`571-272-7822
`
`
`
`
`
`
`Paper # 29
`Entered: July 12, 2023
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`MICRON TECHNOLOGY, INC.; MICRON SEMICONDUCTOR
`PRODUCTS, INC.; and MICRON TECHNOLOGY TEXAS LLC,
`Petitioner,
`
`v.
`
`NETLIST, INC.,
`Patent Owner.
`____________
`
`IPR2022-00418
`Patent 8,301,833 B1
`____________
`
`Record of Oral Hearing
`Held: June 7, 2023
`____________
`
`Before GEORGIANNA W. BRADEN, SHEILA F. McSHANE, and
`KARA L. SZPONDOWSKI, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`JUAN YAQUIAN, ESQUIRE.
`Winston & Strawn LLP
`800 Capitol Street
`Suite 2400
`Houston, Texas 77002-2925 (713) 651-2600
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`REX HWANG, ESQUIRE
`Skiermont Derby, LLP
`633 West Fifth Street
`Suite 5800
`Los Angeles, California 90071
`(213) 788-4300
`
`MICHAEL RICKETTS, ESQUIRE
`RYAN HARGRAVE, ESQUIRE
`Skiermont Derby, LLP
`1601 Elm Street
`Suite 4400
`Dallas, Texas 75201
`(214) 978-6602
`
`
`
`
`The above-entitled matter came on for hearing on Wednesday,
`June 7, 2023, commencing at 1:00 p.m., by video.
`
`
`
`2
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`
`P R O C E E D I N G S
`- - - - -
`JUDGE SZPONDOWSKI: Good afternoon. We have our final
`hearing in IPR2022-00418. Let me introduce the Panel. I'm Judge
`Szpondowski and joining me are Judge McShane and Judge Braden. So,
`let's get started with the party's appearances. First, who do we have here
`from Petitioner?
`MR. YAQUIAN: Hi. May it please the Board, for Petitioner we have
`Juan Yaquian from Winston & Strawn. With me today is Mike Rueckheim,
`he'll be handling the argument. And with me also, who I brought along to
`observe is from the Client Ms. Becky Caisora (phonetic), David Westergard
`(phonetic), Jan Bissy (phonetic), and Casper Larson (phonetic).
`JUDGE SZPONDOWSKI: Okay, thank you. And for Patent Owner?
`MR. HWANG: Yes. May it please the Board, my name is Rex
`Hwang, and with me today are Michael, he goes by Mickey, Ricketts, and
`Ryan Hargrave. And Mickey Ricketts, Mr. Ricketts, will be handling the
`arguments today on behalf of Patent Owner.
`JUDGE SZPONDOWSKI: Okay, thank you. Well, welcome,
`everyone. Obviously, our hearing today is virtual. Given that, we just want
`to start off by clarifying a few items. First, if you encounter any technical
`difficulties that you feel fundamentally undermines your ability to
`adequately represent your client, please, let us know immediately. For
`example, connecting the team members who provided you with connection
`information. Second, when you aren't speaking, please, mute yourself.
`Third, please, identify yourself each time you speak in order to help the
`court reporter prepare an accurate transcript. Fourth, we have the entire
`record, including all of the demonstratives. Please, refer to the
`3
`
`
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`demonstratives papers or exhibits clearly and explicitly by slide or page
`number. That will help prepare an accurate transcript for the hearing.
`Finally, there is a public line, so please, be aware that members of the public
`may be listening as well.
`We set forth the procedure for today's hearing in our May 1st, 2023
`order. Just as a reminder, each party will have a total of 60 minutes to
`present arguments. Petitioner has the burden of proof as to whether the
`challenge claims are unpatentable and will go first. Patent Owner will then
`present opposition arguments. Then, to the extent that Petitioner has
`reserved time, Petitioner will present rebuttal arguments. And then, to the
`extent that Patent Owner has reserved time, Patent Owner will present
`surrebuttal arguments. The rebuttal and surrebuttal time may not be more
`than half of the party's total argument time. We also remind the parties that
`they aren't to interrupt the other party while the other party is presenting its
`arguments and demonstratives. If a party believes that a demonstrative or
`argument presented is objectionable for any reason, you should raise that
`objection or any arguments relating to it only during your own time. Does
`Counsel for Petitioner have any questions before we get started?
`MR. RUECKHEIM: No questions. Thank you.
