Oral argument — ProMOS Technologies, Inc. v. Samsung Electronics Co., Ltd.

Our next case is ProMOS Technologies, Inc. v. Samsung Electronics Co., Ltd. This is case number 19-1343 and we're hearing the arguments of ProMOS Technologies, Inc. v. Samsung Electronics Co., Ltd. case 19-1344, we're hearing in consolidation. And because we consolidated the arguments, as you all know, we've added five minutes of time if you need it in order to address the single common issue here, okay? And, Mr. Jones, I see you've reserved five minutes of your time for rebuttal. That's correct, Your Honor. Okay, we're ready when you are. May it please the Court, Kevin Jones for appellant. Mr. Jones, just a little housekeeping question, do you agree that Independent Claim 10 of the 507 patent is representative of the claims on appeal? Independent Claim 10 is the only independent claim on appeal, so in that sense it's the only claim that we do need to address today. And the only issue and dispute is in that Independent Claim 10, which is the Board's error in construing the maintaining limitation as plain meaning. I'd like to discuss three primary reasons why that was error. First, the specification and claim language itself require that limitation to ensure that the feedback clock signal follow the input clock by less than 180 degrees. Second, the examiner… Where does the phase limitation, phase difference limitation, specify whether the input clock signal has to be ahead or behind the feedback clock signal? I'm looking at the 507, column 4, lines 51, 53. Yes, Your Honor, and that is, so the limitation reads, maintaining the phase difference between the two clock signals approximately 180 degrees. Now, the parties and the Board agree that there's a term missing there, the word within. So with that correction, what it says is maintaining the phase difference within approximately 180 degrees. Now, Samsung would argue, and the Board found, that that would allow the feedback clock signal to be behind or ahead of the input clock signal. Right. So that was my question? Yes. Where does it say that, that it has to be one or the other as opposed to either? So there's three primary reasons, and the specification, which must be part of the interpretation of the claim under Phillips, requires that in order to practice the invention, the feedback must be behind the input clock signal. More specifically, Figure 2, which is on the face of the patent itself, describes this And without delving into the details of all of the technology, I do want to make it clear there's no dispute that the Comparator 30, at the bottom left of Figure 2, will either invert or not the input clock signal based on whether the feedback clock signal is behind by 180 or less, or behind by 180 to 360. And so we're interpreting the claim as required in light of the specification, and that's the basis for our understanding that the feedback clock signal must be within 180 degrees. Now the claim twice states that the number of delay cells needs to be reduced. In fact, the last element of Claim 10 says a number of delay cells is reduced by approximately one half. And that's repeated throughout the 507 specification in a long string site in our opening brief at 13, almost in 10 places. So the purpose of the invention is to reduce the number of delay cells by one half. That's directly tied to the 180 degree phase difference. Thinking back to high school trigonometry, we have a wave which has a 360 degree cycle. The invention is saying, rather than using delay cells to shift the full 360 degrees of delay, we can halve the delay cells, then we only have to shift from 180 to 0, and we can do that because we invert when the clock is 180 to 360 behind. If you invert, it basically shifts it forward by 180, that's what's done by Comparator 30, and then we only have to use half the number of delay cells to get you as close to zero as you can. That's the point of the invention. And that's why the board's application of plain meaning, allowing the feedback to be behind or ahead, doesn't help. Because in that case, you still have to shift a full 360, because you can only delay, you cannot advance, to get to zero. And so that's why the application of the plain meaning, as the board and Samsung argue, doesn't achieve or have anything to do with the invention, in fact. The prosecution history is also enlightening, and the examiner's treatment of the butcher prior art reference. The butcher prior art reference begins in the appendix at 842. When delay occurs, does delay occur in both directions or just one direction? Delay only occurs in one direction. And that's supported by the written description? Yes, it is, Your Honor. I think that would be an undisputed fact, not in the record, but I think by the basis of the term delay line, for example, in the face of the 507, it would be understood only to delay. So you think your opponent would agree with you that this is, we're only speaking about one direction? Yes, Your Honor. I think that would be an undisputed point of technical fact. And in fact, what's