`571-272-7822
`
`
`
`
`Paper 33
`Entered: December 10, 2013
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`INTELLECTUAL VENTURES MANAGEMENT, LLC
`Petitioner,
`
`v.
`
`XILINX, INC.
`Patent Owner.
`____________
`
`Case No. IPR2012-00023
`Patent 7, 994,609
`Case No. IPR2012-00020
`Patent 8,058,897
`____________
`
`Held: November 7, 2013
`____________
`
`Before SALLY C. MEDLEY, JUSTIN T. ARBES and KARL D.
`EASTHOM, Administrative Patent Judges.
`
`
`RECORD OF ORAL HEARING
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`LORI A. GORDON, ESQUIRE
`
`
`ROBERT GREENE STERNE, ESQUIRE
`
`
`OMAR AMIN, ESQUIRE
`
`
`Sterne Kessler Goldstein Fox
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`
`
`1100 New York Avenue, N.W.
`Washington, D.C. 20005
`
`
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`DAVID M. O’DELL, ESQUIRE
`
`
`DAVID L. McCOMBS, ESQUIRE
`
`
`Haynes and Boone, LLP
`
`
`2323 Victory Avenue, Suite 700
`
`
`Dallas, Texas 75219
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`
`16
`
`17
`
`
`
`The above-entitled matter came on for hearing on Thursday,
`November 7, 2013, commencing at 1:30 p.m., at the U.S. Patent and
`Trademark Office, 600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`
`
`
`
`P R O C E E D I N G S
`
`- - - - -
`
`JUDGE MEDLEY: Good afternoon, everyone. This
`
`18
`
`is the hearing for IPR2012 -0020 and IPR2012-00023. So, we'll
`
`19
`
`proceed first with the hearing in IPR2012 -00020, and then we're
`
`20
`
`going to take a short break and let everybody reconvene and get
`
`21
`
`situated, and then we'll begin with the case for 00023.
`
`22
`
`So, at this time we would like the parties to please
`
`23
`
`introduce themselves, starting with the petitioner.
`
`24
`
`MS. GORDON: I'm Lori Gordon, I'll be arguing
`
`25
`
`today on behalf of the petitioner, Intellectual Ventures
`
`26
`
`Management. With me today is Robert Sterne, also from the law
`
` 2
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`firm of Sterne Kessler, and Omar Amin from the law firm of
`
`Sterne Kessler.
`
`JUDGE MEDLEY: Ms. Gordon, will you also be
`
`arguing for 00023?
`
`MS. GORDON: Yes.
`
`JUDGE MEDLEY: For patent owner?
`
`MR. O'DELL: Hello, my name is David O'Dell, I'm
`
`with the law firm Haynes and Boone, I'll be representing the
`
`patent owner XILINX. With me today is my co -counsel, David
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`McCombs, also with Haynes and Boone. Mr. McCombs will be
`
`11
`
`arguing for the first one, matter 00020, and then I will be
`
`12
`
`arguing for the second one, matter 00023.
`
`13
`
`14
`
`JUDGE MEDLEY: Great. Okay, thank you.
`
`So, as you recall from the order, each party gets 20
`
`15
`
`minutes total for the first case, for 00020, and each party may
`
`16
`
`reserve rebuttal time if they wish to.
`
`17
`
`So, we'll begin with the petitioner, and just let us
`
`18
`
`know would you like to reserve rebuttal ti me?
`
`19
`
`MS. GORDON: Yes, we would like to reserve ten
`
`20
`
`minutes.
`
`21
`
`JUDGE MEDLEY: Ten minutes, okay, great. You
`
`22
`
`may begin.
`
`23
`
`MS. GORDON: So, we've prepared demonstratives
`
`24
`
`that we may use to aid the discussion. We uploaded them, per
`
` 3
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`the order yesterday, to PRPS. We have extra copies if you need
`
`them.
`
`JUDGE MEDLEY: I think we're okay.
`
`MS. GORDON: Okay, great, thank you.