`MR. YAQUIAN: Thank you.
`JUDGE SZPONDOWSKI: And does Counsel for Patent Owner have
`any questions?
`MR. RUECKHEIM: One question. I heard a little bit of an echo
`earlier when Mr. Yaquian was speaking. Are we coming in okay?
`JUDGE SZPONDOWSKI: Yes, you sound fine.
`MR. RUECKHEIM: Thank you.
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`
`JUDGE SZPONDOWSKI: And does Counsel for Patent Owner have
`any questions?
`MR. HWANG: No questions, Your Honor.
`JUDGE SZPONDOWSKI: Okay, then I think we're ready to begin.
`Would Petitioner like to reserve any time for rebuttal?
`MR. RUECKHEIM: Twenty minutes, please.
`JUDGE SZPONDOWSKI: Okay, then I will put 40 minutes on the
`clock, and you can begin whenever you are ready.
`MR. RUECKHEIM: Michael Rueckheim for the Micron Petitioners
`and may it please the Board, can we bring up the slides? Slide 2 of the
`Petitioner’s demonstratives.
`This is not your typical IPR. The Board has already determined that
`identical claims, the materially identical claims were invalid for related U.S.
`’831 patent, and the same results should apply here. To simplify the issues
`for the Board, Micron relies upon the same prior art combination that the
`Board has already analyzed and found to be obviate on material identical
`terms -- claims. Netlist counters here are simply unsupportable. These are
`hail Mary type counterarguments, they weren't raised in the prior
`proceedings with the ’831 patent. And they go -- the claim construction
`argument that Netlist relies upon flies in the face of well-established claim
`construction canon. Netlist’s teaching away argument ignores express
`teaching of the references. These arguments were not raised in the prior
`proceeding, and they're not compelling here. If we can turn to slide three.
`As a general roadmap, I'm going to provide a quick overview of the
`challenged patent here, really to orient everybody as to what we're talking
`about. I'm going to provide a very quick summary of the prior art and how it
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`relates to the patents, and then I'm going to go through the primary disputes
`between the parties.
`We have two grounds. Ground one is the collateral estoppel ground
`based on claims 1 and 15 being materially identical to the ‘831 patent.
`Ground two discusses all of the claims and how we've provided the analysis
`and shown the Board how the references at issue here render obvious all of
`the claims. We're going to start the substantive discussion with ground two
`and then leave the collateral estoppel at the end. The Board does not need to
`find collateral estoppel in this case in order to find these claims obvious. All
`the analysis has been provided in the briefing that has been presented to the
`Board. But collateral estoppel, the purpose of collateral estoppel is to allow
`in equity a way to streamline additional lawsuits, stop parties from making
`different arguments after the arguments is made the first time and lost, and
`that’s exactly what happened here.
`If we could go to slide six, slide six shows claim 1 of the challenged
`’833 patent. There's three primary steps in claim 1, and it’s all these steps
`that are repeated in independent claim five, as well as the other independent
`claim. The first step, highlighted in red, is that you have a step of operating
`volatile memory subsystem at a first clock frequency when data is being
`transmitted between a host computer and that volatile memory. Step two, in
`blue, is operating the nonvolatile memory operating a nonvolatile memory
`subsystem at a second clock frequency when data is being written between
`the volatile memory and the nonvolatile memory subsystem. And step three
`is stating that you are going to operate the volatile memory at a reduced third
`clock frequency when you're in this second mode of operation.
`
`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
`
`
`
`6
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`
`As the Board knows, these are comprising type claims. These are not
`the consisting type claims claim recitals, where it's limited only to what's
`recited in the claim. The claim requires the recited features here but are not
`limited to those features.
`If we could turn to slide five. Slide five is depicting figure 1 of the
`‘833 patent and how it relates to the claim we just looked at. There is the
`nonvolatile memory subsystem identified in blue, the volatile memory
`subsystem identified on the top right in purple. And then we also circle the
`host system as well as identified in orange. We also circled in green the
`controller, which is not recited in claim 1 or 15, but a controller is separately
`recited in claim 16, which is a disputed claim between the parties and the
`briefing. So, we circled the controller as well. As the Board will note, there
`are other features that are not highlighted, all in this figure 1 and figure 1 is
`the example memory system compatible with certain embodiments in the
`patent. There's a switch, there's a power bank. We did not circle these
`features as they did not come in -- they did not -- are not separately recited
`in really the claims at issue. There is a switch recited in claim 17, and that is
`shown in this figure. Netlist arguments for claim construction seem to imply
`that a switch is required for the independent claims here to some fashion.
