[00:00:00] Speaker 02: California Institute of Technology versus Broadcom and Apple. [00:00:07] Speaker 02: 2020-22-22. [00:00:08] Speaker 02: Mr. Lee, good morning. [00:00:10] Speaker 01: Thank you, your honor. [00:00:16] Speaker 01: May it please the court. [00:00:18] Speaker 01: My name is Bill Layen, together with my partner, Lauren Fletcher. [00:00:21] Speaker 01: I represent the Pellance Broadcom and Apple. [00:00:24] Speaker 01: The trial court paid numerous legal errors that resulted in a billion dollar verdict that's not supported, as we contend, by the law or the evidence. [00:00:33] Speaker 02: You've argued virtually all of those errors in almost a kitchen sink appeal. [00:00:42] Speaker 02: Does the trial court use that language? [00:00:45] Speaker 01: Your Honor. [00:00:46] Speaker 01: My opponent has used that language. [00:00:48] Speaker 01: There are some appeals that turn on an issue or two. [00:00:51] Speaker 01: There are some appeals that have a host of issues. [00:00:54] Speaker 01: This happens to be one of them. [00:00:56] Speaker 01: And they include claim construction issues, damages issues, substantial evidence issues. [00:01:03] Speaker 01: And notwithstanding the characterization made by Caltech, we would not bring those issues to you unless we thought they were real and significant. [00:01:13] Speaker 01: And in my time today, what I'd like to do is focus on the claim construction and damages issues that infect this really outsized judgment. [00:01:22] Speaker 01: And if I could turn immediately to the term repeat, which has significance for all three of the patents, as you know, we contend that the district court erred in construing repeat to include merely reusing bits. [00:01:37] Speaker 04: I don't understand the significance of that argument. [00:01:40] Speaker 04: I think I do understand. [00:01:42] Speaker 04: Your non-infringement argument, but I don't understand the significance of repeat and reuse. [00:01:50] Speaker 01: So Your Honor, there are two separate issues, so let me come to both issues quickly. [00:01:55] Speaker 01: The repeat issue is significant, because if you were merely reusing, you could not get the irregular duplication that claims to require. [00:02:04] Speaker 04: What's the difference between reusing and repeating? [00:02:07] Speaker 01: Repeating allows you to get multiple copies of the bid. [00:02:11] Speaker 01: And if Your Honor turns to not only the claim limitations, but the columns 2 and 3 in each of the disposed embodiments, all of the embodiments turn on the formula that's in column 2, which is n equals qk. [00:02:28] Speaker 01: And n is the output, k is the input, q is the number of times of repeats. [00:02:35] Speaker 01: To be irregular, q is going to have to vary in number. [00:02:39] Speaker 01: It could not always be 1. [00:02:41] Speaker 01: If you were reusing, you could never have that formula irregularly repeat. [00:02:47] Speaker 01: So the answer to the first part of your question, Your Honor, is if you look at figure 2, if you look at figure 3, if you look at figure 4, if you read the embodiments, they are all consistently predicated upon n equals q times k. q is a number that has to vary. [00:03:06] Speaker 01: It has to vary, or you cannot have irregular repetition. [00:03:10] Speaker 01: And as a consequence, Your Honor, you have to have repeating, not reusing. [00:03:16] Speaker 01: That is precisely what the specification says. [00:03:19] Speaker 01: It's consistent with the claims. [00:03:22] Speaker 01: It's also, as Judge Falzer said, in her pretty well-reasoned opinion on this very issue. [00:03:28] Speaker 01: So it has consequences. [00:03:30] Speaker 01: And Your Honor, when you consider the fact that before the patent office, Caltech distinguished these claims from when? [00:03:38] Speaker 04: I'm sorry. [00:03:38] Speaker 04: I just don't understand the difference. [00:03:40] Speaker 04: I know time is limited. [00:03:41] Speaker 04: Can we turn to the, to the infringement question? [00:03:44] Speaker 04: Because as I understand it, the Caltech theory and their expert testimony said, because the output here is sometimes the same as the input that that creates infringement, which seems to me questionable. [00:04:05] Speaker 04: And their theory, I guess, is that in some circumstances, [00:04:10] Speaker 04: the information bit passes through the process and ends up being the same at the end, but not always. [00:04:20] Speaker 04: How is it that this invention could work if only sometimes the information bit passes through and comes out the other end? [00:04:37] Speaker 04: the same way. [00:04:39] Speaker 04: I thought the whole purpose of the invention here was to repeat all of these bits irregularly. [00:04:48] Speaker 01: Your Honor is exactly correct. [00:04:49] Speaker 01: And if you consider the record, which is extensive, there were three iterations of Kalchek's infringement contention. [00:04:57] Speaker 01: The first was that all the inputs to the end gates, the 972, were the repeated bits. [00:05:05] Speaker 01: That didn't work because, first, they were all the same. [00:05:09] Speaker 01: And the second, they weren't repeated. [00:05:11] Speaker 01: And the second is they weren't irregular. [00:05:14] Speaker 01: They were done 972 times like night