Acceleration Bay LLC v. Electronic Arts Inc. ( 2022 )

IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF DELAWARE ACCELERATION BAY LLC, Plaintiff, C.A. No. 16-454-RGA ELECTRONIC ARTS INC.,, Defendant. MEMORANDUM OPINION Philip A. Rovner, Jonathan A. Choa, POTTER ANDERSON & CORROON LLP, Wilmington, DE; Aaron M. Frankel (argued), KRAMER LEVIN NAFTALIS & FRANKEL LLP, New York, NY; Paul J. Andre, Lisa Kobialka, James R. Hannah, KRAMER LEVIN NAFTALIS & FRANKEL LLP, Menlo Park, CA; Attorneys for Plaintiff. Jack B. Blumenfeld, Cameron P, Clark, MORRIS NICHOLS ARSHT & TUNNELL LLP, Wilmington, DE; Michael A. Tomasulo (argued), David P. Enzminger, Gino Cheng, Joe S. Netikosol, WINSTON & STRAWN LLP, Los Angeles, CA; Louis L. Campbell, WINSTON & STRAWN LLP, Menlo Park, CA; Daniel K. Webb, WINSTON & STRAWN LLP, Chicago, IL; Joseph C. Masullo, WINSTON & STRAWN LLP, Washington, DC; Attorneys for Defendant. October ‘L 2022 bel G Medllac so: Before me is Defendant’s motion for summary judgment of non-infringement by collateral estoppel. (D.I. 580). I have considered the parties’ briefing. (D.I. 581, 582, 586). For the reasons set forth below, I will GRANT Defendant’s motion. I. BACKGROUND On June 17, 2016, Plaintiff Acceleration Bay filed suit against Defendant Electronic Arts (“EA”) alleging infringement of U.S. Patent Nos. 6,701,344 (344 Patent), 6,714,966 (966 Patent), 6,732,147 (147 Patent), 6,829,634 (634 Patent), 6,910,069 (069 Patent), and 6,920,497 (°497 Patent). (D.I. 1 at 410). After prior summary judgment rulings on infringement and on invalidity, two infringement . allegations remain: (1) DOE infringement by all accused products of the ’147 Patent, and (2) literal infringement by the accused NHL and Plants vs. Zombies (“PvZ”) games of the ’344 and □□□□ Patents when EA internally tests those games in the United States. (See D.I. 581 at 1 (citing Acceleration Bay LLC v. Activision Blizzard, Inc., 324 F. Supp. 470, 478-59, 485-87 (D. Del. 2018) (summary judgment opinion addressing claim invalidity); Acceleration Bay LLC v. Elec. Arts Inc., No. 16-454-RGA, 2019 WL 1376036 (D. Del. Mar. 27, 2019) (summary judgment opinion addressing noninfringement by Defendant)); D.I. 582 at 2 n.1 (confirming that Plaintiff “is narrowing its election of asserted claims to no longer include any claims from [the °497 Patent]”)). This case is related to Acceleration Bay LLC v. Take-Two Interactive Software, Inc., No. 16-455-RGA (“Take-Two Case”), where Plaintiff accused online features of three video games— NBA 2K15 and NBA 2K 16 (collectively, “NBA 2K”), and Grand Theft Auto Online (““GTAO”)— of infringing the °344, °966, ’147, °069, and ’497 Patents. See Take-Two Case, 2020 WL 1333131, at *1 (D. Del. Mar. 23, 2020) (“Take-Two SJ Opinion”), appeal dismissed sub nom. Acceleration Bay LLC v. 2K Sports, Inc., 2020 WL 9459373 (Fed. Cir. Oct. 2, 2020) (dismissing cross-appeal), and aff'd in part, dismissed in part sub nom. Acceleration Bay LLC vy. 2K Sports, Inc., 15 F.4th 1069 (Fed. Cir. 2021) (“Take-Two Appeal”). In the Take-Two Case, the defendants moved for summary judgment of non-infringement, which I granted in a detailed opinion. See Take-Two SJ Opinion. Plaintiff appealed. See Take-Two Appeal. On April 21, 2020, I stayed this case pending resolution of Plaintiff's appeal of my summary judgment ruling in the Take-Two Case. (See D.I. 561). The Court of Appeals for the Federal Circuit issued a decision on that appeal on October 4, 2021. See Take-Two Appeal (affirming-in-part the Take-Two SJ opinion and dismissing-in-part Plaintiffs appeal on mootness grounds). With the Take-Two Case’s appeal resolved, Defendant now moves for summary judgment of noninfringement, arguing that Plaintiff is collaterally estopped from relitigating infringement issues it lost in the Take-Two Case. In this case, each remaining asserted claim requires a network that is “m-regular.” I construed “m-regular” to mean “[a] state that the network is configured to maintain, where each [participant or computer] is connected to exactly m neighbor [participants or computers].” □□□□□ 260 at 5). This construction also applied in the Take-Two Case, and Defendant did not appeal this construction. See Take-Two Appeal. I held a hearing on this motion on September 30, 2022. (See D.I. 588). Il. LEGAL STANDARD “The court shall grant summary judgment if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). Material facts are those “that could affect the outcome” of the proceeding, and “a dispute about a material fact is genuine if the evidence is sufficient to permit a reasonable jury to return a verdict for the nonmoving party.” Lamont v. New Jersey, 637 F.3d 177, 181 (3d Cir. 2011) (cleaned up). In deciding a motion for summary judgment, the court must view the evidence in the light most favorable to the non-moving party and draw all reasonable inferences in that party’s favor. See Scott vy. Harris, 550 U.S. 372, 380 (2007); Wishkin v. Potter, 476 F.3d 180, 184 (3d Cir. 2007). Collateral estoppel, also known as issue preclusion, bars parties from relitigating matters that they previously had a full and fair opportunity to litigate. See Montana v. United States, 440 U.S. 147, 153 (1979). This “protects their adversaries from the expense and vexation attending multiple lawsuits, conserves judicial resources, and fosters reliance on judicial action by minimizing the possibility of inconsistent decisions.” Jd. at 153-54. In a patent case, the law of the regional circuit applies to collateral estoppel generally and Federal Circuit precedent applies where the determination of collateral estoppel involves substantive issues of patent law. See Ohio Willow Wood Co. v. Alps South, LLC, 735 F.3d 1333, 1342 (Fed. Cir. 2013). Under Third Circuit law, collateral estoppel applies when “(1) the issue sought to be precluded is the same as that involved in the prior action; (2) that issue was actually litigated; (3) it was determined by a final and valid judgment; and (4) the determination was essential to the prior judgment.” Burlington N. R. Co. v. Hyundai Merch. Marine Co., 63 F.3d 1227, 1231-32 (3d Cir. 1995) (cleaned up). The “essential to the prior judgment” element can be satisfied when the prior judgment was reached through alternative findings. Jean Alexander Cosms., Inc. v. L’Oreal USA, Inc., 458 F.3d 244, 255 (3d Cir. 2006) (“we will follow the traditional view that independently sufficient alternative findings should be given preclusive effect”). Whether the “basic requirements for issue preclusion are satisfied” is a question of law. Id. at 4 □ 248; see also Ohio Willow Wood, 735 F.3d at 1341 (de novo review of the application of collateral estoppel). As is particular to patent law, “an infringement claim in a second suit is the same claim as in an earlier infringement suit if the accused products in the two suits are essentially the same.” Phil-Insul Corp. vy. Airlite Plastics Co., 854 F.3d 1344, 1353 (Fed. Cir. 2017) (cleaned up). “Accused devices are essentially the same where the differences between them are merely colorable or unrelated to the limitations in the claim of the patent.” Jd. (cleaned up). The accused device in a second suit need not be produced by the same company as that considered in a first suit. See Aspex Eyewear, Inc. v. Zenni Optical Inc., 713 F.3d 1377, 1380 (Fed. Cir. 2013) (in holding that “[c]ollateral estoppel precludes a plaintiff from relitigating identical issues by merely switching adversaries|,|” affirming “that [a second defendant’s] accused rimless magnetic clip-on sunglasses are materially indistinguishable from [a first defendant’s] rimless magnetic clip-on sunglasses”) (cleaned up). The alleged infringer “bears the burden of showing that the accused devices are essentially the same as those in the prior litigation.” ArcelorMittal Atlantique et Lorraine v. AK Steel Corp., 908 F.3d 1267, 1274 (Fed. Cir. 2018). Il. DISCUSSION A. Issues Previously Adjudicated in the Take-Two Case The Parties’ dispute centers around three noninfringement issues I decided in the Take- Two SJ Opinion. (See D.I. 581 at 2-3). First, for GTAO, I considered the player movement issue. Plaintiff argued that GTAO infringes the m-regular limitation because the players’ avatars “share more data when they are near each other” thus causing an m-regular network to “arise naturally as the players are moving throughout the game.” Take-Two SJ Opinion at *8 (cleaned up). In rejecting this argument, I held, “Under my claim construction, a network is not m-regular if the participants just happen to connect to the same number of other participants occasionally. Rather, the network must be ‘configured to maintain’ an m-regular state.” Jd. at *8. Based on this, I held that “the [GTAO] players’ actions determine how connections are formed, and the network is not ‘configured to maintain’ any particular