`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`TOSHIBA CORPORATION, TOSHIBA MEMORY CORPORATION, AND
`TOSHIBA AMERICA ELECTRONIC COMPONENTS, INC.,
`Petitioners,
`
`v.
`
`MACRONIX INTERNATIONAL CO., LTD.,
`Patent Owner.
`____________
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`____________
`
`Record of Oral Hearing
`Held: September 14, 2018
`___________
`
`
`
`
`Before KEN B. BARRETT, JENNIFER S. BISK, and JASON M. REPKO,
`Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`KEVIN C. HAMILTON, ESQUIRE
`GERALD T. SEKIMURA, ESQUIRE
`DLA Piper LLP (US)
`401 B Street, Suite 1700
`San Diego, California 92101
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`CHRISTIAN A. CHU, ESQUIRE
`CHRIS DRYER, ESQUIRE
`Fish & Richardson, P.C.
`1000 Maine Avenue SW
`Washington, D.C. 20024
`
`
`
`
`The above-entitled matter came on for hearing on Friday, September
`
`14, 2018, commencing at 1:00 p.m., at the U.S. Patent and Trademark
`Office, Madison Building East, 600 Dulany Street, Alexandria, Virginia,
`22314.
`
`
`
`
`
`
`
`
`
`
`
`
`
`2
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
`
`P R O C E E D I N G S
`- - - - -
` JUDGE BARRETT: Good afternoon, everyone. We are on
`the record for the final hearing in IPR 2017-01632. Toshiba
`versus Macronix International.
` I'm Judge Barrett, and with me at the bench is
`Judges Bisk and Repko.
` Let's get the parties' appearances. Who do we have
` from petitioner?
` MR. HAMILTON: Your Honor, Kevin Hamilton from DLA
`Piper on behalf of petitioner. And with me is Joe Sekimura,
`also from DLA Piper.
` JUDGE BARRETT: Welcome.
` And for patent owner?
` MR. CHU: Christian Chu with Fish & Richardson on
`behalf of Macronix International. With me is my colleague,
`Christopher Dreyer.
` JUDGE BARRETT: We issued a hearing order in this
`case that set forth the procedure, but I'll remind everybody.
` Each party will have 60 minutes total time to
` present arguments. For clarity of the transcript, please
` identify the slides you are referring to. That will make
` for a much cleaner transcript.
` Petitioner will go first, and you may reserve time
` for rebuttal, patent owner will then have an opportunity to
` present its response, and then petitioner may use any
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`3
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` rebuttal time remaining.
` I'll be watching the clock, and I will give counsel
` a warning when you're approaching the end.
` Any questions?
` MR. CHU: No, Your Honor.
` MR. HAMILTON: No, Your Honor.
` JUDGE BARRETT: Okay. Well, with that, you may
`begin.
` MR. HAMILTON: Thank you, Your Honor. I have three
`copies of -- hard copies of petitioner's demonstratives,
`but -- would you like those?
` JUDGE BARRETT: We're okay. We have them pulled up
`on our screen. The court reporter may want a copy before the
`end of the day, though.
` MR. HAMILTON: Excellent.
` Thank you, Your Honor. Good afternoon, Your Honors.
` Good afternoon.
` Kevin Hamilton and Gerald Sekimura from DLA Piper on
` behalf of the petitioner, Toshiba Corporation.
` Your Honor, I would like to reserve 25 minutes for
` rebuttal, if that's possible.
` Your Honor, I'm pulling up Slide 2 of Toshiba's
` demonstratives.
` What is shown on Slide 2 is, on the left we have an
` excerpt from the Background section of the '417 patent, and
` on the right-hand side, we have an excerpt from the
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`4
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` beginning of patent owner's response.
` And what this slide shows is the -- is Macronix's
` position, and frankly, the patent's position regarding what
` the purported invention was.
` And the purported invention was, quite simply, it
` was an output buffer circuit with variable drive strength.
` That's what was the title, and that's what the invention was
` supposed to be.
` These -- both of these excerpts identified problems
` with the prior art, and the problems with the prior art
` included the fact that the prior art, the output buffer
` circuits had a fixed drive strength. You couldn't change
` the drive strength, you could only turn it on or off.
` And what that meant was, if you had -- if a designer
` needed an output buffer circuit, he either had to design an
` output buffer circuit using a one -- one-size-fits-all
` approach, which was not always efficient, or he had to go
` and customize each and every output buffer circuit for its
` particular -- for its particular use. And that was very
` time-consuming, and the specifications had an enormously
` complicated design.