happening inside of a delay line is a series of delay elements. For example, inverters or transistors that basically take your input and delay it by some number of logic gates. It's technically impossible to advance an input clock by adding additional circuitry in series with the input. You don't know that. You'd win a Nobel Prize if you could do it. As I said, Your Honor, it's not a fact in the record. So I've done my best to address the question. The specification demonstrates that the patentee knew how to describe the relationship between the input clock signal and the feedback clock signal in a manner that makes clear which signal lags, one signal lags the other. But it didn't include that language in the claims. Why not? That is correct, Your Honor. In fact, as Samson points out, Claim 13 uses the term follows behind. I think one point to note is that the structure of Claim 13 is different. Here in Claim 10, this limitation is maintaining the phase difference. I think the matter of just plain English speaking and claim drafting, I think it would be extremely awkward to somehow work in the already understood point that the feedback lags by 180. In Claim 13, it's a much more direct stated relationship. In Claim 13, that language, it's found in the determining limitation. So it's determining whether you're 180 behind is effectively what Claim 13 says. Here it's in the maintaining. So you're maintaining the phase relationship within 180 degrees. And I think when read in light of the specification, the claim drafter understood that since that was the invention and what was described, that was the meaning of this element. In the file history, there was a treatment of a butcher reference. The examiner said that butcher taught every element of this claim in its prior form except for this maintaining limitation. So whatever this maintaining limitation means, butcher does not teach it. And I want to make clear up front, there's some ambiguity, I think, in the briefs before this court and also in the board's application of this or treatment of this issue. This is not a prosecution history disclaimer issue. We've never argued that. So the law about clear and unambiguous disclosure is not applicable. This is a case where we're saying the intrinsic record, in this case, the file history, evidences what the examiner understood the term to mean. And specifically, butchers is treating two input clock signals. If I can direct the court to figure one of the butcher reference, which is a... So you're saying there's no disclaimer? That is correct, Your Honor. We're saying it's an implicit definition that was understood by the examiner and the applicant. Now, the two input clocks... But the examiner's statement that none of the prior art teaches doesn't mention whether one signal is ahead or behind the other, correct? And it wasn't endorsed by the patentee. Those two points are correct, Your Honor. The butcher reference, which I'd like to briefly discuss on the details of the technology and why what the examiner must have meant was that the feedback must lag, our conclusion. But to the second point, the applicant's silence does not matter. In fact, in the case that Samsung cites, the Salazar case, it was a case that holds prosecution history disclaimer, or estoppel, cannot apply due to an applicant's silence. That's good law and we don't dispute that. But then Salazar goes on to say that you can absolutely look at statements by the examiner in determining what was understood to a person of skill in the art. And the direct quote from Salazar is, although unilateral statements by an examiner do not give rise to a clear disavowal, it does not necessarily follow that such statements are not pertinent to construing plaintiffs. And that's what we're trying to do. Take the examiner's statement and applicant's silence and interpret the claim accordingly. So your argument is that claim construction would render that limitation meaningless? Yes, Your Honor. So the primary point... So clearly, did you raise this argument in your main brief? It seems like you raised this argument primarily in your server reply. But I want both sides to explain the distinction between the argument on this point that's spelled out in the petition and that's spelled out in the server reply. So Your Honor is talking about the underlying record, not the briefs before this court? Is that your question? Yes. So we did treat this argument directly in our preliminary response at the institution stage. And the quotation from that brief is included in our reply brief at pages 7 and 8. And the quote in the underlying preliminary response is Appendix 831. 831? 