`
`So, there's two disputes that remain in this proceeding
`
`between the parties. The first is whether dependent claims 2
`
`through 7 are obvious over the Wennekamp reference. And the
`
`second issue in dispute is whether independent claim 8 is
`
`obvious over the combination of Wennekamp and Miller. Both
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`these issues are dispositive for all claims under review in this
`
`11
`
`proceeding.
`
`12
`
`So, turning to the first issue, whether claims 2
`
`13
`
`through 7 are obvious over Wennekamp. Claims 2 through 7
`
`14
`
`depend from claim 1. We'll put claim 1 up here for reference.
`
`15
`
`Patent owner does not dispute that claim 1 is unpatentable, based
`
`16
`
`on the grounds instituted for t his trial; however, the only
`
`17
`
`distinction that patent owner is raising relative to dependent
`
`18
`
`claims 2 through 7 is that Wennekamp does not teach or suggest
`
`19
`
`a multi-die IC, and this is a limitation that's only recited in
`
`20
`
`independent claim 1, a claim that pat ent owner does not dispute
`
`21
`
`is unpatentable over either Wennekamp or the combination of
`
`22
`
`Wennekamp and Miller.
`
`23
`
`JUDGE MEDLEY: So, just let me interrupt you real
`
`24
`
`quickly. So, how should we reconcile that? So, we've read in
`
`25
`
`your papers that patent owner ca ncels claim 1 and then perhaps
`
` 4
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`they concede that claim 1 is unpatentable, but then they argue
`
`that claims 2 through 7, which depend on 1, are patentable over
`
`the prior art, based on a feature that is in the cancelled claim 1.
`
`So, how is the Board to recon cile that?
`
`MS. GORDON: Right, and we also struggle with how
`
`to reconcile that. We note that claim 1, there were two grounds
`
`of rejection to claim 1, whether it was obvious over Wennekamp,
`
`standing alone, or obvious over Wennekamp in view of Miller.
`
`Our only way we can make this have any sense is that patent
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`owner may be conceding that the combination of Wennekamp
`
`11
`
`and Miller renders claim 1 unpatentable; however, that they don't
`
`12
`
`believe that Wennekamp standing on its own renders claim 1
`
`13
`
`unpatentable.
`
`14
`
`JUDGE MEDLEY: Okay, and we'll, of course, have
`
`15
`
`an opportunity to ask patent owner their position on that. Thank
`
`16
`
`you.
`
`17
`
`MS. GORDON: So, we're proceeding to address the
`
`18
`
`substance of patent owner's position. So, as we said, the patent
`
`19
`
`owner isn't individu ally arguing any of the features of the
`
`20
`
`dependent claims 2 through 7.
`
`21
`
`The evidence of record in this case, both from
`
`22
`
`Intellectual Ventures Management's expert, Morgan Johnson, and
`
`23
`
`patent owner's expert establishes, in fact, that a person of
`
`24
`
`ordinary skill in the art would have and could have modified the
`
`25
`
`Wennekamp reference as set forth by the petitioner. So, there's
`
` 5
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`no dispute between the parties that Wennekamp teaches multiple
`
`dies. The only dispute is really a person of ordinary skill in the
`
`art would have taken these dies and placed them on the same
`
`substrate in the same IC.
`
`So, as petitioner set forth, multiple die ICs were well
`
`known by the time the '897 patent was filed. IVM's expert not
`
`only said they were well known, he said they were in full bloom
`
`by 2010 when the '897 patent was filed. Patent owner's own
`
`expert, when asked at deposition, agreed that multiple ICs were
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`known for decades, as early as the 1980s or even earlier.
`
`11
`
`Multi-die ICs were well known by the time '897 was filed. Both
`
`12
`
`experts agree that it was within the routine abilities of a person
`
`13
`
`with ordinary skill in the art to make the multi -dies of
`
`14
`
`Wennekamp and put them on the same substrate or the same IC.
`
`15
`
`And Wennekamp provides ample motivation for a
`
`16
`
`person skilled in the art to do so. As petitioner pointed out,
`
`17
`
`Wennekamp describes a situation where long wires or circuits
`
`18
`
`degrades signal quality, and also limit clock frequency. And this
`
`19
`
`is an issue that Wennekamp recognized.