` And so that's what Macronix initially says its
` patent was, an output buffer circuit having variable drive
` strength.
` But, of course, the scope of the patent is
` determined by its claims, so let's look at Claim 1.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`5
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` I've pulled up Slide 3 of petitioner's
` demonstratives on the screen. And here, on the left, we're
` looking at Figure 6 of the '417 patent, and on the right we
` have Claim 1 of the '417 patent.
` There are three independent claims that are
` challenged in this petition; 1, 11, and 18.
` Claim 1 is the narrowest of the claims, and it
` recites the most limitation. And, in fact, every limitation
` recited in Claims 11 and 18 are recited in Claim 1. And so
` if you -- if a prior art reference anticipates Claim 1, then
` it anticipates Claims 11 and 18, as well, and I don't
` believe that there's any dispute between the parties on that
` issue.
` Moving to Toshiba's demonstrative number 4, Slide 4,
` we're looking at the same thing in the previous slide;
` however, this time, on the right-hand side, we have two
` portions of Claim 1 that are highlighted. There's a portion
` of the first "wherein" clause that is highlighted, and that
` represents one of the two disputed limitations between the
` parties, and a portion of the second limitation -- excuse
` me -- a portion of the second "wherein" clause is also
` highlighted, and that shows the second disputed limitation
` between the parties.
` So these are the only things that the parties
` dispute concerning whether the prior art references are
` anticipated or rendered obvious by the -- whether the prior
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`6
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` art references anticipate or render obvious the independent
` claims of the '417 patent.
` Moving to Slide 5. On the right-hand side, again,
` we have Claim 1 of the '417 patent and the first portion of
` the first "wherein" clause, the first disputed limitation is
` highlighted in yellow. On the left-hand side we have Yen
` Figure 3 and Yen Figure 4, and those are both annotated in
` red, as they were in the petition.
` So the limitation requires -- and I'm just going to
` read a portion of this yellow claim -- "The data output
` signal is combined to cross the plurality of output buffer
` circuits to provide a combined data output signal having a
` combined output drive strength."
` Now, Macronix concedes that the red annotations in
` Figures 3 and 4 disclose pluralities of output buffer
` circuits; however, Macronix says that these -- these figures
` are different. They're different because each of the
` figures contains what Macronix calls a default output
` buffer. The default output buffer in Figure 3 is shown at
` the top. It's -- it's a buffer and it's labeled 42. At the
` top. It's not highlighted. It's a triangle.
` And in Yen Figure 4 on the bottom, the output buffer
` circuit -- the default output buffer is the left -- left
` most four stack of transistors that is not highlighted.
` So Macronix coined this term "default output buffer
` circuit". Default output buffer is not something that's
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`7
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` recited in the '417 patent, it's not recited in the Yen
` reference, it's not in the claims, and there's actually
` nothing in the language of the claims that precludes the use
` of a default output buffer circuit.
` The claims recite, "Wherein the data output signal
` is combined across the plurality of output buffer circuits,"
` as shown in red, "to provide a combined output -- a combined
` data output signal having the combined output drive
` strength."
` JUDGE REPKO: So how is that combined? The top one
`isn't 42 combining with it as well, right?
` So it's like -- it seems like when it says, "A data
`output signal having a combined output drive strength," does
`that not include 42, as well?
` MR. HAMILTON: So if -- if you look at the top
`figure, you have, in red, output buffer circuit 44, 48, and
`50. Those are the red ones.
` The outputs of those -- the output signals come out
`of those on the right, and they're combined. They're
`combined in a signal that goes to the output, to the --
` Above and to the right of output buffer circuit 44,
`there's a signal there that contains the combination of those
`three output buffer circuits.
` JUDGE REPKO: So it's the signal that's coming up
`perpendicular to the signal coming from 42? Is that what
`you're saying?
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`8
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` MR. HAMILTON: That's correct. That's one way to
`read it. Okay?
` And in -- in Figure 4, we have the same thing. You
`have a portion of the -- of the horizontal line from 46 that
`couples the two output buffer circuits that are highlighted
`in red. That contains the combined output drive strength of
`those two plurality of output buffer circuits.
` That's not the only way to read this limitation on
`Yen.
