The story about Rambus becomes very clear from Mr. Monahan's opening statement to the Virginia Court - here's the official transcript:

 

MR. MONAHAN:  May it please the Court,

16      counsel, ladies and gentlemen of the jury.  I’m David

17      Monahan.  It’s my privilege to represent the plaintiff

18      in this case, Rambus.

19      This is a case about inventions.  It’s also

20      a case about two brilliant inventors who have

21      revolutionized an industry, an industry driven by

22      technology.

23      It is also a case about infringement.  That

24      is, using some of those inventions’ end product,

25      memory chips, while willfully refusing to pay for that

22

 

1        use.  The inventors are the founders of Rambus, the

2        plaintiff in this case.  The infringer is Infineon,

3        the defendant in this case, until quite recently part

4        of a company, a large multi-national conglomerate

5        called Siemens.

6        Let me tell you how these two inventions

7        came about.  Two teachers meet for dinner.  Back in

8        October 1988, Palo Alto, California, near Stanford,

9        the heart of Silicon Valley, two teachers met for

10      dinner at a restaurant called St. Michaels.

11      One of them, the one who called the meeting,

12      had grown up on a farm in Indiana.  He had gone to

13      Purdue, graduated early from high school, went on to

14      Purdue, an engineering university.  Brilliant there.

15      Then went on and got his Ph.D. in electrical

16      engineering at Stanford.  At this time he was a

17      teacher.  He was a professor at the University of

18      Illinois, and he had an idea.  He had an idea that he

19      had been thinking about for sometime.

20      He basically had a vision that would

21      revolutionize how computers work.  He had a vision

22      that would make them run at speeds that were then

23      unheard of - 500 million cycles per second.  That’s at

24      a time when the fastest were running something like

25      20 million cycles per second.

                                             23

 

1        He not only had a dream for the technology,

2        he had a dream for bringing this technology to the

3        world through licensing, but he knew he couldn’t do it

4        alone.  He needed some help.  And that brings us to

5        the second person at that restaurant meeting at St.

6        Michaels nearly 13 years ago.

7        Another brilliant student, another person

8        who graduated from high school early, went on to the

9        Massachusetts Institute of Technology, MIT, electrical

10      engineering, awarded his Ph.D. from Stanford in

11      electrical engineering.  Also a teacher.  Then, as

12      today, a full-time professor at Stanford.  Is name is

13      Mark Horowitz.  The first man was Mike Farmwald, the

14      inventors.

15      So Mike explained his vision to Mark, and

16      Mark was a little skeptical.  He was skeptical not

17      only of the vision for this technology, he was

18      skeptical on the way that Mike was planning to bring

19      it to the world.  But as he listened, and as he asked

20      questions, he began to think, Maybe, maybe we can do

21      this.

22      So starting with that dinner at St. Michaels

23      Restaurant nearly 13 years ago, these two inventors,

24      teachers, came up with these inventions and then

25      taught the world how to use them.  These are

                                                  24

 

1        inventions that have radically changed the way

2        computers work for the last decade and may change it

3        for years to come in the future.

4        To understand what these inventions do - I

5        could use some of this technology right here - we have

6        to understand a little bit about basic components of

7        the computer.  We all use them now.  We saw your hands

8        go up.  There is something called a CPU on every

9        computer.  Sometimes that’s called a microprocessor.

10      That’s the brains of the computer that does all these

11      complicated functions.

12      There’s something also called main memory.

13      And main memory in your computer is almost always

14      dynamic random access memory.  That’s a mouthful, so

15      I’m going to call it DRAM.  That’s what the products

16      are in this lawsuit.  That’s different from the

17      long-term memory called the hard drive.

18      This is the main memory where you load

19      things, where you type things in, where you read

20      things out on your output device, it says here.

21      That’s your display or monitor or your input device.

22      That’s your computer keys or your mouse.  Isn’t

23      technology great?

24      What is a CPU, again?  That’s that little

25      chip.  You probably know them as microprocessors.

                                                        25

 

1        Intel invents them.  The next one coming out and we

2        will all get one of these days is called P4 or Pentium

3        4.  That’s the brains of the computer.  It has to

4        store things on a short-term basis somewhere and

5        that’s where DRAM or main memory comes in.

6        There was a problem that they were trying to

7        solve.  And it’s called the performance gap.  What

8        that means is that the speed of these CPUs or

9        microprocessors had gone up 200 times in the previous

10      decade.  They were getting very fast very quickly.

11      The problem was that the DRAM main memory that had to

12      communicate with these microprocessors or CPUs weren’t

13      getting much faster, certainly not at that rate.  In

14      the same decade they had gone up perhaps 20 times.

15      So there was a gap that was growing between

16      the speed at which the CPU could run and the speed at

17      which the main memory could run.  And that is where

18      the Rambus inventions were directed.

19      All of the inventions in this suit, and I’m

20      going to give you four names, it’s three patents, are

21      directed at improving the way the CPU communicates

22      with the DRAM main memory.  So think of those arrows

23      like a highway.  And if there’s congestion on that

24      highway, it can slow the communications down.  And

25      that’s what all of these inventions were aimed at.

                                                   26

 

1        I am going to give you some shorthand names

2        to use for these inventions.  First, they work

3        together.  The inventions alone are important, but

4        it’s the inventions in combination that give us these

5        radical speeds, this improvement in speeds between the

6        way the CPU and the DRAM memories operated.

7        The first invention I want to talk about is

8        called the delay time invention.  The delay time

9        invention is like putting a stoplight on each of these

10      DRAM memory chips.  We don’t like stoplights.  They

11      make us stop.  They slow us down.  They aren’t very

12      nice things, but we do know that if they are properly

13      run and properly placed, traffic overall runs better.

14      So they benefit us all.  Same way with these

15      inventions.  By stopping the flow of data at the right

16      times, it makes everything run faster.

17      The second one of these inventions is called

18      variable, sometimes programmable, block size.  That’s

19      block size.  That’s the shorthand name.  And this is

20      like a metering signal on the expressway.  When you

21      have an on ramp and it says stop, and then you go on

22      green.  Well, sometimes they say one car go on green.

23      Sometimes they’ll say two cars go on the green.

24      Sometimes they can say three cars go on the green.

25      And that’s exactly what this variable block size does.

                                                 27

 

1        It says, okay, we’ll give you three bits of

2        information that can go on the green.  Same kind of a

3        deal.  You don’t like stopping, but this is a way of

4        improving the flow of data.

5        The third is we’ll call dual-edge clocking.

6        Sometimes called double dataing.  What that does—I

7        couldn’t think of how to do it with a light, so I’m

8        going to do it with a grandfather clock.  Think as

9        you’re sitting there on that expressway, instead of a

10      light, there’s a giant grandfather clock.  It’s

11      ticking back and forth.  And you can go—every car

12      can go on a tick.

13      All right.  This particular invention,

14      dual-edge clocking says, okay, now we can have a car

15      go on the tick and we can have a car go on the tock.

16      It basically doubled the rate at which we were getting

17      traffic on the expressway.  Now, that sounds simple,

18      but it’s a very complicated job in these devices to

19      make them work.

20      The fourth invention is called a delay lock

21      loop.  Now, the inventors didn’t invent that.  That

22      was a well-known circuit at the time.  Engineers have

23      known that for a number of years, but putting a delay

24      lock loop on a chip, on a DRAM chip, was an invention,

25      and it was their invention, and what it does is it

                                                  28

 

1        makes everything run.  All the circuits run much more

2        precisely than they otherwise would have run.  And

3        people didn’t want to use them because, first, they

4        are complicated and, second, they take up a lot of

5        what they call real estate or space on the chip.