state.” Jd J explained that “if a system is designed to achieve a desired result, one would not normally say the result ‘just arises naturally.’” Jd. Second, for NBA 2K, I considered the all-connected server issue. Plaintiff argued that a server which connects to all the virtual basketball players is not a “participant” in the game and, thus, does not defeat m-regularity. Jd at *9,. I disagreed, finding that, “the server is not playing basketball .... The server is, however, a participant in the network because it transfers data back and forth between other network participants. These patent claims are directed to network management, so what matters is whether the server is a participant in the network, not whether it is making jump shots or grabbing rebounds.” Jd. Third, I considered Plaintiff’s Doctrine of Equivalents (“DOE”) argument that the GTAO network performs substantially the same function as an m-regular network by “optimizing the entire network processing of the network by limiting each participant’s network connections” so that “data are distributed in a balanced fashion over the network[.]” Jd. at *8 (cleaned up); see also id. at *10 (similar DOE argument for NBA 2K). I rejected this DOE argument because it “effectively reads the m-regular limitation out of the patent” when “[t]here is no mention of participants connecting to the same number of other participants.” Jd. at *9; see also id. at *10 (rejecting Plaintiff's DOE argument for NBA 2K for the same reason). Defendant explains that these three issues were actually litigated, determined by final and valid judgment, and essential to the prior judgment. (See D.I. 581 at 18; see also Burlington, 63 F.3d at 1231-32). Plaintiff does not contest that these requirements for collateral estoppel are met. (See generally D.J. 582). I agree with Defendant, and find that these three issues were actually litigated (see generally Take-Two SJ Opinion; Take-Two Appeal), were determined by a final and valid judgment (see generally id; see also Phil-Insul Corp., 854 F.3d at 1357 (“the noninfringement determinations in [a previous case] are final for collateral estoppel purposes by virtue of [a plaintiff’s] failure to appeal them”)), and were essential to the prior judgment (see Take-Two SJ Opinion at *7-10). B. Issues Currently Being Adjudicated Defendant argues that noninfringement issues being considered in this case are identical to the player movement, all-connected server, and DOE issues that were previously adjudicated in the Take-Two Case. 1. Player Movement Issue Defendant argues that, just as with GTAO in the Take-Two Case, Plaintiffs infringement arguments rely on a “claim that the [Defendant’s] networks may or may not be m-regular depending on the players’ actions in the game.” (D.I. 581 at 14). Specifically, Defendant explains that Plaintiff's infringement argument relies on the “voice squelching” feature of the accused games, whereby, “‘When [the] DirtyCast [server] detects that there are four VoIP streams being directed to a participant, it will prevent additional VoIP data from being sent to that participant at that time.’” (D.I. 581 at 15 (quoting D.I. 428, Ex. A-1 (Medvidovic Report) at qqi54, 168, 177, 193)). Plaintiffs expert explains, “Thus, the network is configured so that each participant can only have four VoIP data connections at the same time, making the network m- regular, with m=4.” (DI. 428, Ex. A-1 at 49154, 168, 177, 193)). Considering Plaintiff's argument, Defendant contends, “Here—as in Take-Two—the players’ actions determine whether voice squelching is activated. ... If no players talk, or even if one, two, three, or four players talk, voice squelching is not activated and there is no m-regular network under Acceleration’s theory. Under Acceleration’s theory, only when five players try to talk at the same time does an m-regular network form [which] is exactly like the non-default, transitory, player-generated m-regular network theory that this Court found lacking in Take-Two.” (D.I. 581 at 16). Plaintiff disagrees, arguing that “the voice squelching rules used in the Networks make the broadcast of voice data 4-regular, because each play only has 4 voice data connections, irregardless of the players’ actions.” (D.I. 582 at 13). Plaintiff further asserts, “[Defendant] suggests that these are transient connections that only exist when a player [chooses] to talk, but that is