` JUDGE BISK: Do we have to -- I'm sorry. Let me get
`the microphone. Am I okay?
` JUDGE BARRETT: Yes.
` JUDGE BISK: Do we have to do a claim construction
`for combined -- combined data output signal --
` MR. HAMILTON: No.
` JUDGE BISK: -- to say whether it includes only
`the -- for instance, like buffer enable inputs, or can it
`also include another output?
` MR. HAMILTON: So I don't think a claim construction
`is necessary. I think you read the plain meaning of the --
`of the language that's included in the claim, there's nothing
`that restricts it to "only". Now, let me explain.
` The highlighted limitation towards the second line,
` it provides, "A combined data output signal having the
` combined output drive strength."
` JUDGE BISK: Mm-hmm.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`9
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` MR. HAMILTON: Right? "Having" is --
` So output buffer 46 -- buffer output 46 that is
`shown in Figure 3 and Figure 4 of Yen, that output signal
`itself has the output drive strength of the plurality of
`output buffer circuits in each, but it also has something
`else. But "have" is an open term. Let me give you an
`example.
` So there are 12 months in a year. How many of those
` months are having or have 28 days? They all do. They'll
` have 28 days. Some -- some of the months have more, some of
` them have 30, some of them have 31, but they all have 28
` days.
` But if I were to ask a different question. You have
` 12 months in the year. How many of those months have only
` 28 days? Then that's a limitation that is not in the claim.
` The claim -- the plain language of the claim is
` broader. It just says "has it". Open-ended terminology is
` nothing new in patent law.
` JUDGE REPKO: So you're saying the data output
`signal is the signal from all of the (inaudible) -- including
`42, including 44, including 48, but your year, you're saying,
`is that analogous to the output signal?
` MR. HAMILTON: In the second way that I've shown you
`to look at this, then yes. Buffer output 46 in each of
`the -- of these figures has the combined output drive
`strength of the plurality of output buffer circuits.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`10
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` JUDGE BISK: And that's --
` MR. HAMILTON: But that -- and that's what the claim
`requires.
` JUDGE BISK: That's what was in your petition,
`right?
` MR. HAMILTON: Yes.
` JUDGE BISK: Okay. But using the intermediate
`output, that wasn't in your petition.
` MR. HAMILTON: So the petition says that the
`combined -- that the combined output drive strength of the
`plurality of output buffer circuits, for instance, in
`Figure 3 is -- those are combined to -- to give you that
`combined output drive strength.
` And so if you look at this here, I mean, the --
`fundamentally, I didn't necessarily say it in the same words,
`but the same argument was made.
` JUDGE BISK: Okay.
` MR. HAMILTON: So the claim language itself, it does
`not include that word "only", it's broad, that's the language
`that they chose.
` But if you're not persuaded by my argument and you
` like Macronix's default buffer argument, then it's worth
` taking a look at the Jain reference, and let's do that now.
` Moving to Toshiba's demonstrative Slide 6.
` The right-hand side, again, we have Claim 1 of the
` '417 patent with the first disputed limitation in the first
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`11
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` "wherein" clause is highlighted, and 4 the left-hand side we
` have Figure 6 of Jain.
` So clearly, Figure 6 of Jain discloses, "Plurality
` of output buffer circuits coupled in parallel to provide a
` combined output drive strength, and it does that without a
` default output buffer circuit." I don't think anybody's
` going to argue that that's not true.
` And looking at the limitation in question, "wherein
` the data output signal," which is 22 at the output 4 the
` right of Figure 6, "is combined across a plurality of output
` buffer circuits to provide a combined data output signal
` having the combined output drive strength of that plurality
` of output buffer circuits," and nothing else, no default
` buffer circuit.
` So it's right there. And that is the only thing
` that Toshiba is relying on -- is relying on Jain for is to
` show that an output buffer circuit with a variable drive
` strength and no default output buffer was known in the prior
` art prior to the '417 patent, which directly contradicts
` Macronix's position.
` And also, Toshiba relies upon Jain for the
` proposition that a person of ordinary skill in the art that
` Figure 3 of Yen and Figure 6 of Jain in front of them would
` understand that Yen could be implemented without using a
` default output buffer circuit.
` Now, the patent owner says, "A person of ordinary skill
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`12
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` in the art would not have been motivated to combine those
` two references." Let's take a closer look at that.
