Keynote Transcript

Extreme Ultraviolet Lithography Conference Call

Gordon Moore and Secretary of Energy Federico Pena
September 11, 1997

Operator: Good afternoon everyone, and welcome to this Extreme Ultraviolet Lithography call. This call is being recorded. At this time I'd like to turn the call over to Intel spokesman Howard High. Please go ahead, sir.

Howard High: Thank you for joining us today to discuss the formation of the Extreme Ultraviolet Limited Liability Company. We have with us today U.S. Secretary of Energy Federico Pena and Intel Corporation's Chairman Emeritus Gordon Moore.

We'd like to begin the conference call with comments from Secretary Pena and Dr. Moore, and then we will open up the call for questions and answers. Mr. Secretary?

Federico Pena: Thank you very much to all of you for joining us today for this very exciting announcement. And I'm very honored and thrilled to be here with Dr. Gordon Moore, the Chairman Emeritus and Co-Founder of Intel.

We are announcing a $250 million partnership between the Department of Energy and the U.S. semiconductor industry. This partnership will help create the next generation of computer chips to power the electronics we will use at the dawn of the 21st century.

President Clinton and I fully are committed to ensuring that we have a strong U.S. computer chip manufacturing industry including the companies that make the equipment used to manufacture computer chips. Today's announcement is a step in the right direction to guarantee that we stay competitive in this vital industry.

This partnership is the strongest possible endorsement of the scientific and technical capabilities found in the Department of Energy's National Laboratories. Our partners, Intel, AMD, and Motorola, are on the cutting edge of high technology. But they're relying on the Department of Energy's National Laboratories to make a breakthrough in this critical microchip manufacturing technology.

These new chips will have 100 times the computing power and 1,000 times the storage capacity of today's best chips. We'll be able to store perhaps an entire library on one of these new chips.

The Department of Energy Laboratories have extraordinary capabilities and have produced extra ordinary results. Those results include the recent discovery of a ceramic membrane that will dramatically reduce the cost of producing liquid fuel from natural gas and also a 1997 Discover Magazine Award for remote ultrasound technology that greatly expands the flexibility of this vital medical diagnostic tool. But now our laboratories will help develop the 21st century computer chip.

The public private partnership we are announcing today is necessary because we appear to be reaching the limit of how much computing power we can put onto a chip. By combining the considerable scientific resources of the Department of Energy's National Laboratories and the resources of the high tech companies here today we should be able to break through the barriers that threaten to slow the progress of the information age.

The partnership is a three year $250 million cooperative research and development agreement otherwise referred to as (CRADA) between three of the Department of Energy's National Laboratories and the Extreme Ultraviolet Lithography Limited Liability Company. The company is a consortium of U.S. semiconductor manufacturers including Intel, AMD, and Motorola that are embracing and commercializing the technology developed by the Department of Energy.

The Department of Energy created this technology know as Extreme Ultraviolet Lithography or EUV at three of our national labs -- Livermore, Sandia, and Berkeley. The technology behind EUV was developed as part of our efforts to ensure the safety, reliability and security of our nations nuclear weapons stockpile.

Now there are two reasons why this public private partnership is so unique. First our private sector partners are paying 100 percent of the cost. They'll do this by investing $130 million to cover our expenses at the labs and contributing $120 million in other resources such as materials, equipment and their own personnel. This is the biggest investment the private sector has ever made in a Department of Energy (CRADA).

Secondly, this partnership creates a virtual laboratory for our EUV research consisting of our three national laboratories. Although work will be done at many locations, modern communications systems will enable our scientists and technicians to work side by side through shared computer networks. They will share results, breakthroughs and ideas that will speed the development of this critical technology.

As I mentioned previously, these new chips will have 100 times the computing power and 1,000 times the storage capacity of today's chips. So for example tomorrow's computers will be able to speak and write in perhaps 50 different languages, and they'll have perhaps television quality video that will allow a grandparent here in California to actually have a conversation and an interaction with a grandchild in the state of Florida.