831, Your Honor. And this is, as I say, this is also quoted in our reply brief before this court at pages 7 and 8. As you can see, we make the point that Samsung is arguing that, and this is a quote, any DLL where there are two clocks having the same frequency and therefore the same period must necessarily be within 180 degrees of each other. The result of this implied construction is that any prior art references having two such clocks would meet this element. And the point here, Your Honor, is that if the two frequencies are the same, then of course shifting it 180 degrees forward has the effect of reducing the delay cells by half. If you understand the term as the board did, to require only 180 plus or minus, you're within that full 360 degree wave. And because the two frequencies are the same, you're necessarily within 180 degrees plus or minus. If I can just round out the discussion about the treatment of the butcher reference, the two clocks at issue have equal frequencies, and it ties into the point we just discussed. Because they're equal frequencies, the examiner understood that the feedback clock was plus or minus 180 degrees of the input, and therefore would have met the maintaining limitation. Yet the examiner found that that limitation was not explicitly in the butcher reference. So that sort of completes the intrinsic record, both the specification, claim language, and the examiner's treatment of butcher. I have almost five minutes remaining. If there's no other questions at this time, I'd like to reserve the remaining time. Okay. Thank you, Your Honors. Counselor Modi? Yes. Good morning, Your Honors. Same housekeeping question. Independent claim 10, representative. Yes, Your Honor, it is. Thank you. So PROMOS' challenge to the board's claim construction misses the mark. To avoid invalidity, PROMOS asks this court to import a directionality into the claim. But Judge Wallach, as you recognized with your questions, such a requirement is not found in the plain language or supported by the plain language of the claims, which is key. And more importantly, there is no disclaimer or lexicography that permits the plain meaning to be narrowed in such a manner. PROMOS should not be allowed to escape the language it chose, especially in light of the other claim language that we have here and the intrinsic record. Let me start with the claim language if I can. So if you look at the maintaining limitation that's at A131, it simply requires that the phase difference between the input and feedback clock signals be maintained within 180 degrees, nothing more. Judge Wallach, going back to your question, there is actually testimony in the record that supports the point that you were asking PROMOS' counsel about, which is what does the phase difference tell us in this claim? And if you can look at Appendix A933, Paragraph 22. So that's the testimony from their expert. Your Honor, let me know when you're there. Yeah, yeah. Okay. So if you look at Paragraph 22, their expert tells us in relation to the 507 patent, what is the 507 patent teaching us? It teaches us that when the clock delays or advances, slides left or right, it is called askew. So they are talking about the phase difference. There is testimony in the record that you would actually go both ways. And this goes back to one of the questions Judge Reyna, you asked as well. There's more testimony in the record, not only from their expert. Does it go both ways or does it only go one direction? No, Your Honor, it can go both ways. And I think the testimony that I just pointed to supports that. And there's more testimony in the record. If you look at Samsung's expert testimony from Dr. Baker, and that's at A228. And we can look at that as well. So their Samsung expert is explaining the background of the technology. And in Paragraphs 28 and 29 explains why you would have clock signals and why there would be a phase difference and they could be ahead or they could be behind. So the plain meaning language we believe is fatal to their argument because there is testimony in the record that shows that the phase difference, the term maintaining should be given its plain and ordinary meaning. And if you look at this language, again, it does not import any directionality. All it requires is that there be a maximum phase difference no more than 180 degrees. And there's more here. If you look at Claim 13, which I believe was discussed, if you look at the claim, it tells us that the applicant knew, and that's at Appendix 132, Your Honor, it tells us the applicant knew how to claim directionality. They said a feedback clock signal in the DLL follows within a 180-degree phase difference. So PROMAS clearly knew how to claim this. And what's more, we have four different claims, independent claims here. Claims 1, 8, 10, and 13. And if you look at those claims, PROMAS wanted to claim different embodiments in that claim. And they did use different language. And I think they should be held to that, especially in light of this court's case. So where does that leave us? So that leaves us where, if we go back to the Thorner case, this court's case law is clear that claim terms must be given their plain and ordinary meaning to one skill in the art, unless there are only two exceptions. One, patentee sets forth a definition or acts as his own lexicographer. Two, that there is a disclaimer. And they've conceded that there is no disclaimer. I heard counsel say this is an implicit definition