6        So it’s like the traffic metering lights on

7        the highways.  If they are not set precisely to work

8        precisely right, they hurt the situation instead of

9        help it.  So that’s what the delay locked loop was

10      supposed to do.

11      The thing was when these investors told

12      people they plan to put a delay locked loop or DLL on

13      every one of these chips, they were laughed at.  They

14      were scoffed at.  They said, “No, you can’t do that.

15      There’s a lot of reasons you can’t do that.  That

16      doesn’t make any sense.”  In fact, as late as 1996,

17      engineers were still telling them at some of these

18      major companies, “You can’t do that.  It doesn’t make

19      sense.”

20      But the Rambus inventors knew a dozen years

21      ago what all DRAM designers know today, and that is

22      you can’t get these kind of speeds that they were

23      envisioning without the precise timing that this

24      circuit gives them.  So those are the four inventions

25      in this suit.

                                         29

 

1        The inventors wanted to bring their

2        inventions to the world.  So they had to think of a

3        way to do that.  So they came up with a business

4        model.  More accurately, it came up with them.  These

5        two college professors didn’t have the money to start

6        a company, to compete with the giants in this

7        particular industry.  The industry that makes DRAMs.

8        It costs hundred of millions to build these plants to

9        make these products.  So they knew they weren’t going

10      to bring their inventions to the world that way.

11      Instead, what they came up with is an idea

12      how they would license their inventions, and here’s

13      how it worked:  They would apply for patents.  Patents

14      take years to get.  But in the meantime, they would

15      enter into contracts with these DRAM makers to make

16      what we call or what they called Rambus DRAMs or

17      RDRAMS, and they would license all of their technology

18      to a company such as a DRAM company, and they would

19      not only say, “Here it is,” they would work with them

20      closely.  They would have their engineers work with

21      them and show them how to use their technology into

22      making these chips.

23      They would be partners.  And they called

24      their licensees partners because they would work

25      together to make these products, make their inventions

                                                         30

 

1        work with the products.  And they didn’t make any

2        money for doing that.  Where they made their money was

3        if their licensees, their partners, made chips, sold

4        them to systems companies, such as Dell or Compaq or

5        Nintendo that used these memory chips, then and then

6        only would they make money.  They would make royalties

7        based on those sales of DRAMS to these system

8        companies, and they would come back to Rambus.  That

9        was their business model.  That was their business

10      model from the start.

11      Now, the business model involved patents.

12      Here’s how patents work.  They have been with us since

13      the start of our country.  They are authorized by the

14      Constitution.  They are a legally recognized ownership

15      right, and they are a type of property, and they are a

16      property that comes about this way:  The government

17      encourages people, inventors, to make their inventions

18      public so other people can study them.  They can use

19      them and build on those inventions in the future.  In

20      exchange what the government does is says, okay, we’ll

21      give you something called a patent.  We’ll give you a

22      patent.  And what you can do with this patent is you

23      can have the right for a limited period of years to

24      exclusively use or say who gets to use your

25      inventions.  And typically that’s done by licensing.

                                                    31

 

1        So that’s the reward for applying for a

2        patent.  If you get the patent, you get some property.

3        And it’s called intellectual property.  And it’s

4        property just like a house.  It’s property just like a

5        farm.  It’s not real property.  It’s intangible

6        property called intellectual property, but it’s the

7        same in the sense that you can lease your farm for

8        rent.  You can lease or license, it’s called, your

9        patents for royalties.

10      So Mike and Mark worked for months.  And

11      what they did is they wrote a patent application.  I’m

12      holding it in my hand.  And what it disclosed was all

13      of their inventions.  They put it in one application.

14      150 claims at the end of this.  And they filed this

15      April 18th of 1990, and it disclosed everything.

16      Now, a patent is very much like a patent

17      application in the sense that they both have two

18      parts.  One part of the patent application is called a

19      written description.  The second part is called the

20      claims.  Well, Mike and Mark wrote the written

21      description, and the written description was written

22      for engineers, so they can understand how to make and

23      use these inventions.  And if a patent is issued, then

24      that’s what the engineers turn to.  So that’s what

25      they spent their months on literally at the kitchen

                                                    32

 

1        table writing this patent application.

2        The Patent and Trademark Office says,

3        “You’ve got too many inventions in there.  You have to

4        break them up.”  So ultimately there were a number of

5        patents that resulted from this original patent.  One

6        of those patents is, for example, a patent-in-suit.

7        It’s called—there are seven numbers in a patent.

8        I’ll just use the last three.  It’s called a ‘263

9        patent.  The ‘263 patent has the identical written

10      description as the original.  There’s a few little

11      changes in the drawing, but substantially it’s the

12      same written description, the same teaching, the same

13      disclosure of inventions as in the original filings.

14      It’s 29 pages, and all but the last two pages are the

15      written description, including drawings that tell

16      engineers how to make and use that invention.

17      Now, other patents were issued as a result

18      of this initial filing that aren’t involved in this

19      suit.  And one of them we’ll hear about later is

20      called the ‘703.  The ‘703 has about 25, 26 pages, and

21      again, all but the last two are the same as the

22      patents-in-suit, are the same as the original filing

23      in April of 1990, and that’s the so-called written

24      description part of the patent.

25      Now, they had to, to get licenses to make

                                                   33

 

1        this business model work, they had to get out and they

2        had to disclose their inventions to the industry.  So

3        here are two teachers on a big teaching job.  They are

4        going to teach the industry about their inventions.

5        They decided they would visit every semiconductor

6        company that would listen to them and try and persuade

7        them to license their inventions.  They did this

8        starting in 1989.

9        But to protect against one of these

10      companies stealing their inventions, they relied on

11      written agreements, contracts of a special nature

12      called nondisclosure agreements or, for short, NDAs

13      Before they would say anything to any one of these

14      companies, they would say, “You have got to sign this

15      nondisclosure agreement or NDA.”  And they did.  They

16      signed NDAs with over 60 companies in that first year

17      and a half, two years.

18      Once they signed the NDA, they would tell

19      all.  They would reveal all their documents.  They

20      would reveal something called their technical

21      description that had everything that was in the

22      patents and more.  Some of the companies wanted to see

23      their patent applications that are confidential, and

24      they let them see them under this NDA.

25      They had presentations.  The inventors would

                                              34

 

1        explain this is how our inventions work, and this is

2        how you can make and build products using them.  And

3        these presentations would be attended by groups of

4        engineers for these various companies, and they would

5        ask them questions.  They would delight in trying to

6        stump them, but they weren’t able to stump them much.

7        And so the initial impressions, this can’t

8        happen, you can’t make this kind of a leap, started to

9        melt after they were on the road and talked to all

10      these semiconductor companies.  It melted sufficiently

11      for them to sign up to the biggest in the world.  They

12      originally had what they called a three continent

13      strategy.  They wanted somebody in Europe.  A DRAM

14      company over in Europe.  They wanted a semiconductor

15      company here in the United States, and most of the

16      business was in Asia at that time, so they wanted a

17      couple of the companies over in Asia.  That was their

18      initial strategy.  They wanted to get these partners

19      to work with them to bring their Rambus compatible

20      products out to generate royalties that would again be

21      the only source of revenue for this company they were

22      starting called Rambus.

23      The reaction was expressed in different ways

24      when they first made their pitches.  In fact, one of

25      the Japanese companies, a couple of them, were just

                                                         35

 

1        politely silent when they told them what they could

2        do.  A Korean company wasn’t quite so subtle and

3        actually accused them of lying when they said they

4        were going to make these computers running at 20

5        megahertz at a time go 500 megahertz.