You know I can't think of a stronger vote of confidence in President Clinton and the Department of Energy's technology partnerships than the support that we're receiving from Intel, AMD and Motorola. I want this (CRADA) to serve as a model for future agreements with other U.S. industries that will keep our economy strong and produce quality, high paying jobs here at home.

Staying at the forefront of the worlds semiconductor industry is critical to our nations future for many reasons. The Department of Energy and U.S. tax payers benefit because we'll strengthen our capabilities in EUV research.

EUV has applications in micro-machine manufacturing, advances sensors and other technologies that are needed to perform our missions in national security, environmental clean up and developing new energy technologies. And once the new chips are produced we will have faster and more powerful super computers to work with as we tackle things like global climate change and the insurance, and safety and reliability of our nation's nuclear weapons stockpile.

The U.S. economy benefits because our semiconductor industry will stay in the lead of this world's $1 trillion electronics market creating jobs and opportunities for thousands of Americans. A typical car now has a dozen or more chips that stabilize the ride, control fuel use or help reduce pollution. And the market for chips will only grow as the demand for computers, telephones, televisions and other electronic devices increase around the world.

But perhaps the most important reason why this revolutionary public private partnership is critical to our future is that we'll achieve the information age going full speed. I believe that when historians look back on our era, I have no doubt that they will identify the microchip as the seed and information as the flower that bloomed so dramatically at the last half of the 20th century.

We are doing things that we once thought were impossible. Borders are shrinking, children are being better educated, farms are producing more, diseases are being cured all because of a simple of invention and the way it has been used. The microchip is powering the information age. This public private partnership today is a critical part of that future.

And it is my great pleasure to now introduce an extraordinary American. A gentleman who has formed this extraordinary company, Dr. Gordon Moore, who is with us today. Dr. Moore?

Dr. Gordon Moore: Thank you, Secretary. I'd like to give a little background on why the industry is willing to commit this large amount of money to continue to develop this technology. The semiconductor industry in a lot of respects is unique. But I think that's best shown in what we've done to the cost of our products over the last years.

Since I've been in the business the cost of a transistor has gone down some ten million fold. There's no other industry I can identify in the history of mankind where a similar kind of an improvement in cost has been made, particularly over a relatively short period of time.

The major reason for this is what I consider a violation of Murphy's Law. By making things smaller everything gets better simultaneously. The performance goes up. We can make more reliable systems. But most importantly the cost of an individual transistor drops dramatically. Essentially, we can pack more in a given area on the silicon. We sell real estate. And the more we can put in a given micro-acre the lower the cost of the electronics.

For the last 40 years we've made these patterns optically. We've used optical systems of one sort or another to project the image of the pattern we want onto the wafer.

Originally this was done with 16 millimeter movie camera lenses that we could buy at the local camera store. Today the lenses in the machines we use in production weigh nearly a ton, and the machines themselves cost several million dollars and we buy them by the dozen.

The problem is you can't make images much smaller than the wave length of the light that you're using to make the image. It's fundamental physics that limits this.

We once thought the limit of making patterns optically would be something around one micron. That's one millionth of a meter, about twice the wave length of visible light. But with shorter wave lengths and much better optics we're now starting one-quarter micron feature sizes in production, some four generations of technology beyond where we once thought the limit would be. And we can probably go a couple of more generations, maybe down to something like an eighth of a micron. Beyond that we need something else. We can no longer use ordinary optical systems.

Industry has recognized this problem coming for some time, but no obvious contender has arisen to challenge optical lithography. But we're running out of tricks in what we can do with optical lithography, and we have to move to some other technique, something using much shorter wave lengths than we have today.

The problem is if we go to shorter wave lengths, the standard optics don't work. Materials aren't transparent. There are no longer lenses you can make or masks that we can use in the ordinary way.

We've examined several alternatives in the industry and these are still being pursued. One is to use x-rays of very short wavelengths and use shadow masks so that you can make a pattern on the mask that gets reproduced as a shadow on the wafer. The advantage is a very short wave lengths allowing for very high resolution. But there are severe problems on using masks this way. And a lot of effort is going into this technique worldwide. It is not yet proven to be a real challenge to the optical technology available.