` I pulled Slide 7 of Toshiba's demonstratives up. On
` the left-hand side we're looking at Figure 3 of Yen
` annotated to show plurality of output buffer circuits in
` red, and Jain's Figure 6 is on the right, again, annotated
` to show plurality of output buffer circuits in red.
` Now, Macronix says that, "A person having
` ordinary" -- "A person of ordinary skill in the art looking
` at these two references would not combine them." I think
` that's a tough sell. They're almost identical.
` Other than the labeling of the inputs and outputs,
` the only difference between the two is the fact that Yen
` Figure 3 has an additional output buffer circuit 42 at the
` top that does not have a custom enable signal. It has a
` shared enable signal, but no custom enable signal.
` Now, the person in possession of Yen who looked at
` Jain might understand -- may look at Jain and say, hey,
` these guys did it a little bit different. They -- they
` added this -- this known thing, this custom buffer enable
` signal to the -- to the top most output buffer circuit, and
` that -- that gives me a little bit extra flexibility. Maybe
` I can turn that one off while I turn some of the other ones
` below on.
` That's not -- that's not a big deal. If you wanted
` that extra flexibility, then you could do that.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`13
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` JUDGE BISK: I'm sorry. I have a question which
`might be a little out there. But why isn't -- why wouldn't
`Jain be anticipatory? Why -- why are we combining Yen with
`Jain at all if it's the same --
` MR. HAMILTON: So it is. So the only thing that
`Jain does not expressly disclose it clearly is a -- is the
`final "wherein" clause, which is the buffer enable signal's
`complement controlling pairs of transistors having opposite
`connectivity types. That is not expressly disclosed in Jain.
`That's the only thing.
` JUDGE REPKO: But where is Figure 4 of Yen? How
`does this factor into your analysis?
` MR. HAMILTON: So the way that it factors in is
`Figure 3 and Figure 4 are different levels of abstraction
`showing the same invention. This is a block level diagram
`that does not show transitional level implementation, it just
`shows logic symbols like triangles that represent tri-state
`buffers.
` Going back to Slide No. 5. Figure 4 of Yen is the
` same thing, it's just that one of ordinary skill in the art
` would know that a tri-state buffer is implemented --
` commonly implemented when using a four stack of transistors
` having opposite connectivity types. That's just the way
` that it's done. It's the way that it's been done forever.
` That's the only way that I know of that you can -- that you
` would implement a CMOS tri-state buffer. So there are two
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`14
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` implementations of the same thing.
` So when you're -- when you're looking at Figure 3 of
` Yen, you include Figure 4 of Yen because they're describing
` the same thing, just from different perspectives.
` So you could do the same thing -- the same way that
` you eliminated the output buffer circuit 42 in Figure 3, you
` could also eliminate the output buffer circuit by this
` unhighlighted stack of transistors in Figure 4.
` Okay. So Macronix also says that, "A person of
` ordinary skill in the art would not -- would not combine
` these references because it changes the fundamental nature
` of the Yen reference in such a catastrophic way that it
` wouldn't work properly," but it's not true.
` If you needed -- if you needed Jain to operate in a
` mode that had a default output buffer circuit, all you would
` need to do is tie the ADD0 signal high, and that would
` remain -- that would ensure that the -- in Jain, Figure 6 --
` that the output buffer circuit 60 at the top would stay high
` as long as the output buffer circuit were enabled using the
` output enable signal.
` These -- these two diagrams are so similar in
` nature -- another way to think of it is you could think of
` it as ADD0 as being redundant. You don't need two different
` signals, output enable and ADD0, in order to -- to turn that
` first output buffer circuit off, you just need one or the
` other. And so you could have -- you could have implemented
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`15
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` Jain the way that you did in Yen and saved a little bit of
` area, not used that AND gate. They're interchangeable,
` hardly any difference at all, it's a simple combination, you
` don't lose anything.
` I want to take a look at the second contested
` claim -- or second contested limitation of Claim 1. I'm
` turning to petitioner's demonstrative Slide No. 8.
` Once again, we're looking at Claim 1 of the '417
` patent on the right, and highlighted is the second contested
` limitation in the second "wherein" clause, the final "wherein" clause of the
`claim, and on the left, again, we're looking
` at Figure 6.
` The limitation in question reads, "The buffer enable
` signals and the complements of the buffer enable signals
` control pairs of transistors having opposite connectivity
` types." Yen discloses this. Let's take a look.