case. Well, there is no such implicit definition, either in the specification or the prosecution history. Let me turn to the specification next. If you look at the specification. Is a phase difference language ambiguous in the claim? We do not believe so, Your Honor. We believe, again, if you go back to the testimony that I pointed you to from the experts, including Dr. Baker, Samson's expert, and Mr. Gervasi, their expert, it tells us what they mean by a phase difference, that it could either, the clock signal could either lag or it could be behind. At least in that claim, it could work either way, couldn't it? That's right, Your Honor. In claim time, it could work either way. And we believe if you look at the specification, yes, they do point to the figure 2 embodiment. But as this court's case law makes clear, that's not the be and end all. That's just one embodiment as the board found. And the specification, Your Honor, as you alluded to, does include other language that supports the board's reading of the claim. And that, namely, is at A130. For example, there they used the term behind. It's column 2, lines 63 to 66. They did it again at A131, and that's column 3, lines 3 to 4. And they did it again at A131, column 3, lines 8 to 9. So they clearly knew when to use behind. So they cannot ask this court today to rewrite its claim. And there's support for this language in other places, the board's construction, namely the abstract and the summary of the invention. If you look at the abstract and look at the summary of the invention, it broadly refers to the maintaining limitation. And it, in fact, uses the exact same language that we find in claim 10. What's more, let's go to figure 2. I heard counsel say there is no dispute about the operation of figure 2. There is a dispute about the operation of figure 2. And the dispute, namely, is if you actually look at figure 2 and the way it operates, it actually does not ensure that the feedback clock signal is always behind the input clock within 180 degrees. Let me explain why. So there are instances, for example, and there's testimony in the record on this from their expert, their own expert. We actually went through this scenario with their expert. He admitted, yes, in the situation we gave him, that the circuit would toggle back and forth between switch 1 and switch 2. And you can find that testimony at appendix 1165 to 1167. Now, in their reply brief, they try to come back and say, well, that's not what Mr. Gervasi testified. He did on recross and redirect, he tried to clean up that testimony. But if you actually look at the entirety of that testimony, it will show that Mr. Gervasi was, once he was caught in this sort of, that the circuit of figure 2 doesn't work, he tried to gap fill. But what he did was, and this is at A1172, I think it's important, Your Honor, if you could look at it. At appendix 1172, let me know when you're there. We're there. So there he said, and I'm specifically focusing on page 84, and it's 11 through 13 lines, 11 through 13. He admitted that his understanding was not based on the text of the patent. So he was trying to generalize, and he also said in another place, this unfortunately is not in the appendix. I know the full transcript is certainly available to the court, and that's at Exhibit 1010, pages 81 and lines 8 through 14. He said his understanding, again, he said, was not based on the body of the patent, but instead was an exercise in abstraction and thinking about what the implementation of the abstraction would look like. So figure 2 does not support that. There's more. If we go to figure 3, I think figure 3 is fatal to their argument. They pointed to figure 3 in their reply brief. If I could have the court turn to figure 3, I would like to spend some time on that. And that's at appendix 128. Your Honor, just let me know when you're there. Okay. All right. Okay, so if we look at figure 3, the way figure 3 works is you have the input clock signal on the top, right? That's the CK1, and then you have a feedback clock signal, that's CKF, on the bottom. The way figure 3 works is, and you see there's a shaded region, 32. Do you see that? So the way figure 3 works, and actually let me back up. So the way the phase works in this one is, so if you look at the first rising edge of CK1, right, to the falling edge of CK1, so there's a first rising edge, and then you have a falling edge. That's the 180 degrees, right? And then it goes on. Then you have the 181 to 360 degrees. So the way figure 3 works, it tells us in the specification that if you have a rising edge of the feedback clock signal in the area 32, there would be a switch. The switch would make a switch, right? In figure 2, you would make the switch go from position number 1 to number 2, okay? But more importantly, there is no transition. There is no shift of the switch positions in the areas marked DT2, the non-shaded regions. Well, guess what? What do those areas represent? If you look at their patent, it tells us those, if you do the math, it would be 181 degrees to 