incorrect.” Instead, “The voice connection is implemented through VoIP (Voice Over Internet Protocol) tunnels which create persisting voice connections to certain participants, such as in a squad or group.” (id. (citing D.I. 442, Ex. 1 (Medvidovic Report) at 4133 (including a string cite to source code))). At the close of the hearing, Plaintiff cited additional evidence to support its argument that the VoIP tunnels are “persistent” and m-regular independent of player actions. (See D.I. 429, Ex. A-3 (Medvidovic Reply Report) at J931, 61, 66, 74). During the hearing, Plaintiff also suggested that “the use of [team] channels ... in PvZ, where each participant is directly connected only to the other members of its squad and not to all of the other participants in the game” establishes m-regularity for the PvZ network even in the face of voice squelching. (D.I. 582 at 6-7 (citing D.I. 428, Ex. A-1 at 99128-142, 174-188)).! At the ! While referring to the PvZ team channels in the background section of its briefing, Plaintiff did not include in its arguments that the PvZ team channels defeat Defendant’s summary judgment arguments for voice squelching. (See D.I. 582 at 6-7, 13; see also John Wyeth & Bro. Ltd. v. CIGNA Int'l Corp., 119 F.3d 1070, 1076 n.6 (3d Cir. 1997) (“[A]rguments raised in passing ..., but not squarely argued, are considered waved.) close of the hearing, Plaintiff cited additional evidence to support its argument that the PvZ team channels are m-regular. (See D.I. 429, Ex. A-3 at 9953, 77; D.I. 434, Ex. E-9 at 93:3-23). I agree with Defendant. I find that the accused products in this case and NBA 2K are essentially the same for purposes of the player movement issue. Just as was the case with NBA 2K, for the presently accused products, “it is the players’ actions [in communicating through their microphones that] determine how connections are formed, and the network is not ‘configured to maintain’ any particular state.” (Take-Two SJ Opinion at *8). Particularly, Plaintiffs expert explains that the DirtyCast server enacts voice squelching upon “detect[ing] that there are VoIP streams being directed to a participant.” (D.I. 428, Ex. A-1 at 49154, 168, 177, 193)). Thus, while “the network is configured so that each participant can only have four VoIP data connections at the same time[,]” the network only enacts voice squelching when players are speaking. (/d.). “Under my claim construction, a network is not m-regular if the participants just happen to connect to the same number of other participants occasionally” based on whether players are speaking. (Take-Two SJ Opinion at *8). Further, during the hearing regarding this Motion, Plaintiff could not meaningfully contest that. the voice-squelched VoIP streams are impersistent and dependent on player actions. Particularly, while Plaintiff pointed to paragraphs 31, 61, 66, and 74 of Dr. Medvidovic’s Reply Report during the hearing, these paragraphs do not support the notion that the VoIP tunnels are persistently m-regular. (See D.I. 429, Ex. A-3 at 9931, 61, 66, 74). For example, paragraph 31 describes an “underlying physical network” of VoIP tunnels “through which all of the VoIP traffic flows[,]” where voice squelching cause “some data [to] not [be] sent over some of those [VoIP tunnels] fora period oftime.” (See id at§31). Thus, paragraph 31 suggests that the VoIP tunnels are not m-regular and exist as underlying connections between all players, and that voice squelching limits which tunnels are used based on which players are speaking. That is not a persistent, m-regular network and, instead, any m-regularity would be caused by player actions. Similarly, paragraphs 61, 66, and 74 do not support Plaintiff’s argument that the VoIP tunnels form a persistently m-regular network. Additionally, while Plaintiff suggested that the PvZ team channels—where each player is connected only to the other members of their team—are uniquely m-regular, I disagree. As Plaintiff recognized in its briefing, the PvZ team channels rely on voice squelching. (See D.I. 582 at 6 (noting “the use in the Networks of voice squelching[,]” where “VoIP tunnels operate in the same way for all three Accused Products”); see also D.I. 428, Ex. A-1 at 9131 (explaining that, for PvZ, “not all participants in a given game session are connected through [the] sub-channels. ... For a given channel, a participant can have four neighbor participants at a time