` Turn to Slide 9. On the left I have that same
` second "wherein" clause that contains the second contested --
` the same -- the second disputed limitation of Claim 1, and
` on the right I have Figure 4 of Yen, transistor level
` implementation of the invention, and it has some color
` annotations.
` Let's just look at the claim. "Buffer enable
` signals and the complements of the buffer enable signals."
` So looking over here in -- at Figure 4, on the bottom right
` of the figure, you have out enable, shared enable signal,
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`16
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` OS1 and OS2. Those are customized enable signals. Those
` are buffer enable signals.
` At the top, output enable bar, OS1 bar, and OS2 bar.
` Those are complements of the buffer enable signals. So it's
` right there. Buffer enable -- the claim requires buffer
` enable signals and the complements of the buffer enable
` signals.
` The rest of the limitation reads, "Control pairs of
` transistors having opposite connectivity types." Well,
` let's look at what these signals do.
` OS1 and output enable control transistor MN4 in
` Figure 4 of Yen, and then power being an NMOS device. And
` the complements of those signals control MP4, which is the
` transistor having opposite connectivity type.
` Likewise, signal OS2 and output enable control
` transistor MN6 in the bottom right of Figure 4 of Yen, and
` likewise, the output enable bar and OS2 bar, any buffer
` enable signals control transistor MP6 in the top right of
` Figure 4.
` JUDGE REPKO: So what's your pairs? Plural pairs?
` MR. HAMILTON: So plural pairs is one pair is MN4
`and MP4; that's a pair of transistors having opposite
`connectivity types. And then 6 and MP6 is a second pair of
`transistors having opposite connectivity types.
` That is, those signals, those buffer enable signals
`control pairs -- both pairs of transistors having opposite
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`17
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
`connectivity types. That's it right there. Simple as that.
`Plain reading of the claim.
` JUDGE REPKO: So we have plural signals controlling
`plural pairs.
` MR. HAMILTON: Yes.
` JUDGE REPKO: What about each signal controlling?
`Why should we not read the claim like that?
` MR. HAMILTON: Well, because the claim doesn't say
`that. The word "each" doesn't appear in the claim.
` JUDGE REPKO: Is that consistent with the
`specification?
` MR. HAMILTON: Yes. Let's get into that.
` Let's talk about precisely that here. We'll start
` on the next slide.
` I'm looking at Toshiba's demonstrative Exhibit
` No. 10. And on the top, we're looking at an excerpt from
` the patent owner's response at 34, and on the bottom, again,
` we're looking at that second "wherein" clause that contains
` the second disputed limitation of Claim 1.
` Now, looking at the top here in the patent owner's
` response, it says, "Each buffer enable signal," and then
` Macronix's position, "Each buffer enable signal pair, that
` is a buffer enable signal and its complement signal,
` controls pairs, that is, more than one pair of transistors,
` where the two transistors in each transistor pair have
` opposite connectivity types." So that's Macronix's
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`18
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` position. They say, "Each buffer enable signal in the pair
` control multiple pairs of transistors having an opposite
` connectivity types."
` And -- but the claim doesn't say "each", right? And
` this is -- if you're not careful when you read this, you see
` that the "each" is within a quote. You see it's quoting
` here, it says, "Each buffer enable signals pair."
` If you're not careful, you think that the claim
` actually says that, but the claim doesn't, that's just
` Dickens saying that.
` The claim says, "The buffer enable signals and the
` complements of the buffer enable signals" -- it doesn't say
` "each" -- "control pairs of transistors of having opposite
` connectivity types."
` This "each" interpretation, let's look at where that
` came from.
` I'm looking at -- we're all looking at Toshiba's
` demonstrative Exhibit No. 11. On the left, again, we have
` the same second "wherein" clause that contains the second
` disputed limitation highlighted in yellow, and on the upper
` right we have that same excerpt from the patent owner's
` response that contains the "each" language in quotes.
` On the bottom right we have an excerpt from the
` patent owner's response that cites to Figure 2 of the '417
` patent, and the colors were annotated by Macronix in their
` response.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`19
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` Now, what Macronix is doing to read -- when you take
` a look at how Macronix is reading -- their interpretation of
` how this claim should be read into this -- in this figure.
` So let's start with Z and ZB. Z and ZB are a shared
` enable signal and -- Z is a shared enable signal, ZB is its
` complement. So you have enable signal and its complement.