Another way is to write with electron beam, essentially using an electron microscope backwards, if you wish. This works in that it makes very fine patterns, but it's relatively slow since it writes a line at a time on the wafers. And as we make increasingly complex devices, we essentially negate all the progress that is made in improving the writing time so that we are kind of churning our wheels trying to get more throughput from electron beam systems while making increasingly complex devices.

Today we're talking about a third approach, what has been called EUV or Extreme Ultraviolet. While it certainly isn't obvious at the present time, all three choices are complicated, expensive and technically very difficult. But we think EUV, which has also been called soft x-rays in the past since usually the radiation is kind of intermediate between what we normally think of ultraviolet and as x-rays, is really a very strong contender. But much work remains to be done.

The national labs has done sufficient R&D to establish the proof of concept, to demonstrate the source, to show that the optics can be made at least at some level to use this kind of radiation. But much remains to be done to move this on to a stage where it can really be designed into equipment we could use in our production lines.

The EUV Limited Liability Corporation is going to pick up this program which would otherwise have fallen off the laboratory's plate. And this organization, currently composed of Intel, Motorola and AMD hopes to pursue this technology to the point where it really can be realized as a useful production system.

There's a very strong technical team at the various laboratories that has been combined into a Virtual National Laboratory with representatives of Lawrence Livermore, Lawrence Berkeley and Sandia, and working together with the industrial partners who are taking on some of the technological challenges such as the source of this radiation being pursued by TRW and Northrop Grumman, the optical elements of unprecedented precision that are being undertaken by Tinsley Corporation, the multi-layer coatings that have to be of the highest quality to make the optical systems at these wave lengths being done probably by the 3M Corporation, and then finally the integration of these various state-of-the-art technologies into a system that can be used by the semiconductor manufacturers.

An operation will be taken on by the semiconductors equipment manufacturers including those that are currently the major suppliers to the industry. We think that equipment manufacturers that currently supply over 75 percent of this kind of equipment to the industry will probably participate.

A successful program will allow the semiconductor industry to continue to progress toward more capable and ever cheaper electronics. This program should strengthen and expand the U.S. semiconductor equipment industry and increase the competitiveness of the semiconductor manufacturers in the world economy.

So we think it's a great program that combines technology developed initially for other purposes by the National Laboratories to solve a major problem in a major and growing industry. So we're excited about the program.

Federico Pena: We're happy to take any questions. And by the way, let me say that we have a number of distinguished people that are with us in the room -- representatives of our three laboratories and Intel, so they may be prepared to answer some very technical questions. But who would like to start first?

Operator: The question and answer session will be conducted electronically today. If you do have a question, please signal us by pressing the digit 1 on your touch-tone telephone at this time.

Once again, if you do have a question please press the digit 1 on your touch-tone telephone. And we'll come to you in the order that you signaled.

We'll go first to Colen MacIlwain of Nature.

Colen MacIlwain: Good afternoon. Can you tell me please what steps you are taking to ensure that it is American equipment manufacturers will benefit from this program and that it won't serve to perpetuate the dominance of Japanese equipment manufacturers in lithography technology? I think there is some concern in the Congress about this point.

Federico Pena: Let me begin to answer that question. This is Secretary Pena. First of all remember that the Limited Liability Company is all American. So all the companies involved there are U.S. companies.

And secondly in respect to U.S. equipment manufacturers, today they do not have a predominant share of the world market of this business. As you know international companies do. However we very much want the U.S. equipment manufacturers to be able to participate fully in this venture.

And when the licenses are provided by the Limited Liability Company, they'll be divided on an equal basis for all these companies to participate.

Secondly to the extent that any international company is involved, they would be required in the first two years of the program to actually have the manufacturing done in the United States. So we think that helps.

Thirdly, I have invited representatives of the U.S. companies to our offices in Washington next week to talk to them about how perhaps we could work with them in a creative way to make sure that they are able to fully benefit from these opportunities.