follows the day. [00:05:18] Speaker 01: So they changed positions. [00:05:20] Speaker 01: It said, no, no, no. [00:05:21] Speaker 01: It's the outputs. [00:05:23] Speaker 01: And when the outputs, in some circumstances, depending upon the parity matrix, depending upon whether the enable bit is set to 1, [00:05:34] Speaker 01: In those circumstances, there's a repeated bit. [00:05:37] Speaker 01: It makes no sense. [00:05:38] Speaker 01: The inputs are the same. [00:05:40] Speaker 01: The AND gate performs the same function. [00:05:43] Speaker 04: And all the inputs have to be repeated. [00:05:44] Speaker 01: All the inputs are the same. [00:05:46] Speaker 04: OK, so where do I get from the specification or the claims the concept that all the inputs [00:05:55] Speaker 04: have to be repeated as opposed to occasional repeating of inputs. [00:06:00] Speaker 01: Well, Your Honor, what you get is the claims require that the bits be repeated irregularly. [00:06:09] Speaker 01: And at the input into the AND gates, they're not being repeated irregularly. [00:06:13] Speaker 01: They're all the same inputs. [00:06:16] Speaker 01: If you get to the outputs, they're all coming through the AND gate, just as Your Honor said. [00:06:21] Speaker 01: They are being output by the AND gates. [00:06:25] Speaker 01: after the parity bit matrix, also called the enable bit, is applied. [00:06:31] Speaker 01: And there's an output. [00:06:32] Speaker 01: They're all processed the same way. [00:06:34] Speaker 01: And here, Your Honor, is, I think, the key to answering your question, which is, before the district court, Caltech said, well, when the enable bit is 1, and then there's a repeated bit. [00:06:51] Speaker 01: But when the enable bit is 0, there's no enabled bit. [00:06:56] Speaker 01: They have, as we've pointed out to you at their brief at page 21 to 22, they have said exactly the opposite. [00:07:07] Speaker 01: Before the district court, they said, to sustain this billion-dollar verdict, that when the paribit is zero, there is no repeated bid. [00:07:17] Speaker 01: Before this court, they have come to you and said, no, no, no, that's not right. [00:07:21] Speaker 01: When the paribit is zero, you can't have a repeated bid. [00:07:26] Speaker 01: It cannot be that those are reconcilable. [00:07:30] Speaker 01: They're irreconcilably inconsistent, and they really demonstrate, going to your point, that pointing to some of the outputs is not efficient. [00:07:38] Speaker 04: The testimony is pretty clear that they're suggesting sometimes infringement, sometimes if the information bit is sometimes repeated, [00:07:46] Speaker 04: That constitutes infringement. [00:07:48] Speaker 04: And my question is, how can that be? [00:07:51] Speaker 04: How can this work if it only gets repeated sometimes? [00:07:56] Speaker 04: How is that within the concept of the invention? [00:08:01] Speaker 01: It can't be. [00:08:03] Speaker 01: It can't be because the bits are treated the same way at the AND gate, the XOR gate, the OR gate. [00:08:09] Speaker 01: They're output after being processed the same way. [00:08:12] Speaker 01: This is not a circumstance where there are certain settings to the chip, and the settings can give you different results and different functionality. [00:08:22] Speaker 01: This is the same. [00:08:24] Speaker 01: And the best indication is twofold. [00:08:27] Speaker 01: One is that wasn't their initial contention. [00:08:30] Speaker 01: Their initial contention sort of recognized the problem of saying, well, sometimes it's repeated, sometimes it's not. [00:08:37] Speaker 01: by focusing on the inputs to the end gates. [00:08:39] Speaker 01: But they recognized that that didn't fit because they're not irregular. [00:08:44] Speaker 01: They then switched the trial and said, well, sometimes it is and sometimes it isn't. [00:08:48] Speaker 01: But it's only in those circumstances where the enable bit is 1. [00:08:53] Speaker 01: And their expert explicitly said, Your Honor, that when the enable bit is 0, there is no repeated bit. [00:09:01] Speaker 01: If you look at their brief at page 21 to 22, they give you precisely the same example, but they reach precisely the opposite conclusion. [00:09:11] Speaker 01: That can't sustain a verdict. [00:09:14] Speaker 01: And actually, Your Honor, to go to your point, there is no dispute about how the chip works. [00:09:20] Speaker 01: There is no dispute about how the bits are processed. [00:09:22] Speaker 01: When you look at the undisputed evidence, there's simply no infringement here. [00:09:28] Speaker 01: And that's true. [00:09:30] Speaker 01: no matter what you do with the claim construction of repeat. [00:09:34] Speaker 01: It's the manner in which your two questions converge. [00:09:37] Speaker 01: Let me say one more thing and then use a couple minutes to address damages. [00:09:42] Speaker 01: On the 781 patent, the term repeat is not used, nor is the term a regular repeat. [00:09:50] Speaker 04: Your theory is that at summary judgment, they ask that the 781 be construed to include that requirement. [00:09:59] Speaker 04: And I guess as a result of judicial estoppel or whatever, that they're stuck with that. [00:10:06] Speaker 01: Your Honor, we don't claim judicial estoppel. [00:10:08] Speaker 01: We don't