` The Z controls -- you see here in the red
` (inaudible) -- the Z controls a pair of transistors, and you
` see, because the arrow is pointing different ways, you have
` a PMOS and NMOS device, they're pair of transistors having
` opposite connectivity types.
` ZB, the complement, same thing. It's shown here in
` purple on the bottom left of the annotated figure.
` Transistors 214 and 216 are pairs of transistors
` having opposite connectivity types.
` So Z and ZB, a single pair of transistors having
` opposite -- a single pair -- I'm sorry.
` Z and ZB, a buffer enable signal and its complement,
` in the singular, control pairs of transistors having
` opposite connectivity types.
` But now let's look at what they do with OPON and
` OPONB. This is the customized enable signal.
` So OPON controls directly this -- in the turquoise
` color, the blue -- transistor number 218. We note here that
` it has an arrow pointing out.
` And also, if you follow the output of that signal,
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`20
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` it goes to -- if you follow the output of transistor 218, if
` you follow the blue -- blue wiring, that also goes to
` transistor 204, the bottom right of the figure.
` But look. The arrows on these two transistors point
` the same way, so those are not transistors having opposite
` connectivity types, those transistors have the same
` connectivity type.
` But then if you look at OPONB, it's the -- it's the
` logical dual. OPONB controls transistor 220 that has --
` it's a P-type device. And if you trace down to transistor
` 202, that's also a P-type device.
` So this is how Macronix -- this is just tortured
` interpretation of the claim language to say that this even
` reads on it because of the way that these transistors are
` arranged, but that -- that's their justification. And the
` fact is, the only justification they have for that is that
` this -- this embodiment that is shown in this figure, they
` say it reads -- it reads on the claim language that way.
` But it's an embodiment. They're trying to import
` that tortured interpretation of the embodiment in -- as a
` limitation into the claim, and we know that that's just
` absolutely not proper. You can't read --
` JUDGE REPKO: Are there other embodiments
`that better support your position?
` MR. HAMILTON: I'm sorry?
` JUDGE REPKO: Are there other embodiments in the --
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`21
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
`in the patent that support your position?
` MR. HAMILTON: That support my position?
` JUDGE REPKO: Yeah.
` MR. HAMILTON: Absolutely. Let me give you an
`alternative theory.
` So I'm going to quickly look back at Slide No. 8.
` On Slide No. 8, the left-hand side, you have
` Figure 6 in the '417 patent. Figure 6 contains a plurality
` of output buffer circuits. I think it's 630, 632, 634, and
` 636. And inside of each of these output buffer circuits
` is -- going back to Toshiba Slide 11 -- is the output buffer
` circuit that's shown in Figure 2 of the '417 patent. So
` this is one of them.
` The whole idea of the '417 patent is you have
` multiple -- you have plurality of output buffer circuits,
` multiple output buffer circuits, and you control those
` output buffer circuits with these sets of enable signals and
` their complements.
` And what you can do is you can turn on different
` ones in order to achieve the drive strength that you want.
` What we're talking about here, the enable signal is
` controlling the drivers. Look. This is the driver. These
` two -- I'm looking at Slide 11 on the right-hand side of
` Figure 2.
` The driver are -- is this pair of transistors, 202
` and 204. That's the driver.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`22
`
`
`
`Case IPR2017-01632
`Patent 8,035,417 B1
`
` So when the patent is talking about controlling
` pairs of transistors having opposite connectivity types,
` that's not talking about it pointing to some random
` transistors, it's pointing -- it's talking about pairs of
` transistors having opposite connectivity. It's the driver.
` That's what the whole point of the patent is is to control
` different output drivers.
` This stuff is just logic. And what it does --
` that's not the point, the point is controlling the drivers.
` That -- that is what the patent's talking about in
` this claim when it's talking about controlling pairs of
` transistors having opposite connectivity types.
` JUDGE BARRETT: You have about five minutes left of
`your original 35.
` MR. HAMILTON: Thank you.
` Your Honor, I have put Slide 12 of Toshiba's
` demonstrative exhibits up on the screen.
` And what Slide 12 shows is, again, our familiar
` Claim 1 of the '417 patent. And again, it shows the
` highlighted disputed limitations in the first and second
` "wherein" clauses. But what this also does is this shows you
` the modifications that Macronix is trying to make to the
` claim in order to