`It's shown in the figures. It's not recited in the claim. There is no switch
`separately recited in the independent claims. There is one recited in claim
`17.
`
`If we can turn to slide seven, I'm going to touch on briefly the three
`prior art references here that the Petitioners are relying upon and that the
`Board has reviewed previously with respect to the ‘831 patent IPR. If we
`can turn to slide eight the primary reference that we're relying upon is called
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`the Best reference, that's person is named Best. It includes the same primary
`components that we have been discussing with respect to claim 1. You have
`your volatile memory storage and that's highlighted in purple. They call it
`the DRAM. You have your nonvolatile storage highlighted in blue. That's
`called flash. Notably, in the challenge patent, it also refers to the DRAM
`and the volatile storage system as an example of DRAM and flash is an
`example of the nonvolatile storage system. Notably, Best describes that the
`volatile memory can be run in two different modes, either synchronous or
`asynchronous as explained by the experts, synchronous really refers to
`having a clock frequency set with respect to operations from the volatile
`memory. Turn to slide nine.
`Best is not explicit as to whether the nonvolatile memory is run in
`synchronous or in asynchronous mode. Therefore, we are relying upon the
`Mills reference for the second primary claim element of operating the
`nonvolatile memory at the second clock frequency. When data is transferred
`between the volatile memory and the nonvolatile memory, Mills discloses
`exactly that. Mills involves a flash memory interface, the nonvolatile
`memory and using a clock input as part of that interface. Turn to slide ten.
`Finally, we are relying upon the Bonella reference for the last primary
`claim element that we just discussed. This is operating the volatile memory
`at the third reduced clock frequency when in the second mode of writing
`between the host -- and sorry -- when writing between the volatile and
`nonvolatile memory. At first, as explained by Petitioner's expert, it was very
`well known to reduce volatile memory frequency in order to reduce power
`consumption. And Bonella is just one example of this well-known
`technique as it describes this reduction process as a significant savings in
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`power resulting in major power savings. Bonella teaches this exact
`approach is particularly advantageous when there's a power loss. It teaches
`a right buffer flushing process that can be triggered by a power loss in order
`to, of course, reduce power frequency, which is something that you would
`want to do when you have a power loss. It reduces the power frequency to
`DRAM as part of a power state aware process. Turn to slide 24.
`I'm going to start the discussion with regard to the substance disputes
`here by addressing Netlist claim construction proposals first and then I'm
`going to discuss ground two disputes after that, or maybe, ground two
`disputes. Slide 25. The primary claim construction disputes which were
`not, again, raised in the prior IPR involving material identical claims,
`involve Netlist assertion that the words “first mode of operation” and
`“second mode of operation” in the claims should be construed. So, to start
`the analysis, it's not disputed that it's a fundamental rule of claim
`construction that claims are, given their ordinary meaning absent a clear
`express disavowal in the specification or special lexicography. Here the
`claims are clear on their face. These are not complex technical words. What
`Netlist is trying to construe is the term’s first mode of operation or second
`mode of operation.
`JUDGE McSHANE: So, Mr. Rueckheim, I may be jumping ahead a
`little bit, but are you -- is your exact proposal that the term, for instance, first
`mode of operation -- or first mode should be defined by the claim language
`itself, that is, the recital of the actual operation in the claim language?
`MR. RUECKHEIM: So, Petitioner position is the term first mode of
`operation actually needs no construction. It's clear language on its face and
`we know from the context of the claims, which is the best place to look,
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`what the claim is referring to, the claim is referring to in the first mode of
`operation there must be data communicated between the volatile memory
`subsystem and the host system. But the claim is, again, a comprising type
`claim. It's not limited to only that happening. And of course, that would
`make it -- that wouldn't make sense. There's so many operations that would
`have to occur in a memory system when you're transferring data between a
`host system and a follow the memory subsystem, that if this claim was read
`to be only requiring this first mode, having this data transferred, it would fly
`the face of many of the embodiments of the patent and the challenge patent
`itself. There's a -- more can happen but we know at least the first mode as
`required by the claim, has to have data communicated between the volatile
`memory subsystem and the host system.