need to go that far. [00:10:09] Speaker 01: What we say is they made that contention and the district court adopted that contention. [00:10:13] Speaker 01: They agree today that they've made that contention and it's correct. [00:10:17] Speaker 01: Our position is, [00:10:19] Speaker 01: the court that we then asked to instruct the jury that that was the meaning of the claim and there was no instruction. [00:10:26] Speaker 01: And the only answer to your honors is, well, there was a stipulation that that's a correct instruction. [00:10:31] Speaker 04: And what we would say... Clearly to give the instruction would be harmless error if they're right on infringement of the 710, right? [00:10:42] Speaker 01: Your honor, if they were right on reuse, it still would be [00:10:45] Speaker 01: it still would not be a harmless error, because the jury did not understand the breadth. [00:10:52] Speaker 01: The jury could have misconstrued the breadth of the claim. [00:10:55] Speaker 04: And so it still would be an element. [00:10:57] Speaker 04: If the jury found that element satisfied as to the 710, and if that were correct, [00:11:03] Speaker 04: then why wouldn't the jury have reached the same conclusion with respect to the 781 if they'd been properly instructed? [00:11:10] Speaker 01: Your Honor, because of the manner in which the trial unfolded, which was their expert saying that the patent, 781 patent, required irregular repeating. [00:11:22] Speaker 01: Then our expert got up and gave all his testimony and all the patents, including the 781, and said it requires irregular repeating. [00:11:29] Speaker 01: Then their expert came back and said, no, no, no. [00:11:32] Speaker 01: 781 doesn't have those words. [00:11:34] Speaker 01: It's not required. [00:11:34] Speaker 01: Exactly the opposite of what they said. [00:11:38] Speaker 04: I don't think you're addressing my contention about why is it that if properly instructed, and I think you have a strong argument, they should have been instructed. [00:11:51] Speaker 04: But if properly instructed, why would they have reached a different conclusion with respect to the 781 as opposed to the 710 on this issue? [00:12:02] Speaker 01: The answer to the question is, our experts' reliance on the fact that irregular repeat was required in the 781 was used to attack his opinions across all of the patterns. [00:12:17] Speaker 01: On rebuttal, they came back and said, the 781 doesn't require it. [00:12:21] Speaker 01: And then in closing, they said, oh, the 781 doesn't require it. [00:12:25] Speaker 01: Look at what their expert said. [00:12:26] Speaker 01: You can disregard everything he said on the whole attempt. [00:12:29] Speaker 01: So they used it to attack their expert. [00:12:32] Speaker 01: Yeah. [00:12:33] Speaker 01: It's prejudice as the trial unfolded. [00:12:36] Speaker 01: I know I'm into my rebuttal time. [00:12:37] Speaker 01: Let me make three quick points on damages. [00:12:40] Speaker 01: These two hypothetical negotiations that are offered to the court [00:12:45] Speaker 01: There's no precedent for them. [00:12:47] Speaker 01: There's no evidentiary basis for them. [00:12:49] Speaker 01: They are inconsistent with all the apportionment rules this court has articulated. [00:12:53] Speaker 01: The requirement that the entire market value will not be violated, the smallest stable unit. [00:13:01] Speaker 01: These are the same chips with the same functionality made by the same manufacturer sold, for the most part, for use in the same end product. [00:13:14] Speaker 01: If those end products were never imported in the United States, the royalty is one fifth of what they claim Apple should pay. [00:13:22] Speaker 01: You can't reconcile that with any of those fundamental precepts of apportionment law. [00:13:29] Speaker 01: Point two, you can't solve that by claiming that there are comparable agreements. [00:13:34] Speaker 01: The Caltech agreement with Hughes is not comparable. [00:13:37] Speaker 01: It says on its face it shouldn't be used for a hypothetical negotiation purpose. [00:13:41] Speaker 01: It deals with a different technology, and in fact, excludes the technology of the hypothetical negotiation. [00:13:47] Speaker 01: It's at a different point in time by five years, for sure. [00:13:51] Speaker 01: And perhaps most importantly, it's a $5 million lump sum, not a running royalty. [00:14:00] Speaker 01: You can't claim that that's comparable. [00:14:03] Speaker 01: But even if you did, you have to apportion to address those differences. [00:14:09] Speaker 01: Last point. [00:14:11] Speaker 01: Even if you know all of that and you get to Dr. Kies' imputed royalty rates, he had two bases for increasing the Broadcom rate and the Apple rate. [00:14:23] Speaker 01: In both circumstances, his predicate for increasing those was excluded trial. [00:14:30] Speaker 01: Excluded. [00:14:31] Speaker 01: He had no other basis. [00:14:32] Speaker 01: Yet he went ahead and said, well, it's OK. [00:14:34] Speaker 01: We're going to increase them on a qualitative basis. [00:14:37] Speaker 01: That's insufficient on the discourse, President. [00:14:40] Speaker 01: And I'll reserve my three seconds. [00:14:42] Speaker 02: Before you go, you said you need a full rebuttal time, because there's a lot of issues. [00:14:49] Speaker 