`
`20
`
`At deposition, patent owner's own expert, when
`
`21
`
`explaining what would motivate a designer to use a multi -die IC,
`
`22
`
`focused on these same issues that were presented in Wennekamp,
`
`23
`
`the fact that signal quality was important, and you wanted to
`
`24
`
`maximize clock speed. So, it appears that there's no dispute
`
`25
`
`between the experts from both sides that Wennekamp not only
`
` 6
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`could have been modified to place its multiple dies on a single
`
`substrate, that is in a single IC, but would have been modified by
`
`a person of ordinary skill in the art.
`
`So, moving on to the second i ssue, whether claim 8 is
`
`unpatentable over the combining --
`
`JUDGE MEDLEY: Can we go back to the first issue
`
`first?
`
`MS. GORDON: Sure.
`
`JUDGE MEDLEY: It seems to me that reading their
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`papers, patent owner's papers, that really what they're saying is
`
`11
`
`it's improper to go -- that perhaps we or the petitioner, or
`
`12
`
`whomever, are going beyond the statutory requirements of
`
`13
`
`311(b), 35 USC 311(b), it's not clear to us that really it's patent
`
`14
`
`owner's position that the claims are nonobvious over
`
`15
`
`Wennekamp, they haven't positively come out and said that,
`
`16
`
`they're just saying that the scope of the trial went beyond what
`
`17
`
`was allowed per statute. And I just want to hear your response
`
`18
`
`on that.
`
`19
`
`MS. GORDON: Right. So, in our reply, we addressed
`
`20
`
`patent owner's issue, the single reference 103 accommodation, or
`
`21
`
`single reference 103 rejections are allowed, and we provided
`
`22
`
`some case law from the Federal Circuit, particularly, you know,
`
`23
`
`in the Leapfrog versus Fisher -Price case, which didn't deal with
`
`24
`
`a single reference 103, but dealt with a situation where a
`
`25
`
`reference didn't teach a reader element. And the District Court
`
` 7
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`found that even though neither of the references taught a reader,
`
`that was such a well -known piece of circuitry that a person of
`
`skill in the art would have known that it was part or would have
`
`modified the references to include it. And the Federal Circuit
`
`affirmed, saying that that was, indeed, sufficient to establish a
`
`case of obviousness.
`
`We also supplied additional case law In Re:
`
`O'Farrell, where it supported a single reference 103 where the
`
`Patent Office supplied a missing teaching based on what was
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`well known and within the routine scale of a person of ordinary
`
`11
`
`skill in the art at the time the patent was filed. So, we disagree
`
`12
`
`with the patent owner 's legal position here.
`
`13
`
`JUDGE MEDLEY: And one other question I have
`
`14
`
`before we move on to claim 8 and those claims that depend from
`
`15
`
`claim 8. I believe you said earlier that the patent owner
`
`16
`
`concedes that Wennekamp does show dies, just not multiple dies.
`
`17
`
`18
`
`MS. GORDON: Correct.
`
`JUDGE MEDLEY: On an IC. I didn't understand that
`
`19
`
`that was their position. I thought that they were saying that
`
`20
`
`Wennekamp does not teach dies at all.
`
`21
`
`22
`
`23
`
`MS. GORDON: Yeah, and that --
`
`JUDGE MEDLEY: Is that correct?
`
`MS. GORDON: Yeah, that was not our understanding
`
`24
`
`from the patent owner's response. They never challenged that, or
`
`25
`
`didn't dispute the fact that Wennekamp showed dies. In fact, if
`
` 8
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`we look at, for example, figure 3 of Wennekamp, and it's hard to
`
`read, but page 11 o f our demonstratives, you can see that in each
`
`of these devices here, we have a CSB, which is a chip select
`
`block, or a chip select signal, indicating that each one of these is
`
`an individual chip or a die.
`
`So, our understanding from the briefing was that the
`
`parties didn't have a dispute that Wennekamp, indeed, disclosed
`
`individual dies.