6        But finally, they won them over, and they

7        signed their first DRAM company, which was Toshiba in

8        1990.  And soon it was followed by Fujitsu, and not

9        long after that by NEC.  These are all giants in the

10      industry where probably less than a dozen companies

11      make almost all the supply - the world supply of

12      DRAMs.

13      So there was no mistake about using their

14      inventions improperly before patents were issued, they

15      required their licensees who had signed these

16      nondisclosure agreements to promise in the license not

17      to use their inventions for anything other than this

18      special type of DRAM called a Rambus DRAM or RDRAM.

19      And just for a good measure, they put field of use

20      restrictions in these licenses that says, “You cannot

21      use our inventions or any part of them for any

22      competing parts.  Just the parts these Rambus DRAMs

23      are going to earn revenues for us.”

24      Finally, they signed up everybody.  In fact,

25      even Infineon signed up in 1997.  But I’m getting

                                                    36

 

1        ahead of myself.  One of the first companies, I think

2        the first DRAM company that the inventors talked to

3        was Siemens.  That was in February of 1990 before they

4        even had filed their initial patent applications.

5        What they did is they went to this company.  Again,

6        it’s a large company.  It has hundreds of thousands of

7        employees and does business in the tens of billions of

8        dollars a year, and had at that time something called

9        Siemens Semiconductor.

10      So it went to Siemens Semiconductor and

11      said, “We’d like to tell you about these things, but

12      first, you have to sign a nondisclosure agreement.

13      And if you sign it, then we’re going to give you

14      information that will let you decide whether you want

15      to be one of our partners.  We’d like to have you as

16      our European partner.

17      Of course, Siemens signed.  They signed this

18      agreement in February of 1990.  And over the course of

19      the next almost two years they provided information to

20      Siemens.  They thought it would take Siemens until

21      maybe April of 1990 to make a decision on whether to

22      sign a license with them or not.  Siemens never did

23      sign a license at that particular time.  But they were

24      interested in the technology.

25      After signing a license agreement in

                                                  37

 

1        February of 1990, there were a lot of exchanges.

2        There were personal meetings.  There were

3        communications by phone.  There were communications by

4        letter, fax, for the most part, and it was an exchange

5        of—it wasn’t an exchange, it was a one-way movement

6        of information of the Rambus interface technology.

7        Information about its inventions, all the details

8        about its inventions, how to use those inventions

9        going from Rambus to Siemens.

10      The Siemens people were generally positive

11      with respect to Rambus and encouraged them to give

12      them more and more information, but what we know now,

13      didn’t know then, is that they weren’t interested in a

14      license at the time.  What they were interested in

15      from the outset was using selected portions of these

16      inventions in Siemens memory chips.

17      Here’s a letter.  It’s the 28th of February,

18      1990, and it’s about a visit by Mr. Davidow, and a

19      number of people participated including a Mr. Penzel,

20      the head of Siemens Semiconductor, and Dr. Horninger.

21      Here’s what they found.  Investigations with

22      regard of implementability of Rambus technology into

23      the Siemens particular size DRAM design need to be

24      continued.

25      Continuing on, further investigations need

                                                    38

 

1        to be performed with regard to application of Rambus

2        technology to video DRAM, not a Rambus product.

3        Here’s letter to Dr. Horninger with copy to

4        head of Siemens Semiconductor, Mr. Penzel and others,

5        March 6, 1990.  “Do you think that the Rambus model

6        may be applied to our”—then it goes on to give some

7        numbers—“without changes being necessary to our

8        existing DRAM designs.”

9        So they were interested.  They were

10      interested for the wrong reason.  They were interested

11      in using portions of the Rambus inventions in their

12      chips, not in making Rambus products at that time.

13      Other memos widely distributed to the

14      highest executive levels in Siemens concerned the

15      positive impression they had of Rambus technology and

16      the positive interest they had.

17      After nearly six months of effort without

18      hearing anything from Siemens about a license,

19      Dr. Davidow writes a letter and says, in July of 1990,

20      he says, “Well, I’m sorry to hear that you’re not

21      interested.  Or more accurately, I haven’t heard

22      anything from you for a while.  Will you kindly send

23      our confidential materials, all of it that we’ve sent

24      to you, back.”

25      Immediately he got a letter from the head of

                                                      39

 

1        Siemens semiconductor, “No, no, we’re interested.  We

2        just need more information.”  And Rambus obliged.

3        It sent them their latest technical

4        description.  It made presentations to them.  They had

5        more personal meetings, a number of them.  Everything

6        on top of that line are communications from Rambus to

7        Siemens.  Everything below the line is what was going

8        on at Siemens, or at least what we know was going on

9        at Siemens at the same time.

10      We now know that the inventions in this

11      suit, particularly the delay time register and this

12      block size invention, were of particular interest to

13      Siemens.  They were all over the information, their

14      engineers were.  They even made drawings using parts

15      of the Rambus inventions.

16      Here’s a drawing made by a Mr. Michael.

17      They do their dates a little differently there.  And

18      it’s September 23, 1990, and it shows access time and

19      block size.  These are two of the four inventions that

20      I told you about.  The Rambus inventions.

21      Incorporated not into a Rambus product, but

22      incorporated into a proposed Siemens design.

23      Later in 1991, Siemens brought in—it’s

24      the central research labs.  The central research labs

25      for this very large company.  It wasn’t just for

                                               40

 

1        semiconductor division central research labs, it was

2        central for the entire company.  And they said, “We

3        want you to do a further analysis of what we have

4        here, this Rambus technology.

5        So in a letter, actually a memorandum,

6        July 24, 1991, they get a report back from central

7        research labs.  I’ll read it to you from here.  Here

8        we go.  “In recent years”—this again is Munich,

9        July 24, 1991.  In recent years the access time of

10      DRAM memory chips could not keep up with the

11      requirements of modern microprocessors.  The same

12      problem they were talking about back in St. Michael’s

13      Restaurant in October of 1998.

14      Then it goes on to say, Rambus is a new type

15      of approach for an efficient memory interface.  The

16      semiconductor’s department has asked, and then all

17      that alphabet soup, the central research labs, for an

18      analysis of the suitability of the Rambus concept from

19      a system perspective.

20      They got that analysis.  They continued.

21      The Rambus, as a memory concept, represents a

22      technological step forward.  This is straight from

23      central research labs at Siemens.

24      Here’s a report they gave in July of 1991.

25      And it has a conclusion on the last page that I think

                                                     41

 

1        you might be able to read.  I’ll read it to you.  It’s

2        in the lower left-hand corner.  It’s a slide

3        presentation.  Rambus is technically and conceptually

4        a step further ahead but is that enough for market

5        acceptance?  They were wondering about that then.  And

6        in the bottom they asked the question, this is central

7        research:  “Is Rambus a jewel?”

8        Now, after this years of study, 1990 and

9        1991, by the time they got to 1992 they were ready to

10      act.  And they had an idea, an idea that others

11      suggested to them perhaps, and that was how could they

12      use part of the Rambus inventions in their products,

13      their SDRAM products that they were designing.  And

14      what they did was they had an idea that they were

15      going to make it public domain.  And here’s where they

16      got it.

17      They had a meeting back in April of 1992,

18      and in this meeting there was a report.  Actually it

19      occurred on April 10 of 1992.  And there was a report

20      on the meeting.  The participants included some of the

21      top executives at Siemens.  I’ll just tell you what

22      the report said.  The report basically said—well,

23      here it is.  It’s working now.

24      NEC and Samsung have announced samples for

25      the end of ‘92, and TSB, that’s Toshiba, for ‘93,

                                                         42

 

1        public domain version of a Rambus memory, exclamation

2        point.