`JUDGE McSHANE: Okay, that's a good point. Thank you.
`MR. RUECKHEIM: So, going back, in order to construe a claim and
`for clarity, Micron, the Petitioners here are not proposing these terms first
`mode of operation or second mode of operation need any construction at all.
`There has to be special lexicography or disavow. Here it's undisputed.
`Netlist expert actually admitted there is no special lexicography, there's no
`definition provided in these patents. There's no recital of first mode
`operation in quotations and then a meaning ascribed next to it. There's also
`no express disavowal mentioned in the specification. In fact, Netlist briefing
`has not even offered or used the word disavowal, hasn't offered a disavowal
`argument. Turn to slide 26.
`Instead, Netlist is asking the Board -- and the table on the left hand
`side of the screen on slide 26 shows the two disputed terms and how Netlist
`is trying to construe these terms. And just looking at the proposed
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`constructions alone shows the high hurdle, I would say, that Netlist faces in
`urging this construction. First, the term mode of operation is recited in both
`of the constructions and the claim terms. And so, if you take out the mode
`of operation, Netlist is not arguing that mode of operation needs any special
`construction. Netlist is actually arguing just the term first needs construction
`or the term second needs construction. And as the Board knows, these are
`very common ordinarily understandable terms. And the term first does not
`ordinarily mean, and here we can look at the bolded language on the screen
`that Netlist is proposing, that there is some kind of no backup or restore
`operations are performed. Or does the word first refer to whether
`nonvolatile memory subsystem communicates with the volatile memory
`subsystem? Netlist is trying to turn one word into 16 to 25 different words.
`Now, Netlist had the opportunity to change these terms if it thought
`this was its patent, really, if it invented concept was all about, Netlist could
`have chosen to pursue a motion to amend the claim terms in this IPR
`proceeding. Netlist did not choose to go down that route and Netlist should
`not be allowed to use -- to amend these claims through a claim construction
`process, particularly here claim construction proper that is completely
`unsupportable. Turn to slide 27.
`So, going back to Your Honor's previous question, the context of the
`claims here do provide context and understanding to the extent first mode of
`operation needed any more clarity, what at least needs to happen. There has
`to be these data transfers that happen. But looking at the context of the
`claim also shows the problems with Netlist’s proposal. And with respect to
`the first mode of operation, there is no recital of a backup or restore
`operation per the first or second mode of operation. The claims don't recite,
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`backup, or restore. The claims don't refer at all when it discusses the first
`mode of operation to any limitation on nonvolatile memory. And that's
`exactly what Netlist is proposing. And that's the same for the second mode
`of operation. The claim context here doesn't discuss the host at all, yet
`Netlist’s proposal imposes limitations on the host. Turn to slide 28. So,
`another issue with Netlist’s proposal is that Netlist is trying -- Netlist, the
`crux of Netlist's proposal is that there are certain embodiments in the patent
`specification that discusses isolating volatile memory from nonvolatile
`memory using for example, a switch to decouple these. And that Petitioner
`told me that there are these embodiments in the patent specification. But it's
`another fundamental aspect of claim construction that the Court’s preceding
`Tribunal should not import embodiment language into the claims,
`particularly when there's claim construction, sorry, claim differentiation
`present that would show that importing into an independent claim and
`embodiment language is inappropriate. And that's exactly what we have
`here. Netlist is arguing that the important aspect in the specification that's
`emphasized is decoupling the volatile to nonvolatile memory using a switch.
`Independent claim 16 though recites decoupling a volatile and nonvolatile
`memory. There'd be no point, no reason to write dependent claim 16 if the
`independent claims actually required that feature.
`We actually asked, I asked the Netlist’s expert during deposition to
`explain how claim differentiation applies here and whether his opinion holds
`water under the doctrine of claim differentiation. And the Netlist’s expert
`admitted that he did not consider this doctrine in coming up with his
`opinions and the Board can credit that admission in weighing that testimony.
`And the same is true for claim 28.
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`
`Claim 28 recites that the second mode is a restore operation and that's
`exactly what Netlist is trying to impose into the independent claim under the
`doctrine of claim differentiation. That further shows that Netlist’s proposal
`is incorrect. If we could turn to slide 29.
`There are a number of reasons why Netlist and Netlist’s expert, their
`claim construction proposals here are unreliable. The primary reason, and I
`suspect we're going to get more into this on rebuttal to the extent that Patent
`Owner raises these arguments in its portion of the presentation, but Patent
`Owner is relying upon simply mere embodiment language in the
`specification.