`
`JUDGE MEDLEY: Just not showing how they're
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`arranged on a particular --
`
`11
`
`12
`
`MS. GORDON: Correct.
`
`JUDGE MEDLEY: I see. Okay, thank you. Please
`
`13
`
`proceed, then, with cla im 8 and those claims that depend on
`
`14
`
`claim 8.
`
`15
`
`MS. GORDON: So, in patent owner's response, patent
`
`16
`
`owner argued that Wennekamp did not disclose three specific
`
`17
`
`limitations found in claim 8 of the '897 patent. Specifically,
`
`18
`
`they argued that Wennekamp did n ot determine whether the
`
`19
`
`configuration data comprises segment of configuration data for
`
`20
`
`an additional IC, and they also argued that Wennekamp did not
`
`21
`
`disclose a first die configured to distribute the second segment of
`
`22
`
`the configuration data to the second d ie through the
`
`23
`
`configuration bus, and also that Wennekamp did not disclose that
`
`24
`
`the first die sent the segment of configuration data for the
`
`25
`
`additional IC through the first die configuration data output.
`
` 9
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`Now, these limitations actually correspond to
`
`identical limitations of claim 1, and patent owner did not dispute
`
`on the record that claim 1, that Wennekamp disclosed these
`
`elements when they're present in claim 1.
`
`So, again, I'm not sure how to reconcile that, but we'll
`
`address the merits as to claim 8. So, for the purposes of
`
`argument, we can deal with the distributing and the sending
`
`limitations together, and patent owner's argument for these two
`
`elements is really based on a misinterpretation of the explicit
`
`10
`
`language of claim 8.
`
`11
`
`So, patent owner's position in its response is that we
`
`12
`
`can see that the second segment of configuration data is set on
`
`13
`
`one bus, whereas the segment of configuration data there in the
`
`14
`
`last limitation is set on another bus. But if we look at the
`
`15
`
`explicit language of the claim s, that is not what is claimed.
`
`16
`
`What we see in the claims, instead, is the segment --
`
`17
`
`second segment is sent on a configuration bus, and the segment
`
`18
`
`of configuration data for the additional IC is sent through the
`
`19
`
`configuration data output.
`
`20
`
`The claims do not require two buses, nor do they
`
`21
`
`require two outputs. And when you look at the specification of
`
`22
`
`the '897 patent, they make it clear that a bus and an output are
`
`23
`
`two different things. In fact, when it talks about an output in the
`
`24
`
`context of figure 5, it talks about an output port. So, the output
`
`25
`
`is part of the die, whereas the bus is the transmission path
`
` 10
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`between two die. And the claim does not require two buses, nor
`
`does it require two outputs.
`
`And if we look at claim 8, relative to Wennekamp, we
`
`can see how Wennekamp discloses claim 8, which is broadly
`
`claimed. So, claim 8 requires, again, distributing the second
`
`segment of configuration data, through the configuration bus.
`
`And we see here in Wennekamp, a bus connecting in the first die
`
`and the second die. And the claim says that the configuration
`
`data output of the first die is coupled to the output of the IC.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`And you have to read this in the context of the '897
`
`11
`
`patent, which defines coupling very broadly. Coupling, they
`
`12
`
`explain, is not just a direct connection, it can be indirect through
`
`13
`
`intervening elements.
`
`14
`
`So, we see here, Wennekamp has an output that's part
`
`15
`
`of the die, your PN0, and that's actually coupled to the output of
`
`16
`
`the IC, which would be where that demarcated dashed line is.
`
`17
`
`And the last limitation sends the segment of configuration data
`
`18
`
`for the additional IC through the configuration data output, well
`
`19
`
`that gets sent through the output, which is here.
`
`20
`
`So, we can see that claim 8 is very broadly claimed,
`
`21
`
`and Wennekamp discloses these elements of claim 8, which
`
`22
`
`patent owner argued were not present.
`
`23
`
`JUDGE MEDLEY: So, are you saying that the same
`
`24
`
`elements meet the bus and the output?