3        So, this was Siemens’ view of these new

4        samples.  And the evidence will show that this new was

5        because these new samples, these Sync DRAMs that came

6        out, were apparently using Rambus inventions.  And

7        that’s the conclusion the evidence will show that

8        Siemens drew at that particular time.

9        So, Rambus, they say, currently no action

10      required.  Preliminary results of our design study of

11      Sync DRAM show great similarity with Sync DRAM that is

12      adjacent to Rambus.  Great similarity between Rambus

13      and Sync DRAM.  Then they enclose a comparison between

14      Siemens and IBM positions.

15      Now, the reason that they were working with

16      IBM at that time is because they had a problem.

17      Despite its huge size Siemens had a big problem.  It

18      didn’t have the design engineers to do a new SDRAM

19      design on its own.  It had a number of reasons that it

20      was incapable of doing an SDRAM design on its own.

21      I’ll let Siemens executive, Mr. Bidler, explain that

22      to you in his owns words.

23      (At this time a videotaped segment of Mr.

24      Bidler’s deposition is played for the jury.)

25      MR. MONAHAN:  So what they needed was help,

                                                 43

 

1        and they looked first to a partner.  They had gone

2        into an alliance with IBM.  And they were looking very

3        heavily to IBM to jointly design this new Sync DRAM

4        product with them.  They worked primarily with a

5        fellow named Gordon Kelley at IBM.

6        So they got together with IBM in April of

7        1992, and here’s what they talked about.  Here’s a

8        report of that telephone conference on April 29, 1992.

9        And again the participants were Gordon Kelley, Mr.

10      Meyer, you’ll hear more about Mr. Meyer, Dr. Peisl,

11      and then kept everybody else, including Mr. Penzel,

12      the head of Siemens Semiconductor, informed.

13      Rambus visited the most important in-house

14      users at IBM.  Rambus is still being observed by IBM.

15      Rambus has announced it is demanding 10 million from

16      Samsung because of similarity of SDRAMs with the

17      architecture of Rambus memories.  IBM is therefore

18      seriously considering purchasing a license as soon as

19      possible and as a precaution (at the introductory

20      price.)  So this is what Siemens and IBM were

21      discussing about Rambus back in April of 1992.

22      At this same time, though, there was a

23      confidential report that was issued within Siemens and

24      basically there was a decision made at this point.

25      We’re going to copy Rambus.  Not all of their

                                                  44

 

1        inventions.  We don’t want all of them.  We’re going

2        to copy some of them.

3        And here’s what they reported April 30,

4        1992.  This is a report, April 30, 1992, and a summary

5        appears on the first page.  And they are referring to

6        eliminate the memory bottleneck.  Same kind of thing

7        talked about back at St. Michaels in October of 1998.

8        Next.  Moving to page 2, in the JEDEC

9        Committee on MOS memories, attempts have been made for

10      roughly a year now to define a universally usable

11      successor for the previous DRAM, so-called Synchronous

12      DRAM (SDRAM).

13      Next.  Samsung and NEC have announced

14      samples for the end of 1992 and Toshiba for mid-1993,

15      and they are apparently ready to enter the market

16      without waiting for eventual standardization.

17      These are the same ones referred to earlier

18      in the memo in April.  They weren’t standard.  They

19      weren’t interchangeable.  They were going off on their

20      own with these SDRAM parts.

21      Next please.  The original idea of the SDRAM

22      is based on the basic principles of a simple clock

23      input (IBM toggle pin) and the complex Rambus

24      structure.  NEC (Rambus license holder) was the first

25      to suggest a leaner public domain version based on

                                                                45

 

1        this.  Maintain a synchronous controlled two banks,

2        four-fold, internal data bus, four-word register, the

3        data output, and so on and so on, from the Rambus

4        while leaving off the proprietary Rambus control

5        protocol.

6        So they didn’t want all of it.  They wanted

7        to leave some of it off.  They just wanted to pick

8        some of the inventions.

9        It is possible and it is being discussed to

10      go even further and to restrict it to one bank as well

11      as two-fold data bus synchronous control.  Samsung

12      seems to follow this path.  That’s with their Sync

13      DRAMs.

14      In the meaning time, it’s become clear that

15      a Rambus memory can easily be converted into a SDRAM

16      (one or two banks) or conventional DRAM.

17      Next.  All right.  So what else was going on

18      at this time?  IBM and Siemens were continuing to get

19      together and trying to work out what they were going

20      to do jointly to come up with this SDRAM product.

21      They had an alliance.  An alliance Toshiba joined in

22      1992.  But here it’s just Siemens and IBM.  And here’s

23      what they’re saying.  They’re putting together the

24      pros and cons of going with this Sync DRAM or Rambus

25      DRAM.

                                                                46

 

1        Under the pros, they say, for Sync DRAM,

2        it’s public domain.  Anybody can use that.  The cons,

3        Rambus patents.  That’s the cons they were evaluating

4        in May.  May 6 of 1992.

5        And for the pros under the Rambus DRAM they

6        said, well, it’s faster than a Sync DRAM, but the con

7        is Rambus property.  Willie Meyer, May 6, 1992.

8        Now, what their strategy was, basically—

9        Oh, attached to that report I might add was a block

10      diagram of the proposed joint IBM-Siemens SDRAM, and

11      it had the delay time register, the access time

12      register, the same ones I showed you in the earlier

13      drawing right smack in the middle there.  That’s what

14      they were planning to do.  So that was their SDRAM

15      product they were playing on incorporating selected

16      parts, not all, they didn’t want all of the Rambus

17      inventions.  That’s want they wanted to develop with

18      IBM.

19      Their strategy actually was to use a

20      standards group called JEDEC as a placeholder for

21      their proposal.  Because they were behind the Asian

22      DRAM manufacturers, behind by a mile, and because of

23      this lack of resources that Mr. Meyer explained to

24      you, they wanted to be a player in the world market,

25      the world DRAM market.  They didn’t want to lag too

                                                        47

 

 

1        far behind, although they weren’t leaders, but they

2        needed a strategy so they would be able to close the

3        gap somewhat between the Asian DRAM makers and

4        themselves at the appropriate time.

5        So what they did is they said, okay, we’re

6        going to join this effort to make Rambus public

7        domain.  We, IBM, and Toshiba are going to get

8        together, and we’re going to help put what we want

9        from Rambus into the public domain.  And the way they

10      did that was through a committee called JEDEC that I’m

11      going to talk to you a little bit more about later.

12      But they had some—there was some

13      disagreement.  In fact, at one point they appeared to

14      be upset that they weren’t getting enough credit for

15      putting things into this JEDEC public domain standard

16      that they hoped to see.  So there’s a memorandum, and

17      it’s April 1st of 1993.  And the memorandum basically

18      is complaining—I’ll just read it.

19      Dr. VonZitzewitz, Dr. Beinvogl and myself

20      are somewhat astonished at the procedure that has been

21      agreed upon.  I’m sure there are reasons for it.  Why

22      is there not to be a joint IBM, Siemens, TBS - is for

23      Toshiba - presence?  I suspect that one has given

24      priority to Toshiba as author of the first proposals

25      and Siemens-IBM are supposed to introduce any

                                                 48

 

1        modifications that result from in-house of JEDEC

2        discussions in subsequent JEDEC sessions.

3        All right.  So they’re saying Toshiba

4        apparently ought to be the author to take the lead,

5        and then Siemens and IBM are going to clean up with

6        modifications.  And then they conclude, “That would

7        perhaps not wreak quite as strongly of a cartel and

8        might be better received.”

9        So that’s what their thinking was back in

10      1993 when they were picking and choosing selected

11      parts of Rambus inventions to make this public domain

12      SDRAM.  They put their plans on hold for a little

13      while because they wanted to wait and watch.  They

14      wanted to wait and watch what the market did.