215 degrees in that first instance, and in the second instance, it would be 325 to 359 degrees. So what they are telling us in the patent itself that there will not be a shifting. So in other words, the feedback clock signal does not follow the input clock signal every time when the phase difference is within 180 degrees. So, and I think this also, this argument, and if you look at figure 3, I think it also tells us that, as you know, a clock signal is periodic, right? So it will keep going on. So the input clock signal here will keep going on. So this actually figure, we believe, also supports the board's construction in that it shows that you would have a rising edge in that shaded region or outside of the shaded region. That would be actually ahead of the input clock signal. And this is all supported by Mr. Gervasi's testimony, their expert's testimony, and I'll give you the citation for that. And that's at appendix 1175 to 79. So what figure 3 tells us is that, again, in those windows marked DT2, the rising edge of the feedback clock signal would be ahead of the rising edge of the input clock signal and thus would fall within 180 degrees of the input clock signal and not switch, thus meeting the board's construction. So where does that leave us? They pointed out to the purpose of the invention, right? They pointed out that this reduces the delay lines in the circuit. Well, that's already in claim 10. The prior discloses it. So if you can look at claim 10, that's in, again, going back to appendix 131. So if you look at line, column 4, line 57, it says delaying the input clock signal to compensate for the phase difference wherein a number of delay cells utilized is reduced by approximately one half. That language, there is no dispute that it is taught by both the Kim and Donnelly references that are at issue there. So they cannot say that Samsung is somehow reading out limitations out of the claim. So turning to the prosecution history, I heard counsel say the prosecution history is key here. But as the board found, the prosecution history is anything but clear. Let me explain why. First, as counsel also pointed out, the statements that we have in the prosecution history are only from the examiner. And I think we should look at that because I think it's telling. So if we can turn to appendix 393. Let me know when you're there, Your Honor. Yes, Your Honor. Okay. So if you look at 393, this is the paragraph that they are banking their entire argument on. This paragraph says nothing about whether the maintaining limitation requires the feedback clock signal to follow the input clock signal within 180 degrees. Right? There is no such statement there. They are reading too much into this prosecution history. And that cannot be supported under this court's case law, whether you look at it from a disclaimer perspective, whether you look at it from an implicit definition perspective, or even if you look at it more broadly. And I think what's more here, I just don't want this fact to get lost on the court. Claim 11, application claim 11, is the one where the examiner indicated. So if we actually go to the next page, and that's appendix 394, you'll see there's a paragraph 4 there. And it tells the applicant, look, some of your claims are allowable if you rewrite them in independent form. Right? So that's in paragraph 4. Claim 11 was the claim that resulted. It was a dependent claim. They rewrote it to make it an independent claim. And that's what we have here today. Right? But it's the limitations of claim 11 that presumably the examiner found were the reason for the allowance. But we don't know. We just don't know. The examiner doesn't give us reasons why he allowed that claim. But the maintaining limitation that was in the independent claim, claim 8, in the application was rejected. How does a phase difference work if it's not directional? So that is my point. I believe the phase difference, what happens here is if you look at in a clock circuit, so you have an input clock signal, right, which will have a particular phase, and then you'll have a feedback clock signal with a particular phase. Right? And the goal of the DLL here is, in these circuits, is to align the phases. Right? You want the input and the output generally to have the same phase. Right? So that the rising edge of the input clock matches up with the rising edge of the feedback clock signal. And that's what these circuits do. That's what a DLL does. And if you look at the… So are you saying the delay happens only in one direction? No, Your Honor. It can happen in both directions. So if you had an example where, let's say, the feedback clock signal was ahead of the input clock signal… Your Honor, respectfully, it doesn't. And the reason for that is if you look at the limitation here, it's a very broad limitation. All it tells us is that you maintain the phase difference within 180 degrees. Right? It doesn't tell you the directionality. And that's our point. They should not be allowed to read a directionality requirement into this claim. And I think