So that's the approach that we're taking. I understand that there are some concerns. But we're trying to be as sensitive about those as possible and yet get the job done, which, by the way, is making sure that our U.S. semiconductor industry, which is the world leader and which is the predominant part of this market worldwide, continues to be number one in the world. So that's where our main focus is.

Dr. Gordon Moore: I might add to that -- this is Gordon Moore -- that we're also really developing some new participants in the semiconductor equipment industry by these corporations that are taking on the complicated new technology. The ability to make these mirrors, to coat them, the light sources, these are the major components of new system. And those are being developed by U.S. corporations. And I presume these corporations will become major suppliers to the equipment manufacturers.

So we really are expanding the scope of the semiconductor equipment manufacturer in the U.S. at the same time.

Operator: Go next to Luke Collins of Electronic Times.

Luke Collins: Hi. Thank you. Yes, really a question for Dr. Moore. What do you get with this Extreme UV? I mean, I'm used to hearing now of processes with perhaps of .25 micron, what are we going to get with the Extreme EUV? And can you translate that into perhaps the transistor budget or a microprocessor which would be manufacturable and perhaps memory technology just as a measure of the impact of this technology?

Dr. Gordon Moore: Okay. I didn't hear your question very clearly, but I think I got the gist of it. The first part of it was when are we going to need this technology essentially.

First of all, it's much more convenient to talk of the printed feature size rather. Because that's really what you make with this equipment.

I think the consensus in the industry is we can clearly use an extension of the optical techniques. It's the .18 micron generation. Most people believe we can do it at the .13.

We probably have a very strongly divided opinion in the industry if we can go to 1/10 micron. My personal view is it's unlikely. Our chief technical person on my left is nodding yes to that. Okay, I thought I might be at disagreement with my own people on that one.

So that says we really have two more generations that we can do before we need something like this. Two more generations typically means six years, although its come more nearly to be 2-1/2 years rather than three years a generation recently.

So we really need this capability to be able to start putting into production at about 2002 if we're going to stay on the curve we've been on in the past. And the program is put in place to try to hit that time scale -- a lot of work to be done.

Operator: We'll go next to Bill Loveless of Inside Energy.

Bill Loveless: Thank you very much. I'm interested from both the Secretary and Dr. Moore if you can give us any indication how easy or difficult it is to assemble an agreement like this given its magnitude.

You mentioned a moment ago Secretary Pena, the provision on manufacturing in the U.S.. But were there any other, particularly challenging aspects of this agreement that needed to be addressed in intellectual property or any other issues?

Federico Pena: Well, as you were asking your question, there were a number of people here who were chuckling at your question. I think they were reflecting perhaps that this was a somewhat complicated and challenging agreement. And it is the largest we've had at the Department of Energy.

Let me answer your last question and then perhaps turn to somebody else who was involved in the initial negotiations here to talk about how difficult it was to get done. Okay. Someone just wrote down four months.

Bill Loveless: That's not bad, is it?

Federico Pena: It's quite a challenge. But let me perhaps have one of the people who was directly involved in the negotiations to talk about the complexities. And on the question of ((inaudible)) agreement to ensure that the intellectual property rights are respected and that the companies that are part of this, while they may share information among themselves, will have some protection as respect to other companies that are not part of the limited corporation.

Richard (Freeman): My name's Richard (Freeman). I head up the Virtual National Lab and it was my duty to represent the three national labs in the negotiations that went on with the Limited Liability Corporation to form this (CRADA).

First of all let me say although it took four months there was, good will on all sides otherwise it would have never happened. It's a very complex agreement. But what it essentially does is that it guarantees the Limited Liability Corporation for their participation of funding the operation. It guarantees them exclusive use of the intellectual property in the field of use of EUV.

What is significant from the Department of Energy and what I was representing to the Department of Energy and the national labs is that a lot of intellectual property comes out of this kind of agreement. You have 100 people working full time, bright people working on advanced technology, you get many new things coming out. Those areas that are not unique to EUV revert back essentially to the National Labs to use in future R&D efforts.