15      In a memo October 13 of 1993, they said they

16      are going to look and see what the market does.  They

17      are going to wait and watch.

18      Reasons?  Whether SDRAM remains a niche

19      product, a small segment product, or replaces DRAM

20      will only be decided during the course of 1994.  Both

21      scenarios are equally likely.  And they continue.  A

22      compulsory situation (DRAM no longer sellable) can

23      arise in 1997 at the earliest with two years

24      headstart.  I can’t read the next one.  Start in

25      August ‘94 is sufficient.

                                                               49

 

1        So what they’re saying in ‘93 is we don’t

2        need to go ahead as quickly as we thought we would if

3        we start in ‘94.  That will be plenty.  We’ll be able

4        to maintain the gap behind the Asian DRAM makers at an

5        acceptable distance.

6        Okay.  So that’s what they were thinking in

7        1993.  And here were their plans to make portions of

8        the Rambus DRAM public domain.

9        I told you I’d come back to JEDEC in just a

10      moment, and so I will now.  JEDEC is a

11      standard-setting group.  There are 600

12      standard-setting groups in the United States.  They

13      are not part of the government.  They are private

14      groups that are basically trade associations that act

15      on behalf of their members.

16      But in December of 1991 at the suggestion of

17      Toshiba, one of Siemens’ alliance partners, Rambus

18      began attending meetings of these JEDEC Committees.

19      Now, JEDEC is basically an acronym.  It’s

20      not important what it stands for, but at one point it

21      was Joint Electron Device Engineering Council or

22      something like that.  It’s part of a group.  It’s

23      really just a committee, but it’s part of a group

24      called EIA, a private trade association.  Again, not a

25      government association.  It lobbies the government.

                                                        50

 

1        It does standards work, gets marketing information on

2        behalf of its members.  Rambus never joined the EIA.

3        But getting back to JEDEC, its members were maybe two

4        to 300 at any one time, and on the committees Rambus

5        was attending there would be maybe 50 or 60 attending

6        the these committees, the DRAM or memory committee in

7        particular.

8        The only requirement for going to these

9        committee meetings was you had to pay your dues.  They

10      were held in nice places, usually Hawaii in winter,

11      and it was a good place to see what was going on with

12      your competition, and also to meet personally with

13      your customers.

14      And so Rambus went to several of these

15      different committee meetings starting in December of

16      1991.  The meetings we’re open.  They were

17      non-confidential.  People would bring guests on

18      occasion.

19      Some of the JEDEC members, though, like some

20      of the giant DRAM makers, had a little bit more say-so

21      than others.  The real say-so in JEDEC was held by the

22      council.  Rambus never attended a council meeting, nor

23      did it qualify for membership.  People like Toshiba,

24      Samsung, IBM, NEC, those were the people who ran the

committees and sat on the council and made the

                                 51

 

1        important decisions.  But I told you it was a

2        standards committee.  And some of you may know what

3        standards are.

4        I brought some small standard products so I

5        can put them in my pocket.  This is a flashlight

6        battery.  It’s a standard product because this one is

7        made by Ray-O-Vac, but you know if you go to the

8        store, you find one made by Energizer that you can

9        switch them around and they’ll work pretty much the

10      same way.  So that’s what a standard product is.

11      Since Rambus is a standard product, it took

12      great pains to make sure that whether it’s Toshiba or

13      NEC or Samsung making its Rambus parts, the consumer,

14      Dell, Compaq, Nintendo, wouldn’t know the difference.

15      They would all be interchangeable.  They would act the

16      same.  So that’s how—that’s a standard product.

17      Standards sometimes are set by the market.

18      Somebody gets out in front and then people follow

19      them.  Sometimes they are set by committees.  But the

20      great majority of committee standards don’t result in

21      products.  Much less products with any great volumes.

22      In fact, here we have a committee—

23      May I step up here, Your Honor?

24      THE COURT:  Anything you want to do is all

25      right.

                                                          52

 

1        MR. MONAHAN:  Thank you.

2        Here we have a committee with Rambus first

3        attending as a guest in December of 1991, and that was

4        right after the SDRAM standard setting began.  Rambus

5        noted in May of 1992 that two of their inventions,

6        burst and latency, were being talked about.  And they

7        were asked to comment on their patent situation.  They

8        didn’t have any patents.  The first one wasn’t issued

9        until September of 1993.  But there was a patent

10      policy at this time at JEDEC.

11      In fact, most standards committees would

12      have something called a patent policy.  And that was

13      to try and encourage people to commit in advance to

14      license their patents on reasonable terms and

15      conditions.

16      Now, JEDEC says, “We’re not going to help

17      you determine what reasonable terms and conditions

18      are.”  In fact, the chairman of JEDEC said reason

19      terms and conditions are in the eye of the beholder

20      when asked to explain what that meant.  But the idea

21      was most patents—excuse me.  Most standards are

22      going to have patents.  You’re going to have, you

23      know, cutting-edge technology.  So most patents—I

24      mean, most standards have patents that are applicable,

25      multiple patents that are applicable to them.  And

                                                             53

 

1        what these committees, including JEDEC, were trying to

2        accomplish was, let’s try and get the patents out and

3        people commit to license them on reasonable terms and

4        conditions.  And I’ll say now that Rambus since its

5        inception in 1990 until the present day it’s been in

6        the licensing business, and it has always licensed its

7        patents on reasonable terms and conditions.

8        Some of the people, IBM, for example, one of

9        the committee chairman, Gordon Kelly, said, “We’re

10      going to ignore the patent disclosure rule.”  Said

11      this in March of 1993.

12      Later in ‘93, they just said, “Well, we’re

13      not going to list applicable patents to a particular

14      thing that the committee is talking about.  We’re just

15      not going to do it.”

16      Most of the people weren’t quite so vocal as

17      IBM saying we’re not going to follow.  They just

18      didn’t follow it.  Rambus took the other approach.

19      Rambus disclosed the same month its first patent was

20      issued, the ‘903, it disclosed that to JEDEC.

21      And finally, when it left, it disclosed all

22      of its patents.  Something no other of those companies

23      did.

24      They tried to change the patent standard in

25      October 1993 to include applications.  Up until then

                                                        54

1        and when the standard for the SDRAM was passed it had

2        just been patents.  It had not been applications at

3        all.  But then they kind of confused the situation by

4        every single meeting that Rambus ever attended showing

5        the pre-October 1993 policy that just called for

6        patents, not patent applications.

7        But in any event, Rambus was proud of its

8        first patents and disclosed it immediately at JEDEC,

9        and it’s called a ‘703 patent.  It’s not involved in

10      this lawsuit as a patent-in-suit.  But it’s involved

11      in the sense that all but the last couple of pages is

12      identical with the patents in this suit, and similar,

13      almost identical, for practical purposes with the

14      original filing by the inventors in 1990.  But here is

15      what else the ‘703 did.

16      The ‘703 listed all of the other

17      applications that had the same written description as

18      the patents-in-suit, same written description has the

19      ‘703, same written description as the originally filed

20      patent.  And it says there’s these ten other

21      applications that are now pending in the patent

22      office.  So anybody that looked at this knew that

23      Rambus was pursuing all these other patents.

24      Remember, I said there were so many inventions in the

25      original application that they had to break them up.

                                                     55

1        Rambus was pursuing all these other patents

2        on those same inventions that were disclosed in 1990.

3        Siemens picked it up and looked at it.  In fact, they

4        studied it pretty thoroughly.  But I’m getting ahead

5        of the story at this point.