02: Morning, Ms. [00:14:50] Speaker ?: Sullivan. [00:14:57] Speaker 00: Good morning, Your Honors, and may it please the Court, Kathleen Sullivan for California Institute of Technology. [00:15:03] Speaker 00: The Court should affirm there was no legal error, any possible legal error. [00:15:07] Speaker 04: To some of the infringement questions, the district court [00:15:12] Speaker 04: basically said that our decision in Shaw with respect to the invalidity contentions and the estoppel resulting from the IPR had been overruled by SAS. [00:15:29] Speaker 04: That seems to me to be pretty problematic. [00:15:32] Speaker 00: Not so, Your Honor. [00:15:34] Speaker 00: May I distinguish the cases? [00:15:36] Speaker 04: Well, yeah, go ahead. [00:15:38] Speaker 00: In Shaw, this court said that arguments for [00:15:42] Speaker 00: invalidity to the board that had been petitioned for in an IPR but had not been instituted did not give rise to estoppel. [00:15:50] Speaker 00: Our case is different so Shaw does not control because in our case the judge excluded arguments about validity that could have been made in the petition but were not. [00:16:03] Speaker 04: Yeah, but the reasoning of Shaw is that what could have been raised is not could have been raised in the petition but could have been raised once the [00:16:12] Speaker 04: IPR was instituted. [00:16:14] Speaker 04: And the district judge ignored that rationale and basically dismissed Shaw because of SAS. [00:16:24] Speaker 04: And it just doesn't seem to me that that's correct. [00:16:27] Speaker 04: Even under SAS, there's no requirement of institution. [00:16:31] Speaker 04: You can have situations in which there is institution. [00:16:37] Speaker 04: They dismiss the institution, or they never institute in the first place. [00:16:43] Speaker 04: So there are situations in which the unpatentability contentions won't have been addressed at all. [00:16:52] Speaker 04: But how could it be that there was an intention to create an estoppel under those situations? [00:16:59] Speaker 00: Your Honor, we think that the majority of district court decisions trying to figure out what SAS means for non-instituted grounds, non-petitioned grounds, is correct. [00:17:10] Speaker 00: The district court cited the majority of decisions at appendix pages 42 to 50. [00:17:16] Speaker 00: Your Honor, SAS makes Shaw unlikely to recur because now any petitioned grounds must be instituted [00:17:25] Speaker 00: But this case is different and not controlled by Shaw, because here the problem is that the defendants knew about the Luby 97 and 98 and Richardson 99 references. [00:17:36] Speaker 00: There's no dispute about that. [00:17:38] Speaker 00: And they withheld them from their petition. [00:17:40] Speaker 00: They are now foreclosed by what we think is the plain language of the statute, Your Honor. [00:17:45] Speaker 00: 315E2, a stops new invalidity grounds that defendants reasonably could have raised in their IPR petitions because the emphasis is on unraised. [00:17:54] Speaker 00: And they could have raised grounds they knew about. [00:17:56] Speaker 00: It's like collateral estoppel. [00:17:57] Speaker 04: Have you stuck with the rationale of Shaw eudelutes, right? [00:18:03] Speaker 00: Well, Your Honor, Shaw said that if I could make the analogy. [00:18:08] Speaker 00: Your Honor, I'm going to run out of time to address the other issues, so I hope I may have more time. [00:18:12] Speaker 04: If we were stuck with the rationale of Shaw, you would lose. [00:18:16] Speaker 00: Your Honor, if the rationale of Shaw is that institution is a moment in time when the proceeding begins, then yes, we would lose. [00:18:23] Speaker 00: But we don't think that's key to Shaw. [00:18:25] Speaker 00: We think it's just like collateral estoppel, Your Honor. [00:18:28] Speaker 00: In collateral estoppel, the [00:18:31] Speaker 00: A party is precluded in future cases from raising those grants it raised or could have raised in another proceeding that reached finality. [00:18:42] Speaker 00: There's an exception, of course, if you jurisdictionally could not have raised those issues in the prior case. [00:18:48] Speaker 00: You were barred, and that turned out to be wrong. [00:18:50] Speaker 00: It would be unfair to stop you for reasons that were not your fault. [00:18:54] Speaker 00: Here we read 315E2 exactly the same. [00:18:57] Speaker 00: You are a stop if you raised or could have raised the grounds for invalidity in the prior proceeding. [00:19:04] Speaker 00: You are a stop. [00:19:05] Speaker 00: You're only excused if there was some reason beyond your control that prevented you from raising them. [00:19:12] Speaker 02: And here the- Counsel, I want to ask you about damages. [00:19:15] Speaker 00: Yes, Your Honor. [00:19:15] Speaker 02: About this hypothetical negotiation. [00:19:18] Speaker 02: It is hypothetical. [00:19:19] Speaker 02: It wasn't an actual negotiation where people agreed to these things. [00:19:23] Speaker 02: Now, what reasonable seller to the public would agree to such a different royalty rate for its supplier when the supplier indemnified the seller? [00:19:48] Speaker 02: I mean, these royalty rates [00:19:51] Speaker 02: do not seem reasonable. [00:19:55] Speaker 02: They don't seem rational. [00:19:57] Speaker 