Butcher is key here. Right? They keep on pointing to Butcher and suggesting… that you're importing a limitation if it's ambiguous. I understand, Your Honor. So, Your Honor, we don't… We merely engage in a claim construction. I understand, Your Honor. With all due respect, Your Honor, we do not believe it's ambiguous. We think it is clear based on the expert testimony that I pointed you to before. And in light of the specification. I think their specification tells us if they wanted to be ahead or behind. Certainly, behind is mentioned in their specification. And even their expert agreed that you would have in this type of a situation, signals that are both ahead and that are both behind. So, the phase… That's what this claim is directed to is that you will adjust the phase difference to make sure that the feedback clock signal and the input clock signal is within 180 degrees. The claim requires nothing more. And on the directionality, Judge Arena, just to make sure I get my point across, they did have in Claim 13… If you look at Claim 13, they did say that… And this is at Appendix 132, Column 5. They said they're determining whether feedback clock signal in the DLL follows within a 180-degree phase difference behind the input clock signal. So, they knew how to claim directionality. And I think, Judge Arena, another point to make here is there is nothing in the… Even if you were to conclude it's ambiguous, we would disagree with that. But even if you were to do that, I think when you go to the specification, when you go to the prosecution history, they do not tell us, they do not compel the kind of reading that they want from the claim. It doesn't compel us that it has to be behind as they would have this court construe these claims for the reasons that we discussed. Your argument is that there has to be some sort of disclaimer in the written description. That's right, Your Honor. That's not Phillips, is it? I mean, if we're interpreting, if we're engaging claim construction, then how is… and then the construction is narrower of two constructions, then why is there a requirement that there has to be a disclaimer? So, Your Honor, with all due respect, I certainly think you can look at the prosecution history and specification. It seems to me that our law is that you require a disclaimer or a special type of lexicography whenever you're claiming that a particular term has special meaning and that special meaning is not the plain meaning of the term. Right, and I think they are claiming a special meaning here, Your Honor. They are claiming that within means behind. The plain meaning of within is not behind, right? And I pointed you to expert testimony that we have here, and I think the Torner case is telling, and that's the rubric that the board used here. And I just want to make a quick point on Butcher because they… Well, if the plain meaning of within included behind, you wouldn't have behind in Claim 13. That's right. That's right. And within, in our view, includes both behind and ahead, right? That's the plain meaning of this claim term, and that's what this court should do. And on Butcher, I'm just going to give you one site, which is Column 11, and that's Appendix 843. If you look at Butcher, they keep on saying that the clock frequency in Butcher, they are the same. Sir, if I could just give you one site on Butcher. Did you say 843? Right, Your Honor. That's where Butcher starts, but I'm going to give you the exact page number. So it's Appendix 854, and it's the table in Column 11. So in Butcher, if you look at the way Butcher works, the CLKREF frequency is listed as 10.24 megahertz. That would be the input clock in Butcher. And then you have the received clock frequency. You can see that a few lines down, and that's 80 kilohertz. So Butcher gives you examples where the phase difference is not the same. And we think that's fatal to their argument in terms of the clock, or that the clock signals do not have the same frequency. I think that's fatal to their argument with respect to Butcher. So then what's left? It's a superfluous language, and I want to address, Judge Reyna, your question on waiver. You had asked me a question earlier on waiver. And here, PUMAS did have a waiver. We are not saying that the superfluous argument they didn't make before the board, that's not what we're saying. It's the underlying factual arguments that they're making, that's what they waived. So what they did was, in their third reply before the board for the first time, they pointed to some testimony from our expert and said, if you look at this testimony, this shows that in a DLL, it would always be the case that the input clock signal and the feedback clock signal will always have the same frequency. Well, first of all, if you look at that testimony, it doesn't say that. But that's what the board said was waived, and that's what we're saying was waived on appeal, and they do not challenge the board's evidentiary finding on that issue. So they cannot rely on that testimony was our point. But even if you look at it again, it