`JUDGE BRADEN: Well, Counselor, I would say to you Patent
`Owner's argument that the importance of isolating a nonvolatile memory
`subsystem is not just an embodiment or an example, but something that's
`repeated throughout the entire specification, and therefore, should be or
`would have been considered or known by somebody of skill in the art at the
`time of the filing. And that's why it should be considered in part of claim
`construction.
`MR. RUECKHEIM: So, a few things. One, I would say this
`argument is a new argument, it’s -- that was never raised in the prior
`proceeding. They just came up with it when their first round argument didn't
`work.
`JUDGE BRADEN: But it was raised in their Patent Owner Response
`here. So, whether it was a prior argument in a different proceeding is
`irrelevant, correct?
`MR. RUECKHEIM: Under the doctrine of collateral estoppel, I
`would say --
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`
`JUDGE BRADEN: We're not discussing that doctrine of collateral
`estoppel right now. But it was raised in their Patent Owner Response,
`correct? So it was timely raised in this proceeding.
`MR. RUECKHEIM: So, I'm going to go back to the fundamentals.
`The fundamentals are you have to have disavowal, a clear disavowal or
`lexicography in order to change the ordinary meaning of terms. And there is
`under this realm, this rubric of disavowal, there have been cases that discuss
`when you have clear comparisons that really dictate how claim terms should
`be defined, maybe then the Federal Circuit or District Courts at some point
`have looked at those clear comparisons. These comparisons are not here.
`You have your Toro case cited by Netlist in their briefing that
`describes the specification, discussing some type of unitary structure that
`was required for the features discussed in the claim. The structure here, and
`I believe the same was for Amdocs, the structure here is a switch. There's no
`switch recited in the claims until you get down to dependent claim 17. The
`structure, or at least the feature here of decoupling, that's a separate feature
`they decided to recite in a dependent claim. These are not features that they
`envision as part of their independent invention. If they, if Netlist had viewed
`these features as a real inventive concept, they should have wrote the
`independent claims to cover these features or they could have moved to
`amend during this proceeding, which they did not do.
`JUDGE BRADEN: Thank you, Counsel.
`MR. RUECKHEIM: Okay, slide 30. So, going to the primary
`disputes in ground two, Best -- the first dispute is whether Best discloses this
`claim first and second mode of operation. If we can go to slide 31. This is a
`claim construction dispute. Both sides will admit that the claim construction
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`here is dispositive. If the Board finds that these terms should be given their
`plain and ordinary meaning, the Board should find that these claims are
`invalid. Turn to slide 33.
`Next primary dispute is with respect to dependent claim 16 and
`whether Best discloses or renders obvious this coupling and decoupling
`feature. Claim 16 recites that you have a controller that's configured to
`decouple the nonvolatile and volatile memory or couple the volatile to
`nonvolatile memory. And I don't think it's been disputed but they (phonetic)
`have shown that there's a controller that -- in Best that is directing the traffic
`here and there is no dispute as to whether this controller couples the volatile
`to nonvolatile memory. There's very clear disclosure of writing between the
`nonvolatile and volatile memory in this writeback type operation. The
`dispute really comes to the first part, and we can turn to slide 34, whether or
`not there's disclosure of the controller decoupling the volatiles and
`nonvolatile memory.
`So, what the Board can -- has been shown in the papers is that there's
`really two modes. There's a write back mode where you're writing back
`from volatile into nonvolatile memory at a power loss event, and then there's
`normal mode where the host is writing either to the volatile or the
`nonvolatile memory. There is no write back occurring. And so, based on
`that alone, this feature of decoupling and coupling the nonvolatile and false
`memory has been shown or rendered obvious because there is a mode where
`there is no writing back. They are decoupled at that point and Best meets
`the claim elements.