00: So Your Honor, it's entirely reasonable for there to be two different royalty rates for two different defendants selling two different products at two different places in the supply chain. [00:20:08] Speaker 00: In the same line. [00:20:09] Speaker 00: Yes, Your Honor, because the hypothetical negotiation is a construct. [00:20:13] Speaker 00: It is a defendant-specific construct. [00:20:16] Speaker 00: Georgia-Pacific looks to a host of factors that are specific to the relationship between the willing licensee and the willing licensor. [00:20:22] Speaker 00: Factors 2, 5, 6, 11, they all say, look at the defendant. [00:20:26] Speaker 00: And Your Honor, when we're in that hypothetical negotiation here, there were two negotiations, one with Broadcom, one with Apple. [00:20:33] Speaker 00: We're not thinking during those separate hypothetical negotiations about, well, might there be indemnification or might there be carve-outs. [00:20:42] Speaker 00: We're not thinking about that because we're deciding what would the willing licensor Caltech [00:20:46] Speaker 00: require of a willing licensee Broadcom as a tub on its own bottom and Apple as a tub on its own bottom. [00:20:54] Speaker 00: The answer is different rates. [00:20:56] Speaker 04: But the Apple chips are provided by Broadcom. [00:21:01] Speaker 04: You might have a situation which had two different defendants making two different products, which would end up with a different royalty rate. [00:21:09] Speaker 04: That's not our situation. [00:21:10] Speaker 04: It's the same chip. [00:21:12] Speaker 04: And you're saying that some of the Broadcom chips [00:21:14] Speaker 04: require a higher license fee than others. [00:21:18] Speaker 04: I have difficulty understanding how that could be the case. [00:21:21] Speaker 00: Your Honor, imagine if it were two separate cases. [00:21:24] Speaker 00: If these cases hadn't been tried together, there wouldn't be any argument that the Caltech Hughes license, which was highly comparable because it involves the technology in this case, [00:21:34] Speaker 00: would have led a reasonable hypothetical negotiation rate to be the rate calculated beginning with the Caltech Hughes license. [00:21:43] Speaker 00: Similarly, if we were over in a separate case and there was a hypothetical negotiation being discussed about Caltech and Broadcom, we would have looked to the CSIRO Broadcom license as the starting point. [00:21:56] Speaker 00: Those real world licenses, Your Honor, are the best evidence that it is reasonable to suppose that the same chips [00:22:03] Speaker 00: sold to a manufacturer on the one hand, and supplied to an end user on the other hand, will command different rates. [00:22:10] Speaker 00: The starting points were five times different. [00:22:12] Speaker 00: And Your Honor, the apportionment reference my friend on the other side made is irrelevant here, because this, like CSIRO against Cisco, like Elbit, this is a case where we're looking at comparable benchmark licenses to calculate the rate. [00:22:27] Speaker 00: We're not looking to a royalty base we then have to apportion. [00:22:31] Speaker 00: And so the very fact that the real world license, the Hughes Caltech license, had an imputed royalty rate five times the royalty rate imputed from the CSIRO Broadcom license, shows that end user licensing will command a different value in a hypothetical negotiation than [00:22:53] Speaker 00: of component licensing, even if the component later goes into the calculation. [00:22:59] Speaker 00: And Your Honor, I do want to distinguish laser dynamics. [00:23:01] Speaker 00: This isn't an imputed infringement case, where we might think that the direct infringer should pay the same rate as the person found to have been induced to infringe. [00:23:14] Speaker 00: Sorry, induced infringe. [00:23:16] Speaker 00: It's not an induced infringement case. [00:23:18] Speaker 00: Here the jury was allowed to find direct infringement against both defendants. [00:23:23] Speaker 00: There was no request by the other side to particularize the verdict form into indirect or direct infringement. [00:23:30] Speaker 00: And for all those reasons, Your Honor, this isn't a case where you should think about the negotiation as having [00:23:35] Speaker 00: an indemnification aspect or a carve-out aspect. [00:23:40] Speaker 00: So if you imagine this as two separate cases, we wouldn't be here. [00:23:44] Speaker 00: We wouldn't think there was any problem with looking at the Hughes license for Apple and the CSIRO license for Broadcom. [00:23:52] Speaker 00: And I think it would be very surprising if this court were to announce that a patentee who enjoys a legal monopoly is legally obligated to offer the same royalty rate to all bargaining cards. [00:24:05] Speaker 04: Different cases come out differently, and they're inconsistent. [00:24:09] Speaker 04: But within the same case, you want the approach to be consistent. [00:24:14] Speaker 04: And that's the problem here. [00:24:15] Speaker 04: It seems not to be consistent to say here that some of the Broadcom chips get one license and the others, which are absolutely identical, get a different rate. [00:24:28] Speaker 00: Your Honor, there is no legal precedent that forecloses two different rates to two different