So this was a compromise that we reached with the LLC. And I think we both felt that at the end of the day we had achieved what we wanted to in the negotiations.

Operator: We'll go next to Peter Weiss of Valley Times.

Peter Weiss: Thank you. I'm interested in the impact of the labs in the Livermore area in terms of how much money would go to each of the labs -- and in Berkeley --and what effect would it have? Would it cause the labs to increase hiring? And also I'm interested in how much the industry puts in from their end?

Federico Pena: We have three of the lab representatives here. Why don't we have each of them sort of talk about this? Who would like to start?

(John) Crawford: This is (John) Crawford from Sandia Labs. The money that actually goes to labs as a percentage has been predetermined. Forty-five percent of the funding goes each to Sandia and to Lawrence Livermore and ten percent to Lawrence Berkeley.

I'll speak about the impact at Sandia. And then the others can speak about the impact at the other labs.

This will not cause a major change in our hiring patterns in numbers. It may change a little bit of the kind of people that we hire. But we are not going to increase the general size of the laboratory as a result of this (CRADA). We are pulling people in from other projects, because frankly what we need are people with experience that can go to work right now on the project. And so the people that are working the program at Sandia are moving primarily from other programs and those that have been involved historically in the project over the last number of years when we working on a previous (CRADA).

Charles (Shank): I'm Charles (Shank), Director of the Lawrence Berkeley national lab. The effort that will be taking place at Berkeley will be the utilization of the advanced light source which is the world's brightest source of EUV which will be used to characterize the optics for the program. This will benefit our laboratory and strengthen our scientific program at the advanced light source at the same time providing this benefit for this program. It will not lead to a major change in the size of the laboratory or operations.

Bruce (Tartar): Peter, this is Bruce (Tartar) the Director at Livermore. And I think our story is very similar to that at Sandia. I think we will use primarily people who have been in place.

But I think you're aware and realize this grows out of work we've been doing for the past several years in earlier cooperative agreements. And what this does is provide a focus and a great deal of intensity and a product and a schedule to provide this product. And so I think we're very excited about that aspect. And we can pretty much handle it as an extension and amplification of the work we've already been doing in this area.

Peter Weiss: Dr. (Tartar), if I could follow up with a quick question, for you does this represent a different way of doing business with industry than you have in the past?

Bruce (Tartar): Well, I think we're particularly pleased about the fact that work done with industry has now been sufficiently high quality and relevant that people are going to pay their money to have that work done. I think that's a very rewarding aspect of the work we've been doing over the past number of years.

I'm very pleased with our technologies as a sound value. And I look forward to those results.

I would pick up one other perhaps an ironic comment that I think reflects the earlier remarks Secretary Pena made, that if this is all successful, the product will be a much better super computer. And as the Secretary indicated, that will make my other job of assuring nuclear weapons much easier.

Now the ironic twist in this, of course, is that if this truly successful, a major byproduct of this is we will basically break Moore's Law. And so I think that has its own irony. And we will be able to do better than doubling power every couple of years.

(John) Crawford: Peter, let me follow -- this is (John) Crawford from Sandia -- let me follow up with one other comment. And that is that I believe the opportunity for the three labs to work as truly a virtual National Lab is a very unique opportunity and I think one which all three of us have a great deal of expectation for and pride in the fact that we've got this thing started so that the project and the LLC as a single customer has provided the focus for the three labs to come together in a very unique way and I believe will point the way for future efforts with our industrial partners.

Peter Weiss: So none of the labs are going to hire any new people out of this?

(John) Crawford: Peter, what I would say is that our hiring program will not change a great deal. We are continuously hiring new people. But I don't expect the size of Sandia's operation in Livermore to change as a result of this.

Peter Weiss: But has anyone come up for a figure just for, you know, whether it's even the very minor amount of hiring, has anyone come up with a figure for how many people extra might be needed to make this project work at the labs?