`
` 11
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`MS. GORDON: No, they're actually two different
`
`elements, and you can think of it this way: The output is part of
`
`the die, and the bus is actually part of the substrate, that's the
`
`transmission path.
`
`JUDGE MEDLEY: That connects it.
`
`MS. GORDON: So, if you lifted the die out, the
`
`output would come with it, but the bus would remain on the
`
`substrate.
`
`JUDGE MEDLEY: Okay, thank you.
`
`10
`
`JUDGE ARBES: Counsel, does the specification of
`
`11
`
`the '897 patent refer to a configuration data output in that
`
`12
`
`manner?
`
`13
`
`MS. GORDON: It refers to a configuration -- it refers
`
`14
`
`to a data output port, and in the disc ussion of figure 5. And a
`
`15
`
`separate configuration bus.
`
`16
`
`JUDGE MEDLEY: Okay, I think a few more minutes
`
`17
`
`to wrap up, we asked a lot of questions.
`
`18
`
`MS. GORDON: Just one more point related to patent
`
`19
`
`owner's final argument that Wennekamp doesn't teach
`
`20
`
`determining whether the configuration data comprises a segment
`
`21
`
`intended for the additional IC. So, here we see the disclosure
`
`22
`
`from Wennekamp, and it's talking about this circuit, and this is
`
`23
`
`its parallel daisy chain configuration, where there is the first die,
`
`24
`
`and this is the second die. And Wennekamp clearly states that
`
`25
`
`either the first die or the second die could be implemented using
`
` 12
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`the device of 400. So, let's assume that the device of 400 is
`
`implementing the first die. When we look at the device 400,
`
`Wennekamp very explicitly describes its operation, and it
`
`describes that certain instructions indicate that the device should
`
`enable its output register, POUT, and this enable instruction
`
`means that the data being received by the first die is not targeted
`
`for that device, instead, it's intended for a downstream device.
`
`So, what does this mean? It means that the die in
`
`Wennekamp, the first die, determines whether the data is
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`intended for itself, or it's intended for an additional device
`
`11
`
`downstream, in this case the additional IC.
`
`12
`
`JUDGE MEDLEY: So, what in the first die makes
`
`13
`
`that determination?
`
`14
`
`MS. GORDON: Right. There's controlled logic in the
`
`15
`
`first die that when it receives this enable signal will make the
`
`16
`
`determination, based on the status of the s ignal, whether the data
`
`17
`
`is configured to be -- to stay with it or to go outside the die.
`
`18
`
`JUDGE MEDLEY: And is it correct, because I looked
`
`19
`
`through their papers, and it didn't seem to me that they addressed
`
`20
`
`this embodiment, they seemed to be focused main ly on the
`
`21
`
`embodiment of figure 3. Did they discuss -- I mean, I just
`
`22
`
`didn't -- maybe this is a question for patent owner when they get
`
`23
`
`up, but --
`
` 13
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`
`MS. GORDON: We don't believe they directly
`
`addressed this, but this is the embodiment that was in our
`
`petition, so they had the opportunity to address.
`
`JUDGE MEDLEY: Okay, thank you.
`
`MS. GORDON: Thank you.
`
`JUDGE MEDLEY: Patent owner, please? And,
`
`Counsel, would you like to reserve time?
`
`MR. McCOMBS: I would like to reserve five
`
`minutes, please.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`JUDGE MEDLEY: Five minutes, okay, great. You
`
`11
`
`may begin.
`
`12
`
`MR. McCOMBS: Thank you, good afternoon, Your
`
`13
`
`Honors, I'm David McCombs here for XILINX, and our first
`
`14
`
`issue that we would like to address concerns claims 2 through 7,
`
`15
`
`and whether they are obvious based on the Wennekamp reference
`
`16
`
`alone. You had asked some questions about this procedurally, I
`
`17
`
`would like to first address some procedural points concerning
`
`18
`
`claim 1.