doesn't support them. And the board here, if you look at what the board did with respect to this issue, I think it's very simple. The board said, even if you look at the maintaining limitation, legally, Samsung is not making the argument here that you are ignoring or reading out limitations. With respect to both Kim and Donnelly, Samsung showed how those references, in those references, the input clock signal and the feedback clock signal, they have the same frequency. And then we didn't stop there. We actually showed them how the DLL limitation meets the rest of the limitations of the claim time. So there is simply no reading out of the limitations under this court's case law. Unless your honors have any questions, other questions, we would directly ask that you affirm. Okay. Thank you. Thank you. Thank you. Jones, you have a little over five minutes of time. Thank you, your honor. If you need that much. I'd like to address three or four of the points that counsel just raised. And first is a technical factual point, and I think maybe a misunderstanding or a miscue between Judge Wallach's question and the answer that you got. And that is the question of whether a delay line can advance or delay, or whether a delay line can delay. Now, aside from the fact that a delay line is called a delay line, there's evidence in the 507 spec that tells us that a delay line can only delay. If you look at Column 3, 507, what page? This would be Appendix 130. Excuse me, Appendix 131. Around lines 23 to 26, they're discussing the benefit of the invention, which, again, was to reduce the delay cells by half. Describing how it works, they say at line 23, regardless of the number of register cells, those are delay cells, suitably the shift register starts at a point close to the beginning of delay line with shifting starting to the right. Now, what this delay line does, when the clocks first come in, there is no delay. On the first iteration, you can delay by one register cell, by one delay unit, and each unit has a fixed period of delay based on device dimensions. If that's not enough, the second time through, you delay by two delay cells. If that's still not enough, you delay by three delay cells. Now, you can, if you're at a point where you have delayed by three delay cells, you can move back and delay by only two delay cells, or even move back to one or zero. But you can never take the original input clock and move forward. Delay can't advance. Correct, which seems straightforward when you say it that way. Now, again, if you're in the prior art, which our invention improved upon, but in the prior art, where you can delay up to 360 degrees, of course, delaying by 300 degrees looks like you have just advanced by 60, because it's a repetitive wave. But the critical difference is you require twice the number of delay cells to get to that point. So I think that's important, because it, again, means that the application of maintaining, as they've argued, is not relevant to the entire 507 spec, or the invention, or the rest of the claim. The last element in the claim, 10, says you reduce the delay cells by half. Yet they're arguing that you're always plus or minus 180, and therefore there's no reduction in delay cells at all, because you still would need 360 degrees worth of delay cells in order to get you close to zero. Now, another point that was discussed is figure two, and our expert, Mr. Gervasi's treatment of figure two. I think it bears worth stating up front, clearly, this is a block diagram. These are not circuit elements. This is not a circuit design. And it needs to be understood as a person of skill in the art would understand it. Mr. Gervasi, on his own accord, as the deposition record states, was very uncomfortable with the initial questioning that Samsung's counsel posed to him. He took several minutes at break, again, as the record states, and came back and cleaned up the record with some explanation of what he calls hysteresis in Comparator 30. It's quite a detailed transcript, and rather than walking you through it now, I would just ask the court. Our brief does treat this fairly, our reply brief, but if you read the transcript in context, it's clear that Mr. Gervasi does not agree that figure two jitters back and forth around zero. Figure two absolutely requires the feedback to be behind the input clock signal. And there was a discussion about figure three in the patent as well, and counsel's discussion about the shaded area in the 507 patent. And he says that figure three teaches that the feedback can be ahead of the input clock. The point of these 20% DT2 periods on either side of the shaded area is to avoid false transitions. Column three states that. And that's the basis for Mr. Gervasi saying feedback would not go ahead of the clock input, because that's a false transition, and figure three supports his conclusion on that point. So I'm just approaching the end of my time. I'd welcome any other questions from your honors. We have your argument, and we have the argument of all the parties. We thank you. Thank you.