`If we can turn to slide 37 the next primary dispute. Again, this is a
`dispute that was not raised in the prior ‘831 patent IPR, but it's whether there
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`is a motivation to combine Bonella with Best. And so, Petitioners have
`shown there's a few reasons for why a person of ordinary skill in the art
`would have combined these references and it really relates to reducing
`power to the -- for these nonvolatile flash operations and reducing power
`consumption to the DRAM, which has been shown by the experts and the
`references, was a well-known technique. Turn to slide 38. So, Netlist’s
`arguments against this motivation to combine are not compelling. Netlist
`seems to be arguing that there is no -- there's a teaching away in Best
`because Best discusses just the general concept of having a high speed
`DRAM and that a POSITA would not look at Bonella's power state aware
`teaching when there's a power loss event because Best wants high speed
`DRAM. And, of course, Best envisions high speed DRAM. That's a general
`marketing goal for every type of technology to be faster, smaller, better. But
`that's not going to deter a person of ordinary skill in the art of looking at
`references for specific use cases. And that's exactly what Bonella teaches, a
`specific use case. When you have a power loss event which is envisioned by
`both references, one way you can deal with a power loss event is to reduce
`the power consumption, the clock frequency for DRAM as explained by the
`Petitioners and the expert, the (inaudible) expert. A person of ordinary skill
`would recognize that benefit, which Bonella discusses as a major power
`savings, a significant advantage and apply it to Best. Let's turn to slide 40.
`Finally, as we stated previously, Netlist's two primary arguments,
`claim construction and this motivation to combine argument, were never
`presented in the prior proceeding involving the material identical claims.
`And the Board can apply collateral estoppel to stop Netlist from having its
`second bite of the apple in this proceeding. Finding anything else would
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`result in an inconsistent decision when the Board has already found invalid
`based on the same prior combination material identical claims. Turn to slide
`41.
`
`And we've gone through this in the briefing, but the general idea is
`that all the factors here have been met for collateral estoppel. The prior -692
`IPR presented identical issues. It's the same claims here, material identical
`terms. The IPR was actually litigated and the Judge did not challenge on
`appeal, the final judgment. The IPR necessarily involved determining
`identical issues, what these claim terms mean and should these combination
`of prior art be combined, there was full representation by Netlist in the prior
`proceeding. We can turn to slide 15.
`So, going to the general concept of collateral estoppel more broadly,
`there was discussion in the institution decision as to whether there are minor
`differences in the claims here -- make a difference with respect to collateral
`estoppel. And we've shown in the briefing, particularly with the Ohio
`Willow case, that differences in claims that do not materially alter the
`question of invalidity and collateral estoppel will still apply. We've shown a
`table in our Petition that shows -- what we have shown the Board to be
`material identical claims. The only difference identified by Netlist here is
`with respect to the memory controller and whether the memory controller in
`the ’831 patent is on the module or on the host. And we've shown in
`response to that argument that in the prior proceeding, the memory
`controller was actually mapped to the host system, which is exactly the same
`thing as reciting the ’833 patent. This is a distinction without a difference, it
`does not affect the determination of collateral estoppel, as shown by the
`Ohio Willow case, which actually stated that the Patent Owner had not
`
`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
`
`
`
`IPR 2022-00418
`Patent 8,301,833 B1
`
`shown any difference between the claim terms at issue that was significant
`and applied collateral estoppel. We can turn to slide 18.
`And another distinction without a difference, Patent Owner has
`referred to the BRI standard versus the Phillips standard as a reason why
`collateral estoppel should not apply. And BRI standard is different than the
`Phillips standard in some cases, not in all cases, as the Board knows. And
`we have not been shown any difference here as far as the application at least,
`we have not shown any difference here as far as the application of either
`standard, which, you know, without a showing of a difference, collateral
`estoppel will apply.
`But then, of course, slide 22, Bonella's last argument is that -- to argue
`that there was a lack of incentive to litigate. And we've shown in the papers
`here that this idea of a lack of incentive to litigate they had counsel in the
`prior proceeding, the prior proceeding resulted in them, in Netlist losing one
`of his patents and that's full representation. And regardless, the lack of
`incentive to litigate exception is a rare exception that requires certain things;
`a compelling showing of unfairness, such as a clear and convincing need for
`a new determination. You need to have something more. And that
`something more was not here as Netlist was actually informed in the prior
`proceeding that the ’833 patent and the ’831 patent had materially identical
`claims and with that, I'll rest the remainder of my time.
`JUDGE SZPONDOWSKI: Okay. Thank you. I guess we can go to
`Patent Owner. Would you like to reserve time for a surrebuttal?
`MR. RICKETTS: Yes, Your Honor. I'd like to reserve 15 minutes
`for rebuttal, please.
`
`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
`
`
`
`18
`
`
`
`IPR 2022