such that only four VoIP messages can be heard at a time due to the VoIP squelching rules[.]”), 4182 (explaining that, for PvZ, “the VoIP functionality uses the PvZ logical Sub-Channels”)). Thus, independent of whether PvZ limits the potential voice connections to the members of each team, PvZ’s use of voice squelching causes impersistent connections that are dependent on the players’ actions and, thus, defeats m-regularity. Plaintiff cites to paragraphs 53 and 77 of Dr. Medvidovic’s Reply Report as suggesting otherwise, but, from my review, these paragraphs do not suggest that the voice squelching feature common to each of the accused products works differently based on PvZ’s use of team channels. (See D.I. 429, Ex. A-3 at 9953, 77; see also D.I1. 434, Ex. E-9 at 93:3- 23 (unrelated testimony regarding the DirtyCast server “always [being] a participant” in the accused games)). 10 Thus, Plaintiff is collaterally estopped from proceeding with its literal infringement claims because, as was determined in the Take-Two Case, a network cannot be m-regular if the m- regularity is dependent on players’ actions. 2. All Connected Server Issue Defendant argues that the DirtyCast server, on which each of the accused products relies, is a network participant that is connected to all other participants. (See D.I. 581 at 8-13; see also D.I. 442, Ex. 1 (Medvidovic Report) at 42 (for each of the accused products, asserting, “The participants are [FIFA/NHL/PvZ] software application programs, running on player consoles (PC and Xboxes), and EA’s DirtyCast server”), {79 (“The DirtyCast server is part of the underlying network and is also a participant in the application overlay FIFA Network when it applies game logics and selectively forwards data.”), J986, 93, 154, 168, 177, 193; DI. 434, Ex. E-9 (Medvidovic Dep. Tr.) at 93:3-23 (“Q: ... are you trying to say that the DirtyCast server is sometimes a participant and sometimes not? A: No. It’s always a participant, but the way it participates is by applying the game logics and forwarding data.”)). Through the cited evidence, Defendant strongly conveys that, according to Plaintiff's own expert, the DirtyCast server is a network participant. Plaintiff responds, “[D]efendant’s semantic argument that the EA DirtyCast server, which distributes voice data at the network layer, is a participant at the application layer is unpersuasive. ... [T]he DirtyCast Server is not a participant in the application layer network in the relevant sense because it never processes the voice data in the way that the players, who are the participants at the application layer[,] do[.]” (D.L. 582 at 12 (citing D.I. 434, Ex. E-9 at 93:3-23; DI. 442, Ex. 1 at §168)). Plaintiff emphasizes that, in denying Defendant’s earlier motion for summary judgment of non-infringement, I found that there is a factual dispute regarding “the appropriate 11 ‘layer’ of the network at which to evaluate infringement.” (D.I. 582 at 12-13 (citing Acceleration Bay, 2019 WL 1376036, at *7)). I find Plaintiff's assertion that the DirtyCast server is not a participant “in the relevant sense” to be unsupported attorney argument. Indeed, Plaintiffs cited evidence does not support its argument. For example, paragraph 168 of Dr. Medvidovic’s report does not explain whether the DirtyCast server is a participant in the application layer. (See D.I. 442, Ex. 1 at 9168). Instead, paragraph 168 describes the DirtyCast server’s function in “ensur[ing] that ... player[s] [do] not receive VoIP messaging from all other players ... at the same time” before providing a string cite to the source code. (See id.). I find that Plaintiff's expert Dr. Medvidovic clearly posits that the DirtyCast server is a network participant for each of the accused products. (See, e.g., D.I. 442, Ex. 1 at J92, 79, 86, 93). Dr. Medvidovic plainly states that the DirtyCast server is “always a participant [and] participates ... by applying the game logics and forwarding data.” (D.I. 434, Ex. E-9 at 93:3-23; see also D.J. 442, Ex. 1 at 479 (explaining that, when the DirtyCast server “applies game logics and selectively forwards data,” it is a “participant in the application overlay ... [n]etwork”)). Similarly, in the second paragraph of his expert report, Dr. Medvidovic lists “EA’s DirtyCast server” as a network participant for each of the accused products. (D.I. 442, Ex. 1 at 42). Plaintiffs attorneys cannot limit those statements to avoid summary judgment. Thus, I find the accused products in the present case and NBA 2K in the Take-Two Case to be essentially the same for purposes of the all-connected server issue. See Take-Two SJ Opinion at *9 (“what matters is whether the server is a participant in the network”). This basis for summary judgment of no literal infringement is in addition to the independent basis described in Section III.B.1 above. 