`
`19
`
`And if we look at the institution decision, there were
`
`20
`
`two grounds of challenge that were raised with respect to claim
`
`21
`
`1. First was whether claim 1 was obvious over Wennekamp, and
`
`22
`
`second was whether claim 1 was obvious over Wennekamp in
`
`23
`
`view of Miller. In response to this, on May 7th, what we did was
`
`24
`
`the patent owner submitted its motion to amend, and in our
`
`25
`
`motion to amend, we requested that claim 1 be cancelled, and I
`
` 14
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`understand that you all have not ruled on our motion, but it's our
`
`position that the cancellation of claim 1 renders moot the need to
`
`decide whether claim 1 is patentable in light of the references
`
`that were cited against it.
`
`But what that does is then brings us to claims 2
`
`through 7, and with respect to claims 2 through 7, the only issue
`
`that's pending there, these are dependent claims from claim 1, is
`
`whether those claims are obvious based on Wennekamp alone.
`
`And again, with respect to claim 2, we think that that
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`claim is representative of all of 2 through 7, and the issue there
`
`11
`
`really does boil down to obviousness of the multi -die IC
`
`12
`
`limitation in its base claim, clai m 1. And, so, as we've identified
`
`13
`
`here, the issue is with respect to claim 2, does Wennekamp teach
`
`14
`
`a first IC comprising the master die and the slave die, that's part
`
`15
`
`of claim 2 through its base claim.
`
`16
`
`JUDGE MEDLEY: And I'll ask you the question that
`
`17
`
`I asked petitioner, counsel for petitioner, is the patent owner
`
`18
`
`then conceding that Wennekamp does teach dies, just not
`
`19
`
`multi-die IC, multi-die IC is not taught, which is in the claim,
`
`20
`
`but which everyone understands the claims to mean?
`
`21
`
`MR. McCOMBS: No, that is not our position. Our
`
`22
`
`position is that Wennekamp teaches ICs, it never discusses
`
`23
`
`specifically dies, and our view is that claim 2 is not obvious over
`
`24
`
`Wennekamp. Wennekamp doesn't discuss multi -die ICs at all, it
`
`25
`
`discusses specific IC configurations, in and of themselves. And
`
` 15
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`we believe that claim 2 is patentable over Wennekamp, and we
`
`believe this to be the correct conclusion even when you take into
`
`account the petitioner's evidence of the level of skill in the art
`
`that's been present ed.
`
`We're also well aware of the Federal Circuit's October
`
`30th decision in Randall Manufacturing versus Rea, where the
`
`Federal Circuit was criticizing a failure to take into account
`
`evidence of the state of the art in an obviousness analysis. In
`
`Randall Manufacturing, that case, in particular, all of the art was
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`present, or all of the elements of the claim, that is, were present
`
`11
`
`in the prior art references that were used there, and it was the
`
`12
`
`state of the art was a question having to do with whether tho se
`
`13
`
`references were combinable.
`
`14
`
`So, if we look at this situation, in our case, we want
`
`15
`
`to take with the guidance of Randall Manufacturing, we want to
`
`16
`
`take an unblinkered focus on Wennekamp, and in doing so,
`
`17
`
`consider the evidence of what is the state of th e art and what is
`
`18
`
`ordinary level of skill in the art.
`
`19
`
`And in doing so, we believe that you still don't get
`
`20
`
`there to defeat claim 2. And this is why: First, if you look at
`
`21
`
`the Board's institution decision, the Board was clear there that
`
`22
`
`Wennekamp does n ot describe programmable devices in the
`
`23
`
`context of dies on an IC. There was no discussion of multiple
`
`24
`
`die ICs in Wennekamp. We agree with that. That was stated in
`
` 16
`
`25
`
`the decision.
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`But, it is true, yes, there is evidence in the state of the
`
`art that multi-die ICs generally are known. The Board
`
`recognized this in the institution decision, there was a discussion
`
`of a publication evolution of CPU packaging technology and
`
`future challenges by an author named Oh, and we also have, in
`
`the testimony in this case , we also have our own expert,
`
`Dr. Blanchard, and he recognizes that the general idea of
`
`multi-die ICs is out there, it's in the art.