Federico Pena: Could I -- this is Secretary Pena -- could I approach that question perhaps a little differently? I think that given that this program used to be more substantially funded by the Department and Congress and that this funding would stop about a year and a half ago as I recall, this new (CRADA) will mean that perhaps there may not have been a termination of positions that otherwise would have resulted. So we've been able to maintain the quality scientific talent that we have in the labs.

And secondly, let me say I think our three excellent representatives of the laboratories have been a bit too modest, so let me brag on their behalf. I believe that this partnership and this investment, which is the largest private sector investment in any (CRADA) in the Department of Energy's Laboratories is an enormous vote of confidence and I think a very strong statement by the private sector that government laboratories are quality laboratories, that our scientists, that our engineers are best in the world. And that is a statement you don't hear very often. Sometimes people wonder about the quality of work of their government.

So I want all the laboratories to know that as Secretary of Energy I am very proud of their work and very honored that the private sector has chosen to invest a very substantial amount of their money in government laboratories. So I think that needs to be said.

Peter Weiss: Secretary Pena, do you think that this is a model of working with industry that ought to be carried out at the other labs as well?

Federico Pena: Yes, in two ways. One is the virtual laboratory that has been established among the three labs is something that we're also trying to do among the other labs.

For example, when the President challenged us when he addressed the United Nations on global climate change, we have a conversation with all of our laboratory directors about how they could work together virtually as a virtual climate change lab nationwide to work on new technologies for climate change. So here we have three of our labs putting together this virtual laboratory system and a very unique partnership and (CRADA) with the private sector which I think will be a model for future (CRADAs) in the next several years.

Operator: Well go next to Gene Koprowsky of Wired News.

Gene Koprowsky: Yes, this is a technical question for Dr. Moore. The question would be regarding the relationship between extreme ultraviolet and x-rays. There's some controversy I understand in the scientific community that EUV technology is just a nice new name for x-ray technology with which you've not had much big success so far. So I wanted to get some comment on that first. And then I had another question.

Dr. Gordon Moore: There's certainly not general agreement in the industry which of the various candidate technologies will succeed. Clearly, we at Intel think EUV has some real advantages and is a very strong contender. But there are other companies who have been working on the x-ray shadow masking technology for a long time and, as a result, have made some significant progress, but not enough to take it yet to a production level.

But, you know, this is something that we can't really predetermine. We have to find out if the things that are required in this EUV system are really going to be something that can be made practically on a production scale.

You know, really the quality of the optical surfaces are better than anything that has ever been made previously. And we want to make hundreds of those.

The light source is truly unusual. I think there's lot to learn about that kind of technology.

So we really can't prejudge if this is going to be the replacement for optical lithography. But, you know, that's the nature of advanced development and of research is that you make your best technical questions, and you put strong efforts behind them in an effort to be sure that eventually you do understand and pick the right horses.

Gene Koprowsky: Does this investment mean that the industry is no longer seriously considering the four other alternatives or this is just one additional alternative that you're ratcheting up?

Dr. Gordon Moore: This is one additional alternative. And my hope, certainly, is that the efforts going into the other alternatives will continue until there is a clear winner.

And you know, there are major companies supporting other approaches. IBM for example for years has had a major program in x-rays. They have a (Synchrotron) dedicated completely to the development of the lithography system. Bell laboratories on the other hand has had a strong on-going effort that relates to direct writing with an electron beam.

I hope those programs continue to get major support. And there are several other programs in the industry, not all in the U.S. by any means. Because the semiconductor industry has a major problem to solve here. And eventually we'll figure out what the right choice is.

But we think EUV has a lot of things going for it. And, you know, clearly that's where we're going to put our money at the moment.

Operator: George Avalos of Contra Costa Times News.

George Avalos: Hi. I was wondering if you could talk a little bit about what might be some of the practical applications of the computers that would emanate from this sort of a chip- be based on these sorts of chips not only in- I'm looking at applications there? You just kind of speculate not only applications, you know, in laboratories, or scientific, or medical or industrial design applications but, you know, what can talk about as far as what changes you might see in terms of home computers?