defendants in the same case. [00:24:37] Speaker 00: Laser dynamics doesn't apply. [00:24:39] Speaker 00: This isn't an entire market value apportionment case. [00:24:42] Speaker 00: This isn't a first sale exhaustion case. [00:24:44] Speaker 00: They don't have a single case. [00:24:45] Speaker 02: Of course, the Court of Appeals doesn't need a legal precedent. [00:24:50] Speaker 00: Exactly, Your Honor. [00:24:50] Speaker 02: It needs no contrary law. [00:24:52] Speaker 00: Correct, Your Honor. [00:24:53] Speaker 00: There's no law that forbids what happened here. [00:24:58] Speaker 00: So that means we're in the world of substantial evidence, Your Honor. [00:25:01] Speaker 00: And stunningly, Mr. Lee has gotten up to date. [00:25:04] Speaker 04: I thought there was a legal issue here. [00:25:07] Speaker 00: I would contend, Your Honor, there is no legal bar to having two separate royalty rates for two separate defendants, where it's based on justified comparable licensing methodology, where the comparable licenses involved two different rates. [00:25:21] Speaker 00: And Your Honor, then the question just becomes, is the damages judgment supported by substantial evidence? [00:25:27] Speaker 00: And it clearly is. [00:25:30] Speaker 00: Mr. Lee got up today and started making jury arguments that were conspicuously not made at trial. [00:25:35] Speaker 04: Unless Judge Laurie or Judge Lynn have other questions on this damages issue, before we run out of time, I wanted to talk to you about infringement. [00:25:45] Speaker 04: Because it seems to me pretty clear that the theory [00:25:50] Speaker 04: is that because some of the information bits are the same at the end of the process, some of them, that creates infringement. [00:26:03] Speaker 04: It's a sort of a sometimes infringement argument. [00:26:07] Speaker 04: We don't even know how often that happens. [00:26:10] Speaker 04: I have difficulty in understanding how the idea that the bits only occasionally get repeated brings them within [00:26:20] Speaker 04: the claims here. [00:26:21] Speaker 04: And would you please explain to me how that is consistent with the invention and with the language of the patent, which doesn't say anything about occasional bits being the same at the end of the process as the input? [00:26:41] Speaker 04: Why is that infringement? [00:26:44] Speaker 04: How can that be infringement under this patent? [00:26:47] Speaker 04: Under your theory, [00:26:49] Speaker 04: It could happen 1% of the time, and it would still be infringement. [00:26:53] Speaker 00: Not so, Your Honor. [00:26:54] Speaker 00: Let me explain. [00:26:54] Speaker 00: The evidence clearly supports the finding of infringement, because exactly what you observed, Your Honor, is that not every information bit that is input into the repetition, permutation, summation, accumulation process in these encoders, not every bit [00:27:14] Speaker 00: is repeated the same number of times. [00:27:17] Speaker 00: That is the invention, Your Honor. [00:27:18] Speaker 04: No, no, no. [00:27:19] Speaker 04: Not the same number of times. [00:27:20] Speaker 04: It doesn't get repeated at all. [00:27:22] Speaker 04: Because in some percentage of the cases, which could be 99%, they don't get repeated at all. [00:27:32] Speaker 00: Your Honor, the invention [00:27:34] Speaker 00: is irregular repetition, meaning the information bits are repeated an irregular number of times. [00:27:41] Speaker 04: I think it would help if we could repeat it an irregular number of times, but still repeat it. [00:27:47] Speaker 04: And your problem is that under your own testimony, your witnesses admitted that they don't always get repeated. [00:27:58] Speaker 04: Sometimes, could be many times, they don't get repeated. [00:28:01] Speaker 00: That's not correct, Your Honor. [00:28:03] Speaker 00: I wonder if we could look to the blue brief at page 29 to the chart. [00:28:09] Speaker 00: Maybe I can clear some of this up. [00:28:11] Speaker 00: This is a picture by the defendants of what is happening when the information bits are input. [00:28:19] Speaker 00: And then some are enabled. [00:28:21] Speaker 00: Some are allowed to be repeated at the end. [00:28:23] Speaker 00: And Your Honor, as Mr. Lee and I agree, when the enabled [00:28:29] Speaker 00: bit, when the parity check bit is set to 1 in the center column, then the information bit that is input into the encoder is passed through to the output. [00:28:40] Speaker 00: So on lines 2, information bit 0 passes through an enabled gate parity check bit 1 and comes out as 0. [00:28:48] Speaker 00: And sometimes it is not repeated. [00:28:52] Speaker 00: It is repeated in lines 2 and 4. [00:28:54] Speaker 04: Sometimes it is not repeated because your witness is testified, right? [00:28:59] Speaker 00: Correct, Your Honor, but that's what makes it irregular repetition. [00:29:02] Speaker 00: It means repeated an irregular or unequal number of times. [00:29:07] Speaker 03: But it still has to be repeated. [00:29:09] Speaker 03: And here we have a situation which is not repeated at all. [00:29:13] Speaker 00: Your Honor, every bit is sent into the process. [00:29:17] Speaker 00: Every bit is input. [00:29:19] Speaker 00: The outputs at the end [00:29:21] Speaker 00: allow the information bit to pass