`
`And, of course, we've heard from the petitioner's
`
`10
`
`expert, Morgan Johnson, who asserts numerous times that
`
`11
`
`multi-die ICs are definitely something within the ability of the
`
`12
`
`skilled artisan, but our view is that that doesn't -- that doesn't get
`
`13
`
`you there. What's missing here is the particularized teaching in
`
`14
`
`the claim of an IC configuration, where you have a configureable
`
`15
`
`master die and a configureable slave die. And as we see here --
`
`16
`
`17
`
`18
`
`JUDGE MEDLEY: Can I interrupt you?
`
`MR. McCOMBS: Yes?
`
`JUDGE MEDLEY: It seemed to me in your response,
`
`19
`
`though, that your argument really was that with respect to claims
`
`20
`
`2 through 7, that r eliance on expert testimony showing what was
`
`21
`
`known by a person of ordinary skill in the art isn't proper and is
`
`22
`
`going beyond 35 USC 311(b).
`
`23
`
`So, it's sort of, it didn't seem to us that the paper,
`
`24
`
`your paper, really went out and said that the claims are
`
` 17
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`nonobvious over Wennekamp, showing why they were
`
`nonobvious.
`
`MR. McCOMBS: Well, Your Honor, let me address
`
`that. We don't believe that Wennekamp discusses multi -die ICs
`
`at all, it certainly doesn't discuss the specific implementation in
`
`the claims. Figure 4 of the '897 patent shows the specifically
`
`claimed implementation of an IC having a configureable master
`
`and a configureable slave. What we were also of the view is that
`
`the -- any of this discussion that's in the record of what is -- of
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`what someone of ordinary skill in the art would know, that
`
`11
`
`doesn't get you there. And I am concerned, Your Honor, that this
`
`12
`
`is a situation where it may be very easy under section 311(b) to
`
`13
`
`get into a situation where we're talking about items that are
`
`14
`
`known and used by others, which would be a 102(a) issue, which
`
`15
`
`in this instance would not be the proper scope of a proceeding
`
`16
`
`here, and what we don't want is that this analysis on what is the
`
`17
`
`level of skill in the art to stray into that area.
`
`18
`
`I think that to the extent t hat we have testimony as to
`
`19
`
`what a skilled artisan would know, I don't think that that goes far
`
`20
`
`enough in terms of providing evidence of the specific limitations
`
`21
`
`in these claims. Wennekamp alone does not teach multi -die ICs,
`
`22
`
`it certainly doesn't teach a m ulti-die IC as shown in figure 4
`
`23
`
`here, and the evidence --
`
`24
`
`JUDGE MEDLEY: Well, it's not in anticipation,
`
` 18
`
`25
`
`though.
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`
`MR. McCOMBS: It is not.
`
`JUDGE MEDLEY: It's obviousness. So, if you could
`
`talk to us in that respect, from the obviousness respect.
`
`MR. McCOMBS: Yes.
`
`JUDGE MEDLEY: As opposed to -- because we know
`
`that it's, you know, it's not anticipation. So, we know that.
`
`MR. McCOMBS: Yes. So, if we're looking at the
`
`Wennekamp reference, again, no discussion of multi -die ICs, to
`
`supply the obviousness component of that, by looking at what
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`one of skill in the art would know, the general teaching of
`
`11
`
`multi-die ICs, our view is that that doesn't get you there in terms
`
`12
`
`of the specifics of the claim. To do that goes too far, and would
`
`13
`
`be based on a hindsight reconstruction, using our patent as a
`
`14
`
`guide to get there.
`
`15
`
`16
`
`JUDGE MEDLEY: Did you argue hindsight?
`
`MR. McCOMBS: We argued -- no, Your Honor, we
`
`17
`
`argued that Wennekamp does not teach it.