Dr. Gordon Moore: Okay. Well, you know, this is always a question what's ((inaudible)) of these things going to be. If I look back historically in my record of predicting the important impact, it's really very poor. You essentially kind of extrapolate what's going on today.

But, you know, one thing I see with considerable more computing power in the desktop computers and the like as a very attractive thing is really good speech recognition. Good speech recognition is going to require the recognition being in context of what is said. Once a computer can recognize what you say in context, it can hold an intelligent conversation with you. And then things like language translation of high quality become easily possible.

I think having good speech recognition, first of all, will let a lot more people use computers, because the keyboard is still something that is a barrier to a lot of people. And I think the capability that will give us will really expand convening in a lot of directions that we can't even think of right now.

And I'm sure 20 years from now that won't turn out to have been the most important thing that we get from this. But it's an idea of the kind of thing that becomes possible.

Operator: Jack Robertson, Electronic Buyers News.

Jack Roberts: Dr. Moore, what is the percentage that each of the partners in the EUV corporation is contributing? And when would you hope to have a beta machine- a machine developed for beta testing?

Dr. Gordon Moore: Jack, the first question is not information we're giving out. The second question, when do we hope to have a beta machine, you know, 2002 is when we need something we can really to developing our technology on. Anything earlier than that would be greatly appreciated.

Operator: William Fellows, Computergram.

William Fellows: Hello there. I guess I'm trying- I'd like to understand this from a commoners point of view, if you like. When are we likely to see chips manufactured in these processes? How small are they going to be and how many transistors? Is Moore's Law broken, and is it going to be rewritten?

Does the agreement preclude other investor's coming on board? And what about other types of technology such as (Exemer) laser and so on?

Dr. Gordon Moore: Well, Moore's Law has been stated in so many different ways, one of them is bound to be right. I take credit for all of them.

William Fellows: So how should be think about it now?

Dr. Gordon Moore: You know, there are some fundamental limitations eventually. We have to contend with the atomic nature of matter, for example. And the best laboratory results today suggest that below something like .05 microns the transistors don't behave very well anymore. You know, we may be able to push that a bit. These limits have a habit of receding as you get close to them. But if we stay on the rate of progress we've been on historically, sometime in the 2010 to 2020 range I expect that we'll run up against some real limits.

As far as the question on more players, we'd certainly like to have more players now. We will invite other U.S. companies to participate.

William Fellows: How about between responsibilities?

Dr. Gordon Moore: And hopefully we'll expand the size of the Limited Liability Corporation.

William Fellows: And have other companies been approached already?

Dr. Gordon Moore: Yes. We've approached the companies fairly generally. You know, asking a company to increment its R&D expenditures and the program they went looking at the middle of a budget year is always difficult. This is an expensive program. We hope we'll get more as we start showing favorable results.

William Fellows: Sure. How small can this stuff go? If you use this stuff, how small can you make a chip? And how many transistors could you get on it theoretically? Using this process how small can you make the chips? And for example, how many transistors could you fit on a device manufactured in this process?

Dr. Gordon Moore: Certainly, microprocessors with several hundred million transistors on them would be possible with this technology. I think Secretary Pena had a billion transistors in one of his examples. This is the kind of number that's realistic to look toward. And of course, as this gets applied to memory chips, you'll get huge amounts of memory on a single chip. Maybe the terabit DRAM is a possibility.

Federico Pena: We would all like to thank all of you very much for giving us an opportunity to share this very exciting announcement with you.

Again, I want to congratulate all the partners to this historic agreement and look forward to a very successful implementation of the partnership.

So thank you all very much for being with us this afternoon. And we'll look forward to continuing this dialog in the future. Thank you very much.

Howard High: If anyone would like more information, they can contact Bill (Wicker) at the Department of Energy -- his phone number is 202-586-5806 -- or the representatives that are listed in the press release from the National Labs, or the individual companies at Advanced Micro Devices, Intel or Motorola.

Thanks again for joining us today.

* Other names and brands may be claimed as the property of others.