through an irregular number of times. [00:29:25] Speaker 00: It is that process that creates the ability to detect errors in the messages and to override them, to correct them. [00:29:35] Speaker 00: So there is nothing in the claims that require that every bit that goes in [00:29:40] Speaker 00: Every information bit that goes out is an output. [00:29:43] Speaker 00: What matters is that they are repeated an irregular number of times. [00:29:47] Speaker 00: And the testimony was clear. [00:29:48] Speaker 00: I can refer your honors to the site. [00:29:50] Speaker 04: Is it not true that under your own testimony, some of these bits are not repeated at all? [00:29:57] Speaker 00: Not every bit passes through. [00:29:59] Speaker 00: Every bit is repeated. [00:30:01] Speaker 00: Every bit enters the process and is scrambled, and some are output at the end of the day. [00:30:07] Speaker 00: That's what makes it irregular. [00:30:10] Speaker 04: Is it not true that your witness has testified that some bits are not repeated at all? [00:30:17] Speaker 00: No, Your Honor. [00:30:17] Speaker 00: Our witness testified that all bits are repeated, all inputs, all bits are repeated, not every output [00:30:26] Speaker 00: of the process has the information bit as the output at the end. [00:30:30] Speaker 00: It doesn't require that all input bits be output. [00:30:34] Speaker 04: Let's put it this way. [00:30:35] Speaker 04: Suppose we conclude that, in fact, what I said is correct, that not all of the bits are repeated. [00:30:44] Speaker 04: Do you agree then that there is non-infringement? [00:30:54] Speaker 00: The bits are repeated an irregular number of times. [00:30:58] Speaker 04: No, no. [00:30:58] Speaker 04: Please answer my question. [00:31:00] Speaker 04: If I am correct in that the testimony of your witness is that not all of the information bits are repeated, do you agree that there is no infringement? [00:31:13] Speaker 00: Yes, but that is not what the testimony was. [00:31:15] Speaker 00: The testimony was not that not all bits are repeated. [00:31:19] Speaker 00: The testimony of the expert, which is summarized at Redbrief 22, was that all the products output information bits between 3 and 12 times. [00:31:30] Speaker 00: That is an unequal number of times. [00:31:32] Speaker 00: They are all repeated, Your Honor. [00:31:34] Speaker 00: They are all reused in the process. [00:31:35] Speaker 00: You can see this easily in the 03-2 patent in claim 11 in the Tanner graph, or in the Tanner graphs in the other two patents. [00:31:43] Speaker 00: They show that the information bit being input is repeated an unequal number of times. [00:31:48] Speaker 00: The repetition happens when the information bits are input. [00:31:53] Speaker 00: The fact that they are not all output at the end of the repetition, permutation, summation, and accumulation process, the fact that they are not all output does not mean that they are not all repeated. [00:32:06] Speaker 00: They are all reused. [00:32:07] Speaker 00: And Your Honor, we agree, first of all, [00:32:10] Speaker 00: that the claim construction was that repeat means reuse or duplicate. [00:32:15] Speaker 00: So you're right. [00:32:15] Speaker 00: Nothing turns on Mr. Lee's suggestion that it should be exclusively limited to duplicate, which would violate the claim meaning of the claim. [00:32:22] Speaker 00: All of the bits are repeated in the sense of reused. [00:32:25] Speaker 00: If I could call your attention to the Tanner graph, it's easily accessed in the red brief in the front matter. [00:32:31] Speaker 00: It shows the information bits being repeated an unequal number of times. [00:32:37] Speaker 00: The bits being repeated a [00:32:40] Speaker 00: an irregular or unequal number of times. [00:32:42] Speaker 00: Information bits U1 are repeated twice. [00:32:45] Speaker 00: Information bits below that are repeated three times. [00:32:48] Speaker 00: Information bits down below further are repeated four times. [00:32:53] Speaker 00: The repetition, please don't, Your Honor, equate the fact that not every information bit is output at the end. [00:33:00] Speaker 00: with the notion that they're not being repeated. [00:33:03] Speaker 00: The fact that they are repeated an irregular number of times is what creates the error correction technology. [00:33:09] Speaker 00: That was the advance here. [00:33:10] Speaker 00: We had repeat accumulate codes before. [00:33:13] Speaker 00: We didn't have a regular repeat accumulate codes before, and that was the invention. [00:33:19] Speaker 00: So Your Honor, we think that the substantial evidence plainly supports repetition. [00:33:24] Speaker 00: It would do under our construction or theirs, but the district court's construction was correct. [00:33:29] Speaker 00: conscious that I'm over my time. [00:33:31] Speaker 02: Yes, well, thank you. [00:33:34] Speaker 02: I don't think we have more questions for you. [00:33:37] Speaker 02: Mr. Lee, take three and a half minutes, please. [00:33:40] Speaker 00: Your Honor, do you want any response on the 781 instructional error? [00:33:46] Speaker 00: I think you're right. [00:33:47] Speaker 00: It's harmless. [00:33:48] Speaker 00: It's harmless not only for the reasons, Judge Dyck, that you gave. [00:33:50] Speaker 00: First of