6        Back to JEDEC.  Rambus starting in May of

7        1992 says, They are ripping us off.  They are

8        partners.  In fact, the licensees, people who had

9        signed licenses and promised not to use Rambus

10      technology for competing parts, were feeding this

11      information into JEDEC.  Feeding it into the public

12      domain for competing parts.

13      Toshiba did this.  Siemens and IBM joined

14      them.  So what did Rambus do?  It told its lawyers,

15      “Make sure you get all the protection we’re entitled

16      to from the Patent and Trademark Office.”

17      So its lawyers wrote some more claims,

18      again, based on the original inventions of Rambus, and

19      it also told them, “Don’t worry.  You’re protected.

20      Let’s see how the prosecution goes.  And as long as

21      the prosecution is continuing, we can continue to add

22      claims.  So we don’t have to do everything now.”

23      They didn’t do a great job because most of

24      their work didn’t hit the mark.  In fact, I don’t know

25      that any of it did.  The patents in suit weren’t filed

                                                              56

1        until 1998, 1999, and the issue in late ‘99 and 2000.

2        Those are the claims that apply to the products that

3        we’re dealing with here.

4        The three patents in suit, I told you about

5        the ‘263.  It was filled October 20, 1998, and it was

6        issued September 14, 1999.

7        The second one we’ll call the ‘214 patent.

8        Just the last three digits.  And that was filed

9        February 19, 1999.  Issued February 29, 2000.  And the

10      third patent in suit is called the ‘918 patent.  And

11      it was filed February 19, 1999.  Issued March 7, 2000.

12      And again, these patents have the same

13      written description as the initial filing, and as the

14      ‘703 for that matter.

15      Now, I’ll finish up the JEDEC story.  Rambus

16      in May of 1995 was asked to comment on its patent

17      situation with respect to another standard.  It was

18      called an IEEE SyncLink draft standard.  And they

19      said, “Why are we being asked to comment on this at an

20      JEDEC meeting?  That’s an IEEE proposed standard.”

21      And they said, “Okay.  We’ll get back to you.”  And

22      they did.

23      The very next meeting they got back and

24      here’s what they said:  September 11, 1995, they wrote

25      a letter.  This letter was shown up on a wall just

                                                                57

1        like this.  It was read, and it was attached to the

2        minutes and circulated to every member.  And here’s

3        what they said about their intellectual property.

4        Everybody knew they were a licensing company.

5        Everybody in that room had had a pitch from them.  And

6        most of the DRAM companies were by then their

7        licensees.  And what they said about their

8        intellectual property was at this time Rambus elects

9        to not make a specific comment on our intellectual

10      property position relative to the SyncLink proposal.

11      This was this draft proposal that was still in

12      process.  Our presence or silence at committee

13      meetings does not constitute an endorsement of any

14      proposal under the committee’s consideration, nor does

15      it make any statement regarding potential infringement

16      of Rambus’ intellectual property.  So if it hadn’t

17      been clear before, it was clear then what their

18      position was.

19      They attended their last meeting in December

20      of 1995, and because of these inquiries about

21      SyncLink, this IEEE, there were some other inquiries,

22      they finally re-evaluated their position and said the

23      gain isn’t worth the gamble.  These people wanted

24      commitments on our licensing policies, and we license

25      on reasonable terms and conditions, but we don’t want

                                                     58

1        to have restrictions.  We license on our terms.  And

2        we negotiate with our licensees, and we don’t want a

3        standards company or organization coming in and

4        interfering with that.  So they decided to leave.  So

5        they did leave.  They didn’t pay their dues in 1996.

6        They didn’t go to any meetings after December of 1995.

7        And they decided that they would send a

8        letter to the IEEE and say, okay, here’s all our

9        patents.  And so they did.  They got their lawyers to

10      check them, and they made sure they listed every one

11      of their patents, not just the Farmwald/Horowitz

12      patent, every patent that they have.  They said, “Here

13      they are.  You can evaluate them.  We’re in the

14      licensing business.  If you want to license them from

15      us, you know where we are.”  So they sent that to this

16      IEEE organization.

17      They also sent a letter to JEDEC, and that

18      letter was sent the 16th of June—excuse me.  The

19      17th of June 1995.  And that letter also had a list of

20      patents attached to it, the same one essentially that

21      had gone out before, but it had gone through a number

22      of drafts.  It had been drafted in March.  It had gone

23      through several iterations in March and for some

24      reason didn’t get out for a couple three months, until

25      June, and in the interim, in those three months, one

                                                                59

1        more patent had issued called a ‘327 patent.  And that

2        wasn’t on the list.  And Infineon, you know, will make

3        a big deal about that.

4        But in any event, that was an oversight.  It

5        was a patent that’s never been asserted against

6        Infineon or any DRAM manufacturer and never will be

7        asserted against Infineon or any DRAM manufacturer,

8        and it’s not involved in this lawsuit.

9        But getting back to the letter, here’s what

10      they said.  Rambus plans to continue to license its

11      proprietary technology on terms that are consistent

12      with a business plan of Rambus.

13      And those terms may not be consistent with

14      the terms set by standards bodies, including JEDEC.  A

15      number of major companies are already licensees of

16      Rambus technology.  We trust that you will understand

17      that Rambus reserves all rights regarding its

18      intellectual property.  Reserves all rights regarding

19      its intellectual property.

20      So it last attended December of 1995.  That

21      wasn’t the last comment about Rambus.  In December of

22      1996, they did their first work in JEDEC on something

23      called DDR SDRAM.  That’s one of the two main products

24      that are involved in this lawsuit.  Rambus is gone by

25      this time when they start working on the DDR SDRAM

                                                                60

1        standards, but they aren’t finished talking about

2        Rambus.

3        Here’s a portion of minutes March 13 and 14

4        of 1997.  Rambus has been gone for quite a while at

5        this time.  And it’s NEC DDR SDRAM proposal, Item

6        No. 844, and they say our first showing was made.  You

7        will be hearing what those things are.  And there was

8        a discussion.

9        Some on the committee felt that Rambus had a

10      patent on that type of clock design.  Others felt that

11      the concept predated Rambus by decades.  Some

12      committee members did not feel that the Rambus patent

13      license fit the JEDEC requirement of being reasonable.

14      Rambus has also told JEDEC that they do not

15      intend to comply with JEDEC patent policies.  That was

16      in the letter that we talked about.  So they are gone,

17      but not forgotten at JEDEC.

18      All right.  What happened next?  After these

19      patents in this suit were awarded in 1999 and 2000,

20      Rambus went out to the industry, to these guys who

21      make these DRAM chips, or rather the giant, Toshiba, I

22      think second only to Intel as far as semiconductor

23      manufacture, came to Rambus.  And it came to Rambus

24      and it said, “We want a license.  We want to license

25      our SDRAM and we want to license our DDR SDRAM

                                                                    61

1        products.”

2        So they sat down and did what business

3        people do, and they negotiated back and forth, and

4        after awhile they came up with a license.  In June of

5        2000 Toshiba, one of the first people to put parts of

6        Rambus inventions into this standard, came to Rambus

7        and they signed a license.  Followed by Toshiba was—

8        and I might add that the Toshiba engineer was the

9        chairman this whole time of this JEDEC Committee.

10      Next came NEC, also an industry leader, and

11      somebody took an active role in JEDEC in getting this

12      JEDEC SDRAM standard, and they said, “Okay.  We want a

13      license as well.  So they negotiated and then NEC took

14      a license.

15      Next came Samsung, the biggest DRAM

16      manufacturer in the entire world.  They sat down.

17      They negotiated a license on reasonable terms and

18      conditions with Rambus.

19      Samsung, NEC and Toshiba, these were the

20      companies that had first been out with the Sync DRAM,

21      had first been cherrypicking Rambus inventions and

22      participated in putting Rambus into this JEDEC

23      standard.  They signed licenses.