`
`18
`
`JUDGE ARBES: Counsel, how do you address the
`
`19
`
`petitioner's argument that Wennekamp itself provides a
`
`20
`
`suggestion to have these devices be multi -die ICs? I mean, you
`
`21
`
`obviously have a master device and a slave device, petitioner has
`
`22
`
`argued in their briefing that the reference itself provides a
`
`23
`
`suggestion, and you obviously disagree with that, but is that
`
`24
`
`improper by itself for this reference to disclose certain things
`
`25
`
`and suggest that it could be a multi -die IC?
`
` 19
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`MR. McCOMBS: That's a good question, Your
`
`Honor, and our position on that is that there is a disc ussion in the
`
`Wennekamp reference of a number of circuit configurations, and
`
`we did address that in the record, and it's our view that the idea
`
`of discussing a number of circuit configurations is just that, it's
`
`just a matter of whether there's intermediat e components or
`
`whether there are not intermediate components, and that has
`
`nothing to do with the form factor, whether multi -die ICs are
`
`disclosed or even taught by the reference.
`
`10
`
`And I would direct you to Dr. Blanchard's declaration
`
`11
`
`on that point.
`
`12
`
`JUDGE ARBES: Okay, so the argument isn't that it's
`
`13
`
`improper for this reference to disclose certain things and suggest
`
`14
`
`others for an obviousness analysis, you just disagree as to
`
`15
`
`whether there actually is a suggestion in the reference?
`
`16
`
`MR. McCOMBS: I disagree that there's a suggestion
`
`17
`
`in the reference. I think that discussion of circuit configurations
`
`18
`
`only would be -- only would suggest multi -die ICs if you had our
`
`19
`
`patent in front of you.
`
`20
`
`JUDGE EASTHOM: So, do they just disclose a single
`
`21
`
`die, is that what you're saying, in Wennekamp? Is it a single die
`
`22
`
`IC?
`
`23
`
`MR. McCOMBS: The Wennekamp reference, let me
`
`24
`
`go to a different picture for that. Each of these -- each of these
`
`25
`
`are ICs. I think within the level of skill in the art, as from the
`
` 20
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`testimony in the record, that each one of those ICs could have
`
`multiple dies in it. We don't know. It's not discussed. Certainly
`
`that's within -- certainly within the realm of possibility.
`
`But each of these is a separate IC, and --
`
`JUDGE EASTHOM: Each is a separat e die?
`
`MR. McCOMBS: Each is a separate IC.
`
`JUDGE EASTHOM: Each is a separate IC.
`
`MR. McCOMBS: In Wennekamp.
`
`JUDGE EASTHOM: Each is a separate IC and has
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`separate dies?
`
`11
`
`MR. McCOMBS: Each of these ICs in Wennekamp
`
`12
`
`and could have multiple dies inside of it. There's no disclosure
`
`13
`
`of that.
`
`14
`
`JUDGE EASTHOM: So, the conglomeration would be
`
`15
`
`multiple dies, right? Each one has multiple dies?
`
`16
`
`MR. McCOMBS: Each of these ICs could have
`
`17
`
`multiple dies in them, we don't know.
`
`18
`
`JUDGE EASTHOM: So, they could either be multiple
`
`19
`
`dies or it could be a single die, is that what -- I'm just trying to
`
`20
`
`figure out what --
`
`21
`
`MR. McCOMBS: I think there's a confusion between
`
`22
`
`what a die is and what an IC is.
`
`23
`
`JUDGE EASTHOM: That's what I'm trying to figure
`
`24
`
`out, what you think the difference is, or what is the disclosure in
`
` 21
`
`
`
`
`
`
`
`Case Nos. IPR2012-00023 & IPR2012-00020
`Patents 6,653,215 & 8,058,897
`
`Wennekamp? Is it one die, multiple dies? Let's just answer that,
`
`is it one die or multiple dies?
`
`MR. McCOMBS: Well, each of those are just
`
`discussed as specific devices, and there's, again, there's no
`
`discussion here about placing multiple dies in any separately
`
`packaged IC, other than the particular identified master device,
`
`slave device, et cetera, each of which is an IC, and they're not
`
`put together on a single -- this is not a situation that's disclosing
`
`a single IC comprised of separate dies. It's a disclosure of
`