer orders are taken for expected delivery within the next six months because the average lead time that's out there is four or five months. Certainly there are orders that are being taken that are not looking for delivery till maybe the first quarter of '01. But we don't record financially any bookings that the customer is requesting delivery beyond 12 months. So, the vast majority of that backlog is scheduled to be shipped in the next two quarters, with some in the third quarter and a little bit in a fourth calendar quarter out.

Any other questions? A follow-up question from Ted.

Hi.

. Sorry I didn't announce my name earlier. I had a question with the outsourcing trends. Where do you see your competencies today now that you're outsourcing more? What are you going to keep in-house? And if there's certain modules that you're outsourcing now that you didn't

name, what maybe some of those modules are.

You're talking about manufacturing?

Yes, your manufacturing outsourcing. What do you see your core competencies today? What are you going to keep in-house and what's moving out of house?

I think we're going to keep all specific manufacturing activities that are tied to core chamber technology activities. One, there's an expertise involved in really how you deal with certain critical process chamber parts. And, two, there's also aspects of proprietary design that we really want to maintain a very tight control on. So, there'd be certain aspects of the process chamber that we will not do. There will be, from time to time, certain aspects of other technologies that if we believe that they are either require a manufacturing process control capability or they represent an IP that we want to really maintain tight control we'll keep it. And so, other than that, though, the majority of what will be manufactured will be done by a contract providers. There will be some level of top level validation of integrated modules that we will retain for some period until we can absolutely ensure that the integrity of what's being built at the module level matches the quality standards that we have. But I think that as we go forward we'll develop and demonstrate that competence.

I think the other issue that's really not an IP-related activity is one of having the contract manufacturers demonstrate their ability to respond to the very significant rent requirements that are present in this industry. So, we will maintain with our manufacturing alliance the capability to augment if necessary certain aspects of volume ramp capability so that we can make sure that we continue to respond to the very short cycle time requirements of our customers.

Before you - to make sure you don't walk out with a misunderstanding, if you look at an etch process chamber, whether it's an Exelan or a Versys process chamber, there are only a few pieces of that process chamber that we would consider significantly proprietary. The vast majority of the process chamber - first off, the vast majority of the process chamber is between all of our products now are common. And so we can have a basic process chamber built up to a significant level than we can add parts to it, but make it a metal poly or dielectric system. And there's only a portion of those three parts list that we would consider very proprietary.

OK. Are there any more questions? OK. Thank you for joining us today. We look forward to meeting with you again and updating you in the October timeframe. Thank you. bela: OK. Are there any more questions? OK. Thank you for joining us today. We look forward to meeting with you again and updating you in the October timeframe. Thank you.