24      They were followed soon by Oki, Fujitsu, a

25      new company called Elpida, it’s a Hitachi and NEC

                                                       62

1        joint venture, and Mitsubishi.  So we now have in a

2        very short amount of time after these patents issued

3        nearly half the world supply of DRAMs or at least the

4        makers of them signing up for licenses with Rambus.

5        All right.  What about Infineon?  By this

6        time it’s Infineon.  In April of 1999, Siemens spun

7        off its semiconductor unit and created this company

8        that’s here in court today called Infineon.  Siemens

9        still owns 71 percent of Infineon.

10      But did Infineon have an opportunity to

11      license on these terms that were given to other

12      people?  Well, in June of 2000 they were made an

13      offer.  The opening offer was a little bit higher, but

14      they came down, and Infineon had the same opportunity

15      that Samsung, NEC, Toshiba, and three or four other

16      companies on top of them had to sign license for their

17      SDRAM and their DDR SDRAM products.

18      Now, the evidence will show that they didn’t

19      even really sell their first SDRAM product until 1998.

20      That’s when they started doing this in volume.  They

21      still have yet to sell—maybe that’s not true today,

22      but at some point, pretty quick, they are going to

23      start making volume shipments of DDR SDRAM.  All of

24      these steps were taken after they had this knowledge

25      we have talked about starting in 1990 of the Rambus

                                                                 63

1        inventions, the Rambus business model, and the Rambus

2        patent applications.

3        They were offered the opportunity.  They

4        turned it down.  They were told during the

5        negotiations that if you’re not going to stop using

6        our property, our intellectual property, I mean, if

7        you’re not going to license it, you have to stop.

8        Take your pick.  You can license.  You can stop.  You

9        can’t have it both ways.  They didn’t license.  And

10      they didn’t stop.

11      Basically, what it’s done, what Infineon has

12      done is put its competitors in the same situation as

13      Infineon was trying to avoid.  What it is is this

14      prior competition, these Asian DRAM makers that signed

15      up for licenses and are paying reasonable royalties to

16      Rambus for using these inventions in their SDRAM and

17      DDR SDRAM products.

18      The result is they are now at a competitive

19      disadvantage with Infineon because Infineon is using

20      the inventions and is not paying anything for them.

21      I am going to quickly just touch on a couple

22      of the issues in this lawsuit.  Very quickly.  After

23      Rambus was required to bring this suit to enforce its

24      intellectual property, and that’s what you have to do

25      if someone doesn’t respect your intellectual property,

                                                               64

1        you have to sue in a court like this.  The government

2        gives you the patent, but they don’t help you enforce

3        it.

4        So after Rambus brought this suit, then it

5        was met with a counterclaim that you will hear about.

6        I’m not going to talk about the counterclaim except

7        for just one moment.  This counterclaim accuses these

8        people of fraud, antitrust monopolization.  It accuses

9        them of being racketeers.  All I’m going to say about

10      their counterclaim is during the years of business

11      relationships, during the last few years when they’ve

12      been a Rambus licensee, during the months starting in

13      June of 2000 when they negotiated for a license, not

14      one of these claims or a hint of one of these claims

15      was ever raised.

16      I am now going to address the evidence in

17      Rambus’ lawsuit for patent infringement.  Basically,

18      there’s just three issues.  One issue is:  Are our

19      patents valid?  Did the Patent Trade Mark Office make

20      a mistake in issuing these patents?  And that’s one

21      reason that Infineon says we’re not going to pay you

22      anything for this because we’re not going to take a

23      license on the same terms because your patents are no

24      good.  You don’t own any intellectual property.  The

25      second reason, excuse that they use to excuse their

                                                                   65

1        not taking a license is they don’t infringe.  They

2        say, okay, if you own property, we don’t use it.  Our

3        SDRAM and DDR SDRAM products made to the same

4        standards and Toshiba’s and Samsung’s and NEC’s, and

5        they don’t infringe your—they don’t use your

6        property.

7        We’re going to show they use latency.  They

8        use block size in their DDR part that will be out soon

9        if they are not out already.  They use double data

10      rate or this dual-edge clocking.  And they use the

11      delay locked loop or DLL chip.  We’re going to show by

12      their own data sheet that they use every one of these.

13      They do infringe.

14      And then they finally say, well, we don’t

15      have to take a license because of this JEDEC contract

16      of yours.  Here’s what they say.  That JEDEC and its

17      members, including Infineon, developed and adopted

18      standards in justifiable reliance on Rambus’ silence,

19      conduct and/or misrepresentations, that SDRAMs and DDR

20      SDRAMs manufactured pursuant to such standards did not

21      and would not infringe any patents of Rambus, and/or

22      that such patents would not be asserted against JEDEC

23      members.

24      Well, in English what that says is Rambus

25      hoodwinked us.  It hoodwinked us into building their

                                                                     66

1        parts, into thinking that Rambus’ technology licensing

2        company was either not getting any patents or it was

3        abandoning any right to seek royalties for whatever

4        patents it did get.

5        The evidence will reveal a very different

6        story.  I’m going to touch just on part of it before I

7        conclude.  First, since February of 1990, Rambus has

8        been telling them about their inventions, told them

9        for a year and a half how to use those inventions in

10      end products, and it told them we’re applying for

11      patents.  That’s how we’re going to make our money.

12      We’re in business, and we’re going to obtain our

13      revenues for this business from licensing our

14      technology.  Started telling them that in 1990.

15      In 1992, as I’ve touched on, when Samsung,

16      NEC and Toshiba came out with these Sync DRAM parts,

17      Siemens recognized them as stripped down or leaner, as

18      they called them, versions of Rambus and, remember,

19      IBM said, well, Rambus is demanding data from Samsung,

20      and we’re considering a license.

21      In 1992, two of the three DRAMs, SDRAM

22      alternatives that they were looking at, that is SDRAM

23      and Rambus DRAM, they had the pros and cons charts.

24      You’ll see three of those pros and cons charts.  And

25      the pros are always “SDRAM is public domain,” and the

                                                                   67

1        con for Rambus is “It’s Rambus property and Rambus

2        patents.”  You’ll see those charts.

3        You’ll hear evidence of how in May of 1992

4        right after this pro and con chart and this concern

5        for Rambus patents that they tried to get Rambus to

6        reveal its confidential patent applications at a JEDEC

7        meeting, May 7th of 1992.

8        They did this despite knowing that JEDEC,

9        the JEDEC chairman felt then - I assume he felt then,

10      he felt years later - that people wouldn’t disclose

11      claims in a patent application in a JEDEC meeting.

12      Here’s what he says:  He says that—he

13      says that we have there legal guidelines that call for

14      use of a patented item or process.  That is a standard

15      that calls for use of a patented item or process may

16      not be considered by a JEDEC committee unless all of

17      the relevant technical information covered by the

18      patent or pending patent, he’s talking after

19      October ‘93 here, this is in May of 1996, covered by

20      the patent or pending patent is enough, which is never

21      possible because the person who is processing the

22      patent application won’t disclose the claims because

23      it corrupts his ability to protect his claims.  That’s

24      what the chairman, Mr. Townsend, of this committee

25      felt in 1996.

                                                       68

1        And so the Rambus representative didn’t

2        fall—didn’t take the bait, and he didn’t reveal the

3        patent position or confidential patent position at

4        that meeting.

5        But let’s go on.  Were they concerned after

6        the meeting about Rambus patents?  On June 11th of

7        1992 in a telephone memo of a conference that took

8        place on June 10, 1992 between Willie Meyer of Siemens

9        and Gordon Kelly of IBM, they talked about this table

10      of alternatives.  Again, another pro and con table.