all, it wasn't error because the claim plainly sets forth what variable number of subsets means. [00:33:59] Speaker 04: And Your Honor, did you ask for the construction that read in the irregular repeat requirement, right? [00:34:06] Speaker 04: In connection with the 101 issue? [00:34:13] Speaker 00: Your Honor, the only point I wanted to make about why it's- No, no, no, no, no. [00:34:17] Speaker 00: I'm sorry. [00:34:17] Speaker 00: I don't understand the question, Your Honor. [00:34:18] Speaker 04: OK. [00:34:20] Speaker 04: Did you not, in connection with the 101 issue, ask for a construction of the 781 that included irregular repeat? [00:34:29] Speaker 00: Included a regular repetition? [00:34:32] Speaker 00: Yes, we did, Your Honor. [00:34:34] Speaker 00: And we don't dispute that irregular repetition is required. [00:34:37] Speaker 04: So why shouldn't the district court have instructed the jury that that was a requirement? [00:34:41] Speaker 00: Because it was already clear from the patent, Your Honor. [00:34:44] Speaker 00: And he didn't want to confuse it by connecting it to the 781. [00:34:47] Speaker 00: Clear from the patent? [00:34:48] Speaker 00: Yes, Your Honor. [00:34:49] Speaker 00: That was the ruling. [00:34:50] Speaker 00: We think it's fair. [00:34:51] Speaker 00: But Your Honor, I want to just point out that our experts said to the jury, this is the irregular repetition part. [00:34:59] Speaker 00: So there can't be prejudice. [00:35:00] Speaker 00: It's harmless error. [00:35:01] Speaker 00: because we didn't argue it wasn't a regular repetition. [00:35:04] Speaker 00: So there's no instructional error on the 781. [00:35:06] Speaker 00: And if there was, it was harmless. [00:35:08] Speaker 00: Thank you. [00:35:09] Speaker 00: Thank you, Your Honor. [00:35:09] Speaker 02: Mr. Lee, you have up to four minutes if you need it. [00:35:14] Speaker 01: Yes, Your Honor. [00:35:15] Speaker 01: Just four points. [00:35:16] Speaker 01: Let me start with the infringement point. [00:35:19] Speaker 01: Judge Dyke, to your question, their expert testified that only sometimes were the bits repeated. [00:35:25] Speaker 01: And that's at pages 303-3, 304-1. [00:35:29] Speaker 01: And they're explicitly asked that question at 303-6. [00:35:32] Speaker 01: So we take Ms. [00:35:33] Speaker 01: Sullivan at her word, which I do, which is if it's only repeated sometimes, there's no infringement. [00:35:38] Speaker 01: There is no infringement here. [00:35:40] Speaker 01: And the infringement theory I heard today [00:35:43] Speaker 01: is, again, different than even the one that's in the briefs. [00:35:46] Speaker 01: And I would just ask the court to consider their brief at page 21 and 22 and compare it to A3038, where literally what they say in the brief to you is the repeated bid is the opposite of what they said at trial. [00:36:05] Speaker 01: It cannot be that that sustains a verdict of this magnitude or any verdict. [00:36:10] Speaker 01: Second point. [00:36:13] Speaker 01: The chronology is important. [00:36:14] Speaker 01: Section 315E2 is enacted, saying during an interparties review. [00:36:19] Speaker 01: Shaw has decided there's no question that it means during interparties review, not in the petition. [00:36:26] Speaker 01: And the district courts read it consistently. [00:36:29] Speaker 01: SAS has decided that Congress doesn't change the language to end the petition. [00:36:35] Speaker 01: The district court changes it, in his opinion. [00:36:38] Speaker 01: But Congress didn't change the statute [00:36:42] Speaker 01: The district courts cannot change the words of the statute simply because they think it might not be as necessary now. [00:36:50] Speaker 01: And there are still circumstances where it's necessary. [00:36:53] Speaker 01: On damages, just two points. [00:36:57] Speaker 01: There was an inducement claim. [00:36:59] Speaker 01: The specific claim was that Broadcom induced Apple to infringe. [00:37:03] Speaker 01: It is very much like the Lucent case. [00:37:07] Speaker 01: And then the answer to Ms. [00:37:08] Speaker 01: Sullivan's question about two [00:37:11] Speaker 01: Cases is threefold. [00:37:13] Speaker 01: There weren't two cases. [00:37:14] Speaker 01: There's a single case with the same product, with the same functionality, made by the same manufacturer, sold to the same end products for the most part, and the only difference is where they were sold. [00:37:29] Speaker 01: It's the same case, and the incremental value should be the same. [00:37:35] Speaker 01: The second answer to her question is, she kept saying that you start with the end value of the product, [00:37:40] Speaker 01: No, the entire market value rule says you cannot do that. [00:37:44] Speaker 01: That's why the entire market value rule exists. [00:37:47] Speaker 01: That's why the smallest saleable unit will exist. [00:37:49] Speaker 01: And the fact they claim the Hughes Caltech license is comparable, which it's not, excuses that adherence to the apportionment principles that the Supreme Court and this court requires is really nonsensical. [00:38:06] Speaker 01: Thank you, Your Honor. [00:38:08] Speaker 02: Thank you, Mr. Lee and Ms. [00:38:10] Speaker 02: Sullivan. [00:38:11] Speaker 02: The case well argued will take place on the submission.