12 For these reasons, I find that Plaintiff is collaterally estopped from advancing its remaining literal infringement theories in this case. 3. Doctrine of Equivalents Issue Defendant argues that the DOE issues being considered in this case are identical to the DOE issues that were previously adjudicated for GTAO and NBA 2K, particularly considering that for the ’147, ’344, and ’966 Patent—the only patents still at play in this case—“the patentee added the m-regular limitation during prosecution [thus barring Plaintiff] by prosecution history estoppel from now attempting to erase that limitation from the patents.” (Take-Two SJ Opinion at *9; see D.I. 581 at 17-18). Defendant asserts that estoppel “applies because the features of the accused EA games alleged to infringe by equivalents are essentially the same as those rejected by this Court for both NBA 2K and GTAO [where] [i]n both cases, [Plaintiffs] experts used nearly identical language in alleging infringement by equivalents.” (D.I. 581 at 17 (citing D.I. 428, Ex. A-1 (Medvidovic report for the present case) at J199-201, 211-213, 226-228; Take-Two Case, D.I. 464, Ex. A-1 (Medvidovic report for the Take-Two case) at J9192-194)). Without citing any evidence, Plaintiff disagrees, arguing that the DOE issues are not essentially the same because “[t]he [EA] Networks are m-regular through the imposition of specific rules, such as the voice squelching, which makes the Networks 4-regular for voice data[,] and team channels, which makes the Network m-regular based on the number of players in each team.” (D.I. 582 at 13-14). In granting summary judgment in the Take-Two SJ Opinion, I explained that Plaintiff's DOE arguments for both GTAO and NBA 2K were flawed because they attempt “to remove inconvenient claim elements, such as the m-regular limitation.” (Take-Two SJ Opinion at *10). This reasoning was underscored by the fact that “for the °344, ’966, and °147 patents ... the 13 patentee added the m-regular limitation during prosecution” to overcome “a specific prior art reference[,]” thus barring Plaintiff “by prosecution history estoppel from now attempting to erase that limitation from the patents.” (/d. at *9). Iagree with Defendant. The portions of Plaintiffs expert’s reports cited by Defendant show that there are no material differences between the accused products in this case and Take-Two as they relate to Plaintiffs rejected DOE arguments. My belief that the accused products are essentially the same is buttressed by Plaintiff's inability to point to any evidence supporting meaningful differences in the accused products as they relate to infringement of the m-regular limitation under DOE. (See D.I. 582 at 13-14; see also Phil-Insul Corp., 854 F.3d at 1353; Transocean Offshore Deepwater Drilling, Inc. vy. Maersk Contractors USA, Inc., 617 F.3d 1296, 1312 (Fed. Cir. 2010) (applying collateral estoppel when the infringement issue in a second case was identical to the issue considered in an earlier case, and product differences “[did] not change the fact that the [allegedly infringing product] does not infringe”)). For these reasons, where each of the elements for collateral estoppel is met, I find that Plaintiff is collaterally estopped from advancing its DOE theories in this case. 4, Summary Judgment Absent Collateral Estoppel Defendant also argues, “if this Court finds that collateral estoppel does not apply[,]” summary judgment of non-infringement should still be granted because “no reasonable jury could find infringement under [Plaintiffs theories[.]” (D.I. 581 at 19). That may be true, but Defendant previously moved for summary judgment of non-infringement, however, and I ruled on that motion. (See D.I. 546). Thus, Defendant’s present request for summary judgment of non- infringement is, in essence, a motion for reconsideration. I deny summary judgment on this alternative basis because Defendant has not met the standard for reconsideration. 14 IV. CONCLUSION A separate order will be entered. 15