11      Next page, please.  And here it is.  Here is

12      comparing the Sync DRAM with Rambus and one third

13      player that never made the cut.  And this is the

14      IBM/Siemens documents.

15      Would you go back, please?  And here it is.

16      Compare the alternatives.  Sync DRAM, the pro, is

17      public domain.  The con is patent problems.  You have

18      got Motorola here, as well as Rambus.

19      Let’s go down to the Rambus cons.  Rambus

20      property.  This June 11th of 1992, over a month after

21      this May meeting in JEDEC where they tried to get the

22      Rambus representative to comment.  And he didn’t go

23      for it.

24      They weren’t through.  There was a

25      presentation in September of 1992.  And at that

                                                             69

1        presentation they had slides.  And they showed these

2        transparencies up on the wall.  There were a number of

3        people there, but not one person has owned up to being

4        in this audience that was assembled, but here’s what

5        they were shown.

6        Siemens is making this presentation.  Mr.

7        Meyer.  September of 1992.  And he asked, “What is

8        Rambus?”  He says, “A small firm without a fab,” and

9        you’re going to hear that if you don’t make chips,

10      basically, you’re not one of the club.  So that’s kind

11      of a derogatory comment about Rambus at this point.

12      But then they go back to the Rambus pros and

13      cons.  So here they are in September of 1992 with

14      almost two years of studying Rambus, and what do they

15      conclude?  A fascinating technical concept.  I assume

16      that’s a pro.  We’ve got a con.  “Deadly menace to the

17      established computer industry.”

18      Now, keep in mind this is a company that

19      maybe has 40 people.  This is a company with hundreds

20      of thousands of employees.  This is a company that’s

21      told the world, “We’re not going to make chips and

22      sell them.  We’re not going to have chips made and

23      sell them.”  The evidence will show that this deadly

24      menace is because of their business model, licensing

25      patents.  So that’s the way they were viewed.

                                                               70

1        In case there was any doubt about that and

2        about the awareness and continued awareness of Rambus

3        patents—excuse me.  I left off the alternatives.

4        After they went through the pros and cons,

5        they had some alternatives.  They said, okay, here’s

6        what Rambus is.  Here are the pros and cons.  They are

7        this deadly menace.  Their patents are a concern to

8        us.  The pro is it’s fascinating.  So what do they

9        suggest as alternatives?  These are alternatives for

10      Siemens.

11      So what do they do?  This is in September of

12      1992.  Here are the four alternatives that they offer.

13      The first alternative is push the computer technology.

14      Join Rambus.  Get on the cutting edge of the wave.

15      They rejected that alternative.

16      The next one was protect your own computer

17      industry.  Protect it from this deadly menace.  Buy

18      Rambus and dump it.  They didn’t take that

19      alternative.

20      Support the Japs.  Wait and watch.  They

21      took that in part.  Remember that memo.  Let’s wait

22      and see what it goes.  We don’t want this to be just a

23      niche product.  But then here was their predominant

24      alternative.  Make it Rambus.  Public domain.  Join

25      Sync DRAM, exclamation point, exclamation point,

                                                                71

1        exclamation point.  That was their position in

2        September 1992.

3        Now, they didn’t forget about Rambus

4        patents, and again, we’ve been through the disclosure

5        of the ‘703 in 1993, and the comments that were made

6        about their intellectual property.

7        Mr. Meyer was at that meeting where the ‘703

8        was disclosed, and he said, “Somebody made a comment

9        about Rambus pending patent applications.  I don’t

10      remember their name.  I didn’t check out anything that

11      they said.  But here’s what he said.  I’ll just use

12      his words.

13      This is September 1993 after the Rambus ‘703

14      patent is disclosed.  “Did someone during a meeting

15      after the ‘703 was disclosed indicate that there were

16      difficulties with an application that had been filed

17      by Rambus in the U.S. Patent Office?”

18      “Answer:  Yes.”

19      “Question:  What was said about that?

20      “The word stuck was used.”

21      “Question:  What’s the best recollection you

22      have of what was said about an application being

23      stuck?”

24      “Answer:  They were reported, the patent

25      applications, they were reported to be stuck because

                                                                     72

 

1        they can never be patented as being a collection of

2        prior art.”

3        “Question:  What was stuck?  Was it a patent

4        application by Rambus?”

5        “Yes.”

6        So we know the company was aware from 1990

7        up until the time it was put on notice of infringement

8        of this lawsuit of a couple of things.  First, that

9        Rambus had inventions, and second that they were going

10      to seek reasonable royalties to license those

11      inventions and the patents for those inventions.

12      1994 in a memo to its lawyer, Eric Kempfel,

13      Mr. Meyer put up a number of patent holders, quoted a

14      number of patents that will comply to any standard,

15      SDRAM standard notwithstanding, of other companies and

16      patents and what products they apply to.  And most of

17      these were SDRAM products.  And it gets over to

18      Rambus, and it has the US5243703.  The patent I told

19      you about that they said earlier didn’t have anything

20      to do with SDRAMs.  Product, SDRAM.

21      And they don’t stop there.  You see this

22      seven question marks preceded by US.  All U.S. patents

23      have seven numbers.  Just like the ‘703.  And so below

24      the ‘703, Rambus, they put seven question marks,

25      diverse.  Many.

                                                               73

1        So that’s what they thought about Rambus’

2        intellectual property in 1994.  That’s not all they

3        thought.  They weren’t hoodwinked.  They were aware of

4        our inventions and our intention to get patent

5        protection, the fullest patent protection we could get

6        under the law for those inventions, and license those

7        inventions for reasonable royalty.

8        They also, as we’ll show by the evidence,

9        starting in 1992 set out on a course to cheat Rambus

10      out of royalties, at least royalties from Siemens,

11      later Infineon, and decided to use selected portions

12      of the inventions without any royalty payments.

13      Since then they have knowingly and willfully

14      infringed Rambus patents.  They haven’t stopped to

15      this day.  In a technology driven field like this is,

16      you don’t know what’s going to happen tomorrow, but

17      you know for sure that today’s cutting edge technology

18      is obsolete tomorrow.  But sometimes when you look

19      into the future, you can be accurate, and we know what

20      Siemens thought when it looked into the future back in

21      1994.

22      In a memo from Mr. Meyer to Mr. Penzel, the

23      head of Siemens Semiconductor, March 31, 1994, this is

24      the way Mr. Meyer concluded.  And this is the way he

25      saw the future.  “One day all computers will have to

                                                               74

1        be built like this.”  He’s talking about Rambus.  “But

2        hopefully without the royalties going to Rambus.”

3        Thank you for your attention.

4        THE COURT:  All right.  Ladies and

5        gentlemen, we’ll take about a 20-minute recess to let

6        you move around a little bit, and then we’ll be back

7        in 20 minutes by that clock, please, Mr. Mack.

8        All right.  All please remain seated while

9        the jury is being excused.  You can take your pads

10      with you if you’d like to.  And you may as well write

11      your name on them because we will collect them at the

12      end of the day and return them to you tomorrow.

13      Anybody, ladies and gentlemen, who needs to

14      stand up at any time, remember I told you you’re free

15      to do it.

16      (The jury is excused from the courtroom.)

17      THE COURT:  Mr. Mack, are the first two rows

18      there for the lawyers?

19      MR. MACK:  Yes, sir.

20      THE COURT:  When you come back in, if you’re

21      going to come back in, make sure you’re back here and

22      seated in 20 minutes by that clock because we’re going

23      to immediately at that time bring the jury back in,

24      and I don’t want to hold up the jury’s consideration.

25      We’ll be in recess.